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/5b50a4b0fd05b0f4_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: c2f2664d-127b-49f9-976d-7c0414ad11de
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: c2f2664d-127b-49f9-976d-7c0414ad11de
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # 8b9d2c85-f951-4fe6-b561-c59e8289ac66
 
 This model is a fine-tuned version of [aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct](https://huggingface.co/aisingapore/llama3-8b-cpt-sea-lionv2.1-instruct) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.9738
 
 ## 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        | 0.0008 | 1    | 3.6270          |
+| 3.387         | 0.0075 | 10   | 3.3811          |
+| 3.0071        | 0.0150 | 20   | 3.0675          |
+| 2.939         | 0.0226 | 30   | 3.0065          |
+| 2.6426        | 0.0301 | 40   | 2.9789          |
+| 2.7795        | 0.0376 | 50   | 2.9738          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed26768ad578e9cd325dcb8bb184990aa44a505edd3eab095a444aa3f57c384d
+oid sha256:934393cb13f4919ec2a3e1d186b667885a229bdcf9de681b6e4e8976e7b4654a
 size 84047370