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/6f3649c1a2a2b72c_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 19636305-21ab-4ea1-bd58-1ed1ac9c4a9b
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 19636305-21ab-4ea1-bd58-1ed1ac9c4a9b
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 0e78ea43-b2e7-4c16-8c9c-cf09a6df9733
 
 This model is a fine-tuned version of [NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Alternate-Tokenizer) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6494
 
 ## 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        | 0.0008 | 1    | 3.5504          |
+| 3.4471        | 0.0080 | 10   | 2.9990          |
+| 1.9901        | 0.0160 | 20   | 1.8979          |
+| 1.6579        | 0.0239 | 30   | 1.7123          |
+| 1.6844        | 0.0319 | 40   | 1.6592          |
+| 1.6745        | 0.0399 | 50   | 1.6494          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c409de60c4b23f91aaf5fc1078fe27f73726f8f80fff3e1d622d71e65caf05d
+oid sha256:2a7529c4b56bfba5ae575800fc1626e187e5a070fe0b3ebd7582f1cc6def991d
 size 84047370