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/0e2c0576f93a90d8_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 0739603f-1adf-470c-961b-f4f2cd088074
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 0739603f-1adf-470c-961b-f4f2cd088074
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 23b30273-cd15-b1c6-cc23-05e44ecfee97
 
 This model is a fine-tuned version of [OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5](https://huggingface.co/OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.7351
 
 ## 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        | 0.0003 | 1    | 2.0185          |
+| 7.9446        | 0.0032 | 10   | 1.9448          |
+| 7.0552        | 0.0064 | 20   | 1.8021          |
+| 6.9021        | 0.0096 | 30   | 1.7518          |
+| 7.1911        | 0.0129 | 40   | 1.7379          |
+| 6.687         | 0.0161 | 50   | 1.7351          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da69a19a36f57ed304ccef277b6f51c3017c07f587584105dbb4c3f6c35d1a79
+oid sha256:494c24e6bd8aea92ec887a46f4dd568c719bbae2a97f0edfcf089da509557481
 size 94477642