l3.1-8b-inst-fft-induction-barc-heavy-200k-old-200k-lr1e-5-ep3

This model is a fine-tuned version of meta-llama/Meta-Llama-3.1-8B-Instruct on the barc0/induction_heavy_100k_jsonl, the barc0/induction_heavy_suggestfunction_100k_jsonl, the barc0/induction_100k-gpt4-description-gpt4omini-code_generated_problems_messages_format_0.3 and the barc0/induction_100k_gpt4o-mini_generated_problems_seed100.jsonl_messages_format_0.3 datasets. It achieves the following results on the evaluation set:

• Loss: 0.3711

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: 16
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• total_eval_batch_size: 32
• 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.4395	1.0	2995	0.4231
0.3529	2.0	5990	0.3733
0.2858	3.0	8985	0.3711

Framework versions

• Transformers 4.46.2
• Pytorch 2.4.0+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3