Llama-31-8B_task-2_60-samples_config-1_full_auto

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the GaetanMichelet/chat-60_ft_task-2_auto dataset. It achieves the following results on the evaluation set:

• Loss: 1.0304

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: 0.0001
• 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: 50

Training results

Training Loss	Epoch	Step	Validation Loss
1.4989	0.8696	5	1.5039
1.4648	1.9130	11	1.4134
1.3625	2.9565	17	1.3141
1.2061	4.0	23	1.2032
1.078	4.8696	28	1.1121
0.9963	5.9130	34	1.0649
0.9285	6.9565	40	1.0479
0.9295	8.0	46	1.0360
0.9178	8.8696	51	1.0304
0.9031	9.9130	57	1.0313
0.8224	10.9565	63	1.0367
0.7715	12.0	69	1.0482
0.7775	12.8696	74	1.0640
0.6631	13.9130	80	1.1065
0.6127	14.9565	86	1.1382
0.6162	16.0	92	1.1660

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.1.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1