| { | |
| "architectures": [ | |
| "DbrxForCausalLM" | |
| ], | |
| "attn_config": { | |
| "attn_pdrop": 0.0, | |
| "clip_qkv": 8, | |
| "kv_n_heads": 1, | |
| "model_type": "", | |
| "rope_theta": 500000 | |
| }, | |
| "bos_token_id": 50256, | |
| "d_model": 32, | |
| "dtype": "float32", | |
| "emb_pdrop": 0.0, | |
| "eos_token_id": 50256, | |
| "ffn_config": { | |
| "ffn_act_fn": { | |
| "name": "gelu" | |
| }, | |
| "ffn_hidden_size": 64, | |
| "hidden_size": 32, | |
| "model_type": "", | |
| "moe_jitter_eps": 0.0, | |
| "moe_loss_weight": 0.05, | |
| "moe_normalize_expert_weights": 1.0, | |
| "moe_num_experts": 8, | |
| "moe_top_k": 4 | |
| }, | |
| "initializer_range": 0.02, | |
| "is_decoder": true, | |
| "max_seq_len": 512, | |
| "model_type": "dbrx", | |
| "n_heads": 2, | |
| "n_layers": 2, | |
| "num_key_value_heads": 1, | |
| "output_router_logits": false, | |
| "pad_token_id": 0, | |
| "resid_pdrop": 0.0, | |
| "rope_parameters": { | |
| "rope_theta": 10000.0, | |
| "rope_type": "default" | |
| }, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "5.16.0.dev0", | |
| "use_cache": true, | |
| "vocab_size": 50257 | |
| } | |