| { | |
| "model_type": "gpt-from-scratch", | |
| "n_layer": 8, | |
| "n_head": 8, | |
| "n_embd": 512, | |
| "block_size": 1024, | |
| "vocab_size": 50257, | |
| "dropout": 0.0, | |
| "tokenizer": "gpt2", | |
| "params_millions": 77.2, | |
| "trained_steps": 5000, | |
| "tokens_seen_millions": 80, | |
| "val_loss": 5.2376, | |
| "precision": "fp32 (trained in bf16)" | |
| } |