nikitastheo/v6-mixed-25k-lower-shared-ewc10-deu-ell-sequential
Trained with train_clm.py, a Hugging Face Accelerate causal-LM training script (no Trainer).
Training details
- Base config:
configurations/gpt_base_config.json - Tokenizer:
nikitastheo/goldfish-25k-deu-lower-tokenizer - Max steps: 15610
- Learning rate: 0.0001
- LR scheduler: linear
- Warmup steps: 1561
- Batch size (per device): 32
- Gradient accumulation steps: 1
- Total train batch size: 32
- Language switch epoch: 10