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/e6c3ac17f3dacb5a_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 9bfec056-5c30-47fb-90f2-3264fdae9205
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 9bfec056-5c30-47fb-90f2-3264fdae9205
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 553c68d5-3f0f-dca4-213d-f5a85b2339d0
 
 This model is a fine-tuned version of [echarlaix/tiny-random-mistral](https://huggingface.co/echarlaix/tiny-random-mistral) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 10.3718
 
 ## 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        | 0.0002 | 1    | 10.3769         |
+| 41.4991       | 0.0024 | 10   | 10.3761         |
+| 41.5071       | 0.0047 | 20   | 10.3743         |
+| 41.493        | 0.0071 | 30   | 10.3727         |
+| 41.5          | 0.0094 | 40   | 10.3719         |
+| 41.4824       | 0.0118 | 50   | 10.3718         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7135a7308f6b7fc588170b80464ff1cd0e5a2a6923ed747ceca967d24fc3ebc
+oid sha256:c83d3fa6b9a8245742abb60f7078dae15f3a5d6f41c751d416f9c4bdfc4306dd
 size 37506