Views
No views yet
Disclaimer: Decensored model.
- Safe input → Safe output (rare exceptions may apply)
- Harmful input → Harmful output
- Best for testing; avoid public-facing applications.
Note:
- Intermediate product of random257/Qwen3.5-122B-A10B-trohrbaugh-Heretic-MTP.
- Experimental conversions; minor tensor misalignments, loading issues, or long-context quirks may occur. Use at your own risk.
BF16: Converted from trohrbaugh/Qwen3.5-122B-A10B-heretic (post-2605 version) to GGUF format.Modifications:
- Removed redundant
moe-fusedcomponents, which were causing conversion failures.- Excluded the MTP block.
- Updated the
general.namemetadata and split the model into multiple files.
1: UINT32 | 1 | GGUF.version = 3
2: UINT64 | 1 | GGUF.tensor_count = 879
3: UINT64 | 1 | GGUF.kv_count = 44
4: STRING | 1 | general.architecture = 'qwen35moe'
5: STRING | 1 | general.type = 'model'
6: INT32 | 1 | general.sampling.top_k = 20
7: FLOAT32 | 1 | general.sampling.top_p = 0.949999988079071
8: FLOAT32 | 1 | general.sampling.temp = 0.6000000238418579
9: STRING | 1 | general.name = 'Qwen3.5 122B A10B Heretic No MTP'
10: STRING | 1 | general.finetune = 'heretic'
11: STRING | 1 | general.basename = 'Qwen3.5'
12: STRING | 1 | general.size_label = '122B-A10B'
13: STRING | 1 | general.license = 'apache-2.0'
14: STRING | 1 | general.license.link = 'https://huggingface.co/Qwen/Qwen3.5-122B-A10B/blob/main/L...'
15: [STRING] | 5 | general.tags = ['heretic', 'uncensored', 'decensored', 'abliterated', 'image-text-to-text']
16: UINT32 | 1 | qwen35moe.block_count = 48
17: UINT32 | 1 | qwen35moe.context_length = 262144
18: UINT32 | 1 | qwen35moe.embedding_length = 3072
19: UINT32 | 1 | qwen35moe.attention.head_count = 32
20: UINT32 | 1 | qwen35moe.attention.head_count_kv = 2
21: [INT32] | 4 | qwen35moe.rope.dimension_sections = [11, 11, 10, 0]
22: FLOAT32 | 1 | qwen35moe.rope.freq_base = 10000000.0
23: FLOAT32 | 1 | qwen35moe.attention.layer_norm_rms_epsilon = 9.999999974752427e-07
24: UINT32 | 1 | qwen35moe.expert_count = 256
25: UINT32 | 1 | qwen35moe.expert_used_count = 8
26: UINT32 | 1 | qwen35moe.attention.key_length = 256
27: UINT32 | 1 | qwen35moe.attention.value_length = 256
28: UINT32 | 1 | general.file_type = 32
29: UINT32 | 1 | qwen35moe.expert_feed_forward_length = 1024
30: UINT32 | 1 | qwen35moe.expert_shared_feed_forward_length = 1024
31: UINT32 | 1 | qwen35moe.ssm.conv_kernel = 4
32: UINT32 | 1 | qwen35moe.ssm.state_size = 128
33: UINT32 | 1 | qwen35moe.ssm.group_count = 16
34: UINT32 | 1 | qwen35moe.ssm.time_step_rank = 64
35: UINT32 | 1 | qwen35moe.ssm.inner_size = 8192
36: UINT32 | 1 | qwen35moe.full_attention_interval = 4
37: UINT32 | 1 | qwen35moe.rope.dimension_count = 64
38: UINT32 | 1 | general.quantization_version = 2
39: STRING | 1 | tokenizer.ggml.model = 'gpt2'
40: STRING | 1 | tokenizer.ggml.pre = 'qwen35'
41: [STRING] | 248320 | tokenizer.ggml.tokens = ['!', '"', '#', '$', '%', '&', ...]
42: [INT32] | 248320 | tokenizer.ggml.token_type = [1, 1, 1, 1, 1, 1, ...]
43: [STRING] | 247587 | tokenizer.ggml.merges = ['Ġ Ġ', 'ĠĠ ĠĠ', 'i n', 'Ġ t', 'ĠĠĠĠ ĠĠĠĠ', 'e r', ...]
44: UINT32 | 1 | tokenizer.ggml.eos_token_id = 248046
45: UINT32 | 1 | tokenizer.ggml.padding_token_id = 248044
46: BOOL | 1 | tokenizer.ggml.add_bos_token = False
47: STRING | 1 | tokenizer.chat_template = '{%- set image_count = namespace(value=0) %}\n{%- set video...'
1: 762839040 | 3072, 248320, 1, 1 | BF16 | output.weight
2: 762839040 | 3072, 248320, 1, 1 | BF16 | token_embd.weight
3: 3072 | 3072, 1, 1, 1 | F32 | blk.0.attn_norm.weight
4: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_a
5: 49152 | 4, 12288, 1, 1 | F32 | blk.0.ssm_conv1d.weight
6: 64 | 64, 1, 1, 1 | F32 | blk.0.ssm_dt.bias
7: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_alpha.weight
8: 196608 | 3072, 64, 1, 1 | BF16 | blk.0.ssm_beta.weight
9: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.0.attn_qkv.weight
10: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.0.attn_gate.weight
11: 128 | 128, 1, 1, 1 | F32 | blk.0.ssm_norm.weight
12: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.0.ssm_out.weight
13: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.0.ffn_down_exps.weight
14: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_gate_exps.weight
15: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.0.ffn_up_exps.weight
16: 786432 | 3072, 256, 1, 1 | F32 | blk.0.ffn_gate_inp.weight
17: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.0.ffn_down_shexp.weight
18: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_gate_shexp.weight
19: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.0.ffn_up_shexp.weight
20: 3072 | 3072, 1, 1, 1 | F32 | blk.0.ffn_gate_inp_shexp.weight
21: 3072 | 3072, 1, 1, 1 | F32 | blk.0.post_attention_norm.weight
22: 3072 | 3072, 1, 1, 1 | F32 | blk.1.attn_norm.weight
23: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_a
24: 49152 | 4, 12288, 1, 1 | F32 | blk.1.ssm_conv1d.weight
25: 64 | 64, 1, 1, 1 | F32 | blk.1.ssm_dt.bias
26: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_alpha.weight
27: 196608 | 3072, 64, 1, 1 | BF16 | blk.1.ssm_beta.weight
28: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.1.attn_qkv.weight
29: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.1.attn_gate.weight
30: 128 | 128, 1, 1, 1 | F32 | blk.1.ssm_norm.weight
31: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.1.ssm_out.weight
32: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.1.ffn_down_exps.weight
33: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_gate_exps.weight
34: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.1.ffn_up_exps.weight
35: 786432 | 3072, 256, 1, 1 | F32 | blk.1.ffn_gate_inp.weight
36: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.1.ffn_down_shexp.weight
37: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_gate_shexp.weight
38: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.1.ffn_up_shexp.weight
39: 3072 | 3072, 1, 1, 1 | F32 | blk.1.ffn_gate_inp_shexp.weight
40: 3072 | 3072, 1, 1, 1 | F32 | blk.1.post_attention_norm.weight
41: 3072 | 3072, 1, 1, 1 | F32 | blk.2.attn_norm.weight
42: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_a
43: 49152 | 4, 12288, 1, 1 | F32 | blk.2.ssm_conv1d.weight
44: 64 | 64, 1, 1, 1 | F32 | blk.2.ssm_dt.bias
45: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_alpha.weight
46: 196608 | 3072, 64, 1, 1 | BF16 | blk.2.ssm_beta.weight
47: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.2.attn_qkv.weight
48: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.2.attn_gate.weight
49: 128 | 128, 1, 1, 1 | F32 | blk.2.ssm_norm.weight
50: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.2.ssm_out.weight
51: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.2.ffn_down_exps.weight
52: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_gate_exps.weight
53: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.2.ffn_up_exps.weight
54: 786432 | 3072, 256, 1, 1 | F32 | blk.2.ffn_gate_inp.weight
55: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.2.ffn_down_shexp.weight
56: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_gate_shexp.weight
57: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.2.ffn_up_shexp.weight
58: 3072 | 3072, 1, 1, 1 | F32 | blk.2.ffn_gate_inp_shexp.weight
59: 3072 | 3072, 1, 1, 1 | F32 | blk.2.post_attention_norm.weight
60: 3072 | 3072, 1, 1, 1 | F32 | blk.3.attn_norm.weight
61: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.3.ffn_down_exps.weight
62: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_gate_exps.weight
63: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.3.ffn_up_exps.weight
64: 786432 | 3072, 256, 1, 1 | F32 | blk.3.ffn_gate_inp.weight
65: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.3.ffn_down_shexp.weight
66: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_gate_shexp.weight
67: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.3.ffn_up_shexp.weight
68: 3072 | 3072, 1, 1, 1 | F32 | blk.3.ffn_gate_inp_shexp.weight
69: 3072 | 3072, 1, 1, 1 | F32 | blk.3.post_attention_norm.weight
