train_cb_42_1760637133

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

• Loss: 0.2378
• Num Input Tokens Seen: 158656

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: 42
• 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
1.1162	0.5	25	0.7037	14720
0.2568	1.0	50	0.2891	31456
0.254	1.5	75	0.2438	47168
0.1647	2.0	100	0.3600	63168
0.3223	2.5	125	0.2357	79424
0.2999	3.0	150	0.2933	95168
0.244	3.5	175	0.2299	111392
0.2501	4.0	200	0.2389	127136
0.1977	4.5	225	0.2310	141632
0.1052	5.0	250	0.2378	158656

Framework versions

• PEFT 0.17.1
• Transformers 4.51.3
• Pytorch 2.9.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4