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/7e74a6d550157d47_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: d312c45a-01ef-4b46-8176-553e264ca691
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: d312c45a-01ef-4b46-8176-553e264ca691
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # bfd6c8f5-e081-4f6e-94eb-037bd2a6a634
 
 This model is a fine-tuned version of [echarlaix/tiny-random-mistral](https://huggingface.co/echarlaix/tiny-random-mistral) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 10.3746
 
 ## Model description
 
@@ -129,14 +131,19 @@ 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        | 0.0000 | 1    | 10.3772         |
+| 41.5104       | 0.0004 | 10   | 10.3768         |
+| 41.5043       | 0.0009 | 20   | 10.3758         |
+| 41.5139       | 0.0013 | 30   | 10.3751         |
+| 41.5075       | 0.0017 | 40   | 10.3747         |
+| 41.4827       | 0.0022 | 50   | 10.3746         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:28ed34fd65a2cf085c50d39660856e353b25cefe90f1587c540dfb4311596aca
+oid sha256:b1ef2a50dfde49194a5ffbe8246faa9b5f85fa737de3baa38a4af69cb16733da
 size 37506