Training in progress, step 393, checkpoint
Browse files- last-checkpoint/README.md +202 -0
- last-checkpoint/adapter_config.json +30 -0
- last-checkpoint/adapter_model.safetensors +3 -0
- last-checkpoint/added_tokens.json +4 -0
- last-checkpoint/merges.txt +0 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +30 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +205 -0
- last-checkpoint/trainer_state.json +2800 -0
- last-checkpoint/training_args.bin +3 -0
- last-checkpoint/vocab.json +0 -0
last-checkpoint/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: katuni4ka/tiny-random-dbrx
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.13.2
|
last-checkpoint/adapter_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "katuni4ka/tiny-random-dbrx",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": null,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 8,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"layer",
|
| 24 |
+
"out_proj",
|
| 25 |
+
"Wqkv"
|
| 26 |
+
],
|
| 27 |
+
"task_type": "CAUSAL_LM",
|
| 28 |
+
"use_dora": false,
|
| 29 |
+
"use_rslora": false
|
| 30 |
+
}
|
last-checkpoint/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66bfadf366badbf014115124d5dfc9ad68216430505e0bc72295fab091c16206
|
| 3 |
+
size 5752
|
last-checkpoint/added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|im_end|>": 100279,
|
| 3 |
+
"<|im_start|>": 100278
|
| 4 |
+
}
|
last-checkpoint/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67bf8fddc9546cec40911f74a098f850dddc78879a74d9fe836bd7dc29d40b08
|
| 3 |
+
size 15814
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2800f540ca036b216a08382c344570bf4cc88442e6d1b53b2fcc988d87a1b62d
|
| 3 |
+
size 14244
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b8773234ba45d13c3b7d8e7efcb7f6c34dfa1b070ab841033b2610c341439c8
|
| 3 |
+
size 1064
|
last-checkpoint/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|pad|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,205 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"100256": {
|
| 5 |
+
"content": "<||_unused_0_||>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"100257": {
|
| 13 |
+
"content": "<|endoftext|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"100258": {
|
| 21 |
+
"content": "<|fim_prefix|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"100259": {
|
| 29 |
+
"content": "<|fim_middle|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"100260": {
|
| 37 |
+
"content": "<|fim_suffix|>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"100261": {
|
| 45 |
+
"content": "<||_unused_1_||>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"100262": {
|
| 53 |
+
"content": "<||_unused_2_||>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"100263": {
|
| 61 |
+
"content": "<||_unused_3_||>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"100264": {
|
| 69 |
+
"content": "<||_unused_4_||>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"100265": {
|
| 77 |
+
"content": "<||_unused_5_||>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"100266": {
|
| 85 |
+
"content": "<||_unused_6_||>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"100267": {
|
| 93 |
+
"content": "<||_unused_7_||>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"100268": {
|
| 101 |
+
"content": "<||_unused_8_||>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"100269": {
|
| 109 |
+
"content": "<||_unused_9_||>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"100270": {
|
| 117 |
+
"content": "<||_unused_10_||>",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": false,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": true
|
| 123 |
+
},
|
| 124 |
+
"100271": {
|
| 125 |
+
"content": "<||_unused_11_||>",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": false,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": true
|
| 131 |
+
},
|
| 132 |
+
"100272": {
|
| 133 |
+
"content": "<||_unused_12_||>",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": false,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": true
|
| 139 |
+
},
|
| 140 |
+
"100273": {
|
| 141 |
+
"content": "<||_unused_13_||>",
|
| 142 |
+
"lstrip": false,
|
| 143 |
+
"normalized": false,
|
| 144 |
+
"rstrip": false,
|
| 145 |
+
"single_word": false,
|
| 146 |
+
"special": true
|
| 147 |
+
},
|
| 148 |
+
"100274": {
|
| 149 |
+
"content": "<||_unused_14_||>",
|
| 150 |
+
"lstrip": false,
|
| 151 |
+
"normalized": false,
|
| 152 |
+
"rstrip": false,
|
| 153 |
+
"single_word": false,
|
| 154 |
+
"special": true
|
| 155 |
+
},
|
| 156 |
+
"100275": {
|
| 157 |
+
"content": "<||_unused_15_||>",
|
| 158 |
+
"lstrip": false,
|
| 159 |
+
"normalized": false,
|
| 160 |
+
"rstrip": false,
|
| 161 |
+
"single_word": false,
|
| 162 |
+
"special": true
|
| 163 |
+
},
|
| 164 |
+
"100276": {
|
| 165 |
+
"content": "<|endofprompt|>",
|
| 166 |
+
"lstrip": false,
|
| 167 |
+
"normalized": false,
|
| 168 |
+
"rstrip": false,
|
| 169 |
+
"single_word": false,
|
| 170 |
+
"special": true
|
| 171 |
+
},
|
| 172 |
+
"100277": {
|
| 173 |
+
"content": "<|pad|>",
|
| 174 |
+
"lstrip": false,
|
| 175 |
+
"normalized": false,
|
| 176 |
+
"rstrip": false,
|
| 177 |
+
"single_word": false,
|
| 178 |
+
"special": true
|
| 179 |
+
},
|
| 180 |
+
"100278": {
|
| 181 |
+
"content": "<|im_start|>",
|
| 182 |
+
"lstrip": false,
|
| 183 |
+
"normalized": false,
|
| 184 |
+
"rstrip": false,
|
| 185 |
+
"single_word": false,
|
| 186 |
+
"special": true
|
| 187 |
+
},
|
| 188 |
+
"100279": {
|
| 189 |
+
"content": "<|im_end|>",
|
| 190 |
+
"lstrip": false,
|
| 191 |
+
"normalized": false,
|
| 192 |
+
"rstrip": false,
|
| 193 |
+
"single_word": false,
|
| 194 |
+
"special": true
|
| 195 |
+
}
|
| 196 |
+
},
|
| 197 |
+
"bos_token": "<|endoftext|>",
|
| 198 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
|
| 199 |
+
"clean_up_tokenization_spaces": true,
|
| 200 |
+
"eos_token": "<|endoftext|>",
|
| 201 |
+
"model_max_length": 32768,
|
| 202 |
+
"pad_token": "<|pad|>",
|
| 203 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 204 |
+
"unk_token": "<|endoftext|>"
|
| 205 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,2800 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.00791086687399982,
|
| 5 |
+
"eval_steps": 393,
|
| 6 |
+
"global_step": 393,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 2.0129432249363408e-05,
|
| 13 |
+
"grad_norm": 1.2282008356123697e-05,
|
| 14 |
+
"learning_rate": 2e-05,
|
| 15 |
+
"loss": 46.0,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 2.0129432249363408e-05,
|
| 20 |
+
"eval_loss": 11.5,
|
| 21 |
+
"eval_runtime": 130.2655,
|
| 22 |
+
"eval_samples_per_second": 160.58,
|
| 23 |
+
"eval_steps_per_second": 80.29,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 4.0258864498726816e-05,
|
| 28 |
+
"grad_norm": 2.1924333850620314e-05,
|
| 29 |
+
"learning_rate": 4e-05,
|
| 30 |
+
"loss": 46.0,
|
| 31 |
+
"step": 2
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 6.038829674809022e-05,
|
| 35 |
+
"grad_norm": 1.5113248082343489e-05,
|
| 36 |
+
"learning_rate": 6e-05,
|
| 37 |
+
"loss": 46.0,
|
| 38 |
+
"step": 3
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 8.051772899745363e-05,
|
| 42 |
+
"grad_norm": 1.604706449143123e-05,
|
| 43 |
+
"learning_rate": 8e-05,
|
| 44 |
+
"loss": 46.0,
|
| 45 |
+
"step": 4
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00010064716124681703,
|
| 49 |
+
"grad_norm": 2.237814805994276e-05,
|
| 50 |
+
"learning_rate": 0.0001,
|
| 51 |
+
"loss": 46.0,
|
| 52 |
+
"step": 5
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00012077659349618043,
|
| 56 |
+
"grad_norm": 2.265493640152272e-05,
|
| 57 |
+
"learning_rate": 0.00012,
|
| 58 |
+
"loss": 46.0,
|
| 59 |
+
"step": 6
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00014090602574554385,
|
| 63 |
+
"grad_norm": 1.7647287677391432e-05,
|
| 64 |
+
"learning_rate": 0.00014,
|
| 65 |
+
"loss": 46.0,
|
| 66 |
+
"step": 7
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00016103545799490726,
|
| 70 |
+
"grad_norm": 2.9863820600439794e-05,
|
| 71 |
+
"learning_rate": 0.00016,
|
| 72 |
+
"loss": 46.0,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.00018116489024427065,
|
| 77 |
+
"grad_norm": 1.1106193596788216e-05,
|
| 78 |
+
"learning_rate": 0.00018,
|
| 79 |
+
"loss": 46.0,
|
| 80 |
+
"step": 9
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00020129432249363407,
|
| 84 |
+
"grad_norm": 1.795052594388835e-05,
|
| 85 |
+
"learning_rate": 0.0002,
|
| 86 |
+
"loss": 46.0,
|
| 87 |
+
"step": 10
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00022142375474299748,
|
| 91 |
+
"grad_norm": 1.9582959794206545e-05,
|
| 92 |
+
"learning_rate": 0.0001999997969619787,
|
| 93 |
+
"loss": 46.0,
|
| 94 |
+
"step": 11
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00024155318699236087,
|
| 98 |
+
"grad_norm": 3.401090725674294e-05,
|
| 99 |
+
"learning_rate": 0.00019999918784873927,
|
| 100 |
+
"loss": 46.0,
|
| 101 |
+
"step": 12
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0002616826192417243,
|
| 105 |
+
"grad_norm": 1.5809200704097748e-05,
|
| 106 |
+
"learning_rate": 0.00019999817266275517,
|
| 107 |
+
"loss": 46.0,
|
| 108 |
+
"step": 13
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0002818120514910877,
|
| 112 |
+
"grad_norm": 1.1195018487342168e-05,
|
| 113 |
+
"learning_rate": 0.00019999675140814887,
|
| 114 |
+
"loss": 46.0,
|
| 115 |
+
"step": 14
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.0003019414837404511,
|
| 119 |
+
"grad_norm": 4.638621976482682e-05,
|
| 120 |
+
"learning_rate": 0.00019999492409069166,
|
| 121 |
+
"loss": 46.0,
|
| 122 |
+
"step": 15
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.00032207091598981453,
|
| 126 |
+
"grad_norm": 2.1349093003664166e-05,
|
| 127 |
+
"learning_rate": 0.00019999269071780394,
|
| 128 |
+
"loss": 46.0,
|
| 129 |
+
"step": 16
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0003422003482391779,
|
| 133 |
+
"grad_norm": 1.8559087038738653e-05,
|
| 134 |
+
"learning_rate": 0.0001999900512985548,
|
| 135 |
+
"loss": 46.0,
|
| 136 |
+
"step": 17
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0003623297804885413,
|
| 140 |
+
"grad_norm": 9.40681820793543e-06,
|
| 141 |
+
"learning_rate": 0.00019998700584366238,
|
| 142 |
+
"loss": 46.0,
|
| 143 |
+
"step": 18
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.00038245921273790474,
|
| 147 |
+
"grad_norm": 1.6563231838517822e-05,
|
| 148 |
+
"learning_rate": 0.0001999835543654935,
|
| 149 |
+
"loss": 46.0,
|
| 150 |
+
"step": 19
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.00040258864498726813,
|
| 154 |
+
"grad_norm": 1.7772108549252152e-05,
|
| 155 |
+
"learning_rate": 0.0001999796968780638,
|
| 156 |
+
"loss": 46.0,
|
| 157 |
+
"step": 20
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.0004227180772366315,
|
| 161 |
+
"grad_norm": 3.5355504223844036e-05,
|
| 162 |
+
"learning_rate": 0.00019997543339703757,
|
| 163 |
+
"loss": 46.0,
|
| 164 |
+
"step": 21
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.00044284750948599496,
|
| 168 |
+
"grad_norm": 2.7386544388718903e-05,
|
| 169 |
+
"learning_rate": 0.00019997076393972783,
|
| 170 |
+
"loss": 46.0,
|
| 171 |
+
"step": 22
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.00046297694173535835,
|
| 175 |
+
"grad_norm": 2.447705810482148e-05,
|
| 176 |
+
"learning_rate": 0.0001999656885250961,
|
| 177 |
+
"loss": 46.0,
|
| 178 |
+
"step": 23
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.00048310637398472174,
|
| 182 |
+
"grad_norm": 2.7531290470506065e-05,
|
| 183 |
+
"learning_rate": 0.00019996020717375247,
|
| 184 |
+
"loss": 46.0,
|
| 185 |
+
"step": 24
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0005032358062340851,
|
| 189 |
+