70: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_k_norm.weight
71: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_k.weight
72: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.3.attn_output.weight
73: 256 | 256, 1, 1, 1 | F32 | blk.3.attn_q_norm.weight
74: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.3.attn_q.weight
75: 1572864 | 3072, 512, 1, 1 | BF16 | blk.3.attn_v.weight
76: 3072 | 3072, 1, 1, 1 | F32 | blk.4.attn_norm.weight
77: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_a
78: 49152 | 4, 12288, 1, 1 | F32 | blk.4.ssm_conv1d.weight
79: 64 | 64, 1, 1, 1 | F32 | blk.4.ssm_dt.bias
80: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_alpha.weight
81: 196608 | 3072, 64, 1, 1 | BF16 | blk.4.ssm_beta.weight
82: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.4.attn_qkv.weight
83: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.4.attn_gate.weight
84: 128 | 128, 1, 1, 1 | F32 | blk.4.ssm_norm.weight
85: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.4.ssm_out.weight
86: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.4.ffn_down_exps.weight
87: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_gate_exps.weight
88: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.4.ffn_up_exps.weight
89: 786432 | 3072, 256, 1, 1 | F32 | blk.4.ffn_gate_inp.weight
90: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.4.ffn_down_shexp.weight
91: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_gate_shexp.weight
92: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.4.ffn_up_shexp.weight
93: 3072 | 3072, 1, 1, 1 | F32 | blk.4.ffn_gate_inp_shexp.weight
94: 3072 | 3072, 1, 1, 1 | F32 | blk.4.post_attention_norm.weight
95: 3072 | 3072, 1, 1, 1 | F32 | blk.5.attn_norm.weight
96: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_a
97: 49152 | 4, 12288, 1, 1 | F32 | blk.5.ssm_conv1d.weight
98: 64 | 64, 1, 1, 1 | F32 | blk.5.ssm_dt.bias
99: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_alpha.weight
100: 196608 | 3072, 64, 1, 1 | BF16 | blk.5.ssm_beta.weight
101: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.5.attn_qkv.weight
102: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.5.attn_gate.weight
103: 128 | 128, 1, 1, 1 | F32 | blk.5.ssm_norm.weight
104: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.5.ssm_out.weight
105: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.5.ffn_down_exps.weight
106: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_gate_exps.weight
107: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.5.ffn_up_exps.weight
108: 786432 | 3072, 256, 1, 1 | F32 | blk.5.ffn_gate_inp.weight
109: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.5.ffn_down_shexp.weight
110: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_gate_shexp.weight
111: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.5.ffn_up_shexp.weight
112: 3072 | 3072, 1, 1, 1 | F32 | blk.5.ffn_gate_inp_shexp.weight
113: 3072 | 3072, 1, 1, 1 | F32 | blk.5.post_attention_norm.weight
114: 3072 | 3072, 1, 1, 1 | F32 | blk.6.attn_norm.weight
115: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_a
116: 49152 | 4, 12288, 1, 1 | F32 | blk.6.ssm_conv1d.weight
117: 64 | 64, 1, 1, 1 | F32 | blk.6.ssm_dt.bias
118: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_alpha.weight
119: 196608 | 3072, 64, 1, 1 | BF16 | blk.6.ssm_beta.weight
120: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.6.attn_qkv.weight
121: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.6.attn_gate.weight
122: 128 | 128, 1, 1, 1 | F32 | blk.6.ssm_norm.weight
123: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.6.ssm_out.weight
124: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.6.ffn_down_exps.weight
125: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_gate_exps.weight
126: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.6.ffn_up_exps.weight
127: 786432 | 3072, 256, 1, 1 | F32 | blk.6.ffn_gate_inp.weight
128: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.6.ffn_down_shexp.weight
129: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_gate_shexp.weight
130: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.6.ffn_up_shexp.weight
131: 3072 | 3072, 1, 1, 1 | F32 | blk.6.ffn_gate_inp_shexp.weight
132: 3072 | 3072, 1, 1, 1 | F32 | blk.6.post_attention_norm.weight
133: 3072 | 3072, 1, 1, 1 | F32 | blk.7.attn_norm.weight
134: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.7.ffn_down_exps.weight
135: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_gate_exps.weight
136: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.7.ffn_up_exps.weight
137: 786432 | 3072, 256, 1, 1 | F32 | blk.7.ffn_gate_inp.weight
138: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.7.ffn_down_shexp.weight
139: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_gate_shexp.weight
140: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.7.ffn_up_shexp.weight
141: 3072 | 3072, 1, 1, 1 | F32 | blk.7.ffn_gate_inp_shexp.weight
142: 3072 | 3072, 1, 1, 1 | F32 | blk.7.post_attention_norm.weight
143: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_k_norm.weight
144: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_k.weight
145: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.7.attn_output.weight
146: 256 | 256, 1, 1, 1 | F32 | blk.7.attn_q_norm.weight
147: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.7.attn_q.weight
148: 1572864 | 3072, 512, 1, 1 | BF16 | blk.7.attn_v.weight
149: 3072 | 3072, 1, 1, 1 | F32 | blk.8.attn_norm.weight
150: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_a
151: 49152 | 4, 12288, 1, 1 | F32 | blk.8.ssm_conv1d.weight
152: 64 | 64, 1, 1, 1 | F32 | blk.8.ssm_dt.bias
153: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_alpha.weight
154: 196608 | 3072, 64, 1, 1 | BF16 | blk.8.ssm_beta.weight
155: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.8.attn_qkv.weight
156: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.8.attn_gate.weight
157: 128 | 128, 1, 1, 1 | F32 | blk.8.ssm_norm.weight
158: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.8.ssm_out.weight
159: 3072 | 3072, 1, 1, 1 | F32 | blk.10.attn_norm.weight
160: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_a
161: 49152 | 4, 12288, 1, 1 | F32 | blk.10.ssm_conv1d.weight
162: 64 | 64, 1, 1, 1 | F32 | blk.10.ssm_dt.bias
163: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_alpha.weight
164: 196608 | 3072, 64, 1, 1 | BF16 | blk.10.ssm_beta.weight
165: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.10.attn_qkv.weight
166: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.10.attn_gate.weight
167: 128 | 128, 1, 1, 1 | F32 | blk.10.ssm_norm.weight
168: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.10.ssm_out.weight
169: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.8.ffn_down_exps.weight
170: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_gate_exps.weight
171: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.8.ffn_up_exps.weight
172: 786432 | 3072, 256, 1, 1 | F32 | blk.8.ffn_gate_inp.weight
173: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.8.ffn_down_shexp.weight
174: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_gate_shexp.weight
175: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.8.ffn_up_shexp.weight
176: 3072 | 3072, 1, 1, 1 | F32 | blk.8.ffn_gate_inp_shexp.weight
177: 3072 | 3072, 1, 1, 1 | F32 | blk.8.post_attention_norm.weight
178: 3072 | 3072, 1, 1, 1 | F32 | blk.9.attn_norm.weight
179: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_a
180: 49152 | 4, 12288, 1, 1 | F32 | blk.9.ssm_conv1d.weight
181: 64 | 64, 1, 1, 1 | F32 | blk.9.ssm_dt.bias
182: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_alpha.weight
183: 196608 | 3072, 64, 1, 1 | BF16 | blk.9.ssm_beta.weight
184: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.9.attn_qkv.weight
185: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.9.attn_gate.weight
186: 128 | 128, 1, 1, 1 | F32 | blk.9.ssm_norm.weight
187: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.9.ssm_out.weight
