hansken_human_hql_v3

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the hansh/hansken_hql_cot dataset. It achieves the following results on the evaluation set:

• Loss: 0.5017

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0002
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss
0.6267	1.0	469	0.6078
0.5094	2.0	938	0.5335
0.513	3.0	1407	0.5142
0.4306	4.0	1876	0.5044
0.4128	5.0	2345	0.5017
0.3924	6.0	2814	0.5093
0.3684	7.0	3283	0.5168
0.3403	8.0	3752	0.5338
0.311	9.0	4221	0.5566
0.2853	10.0	4690	0.5920

Framework versions

• PEFT 0.12.0
• Transformers 4.43.1
• Pytorch 2.1.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1