"grad_norm": 1.6084130038507283e-05,
|
| 190 |
+
"learning_rate": 0.00019995431990795531,
|
| 191 |
+
"loss": 46.0,
|
| 192 |
+
"step": 25
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0005233652384834486,
|
| 196 |
+
"grad_norm": 1.0257011126668658e-05,
|
| 197 |
+
"learning_rate": 0.00019994802675161148,
|
| 198 |
+
"loss": 46.0,
|
| 199 |
+
"step": 26
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.000543494670732812,
|
| 203 |
+
"grad_norm": 1.4123150322120637e-05,
|
| 204 |
+
"learning_rate": 0.00019994132773027597,
|
| 205 |
+
"loss": 46.0,
|
| 206 |
+
"step": 27
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.0005636241029821754,
|
| 210 |
+
"grad_norm": 2.4509085051249713e-05,
|
| 211 |
+
"learning_rate": 0.00019993422287115185,
|
| 212 |
+
"loss": 46.0,
|
| 213 |
+
"step": 28
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0005837535352315388,
|
| 217 |
+
"grad_norm": 1.926114782691002e-05,
|
| 218 |
+
"learning_rate": 0.0001999267122030903,
|
| 219 |
+
"loss": 46.0,
|
| 220 |
+
"step": 29
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.0006038829674809022,
|
| 224 |
+
"grad_norm": 2.307241447851993e-05,
|
| 225 |
+
"learning_rate": 0.0001999187957565903,
|
| 226 |
+
"loss": 46.0,
|
| 227 |
+
"step": 30
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.0006240123997302656,
|
| 231 |
+
"grad_norm": 1.905815770442132e-05,
|
| 232 |
+
"learning_rate": 0.00019991047356379866,
|
| 233 |
+
"loss": 46.0,
|
| 234 |
+
"step": 31
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0006441418319796291,
|
| 238 |
+
"grad_norm": 3.1321374990511686e-05,
|
| 239 |
+
"learning_rate": 0.00019990174565850984,
|
| 240 |
+
"loss": 46.0,
|
| 241 |
+
"step": 32
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0006642712642289924,
|
| 245 |
+
"grad_norm": 1.4927420124877244e-05,
|
| 246 |
+
"learning_rate": 0.00019989261207616573,
|
| 247 |
+
"loss": 46.0,
|
| 248 |
+
"step": 33
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0006844006964783558,
|
| 252 |
+
"grad_norm": 6.0364993260009214e-05,
|
| 253 |
+
"learning_rate": 0.00019988307285385566,
|
| 254 |
+
"loss": 46.0,
|
| 255 |
+
"step": 34
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.0007045301287277192,
|
| 259 |
+
"grad_norm": 3.057021604035981e-05,
|
| 260 |
+
"learning_rate": 0.00019987312803031607,
|
| 261 |
+
"loss": 46.0,
|
| 262 |
+
"step": 35
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.0007246595609770826,
|
| 266 |
+
"grad_norm": 2.576452425273601e-05,
|
| 267 |
+
"learning_rate": 0.00019986277764593057,
|
| 268 |
+
"loss": 46.0,
|
| 269 |
+
"step": 36
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.0007447889932264461,
|
| 273 |
+
"grad_norm": 1.8830834960681386e-05,
|
| 274 |
+
"learning_rate": 0.00019985202174272956,
|
| 275 |
+
"loss": 46.0,
|
| 276 |
+
"step": 37
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.0007649184254758095,
|
| 280 |
+
"grad_norm": 2.4923951059463434e-05,
|
| 281 |
+
"learning_rate": 0.0001998408603643902,
|
| 282 |
+
"loss": 46.0,
|
| 283 |
+
"step": 38
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0007850478577251729,
|
| 287 |
+
"grad_norm": 2.467411286488641e-05,
|
| 288 |
+
"learning_rate": 0.00019982929355623615,
|
| 289 |
+
"loss": 46.0,
|
| 290 |
+
"step": 39
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.0008051772899745363,
|
| 294 |
+
"grad_norm": 4.072063529747538e-05,
|
| 295 |
+
"learning_rate": 0.00019981732136523746,
|
| 296 |
+
"loss": 46.0,
|
| 297 |
+
"step": 40
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.0008253067222238997,
|
| 301 |
+
"grad_norm": 2.181486888730433e-05,
|
| 302 |
+
"learning_rate": 0.0001998049438400103,
|
| 303 |
+
"loss": 46.0,
|
| 304 |
+
"step": 41
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.000845436154473263,
|
| 308 |
+
"grad_norm": 1.713978599582333e-05,
|
| 309 |
+
"learning_rate": 0.0001997921610308169,
|
| 310 |
+
"loss": 46.0,
|
| 311 |
+
"step": 42
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.0008655655867226265,
|
| 315 |
+
"grad_norm": 1.962662281584926e-05,
|
| 316 |
+
"learning_rate": 0.00019977897298956515,
|
| 317 |
+
"loss": 46.0,
|
| 318 |
+
"step": 43
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0008856950189719899,
|
| 322 |
+
"grad_norm": 2.0111776393605396e-05,
|
| 323 |
+
"learning_rate": 0.0001997653797698085,
|
| 324 |
+
"loss": 46.0,
|
| 325 |
+
"step": 44
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0009058244512213533,
|
| 329 |
+
"grad_norm": 3.424846727284603e-05,
|
| 330 |
+
"learning_rate": 0.0001997513814267458,
|
| 331 |
+
"loss": 46.0,
|
| 332 |
+
"step": 45
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.0009259538834707167,
|
| 336 |
+
"grad_norm": 2.3754591893521138e-05,
|
| 337 |
+
"learning_rate": 0.00019973697801722095,
|
| 338 |
+
"loss": 46.0,
|
| 339 |
+
"step": 46
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.0009460833157200801,
|
| 343 |
+
"grad_norm": 3.0731138394912705e-05,
|
| 344 |
+
"learning_rate": 0.00019972216959972274,
|
| 345 |
+
"loss": 46.0,
|
| 346 |
+
"step": 47
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.0009662127479694435,
|
| 350 |
+
"grad_norm": 2.0918800146318972e-05,
|
| 351 |
+
"learning_rate": 0.0001997069562343846,
|
| 352 |
+
"loss": 46.0,
|
| 353 |
+
"step": 48
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.000986342180218807,
|
| 357 |
+
"grad_norm": 3.504705455270596e-05,
|
| 358 |
+
"learning_rate": 0.0001996913379829844,
|
| 359 |
+
"loss": 46.0,
|
| 360 |
+
"step": 49
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.0010064716124681702,
|
| 364 |
+
"grad_norm": 2.918963582487777e-05,
|
| 365 |
+
"learning_rate": 0.0001996753149089441,
|
| 366 |
+
"loss": 46.0,
|
| 367 |
+
"step": 50
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.0010266010447175337,
|
| 371 |
+
"grad_norm": 3.13876080326736e-05,
|
| 372 |
+
"learning_rate": 0.00019965888707732953,
|
| 373 |
+
"loss": 46.0,
|
| 374 |
+
"step": 51
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.0010467304769668972,
|
| 378 |
+
"grad_norm": 1.2979964594705962e-05,
|
| 379 |
+
"learning_rate": 0.0001996420545548502,
|
| 380 |
+
"loss": 46.0,
|
| 381 |
+
"step": 52
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.0010668599092162605,
|
| 385 |
+
"grad_norm": 2.1166308215470053e-05,
|
| 386 |
+
"learning_rate": 0.00019962481740985895,
|
| 387 |
+
"loss": 46.0,
|
| 388 |
+
"step": 53
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.001086989341465624,
|
| 392 |
+
"grad_norm": 2.2731392164132558e-05,
|
| 393 |
+
"learning_rate": 0.00019960717571235173,
|
| 394 |
+
"loss": 46.0,
|
| 395 |
+
"step": 54
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.0011071187737149873,
|
| 399 |
+
"grad_norm": 2.4999253582791425e-05,
|
| 400 |
+
"learning_rate": 0.00019958912953396723,
|
| 401 |
+
"loss": 46.0,
|
| 402 |
+
"step": 55
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.0011272482059643508,
|
| 406 |
+
"grad_norm": 6.292035686783493e-05,
|
| 407 |
+
"learning_rate": 0.00019957067894798663,
|
| 408 |
+
"loss": 46.0,
|
| 409 |
+
"step": 56
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.0011473776382137143,
|
| 413 |
+
"grad_norm": 1.5206553143798374e-05,
|
| 414 |
+
"learning_rate": 0.00019955182402933334,
|
| 415 |
+
"loss": 46.0,
|
| 416 |
+
"step": 57
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.0011675070704630776,
|
| 420 |
+
"grad_norm": 3.7179495848249644e-05,
|
| 421 |
+
"learning_rate": 0.0001995325648545727,
|
| 422 |
+
"loss": 46.0,
|
| 423 |
+
"step": 58
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.001187636502712441,
|
| 427 |
+
"grad_norm": 2.6544912543613464e-05,
|
| 428 |
+
"learning_rate": 0.00019951290150191158,
|
| 429 |
+
"loss": 46.0,
|
| 430 |
+
"step": 59
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.0012077659349618043,
|
| 434 |
+
"grad_norm": 1.6469433830934577e-05,
|
| 435 |
+
"learning_rate": 0.00019949283405119815,
|
| 436 |
+
"loss": 46.0,
|
| 437 |
+
"step": 60
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.0012278953672111678,
|
| 441 |
+
"grad_norm": 2.489179496478755e-05,
|
| 442 |
+
"learning_rate": 0.00019947236258392154,
|
| 443 |
+
"loss": 46.0,
|
| 444 |
+
"step": 61
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.0012480247994605311,
|
| 448 |
+
"grad_norm": 1.7665654013399035e-05,
|
| 449 |
+
"learning_rate": 0.00019945148718321143,
|
| 450 |
+
"loss": 46.0,
|
| 451 |
+
"step": 62
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.0012681542317098946,
|
| 455 |
+
"grad_norm": 2.1060941435280256e-05,
|
| 456 |
+
"learning_rate": 0.00019943020793383785,
|
| 457 |
+
"loss": 46.0,
|
| 458 |
+
"step": 63
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0012882836639592581,
|
| 462 |
+
"grad_norm": 1.269436688744463e-05,
|
| 463 |
+
"learning_rate": 0.00019940852492221075,
|
| 464 |
+
"loss": 46.0,
|
| 465 |
+
"step": 64
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0013084130962086214,
|
| 469 |
+
"grad_norm": 2.0220479200361297e-05,
|
| 470 |
+
"learning_rate": 0.0001993864382363796,
|
| 471 |
+
"loss": 46.0,
|
| 472 |
+
"step": 65
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.0013285425284579849,
|
| 476 |
+
"grad_norm": 2.53070866165217e-05,
|
| 477 |
+
"learning_rate": 0.00019936394796603318,
|
| 478 |
+
"loss": 46.0,
|
| 479 |
+
"step": 66
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.0013486719607073482,
|
| 483 |
+
"grad_norm": 3.738961459021084e-05,
|
| 484 |
+
"learning_rate": 0.00019934105420249908,
|
| 485 |
+
"loss": 46.0,
|
| 486 |
+
"step": 67
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.0013688013929567117,
|
| 490 |
+
"grad_norm": 2.57118354056729e-05,
|
| 491 |
+
"learning_rate": 0.0001993177570387434,
|
| 492 |
+
"loss": 46.0,
|
| 493 |
+
"step": 68
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0013889308252060752,
|
| 497 |
+
"grad_norm": 1.631179293326568e-05,
|
| 498 |
+
"learning_rate": 0.00019929405656937032,
|
| 499 |
+
"loss": 46.0,
|
| 500 |
+
"step": 69
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.0014090602574554384,
|
| 504 |
+
"grad_norm": 2.4105151169351302e-05,
|
| 505 |
+
"learning_rate": 0.00019926995289062176,
|
| 506 |
+
"loss": 46.0,
|
| 507 |
+
"step": 70
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.001429189689704802,
|
| 511 |
+
"grad_norm": 6.657966878265142e-05,
|
| 512 |
+
"learning_rate": 0.00019924544610037698,
|
| 513 |
+
"loss": 46.0,
|
| 514 |
+
"step": 71
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.0014493191219541652,
|
| 518 |
+
"grad_norm": 2.2929831175133586e-05,
|
| 519 |
+
"learning_rate": 0.00019922053629815224,
|
| 520 |
+
"loss": 46.0,
|
| 521 |
+
"step": 72
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.0014694485542035287,
|
| 525 |
+
"grad_norm": 2.9039501896477304e-05,
|
| 526 |
+
"learning_rate": 0.00019919522358510024,
|
| 527 |
+
"loss": 46.0,
|
| 528 |
+
"step": 73
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.0014895779864528922,
|
| 532 |
+
"grad_norm": 5.170597432879731e-05,
|
| 533 |
+
"learning_rate": 0.00019916950806400983,
|
| 534 |
+
"loss": 46.0,
|
| 535 |
+
"step": 74
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.0015097074187022555,
|
| 539 |
+
"grad_norm": 2.5796563932090066e-05,
|
| 540 |
+
"learning_rate": 0.00019914338983930557,
|
| 541 |
+
"loss": 46.0,
|
| 542 |
+
"step": 75
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.001529836850951619,
|
| 546 |
+
"grad_norm": 2.303555083926767e-05,
|
| 547 |
+
"learning_rate": 0.0001991168690170474,
|
| 548 |
+
"loss": 46.0,
|
| 549 |
+
"step": 76
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.0015499662832009823,
|
| 553 |
+
"grad_norm": 3.512139664962888e-05,
|
| 554 |
+
"learning_rate": 0.00019908994570492993,
|
| 555 |
+
"loss": 46.0,
