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: 2
 mlflow_experiment_name: /tmp/08d5261bf693cde0_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 0d5df66a-fe13-409b-a177-c353c2baadb3
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 0d5df66a-fe13-409b-a177-c353c2baadb3
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 0d5df66a-fe13-409b-a177-c353c2baadb3
 
 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: 0.0334
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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        | 0.0036 | 1    | 2.6167          |
+| 2.3494        | 0.0365 | 10   | 1.9993          |
+| 1.1345        | 0.0730 | 20   | 0.8235          |
+| 0.3124        | 0.1095 | 30   | 0.1790          |
+| 0.0655        | 0.1460 | 40   | 0.0458          |
+| 0.0427        | 0.1825 | 50   | 0.0334          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9057395f9ee2b6101ce175f1ea5e47bc5d0180844ea84a610a09a8baba61119
+oid sha256:567d14f2dee330f774277b8ebec80c29c53403ae47e6081e1c61a790ae6fe004
 size 41664442