Models
Turbo
Sign in
Download
Models
Download
Sign in
granite-code
:34b
233.2K
Downloads
Updated
11 months ago
A family of open foundation models by IBM for Code Intelligence
A family of open foundation models by IBM for Code Intelligence
Cancel
3b
8b
20b
34b
granite-code:34b
...
/
model
85aa034aee49 · 19GB
Metadata
general.architecture
starcoder
starcoder
general.file_type
Q4_0
Q4_0
starcoder.attention.head_count
48
48
starcoder.attention.head_count_kv
1
1
starcoder.attention.layer_norm_epsilon
1e-05
1e-05
starcoder.block_count
88
88
starcoder.context_length
8192
8192
starcoder.embedding_length
6144
6144
starcoder.feed_forward_length
24576
24576
tokenizer.ggml.bos_token_id
0
0
tokenizer.ggml.eos_token_id
0
0
tokenizer.ggml.merges
[Ġ Ġ, ĠĠ ĠĠ, ĠĠĠĠ ĠĠĠĠ, ĠĠ Ġ, e r, ...]
[Ġ Ġ, ĠĠ ĠĠ, ĠĠĠĠ ĠĠĠĠ, ĠĠ Ġ, e r, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
0
0
tokenizer.ggml.pre
refact
refact
tokenizer.ggml.token_type
[3, 3, 3, 3, 3, ...]
[3, 3, 3, 3, 3, ...]
tokenizer.ggml.tokens
[<|endoftext|>, <fim_prefix>, <fim_middle>, <fim_suffix>, <fim_pad>, ...]
[<|endoftext|>, <fim_prefix>, <fim_middle>, <fim_suffix>, <fim_pad>, ...]
tokenizer.ggml.unknown_token_id
0
0
Tensor
Name
Type
Shape
token_embd.weight
Q6_K
Q6_K
[6144, 49152]
blk.0
blk.0.attn_norm.bias
F32
F32
[6144]
blk.0.attn_norm.weight
F32
F32
[6144]
blk.0.attn_output.bias
F32
F32
[6144]
blk.0.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.0.attn_qkv.bias
F32
F32
[6400]
blk.0.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.0.ffn_down.bias
F32
F32
[6144]
blk.0.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.0.ffn_norm.bias
F32
F32
[6144]
blk.0.ffn_norm.weight
F32
F32
[6144]
blk.0.ffn_up.bias
F32
F32
[24576]
blk.0.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.1
blk.1.attn_norm.bias
F32
F32
[6144]
blk.1.attn_norm.weight
F32
F32
[6144]
blk.1.attn_output.bias
F32
F32
[6144]
blk.1.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.1.attn_qkv.bias
F32
F32
[6400]
blk.1.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.1.ffn_down.bias
F32
F32
[6144]
blk.1.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.1.ffn_norm.bias
F32
F32
[6144]
blk.1.ffn_norm.weight
F32
F32
[6144]
blk.1.ffn_up.bias
F32
F32
[24576]
blk.1.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.2
blk.2.attn_norm.bias
F32
F32
[6144]
blk.2.attn_norm.weight
F32
F32
[6144]
blk.2.attn_output.bias
F32
F32
[6144]
blk.2.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.2.attn_qkv.bias
F32
F32
[6400]
blk.2.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.2.ffn_down.bias
F32
F32
[6144]
blk.2.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.2.ffn_norm.bias
F32
F32
[6144]
blk.2.ffn_norm.weight
F32
F32
[6144]
blk.2.ffn_up.bias
F32
F32
[24576]
blk.2.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.3
blk.3.attn_norm.bias
F32
F32
[6144]
blk.3.attn_norm.weight
F32
F32
[6144]
blk.3.attn_output.bias
F32
F32
[6144]
blk.3.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.3.attn_qkv.bias
F32
F32
[6400]
blk.3.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.3.ffn_down.bias
F32
F32
[6144]
blk.3.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.3.ffn_norm.bias
F32
F32
[6144]
blk.3.ffn_norm.weight
F32
F32
[6144]
blk.3.ffn_up.bias
F32
F32
[24576]
blk.3.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.4
blk.4.attn_norm.bias
F32
F32
[6144]
blk.4.attn_norm.weight
F32
F32
[6144]
blk.4.attn_output.bias
F32
F32
[6144]
blk.4.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.4.attn_qkv.bias
F32
F32
[6400]
blk.4.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.4.ffn_down.bias
F32
F32
[6144]
blk.4.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.4.ffn_norm.bias
F32
F32
[6144]
blk.4.ffn_norm.weight
F32
F32
[6144]
blk.4.ffn_up.bias
F32
F32
[24576]
blk.4.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.5
blk.5.attn_norm.bias
F32
F32
[6144]
blk.5.attn_norm.weight
F32
F32
[6144]
blk.5.attn_output.bias
F32
F32
[6144]
blk.5.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.5.attn_qkv.bias
F32
F32
[6400]
blk.5.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.5.ffn_down.bias
F32
F32
[6144]
blk.5.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.5.ffn_norm.bias
F32
F32
[6144]
blk.5.ffn_norm.weight
F32
F32
[6144]
blk.5.ffn_up.bias
F32
F32
[24576]
blk.5.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.6
blk.6.attn_norm.bias
F32
F32
[6144]
blk.6.attn_norm.weight
F32
F32
[6144]
blk.6.attn_output.bias
F32
F32
[6144]
blk.6.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.6.attn_qkv.bias
F32
F32
[6400]
blk.6.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.6.ffn_down.bias
F32
F32
[6144]
blk.6.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.6.ffn_norm.bias
F32
F32
[6144]
blk.6.ffn_norm.weight
F32
F32
[6144]
blk.6.ffn_up.bias
F32
F32
[24576]
