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/89dbf2c6d7e7022e_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 883e0faa-f4ff-4a6a-bc43-ce8be26f9d7f
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 883e0faa-f4ff-4a6a-bc43-ce8be26f9d7f
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 883e0faa-f4ff-4a6a-bc43-ce8be26f9d7f
 
 This model is a fine-tuned version of [unsloth/OpenHermes-2.5-Mistral-7B](https://huggingface.co/unsloth/OpenHermes-2.5-Mistral-7B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7368
 
 ## 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        | 0.0012 | 1    | 1.0530          |
+| 3.2635        | 0.0124 | 10   | 0.8407          |
+| 3.2063        | 0.0248 | 20   | 0.7789          |
+| 3.2268        | 0.0372 | 30   | 0.7557          |
+| 2.7974        | 0.0496 | 40   | 0.7392          |
+| 2.9837        | 0.0620 | 50   | 0.7368          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5265efb7f7dfce0e24f94cbd943f1c118e6d872c23cd5cbea5f0056ceb4ae010
+oid sha256:3c3b0cf84bc1ce2fa9e1b9cd910f337dad837a7e7c06734885ce726461fa81e8
 size 84047370