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/ba9fa334fe8cd172_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 6a3ceda6-4b19-4ab8-a78c-63864ff6668b
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 6a3ceda6-4b19-4ab8-a78c-63864ff6668b
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 6a3ceda6-4b19-4ab8-a78c-63864ff6668b
 
 This model is a fine-tuned version of [echarlaix/tiny-random-PhiForCausalLM](https://huggingface.co/echarlaix/tiny-random-PhiForCausalLM) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 6.9195
 
 ## Model description
 
@@ -129,14 +131,19 @@ 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        | 0.0000 | 1    | 6.9395          |
+| No log        | 0.0000 | 1    | 6.9394          |
+| 6.9377        | 0.0002 | 10   | 6.9362          |
+| 6.9316        | 0.0003 | 20   | 6.9289          |
+| 6.9242        | 0.0005 | 30   | 6.9231          |
+| 6.9202        | 0.0007 | 40   | 6.9201          |
+| 6.9172        | 0.0008 | 50   | 6.9195          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66602de649df6b3d9ddad3f40624f9d6468b171546f54e3cfdad195f5a796416
+oid sha256:c5820d591ec7c0af3a987bc0cea29a389d8530163cfe813ef0edb9600b7c3606
 size 168063