|
| 556 |
+
"step": 77
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0015700957154503458,
|
| 560 |
+
"grad_norm": 3.3836728107417e-05,
|
| 561 |
+
"learning_rate": 0.00019906262001228228,
|
| 562 |
+
"loss": 46.0,
|
| 563 |
+
"step": 78
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.001590225147699709,
|
| 567 |
+
"grad_norm": 2.430832500976976e-05,
|
| 568 |
+
"learning_rate": 0.00019903489205006764,
|
| 569 |
+
"loss": 46.0,
|
| 570 |
+
"step": 79
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.0016103545799490725,
|
| 574 |
+
"grad_norm": 3.29008289554622e-05,
|
| 575 |
+
"learning_rate": 0.0001990067619308825,
|
| 576 |
+
"loss": 46.0,
|
| 577 |
+
"step": 80
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.001630484012198436,
|
| 581 |
+
"grad_norm": 2.6406103643239476e-05,
|
| 582 |
+
"learning_rate": 0.00019897822976895665,
|
| 583 |
+
"loss": 46.0,
|
| 584 |
+
"step": 81
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.0016506134444477993,
|
| 588 |
+
"grad_norm": 1.8888900740421377e-05,
|
| 589 |
+
"learning_rate": 0.00019894929568015226,
|
| 590 |
+
"loss": 46.0,
|
| 591 |
+
"step": 82
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.0016707428766971628,
|
| 595 |
+
"grad_norm": 3.3854525099741295e-05,
|
| 596 |
+
"learning_rate": 0.0001989199597819638,
|
| 597 |
+
"loss": 46.0,
|
| 598 |
+
"step": 83
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.001690872308946526,
|
| 602 |
+
"grad_norm": 2.4574126655352302e-05,
|
| 603 |
+
"learning_rate": 0.0001988902221935173,
|
| 604 |
+
"loss": 46.0,
|
| 605 |
+
"step": 84
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.0017110017411958896,
|
| 609 |
+
"grad_norm": 1.5161581359279808e-05,
|
| 610 |
+
"learning_rate": 0.00019886008303557,
|
| 611 |
+
"loss": 46.0,
|
| 612 |
+
"step": 85
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.001731131173445253,
|
| 616 |
+
"grad_norm": 2.6385316232335754e-05,
|
| 617 |
+
"learning_rate": 0.00019882954243050972,
|
| 618 |
+
"loss": 46.0,
|
| 619 |
+
"step": 86
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.0017512606056946164,
|
| 623 |
+
"grad_norm": 1.925287324411329e-05,
|
| 624 |
+
"learning_rate": 0.00019879860050235469,
|
| 625 |
+
"loss": 46.0,
|
| 626 |
+
"step": 87
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.0017713900379439798,
|
| 630 |
+
"grad_norm": 1.4624111827288289e-05,
|
| 631 |
+
"learning_rate": 0.00019876725737675254,
|
| 632 |
+
"loss": 46.0,
|
| 633 |
+
"step": 88
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0017915194701933431,
|
| 637 |
+
"grad_norm": 2.154901812900789e-05,
|
| 638 |
+
"learning_rate": 0.00019873551318098026,
|
| 639 |
+
"loss": 46.0,
|
| 640 |
+
"step": 89
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.0018116489024427066,
|
| 644 |
+
"grad_norm": 1.7668218788458034e-05,
|
| 645 |
+
"learning_rate": 0.00019870336804394338,
|
| 646 |
+
"loss": 46.0,
|
| 647 |
+
"step": 90
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.0018317783346920701,
|
| 651 |
+
"grad_norm": 2.4439837943646125e-05,
|
| 652 |
+
"learning_rate": 0.00019867082209617563,
|
| 653 |
+
"loss": 46.0,
|
| 654 |
+
"step": 91
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.0018519077669414334,
|
| 658 |
+
"grad_norm": 2.07311622943962e-05,
|
| 659 |
+
"learning_rate": 0.00019863787546983832,
|
| 660 |
+
"loss": 46.0,
|
| 661 |
+
"step": 92
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.001872037199190797,
|
| 665 |
+
"grad_norm": 2.8900734832859598e-05,
|
| 666 |
+
"learning_rate": 0.00019860452829871975,
|
| 667 |
+
"loss": 46.0,
|
| 668 |
+
"step": 93
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.0018921666314401602,
|
| 672 |
+
"grad_norm": 2.661554935912136e-05,
|
| 673 |
+
"learning_rate": 0.00019857078071823484,
|
| 674 |
+
"loss": 46.0,
|
| 675 |
+
"step": 94
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.0019122960636895237,
|
| 679 |
+
"grad_norm": 3.1965726520866156e-05,
|
| 680 |
+
"learning_rate": 0.00019853663286542442,
|
| 681 |
+
"loss": 46.0,
|
| 682 |
+
"step": 95
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.001932425495938887,
|
| 686 |
+
"grad_norm": 2.5141018340946175e-05,
|
| 687 |
+
"learning_rate": 0.00019850208487895475,
|
| 688 |
+
"loss": 46.0,
|
| 689 |
+
"step": 96
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.0019525549281882504,
|
| 693 |
+
"grad_norm": 2.246517760795541e-05,
|
| 694 |
+
"learning_rate": 0.0001984671368991169,
|
| 695 |
+
"loss": 46.0,
|
| 696 |
+
"step": 97
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.001972684360437614,
|
| 700 |
+
"grad_norm": 1.959139626706019e-05,
|
| 701 |
+
"learning_rate": 0.00019843178906782624,
|
| 702 |
+
"loss": 46.0,
|
| 703 |
+
"step": 98
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.0019928137926869772,
|
| 707 |
+
"grad_norm": 1.0403034139017109e-05,
|
| 708 |
+
"learning_rate": 0.0001983960415286219,
|
| 709 |
+
"loss": 46.0,
|
| 710 |
+
"step": 99
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.0020129432249363405,
|
| 714 |
+
"grad_norm": 2.894277531595435e-05,
|
| 715 |
+
"learning_rate": 0.000198359894426666,
|
| 716 |
+
"loss": 46.0,
|
| 717 |
+
"step": 100
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.002033072657185704,
|
| 721 |
+
"grad_norm": 2.3059441446093842e-05,
|
| 722 |
+
"learning_rate": 0.00019832334790874332,
|
| 723 |
+
"loss": 46.0,
|
| 724 |
+
"step": 101
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0020532020894350675,
|
| 728 |
+
"grad_norm": 3.624501550802961e-05,
|
| 729 |
+
"learning_rate": 0.00019828640212326046,
|
| 730 |
+
"loss": 46.0,
|
| 731 |
+
"step": 102
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.0020733315216844308,
|
| 735 |
+
"grad_norm": 2.7872463761013933e-05,
|
| 736 |
+
"learning_rate": 0.00019824905722024542,
|
| 737 |
+
"loss": 46.0,
|
| 738 |
+
"step": 103
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.0020934609539337945,
|
| 742 |
+
"grad_norm": 4.196954978397116e-05,
|
| 743 |
+
"learning_rate": 0.00019821131335134696,
|
| 744 |
+
"loss": 46.0,
|
| 745 |
+
"step": 104
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.0021135903861831578,
|
| 749 |
+
"grad_norm": 4.869971962762065e-05,
|
| 750 |
+
"learning_rate": 0.00019817317066983382,
|
| 751 |
+
"loss": 46.0,
|
| 752 |
+
"step": 105
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.002133719818432521,
|
| 756 |
+
"grad_norm": 2.689683788048569e-05,
|
| 757 |
+
"learning_rate": 0.00019813462933059435,
|
| 758 |
+
"loss": 46.0,
|
| 759 |
+
"step": 106
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.0021538492506818843,
|
| 763 |
+
"grad_norm": 2.0828028937103227e-05,
|
| 764 |
+
"learning_rate": 0.0001980956894901356,
|
| 765 |
+
"loss": 46.0,
|
| 766 |
+
"step": 107
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.002173978682931248,
|
| 770 |
+
"grad_norm": 1.8326103599974886e-05,
|
| 771 |
+
"learning_rate": 0.00019805635130658306,
|
| 772 |
+
"loss": 46.0,
|
| 773 |
+
"step": 108
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.0021941081151806113,
|
| 777 |
+
"grad_norm": 2.5773550078156404e-05,
|
| 778 |
+
"learning_rate": 0.00019801661493967955,
|
| 779 |
+
"loss": 46.0,
|
| 780 |
+
"step": 109
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.0022142375474299746,
|
| 784 |
+
"grad_norm": 1.7421609300072305e-05,
|
| 785 |
+
"learning_rate": 0.00019797648055078503,
|
| 786 |
+
"loss": 46.0,
|
| 787 |
+
"step": 110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.0022343669796793383,
|
| 791 |
+
"grad_norm": 2.473703534633387e-05,
|
| 792 |
+
"learning_rate": 0.0001979359483028756,
|
| 793 |
+
"loss": 46.0,
|
| 794 |
+
"step": 111
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.0022544964119287016,
|
| 798 |
+
"grad_norm": 4.040896601509303e-05,
|
| 799 |
+
"learning_rate": 0.00019789501836054297,
|
| 800 |
+
"loss": 46.0,
|
| 801 |
+
"step": 112
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.002274625844178065,
|
| 805 |
+
"grad_norm": 3.786892557400279e-05,
|
| 806 |
+
"learning_rate": 0.00019785369088999387,
|
| 807 |
+
"loss": 46.0,
|
| 808 |
+
"step": 113
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.0022947552764274286,
|
| 812 |
+
"grad_norm": 2.3734346541459672e-05,
|
| 813 |
+
"learning_rate": 0.0001978119660590493,
|
| 814 |
+
"loss": 46.0,
|
| 815 |
+
"step": 114
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.002314884708676792,
|
| 819 |
+
"grad_norm": 4.281644578441046e-05,
|
| 820 |
+
"learning_rate": 0.0001977698440371437,
|
| 821 |
+
"loss": 46.0,
|
| 822 |
+
"step": 115
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.002335014140926155,
|
| 826 |
+
"grad_norm": 6.363167631207034e-05,
|
| 827 |
+
"learning_rate": 0.0001977273249953246,
|
| 828 |
+
"loss": 46.0,
|
| 829 |
+
"step": 116
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.0023551435731755184,
|
| 833 |
+
"grad_norm": 2.2506428649649024e-05,
|
| 834 |
+
"learning_rate": 0.00019768440910625162,
|
| 835 |
+
"loss": 46.0,
|
| 836 |
+
"step": 117
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.002375273005424882,
|
| 840 |
+
"grad_norm": 2.2376898414222524e-05,
|
| 841 |
+
"learning_rate": 0.00019764109654419584,
|
| 842 |
+
"loss": 46.0,
|
| 843 |
+
"step": 118
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.0023954024376742454,
|
| 847 |
+
"grad_norm": 4.303127570892684e-05,
|
| 848 |
+
"learning_rate": 0.0001975973874850393,
|
| 849 |
+
"loss": 46.0,
|
| 850 |
+
"step": 119
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.0024155318699236087,
|
| 854 |
+
"grad_norm": 2.1347035726648755e-05,
|
| 855 |
+
"learning_rate": 0.00019755328210627394,
|
| 856 |
+
"loss": 46.0,
|
| 857 |
+
"step": 120
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.0024356613021729724,
|
| 861 |
+
"grad_norm": 3.3754025935195386e-05,
|
| 862 |
+
"learning_rate": 0.00019750878058700117,
|
| 863 |
+
"loss": 46.0,
|
| 864 |
+
"step": 121
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.0024557907344223357,
|
| 868 |
+
"grad_norm": 1.353104380541481e-05,
|
| 869 |
+
"learning_rate": 0.000197463883107931,
|
| 870 |
+
"loss": 46.0,
|
| 871 |
+
"step": 122
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.002475920166671699,
|
| 875 |
+
"grad_norm": 4.797324072569609e-05,
|
| 876 |
+
"learning_rate": 0.00019741858985138132,
|
| 877 |
+
"loss": 46.0,
|
| 878 |
+
"step": 123
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.0024960495989210622,
|
| 882 |
+
"grad_norm": 2.413903348497115e-05,
|
| 883 |
+
"learning_rate": 0.00019737290100127722,
|
| 884 |
+
"loss": 46.0,
|
| 885 |
+
"step": 124
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.002516179031170426,
|
| 889 |
+
"grad_norm": 1.5329667803598568e-05,
|
| 890 |
+
"learning_rate": 0.00019732681674315014,
|
| 891 |
+
"loss": 46.0,
|
| 892 |
+
"step": 125
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.0025363084634197892,
|
| 896 |
+
"grad_norm": 1.7863807443063706e-05,
|
| 897 |
+
"learning_rate": 0.00019728033726413723,
|
| 898 |
+
"loss": 46.0,
|
| 899 |
+
"step": 126
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.0025564378956691525,
|
| 903 |
+
"grad_norm": 3.0265011446317658e-05,
|
| 904 |
+
"learning_rate": 0.00019723346275298052,
|
| 905 |
+
"loss": 46.0,
|
| 906 |
+
"step": 127
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.0025765673279185162,
|
| 910 |
+
"grad_norm": 2.1787187506561168e-05,
|
| 911 |
+
"learning_rate": 0.00019718619340002618,
|
| 912 |
+
"loss": 46.0,
|
| 913 |
+
"step": 128
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.0025966967601678795,
|
| 917 |
+
"grad_norm": 2.615424818941392e-05,
|
| 918 |
+
"learning_rate": 0.0001971385293972237,
|
| 919 |
+
"loss": 46.0,
|
| 920 |
+
"step": 129
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.0026168261924172428,
