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/590c61e95721fac8_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 811718e1-2da8-4082-a53e-4ce832a92c02
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 811718e1-2da8-4082-a53e-4ce832a92c02
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 811718e1-2da8-4082-a53e-4ce832a92c02
 
 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: 1.3965
 
 ## 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    | 6.2469          |
+| 14.923        | 0.0012 | 10   | 2.5303          |
+| 6.52          | 0.0023 | 20   | 1.5055          |
+| 5.7146        | 0.0035 | 30   | 1.4313          |
+| 5.6891        | 0.0047 | 40   | 1.4018          |
+| 5.4539        | 0.0058 | 50   | 1.3965          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6538fc99c89bf38a201566384bda0c1ec90c67777f617f809e26c838507521a3
+oid sha256:c7c8d108841db257f32229fbd530216751442a477d9d101688fa85e109cb567f
 size 84047370