blk.6.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.7
blk.7.attn_norm.bias
F32
F32
[6144]
blk.7.attn_norm.weight
F32
F32
[6144]
blk.7.attn_output.bias
F32
F32
[6144]
blk.7.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.7.attn_qkv.bias
F32
F32
[6400]
blk.7.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.7.ffn_down.bias
F32
F32
[6144]
blk.7.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.7.ffn_norm.bias
F32
F32
[6144]
blk.7.ffn_norm.weight
F32
F32
[6144]
blk.7.ffn_up.bias
F32
F32
[24576]
blk.7.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.8
blk.8.attn_norm.bias
F32
F32
[6144]
blk.8.attn_norm.weight
F32
F32
[6144]
blk.8.attn_output.bias
F32
F32
[6144]
blk.8.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.8.attn_qkv.bias
F32
F32
[6400]
blk.8.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.8.ffn_down.bias
F32
F32
[6144]
blk.8.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.8.ffn_norm.bias
F32
F32
[6144]
blk.8.ffn_norm.weight
F32
F32
[6144]
blk.8.ffn_up.bias
F32
F32
[24576]
blk.8.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.9
blk.9.attn_norm.bias
F32
F32
[6144]
blk.9.attn_norm.weight
F32
F32
[6144]
blk.9.attn_output.bias
F32
F32
[6144]
blk.9.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.9.attn_qkv.bias
F32
F32
[6400]
blk.9.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.9.ffn_down.bias
F32
F32
[6144]
blk.9.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.9.ffn_norm.bias
F32
F32
[6144]
blk.9.ffn_norm.weight
F32
F32
[6144]
blk.9.ffn_up.bias
F32
F32
[24576]
blk.9.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.10
blk.10.attn_norm.bias
F32
F32
[6144]
blk.10.attn_norm.weight
F32
F32
[6144]
blk.10.attn_output.bias
F32
F32
[6144]
blk.10.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.10.attn_qkv.bias
F32
F32
[6400]
blk.10.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.10.ffn_down.bias
F32
F32
[6144]
blk.10.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.10.ffn_norm.bias
F32
F32
[6144]
blk.10.ffn_norm.weight
F32
F32
[6144]
blk.10.ffn_up.bias
F32
F32
[24576]
blk.10.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.11
blk.11.attn_norm.bias
F32
F32
[6144]
blk.11.attn_norm.weight
F32
F32
[6144]
blk.11.attn_output.bias
F32
F32
[6144]
blk.11.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.11.attn_qkv.bias
F32
F32
[6400]
blk.11.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.11.ffn_down.bias
F32
F32
[6144]
blk.11.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.11.ffn_norm.bias
F32
F32
[6144]
blk.11.ffn_norm.weight
F32
F32
[6144]
blk.11.ffn_up.bias
F32
F32
[24576]
blk.11.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.12
blk.12.attn_norm.bias
F32
F32
[6144]
blk.12.attn_norm.weight
F32
F32
[6144]
blk.12.attn_output.bias
F32
F32
[6144]
blk.12.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.12.attn_qkv.bias
F32
F32
[6400]
blk.12.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.12.ffn_down.bias
F32
F32
[6144]
blk.12.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.12.ffn_norm.bias
F32
F32
[6144]
blk.12.ffn_norm.weight
F32
F32
[6144]
blk.12.ffn_up.bias
F32
F32
[24576]
blk.12.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.13
blk.13.attn_norm.bias
F32
F32
[6144]
blk.13.attn_norm.weight
F32
F32
[6144]
blk.13.attn_output.bias
F32
F32
[6144]
blk.13.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.13.attn_qkv.bias
F32
F32
[6400]
blk.13.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.13.ffn_down.bias
F32
F32
[6144]
blk.13.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.13.ffn_norm.bias
F32
F32
[6144]
blk.13.ffn_norm.weight
F32
F32
[6144]
blk.13.ffn_up.bias
F32
F32
[24576]
blk.13.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.14
blk.14.attn_norm.bias
F32
F32
[6144]
blk.14.attn_norm.weight
F32
F32
[6144]
blk.14.attn_output.bias
F32
F32
[6144]
blk.14.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.14.attn_qkv.bias
F32
F32
[6400]
blk.14.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.14.ffn_down.bias
F32
F32
[6144]
blk.14.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.14.ffn_norm.bias
F32
F32
[6144]
blk.14.ffn_norm.weight
F32
F32
[6144]
blk.14.ffn_up.bias
F32
F32
[24576]
blk.14.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.15
blk.15.attn_norm.bias
F32
F32
[6144]
blk.15.attn_norm.weight
F32
F32
[6144]
blk.15.attn_output.bias
F32
F32
[6144]
blk.15.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.15.attn_qkv.bias
F32
F32
[6400]
blk.15.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.15.ffn_down.bias
F32
F32
[6144]
blk.15.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.15.ffn_norm.bias
F32
F32
[6144]
blk.15.ffn_norm.weight
F32
F32
[6144]
blk.15.ffn_up.bias
F32
F32
[24576]
blk.15.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.16
blk.16.attn_norm.bias
F32
F32
[6144]
