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/da433f4b49cbfbfc_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 48230c59-9dee-4f6f-a7d8-47950f14e20e
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 48230c59-9dee-4f6f-a7d8-47950f14e20e
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 48230c59-9dee-4f6f-a7d8-47950f14e20e
 
 This model is a fine-tuned version of [defog/llama-3-sqlcoder-8b](https://huggingface.co/defog/llama-3-sqlcoder-8b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2578
 
 ## 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        | 0.0013 | 1    | 16.5815         |
+| 11.5695       | 0.0133 | 10   | 5.2219          |
+| 0.3263        | 0.0267 | 20   | 0.4066          |
+| 0.4301        | 0.0400 | 30   | 0.3173          |
+| 0.3096        | 0.0534 | 40   | 0.2830          |
+| 0.1974        | 0.0667 | 50   | 0.2578          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87adc3ea712d34830f38088d3d72cb829bfd91585dee464defe53d511f6a8727
+oid sha256:1be25e5ae723dbaa1d8ce93cb9fb19b48970a622aafa3babde46b1f5b3c9eddf
 size 84047370