train_multirc_789_1770096768

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: 1.6388
• Num Input Tokens Seen: 235033616

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: 789
• 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
0.2758	2.0	10898	0.2922	23502720
0.2528	4.0	21796	0.2865	46986144
0.4294	6.0	32694	0.2955	70499968
0.1299	8.0	43592	0.3384	94033824
0.1786	10.0	54490	0.5955	117557568
0.011	12.0	65388	0.9427	141038096
0.0001	14.0	76286	1.3824	164567856
0.0	16.0	87184	1.5856	188068288
0.0	18.0	98082	1.6488	211547568
0.0	20.0	108980	1.6388	235033616

Framework versions

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