End of training

README.md

@@ -65,7 +65,7 @@ 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/4ca900f47590699f_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: d03b1aa7-3120-4eb3-9690-a45cc19e5306
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: d03b1aa7-3120-4eb3-9690-a45cc19e5306
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # d03b1aa7-3120-4eb3-9690-a45cc19e5306
 
 This model is a fine-tuned version of [NousResearch/Nous-Hermes-2-Mistral-7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.9541
 
 ## Model description
 
@@ -127,14 +129,19 @@ 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        | 0.0018 | 1    | 1.5324          |
+| 5.5289        | 0.0180 | 10   | 1.2460          |
+| 4.1946        | 0.0361 | 20   | 1.0456          |
+| 4.0129        | 0.0541 | 30   | 0.9842          |
+| 3.7932        | 0.0721 | 40   | 0.9596          |
+| 3.6417        | 0.0902 | 50   | 0.9541          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84a24ec5a488fbfff0d7c4e8cb0549c54029083e6d855f71c28164839f7793c0
+oid sha256:c5c2f3e49bcf9946a3881cee0851291706aba40e2a52fdc7dd30e0dcc555e8e2
 size 84047370