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/4c4b009282d6d05b_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: 34ad7f9d-ff58-4068-8537-2d15a40438a9
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 34ad7f9d-ff58-4068-8537-2d15a40438a9
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # 70648ed1-68ff-4b27-bdb2-a18dec185c76
 
 This model is a fine-tuned version of [HuggingFaceH4/tiny-random-LlamaForCausalLM](https://huggingface.co/HuggingFaceH4/tiny-random-LlamaForCausalLM) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 10.3763
 
 ## 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        | 0.0004 | 1    | 10.3774         |
+| 10.3766       | 0.0043 | 10   | 10.3772         |
+| 10.3777       | 0.0085 | 20   | 10.3768         |
+| 10.377        | 0.0128 | 30   | 10.3765         |
+| 10.376        | 0.0170 | 40   | 10.3763         |
+| 10.3775       | 0.0213 | 50   | 10.3763         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84bb4a7ec4fcdd6f077ecac27c6cc8089d14a1a4ddcb9f8737d9b8164df92975
+oid sha256:06775aa99aa57bcce7808df9c60e360d6d92273c6bb8c896109e1b3d1de9e346
 size 33666