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/e24fd71460cf9a49_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: e89008ec-0ddd-42bd-bbd1-7f9ce3f4ee10
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: e89008ec-0ddd-42bd-bbd1-7f9ce3f4ee10
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # e89008ec-0ddd-42bd-bbd1-7f9ce3f4ee10
 
 This model is a fine-tuned version of [TinyLlama/TinyLlama_v1.1](https://huggingface.co/TinyLlama/TinyLlama_v1.1) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.9998
 
 ## 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        | 0.0041 | 1    | 7.3914          |
+| 6.7943        | 0.0408 | 10   | 6.0374          |
+| 3.7522        | 0.0816 | 20   | 3.0888          |
+| 2.3264        | 0.1224 | 30   | 2.1280          |
+| 1.9886        | 0.1633 | 40   | 2.0036          |
+| 1.843         | 0.2041 | 50   | 1.9998          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a269dcff0f71dc7f77eb5129d906a95c8a85467ddb0b30b0d0b8ade348b85d6
+oid sha256:8f34cd8ad8336057cc20a305f24c5406d4793a5e9a8945621c4f4c910ee5632d
 size 25342042