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: 8
 mlflow_experiment_name: /tmp/c336792bc8fa053d_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 37567b39-1c8e-40ac-aa59-12887bc56d68
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 37567b39-1c8e-40ac-aa59-12887bc56d68
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 37567b39-1c8e-40ac-aa59-12887bc56d68
 
 This model is a fine-tuned version of [Maykeye/TinyLLama-v0](https://huggingface.co/Maykeye/TinyLLama-v0) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 6.9658
 
 ## Model description
 
@@ -131,13 +133,18 @@ The following hyperparameters were used during training:
 - 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
+- training_steps: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0097 | 1    | 10.8824         |
+| 10.4561       | 0.0966 | 10   | 9.4416          |
+| 8.7933        | 0.1932 | 20   | 8.1607          |
+| 7.8034        | 0.2899 | 30   | 7.3733          |
+| 7.1231        | 0.3865 | 40   | 7.0240          |
+| 6.9844        | 0.4831 | 50   | 6.9658          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e676eed511b0a7bab8cce7d2381a3a6c1b81abcd6898ce2faa8f42a04320d499
+oid sha256:bbafd452c5712beb65505794fa4a7bb8c4bf932ba96a07f1ca4eef801a0ce15b
 size 416906