|
| 924 |
+
"grad_norm": 2.6950148821924813e-05,
|
| 925 |
+
"learning_rate": 0.0001970904709381252,
|
| 926 |
+
"loss": 46.0,
|
| 927 |
+
"step": 130
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.0026369556246666065,
|
| 931 |
+
"grad_norm": 2.9200287826824933e-05,
|
| 932 |
+
"learning_rate": 0.00019704201821788456,
|
| 933 |
+
"loss": 46.0,
|
| 934 |
+
"step": 131
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.0026570850569159698,
|
| 938 |
+
"grad_norm": 4.081759470864199e-05,
|
| 939 |
+
"learning_rate": 0.00019699317143325666,
|
| 940 |
+
"loss": 46.0,
|
| 941 |
+
"step": 132
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.002677214489165333,
|
| 945 |
+
"grad_norm": 3.662336166598834e-05,
|
| 946 |
+
"learning_rate": 0.0001969439307825966,
|
| 947 |
+
"loss": 46.0,
|
| 948 |
+
"step": 133
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.0026973439214146963,
|
| 952 |
+
"grad_norm": 3.1114377634366974e-05,
|
| 953 |
+
"learning_rate": 0.0001968942964658589,
|
| 954 |
+
"loss": 46.0,
|
| 955 |
+
"step": 134
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.00271747335366406,
|
| 959 |
+
"grad_norm": 2.8435193598852493e-05,
|
| 960 |
+
"learning_rate": 0.00019684426868459655,
|
| 961 |
+
"loss": 46.0,
|
| 962 |
+
"step": 135
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.0027376027859134233,
|
| 966 |
+
"grad_norm": 2.6921919925371185e-05,
|
| 967 |
+
"learning_rate": 0.00019679384764196046,
|
| 968 |
+
"loss": 46.0,
|
| 969 |
+
"step": 136
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0027577322181627866,
|
| 973 |
+
"grad_norm": 2.3536973458249122e-05,
|
| 974 |
+
"learning_rate": 0.00019674303354269833,
|
| 975 |
+
"loss": 46.0,
|
| 976 |
+
"step": 137
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.0027778616504121503,
|
| 980 |
+
"grad_norm": 1.310581592406379e-05,
|
| 981 |
+
"learning_rate": 0.00019669182659315412,
|
| 982 |
+
"loss": 46.0,
|
| 983 |
+
"step": 138
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.0027979910826615136,
|
| 987 |
+
"grad_norm": 3.121258123428561e-05,
|
| 988 |
+
"learning_rate": 0.00019664022700126695,
|
| 989 |
+
"loss": 46.0,
|
| 990 |
+
"step": 139
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.002818120514910877,
|
| 994 |
+
"grad_norm": 3.1170515285339206e-05,
|
| 995 |
+
"learning_rate": 0.00019658823497657038,
|
| 996 |
+
"loss": 46.0,
|
| 997 |
+
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.00283824994716024,
|
| 1001 |
+
"grad_norm": 2.6195963073405437e-05,
|
| 1002 |
+
"learning_rate": 0.0001965358507301916,
|
| 1003 |
+
"loss": 46.0,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.002858379379409604,
|
| 1008 |
+
"grad_norm": 2.5877065127133392e-05,
|
| 1009 |
+
"learning_rate": 0.00019648307447485048,
|
| 1010 |
+
"loss": 46.0,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.002878508811658967,
|
| 1015 |
+
"grad_norm": 2.742213837336749e-05,
|
| 1016 |
+
"learning_rate": 0.00019642990642485875,
|
| 1017 |
+
"loss": 46.0,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.0028986382439083304,
|
| 1022 |
+
"grad_norm": 3.452887904131785e-05,
|
| 1023 |
+
"learning_rate": 0.0001963763467961191,
|
| 1024 |
+
"loss": 46.0,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.002918767676157694,
|
| 1029 |
+
"grad_norm": 3.4890996175818145e-05,
|
| 1030 |
+
"learning_rate": 0.00019632239580612436,
|
| 1031 |
+
"loss": 46.0,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.0029388971084070574,
|
| 1036 |
+
"grad_norm": 2.4831771952449344e-05,
|
| 1037 |
+
"learning_rate": 0.0001962680536739566,
|
| 1038 |
+
"loss": 46.0,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.0029590265406564207,
|
| 1043 |
+
"grad_norm": 2.926559500338044e-05,
|
| 1044 |
+
"learning_rate": 0.00019621332062028617,
|
| 1045 |
+
"loss": 46.0,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.0029791559729057844,
|
| 1050 |
+
"grad_norm": 3.801756975008175e-05,
|
| 1051 |
+
"learning_rate": 0.00019615819686737092,
|
| 1052 |
+
"loss": 46.0,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.0029992854051551477,
|
| 1057 |
+
"grad_norm": 4.2626343201845884e-05,
|
| 1058 |
+
"learning_rate": 0.00019610268263905515,
|
| 1059 |
+
"loss": 46.0,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.003019414837404511,
|
| 1064 |
+
"grad_norm": 5.3078922064742073e-05,
|
| 1065 |
+
"learning_rate": 0.00019604677816076888,
|
| 1066 |
+
"loss": 46.0,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.0030395442696538742,
|
| 1071 |
+
"grad_norm": 3.865420876536518e-05,
|
| 1072 |
+
"learning_rate": 0.00019599048365952682,
|
| 1073 |
+
"loss": 46.0,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.003059673701903238,
|
| 1078 |
+
"grad_norm": 2.114846029144246e-05,
|
| 1079 |
+
"learning_rate": 0.00019593379936392742,
|
| 1080 |
+
"loss": 46.0,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.0030798031341526012,
|
| 1085 |
+
"grad_norm": 2.966085958178155e-05,
|
| 1086 |
+
"learning_rate": 0.00019587672550415203,
|
| 1087 |
+
"loss": 46.0,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.0030999325664019645,
|
| 1092 |
+
"grad_norm": 2.12357390410034e-05,
|
| 1093 |
+
"learning_rate": 0.00019581926231196391,
|
| 1094 |
+
"loss": 46.0,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.0031200619986513282,
|
| 1099 |
+
"grad_norm": 2.5047153030755e-05,
|
| 1100 |
+
"learning_rate": 0.00019576141002070738,
|
| 1101 |
+
"loss": 46.0,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.0031401914309006915,
|
| 1106 |
+
"grad_norm": 2.9844281016266905e-05,
|
| 1107 |
+
"learning_rate": 0.00019570316886530665,
|
| 1108 |
+
"loss": 46.0,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.003160320863150055,
|
| 1113 |
+
"grad_norm": 2.7546439014258794e-05,
|
| 1114 |
+
"learning_rate": 0.00019564453908226515,
|
| 1115 |
+
"loss": 46.0,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.003180450295399418,
|
| 1120 |
+
"grad_norm": 5.0760227168211713e-05,
|
| 1121 |
+
"learning_rate": 0.00019558552090966435,
|
| 1122 |
+
"loss": 46.0,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.0032005797276487818,
|
| 1127 |
+
"grad_norm": 4.231133061693981e-05,
|
| 1128 |
+
"learning_rate": 0.00019552611458716296,
|
| 1129 |
+
"loss": 46.0,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.003220709159898145,
|
| 1134 |
+
"grad_norm": 5.593464447883889e-05,
|
| 1135 |
+
"learning_rate": 0.0001954663203559958,
|
| 1136 |
+
"loss": 46.0,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.0032408385921475083,
|
| 1141 |
+
"grad_norm": 3.7881276512052864e-05,
|
| 1142 |
+
"learning_rate": 0.00019540613845897288,
|
| 1143 |
+
"loss": 46.0,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.003260968024396872,
|
| 1148 |
+
"grad_norm": 3.1589570426149294e-05,
|
| 1149 |
+
"learning_rate": 0.00019534556914047851,
|
| 1150 |
+
"loss": 46.0,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.0032810974566462353,
|
| 1155 |
+
"grad_norm": 7.080697105266154e-05,
|
| 1156 |
+
"learning_rate": 0.00019528461264647014,
|
| 1157 |
+
"loss": 46.0,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.0033012268888955986,
|
| 1162 |
+
"grad_norm": 6.128560198703781e-05,
|
| 1163 |
+
"learning_rate": 0.00019522326922447755,
|
| 1164 |
+
"loss": 46.0,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.0033213563211449623,
|
| 1169 |
+
"grad_norm": 6.614374433411285e-05,
|
| 1170 |
+
"learning_rate": 0.00019516153912360165,
|
| 1171 |
+
"loss": 46.0,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.0033414857533943256,
|
| 1176 |
+
"grad_norm": 4.147323124925606e-05,
|
| 1177 |
+
"learning_rate": 0.00019509942259451357,
|
| 1178 |
+
"loss": 46.0,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.003361615185643689,
|
| 1183 |
+
"grad_norm": 3.474095865385607e-05,
|
| 1184 |
+
"learning_rate": 0.00019503691988945367,
|
| 1185 |
+
"loss": 46.0,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.003381744617893052,
|
| 1190 |
+
"grad_norm": 2.644029700604733e-05,
|
| 1191 |
+
"learning_rate": 0.00019497403126223048,
|
| 1192 |
+
"loss": 46.0,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.003401874050142416,
|
| 1197 |
+
"grad_norm": 2.13767089007888e-05,
|
| 1198 |
+
"learning_rate": 0.00019491075696821962,
|
| 1199 |
+
"loss": 46.0,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.003422003482391779,
|
| 1204 |
+
"grad_norm": 2.432280598441139e-05,
|
| 1205 |
+
"learning_rate": 0.00019484709726436282,
|
| 1206 |
+
"loss": 46.0,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.0034421329146411424,
|
| 1211 |
+
"grad_norm": 2.13755301956553e-05,
|
| 1212 |
+
"learning_rate": 0.00019478305240916698,
|
| 1213 |
+
"loss": 46.0,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.003462262346890506,
|
| 1218 |
+
"grad_norm": 2.9935839847894385e-05,
|
| 1219 |
+
"learning_rate": 0.0001947186226627028,
|
| 1220 |
+
"loss": 46.0,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.0034823917791398694,
|
| 1225 |
+
"grad_norm": 6.0657053836621344e-05,
|
| 1226 |
+
"learning_rate": 0.0001946538082866041,
|
| 1227 |
+
"loss": 46.0,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.0035025212113892327,
|
| 1232 |
+
"grad_norm": 3.512473631417379e-05,
|
| 1233 |
+
"learning_rate": 0.00019458860954406655,
|
| 1234 |
+
"loss": 46.0,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.003522650643638596,
|
| 1239 |
+
"grad_norm": 3.1507472158409655e-05,
|
| 1240 |
+
"learning_rate": 0.00019452302669984662,
|
| 1241 |
+
"loss": 46.0,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.0035427800758879597,
|
| 1246 |
+
"grad_norm": 9.249807771993801e-05,
|
| 1247 |
+
"learning_rate": 0.00019445706002026048,
|
| 1248 |
+
"loss": 46.0,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.003562909508137323,
|
| 1253 |
+
"grad_norm": 2.893183773267083e-05,
|
| 1254 |
+
"learning_rate": 0.000194390709773183,
|
| 1255 |
+
"loss": 46.0,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.0035830389403866863,
|
| 1260 |
+
"grad_norm": 6.318593659671023e-05,
|
| 1261 |
+
"learning_rate": 0.00019432397622804674,
|
| 1262 |
+
"loss": 46.0,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.00360316837263605,
|
| 1267 |
+
"grad_norm": 5.4799009376438335e-05,
|
| 1268 |
+
"learning_rate": 0.00019425685965584056,
|
| 1269 |
+
"loss": 46.0,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.0036232978048854132,
|
| 1274 |
+
"grad_norm": 5.187106944504194e-05,
|
| 1275 |
+
"learning_rate": 0.0001941893603291088,
|
| 1276 |
+
"loss": 46.0,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.0036434272371347765,
|
| 1281 |
+
"grad_norm": 3.32783383782953e-05,
|
| 1282 |
+
"learning_rate": 0.00019412147852195007,
|
| 1283 |
+
"loss": 46.0,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.0036635566693841402,
|
| 1288 |
+
"grad_norm": 2.430532913422212e-05,
|
| 1289 |
+
"learning_rate": 0.00019405321451001605,
|
| 1290 |
+
"loss": 46.0,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.0036836861016335035,
|
| 1295 |
+
"grad_norm": 5.952031642664224e-05,
|
| 1296 |
+
"learning_rate": 0.00019398456857051065,
|
| 1297 |
+
"loss": 46.0,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.003703815533882867,
|
| 1302 |
+
"grad_norm": 3.728271258296445e-05,
|
| 1303 |
+
"learning_rate": 0.00019391554098218853,
|
| 1304 |
+
"loss": 46.0,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.00372394496613223,
|
| 1309 |
+
"grad_norm": 4.849689139518887e-05,
|
| 1310 |
+
"learning_rate": 0.00019384613202535418,
|
| 1311 |
+
"loss": 46.0,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.003744074398381594,
|
| 1316 |
+
"grad_norm": 6.212339212652296e-05,
|
| 1317 |
+
"learning_rate": 0.00019377634198186077,
|
| 1318 |
+
