| { | |
| "_name_or_path": "plankton-100M", | |
| "architectures": [ | |
| "PlanktonForCausalLM" | |
| ], | |
| "bos_token_id": 0, | |
| "eos_token_id": 1, | |
| "hidden_act": "silu", | |
| "hidden_size": 600, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 2000, | |
| "max_position_embeddings": 4096, | |
| "model_type": "PLANKTON", | |
| "num_attention_heads": 10, | |
| "num_hidden_layers": 10, | |
| "num_key_value_heads": 10, | |
| "pad_token_id": 3, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.33.1", | |
| "unk_token_id": 2, | |
| "use_cache": true, | |
| "vocab_size": 42000 | |
| } | |