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/8a752da9fdd4976e_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 7fd3020e-17c6-4bcc-926f-1357afe7894c
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 7fd3020e-17c6-4bcc-926f-1357afe7894c
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # fe587437-34b7-f371-c523-518aa6dcf09b
 
 This model is a fine-tuned version of [unsloth/gemma-7b-it](https://huggingface.co/unsloth/gemma-7b-it) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.8229
 
 ## 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        | 0.0009 | 1    | 6.6454          |
+| 4.4934        | 0.0091 | 10   | 3.8041          |
+| 2.5895        | 0.0182 | 20   | 2.1505          |
+| 2.416         | 0.0273 | 30   | 1.9078          |
+| 1.9629        | 0.0364 | 40   | 1.8378          |
+| 1.9387        | 0.0455 | 50   | 1.8229          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b986420e635c276b162e5974c4c6c450998c7705d27e4b6a995b7af2753d970f
+oid sha256:3d40647e7549225e02e5b61816fa65cf2a0d33f69508458e3cdca67cccba547d
 size 100149034