188: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.9.ffn_down_exps.weight
189: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_gate_exps.weight
190: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.9.ffn_up_exps.weight
191: 786432 | 3072, 256, 1, 1 | F32 | blk.9.ffn_gate_inp.weight
192: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.9.ffn_down_shexp.weight
193: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_gate_shexp.weight
194: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.9.ffn_up_shexp.weight
195: 3072 | 3072, 1, 1, 1 | F32 | blk.9.ffn_gate_inp_shexp.weight
196: 3072 | 3072, 1, 1, 1 | F32 | blk.9.post_attention_norm.weight
197: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.10.ffn_down_exps.weight
198: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_gate_exps.weight
199: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.10.ffn_up_exps.weight
200: 786432 | 3072, 256, 1, 1 | F32 | blk.10.ffn_gate_inp.weight
201: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.10.ffn_down_shexp.weight
202: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_gate_shexp.weight
203: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.10.ffn_up_shexp.weight
204: 3072 | 3072, 1, 1, 1 | F32 | blk.10.ffn_gate_inp_shexp.weight
205: 3072 | 3072, 1, 1, 1 | F32 | blk.10.post_attention_norm.weight
206: 3072 | 3072, 1, 1, 1 | F32 | blk.11.attn_norm.weight
207: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.11.ffn_down_exps.weight
208: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_gate_exps.weight
209: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.11.ffn_up_exps.weight
210: 786432 | 3072, 256, 1, 1 | F32 | blk.11.ffn_gate_inp.weight
211: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.11.ffn_down_shexp.weight
212: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_gate_shexp.weight
213: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.11.ffn_up_shexp.weight
214: 3072 | 3072, 1, 1, 1 | F32 | blk.11.ffn_gate_inp_shexp.weight
215: 3072 | 3072, 1, 1, 1 | F32 | blk.11.post_attention_norm.weight
216: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_k_norm.weight
217: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_k.weight
218: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.11.attn_output.weight
219: 256 | 256, 1, 1, 1 | F32 | blk.11.attn_q_norm.weight
220: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.11.attn_q.weight
221: 1572864 | 3072, 512, 1, 1 | BF16 | blk.11.attn_v.weight
222: 3072 | 3072, 1, 1, 1 | F32 | blk.12.attn_norm.weight
223: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_a
224: 49152 | 4, 12288, 1, 1 | F32 | blk.12.ssm_conv1d.weight
225: 64 | 64, 1, 1, 1 | F32 | blk.12.ssm_dt.bias
226: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_alpha.weight
227: 196608 | 3072, 64, 1, 1 | BF16 | blk.12.ssm_beta.weight
228: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.12.attn_qkv.weight
229: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.12.attn_gate.weight
230: 128 | 128, 1, 1, 1 | F32 | blk.12.ssm_norm.weight
231: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.12.ssm_out.weight
232: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.12.ffn_down_exps.weight
233: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_gate_exps.weight
234: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.12.ffn_up_exps.weight
235: 786432 | 3072, 256, 1, 1 | F32 | blk.12.ffn_gate_inp.weight
236: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.12.ffn_down_shexp.weight
237: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_gate_shexp.weight
238: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.12.ffn_up_shexp.weight
239: 3072 | 3072, 1, 1, 1 | F32 | blk.12.ffn_gate_inp_shexp.weight
240: 3072 | 3072, 1, 1, 1 | F32 | blk.12.post_attention_norm.weight
241: 3072 | 3072, 1, 1, 1 | F32 | blk.13.attn_norm.weight
242: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_a
243: 49152 | 4, 12288, 1, 1 | F32 | blk.13.ssm_conv1d.weight
244: 64 | 64, 1, 1, 1 | F32 | blk.13.ssm_dt.bias
245: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_alpha.weight
246: 196608 | 3072, 64, 1, 1 | BF16 | blk.13.ssm_beta.weight
247: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.13.attn_qkv.weight
248: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.13.attn_gate.weight
249: 128 | 128, 1, 1, 1 | F32 | blk.13.ssm_norm.weight
250: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.13.ssm_out.weight
251: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.13.ffn_down_exps.weight
252: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_gate_exps.weight
253: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.13.ffn_up_exps.weight
254: 786432 | 3072, 256, 1, 1 | F32 | blk.13.ffn_gate_inp.weight
255: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.13.ffn_down_shexp.weight
256: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_gate_shexp.weight
257: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.13.ffn_up_shexp.weight
258: 3072 | 3072, 1, 1, 1 | F32 | blk.13.ffn_gate_inp_shexp.weight
259: 3072 | 3072, 1, 1, 1 | F32 | blk.13.post_attention_norm.weight
260: 3072 | 3072, 1, 1, 1 | F32 | blk.14.attn_norm.weight
261: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_a
262: 49152 | 4, 12288, 1, 1 | F32 | blk.14.ssm_conv1d.weight
263: 64 | 64, 1, 1, 1 | F32 | blk.14.ssm_dt.bias
264: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_alpha.weight
265: 196608 | 3072, 64, 1, 1 | BF16 | blk.14.ssm_beta.weight
266: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.14.attn_qkv.weight
267: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.14.attn_gate.weight
268: 128 | 128, 1, 1, 1 | F32 | blk.14.ssm_norm.weight
269: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.14.ssm_out.weight
270: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.14.ffn_down_exps.weight
271: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_gate_exps.weight
272: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.14.ffn_up_exps.weight
273: 786432 | 3072, 256, 1, 1 | F32 | blk.14.ffn_gate_inp.weight
274: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.14.ffn_down_shexp.weight
275: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_gate_shexp.weight
276: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.14.ffn_up_shexp.weight
277: 3072 | 3072, 1, 1, 1 | F32 | blk.14.ffn_gate_inp_shexp.weight
278: 3072 | 3072, 1, 1, 1 | F32 | blk.14.post_attention_norm.weight
279: 3072 | 3072, 1, 1, 1 | F32 | blk.15.attn_norm.weight
280: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.15.ffn_down_exps.weight
281: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_gate_exps.weight
282: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.15.ffn_up_exps.weight
283: 786432 | 3072, 256, 1, 1 | F32 | blk.15.ffn_gate_inp.weight
284: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.15.ffn_down_shexp.weight
285: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_gate_shexp.weight
286: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.15.ffn_up_shexp.weight
287: 3072 | 3072, 1, 1, 1 | F32 | blk.15.ffn_gate_inp_shexp.weight
288: 3072 | 3072, 1, 1, 1 | F32 | blk.15.post_attention_norm.weight
289: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_k_norm.weight
290: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_k.weight
291: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.15.attn_output.weight
292: 256 | 256, 1, 1, 1 | F32 | blk.15.attn_q_norm.weight
293: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.15.attn_q.weight
294: 1572864 | 3072, 512, 1, 1 | BF16 | blk.15.attn_v.weight
295: 3072 | 3072, 1, 1, 1 | F32 | blk.16.attn_norm.weight
296: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_a
297: 49152 | 4, 12288, 1, 1 | F32 | blk.16.ssm_conv1d.weight
298: 64 | 64, 1, 1, 1 | F32 | blk.16.ssm_dt.bias
299: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_alpha.weight
300: 196608 | 3072, 64, 1, 1 | BF16 | blk.16.ssm_beta.weight
301: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.16.attn_qkv.weight
302: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.16.attn_gate.weight
