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/bb1be9d7d28607ee_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 48cd521d-1f66-41d7-88a6-2b0d24cf07b1
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 48cd521d-1f66-41d7-88a6-2b0d24cf07b1
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 93bd4aa5-744c-b235-c2f2-07a6c9098d12
 
 This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v0.6](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8212
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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        | 0.0001 | 1    | 0.8483          |
+| 0.8052        | 0.0006 | 10   | 0.8474          |
+| 0.8188        | 0.0012 | 20   | 0.8384          |
+| 0.8191        | 0.0018 | 30   | 0.8269          |
+| 0.8342        | 0.0024 | 40   | 0.8220          |
+| 0.797         | 0.0029 | 50   | 0.8212          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43009314905ce23e314342d2975380c8c61ff3b6f70950ee23f570b07da30474
+oid sha256:afccaf5cb4c01958b560d83265007a74071003ba79b12b9f39e60b5f55ac4a0a
 size 25342042