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/a16c9edca92af56e_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: b4b4193f-f7bc-4e75-87fa-eca0baf67f2c
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: b4b4193f-f7bc-4e75-87fa-eca0baf67f2c
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 024279b3-5062-4b90-840c-cfb4054486ce
 
 This model is a fine-tuned version of [unsloth/gemma-2-2b](https://huggingface.co/unsloth/gemma-2-2b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.7137
 
 ## 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        | 0.0025 | 1    | 1.8971          |
+| 2.0113        | 0.0254 | 10   | 1.8218          |
+| 1.8522        | 0.0509 | 20   | 1.7397          |
+| 1.8622        | 0.0763 | 30   | 1.7218          |
+| 1.7949        | 0.1017 | 40   | 1.7146          |
+| 1.8294        | 0.1271 | 50   | 1.7137          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e83254e0a35f89393e9407269e6e136af26e0cf1b24fbeef83a9a404191e2c8
+oid sha256:b5b3a670f7ce50eafa879d0a67dc70e8cbfdafe63db090d944924c9a80416685
 size 41664442