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/a1279926648ef033_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: f7b9da4f-b9ac-4e44-972b-27465e6c85df
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: f7b9da4f-b9ac-4e44-972b-27465e6c85df
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 3dace872-6e44-c483-c233-b41afd4a6d4f
 
 This model is a fine-tuned version of [NousResearch/Llama-3.2-1B](https://huggingface.co/NousResearch/Llama-3.2-1B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.7537
 
 ## 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        | 0.0004 | 1    | 2.1373          |
+| 1.4674        | 0.0036 | 10   | 2.1022          |
+| 1.5702        | 0.0071 | 20   | 1.9243          |
+| 1.4202        | 0.0107 | 30   | 1.7890          |
+| 1.4122        | 0.0142 | 40   | 1.7567          |
+| 1.4053        | 0.0178 | 50   | 1.7537          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80ba3acfe88279714425d292eabec4d0414cee5278c8457c78a5d5d66bc49ba6
+oid sha256:0455722d83f8b4d650f8da8653276936eb096c7d242275376b14d62048dd5992
 size 22624970