blk.16.attn_norm.weight
F32
F32
[6144]
blk.16.attn_output.bias
F32
F32
[6144]
blk.16.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.16.attn_qkv.bias
F32
F32
[6400]
blk.16.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.16.ffn_down.bias
F32
F32
[6144]
blk.16.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.16.ffn_norm.bias
F32
F32
[6144]
blk.16.ffn_norm.weight
F32
F32
[6144]
blk.16.ffn_up.bias
F32
F32
[24576]
blk.16.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.17
blk.17.attn_norm.bias
F32
F32
[6144]
blk.17.attn_norm.weight
F32
F32
[6144]
blk.17.attn_output.bias
F32
F32
[6144]
blk.17.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.17.attn_qkv.bias
F32
F32
[6400]
blk.17.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.17.ffn_down.bias
F32
F32
[6144]
blk.17.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.17.ffn_norm.bias
F32
F32
[6144]
blk.17.ffn_norm.weight
F32
F32
[6144]
blk.17.ffn_up.bias
F32
F32
[24576]
blk.17.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.18
blk.18.attn_norm.bias
F32
F32
[6144]
blk.18.attn_norm.weight
F32
F32
[6144]
blk.18.attn_output.bias
F32
F32
[6144]
blk.18.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.18.attn_qkv.bias
F32
F32
[6400]
blk.18.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.18.ffn_down.bias
F32
F32
[6144]
blk.18.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.18.ffn_norm.bias
F32
F32
[6144]
blk.18.ffn_norm.weight
F32
F32
[6144]
blk.18.ffn_up.bias
F32
F32
[24576]
blk.18.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.19
blk.19.attn_norm.bias
F32
F32
[6144]
blk.19.attn_norm.weight
F32
F32
[6144]
blk.19.attn_output.bias
F32
F32
[6144]
blk.19.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.19.attn_qkv.bias
F32
F32
[6400]
blk.19.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.19.ffn_down.bias
F32
F32
[6144]
blk.19.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.19.ffn_norm.bias
F32
F32
[6144]
blk.19.ffn_norm.weight
F32
F32
[6144]
blk.19.ffn_up.bias
F32
F32
[24576]
blk.19.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.20
blk.20.attn_norm.bias
F32
F32
[6144]
blk.20.attn_norm.weight
F32
F32
[6144]
blk.20.attn_output.bias
F32
F32
[6144]
blk.20.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.20.attn_qkv.bias
F32
F32
[6400]
blk.20.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.20.ffn_down.bias
F32
F32
[6144]
blk.20.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.20.ffn_norm.bias
F32
F32
[6144]
blk.20.ffn_norm.weight
F32
F32
[6144]
blk.20.ffn_up.bias
F32
F32
[24576]
blk.20.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.21
blk.21.attn_norm.bias
F32
F32
[6144]
blk.21.attn_norm.weight
F32
F32
[6144]
blk.21.attn_output.bias
F32
F32
[6144]
blk.21.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.21.attn_qkv.bias
F32
F32
[6400]
blk.21.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.21.ffn_down.bias
F32
F32
[6144]
blk.21.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.21.ffn_norm.bias
F32
F32
[6144]
blk.21.ffn_norm.weight
F32
F32
[6144]
blk.21.ffn_up.bias
F32
F32
[24576]
blk.21.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.22
blk.22.attn_norm.bias
F32
F32
[6144]
blk.22.attn_norm.weight
F32
F32
[6144]
blk.22.attn_output.bias
F32
F32
[6144]
blk.22.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.22.attn_qkv.bias
F32
F32
[6400]
blk.22.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.22.ffn_down.bias
F32
F32
[6144]
blk.22.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.22.ffn_norm.bias
F32
F32
[6144]
blk.22.ffn_norm.weight
F32
F32
[6144]
blk.22.ffn_up.bias
F32
F32
[24576]
blk.22.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.23
blk.23.attn_norm.bias
F32
F32
[6144]
blk.23.attn_norm.weight
F32
F32
[6144]
blk.23.attn_output.bias
F32
F32
[6144]
blk.23.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.23.attn_qkv.bias
F32
F32
[6400]
blk.23.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.23.ffn_down.bias
F32
F32
[6144]
blk.23.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.23.ffn_norm.bias
F32
F32
[6144]
blk.23.ffn_norm.weight
F32
F32
[6144]
blk.23.ffn_up.bias
F32
F32
[24576]
blk.23.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.24
blk.24.attn_norm.bias
F32
F32
[6144]
blk.24.attn_norm.weight
F32
F32
[6144]
blk.24.attn_output.bias
F32
F32
[6144]
blk.24.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.24.attn_qkv.bias
F32
F32
[6400]
blk.24.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.24.ffn_down.bias
F32
F32
[6144]
blk.24.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.24.ffn_norm.bias
F32
F32
[6144]
blk.24.ffn_norm.weight
F32
F32
[6144]
blk.24.ffn_up.bias
F32
F32
[24576]
blk.24.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.25
blk.25.attn_norm.bias
F32
F32
[6144]
blk.25.attn_norm.weight
F32
F32
[6144]
blk.25.attn_output.bias
