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/95f83689ef25550f_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: b4543482-e72d-4cec-aef5-ca81787e529a
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: b4543482-e72d-4cec-aef5-ca81787e529a
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # cf78d9f8-2913-3a5c-70e4-a0106a03b7e8
 
 This model is a fine-tuned version of [unsloth/mistral-7b-instruct-v0.3](https://huggingface.co/unsloth/mistral-7b-instruct-v0.3) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5461
 
 ## 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        | 0.0039 | 1    | 0.9152          |
+| 2.8376        | 0.0390 | 10   | 0.6592          |
+| 2.1865        | 0.0780 | 20   | 0.5718          |
+| 2.3561        | 0.1170 | 30   | 0.5550          |
+| 2.1427        | 0.1559 | 40   | 0.5474          |
+| 2.182         | 0.1949 | 50   | 0.5461          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5dff9550d5e9ed587bf69f5e29e59048390f0326d1d825aba53686d8010aea9
+oid sha256:85b553ecefa06b271be82024b5b7f063902fd8b5111fa0dc74ba0c4cd651099a
 size 84047370