303: 128 | 128, 1, 1, 1 | F32 | blk.16.ssm_norm.weight
304: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.16.ssm_out.weight
305: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.16.ffn_down_exps.weight
306: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_gate_exps.weight
307: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.16.ffn_up_exps.weight
308: 786432 | 3072, 256, 1, 1 | F32 | blk.16.ffn_gate_inp.weight
309: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.16.ffn_down_shexp.weight
310: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_gate_shexp.weight
311: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.16.ffn_up_shexp.weight
312: 3072 | 3072, 1, 1, 1 | F32 | blk.16.ffn_gate_inp_shexp.weight
313: 3072 | 3072, 1, 1, 1 | F32 | blk.16.post_attention_norm.weight
314: 3072 | 3072, 1, 1, 1 | F32 | blk.17.attn_norm.weight
315: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_a
316: 49152 | 4, 12288, 1, 1 | F32 | blk.17.ssm_conv1d.weight
317: 64 | 64, 1, 1, 1 | F32 | blk.17.ssm_dt.bias
318: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_alpha.weight
319: 196608 | 3072, 64, 1, 1 | BF16 | blk.17.ssm_beta.weight
320: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.17.attn_qkv.weight
321: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.17.attn_gate.weight
322: 128 | 128, 1, 1, 1 | F32 | blk.17.ssm_norm.weight
323: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.17.ssm_out.weight
324: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.17.ffn_down_exps.weight
325: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_gate_exps.weight
326: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.17.ffn_up_exps.weight
327: 786432 | 3072, 256, 1, 1 | F32 | blk.17.ffn_gate_inp.weight
328: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.17.ffn_down_shexp.weight
329: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_gate_shexp.weight
330: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.17.ffn_up_shexp.weight
331: 3072 | 3072, 1, 1, 1 | F32 | blk.17.ffn_gate_inp_shexp.weight
332: 3072 | 3072, 1, 1, 1 | F32 | blk.17.post_attention_norm.weight
333: 3072 | 3072, 1, 1, 1 | F32 | blk.18.attn_norm.weight
334: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_a
335: 49152 | 4, 12288, 1, 1 | F32 | blk.18.ssm_conv1d.weight
336: 64 | 64, 1, 1, 1 | F32 | blk.18.ssm_dt.bias
337: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_alpha.weight
338: 196608 | 3072, 64, 1, 1 | BF16 | blk.18.ssm_beta.weight
339: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.18.attn_qkv.weight
340: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.18.attn_gate.weight
341: 128 | 128, 1, 1, 1 | F32 | blk.18.ssm_norm.weight
342: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.18.ssm_out.weight
343: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.18.ffn_down_exps.weight
344: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_gate_exps.weight
345: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.18.ffn_up_exps.weight
346: 786432 | 3072, 256, 1, 1 | F32 | blk.18.ffn_gate_inp.weight
347: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.18.ffn_down_shexp.weight
348: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_gate_shexp.weight
349: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.18.ffn_up_shexp.weight
350: 3072 | 3072, 1, 1, 1 | F32 | blk.18.ffn_gate_inp_shexp.weight
351: 3072 | 3072, 1, 1, 1 | F32 | blk.18.post_attention_norm.weight
352: 3072 | 3072, 1, 1, 1 | F32 | blk.19.attn_norm.weight
353: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.19.ffn_down_exps.weight
354: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_gate_exps.weight
355: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.19.ffn_up_exps.weight
356: 786432 | 3072, 256, 1, 1 | F32 | blk.19.ffn_gate_inp.weight
357: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.19.ffn_down_shexp.weight
358: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_gate_shexp.weight
359: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.19.ffn_up_shexp.weight
360: 3072 | 3072, 1, 1, 1 | F32 | blk.19.ffn_gate_inp_shexp.weight
361: 3072 | 3072, 1, 1, 1 | F32 | blk.19.post_attention_norm.weight
362: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_k_norm.weight
363: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_k.weight
364: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.19.attn_output.weight
365: 256 | 256, 1, 1, 1 | F32 | blk.19.attn_q_norm.weight
366: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.19.attn_q.weight
367: 1572864 | 3072, 512, 1, 1 | BF16 | blk.19.attn_v.weight
368: 3072 | 3072, 1, 1, 1 | F32 | blk.20.attn_norm.weight
369: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_a
370: 49152 | 4, 12288, 1, 1 | F32 | blk.20.ssm_conv1d.weight
371: 64 | 64, 1, 1, 1 | F32 | blk.20.ssm_dt.bias
372: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_alpha.weight
373: 196608 | 3072, 64, 1, 1 | BF16 | blk.20.ssm_beta.weight
374: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.20.attn_qkv.weight
375: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.20.attn_gate.weight
376: 128 | 128, 1, 1, 1 | F32 | blk.20.ssm_norm.weight
377: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.20.ssm_out.weight
378: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.20.ffn_down_exps.weight
379: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_gate_exps.weight
380: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.20.ffn_up_exps.weight
381: 786432 | 3072, 256, 1, 1 | F32 | blk.20.ffn_gate_inp.weight
382: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.20.ffn_down_shexp.weight
383: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_gate_shexp.weight
384: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.20.ffn_up_shexp.weight
385: 3072 | 3072, 1, 1, 1 | F32 | blk.20.ffn_gate_inp_shexp.weight
386: 3072 | 3072, 1, 1, 1 | F32 | blk.20.post_attention_norm.weight
387: 3072 | 3072, 1, 1, 1 | F32 | blk.21.attn_norm.weight
388: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_a
389: 49152 | 4, 12288, 1, 1 | F32 | blk.21.ssm_conv1d.weight
390: 64 | 64, 1, 1, 1 | F32 | blk.21.ssm_dt.bias
391: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_alpha.weight
392: 196608 | 3072, 64, 1, 1 | BF16 | blk.21.ssm_beta.weight
393: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.21.attn_qkv.weight
394: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.21.attn_gate.weight
395: 128 | 128, 1, 1, 1 | F32 | blk.21.ssm_norm.weight
396: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.21.ssm_out.weight
397: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.21.ffn_down_exps.weight
398: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_gate_exps.weight
399: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.21.ffn_up_exps.weight
400: 786432 | 3072, 256, 1, 1 | F32 | blk.21.ffn_gate_inp.weight
401: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.21.ffn_down_shexp.weight
402: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_gate_shexp.weight
403: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.21.ffn_up_shexp.weight
404: 3072 | 3072, 1, 1, 1 | F32 | blk.21.ffn_gate_inp_shexp.weight
405: 3072 | 3072, 1, 1, 1 | F32 | blk.21.post_attention_norm.weight
406: 3072 | 3072, 1, 1, 1 | F32 | blk.22.attn_norm.weight
407: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_a
408: 49152 | 4, 12288, 1, 1 | F32 | blk.22.ssm_conv1d.weight
409: 64 | 64, 1, 1, 1 | F32 | blk.22.ssm_dt.bias
410: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_alpha.weight
411: 196608 | 3072, 64, 1, 1 | BF16 | blk.22.ssm_beta.weight
412: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.22.attn_qkv.weight
413: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.22.attn_gate.weight
414: 128 | 128, 1, 1, 1 | F32 | blk.22.ssm_norm.weight
415: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.22.ssm_out.weight
416: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.22.ffn_down_exps.weight
417: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_gate_exps.weight