F32
F32
[6144]
blk.25.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.25.attn_qkv.bias
F32
F32
[6400]
blk.25.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.25.ffn_down.bias
F32
F32
[6144]
blk.25.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.25.ffn_norm.bias
F32
F32
[6144]
blk.25.ffn_norm.weight
F32
F32
[6144]
blk.25.ffn_up.bias
F32
F32
[24576]
blk.25.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.26
blk.26.attn_norm.bias
F32
F32
[6144]
blk.26.attn_norm.weight
F32
F32
[6144]
blk.26.attn_output.bias
F32
F32
[6144]
blk.26.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.26.attn_qkv.bias
F32
F32
[6400]
blk.26.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.26.ffn_down.bias
F32
F32
[6144]
blk.26.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.26.ffn_norm.bias
F32
F32
[6144]
blk.26.ffn_norm.weight
F32
F32
[6144]
blk.26.ffn_up.bias
F32
F32
[24576]
blk.26.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.27
blk.27.attn_norm.bias
F32
F32
[6144]
blk.27.attn_norm.weight
F32
F32
[6144]
blk.27.attn_output.bias
F32
F32
[6144]
blk.27.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.27.attn_qkv.bias
F32
F32
[6400]
blk.27.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.27.ffn_down.bias
F32
F32
[6144]
blk.27.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.27.ffn_norm.bias
F32
F32
[6144]
blk.27.ffn_norm.weight
F32
F32
[6144]
blk.27.ffn_up.bias
F32
F32
[24576]
blk.27.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.28
blk.28.attn_norm.bias
F32
F32
[6144]
blk.28.attn_norm.weight
F32
F32
[6144]
blk.28.attn_output.bias
F32
F32
[6144]
blk.28.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.28.attn_qkv.bias
F32
F32
[6400]
blk.28.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.28.ffn_down.bias
F32
F32
[6144]
blk.28.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.28.ffn_norm.bias
F32
F32
[6144]
blk.28.ffn_norm.weight
F32
F32
[6144]
blk.28.ffn_up.bias
F32
F32
[24576]
blk.28.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.29
blk.29.attn_norm.bias
F32
F32
[6144]
blk.29.attn_norm.weight
F32
F32
[6144]
blk.29.attn_output.bias
F32
F32
[6144]
blk.29.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.29.attn_qkv.bias
F32
F32
[6400]
blk.29.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.29.ffn_down.bias
F32
F32
[6144]
blk.29.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.29.ffn_norm.bias
F32
F32
[6144]
blk.29.ffn_norm.weight
F32
F32
[6144]
blk.29.ffn_up.bias
F32
F32
[24576]
blk.29.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.30
blk.30.attn_norm.bias
F32
F32
[6144]
blk.30.attn_norm.weight
F32
F32
[6144]
blk.30.attn_output.bias
F32
F32
[6144]
blk.30.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.30.attn_qkv.bias
F32
F32
[6400]
blk.30.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.30.ffn_down.bias
F32
F32
[6144]
blk.30.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.30.ffn_norm.bias
F32
F32
[6144]
blk.30.ffn_norm.weight
F32
F32
[6144]
blk.30.ffn_up.bias
F32
F32
[24576]
blk.30.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.31
blk.31.attn_norm.bias
F32
F32
[6144]
blk.31.attn_norm.weight
F32
F32
[6144]
blk.31.attn_output.bias
F32
F32
[6144]
blk.31.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.31.attn_qkv.bias
F32
F32
[6400]
blk.31.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.31.ffn_down.bias
F32
F32
[6144]
blk.31.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.31.ffn_norm.bias
F32
F32
[6144]
blk.31.ffn_norm.weight
F32
F32
[6144]
blk.31.ffn_up.bias
F32
F32
[24576]
blk.31.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.32
blk.32.attn_norm.bias
F32
F32
[6144]
blk.32.attn_norm.weight
F32
F32
[6144]
blk.32.attn_output.bias
F32
F32
[6144]
blk.32.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.32.attn_qkv.bias
F32
F32
[6400]
blk.32.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.32.ffn_down.bias
F32
F32
[6144]
blk.32.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.32.ffn_norm.bias
F32
F32
[6144]
blk.32.ffn_norm.weight
F32
F32
[6144]
blk.32.ffn_up.bias
F32
F32
[24576]
blk.32.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.33
blk.33.attn_norm.bias
F32
F32
[6144]
blk.33.attn_norm.weight
F32
F32
[6144]
blk.33.attn_output.bias
F32
F32
[6144]
blk.33.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.33.attn_qkv.bias
F32
F32
[6400]
blk.33.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.33.ffn_down.bias
F32
F32
[6144]
blk.33.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.33.ffn_norm.bias
F32
F32
[6144]
blk.33.ffn_norm.weight
F32
F32
[6144]
blk.33.ffn_up.bias
F32
F32
[24576]
blk.33.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.34
blk.34.attn_norm.bias
F32
F32
[6144]
blk.34.attn_norm.weight
F32
F32
[6144]
blk.34.attn_output.bias
F32
F32
[6144]
