End of training

README.md

@@ -64,7 +64,7 @@ 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/67c40f5b4e50e6b5_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: d036b0f5-2a10-4e6f-8ea9-a2702daf7d6d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: d036b0f5-2a10-4e6f-8ea9-a2702daf7d6d
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,6 +100,8 @@ xformers_attention: null
 # d036b0f5-2a10-4e6f-8ea9-a2702daf7d6d
 
 This model is a fine-tuned version of [trl-internal-testing/tiny-random-LlamaForCausalLM](https://huggingface.co/trl-internal-testing/tiny-random-LlamaForCausalLM) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 10.3818
 
 ## Model description
 
@@ -127,13 +129,18 @@ 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: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0027 | 1    | 10.3826         |
+| 10.3825       | 0.0274 | 10   | 10.3824         |
+| 10.3824       | 0.0547 | 20   | 10.3821         |
+| 10.3818       | 0.0821 | 30   | 10.3819         |
+| 10.382        | 0.1095 | 40   | 10.3818         |
+| 10.3821       | 0.1368 | 50   | 10.3818         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9ad185c21eb75cbf5b37f7305182d389f60948e7eb223be97f5396c25394a04
+oid sha256:1e108da8e783660f949b6c0edc49b84d0b3a5b2f76ec1960a81e3500df13d5f1
 size 33666