train_multirc_123_1764892185

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the multirc dataset. It achieves the following results on the evaluation set:

• Loss: 0.6626
• Num Input Tokens Seen: 235053712

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 123
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.2708	2.0	10898	0.2877	23516768
0.0961	4.0	21796	0.1894	47040736
0.0933	6.0	32694	0.1662	70557920
0.0005	8.0	43592	0.2701	94040064
0.0001	10.0	54490	0.3495	117532800
0.0002	12.0	65388	0.3732	141036960
0.0	14.0	76286	0.5068	164523360
0.0	16.0	87184	0.6248	188023968
0.0	18.0	98082	0.6490	211546832
0.0	20.0	108980	0.6626	235053712

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1