blk.34.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.34.attn_qkv.bias
F32
F32
[6400]
blk.34.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.34.ffn_down.bias
F32
F32
[6144]
blk.34.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.34.ffn_norm.bias
F32
F32
[6144]
blk.34.ffn_norm.weight
F32
F32
[6144]
blk.34.ffn_up.bias
F32
F32
[24576]
blk.34.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.35
blk.35.attn_norm.bias
F32
F32
[6144]
blk.35.attn_norm.weight
F32
F32
[6144]
blk.35.attn_output.bias
F32
F32
[6144]
blk.35.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.35.attn_qkv.bias
F32
F32
[6400]
blk.35.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.35.ffn_down.bias
F32
F32
[6144]
blk.35.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.35.ffn_norm.bias
F32
F32
[6144]
blk.35.ffn_norm.weight
F32
F32
[6144]
blk.35.ffn_up.bias
F32
F32
[24576]
blk.35.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.36
blk.36.attn_norm.bias
F32
F32
[6144]
blk.36.attn_norm.weight
F32
F32
[6144]
blk.36.attn_output.bias
F32
F32
[6144]
blk.36.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.36.attn_qkv.bias
F32
F32
[6400]
blk.36.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.36.ffn_down.bias
F32
F32
[6144]
blk.36.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.36.ffn_norm.bias
F32
F32
[6144]
blk.36.ffn_norm.weight
F32
F32
[6144]
blk.36.ffn_up.bias
F32
F32
[24576]
blk.36.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.37
blk.37.attn_norm.bias
F32
F32
[6144]
blk.37.attn_norm.weight
F32
F32
[6144]
blk.37.attn_output.bias
F32
F32
[6144]
blk.37.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.37.attn_qkv.bias
F32
F32
[6400]
blk.37.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.37.ffn_down.bias
F32
F32
[6144]
blk.37.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.37.ffn_norm.bias
F32
F32
[6144]
blk.37.ffn_norm.weight
F32
F32
[6144]
blk.37.ffn_up.bias
F32
F32
[24576]
blk.37.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.38
blk.38.attn_norm.bias
F32
F32
[6144]
blk.38.attn_norm.weight
F32
F32
[6144]
blk.38.attn_output.bias
F32
F32
[6144]
blk.38.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.38.attn_qkv.bias
F32
F32
[6400]
blk.38.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.38.ffn_down.bias
F32
F32
[6144]
blk.38.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.38.ffn_norm.bias
F32
F32
[6144]
blk.38.ffn_norm.weight
F32
F32
[6144]
blk.38.ffn_up.bias
F32
F32
[24576]
blk.38.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.39
blk.39.attn_norm.bias
F32
F32
[6144]
blk.39.attn_norm.weight
F32
F32
[6144]
blk.39.attn_output.bias
F32
F32
[6144]
blk.39.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.39.attn_qkv.bias
F32
F32
[6400]
blk.39.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.39.ffn_down.bias
F32
F32
[6144]
blk.39.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.39.ffn_norm.bias
F32
F32
[6144]
blk.39.ffn_norm.weight
F32
F32
[6144]
blk.39.ffn_up.bias
F32
F32
[24576]
blk.39.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.40
blk.40.attn_norm.bias
F32
F32
[6144]
blk.40.attn_norm.weight
F32
F32
[6144]
blk.40.attn_output.bias
F32
F32
[6144]
blk.40.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.40.attn_qkv.bias
F32
F32
[6400]
blk.40.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.40.ffn_down.bias
F32
F32
[6144]
blk.40.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.40.ffn_norm.bias
F32
F32
[6144]
blk.40.ffn_norm.weight
F32
F32
[6144]
blk.40.ffn_up.bias
F32
F32
[24576]
blk.40.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.41
blk.41.attn_norm.bias
F32
F32
[6144]
blk.41.attn_norm.weight
F32
F32
[6144]
blk.41.attn_output.bias
F32
F32
[6144]
blk.41.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.41.attn_qkv.bias
F32
F32
[6400]
blk.41.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.41.ffn_down.bias
F32
F32
[6144]
blk.41.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.41.ffn_norm.bias
F32
F32
[6144]
blk.41.ffn_norm.weight
F32
F32
[6144]
blk.41.ffn_up.bias
F32
F32
[24576]
blk.41.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.42
blk.42.attn_norm.bias
F32
F32
[6144]
blk.42.attn_norm.weight
F32
F32
[6144]
blk.42.attn_output.bias
F32
F32
[6144]
blk.42.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.42.attn_qkv.bias
F32
F32
[6400]
blk.42.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.42.ffn_down.bias
F32
F32
[6144]
blk.42.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.42.ffn_norm.bias
F32
F32
[6144]
blk.42.ffn_norm.weight
F32
F32
[6144]
blk.42.ffn_up.bias
F32
F32
[24576]
blk.42.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.43
blk.43.attn_norm.bias
F32
F32
[6144]
blk.43.attn_norm.weight
F32
F32
[6144]
blk.43.attn_output.bias
F32
F32
[6144]
blk.43.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.43.attn_qkv.bias
F32
F32
[6400]