"loss": 46.0,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.003764203830630957,
|
| 1323 |
+
"grad_norm": 0.00010154359188163653,
|
| 1324 |
+
"learning_rate": 0.0001937061711351089,
|
| 1325 |
+
"loss": 46.0,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.0037843332628803203,
|
| 1330 |
+
"grad_norm": 4.9923335609491915e-05,
|
| 1331 |
+
"learning_rate": 0.00019363561977004564,
|
| 1332 |
+
"loss": 46.0,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.003804462695129684,
|
| 1337 |
+
"grad_norm": 0.00011626673949649557,
|
| 1338 |
+
"learning_rate": 0.00019356468817316311,
|
| 1339 |
+
"loss": 46.0,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.0038245921273790473,
|
| 1344 |
+
"grad_norm": 9.168522228719667e-05,
|
| 1345 |
+
"learning_rate": 0.0001934933766324976,
|
| 1346 |
+
"loss": 46.0,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.0038447215596284106,
|
| 1351 |
+
"grad_norm": 3.176367317792028e-05,
|
| 1352 |
+
"learning_rate": 0.00019342168543762814,
|
| 1353 |
+
"loss": 46.0,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.003864850991877774,
|
| 1358 |
+
"grad_norm": 5.836347190779634e-05,
|
| 1359 |
+
"learning_rate": 0.0001933496148796755,
|
| 1360 |
+
"loss": 46.0,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.0038849804241271376,
|
| 1365 |
+
"grad_norm": 2.9238137358333915e-05,
|
| 1366 |
+
"learning_rate": 0.00019327716525130098,
|
| 1367 |
+
"loss": 46.0,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.003905109856376501,
|
| 1372 |
+
"grad_norm": 3.159691550536081e-05,
|
| 1373 |
+
"learning_rate": 0.00019320433684670514,
|
| 1374 |
+
"loss": 46.0,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.003925239288625864,
|
| 1379 |
+
"grad_norm": 3.837713666143827e-05,
|
| 1380 |
+
"learning_rate": 0.00019313112996162667,
|
| 1381 |
+
"loss": 46.0,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.003945368720875228,
|
| 1386 |
+
"grad_norm": 3.1190254958346486e-05,
|
| 1387 |
+
"learning_rate": 0.00019305754489334125,
|
| 1388 |
+
"loss": 46.0,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.003965498153124591,
|
| 1393 |
+
"grad_norm": 9.271525777876377e-05,
|
| 1394 |
+
"learning_rate": 0.00019298358194066016,
|
| 1395 |
+
"loss": 46.0,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.0039856275853739544,
|
| 1400 |
+
"grad_norm": 3.302880941191688e-05,
|
| 1401 |
+
"learning_rate": 0.00019290924140392921,
|
| 1402 |
+
"loss": 46.0,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.004005757017623318,
|
| 1407 |
+
"grad_norm": 3.4837332350434735e-05,
|
| 1408 |
+
"learning_rate": 0.0001928345235850276,
|
| 1409 |
+
"loss": 46.0,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.004025886449872681,
|
| 1414 |
+
"grad_norm": 2.956252865260467e-05,
|
| 1415 |
+
"learning_rate": 0.0001927594287873664,
|
| 1416 |
+
"loss": 46.0,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.004046015882122045,
|
| 1421 |
+
"grad_norm": 4.961419472238049e-05,
|
| 1422 |
+
"learning_rate": 0.00019268395731588764,
|
| 1423 |
+
"loss": 46.0,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.004066145314371408,
|
| 1428 |
+
"grad_norm": 7.009352702880278e-05,
|
| 1429 |
+
"learning_rate": 0.00019260810947706287,
|
| 1430 |
+
"loss": 46.0,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.004086274746620771,
|
| 1435 |
+
"grad_norm": 4.309253199608065e-05,
|
| 1436 |
+
"learning_rate": 0.000192531885578892,
|
| 1437 |
+
"loss": 46.0,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.004106404178870135,
|
| 1442 |
+
"grad_norm": 0.00012773476191796362,
|
| 1443 |
+
"learning_rate": 0.00019245528593090204,
|
| 1444 |
+
"loss": 46.0,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.004126533611119499,
|
| 1449 |
+
"grad_norm": 2.8066795493941754e-05,
|
| 1450 |
+
"learning_rate": 0.00019237831084414577,
|
| 1451 |
+
"loss": 46.0,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.0041466630433688615,
|
| 1456 |
+
"grad_norm": 5.905181751586497e-05,
|
| 1457 |
+
"learning_rate": 0.0001923009606312006,
|
| 1458 |
+
"loss": 46.0,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.004166792475618225,
|
| 1463 |
+
"grad_norm": 4.193755376036279e-05,
|
| 1464 |
+
"learning_rate": 0.0001922232356061672,
|
| 1465 |
+
"loss": 46.0,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.004186921907867589,
|
| 1470 |
+
"grad_norm": 4.6146677050273865e-05,
|
| 1471 |
+
"learning_rate": 0.00019214513608466826,
|
| 1472 |
+
"loss": 46.0,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.004207051340116952,
|
| 1477 |
+
"grad_norm": 2.824837065418251e-05,
|
| 1478 |
+
"learning_rate": 0.00019206666238384728,
|
| 1479 |
+
"loss": 46.0,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.0042271807723663155,
|
| 1484 |
+
"grad_norm": 3.296473005320877e-05,
|
| 1485 |
+
"learning_rate": 0.0001919878148223671,
|
| 1486 |
+
"loss": 46.0,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.004247310204615679,
|
| 1491 |
+
"grad_norm": 3.6901357816532254e-05,
|
| 1492 |
+
"learning_rate": 0.00019190859372040882,
|
| 1493 |
+
"loss": 46.0,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.004267439636865042,
|
| 1498 |
+
"grad_norm": 4.644416912924498e-05,
|
| 1499 |
+
"learning_rate": 0.00019182899939967034,
|
| 1500 |
+
"loss": 46.0,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.004287569069114406,
|
| 1505 |
+
"grad_norm": 5.2605690143536776e-05,
|
| 1506 |
+
"learning_rate": 0.00019174903218336511,
|
| 1507 |
+
"loss": 46.0,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.004307698501363769,
|
| 1512 |
+
"grad_norm": 2.7000423870049417e-05,
|
| 1513 |
+
"learning_rate": 0.00019166869239622085,
|
| 1514 |
+
"loss": 46.0,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.004327827933613132,
|
| 1519 |
+
"grad_norm": 4.584537600749172e-05,
|
| 1520 |
+
"learning_rate": 0.00019158798036447822,
|
| 1521 |
+
"loss": 46.0,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.004347957365862496,
|
| 1526 |
+
"grad_norm": 2.657137156347744e-05,
|
| 1527 |
+
"learning_rate": 0.0001915068964158894,
|
| 1528 |
+
"loss": 46.0,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.004368086798111859,
|
| 1533 |
+
"grad_norm": 4.8142963350983337e-05,
|
| 1534 |
+
"learning_rate": 0.00019142544087971693,
|
| 1535 |
+
"loss": 46.0,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.004388216230361223,
|
| 1540 |
+
"grad_norm": 3.945273419958539e-05,
|
| 1541 |
+
"learning_rate": 0.00019134361408673216,
|
| 1542 |
+
"loss": 46.0,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.004408345662610586,
|
| 1547 |
+
"grad_norm": 6.008298441884108e-05,
|
| 1548 |
+
"learning_rate": 0.00019126141636921414,
|
| 1549 |
+
"loss": 46.0,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.004428475094859949,
|
| 1554 |
+
"grad_norm": 5.5544471251778305e-05,
|
| 1555 |
+
"learning_rate": 0.0001911788480609481,
|
| 1556 |
+
"loss": 46.0,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.004448604527109313,
|
| 1561 |
+
"grad_norm": 5.3515970648732036e-05,
|
| 1562 |
+
"learning_rate": 0.00019109590949722413,
|
| 1563 |
+
"loss": 46.0,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.004468733959358677,
|
| 1568 |
+
"grad_norm": 4.5226741349324584e-05,
|
| 1569 |
+
"learning_rate": 0.00019101260101483592,
|
| 1570 |
+
"loss": 46.0,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.0044888633916080395,
|
| 1575 |
+
"grad_norm": 5.253091512713581e-05,
|
| 1576 |
+
"learning_rate": 0.0001909289229520792,
|
| 1577 |
+
"loss": 46.0,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.004508992823857403,
|
| 1582 |
+
"grad_norm": 7.821829058229923e-05,
|
| 1583 |
+
"learning_rate": 0.0001908448756487506,
|
| 1584 |
+
"loss": 46.0,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.004529122256106767,
|
| 1589 |
+
"grad_norm": 2.5970837668864988e-05,
|
| 1590 |
+
"learning_rate": 0.00019076045944614603,
|
| 1591 |
+
"loss": 46.0,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.00454925168835613,
|
| 1596 |
+
"grad_norm": 8.79172730492428e-05,
|
| 1597 |
+
"learning_rate": 0.0001906756746870595,
|
| 1598 |
+
"loss": 46.0,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.0045693811206054934,
|
| 1603 |
+
"grad_norm": 5.294303991831839e-05,
|
| 1604 |
+
"learning_rate": 0.00019059052171578155,
|
| 1605 |
+
"loss": 46.0,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.004589510552854857,
|
| 1610 |
+
"grad_norm": 3.699533408507705e-05,
|
| 1611 |
+
"learning_rate": 0.00019050500087809807,
|
| 1612 |
+
"loss": 46.0,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.00460963998510422,
|
| 1617 |
+
"grad_norm": 5.711496851290576e-05,
|
| 1618 |
+
"learning_rate": 0.00019041911252128864,
|
| 1619 |
+
"loss": 46.0,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.004629769417353584,
|
| 1624 |
+
"grad_norm": 5.045397483627312e-05,
|
| 1625 |
+
"learning_rate": 0.00019033285699412533,
|
| 1626 |
+
"loss": 46.0,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.0046498988496029466,
|
| 1631 |
+
"grad_norm": 7.707828626735136e-05,
|
| 1632 |
+
"learning_rate": 0.00019024623464687114,
|
| 1633 |
+
"loss": 46.0,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.00467002828185231,
|
| 1638 |
+
"grad_norm": 4.182373595540412e-05,
|
| 1639 |
+
"learning_rate": 0.00019015924583127872,
|
| 1640 |
+
"loss": 46.0,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.004690157714101674,
|
| 1645 |
+
"grad_norm": 4.5557317207567394e-05,
|
| 1646 |
+
"learning_rate": 0.00019007189090058878,
|
| 1647 |
+
"loss": 46.0,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.004710287146351037,
|
| 1652 |
+
"grad_norm": 4.820396861759946e-05,
|
| 1653 |
+
"learning_rate": 0.0001899841702095287,
|
| 1654 |
+
"loss": 46.0,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.0047304165786004005,
|
| 1659 |
+
"grad_norm": 5.630170926451683e-05,
|
| 1660 |
+
"learning_rate": 0.00018989608411431135,
|
| 1661 |
+
"loss": 46.0,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.004750546010849764,
|
| 1666 |
+
"grad_norm": 3.503153857309371e-05,
|
| 1667 |
+
"learning_rate": 0.0001898076329726331,
|
| 1668 |
+
"loss": 46.0,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.004770675443099127,
|
| 1673 |
+
"grad_norm": 8.119984704535455e-05,
|
| 1674 |
+
"learning_rate": 0.00018971881714367295,
|
| 1675 |
+
"loss": 46.0,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.004790804875348491,
|
| 1680 |
+
"grad_norm": 0.0001170546529465355,
|
| 1681 |
+
"learning_rate": 0.00018962963698809063,
|
| 1682 |
+
"loss": 46.0,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.0048109343075978545,
|
| 1687 |
+
"grad_norm": 3.3851210901048034e-05,
|
| 1688 |
+
"learning_rate": 0.00018954009286802545,
|
| 1689 |
+
"loss": 46.0,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.004831063739847217,
|
| 1694 |
+
"grad_norm": 4.396312215249054e-05,
|
| 1695 |
+
"learning_rate": 0.0001894501851470946,
|
| 1696 |
+
"loss": 46.0,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.004851193172096581,
|
| 1701 |
+
"grad_norm": 6.271849269978702e-05,
|
| 1702 |
+
"learning_rate": 0.00018935991419039176,
|
| 1703 |
+
"loss": 46.0,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.004871322604345945,
|
| 1708 |
+
"grad_norm": 4.544670082395896e-05,
|
| 1709 |
+
"learning_rate": 0.00018926928036448572,
|
| 1710 |
+
"loss": 46.0,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.004891452036595308,
|
| 1715 |
+
"grad_norm": 3.364813528605737e-05,
|
| 1716 |
+
"learning_rate": 0.0001891782840374187,
|
| 1717 |
+
"loss": 46.0,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.004911581468844671,
|
| 1722 |
+
"grad_norm": 3.067620491492562e-05,
|
| 1723 |
+
"learning_rate": 0.000189086925578705,
|
| 1724 |
+
"loss": 46.0,
