| { | |
| "apply_residual_connection_post_layernorm": false, | |
| "architectures": [ | |
| "GPTNeoXForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 0, | |
| "dht_prefix": "Kide2006/test-bloomd-6b3", | |
| "eos_token_id": 0, | |
| "hidden_act": "gelu", | |
| "hidden_dropout": 0.0, | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8192, | |
| "layer_norm_eps": 1e-05, | |
| "layer_norm_epsilon": 1e-05, | |
| "max_position_embeddings": 2048, | |
| "model_type": "bloom", | |
| "n_head": 8, | |
| "n_layer": 16, | |
| "pretraining_tp": 1, | |
| "rotary_emb_base": 10000, | |
| "rotary_pct": 0.25, | |
| "slow_but_exact": false, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.28.0.dev0", | |
| "use_cache": true, | |
| "use_parallel_residual": true, | |
| "vocab_size": 50304 | |
| } | |