• Model creator: https://huggingface.co/barc0/
• Original model: https://huggingface.co/barc0/transduction-40k-seed100-instruct-fft_lr1e-5_epoch2/

Original model description:

• alignment-handbook
• trl
• sft
• generated_from_trainer
• trl
• sft
• generated_from_trainer datasets:
• barc0/transduction_100k_gpt4o-mini_generated_problems_seed100.jsonl_messages_format_0.3 model-index:
• name: transduction-40k-seed100-instruct-fft_lr1e-5_epoch2 results: []

transduction-40k-seed100-instruct-fft_lr1e-5_epoch2

• Loss: 0.0587

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

• learning_rate: 1e-05
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 2

Training results

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 3.0.1
• Tokenizers 0.19.1