blk.43.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.43.ffn_down.bias
F32
F32
[6144]
blk.43.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.43.ffn_norm.bias
F32
F32
[6144]
blk.43.ffn_norm.weight
F32
F32
[6144]
blk.43.ffn_up.bias
F32
F32
[24576]
blk.43.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.44
blk.44.attn_norm.bias
F32
F32
[6144]
blk.44.attn_norm.weight
F32
F32
[6144]
blk.44.attn_output.bias
F32
F32
[6144]
blk.44.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.44.attn_qkv.bias
F32
F32
[6400]
blk.44.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.44.ffn_down.bias
F32
F32
[6144]
blk.44.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.44.ffn_norm.bias
F32
F32
[6144]
blk.44.ffn_norm.weight
F32
F32
[6144]
blk.44.ffn_up.bias
F32
F32
[24576]
blk.44.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.45
blk.45.attn_norm.bias
F32
F32
[6144]
blk.45.attn_norm.weight
F32
F32
[6144]
blk.45.attn_output.bias
F32
F32
[6144]
blk.45.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.45.attn_qkv.bias
F32
F32
[6400]
blk.45.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.45.ffn_down.bias
F32
F32
[6144]
blk.45.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.45.ffn_norm.bias
F32
F32
[6144]
blk.45.ffn_norm.weight
F32
F32
[6144]
blk.45.ffn_up.bias
F32
F32
[24576]
blk.45.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.46
blk.46.attn_norm.bias
F32
F32
[6144]
blk.46.attn_norm.weight
F32
F32
[6144]
blk.46.attn_output.bias
F32
F32
[6144]
blk.46.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.46.attn_qkv.bias
F32
F32
[6400]
blk.46.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.46.ffn_down.bias
F32
F32
[6144]
blk.46.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.46.ffn_norm.bias
F32
F32
[6144]
blk.46.ffn_norm.weight
F32
F32
[6144]
blk.46.ffn_up.bias
F32
F32
[24576]
blk.46.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.47
blk.47.attn_norm.bias
F32
F32
[6144]
blk.47.attn_norm.weight
F32
F32
[6144]
blk.47.attn_output.bias
F32
F32
[6144]
blk.47.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.47.attn_qkv.bias
F32
F32
[6400]
blk.47.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.47.ffn_down.bias
F32
F32
[6144]
blk.47.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.47.ffn_norm.bias
F32
F32
[6144]
blk.47.ffn_norm.weight
F32
F32
[6144]
blk.47.ffn_up.bias
F32
F32
[24576]
blk.47.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.48
blk.48.attn_norm.bias
F32
F32
[6144]
blk.48.attn_norm.weight
F32
F32
[6144]
blk.48.attn_output.bias
F32
F32
[6144]
blk.48.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.48.attn_qkv.bias
F32
F32
[6400]
blk.48.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.48.ffn_down.bias
F32
F32
[6144]
blk.48.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.48.ffn_norm.bias
F32
F32
[6144]
blk.48.ffn_norm.weight
F32
F32
[6144]
blk.48.ffn_up.bias
F32
F32
[24576]
blk.48.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.49
blk.49.attn_norm.bias
F32
F32
[6144]
blk.49.attn_norm.weight
F32
F32
[6144]
blk.49.attn_output.bias
F32
F32
[6144]
blk.49.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.49.attn_qkv.bias
F32
F32
[6400]
blk.49.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.49.ffn_down.bias
F32
F32
[6144]
blk.49.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.49.ffn_norm.bias
F32
F32
[6144]
blk.49.ffn_norm.weight
F32
F32
[6144]
blk.49.ffn_up.bias
F32
F32
[24576]
blk.49.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.50
blk.50.attn_norm.bias
F32
F32
[6144]
blk.50.attn_norm.weight
F32
F32
[6144]
blk.50.attn_output.bias
F32
F32
[6144]
blk.50.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.50.attn_qkv.bias
F32
F32
[6400]
blk.50.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.50.ffn_down.bias
F32
F32
[6144]
blk.50.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.50.ffn_norm.bias
F32
F32
[6144]
blk.50.ffn_norm.weight
F32
F32
[6144]
blk.50.ffn_up.bias
F32
F32
[24576]
blk.50.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.51
blk.51.attn_norm.bias
F32
F32
[6144]
blk.51.attn_norm.weight
F32
F32
[6144]
blk.51.attn_output.bias
F32
F32
[6144]
blk.51.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.51.attn_qkv.bias
F32
F32
[6400]
blk.51.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.51.ffn_down.bias
F32
F32
[6144]
blk.51.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.51.ffn_norm.bias
F32
F32
[6144]
blk.51.ffn_norm.weight
F32
F32
[6144]
blk.51.ffn_up.bias
F32
F32
[24576]
blk.51.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.52
blk.52.attn_norm.bias
F32
F32
[6144]
blk.52.attn_norm.weight
F32
F32
[6144]
blk.52.attn_output.bias
F32
F32
[6144]
blk.52.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.52.attn_qkv.bias
F32
F32
[6400]
blk.52.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.52.ffn_down.bias
