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/5fb8db7f149b506f_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: d3c2cc22-fd17-4205-b721-afe8dce56b58
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: d3c2cc22-fd17-4205-b721-afe8dce56b58
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # be9508db-b041-0e4f-a796-9c3f891f2986
 
 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: 0.7383
 
 ## 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        | 0.0004 | 1    | 1.0386          |
+| 3.7081        | 0.0041 | 10   | 0.9729          |
+| 3.3143        | 0.0082 | 20   | 0.8414          |
+| 3.2341        | 0.0122 | 30   | 0.7632          |
+| 2.8461        | 0.0163 | 40   | 0.7422          |
+| 2.915         | 0.0204 | 50   | 0.7383          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0b2c8c10309adeca16d403a79f8d843910f61c035ab51b0a6c3fc896d95c9e8
+oid sha256:9d7169443115866b0f7da048a625772035ed6184b0bdb847eb940d5ea3c47824
 size 94477642