Hugging Face's logo

mrferr3t
/
ee29425a-d835-495b-8fb5-e11141a51ea1

PEFT
Safetensors
llama
axolotl
Generated from Trainer
custom_code
Model card Files Community
mrferr3t commited on
Commit
df1fe57
·
verified ·
1 Parent(s): e01f90a

End of training

Browse files
Files changed (3) hide show
  1. README.md +200 -0
  2. adapter_model.bin +3 -0
  3. adapter_model.safetensors +1 -1
README.md ADDED
@@ -0,0 +1,200 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: NousResearch/CodeLlama-7b-hf
4
+ tags:
5
+ - axolotl
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: ee29425a-d835-495b-8fb5-e11141a51ea1
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.4.1`
19
+ ```yaml
20
+ adapter: lora
21
+ auto_find_batch_size: true
22
+ base_model: NousResearch/CodeLlama-7b-hf
23
+ bf16: auto
24
+ chat_template: llama3
25
+ dataloader_num_workers: 12
26
+ dataset_prepared_path: null
27
+ datasets:
28
+ - data_files:
29
+ - 77f4d29f468bc2c0_train_data.json
30
+ ds_type: json
31
+ format: custom
32
+ path: /workspace/input_data/77f4d29f468bc2c0_train_data.json
33
+ type:
34
+ field_input: input
35
+ field_instruction: instruction
36
+ field_output: response
37
+ format: '{instruction} {input}'
38
+ no_input_format: '{instruction}'
39
+ system_format: '{system}'
40
+ system_prompt: ''
41
+ debug: null
42
+ deepspeed: null
43
+ early_stopping_patience: 3
44
+ early_stopping_threshold: 0.001
45
+ eval_max_new_tokens: 128
46
+ eval_steps: 40
47
+ flash_attention: false
48
+ fp16: null
49
+ fsdp: null
50
+ fsdp_config: null
51
+ gradient_accumulation_steps: 2
52
+ gradient_checkpointing: false
53
+ group_by_length: false
54
+ hub_model_id: mrferr3t/ee29425a-d835-495b-8fb5-e11141a51ea1
55
+ hub_repo: null
56
+ hub_strategy: checkpoint
57
+ hub_token: null
58
+ learning_rate: 0.0003
59
+ load_in_4bit: false
60
+ load_in_8bit: false
61
+ local_rank: null
62
+ logging_steps: 100
63
+ lora_alpha: 16
64
+ lora_dropout: 0.05
65
+ lora_fan_in_fan_out: null
66
+ lora_model_dir: null
67
+ lora_r: 8
68
+ lora_target_linear: true
69
+ lr_scheduler: cosine
70
+ micro_batch_size: 32
71
+ mlflow_experiment_name: /tmp/77f4d29f468bc2c0_train_data.json
72
+ model_type: AutoModelForCausalLM
73
+ num_epochs: 50
74
+ optimizer: adamw_bnb_8bit
75
+ output_dir: miner_id_24
76
+ pad_to_sequence_len: true
77
+ s2_attention: null
78
+ sample_packing: false
79
+ save_steps: 40
80
+ saves_per_epoch: 0
81
+ sequence_len: 512
82
+ special_tokens:
83
+ pad_token: </s>
84
+ strict: false
85
+ tf32: false
86
+ tokenizer_type: AutoTokenizer
87
+ train_on_inputs: false
88
+ trust_remote_code: true
89
+ val_set_size: 0.02
90
+ wandb_entity: null
91
+ wandb_mode: online
92
+ wandb_name: 48c3c8d8-6645-4a88-adc5-70cd656a2562
93
+ wandb_project: Gradients-On-Demand
94
+ wandb_run: your_name
95
+ wandb_runid: 48c3c8d8-6645-4a88-adc5-70cd656a2562
96
+ warmup_ratio: 0.05
97
+ weight_decay: 0.0
98
+ xformers_attention: null
99
+
100
+ ```
101
+
102
+ </details><br>
103
+
104
+ # ee29425a-d835-495b-8fb5-e11141a51ea1
105
+
106
+ This model is a fine-tuned version of [NousResearch/CodeLlama-7b-hf](https://huggingface.co/NousResearch/CodeLlama-7b-hf) on the None dataset.
107
+ It achieves the following results on the evaluation set:
108
+ - Loss: 0.6540
109
+
110
+ ## Model description
111
+
112
+ More information needed
113
+
114
+ ## Intended uses & limitations
115
+
116