418: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.22.ffn_up_exps.weight
419: 786432 | 3072, 256, 1, 1 | F32 | blk.22.ffn_gate_inp.weight
420: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.22.ffn_down_shexp.weight
421: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_gate_shexp.weight
422: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.22.ffn_up_shexp.weight
423: 3072 | 3072, 1, 1, 1 | F32 | blk.22.ffn_gate_inp_shexp.weight
424: 3072 | 3072, 1, 1, 1 | F32 | blk.22.post_attention_norm.weight
425: 3072 | 3072, 1, 1, 1 | F32 | blk.23.attn_norm.weight
426: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.23.ffn_down_exps.weight
427: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_gate_exps.weight
428: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.23.ffn_up_exps.weight
429: 786432 | 3072, 256, 1, 1 | F32 | blk.23.ffn_gate_inp.weight
430: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.23.ffn_down_shexp.weight
431: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_gate_shexp.weight
432: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.23.ffn_up_shexp.weight
433: 3072 | 3072, 1, 1, 1 | F32 | blk.23.ffn_gate_inp_shexp.weight
434: 3072 | 3072, 1, 1, 1 | F32 | blk.23.post_attention_norm.weight
435: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_k_norm.weight
436: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_k.weight
437: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.23.attn_output.weight
438: 256 | 256, 1, 1, 1 | F32 | blk.23.attn_q_norm.weight
439: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.23.attn_q.weight
440: 1572864 | 3072, 512, 1, 1 | BF16 | blk.23.attn_v.weight
441: 3072 | 3072, 1, 1, 1 | F32 | blk.24.attn_norm.weight
442: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_a
443: 49152 | 4, 12288, 1, 1 | F32 | blk.24.ssm_conv1d.weight
444: 64 | 64, 1, 1, 1 | F32 | blk.24.ssm_dt.bias
445: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_alpha.weight
446: 196608 | 3072, 64, 1, 1 | BF16 | blk.24.ssm_beta.weight
447: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.24.attn_qkv.weight
448: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.24.attn_gate.weight
449: 128 | 128, 1, 1, 1 | F32 | blk.24.ssm_norm.weight
450: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.24.ssm_out.weight
451: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.24.ffn_down_exps.weight
452: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_gate_exps.weight
453: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.24.ffn_up_exps.weight
454: 786432 | 3072, 256, 1, 1 | F32 | blk.24.ffn_gate_inp.weight
455: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.24.ffn_down_shexp.weight
456: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_gate_shexp.weight
457: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.24.ffn_up_shexp.weight
458: 3072 | 3072, 1, 1, 1 | F32 | blk.24.ffn_gate_inp_shexp.weight
459: 3072 | 3072, 1, 1, 1 | F32 | blk.24.post_attention_norm.weight
460: 3072 | 3072, 1, 1, 1 | F32 | blk.25.attn_norm.weight
461: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_a
462: 49152 | 4, 12288, 1, 1 | F32 | blk.25.ssm_conv1d.weight
463: 64 | 64, 1, 1, 1 | F32 | blk.25.ssm_dt.bias
464: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_alpha.weight
465: 196608 | 3072, 64, 1, 1 | BF16 | blk.25.ssm_beta.weight
466: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.25.attn_qkv.weight
467: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.25.attn_gate.weight
468: 128 | 128, 1, 1, 1 | F32 | blk.25.ssm_norm.weight
469: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.25.ssm_out.weight
470: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.25.ffn_down_exps.weight
471: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_gate_exps.weight
472: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.25.ffn_up_exps.weight
473: 786432 | 3072, 256, 1, 1 | F32 | blk.25.ffn_gate_inp.weight
474: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.25.ffn_down_shexp.weight
475: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_gate_shexp.weight
476: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.25.ffn_up_shexp.weight
477: 3072 | 3072, 1, 1, 1 | F32 | blk.25.ffn_gate_inp_shexp.weight
478: 3072 | 3072, 1, 1, 1 | F32 | blk.25.post_attention_norm.weight
479: 3072 | 3072, 1, 1, 1 | F32 | blk.26.attn_norm.weight
480: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_a
481: 49152 | 4, 12288, 1, 1 | F32 | blk.26.ssm_conv1d.weight
482: 64 | 64, 1, 1, 1 | F32 | blk.26.ssm_dt.bias
483: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_alpha.weight
484: 196608 | 3072, 64, 1, 1 | BF16 | blk.26.ssm_beta.weight
485: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.26.attn_qkv.weight
486: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.26.attn_gate.weight
487: 128 | 128, 1, 1, 1 | F32 | blk.26.ssm_norm.weight
488: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.26.ssm_out.weight
489: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.26.ffn_down_exps.weight
490: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_gate_exps.weight
491: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.26.ffn_up_exps.weight
492: 786432 | 3072, 256, 1, 1 | F32 | blk.26.ffn_gate_inp.weight
493: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.26.ffn_down_shexp.weight
494: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_gate_shexp.weight
495: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.26.ffn_up_shexp.weight
496: 3072 | 3072, 1, 1, 1 | F32 | blk.26.ffn_gate_inp_shexp.weight
497: 3072 | 3072, 1, 1, 1 | F32 | blk.26.post_attention_norm.weight
498: 3072 | 3072, 1, 1, 1 | F32 | blk.27.attn_norm.weight
499: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.27.ffn_down_exps.weight
500: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_gate_exps.weight
501: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.27.ffn_up_exps.weight
502: 786432 | 3072, 256, 1, 1 | F32 | blk.27.ffn_gate_inp.weight
503: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.27.ffn_down_shexp.weight
504: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_gate_shexp.weight
505: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.27.ffn_up_shexp.weight
506: 3072 | 3072, 1, 1, 1 | F32 | blk.27.ffn_gate_inp_shexp.weight
507: 3072 | 3072, 1, 1, 1 | F32 | blk.27.post_attention_norm.weight
508: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_k_norm.weight
509: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_k.weight
510: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.27.attn_output.weight
511: 256 | 256, 1, 1, 1 | F32 | blk.27.attn_q_norm.weight
512: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.27.attn_q.weight
513: 1572864 | 3072, 512, 1, 1 | BF16 | blk.27.attn_v.weight
514: 3072 | 3072, 1, 1, 1 | F32 | blk.28.attn_norm.weight
515: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_a
516: 49152 | 4, 12288, 1, 1 | F32 | blk.28.ssm_conv1d.weight
517: 64 | 64, 1, 1, 1 | F32 | blk.28.ssm_dt.bias
518: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_alpha.weight
519: 196608 | 3072, 64, 1, 1 | BF16 | blk.28.ssm_beta.weight
520: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.28.attn_qkv.weight
521: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.28.attn_gate.weight
522: 128 | 128, 1, 1, 1 | F32 | blk.28.ssm_norm.weight
523: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.28.ssm_out.weight
524: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.28.ffn_down_exps.weight
525: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_gate_exps.weight
526: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.28.ffn_up_exps.weight
527: 786432 | 3072, 256, 1, 1 | F32 | blk.28.ffn_gate_inp.weight
528: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.28.ffn_down_shexp.weight
529: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_gate_shexp.weight
530: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.28.ffn_up_shexp.weight
531: 3072 | 3072, 1, 1, 1 | F32 | blk.28.ffn_gate_inp_shexp.weight
532: 3072 | 3072, 1, 1, 1 | F32 | blk.28.post_attention_norm.weight
