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/9092961d53ecf111_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: f0b61e07-9d2c-49db-9d6b-55033bff6730
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: f0b61e07-9d2c-49db-9d6b-55033bff6730
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # f0b61e07-9d2c-49db-9d6b-55033bff6730
 
 This model is a fine-tuned version of [unsloth/codegemma-7b](https://huggingface.co/unsloth/codegemma-7b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.4299
 
 ## Model description
 
@@ -129,13 +131,18 @@ 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: 30
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0333 | 1    | 1.7955          |
+| No log        | 0.2    | 6    | 1.5877          |
+| 1.6572        | 0.4    | 12   | 1.4679          |
+| 1.6572        | 0.6    | 18   | 1.4380          |
+| 1.4574        | 0.8    | 24   | 1.4285          |
+| 1.438         | 1.0    | 30   | 1.4299          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5c3cd3b522111797584ca955170ab3748ad66721e9b4711adca6d211eb0acf1
+oid sha256:56b21cfe1235857aa14623309fb28f5c96e4d8395d9067aa13da1a537314ec3d
 size 100149034