|
| 1725 |
+
"step": 244
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.004931710901094035,
|
| 1729 |
+
"grad_norm": 3.76120260625612e-05,
|
| 1730 |
+
"learning_rate": 0.00018899520535932938,
|
| 1731 |
+
"loss": 46.0,
|
| 1732 |
+
"step": 245
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.004951840333343398,
|
| 1736 |
+
"grad_norm": 5.46066730748862e-05,
|
| 1737 |
+
"learning_rate": 0.00018890312375174578,
|
| 1738 |
+
"loss": 46.0,
|
| 1739 |
+
"step": 246
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.004971969765592762,
|
| 1743 |
+
"grad_norm": 4.003910726169124e-05,
|
| 1744 |
+
"learning_rate": 0.0001888106811298755,
|
| 1745 |
+
"loss": 46.0,
|
| 1746 |
+
"step": 247
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.0049920991978421245,
|
| 1750 |
+
"grad_norm": 6.587472307728603e-05,
|
| 1751 |
+
"learning_rate": 0.00018871787786910583,
|
| 1752 |
+
"loss": 46.0,
|
| 1753 |
+
"step": 248
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.005012228630091488,
|
| 1757 |
+
"grad_norm": 3.237745113437995e-05,
|
| 1758 |
+
"learning_rate": 0.0001886247143462886,
|
| 1759 |
+
"loss": 46.0,
|
| 1760 |
+
"step": 249
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.005032358062340852,
|
| 1764 |
+
"grad_norm": 4.901425199932419e-05,
|
| 1765 |
+
"learning_rate": 0.00018853119093973863,
|
| 1766 |
+
"loss": 46.0,
|
| 1767 |
+
"step": 250
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.005052487494590215,
|
| 1771 |
+
"grad_norm": 0.00016870767285581678,
|
| 1772 |
+
"learning_rate": 0.00018843730802923202,
|
| 1773 |
+
"loss": 46.0,
|
| 1774 |
+
"step": 251
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.0050726169268395785,
|
| 1778 |
+
"grad_norm": 5.0396236474625766e-05,
|
| 1779 |
+
"learning_rate": 0.00018834306599600472,
|
| 1780 |
+
"loss": 46.0,
|
| 1781 |
+
"step": 252
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.005092746359088942,
|
| 1785 |
+
"grad_norm": 4.4291067752055824e-05,
|
| 1786 |
+
"learning_rate": 0.00018824846522275113,
|
| 1787 |
+
"loss": 46.0,
|
| 1788 |
+
"step": 253
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.005112875791338305,
|
| 1792 |
+
"grad_norm": 4.502312367549166e-05,
|
| 1793 |
+
"learning_rate": 0.0001881535060936223,
|
| 1794 |
+
"loss": 46.0,
|
| 1795 |
+
"step": 254
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.005133005223587669,
|
| 1799 |
+
"grad_norm": 9.577722084941342e-05,
|
| 1800 |
+
"learning_rate": 0.00018805818899422447,
|
| 1801 |
+
"loss": 46.0,
|
| 1802 |
+
"step": 255
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.0051531346558370324,
|
| 1806 |
+
"grad_norm": 3.933937841793522e-05,
|
| 1807 |
+
"learning_rate": 0.0001879625143116176,
|
| 1808 |
+
"loss": 46.0,
|
| 1809 |
+
"step": 256
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.005173264088086395,
|
| 1813 |
+
"grad_norm": 5.2159059123368934e-05,
|
| 1814 |
+
"learning_rate": 0.00018786648243431363,
|
| 1815 |
+
"loss": 46.0,
|
| 1816 |
+
"step": 257
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.005193393520335759,
|
| 1820 |
+
"grad_norm": 6.076386125641875e-05,
|
| 1821 |
+
"learning_rate": 0.000187770093752275,
|
| 1822 |
+
"loss": 46.0,
|
| 1823 |
+
"step": 258
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.005213522952585123,
|
| 1827 |
+
"grad_norm": 7.954567990964279e-05,
|
| 1828 |
+
"learning_rate": 0.0001876733486569131,
|
| 1829 |
+
"loss": 46.0,
|
| 1830 |
+
"step": 259
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.0052336523848344856,
|
| 1834 |
+
"grad_norm": 9.186089300783351e-05,
|
| 1835 |
+
"learning_rate": 0.0001875762475410865,
|
| 1836 |
+
"loss": 46.0,
|
| 1837 |
+
"step": 260
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.005253781817083849,
|
| 1841 |
+
"grad_norm": 6.62936654407531e-05,
|
| 1842 |
+
"learning_rate": 0.00018747879079909963,
|
| 1843 |
+
"loss": 46.0,
|
| 1844 |
+
"step": 261
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.005273911249333213,
|
| 1848 |
+
"grad_norm": 8.344445814145729e-05,
|
| 1849 |
+
"learning_rate": 0.00018738097882670097,
|
| 1850 |
+
"loss": 46.0,
|
| 1851 |
+
"step": 262
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.005294040681582576,
|
| 1855 |
+
"grad_norm": 4.7971618187148124e-05,
|
| 1856 |
+
"learning_rate": 0.0001872828120210815,
|
| 1857 |
+
"loss": 46.0,
|
| 1858 |
+
"step": 263
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.0053141701138319395,
|
| 1862 |
+
"grad_norm": 6.027764902682975e-05,
|
| 1863 |
+
"learning_rate": 0.00018718429078087306,
|
| 1864 |
+
"loss": 46.0,
|
| 1865 |
+
"step": 264
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.005334299546081302,
|
| 1869 |
+
"grad_norm": 3.437638588366099e-05,
|
| 1870 |
+
"learning_rate": 0.00018708541550614688,
|
| 1871 |
+
"loss": 46.0,
|
| 1872 |
+
"step": 265
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.005354428978330666,
|
| 1876 |
+
"grad_norm": 0.00015669086133129895,
|
| 1877 |
+
"learning_rate": 0.00018698618659841168,
|
| 1878 |
+
"loss": 46.0,
|
| 1879 |
+
"step": 266
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.00537455841058003,
|
| 1883 |
+
"grad_norm": 5.7055072829825804e-05,
|
| 1884 |
+
"learning_rate": 0.00018688660446061235,
|
| 1885 |
+
"loss": 46.0,
|
| 1886 |
+
"step": 267
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.005394687842829393,
|
| 1890 |
+
"grad_norm": 9.243898966815323e-05,
|
| 1891 |
+
"learning_rate": 0.00018678666949712805,
|
| 1892 |
+
"loss": 46.0,
|
| 1893 |
+
"step": 268
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.005414817275078756,
|
| 1897 |
+
"grad_norm": 6.301647226791829e-05,
|
| 1898 |
+
"learning_rate": 0.00018668638211377075,
|
| 1899 |
+
"loss": 46.0,
|
| 1900 |
+
"step": 269
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.00543494670732812,
|
| 1904 |
+
"grad_norm": 5.150353172211908e-05,
|
| 1905 |
+
"learning_rate": 0.00018658574271778345,
|
| 1906 |
+
"loss": 46.0,
|
| 1907 |
+
"step": 270
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.005455076139577483,
|
| 1911 |
+
"grad_norm": 7.764642214169726e-05,
|
| 1912 |
+
"learning_rate": 0.0001864847517178387,
|
| 1913 |
+
"loss": 46.0,
|
| 1914 |
+
"step": 271
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.005475205571826847,
|
| 1918 |
+
"grad_norm": 6.848713383078575e-05,
|
| 1919 |
+
"learning_rate": 0.0001863834095240367,
|
| 1920 |
+
"loss": 46.0,
|
| 1921 |
+
"step": 272
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.00549533500407621,
|
| 1925 |
+
"grad_norm": 6.603610381716862e-05,
|
| 1926 |
+
"learning_rate": 0.00018628171654790383,
|
| 1927 |
+
"loss": 46.0,
|
| 1928 |
+
"step": 273
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.005515464436325573,
|
| 1932 |
+
"grad_norm": 4.008671749033965e-05,
|
| 1933 |
+
"learning_rate": 0.00018617967320239088,
|
| 1934 |
+
"loss": 46.0,
|
| 1935 |
+
"step": 274
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.005535593868574937,
|
| 1939 |
+
"grad_norm": 0.00014578885748051107,
|
| 1940 |
+
"learning_rate": 0.00018607727990187147,
|
| 1941 |
+
"loss": 46.0,
|
| 1942 |
+
"step": 275
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.005555723300824301,
|
| 1946 |
+
"grad_norm": 0.0001186303561553359,
|
| 1947 |
+
"learning_rate": 0.00018597453706214025,
|
| 1948 |
+
"loss": 46.0,
|
| 1949 |
+
"step": 276
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.0055758527330736635,
|
| 1953 |
+
"grad_norm": 6.097505320212804e-05,
|
| 1954 |
+
"learning_rate": 0.00018587144510041128,
|
| 1955 |
+
"loss": 46.0,
|
| 1956 |
+
"step": 277
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.005595982165323027,
|
| 1960 |
+
"grad_norm": 3.146566086797975e-05,
|
| 1961 |
+
"learning_rate": 0.0001857680044353163,
|
| 1962 |
+
"loss": 46.0,
|
| 1963 |
+
"step": 278
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.005616111597572391,
|
| 1967 |
+
"grad_norm": 3.7260102544678375e-05,
|
| 1968 |
+
"learning_rate": 0.0001856642154869031,
|
| 1969 |
+
"loss": 46.0,
|
| 1970 |
+
"step": 279
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.005636241029821754,
|
| 1974 |
+
"grad_norm": 9.276531636714935e-05,
|
| 1975 |
+
"learning_rate": 0.0001855600786766337,
|
| 1976 |
+
"loss": 46.0,
|
| 1977 |
+
"step": 280
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.0056563704620711175,
|
| 1981 |
+
"grad_norm": 4.5973942178534344e-05,
|
| 1982 |
+
"learning_rate": 0.00018545559442738273,
|
| 1983 |
+
"loss": 46.0,
|
| 1984 |
+
"step": 281
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.00567649989432048,
|
| 1988 |
+
"grad_norm": 6.683785613859072e-05,
|
| 1989 |
+
"learning_rate": 0.00018535076316343575,
|
| 1990 |
+
"loss": 46.0,
|
| 1991 |
+
"step": 282
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.005696629326569844,
|
| 1995 |
+
"grad_norm": 8.007440192159265e-05,
|
| 1996 |
+
"learning_rate": 0.00018524558531048737,
|
| 1997 |
+
"loss": 46.0,
|
| 1998 |
+
"step": 283
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.005716758758819208,
|
| 2002 |
+
"grad_norm": 4.978142897016369e-05,
|
| 2003 |
+
"learning_rate": 0.00018514006129563966,
|
| 2004 |
+
"loss": 46.0,
|
| 2005 |
+
"step": 284
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.005736888191068571,
|
| 2009 |
+
"grad_norm": 0.00010632863268256187,
|
| 2010 |
+
"learning_rate": 0.00018503419154740035,
|
| 2011 |
+
"loss": 46.0,
|
| 2012 |
+
"step": 285
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.005757017623317934,
|
| 2016 |
+
"grad_norm": 2.972048423544038e-05,
|
| 2017 |
+
"learning_rate": 0.00018492797649568115,
|
| 2018 |
+
"loss": 46.0,
|
| 2019 |
+
"step": 286
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.005777147055567298,
|
| 2023 |
+
"grad_norm": 4.370976603240706e-05,
|
| 2024 |
+
"learning_rate": 0.00018482141657179594,
|
| 2025 |
+
"loss": 46.0,
|
| 2026 |
+
"step": 287
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.005797276487816661,
|
| 2030 |
+
"grad_norm": 7.786518835928291e-05,
|
| 2031 |
+
"learning_rate": 0.00018471451220845902,
|
| 2032 |
+
"loss": 46.0,
|
| 2033 |
+
"step": 288
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.0058174059200660246,
|
| 2037 |
+
"grad_norm": 0.00010520989599172026,
|
| 2038 |
+
"learning_rate": 0.00018460726383978337,
|
| 2039 |
+
"loss": 46.0,
|
| 2040 |
+
"step": 289
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.005837535352315388,
|
| 2044 |
+
"grad_norm": 7.304285827558488e-05,
|
| 2045 |
+
"learning_rate": 0.000184499671901279,
|
| 2046 |
+
"loss": 46.0,
|
| 2047 |
+
"step": 290
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.005857664784564751,
|
| 2051 |
+
"grad_norm": 3.885753540089354e-05,
|
| 2052 |
+
"learning_rate": 0.00018439173682985094,
|
| 2053 |
+
"loss": 46.0,
|
| 2054 |
+
"step": 291
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.005877794216814115,
|
| 2058 |
+
"grad_norm": 9.42759434110485e-05,
|
| 2059 |
+
"learning_rate": 0.00018428345906379767,
|
| 2060 |
+
"loss": 46.0,
|
| 2061 |
+
"step": 292
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.0058979236490634785,
|
| 2065 |
+
"grad_norm": 7.605970313306898e-05,
|
| 2066 |
+
"learning_rate": 0.00018417483904280925,
|
| 2067 |
+
"loss": 46.0,
|
| 2068 |
+
"step": 293
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.005918053081312841,
|
| 2072 |
+
"grad_norm": 7.958237983984873e-05,
|
| 2073 |
+
"learning_rate": 0.00018406587720796555,
|
| 2074 |
+
"loss": 46.0,
|
| 2075 |
+
"step": 294
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.005938182513562205,
|
| 2079 |
+
"grad_norm": 0.00015608093235641718,
|
| 2080 |
+
"learning_rate": 0.00018395657400173453,
|
| 2081 |
+
"loss": 46.0,
|
| 2082 |
+
"step": 295
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.005958311945811569,
|