533: 3072 | 3072, 1, 1, 1 | F32 | blk.29.attn_norm.weight
534: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_a
535: 49152 | 4, 12288, 1, 1 | F32 | blk.29.ssm_conv1d.weight
536: 64 | 64, 1, 1, 1 | F32 | blk.29.ssm_dt.bias
537: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_alpha.weight
538: 196608 | 3072, 64, 1, 1 | BF16 | blk.29.ssm_beta.weight
539: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.29.attn_qkv.weight
540: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.29.attn_gate.weight
541: 128 | 128, 1, 1, 1 | F32 | blk.29.ssm_norm.weight
542: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.29.ssm_out.weight
543: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.29.ffn_down_exps.weight
544: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_gate_exps.weight
545: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.29.ffn_up_exps.weight
546: 786432 | 3072, 256, 1, 1 | F32 | blk.29.ffn_gate_inp.weight
547: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.29.ffn_down_shexp.weight
548: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_gate_shexp.weight
549: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.29.ffn_up_shexp.weight
550: 3072 | 3072, 1, 1, 1 | F32 | blk.29.ffn_gate_inp_shexp.weight
551: 3072 | 3072, 1, 1, 1 | F32 | blk.29.post_attention_norm.weight
552: 3072 | 3072, 1, 1, 1 | F32 | blk.30.attn_norm.weight
553: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_a
554: 49152 | 4, 12288, 1, 1 | F32 | blk.30.ssm_conv1d.weight
555: 64 | 64, 1, 1, 1 | F32 | blk.30.ssm_dt.bias
556: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_alpha.weight
557: 196608 | 3072, 64, 1, 1 | BF16 | blk.30.ssm_beta.weight
558: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.30.attn_qkv.weight
559: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.30.attn_gate.weight
560: 128 | 128, 1, 1, 1 | F32 | blk.30.ssm_norm.weight
561: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.30.ssm_out.weight
562: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.30.ffn_down_exps.weight
563: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_gate_exps.weight
564: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.30.ffn_up_exps.weight
565: 786432 | 3072, 256, 1, 1 | F32 | blk.30.ffn_gate_inp.weight
566: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.30.ffn_down_shexp.weight
567: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_gate_shexp.weight
568: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.30.ffn_up_shexp.weight
569: 3072 | 3072, 1, 1, 1 | F32 | blk.30.ffn_gate_inp_shexp.weight
570: 3072 | 3072, 1, 1, 1 | F32 | blk.30.post_attention_norm.weight
571: 3072 | 3072, 1, 1, 1 | F32 | blk.31.attn_norm.weight
572: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.31.ffn_down_exps.weight
573: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_gate_exps.weight
574: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.31.ffn_up_exps.weight
575: 786432 | 3072, 256, 1, 1 | F32 | blk.31.ffn_gate_inp.weight
576: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.31.ffn_down_shexp.weight
577: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_gate_shexp.weight
578: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.31.ffn_up_shexp.weight
579: 3072 | 3072, 1, 1, 1 | F32 | blk.31.ffn_gate_inp_shexp.weight
580: 3072 | 3072, 1, 1, 1 | F32 | blk.31.post_attention_norm.weight
581: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_k_norm.weight
582: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_k.weight
583: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.31.attn_output.weight
584: 256 | 256, 1, 1, 1 | F32 | blk.31.attn_q_norm.weight
585: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.31.attn_q.weight
586: 1572864 | 3072, 512, 1, 1 | BF16 | blk.31.attn_v.weight
587: 3072 | 3072, 1, 1, 1 | F32 | blk.32.attn_norm.weight
588: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_a
589: 49152 | 4, 12288, 1, 1 | F32 | blk.32.ssm_conv1d.weight
590: 64 | 64, 1, 1, 1 | F32 | blk.32.ssm_dt.bias
591: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_alpha.weight
592: 196608 | 3072, 64, 1, 1 | BF16 | blk.32.ssm_beta.weight
593: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.32.attn_qkv.weight
594: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.32.attn_gate.weight
595: 128 | 128, 1, 1, 1 | F32 | blk.32.ssm_norm.weight
596: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.32.ssm_out.weight
597: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.32.ffn_down_exps.weight
598: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_gate_exps.weight
599: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.32.ffn_up_exps.weight
600: 786432 | 3072, 256, 1, 1 | F32 | blk.32.ffn_gate_inp.weight
601: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.32.ffn_down_shexp.weight
602: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_gate_shexp.weight
603: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.32.ffn_up_shexp.weight
604: 3072 | 3072, 1, 1, 1 | F32 | blk.32.ffn_gate_inp_shexp.weight
605: 3072 | 3072, 1, 1, 1 | F32 | blk.32.post_attention_norm.weight
606: 3072 | 3072, 1, 1, 1 | F32 | blk.33.attn_norm.weight
607: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_a
608: 49152 | 4, 12288, 1, 1 | F32 | blk.33.ssm_conv1d.weight
609: 64 | 64, 1, 1, 1 | F32 | blk.33.ssm_dt.bias
610: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_alpha.weight
611: 196608 | 3072, 64, 1, 1 | BF16 | blk.33.ssm_beta.weight
612: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.33.attn_qkv.weight
613: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.33.attn_gate.weight
614: 128 | 128, 1, 1, 1 | F32 | blk.33.ssm_norm.weight
615: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.33.ssm_out.weight
616: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.33.ffn_down_exps.weight
617: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_gate_exps.weight
618: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.33.ffn_up_exps.weight
619: 786432 | 3072, 256, 1, 1 | F32 | blk.33.ffn_gate_inp.weight
620: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.33.ffn_down_shexp.weight
621: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_gate_shexp.weight
622: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.33.ffn_up_shexp.weight
623: 3072 | 3072, 1, 1, 1 | F32 | blk.33.ffn_gate_inp_shexp.weight
624: 3072 | 3072, 1, 1, 1 | F32 | blk.33.post_attention_norm.weight
625: 3072 | 3072, 1, 1, 1 | F32 | blk.34.attn_norm.weight
626: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_a
627: 49152 | 4, 12288, 1, 1 | F32 | blk.34.ssm_conv1d.weight
628: 64 | 64, 1, 1, 1 | F32 | blk.34.ssm_dt.bias
629: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_alpha.weight
630: 196608 | 3072, 64, 1, 1 | BF16 | blk.34.ssm_beta.weight
631: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.34.attn_qkv.weight
632: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.34.attn_gate.weight
633: 128 | 128, 1, 1, 1 | F32 | blk.34.ssm_norm.weight
634: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.34.ssm_out.weight
635: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.34.ffn_down_exps.weight
636: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_gate_exps.weight
637: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.34.ffn_up_exps.weight
638: 786432 | 3072, 256, 1, 1 | F32 | blk.34.ffn_gate_inp.weight
639: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.34.ffn_down_shexp.weight
640: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_gate_shexp.weight
641: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.34.ffn_up_shexp.weight
642: 3072 | 3072, 1, 1, 1 | F32 | blk.34.ffn_gate_inp_shexp.weight
643: 3072 | 3072, 1, 1, 1 | F32 | blk.34.post_attention_norm.weight
644: 3072 | 3072, 1, 1, 1 | F32 | blk.35.attn_norm.weight
645: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.35.ffn_down_exps.weight
646: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_gate_exps.weight
647: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.35.ffn_up_exps.weight
