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: 2
 mlflow_experiment_name: /tmp/65b2c1fff352fd70_train_data.json
 model_type: AutoModelForCausalLM
@@ -89,7 +89,7 @@ wandb_name: 8a67a667-0208-4681-bace-396b2e10c72c
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 8a67a667-0208-4681-bace-396b2e10c72c
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -100,6 +100,8 @@ xformers_attention: null
 # 8a67a667-0208-4681-bace-396b2e10c72c
 
 This model is a fine-tuned version of [peft-internal-testing/tiny-dummy-qwen2](https://huggingface.co/peft-internal-testing/tiny-dummy-qwen2) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.9302
 
 ## Model description
 
@@ -126,14 +128,19 @@ 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    | 11.9304         |
+| 11.931        | 0.0007 | 10   | 11.9303         |
+| 11.928        | 0.0015 | 20   | 11.9303         |
+| 11.9303       | 0.0022 | 30   | 11.9302         |
+| 11.9328       | 0.0030 | 40   | 11.9302         |
+| 11.9285       | 0.0037 | 50   | 11.9302         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e1aa2e812e75408660a19c6bb94b0f39a9686d431edc714a3841bbc4b28d5cd
+oid sha256:33dc7dbc361f4b8302c0914d51381bd5da53de35f923dad2505d75645150594f
 size 21378