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/f35d76710dd4624f_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 3a33c715-523a-4987-9e25-23b57afcd0dd
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 3a33c715-523a-4987-9e25-23b57afcd0dd
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 3a33c715-523a-4987-9e25-23b57afcd0dd
 
 This model is a fine-tuned version of [llamafactory/tiny-random-Llama-3](https://huggingface.co/llamafactory/tiny-random-Llama-3) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.7654
 
 ## Model description
 
@@ -130,14 +132,19 @@ 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        | 0.0008 | 1    | 11.7664         |
+| 11.7637       | 0.0078 | 10   | 11.7662         |
+| 11.7622       | 0.0156 | 20   | 11.7659         |
+| 11.7613       | 0.0235 | 30   | 11.7656         |
+| 11.7617       | 0.0313 | 40   | 11.7654         |
+| 11.7621       | 0.0391 | 50   | 11.7654         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fde731563bc7a058c7178f053acfe9712c2ab43ba378310a2bdecf40161fc31
+oid sha256:f8d1b9021257822cf28d92ae3d18e75f106fa320fb6959b9511518a563d98691
 size 33666