648: 786432 | 3072, 256, 1, 1 | F32 | blk.35.ffn_gate_inp.weight
649: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.35.ffn_down_shexp.weight
650: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_gate_shexp.weight
651: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.35.ffn_up_shexp.weight
652: 3072 | 3072, 1, 1, 1 | F32 | blk.35.ffn_gate_inp_shexp.weight
653: 3072 | 3072, 1, 1, 1 | F32 | blk.35.post_attention_norm.weight
654: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_k_norm.weight
655: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_k.weight
656: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.35.attn_output.weight
657: 256 | 256, 1, 1, 1 | F32 | blk.35.attn_q_norm.weight
658: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.35.attn_q.weight
659: 1572864 | 3072, 512, 1, 1 | BF16 | blk.35.attn_v.weight
660: 3072 | 3072, 1, 1, 1 | F32 | blk.36.attn_norm.weight
661: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_a
662: 49152 | 4, 12288, 1, 1 | F32 | blk.36.ssm_conv1d.weight
663: 64 | 64, 1, 1, 1 | F32 | blk.36.ssm_dt.bias
664: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_alpha.weight
665: 196608 | 3072, 64, 1, 1 | BF16 | blk.36.ssm_beta.weight
666: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.36.attn_qkv.weight
667: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.36.attn_gate.weight
668: 128 | 128, 1, 1, 1 | F32 | blk.36.ssm_norm.weight
669: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.36.ssm_out.weight
670: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.36.ffn_down_exps.weight
671: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_gate_exps.weight
672: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.36.ffn_up_exps.weight
673: 786432 | 3072, 256, 1, 1 | F32 | blk.36.ffn_gate_inp.weight
674: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.36.ffn_down_shexp.weight
675: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_gate_shexp.weight
676: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.36.ffn_up_shexp.weight
677: 3072 | 3072, 1, 1, 1 | F32 | blk.36.ffn_gate_inp_shexp.weight
678: 3072 | 3072, 1, 1, 1 | F32 | blk.36.post_attention_norm.weight
679: 3072 | 3072, 1, 1, 1 | F32 | blk.37.attn_norm.weight
680: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_a
681: 49152 | 4, 12288, 1, 1 | F32 | blk.37.ssm_conv1d.weight
682: 64 | 64, 1, 1, 1 | F32 | blk.37.ssm_dt.bias
683: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_alpha.weight
684: 196608 | 3072, 64, 1, 1 | BF16 | blk.37.ssm_beta.weight
685: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.37.attn_qkv.weight
686: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.37.attn_gate.weight
687: 128 | 128, 1, 1, 1 | F32 | blk.37.ssm_norm.weight
688: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.37.ssm_out.weight
689: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.37.ffn_down_exps.weight
690: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_gate_exps.weight
691: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.37.ffn_up_exps.weight
692: 786432 | 3072, 256, 1, 1 | F32 | blk.37.ffn_gate_inp.weight
693: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.37.ffn_down_shexp.weight
694: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_gate_shexp.weight
695: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.37.ffn_up_shexp.weight
696: 3072 | 3072, 1, 1, 1 | F32 | blk.37.ffn_gate_inp_shexp.weight
697: 3072 | 3072, 1, 1, 1 | F32 | blk.37.post_attention_norm.weight
698: 3072 | 3072, 1, 1, 1 | F32 | blk.38.attn_norm.weight
699: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_a
700: 49152 | 4, 12288, 1, 1 | F32 | blk.38.ssm_conv1d.weight
701: 64 | 64, 1, 1, 1 | F32 | blk.38.ssm_dt.bias
702: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_alpha.weight
703: 196608 | 3072, 64, 1, 1 | BF16 | blk.38.ssm_beta.weight
704: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.38.attn_qkv.weight
705: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.38.attn_gate.weight
706: 128 | 128, 1, 1, 1 | F32 | blk.38.ssm_norm.weight
707: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.38.ssm_out.weight
708: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.38.ffn_down_exps.weight
709: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_gate_exps.weight
710: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.38.ffn_up_exps.weight
711: 786432 | 3072, 256, 1, 1 | F32 | blk.38.ffn_gate_inp.weight
712: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.38.ffn_down_shexp.weight
713: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_gate_shexp.weight
714: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.38.ffn_up_shexp.weight
715: 3072 | 3072, 1, 1, 1 | F32 | blk.38.ffn_gate_inp_shexp.weight
716: 3072 | 3072, 1, 1, 1 | F32 | blk.38.post_attention_norm.weight
717: 3072 | 3072, 1, 1, 1 | F32 | blk.39.attn_norm.weight
718: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.39.ffn_down_exps.weight
719: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_gate_exps.weight
720: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.39.ffn_up_exps.weight
721: 786432 | 3072, 256, 1, 1 | F32 | blk.39.ffn_gate_inp.weight
722: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.39.ffn_down_shexp.weight
723: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_gate_shexp.weight
724: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.39.ffn_up_shexp.weight
725: 3072 | 3072, 1, 1, 1 | F32 | blk.39.ffn_gate_inp_shexp.weight
726: 3072 | 3072, 1, 1, 1 | F32 | blk.39.post_attention_norm.weight
727: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_k_norm.weight
728: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_k.weight
729: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.39.attn_output.weight
730: 256 | 256, 1, 1, 1 | F32 | blk.39.attn_q_norm.weight
731: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.39.attn_q.weight
732: 1572864 | 3072, 512, 1, 1 | BF16 | blk.39.attn_v.weight
733: 3072 | 3072, 1, 1, 1 | F32 | blk.40.attn_norm.weight
734: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_a
735: 49152 | 4, 12288, 1, 1 | F32 | blk.40.ssm_conv1d.weight
736: 64 | 64, 1, 1, 1 | F32 | blk.40.ssm_dt.bias
737: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_alpha.weight
738: 196608 | 3072, 64, 1, 1 | BF16 | blk.40.ssm_beta.weight
739: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.40.attn_qkv.weight
740: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.40.attn_gate.weight
741: 128 | 128, 1, 1, 1 | F32 | blk.40.ssm_norm.weight
742: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.40.ssm_out.weight
743: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.40.ffn_down_exps.weight
744: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_gate_exps.weight
745: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.40.ffn_up_exps.weight
746: 786432 | 3072, 256, 1, 1 | F32 | blk.40.ffn_gate_inp.weight
747: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.40.ffn_down_shexp.weight
748: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_gate_shexp.weight
749: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.40.ffn_up_shexp.weight
750: 3072 | 3072, 1, 1, 1 | F32 | blk.40.ffn_gate_inp_shexp.weight
751: 3072 | 3072, 1, 1, 1 | F32 | blk.40.post_attention_norm.weight
752: 3072 | 3072, 1, 1, 1 | F32 | blk.41.attn_norm.weight
753: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_a
754: 49152 | 4, 12288, 1, 1 | F32 | blk.41.ssm_conv1d.weight
755: 64 | 64, 1, 1, 1 | F32 | blk.41.ssm_dt.bias
756: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_alpha.weight
757: 196608 | 3072, 64, 1, 1 | BF16 | blk.41.ssm_beta.weight
758: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.41.attn_qkv.weight
759: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.41.attn_gate.weight
760: 128 | 128, 1, 1, 1 | F32 | blk.41.ssm_norm.weight
761: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.41.ssm_out.weight
762: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.41.ffn_down_exps.weight
763: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_gate_exps.weight
