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/f8232566451d90d2_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: c2e82a22-8bc3-4a6a-8590-0acbfe70412e
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: c2e82a22-8bc3-4a6a-8590-0acbfe70412e
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 0e9ea84e-3c6a-4479-bc96-ae0c9697201f
 
 This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-7B-Instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.7086
 
 ## 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        | 0.0001 | 1    | 4.9000          |
+| 4.705         | 0.0012 | 10   | 4.8250          |
+| 3.7779        | 0.0024 | 20   | 4.1533          |
+| 3.589         | 0.0037 | 30   | 3.8432          |
+| 3.9802        | 0.0049 | 40   | 3.7290          |
+| 3.8593        | 0.0061 | 50   | 3.7086          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c125df8f18fd1e6040ae14362eb9160d4e5a4dbbad5a132a2351b0d0bfc1bc15
+oid sha256:d20d2bca0a867e79c195a52163b71c188d95a1875e0fedbb65c06db37e7f7c3b
 size 80881450