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/22877a35a54297fd_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 3667de4d-f009-44b2-ba5c-c58a3f3b0f7d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 3667de4d-f009-44b2-ba5c-c58a3f3b0f7d
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 3667de4d-f009-44b2-ba5c-c58a3f3b0f7d
 
 This model is a fine-tuned version of [unsloth/gemma-2b-it](https://huggingface.co/unsloth/gemma-2b-it) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0000
 
 ## 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        | 0.0001 | 1    | 8.6423          |
+| 4.7642        | 0.0011 | 10   | 1.3221          |
+| 0.0006        | 0.0022 | 20   | 0.0000          |
+| 0.0           | 0.0033 | 30   | 0.0000          |
+| 0.0           | 0.0044 | 40   | 0.0000          |
+| 0.0           | 0.0055 | 50   | 0.0000          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80e4f86b53409f3012ee86615ac85635d0e91a76b0db9d15bfabb028b20a39e3
+oid sha256:d6611eebc293407effa4f442786f9b6c87506b15e8260defd7e885dab00e61b3
 size 39313978