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/6d7606ae1dc9acbd_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 603350de-8085-46cc-b1ad-6348c50fb71d
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 603350de-8085-46cc-b1ad-6348c50fb71d
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 603350de-8085-46cc-b1ad-6348c50fb71d
 
 This model is a fine-tuned version of [openlm-research/open_llama_3b](https://huggingface.co/openlm-research/open_llama_3b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.3587
 
 ## 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        | 0.0004 | 1    | 9.4619          |
+| 9.4056        | 0.0039 | 10   | 8.9513          |
+| 6.2763        | 0.0078 | 20   | 4.9152          |
+| 2.59          | 0.0116 | 30   | 2.0999          |
+| 1.6106        | 0.0155 | 40   | 1.4494          |
+| 1.3891        | 0.0194 | 50   | 1.3587          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfc564b03b10024ac032a9467e090c3aefa5120536aeb8d9ae2999f560f520f8
+oid sha256:8498a3ceff05e422833f2a88ccf3678218284b3b005390fb5aa58b67ff263494
 size 50982842