F32
F32
[6144]
blk.52.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.52.ffn_norm.bias
F32
F32
[6144]
blk.52.ffn_norm.weight
F32
F32
[6144]
blk.52.ffn_up.bias
F32
F32
[24576]
blk.52.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.53
blk.53.attn_norm.bias
F32
F32
[6144]
blk.53.attn_norm.weight
F32
F32
[6144]
blk.53.attn_output.bias
F32
F32
[6144]
blk.53.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.53.attn_qkv.bias
F32
F32
[6400]
blk.53.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.53.ffn_down.bias
F32
F32
[6144]
blk.53.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.53.ffn_norm.bias
F32
F32
[6144]
blk.53.ffn_norm.weight
F32
F32
[6144]
blk.53.ffn_up.bias
F32
F32
[24576]
blk.53.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.54
blk.54.attn_norm.bias
F32
F32
[6144]
blk.54.attn_norm.weight
F32
F32
[6144]
blk.54.attn_output.bias
F32
F32
[6144]
blk.54.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.54.attn_qkv.bias
F32
F32
[6400]
blk.54.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.54.ffn_down.bias
F32
F32
[6144]
blk.54.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.54.ffn_norm.bias
F32
F32
[6144]
blk.54.ffn_norm.weight
F32
F32
[6144]
blk.54.ffn_up.bias
F32
F32
[24576]
blk.54.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.55
blk.55.attn_norm.bias
F32
F32
[6144]
blk.55.attn_norm.weight
F32
F32
[6144]
blk.55.attn_output.bias
F32
F32
[6144]
blk.55.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.55.attn_qkv.bias
F32
F32
[6400]
blk.55.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.55.ffn_down.bias
F32
F32
[6144]
blk.55.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.55.ffn_norm.bias
F32
F32
[6144]
blk.55.ffn_norm.weight
F32
F32
[6144]
blk.55.ffn_up.bias
F32
F32
[24576]
blk.55.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.56
blk.56.attn_norm.bias
F32
F32
[6144]
blk.56.attn_norm.weight
F32
F32
[6144]
blk.56.attn_output.bias
F32
F32
[6144]
blk.56.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.56.attn_qkv.bias
F32
F32
[6400]
blk.56.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.56.ffn_down.bias
F32
F32
[6144]
blk.56.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.56.ffn_norm.bias
F32
F32
[6144]
blk.56.ffn_norm.weight
F32
F32
[6144]
blk.56.ffn_up.bias
F32
F32
[24576]
blk.56.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.57
blk.57.attn_norm.bias
F32
F32
[6144]
blk.57.attn_norm.weight
F32
F32
[6144]
blk.57.attn_output.bias
F32
F32
[6144]
blk.57.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.57.attn_qkv.bias
F32
F32
[6400]
blk.57.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.57.ffn_down.bias
F32
F32
[6144]
blk.57.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.57.ffn_norm.bias
F32
F32
[6144]
blk.57.ffn_norm.weight
F32
F32
[6144]
blk.57.ffn_up.bias
F32
F32
[24576]
blk.57.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.58
blk.58.attn_norm.bias
F32
F32
[6144]
blk.58.attn_norm.weight
F32
F32
[6144]
blk.58.attn_output.bias
F32
F32
[6144]
blk.58.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.58.attn_qkv.bias
F32
F32
[6400]
blk.58.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.58.ffn_down.bias
F32
F32
[6144]
blk.58.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.58.ffn_norm.bias
F32
F32
[6144]
blk.58.ffn_norm.weight
F32
F32
[6144]
blk.58.ffn_up.bias
F32
F32
[24576]
blk.58.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.59
blk.59.attn_norm.bias
F32
F32
[6144]
blk.59.attn_norm.weight
F32
F32
[6144]
blk.59.attn_output.bias
F32
F32
[6144]
blk.59.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.59.attn_qkv.bias
F32
F32
[6400]
blk.59.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.59.ffn_down.bias
F32
F32
[6144]
blk.59.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.59.ffn_norm.bias
F32
F32
[6144]
blk.59.ffn_norm.weight
F32
F32
[6144]
blk.59.ffn_up.bias
F32
F32
[24576]
blk.59.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.60
blk.60.attn_norm.bias
F32
F32
[6144]
blk.60.attn_norm.weight
F32
F32
[6144]
blk.60.attn_output.bias
F32
F32
[6144]
blk.60.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.60.attn_qkv.bias
F32
F32
[6400]
blk.60.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.60.ffn_down.bias
F32
F32
[6144]
blk.60.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.60.ffn_norm.bias
F32
F32
[6144]
blk.60.ffn_norm.weight
F32
F32
[6144]
blk.60.ffn_up.bias
F32
F32
[24576]
blk.60.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.61
blk.61.attn_norm.bias
F32
F32
[6144]
blk.61.attn_norm.weight
F32
F32
[6144]
blk.61.attn_output.bias
F32
F32
[6144]
blk.61.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.61.attn_qkv.bias
F32
F32
[6400]
blk.61.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.61.ffn_down.bias
F32
F32
[6144]