| 2086 |
+
"grad_norm": 6.675553595414385e-05,
|
| 2087 |
+
"learning_rate": 0.00018384692986797026,
|
| 2088 |
+
"loss": 46.0,
|
| 2089 |
+
"step": 296
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.005978441378060932,
|
| 2093 |
+
"grad_norm": 4.7439232730539516e-05,
|
| 2094 |
+
"learning_rate": 0.00018373694525191138,
|
| 2095 |
+
"loss": 46.0,
|
| 2096 |
+
"step": 297
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.005998570810310295,
|
| 2100 |
+
"grad_norm": 7.789856317685917e-05,
|
| 2101 |
+
"learning_rate": 0.00018362662060017896,
|
| 2102 |
+
"loss": 46.0,
|
| 2103 |
+
"step": 298
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.006018700242559658,
|
| 2107 |
+
"grad_norm": 5.332715591066517e-05,
|
| 2108 |
+
"learning_rate": 0.00018351595636077509,
|
| 2109 |
+
"loss": 46.0,
|
| 2110 |
+
"step": 299
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.006038829674809022,
|
| 2114 |
+
"grad_norm": 9.82348938123323e-05,
|
| 2115 |
+
"learning_rate": 0.00018340495298308063,
|
| 2116 |
+
"loss": 46.0,
|
| 2117 |
+
"step": 300
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.006058959107058386,
|
| 2121 |
+
"grad_norm": 0.00010532526357565075,
|
| 2122 |
+
"learning_rate": 0.0001832936109178538,
|
| 2123 |
+
"loss": 46.0,
|
| 2124 |
+
"step": 301
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.0060790885393077485,
|
| 2128 |
+
"grad_norm": 5.648566730087623e-05,
|
| 2129 |
+
"learning_rate": 0.00018318193061722795,
|
| 2130 |
+
"loss": 46.0,
|
| 2131 |
+
"step": 302
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.006099217971557112,
|
| 2135 |
+
"grad_norm": 0.00012587102537509054,
|
| 2136 |
+
"learning_rate": 0.00018306991253471013,
|
| 2137 |
+
"loss": 46.0,
|
| 2138 |
+
"step": 303
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.006119347403806476,
|
| 2142 |
+
"grad_norm": 0.00011527648894116282,
|
| 2143 |
+
"learning_rate": 0.00018295755712517887,
|
| 2144 |
+
"loss": 46.0,
|
| 2145 |
+
"step": 304
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.006139476836055839,
|
| 2149 |
+
"grad_norm": 6.82896061334759e-05,
|
| 2150 |
+
"learning_rate": 0.00018284486484488257,
|
| 2151 |
+
"loss": 46.0,
|
| 2152 |
+
"step": 305
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.0061596062683052025,
|
| 2156 |
+
"grad_norm": 6.256354390643537e-05,
|
| 2157 |
+
"learning_rate": 0.00018273183615143764,
|
| 2158 |
+
"loss": 46.0,
|
| 2159 |
+
"step": 306
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.006179735700554566,
|
| 2163 |
+
"grad_norm": 6.483913603005931e-05,
|
| 2164 |
+
"learning_rate": 0.00018261847150382644,
|
| 2165 |
+
"loss": 46.0,
|
| 2166 |
+
"step": 307
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.006199865132803929,
|
| 2170 |
+
"grad_norm": 9.90207918221131e-05,
|
| 2171 |
+
"learning_rate": 0.00018250477136239572,
|
| 2172 |
+
"loss": 46.0,
|
| 2173 |
+
"step": 308
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.006219994565053293,
|
| 2177 |
+
"grad_norm": 8.644862100481987e-05,
|
| 2178 |
+
"learning_rate": 0.00018239073618885447,
|
| 2179 |
+
"loss": 46.0,
|
| 2180 |
+
"step": 309
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.0062401239973026565,
|
| 2184 |
+
"grad_norm": 5.34062746737618e-05,
|
| 2185 |
+
"learning_rate": 0.00018227636644627224,
|
| 2186 |
+
"loss": 46.0,
|
| 2187 |
+
"step": 310
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.006260253429552019,
|
| 2191 |
+
"grad_norm": 5.9164103731745854e-05,
|
| 2192 |
+
"learning_rate": 0.00018216166259907713,
|
| 2193 |
+
"loss": 46.0,
|
| 2194 |
+
"step": 311
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.006280382861801383,
|
| 2198 |
+
"grad_norm": 0.00011165209434693679,
|
| 2199 |
+
"learning_rate": 0.000182046625113054,
|
| 2200 |
+
"loss": 46.0,
|
| 2201 |
+
"step": 312
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.006300512294050747,
|
| 2205 |
+
"grad_norm": 9.209036215906963e-05,
|
| 2206 |
+
"learning_rate": 0.00018193125445534252,
|
| 2207 |
+
"loss": 46.0,
|
| 2208 |
+
"step": 313
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.00632064172630011,
|
| 2212 |
+
"grad_norm": 9.060541924554855e-05,
|
| 2213 |
+
"learning_rate": 0.00018181555109443527,
|
| 2214 |
+
"loss": 46.0,
|
| 2215 |
+
"step": 314
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.006340771158549473,
|
| 2219 |
+
"grad_norm": 0.00010777592979138717,
|
| 2220 |
+
"learning_rate": 0.0001816995155001759,
|
| 2221 |
+
"loss": 46.0,
|
| 2222 |
+
"step": 315
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.006360900590798836,
|
| 2226 |
+
"grad_norm": 6.211627623997629e-05,
|
| 2227 |
+
"learning_rate": 0.00018158314814375716,
|
| 2228 |
+
"loss": 46.0,
|
| 2229 |
+
"step": 316
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.0063810300230482,
|
| 2233 |
+
"grad_norm": 0.00011468301818240434,
|
| 2234 |
+
"learning_rate": 0.000181466449497719,
|
| 2235 |
+
"loss": 46.0,
|
| 2236 |
+
"step": 317
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.0064011594552975636,
|
| 2240 |
+
"grad_norm": 0.00011545659799594432,
|
| 2241 |
+
"learning_rate": 0.00018134942003594665,
|
| 2242 |
+
"loss": 46.0,
|
| 2243 |
+
"step": 318
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.006421288887546926,
|
| 2247 |
+
"grad_norm": 7.793011172907427e-05,
|
| 2248 |
+
"learning_rate": 0.00018123206023366875,
|
| 2249 |
+
"loss": 46.0,
|
| 2250 |
+
"step": 319
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.00644141831979629,
|
| 2254 |
+
"grad_norm": 6.144792860141024e-05,
|
| 2255 |
+
"learning_rate": 0.00018111437056745532,
|
| 2256 |
+
"loss": 46.0,
|
| 2257 |
+
"step": 320
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.006461547752045654,
|
| 2261 |
+
"grad_norm": 7.264247687999159e-05,
|
| 2262 |
+
"learning_rate": 0.00018099635151521586,
|
| 2263 |
+
"loss": 46.0,
|
| 2264 |
+
"step": 321
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.006481677184295017,
|
| 2268 |
+
"grad_norm": 0.0001523289829492569,
|
| 2269 |
+
"learning_rate": 0.00018087800355619753,
|
| 2270 |
+
"loss": 46.0,
|
| 2271 |
+
"step": 322
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.00650180661654438,
|
| 2275 |
+
"grad_norm": 0.00013660687545780092,
|
| 2276 |
+
"learning_rate": 0.000180759327170983,
|
| 2277 |
+
"loss": 46.0,
|
| 2278 |
+
"step": 323
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.006521936048793744,
|
| 2282 |
+
"grad_norm": 0.0001787421788321808,
|
| 2283 |
+
"learning_rate": 0.00018064032284148868,
|
| 2284 |
+
"loss": 46.0,
|
| 2285 |
+
"step": 324
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.006542065481043107,
|
| 2289 |
+
"grad_norm": 0.00017624157771933824,
|
| 2290 |
+
"learning_rate": 0.0001805209910509626,
|
| 2291 |
+
"loss": 46.0,
|
| 2292 |
+
"step": 325
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.006562194913292471,
|
| 2296 |
+
"grad_norm": 7.056714821374044e-05,
|
| 2297 |
+
"learning_rate": 0.0001804013322839826,
|
| 2298 |
+
"loss": 46.0,
|
| 2299 |
+
"step": 326
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.006582324345541834,
|
| 2303 |
+
"grad_norm": 0.00016265243175439537,
|
| 2304 |
+
"learning_rate": 0.00018028134702645425,
|
| 2305 |
+
"loss": 46.0,
|
| 2306 |
+
"step": 327
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.006602453777791197,
|
| 2310 |
+
"grad_norm": 0.00015621079364791512,
|
| 2311 |
+
"learning_rate": 0.00018016103576560895,
|
| 2312 |
+
"loss": 46.0,
|
| 2313 |
+
"step": 328
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.006622583210040561,
|
| 2317 |
+
"grad_norm": 0.00015462673036381602,
|
| 2318 |
+
"learning_rate": 0.00018004039899000186,
|
| 2319 |
+
"loss": 46.0,
|
| 2320 |
+
"step": 329
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.006642712642289925,
|
| 2324 |
+
"grad_norm": 9.932387911248952e-05,
|
| 2325 |
+
"learning_rate": 0.0001799194371895101,
|
| 2326 |
+
"loss": 46.0,
|
| 2327 |
+
"step": 330
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.0066628420745392875,
|
| 2331 |
+
"grad_norm": 0.0001572413748363033,
|
| 2332 |
+
"learning_rate": 0.00017979815085533048,
|
| 2333 |
+
"loss": 46.0,
|
| 2334 |
+
"step": 331
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.006682971506788651,
|
| 2338 |
+
"grad_norm": 0.00013717268302571028,
|
| 2339 |
+
"learning_rate": 0.00017967654047997784,
|
| 2340 |
+
"loss": 46.0,
|
| 2341 |
+
"step": 332
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.006703100939038014,
|
| 2345 |
+
"grad_norm": 0.00017757757450453937,
|
| 2346 |
+
"learning_rate": 0.0001795546065572827,
|
| 2347 |
+
"loss": 46.0,
|
| 2348 |
+
"step": 333
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.006723230371287378,
|
| 2352 |
+
"grad_norm": 9.130597027251497e-05,
|
| 2353 |
+
"learning_rate": 0.00017943234958238952,
|
| 2354 |
+
"loss": 46.0,
|
| 2355 |
+
"step": 334
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.0067433598035367415,
|
| 2359 |
+
"grad_norm": 6.078862497815862e-05,
|
| 2360 |
+
"learning_rate": 0.00017930977005175465,
|
| 2361 |
+
"loss": 46.0,
|
| 2362 |
+
"step": 335
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.006763489235786104,
|
| 2366 |
+
"grad_norm": 0.00018025643657892942,
|
| 2367 |
+
"learning_rate": 0.0001791868684631441,
|
| 2368 |
+
"loss": 46.0,
|
| 2369 |
+
"step": 336
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.006783618668035468,
|
| 2373 |
+
"grad_norm": 0.0001726304617477581,
|
| 2374 |
+
"learning_rate": 0.00017906364531563185,
|
| 2375 |
+
"loss": 46.0,
|
| 2376 |
+
"step": 337
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.006803748100284832,
|
| 2380 |
+
"grad_norm": 8.799460192676634e-05,
|
| 2381 |
+
"learning_rate": 0.00017894010110959755,
|
| 2382 |
+
"loss": 46.0,
|
| 2383 |
+
"step": 338
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.006823877532534195,
|
| 2387 |
+
"grad_norm": 0.00012871818034909666,
|
| 2388 |
+
"learning_rate": 0.00017881623634672465,
|
| 2389 |
+
"loss": 46.0,
|
| 2390 |
+
"step": 339
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.006844006964783558,
|
| 2394 |
+
"grad_norm": 0.00010133467003470287,
|
| 2395 |
+
"learning_rate": 0.00017869205152999822,
|
| 2396 |
+
"loss": 46.0,
|
| 2397 |
+
"step": 340
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.006864136397032922,
|
| 2401 |
+
"grad_norm": 0.00010885829397011548,
|
| 2402 |
+
"learning_rate": 0.00017856754716370313,
|
| 2403 |
+
"loss": 46.0,
|
| 2404 |
+
"step": 341
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.006884265829282285,
|
| 2408 |
+
"grad_norm": 9.338463132735342e-05,
|
| 2409 |
+
"learning_rate": 0.0001784427237534217,
|
| 2410 |
+
"loss": 46.0,
|
| 2411 |
+
"step": 342
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.006904395261531649,
|
| 2415 |
+
"grad_norm": 9.378411778016016e-05,
|
| 2416 |
+
"learning_rate": 0.00017831758180603195,
|
| 2417 |
+
"loss": 46.0,
|
| 2418 |
+
"step": 343
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.006924524693781012,
|
| 2422 |
+
"grad_norm": 0.00012989221431780607,
|
| 2423 |
+
"learning_rate": 0.00017819212182970535,
|
| 2424 |
+
"loss": 46.0,
|
| 2425 |
+
"step": 344
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.006944654126030375,
|
| 2429 |
+
"grad_norm": 0.00017901930550578982,
|
| 2430 |
+
"learning_rate": 0.00017806634433390476,
|
| 2431 |
+
"loss": 46.0,
|
| 2432 |
+
"step": 345
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.006964783558279739,
|
| 2436 |
+
"grad_norm": 9.102857438847423e-05,
|
| 2437 |
+
"learning_rate": 0.00017794024982938252,
|
| 2438 |
+
"loss": 46.0,
|
| 2439 |
+
"step": 346
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.0069849129905291026,
|
| 2443 |
+
"grad_norm": 7.371963874902576e-05,
|
| 2444 |
+
"learning_rate": 0.00017781383882817811,
|
