End of training

README.md

@@ -66,7 +66,7 @@ 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/c78d73cb9578957c_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 0e9a71c2-29a0-42c7-846d-e0ce6e6fcb18
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 0e9a71c2-29a0-42c7-846d-e0ce6e6fcb18
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 254d521e-51bc-8362-8ddf-36ee95e5e56f
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3278
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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    | 3.2057          |
+| 2.8572        | 0.0004 | 10   | 3.0713          |
+| 2.6758        | 0.0007 | 20   | 2.7240          |
+| 2.3199        | 0.0011 | 30   | 2.3954          |
+| 2.2798        | 0.0015 | 40   | 2.3342          |
+| 2.4558        | 0.0018 | 50   | 2.3278          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:363185fa08104fdc21e92a5166cdaedeeb294e8723f36a2884bdd4928e48bf78
+oid sha256:55515e31eec3d252d7fdd6e02978f641c18b7763f9c5da5d4007bd30471c20ef
 size 80881450