test_2 / config.json
BICORP's picture
Upload 2 files
a7ee440 verified
raw
history blame contribute delete
526 Bytes
{"max_batch_size": 2, "max_seq_len": 8192, "dtype": "bf16", "vocab_size": 5557, "dim": 2096, "inter_dim": 1048, "moe_inter_dim": 548, "n_layers": 12, "n_dense_layers": 1, "n_heads": 16, "n_routed_experts": 32, "n_shared_experts": 2, "n_activated_experts": 6, "score_func": "softmax", "route_scale": 1.0, "q_lora_rank": 0, "kv_lora_rank": 512, "qk_nope_head_dim": 128, "qk_rope_head_dim": 64, "v_head_dim": 128, "original_seq_len": 4096, "rope_theta": 10000.0, "rope_factor": 40, "beta_fast": 32, "beta_slow": 1, "mscale": 1.0}