764: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.41.ffn_up_exps.weight
765: 786432 | 3072, 256, 1, 1 | F32 | blk.41.ffn_gate_inp.weight
766: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.41.ffn_down_shexp.weight
767: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_gate_shexp.weight
768: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.41.ffn_up_shexp.weight
769: 3072 | 3072, 1, 1, 1 | F32 | blk.41.ffn_gate_inp_shexp.weight
770: 3072 | 3072, 1, 1, 1 | F32 | blk.41.post_attention_norm.weight
771: 3072 | 3072, 1, 1, 1 | F32 | blk.42.attn_norm.weight
772: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_a
773: 49152 | 4, 12288, 1, 1 | F32 | blk.42.ssm_conv1d.weight
774: 64 | 64, 1, 1, 1 | F32 | blk.42.ssm_dt.bias
775: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_alpha.weight
776: 196608 | 3072, 64, 1, 1 | BF16 | blk.42.ssm_beta.weight
777: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.42.attn_qkv.weight
778: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.42.attn_gate.weight
779: 128 | 128, 1, 1, 1 | F32 | blk.42.ssm_norm.weight
780: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.42.ssm_out.weight
781: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.42.ffn_down_exps.weight
782: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_gate_exps.weight
783: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.42.ffn_up_exps.weight
784: 786432 | 3072, 256, 1, 1 | F32 | blk.42.ffn_gate_inp.weight
785: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.42.ffn_down_shexp.weight
786: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_gate_shexp.weight
787: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.42.ffn_up_shexp.weight
788: 3072 | 3072, 1, 1, 1 | F32 | blk.42.ffn_gate_inp_shexp.weight
789: 3072 | 3072, 1, 1, 1 | F32 | blk.42.post_attention_norm.weight
790: 3072 | 3072, 1, 1, 1 | F32 | blk.43.attn_norm.weight
791: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.43.ffn_down_exps.weight
792: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_gate_exps.weight
793: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.43.ffn_up_exps.weight
794: 786432 | 3072, 256, 1, 1 | F32 | blk.43.ffn_gate_inp.weight
795: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.43.ffn_down_shexp.weight
796: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_gate_shexp.weight
797: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.43.ffn_up_shexp.weight
798: 3072 | 3072, 1, 1, 1 | F32 | blk.43.ffn_gate_inp_shexp.weight
799: 3072 | 3072, 1, 1, 1 | F32 | blk.43.post_attention_norm.weight
800: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_k_norm.weight
801: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_k.weight
802: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.43.attn_output.weight
803: 256 | 256, 1, 1, 1 | F32 | blk.43.attn_q_norm.weight
804: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.43.attn_q.weight
805: 1572864 | 3072, 512, 1, 1 | BF16 | blk.43.attn_v.weight
806: 3072 | 3072, 1, 1, 1 | F32 | blk.44.attn_norm.weight
807: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_a
808: 49152 | 4, 12288, 1, 1 | F32 | blk.44.ssm_conv1d.weight
809: 64 | 64, 1, 1, 1 | F32 | blk.44.ssm_dt.bias
810: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_alpha.weight
811: 196608 | 3072, 64, 1, 1 | BF16 | blk.44.ssm_beta.weight
812: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.44.attn_qkv.weight
813: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.44.attn_gate.weight
814: 128 | 128, 1, 1, 1 | F32 | blk.44.ssm_norm.weight
815: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.44.ssm_out.weight
816: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.44.ffn_down_exps.weight
817: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_gate_exps.weight
818: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.44.ffn_up_exps.weight
819: 786432 | 3072, 256, 1, 1 | F32 | blk.44.ffn_gate_inp.weight
820: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.44.ffn_down_shexp.weight
821: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_gate_shexp.weight
822: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.44.ffn_up_shexp.weight
823: 3072 | 3072, 1, 1, 1 | F32 | blk.44.ffn_gate_inp_shexp.weight
824: 3072 | 3072, 1, 1, 1 | F32 | blk.44.post_attention_norm.weight
825: 3072 | 3072, 1, 1, 1 | F32 | blk.45.attn_norm.weight
826: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_a
827: 49152 | 4, 12288, 1, 1 | F32 | blk.45.ssm_conv1d.weight
828: 64 | 64, 1, 1, 1 | F32 | blk.45.ssm_dt.bias
829: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_alpha.weight
830: 196608 | 3072, 64, 1, 1 | BF16 | blk.45.ssm_beta.weight
831: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.45.attn_qkv.weight
832: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.45.attn_gate.weight
833: 128 | 128, 1, 1, 1 | F32 | blk.45.ssm_norm.weight
834: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.45.ssm_out.weight
835: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.45.ffn_down_exps.weight
836: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_gate_exps.weight
837: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.45.ffn_up_exps.weight
838: 786432 | 3072, 256, 1, 1 | F32 | blk.45.ffn_gate_inp.weight
839: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.45.ffn_down_shexp.weight
840: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_gate_shexp.weight
841: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.45.ffn_up_shexp.weight
842: 3072 | 3072, 1, 1, 1 | F32 | blk.45.ffn_gate_inp_shexp.weight
843: 3072 | 3072, 1, 1, 1 | F32 | blk.45.post_attention_norm.weight
844: 3072 | 3072, 1, 1, 1 | F32 | blk.46.attn_norm.weight
845: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_a
846: 49152 | 4, 12288, 1, 1 | F32 | blk.46.ssm_conv1d.weight
847: 64 | 64, 1, 1, 1 | F32 | blk.46.ssm_dt.bias
848: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_alpha.weight
849: 196608 | 3072, 64, 1, 1 | BF16 | blk.46.ssm_beta.weight
850: 37748736 | 3072, 12288, 1, 1 | BF16 | blk.46.attn_qkv.weight
851: 25165824 | 3072, 8192, 1, 1 | BF16 | blk.46.attn_gate.weight
852: 128 | 128, 1, 1, 1 | F32 | blk.46.ssm_norm.weight
853: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.46.ssm_out.weight
854: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.46.ffn_down_exps.weight
855: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_gate_exps.weight
856: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.46.ffn_up_exps.weight
857: 786432 | 3072, 256, 1, 1 | F32 | blk.46.ffn_gate_inp.weight
858: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.46.ffn_down_shexp.weight
859: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_gate_shexp.weight
860: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.46.ffn_up_shexp.weight
861: 3072 | 3072, 1, 1, 1 | F32 | blk.46.ffn_gate_inp_shexp.weight
862: 3072 | 3072, 1, 1, 1 | F32 | blk.46.post_attention_norm.weight
863: 3072 | 3072, 1, 1, 1 | F32 | blk.47.attn_norm.weight
864: 805306368 | 1024, 3072, 256, 1 | BF16 | blk.47.ffn_down_exps.weight
865: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_gate_exps.weight
866: 805306368 | 3072, 1024, 256, 1 | BF16 | blk.47.ffn_up_exps.weight
867: 786432 | 3072, 256, 1, 1 | F32 | blk.47.ffn_gate_inp.weight
868: 3145728 | 1024, 3072, 1, 1 | BF16 | blk.47.ffn_down_shexp.weight
869: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_gate_shexp.weight
870: 3145728 | 3072, 1024, 1, 1 | BF16 | blk.47.ffn_up_shexp.weight
871: 3072 | 3072, 1, 1, 1 | F32 | blk.47.ffn_gate_inp_shexp.weight
872: 3072 | 3072, 1, 1, 1 | F32 | blk.47.post_attention_norm.weight
873: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_k_norm.weight
874: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_k.weight
875: 25165824 | 8192, 3072, 1, 1 | BF16 | blk.47.attn_output.weight
876: 256 | 256, 1, 1, 1 | F32 | blk.47.attn_q_norm.weight
877: 50331648 | 3072, 16384, 1, 1 | BF16 | blk.47.attn_q.weight
878: 1572864 | 3072, 512, 1, 1 | BF16 | blk.47.attn_v.weight
879: 3072 | 3072, 1, 1, 1 | F32 | output_norm.weight