End of training

README.md

@@ -65,7 +65,7 @@ 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/f4030b48c4e6cfd6_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 2b0e28e5-e8ba-4372-9369-fedde1e2be48
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 2b0e28e5-e8ba-4372-9369-fedde1e2be48
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 2b0e28e5-e8ba-4372-9369-fedde1e2be48
 
 This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v0.6](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0339
 
 ## Model description
 
@@ -127,14 +129,19 @@ 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        | 0.0002 | 1    | 1.0664          |
+| 0.9289        | 0.0016 | 10   | 1.0648          |
+| 1.0568        | 0.0033 | 20   | 1.0532          |
+| 0.9845        | 0.0049 | 30   | 1.0406          |
+| 1.0015        | 0.0065 | 40   | 1.0349          |
+| 0.9758        | 0.0082 | 50   | 1.0339          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0772ebf7e2f0918438223002045481840ad87a1a56cca6e28df3e446a510bfb9
+oid sha256:840a9f462206de0c65e33a3d3a8066121150a19e5895f7e187f78d7aa916dd9c
 size 25342042