End of training

README.md

@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/358b7fae411d76fd_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 8b06a0f5-97d8-4422-8478-2e5f336409ab
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 8b06a0f5-97d8-4422-8478-2e5f336409ab
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # d85bf11b-9a09-4709-5bd9-00e44827f46a
 
 This model is a fine-tuned version of [NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0037
 
 ## Model description
 
@@ -128,14 +130,19 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_steps: 2
-- training_steps: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0003 | 1    | 1.4694          |
+| 1.4291        | 0.0033 | 10   | 1.3230          |
+| 1.0152        | 0.0067 | 20   | 1.1381          |
+| 1.0069        | 0.0100 | 30   | 1.0423          |
+| 0.8803        | 0.0133 | 40   | 1.0084          |
+| 0.9714        | 0.0167 | 50   | 1.0037          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc38364a73aa8fd5351f36368a9bc8d5afe4b96ef1525e7fbeee7e2d8f4dae12
+oid sha256:e0fe72360e71bd635fbd398e1753bf9a148047eaf6a17dce17beb97fe0b7cf34
 size 84047370