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/1fd166178dc611b1_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: d2e73c64-4432-4749-92a6-b96236c9a7bb
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: d2e73c64-4432-4749-92a6-b96236c9a7bb
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # d907b078-43f5-47f0-856a-1b45b2603181
 
 This model is a fine-tuned version of [fxmarty/really-tiny-falcon-testing](https://huggingface.co/fxmarty/really-tiny-falcon-testing) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.0541
 
 ## 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        | 0.0033 | 1    | 11.0767         |
+| 44.3017       | 0.0326 | 10   | 11.0749         |
+| 44.2679       | 0.0653 | 20   | 11.0665         |
+| 44.25         | 0.0979 | 30   | 11.0591         |
+| 44.222        | 0.1305 | 40   | 11.0549         |
+| 44.216        | 0.1631 | 50   | 11.0541         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09fa7446db1f2b26849da40667e9046650c5024d9d8b77823d07f5dcc728cd74
+oid sha256:8c9bf9e49e3bab2b245e85fd8f33e13948e2555dd0c49a9b0fa19ee5cc8bcd4b
 size 37022