+ More information needed
117
+
118
+ ## Training and evaluation data
119
+
120
+ More information needed
121
+
122
+ ## Training procedure
123
+
124
+ ### Training hyperparameters
125
+
126
+ The following hyperparameters were used during training:
127
+ - learning_rate: 0.0003
128
+ - train_batch_size: 32
129
+ - eval_batch_size: 32
130
+ - seed: 42
131
+ - gradient_accumulation_steps: 2
132
+ - total_train_batch_size: 64
133
+ - optimizer: Use adamw_bnb_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
134
+ - lr_scheduler_type: cosine
135
+ - lr_scheduler_warmup_steps: 218
136
+ - num_epochs: 50
137
+
138
+ ### Training results
139
+
140
+ | Training Loss | Epoch | Step | Validation Loss |
141
+ |:-------------:|:------:|:----:|:---------------:|
142
+ | No log | 0.0014 | 1 | 2.6242 |
143
+ | No log | 0.0571 | 40 | 2.2860 |
144
+ | No log | 0.1142 | 80 | 1.9066 |
145
+ | 4.9536 | 0.1713 | 120 | 1.7338 |
146
+ | 4.9536 | 0.2284 | 160 | 1.6253 |
147
+ | 3.6406 | 0.2855 | 200 | 1.5644 |
148
+ | 3.6406 | 0.3426 | 240 | 1.4934 |
149
+ | 3.6406 | 0.3997 | 280 | 1.4442 |
150
+ | 3.1356 | 0.4568 | 320 | 1.3714 |
151
+ | 3.1356 | 0.5139 | 360 | 1.3406 |
152
+ | 2.8521 | 0.5710 | 400 | 1.3140 |
153
+ | 2.8521 | 0.6281 | 440 | 1.2290 |
154
+ | 2.8521 | 0.6852 | 480 | 1.1974 |
155
+ | 2.7573 | 0.7423 | 520 | 1.1788 |
156
+ | 2.7573 | 0.7994 | 560 | 1.1294 |
157
+ | 2.3985 | 0.8565 | 600 | 1.1246 |
158
+ | 2.3985 | 0.9136 | 640 | 1.0637 |
159
+ | 2.3985 | 0.9707 | 680 | 1.0167 |
160
+ | 2.2745 | 1.0278 | 720 | 1.0103 |
161
+ | 2.2745 | 1.0849 | 760 | 0.9873 |
162
+ | 1.8141 | 1.1420 | 800 | 0.9797 |
163
+ | 1.8141 | 1.1991 | 840 | 0.9597 |
164
+ | 1.8141 | 1.2562 | 880 | 0.9458 |
165
+ | 1.5585 | 1.3133 | 920 | 0.9204 |
166
+ | 1.5585 | 1.3704 | 960 | 0.9172 |
167
+ | 1.5745 | 1.4276 | 1000 | 0.9020 |
168
+ | 1.5745 | 1.4847 | 1040 | 0.8774 |
169
+ | 1.5745 | 1.5418 | 1080 | 0.8412 |
170
+ | 1.3623 | 1.5989 | 1120 | 0.8341 |
171
+ | 1.3623 | 1.6560 | 1160 | 0.8323 |
172
+ | 1.3262 | 1.7131 | 1200 | 0.7996 |
173
+ | 1.3262 | 1.7702 | 1240 | 0.8067 |
174
+ | 1.3262 | 1.8273 | 1280 | 0.7752 |
175
+ | 1.2514 | 1.8844 | 1320 | 0.7176 |
176
+ | 1.2514 | 1.9415 | 1360 | 0.7170 |
177
+ | 1.2277 | 1.9986 | 1400 | 0.7039 |
178
+ | 1.2277 | 2.0557 | 1440 | 0.7033 |
179
+ | 1.2277 | 2.1128 | 1480 | 0.7025 |
180
+ | 0.68 | 2.1699 | 1520 | 0.7079 |
181
+ | 0.68 | 2.2270 | 1560 | 0.7214 |
182
+ | 0.809 | 2.2841 | 1600 | 0.6897 |
183
+ | 0.809 | 2.3412 | 1640 | 0.7047 |
184
+ | 0.809 | 2.3983 | 1680 | 0.7042 |
185
+ | 0.7931 | 2.4554 | 1720 | 0.6792 |
186
+ | 0.7931 | 2.5125 | 1760 | 0.6777 |
187
+ | 0.6957 | 2.5696 | 1800 | 0.6809 |
188
+ | 0.6957 | 2.6267 | 1840 | 0.6466 |
189
+ | 0.6957 | 2.6838 | 1880 | 0.6705 |
190
+ | 0.7611 | 2.7409 | 1920 | 0.6549 |
191
+ | 0.7611 | 2.7980 | 1960 | 0.6540 |
192
+
193
+
194
+ ### Framework versions
195
+
196
+ - PEFT 0.13.2
197
+ - Transformers 4.46.0
198
+ - Pytorch 2.3.1+cu121
199
+ - Datasets 3.0.1
200
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a3b60c971ea35e6e6f563f6ee4c4d6aefbc8f23fb37cb97f7d3ebb9a58fc8e6
3
+ size 80115210
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6f6a966a7b6dc94fce248a3fd714b53cfa8886f737dc508e97110b7be08da6a8
3
  size 80013120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5b418f9ffd77d79715b6b6d85f2fb4a1990a6b1e03c652f023e7cdc6736f6c2
3
  size 80013120