| 2445 |
+
"loss": 46.0,
|
| 2446 |
+
"step": 347
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.007005042422778465,
|
| 2450 |
+
"grad_norm": 7.886077219154686e-05,
|
| 2451 |
+
"learning_rate": 0.00017768711184361645,
|
| 2452 |
+
"loss": 46.0,
|
| 2453 |
+
"step": 348
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.007025171855027829,
|
| 2457 |
+
"grad_norm": 0.0001304990437347442,
|
| 2458 |
+
"learning_rate": 0.00017756006939030535,
|
| 2459 |
+
"loss": 46.0,
|
| 2460 |
+
"step": 349
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.007045301287277192,
|
| 2464 |
+
"grad_norm": 0.00014258353621698916,
|
| 2465 |
+
"learning_rate": 0.00017743271198413386,
|
| 2466 |
+
"loss": 46.0,
|
| 2467 |
+
"step": 350
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.007065430719526556,
|
| 2471 |
+
"grad_norm": 0.00015988641825970262,
|
| 2472 |
+
"learning_rate": 0.00017730504014226982,
|
| 2473 |
+
"loss": 46.0,
|
| 2474 |
+
"step": 351
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.007085560151775919,
|
| 2478 |
+
"grad_norm": 8.829456783132628e-05,
|
| 2479 |
+
"learning_rate": 0.00017717705438315804,
|
| 2480 |
+
"loss": 46.0,
|
| 2481 |
+
"step": 352
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.007105689584025282,
|
| 2485 |
+
"grad_norm": 0.00018408888718113303,
|
| 2486 |
+
"learning_rate": 0.00017704875522651806,
|
| 2487 |
+
"loss": 46.0,
|
| 2488 |
+
"step": 353
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.007125819016274646,
|
| 2492 |
+
"grad_norm": 0.0001536206982564181,
|
| 2493 |
+
"learning_rate": 0.0001769201431933419,
|
| 2494 |
+
"loss": 46.0,
|
| 2495 |
+
"step": 354
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.00714594844852401,
|
| 2499 |
+
"grad_norm": 0.00015358305245172232,
|
| 2500 |
+
"learning_rate": 0.00017679121880589236,
|
| 2501 |
+
"loss": 46.0,
|
| 2502 |
+
"step": 355
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.0071660778807733725,
|
| 2506 |
+
"grad_norm": 0.00028956442838534713,
|
| 2507 |
+
"learning_rate": 0.00017666198258770038,
|
| 2508 |
+
"loss": 46.0,
|
| 2509 |
+
"step": 356
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.007186207313022736,
|
| 2513 |
+
"grad_norm": 0.0001954118488356471,
|
| 2514 |
+
"learning_rate": 0.00017653243506356332,
|
| 2515 |
+
"loss": 46.0,
|
| 2516 |
+
"step": 357
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.0072063367452721,
|
| 2520 |
+
"grad_norm": 0.0001885920064523816,
|
| 2521 |
+
"learning_rate": 0.00017640257675954264,
|
| 2522 |
+
"loss": 46.0,
|
| 2523 |
+
"step": 358
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.007226466177521463,
|
| 2527 |
+
"grad_norm": 7.952122541610152e-05,
|
| 2528 |
+
"learning_rate": 0.00017627240820296177,
|
| 2529 |
+
"loss": 46.0,
|
| 2530 |
+
"step": 359
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.0072465956097708265,
|
| 2534 |
+
"grad_norm": 6.53123643132858e-05,
|
| 2535 |
+
"learning_rate": 0.00017614192992240413,
|
| 2536 |
+
"loss": 46.0,
|
| 2537 |
+
"step": 360
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.00726672504202019,
|
| 2541 |
+
"grad_norm": 0.0001405112270731479,
|
| 2542 |
+
"learning_rate": 0.00017601114244771067,
|
| 2543 |
+
"loss": 46.0,
|
| 2544 |
+
"step": 361
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.007286854474269553,
|
| 2548 |
+
"grad_norm": 0.00019535009050741792,
|
| 2549 |
+
"learning_rate": 0.000175880046309978,
|
| 2550 |
+
"loss": 46.0,
|
| 2551 |
+
"step": 362
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.007306983906518917,
|
| 2555 |
+
"grad_norm": 0.0003268709115218371,
|
| 2556 |
+
"learning_rate": 0.00017574864204155614,
|
| 2557 |
+
"loss": 46.0,
|
| 2558 |
+
"step": 363
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.0073271133387682805,
|
| 2562 |
+
"grad_norm": 0.00010955316975014284,
|
| 2563 |
+
"learning_rate": 0.00017561693017604637,
|
| 2564 |
+
"loss": 46.0,
|
| 2565 |
+
"step": 364
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.007347242771017643,
|
| 2569 |
+
"grad_norm": 0.00027552066603675485,
|
| 2570 |
+
"learning_rate": 0.000175484911248299,
|
| 2571 |
+
"loss": 46.0,
|
| 2572 |
+
"step": 365
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.007367372203267007,
|
| 2576 |
+
"grad_norm": 0.00019117463671136647,
|
| 2577 |
+
"learning_rate": 0.0001753525857944112,
|
| 2578 |
+
"loss": 46.0,
|
| 2579 |
+
"step": 366
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.00738750163551637,
|
| 2583 |
+
"grad_norm": 9.778481035027653e-05,
|
| 2584 |
+
"learning_rate": 0.00017521995435172504,
|
| 2585 |
+
"loss": 46.0,
|
| 2586 |
+
"step": 367
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.007407631067765734,
|
| 2590 |
+
"grad_norm": 3.404769449844025e-05,
|
| 2591 |
+
"learning_rate": 0.000175087017458825,
|
| 2592 |
+
"loss": 46.0,
|
| 2593 |
+
"step": 368
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.007427760500015097,
|
| 2597 |
+
"grad_norm": 0.00010666289017535746,
|
| 2598 |
+
"learning_rate": 0.00017495377565553594,
|
| 2599 |
+
"loss": 46.0,
|
| 2600 |
+
"step": 369
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.00744788993226446,
|
| 2604 |
+
"grad_norm": 0.0001144933485193178,
|
| 2605 |
+
"learning_rate": 0.0001748202294829209,
|
| 2606 |
+
"loss": 46.0,
|
| 2607 |
+
"step": 370
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.007468019364513824,
|
| 2611 |
+
"grad_norm": 0.00013504338858183473,
|
| 2612 |
+
"learning_rate": 0.00017468637948327894,
|
| 2613 |
+
"loss": 46.0,
|
| 2614 |
+
"step": 371
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.007488148796763188,
|
| 2618 |
+
"grad_norm": 0.00010847981320694089,
|
| 2619 |
+
"learning_rate": 0.00017455222620014276,
|
| 2620 |
+
"loss": 46.0,
|
| 2621 |
+
"step": 372
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.00750827822901255,
|
| 2625 |
+
"grad_norm": 0.00020060865790583193,
|
| 2626 |
+
"learning_rate": 0.00017441777017827677,
|
| 2627 |
+
"loss": 46.0,
|
| 2628 |
+
"step": 373
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.007528407661261914,
|
| 2632 |
+
"grad_norm": 0.00019206189608667046,
|
| 2633 |
+
"learning_rate": 0.00017428301196367464,
|
| 2634 |
+
"loss": 46.0,
|
| 2635 |
+
"step": 374
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.007548537093511278,
|
| 2639 |
+
"grad_norm": 0.00011548047768883407,
|
| 2640 |
+
"learning_rate": 0.0001741479521035572,
|
| 2641 |
+
"loss": 46.0,
|
| 2642 |
+
"step": 375
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.007568666525760641,
|
| 2646 |
+
"grad_norm": 0.0001220878621097654,
|
| 2647 |
+
"learning_rate": 0.00017401259114637014,
|
| 2648 |
+
"loss": 46.0,
|
| 2649 |
+
"step": 376
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.007588795958010004,
|
| 2653 |
+
"grad_norm": 0.00027704107924364507,
|
| 2654 |
+
"learning_rate": 0.00017387692964178198,
|
| 2655 |
+
"loss": 46.0,
|
| 2656 |
+
"step": 377
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.007608925390259368,
|
| 2660 |
+
"grad_norm": 8.742011414142326e-05,
|
| 2661 |
+
"learning_rate": 0.00017374096814068145,
|
| 2662 |
+
"loss": 46.0,
|
| 2663 |
+
"step": 378
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.007629054822508731,
|
| 2667 |
+
"grad_norm": 0.0001393141719745472,
|
| 2668 |
+
"learning_rate": 0.00017360470719517577,
|
| 2669 |
+
"loss": 46.0,
|
| 2670 |
+
"step": 379
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.007649184254758095,
|
| 2674 |
+
"grad_norm": 0.00011780137720052153,
|
| 2675 |
+
"learning_rate": 0.00017346814735858792,
|
| 2676 |
+
"loss": 46.0,
|
| 2677 |
+
"step": 380
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.007669313687007458,
|
| 2681 |
+
"grad_norm": 0.00013582094106823206,
|
| 2682 |
+
"learning_rate": 0.0001733312891854547,
|
| 2683 |
+
"loss": 46.0,
|
| 2684 |
+
"step": 381
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.007689443119256821,
|
| 2688 |
+
"grad_norm": 7.030325650703162e-05,
|
| 2689 |
+
"learning_rate": 0.00017319413323152436,
|
| 2690 |
+
"loss": 46.0,
|
| 2691 |
+
"step": 382
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.007709572551506185,
|
| 2695 |
+
"grad_norm": 0.00016730479546822608,
|
| 2696 |
+
"learning_rate": 0.00017305668005375435,
|
| 2697 |
+
"loss": 46.0,
|
| 2698 |
+
"step": 383
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.007729701983755548,
|
| 2702 |
+
"grad_norm": 0.00027646831586025655,
|
| 2703 |
+
"learning_rate": 0.00017291893021030913,
|
| 2704 |
+
"loss": 46.0,
|
| 2705 |
+
"step": 384
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.0077498314160049115,
|
| 2709 |
+
"grad_norm": 0.0001369678502669558,
|
| 2710 |
+
"learning_rate": 0.0001727808842605578,
|
| 2711 |
+
"loss": 46.0,
|
| 2712 |
+
"step": 385
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.007769960848254275,
|
| 2716 |
+
"grad_norm": 0.00011632432142505422,
|
| 2717 |
+
"learning_rate": 0.00017264254276507188,
|
| 2718 |
+
"loss": 46.0,
|
| 2719 |
+
"step": 386
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.007790090280503638,
|
| 2723 |
+
"grad_norm": 0.00013326382031664252,
|
| 2724 |
+
"learning_rate": 0.00017250390628562303,
|
| 2725 |
+
"loss": 46.0,
|
| 2726 |
+
"step": 387
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.007810219712753002,
|
| 2730 |
+
"grad_norm": 0.0002664460625965148,
|
| 2731 |
+
"learning_rate": 0.00017236497538518082,
|
| 2732 |
+
"loss": 46.0,
|
| 2733 |
+
"step": 388
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.007830349145002365,
|
| 2737 |
+
"grad_norm": 0.00013524248788598925,
|
| 2738 |
+
"learning_rate": 0.00017222575062791033,
|
| 2739 |
+
"loss": 46.0,
|
| 2740 |
+
"step": 389
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.007850478577251728,
|
| 2744 |
+
"grad_norm": 0.00023887053248472512,
|
| 2745 |
+
"learning_rate": 0.00017208623257916993,
|
| 2746 |
+
"loss": 46.0,
|
| 2747 |
+
"step": 390
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.007870608009501092,
|
| 2751 |
+
"grad_norm": 0.0002952871145680547,
|
| 2752 |
+
"learning_rate": 0.000171946421805509,
|
| 2753 |
+
"loss": 46.0,
|
| 2754 |
+
"step": 391
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.007890737441750456,
|
| 2758 |
+
"grad_norm": 0.00014817963528912514,
|
| 2759 |
+
"learning_rate": 0.00017180631887466562,
|
| 2760 |
+
"loss": 46.0,
|
| 2761 |
+
"step": 392
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.00791086687399982,
|
| 2765 |
+
"grad_norm": 5.316930764820427e-05,
|
| 2766 |
+
"learning_rate": 0.0001716659243555642,
|
| 2767 |
+
"loss": 46.0,
|
| 2768 |
+
"step": 393
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.00791086687399982,
|
| 2772 |
+
"eval_loss": 11.5,
|
| 2773 |
+
"eval_runtime": 130.3433,
|
| 2774 |
+
"eval_samples_per_second": 160.484,
|
| 2775 |
+
"eval_steps_per_second": 80.242,
|
| 2776 |
+
"step": 393
|
| 2777 |
+
}
|
| 2778 |
+
],
|
| 2779 |
+
"logging_steps": 1,
|
| 2780 |
+
"max_steps": 1569,
|
| 2781 |
+
"num_input_tokens_seen": 0,
|
| 2782 |
+
"num_train_epochs": 1,
|
| 2783 |
+
"save_steps": 393,
|
| 2784 |
+
"stateful_callbacks": {
|
| 2785 |
+
"TrainerControl": {
|
| 2786 |
+
"args": {
|
| 2787 |
+
"should_epoch_stop": false,
|
| 2788 |
+
"should_evaluate": false,
|
| 2789 |
+
"should_log": false,
|
| 2790 |
+
"should_save": true,
|
| 2791 |
+
"should_training_stop": false
|
| 2792 |
+
},
|
| 2793 |
+
"attributes": {}
|
| 2794 |
+
}
|
| 2795 |
+
},
|
| 2796 |
+
"total_flos": 9074871140352.0,
|
| 2797 |
+
"train_batch_size": 2,
|
| 2798 |
+
"trial_name": null,
|
| 2799 |
+
"trial_params": null
|
| 2800 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17aa2810ee34df32858b81505baecfd1b4a435a05c73ea412e1baa1aca3972b6
|
| 3 |
+
size 6776
|
last-checkpoint/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|