| { | |
| "model_type": "gpt2", | |
| "architectures": [ | |
| "ScratchTransformer" | |
| ], | |
| "_scratch_model": true, | |
| "vocab_size": 8000, | |
| "n_embd": 256, | |
| "n_layer": 16, | |
| "n_head": 16, | |
| "n_inner": 512, | |
| "n_positions": 4096, | |
| "resid_pdrop": 0.1, | |
| "embd_pdrop": 0.1, | |
| "attn_pdrop": 0.1, | |
| "d_model": 256, | |
| "n_layers": 16, | |
| "n_heads": 16, | |
| "d_ff": 512, | |
| "max_seq_len": 4096, | |
| "dropout": 0.1, | |
| "name": "scratch-model" | |
| } |