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/2be6cbcc03e4d1d0_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: 17e51ae1-bb59-4895-8f4e-b7862ee504bb
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 17e51ae1-bb59-4895-8f4e-b7862ee504bb
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 2537e6b4-124f-4bab-9685-de11d32f1b5f
 
 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: 0.5821
 
 ## 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        | 0.0001 | 1    | 1.8562          |
+| 6.9662        | 0.0008 | 10   | 1.3146          |
+| 3.2653        | 0.0015 | 20   | 0.7487          |
+| 2.6388        | 0.0023 | 30   | 0.6180          |
+| 2.4012        | 0.0030 | 40   | 0.5869          |
+| 2.2431        | 0.0038 | 50   | 0.5821          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3891eed21ee6790323fbc753c08600ffc765efbd11157d65dae7154293a4ef36
+oid sha256:cd5a8b38d80edab29dbb370ab7090c446b63652bc46b383abdc97b891a5c69b1
 size 94477642