blk.61.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.61.ffn_norm.bias
F32
F32
[6144]
blk.61.ffn_norm.weight
F32
F32
[6144]
blk.61.ffn_up.bias
F32
F32
[24576]
blk.61.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.62
blk.62.attn_norm.bias
F32
F32
[6144]
blk.62.attn_norm.weight
F32
F32
[6144]
blk.62.attn_output.bias
F32
F32
[6144]
blk.62.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.62.attn_qkv.bias
F32
F32
[6400]
blk.62.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.62.ffn_down.bias
F32
F32
[6144]
blk.62.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.62.ffn_norm.bias
F32
F32
[6144]
blk.62.ffn_norm.weight
F32
F32
[6144]
blk.62.ffn_up.bias
F32
F32
[24576]
blk.62.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.63
blk.63.attn_norm.bias
F32
F32
[6144]
blk.63.attn_norm.weight
F32
F32
[6144]
blk.63.attn_output.bias
F32
F32
[6144]
blk.63.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.63.attn_qkv.bias
F32
F32
[6400]
blk.63.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.63.ffn_down.bias
F32
F32
[6144]
blk.63.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.63.ffn_norm.bias
F32
F32
[6144]
blk.63.ffn_norm.weight
F32
F32
[6144]
blk.63.ffn_up.bias
F32
F32
[24576]
blk.63.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.64
blk.64.attn_norm.bias
F32
F32
[6144]
blk.64.attn_norm.weight
F32
F32
[6144]
blk.64.attn_output.bias
F32
F32
[6144]
blk.64.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.64.attn_qkv.bias
F32
F32
[6400]
blk.64.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.64.ffn_down.bias
F32
F32
[6144]
blk.64.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.64.ffn_norm.bias
F32
F32
[6144]
blk.64.ffn_norm.weight
F32
F32
[6144]
blk.64.ffn_up.bias
F32
F32
[24576]
blk.64.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.65
blk.65.attn_norm.bias
F32
F32
[6144]
blk.65.attn_norm.weight
F32
F32
[6144]
blk.65.attn_output.bias
F32
F32
[6144]
blk.65.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.65.attn_qkv.bias
F32
F32
[6400]
blk.65.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.65.ffn_down.bias
F32
F32
[6144]
blk.65.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.65.ffn_norm.bias
F32
F32
[6144]
blk.65.ffn_norm.weight
F32
F32
[6144]
blk.65.ffn_up.bias
F32
F32
[24576]
blk.65.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.66
blk.66.attn_norm.bias
F32
F32
[6144]
blk.66.attn_norm.weight
F32
F32
[6144]
blk.66.attn_output.bias
F32
F32
[6144]
blk.66.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.66.attn_qkv.bias
F32
F32
[6400]
blk.66.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.66.ffn_down.bias
F32
F32
[6144]
blk.66.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.66.ffn_norm.bias
F32
F32
[6144]
blk.66.ffn_norm.weight
F32
F32
[6144]
blk.66.ffn_up.bias
F32
F32
[24576]
blk.66.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.67
blk.67.attn_norm.bias
F32
F32
[6144]
blk.67.attn_norm.weight
F32
F32
[6144]
blk.67.attn_output.bias
F32
F32
[6144]
blk.67.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.67.attn_qkv.bias
F32
F32
[6400]
blk.67.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.67.ffn_down.bias
F32
F32
[6144]
blk.67.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.67.ffn_norm.bias
F32
F32
[6144]
blk.67.ffn_norm.weight
F32
F32
[6144]
blk.67.ffn_up.bias
F32
F32
[24576]
blk.67.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.68
blk.68.attn_norm.bias
F32
F32
[6144]
blk.68.attn_norm.weight
F32
F32
[6144]
blk.68.attn_output.bias
F32
F32
[6144]
blk.68.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.68.attn_qkv.bias
F32
F32
[6400]
blk.68.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.68.ffn_down.bias
F32
F32
[6144]
blk.68.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.68.ffn_norm.bias
F32
F32
[6144]
blk.68.ffn_norm.weight
F32
F32
[6144]
blk.68.ffn_up.bias
F32
F32
[24576]
blk.68.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.69
blk.69.attn_norm.bias
F32
F32
[6144]
blk.69.attn_norm.weight
F32
F32
[6144]
blk.69.attn_output.bias
F32
F32
[6144]
blk.69.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.69.attn_qkv.bias
F32
F32
[6400]
blk.69.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.69.ffn_down.bias
F32
F32
[6144]
blk.69.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.69.ffn_norm.bias
F32
F32
[6144]
blk.69.ffn_norm.weight
F32
F32
[6144]
blk.69.ffn_up.bias
F32
F32
[24576]
blk.69.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.70
blk.70.attn_norm.bias
F32
F32
[6144]
blk.70.attn_norm.weight
F32
F32
[6144]
blk.70.attn_output.bias
F32
F32
[6144]
blk.70.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.70.attn_qkv.bias
F32
F32
[6400]
blk.70.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.70.ffn_down.bias
F32
F32
[6144]
blk.70.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.70.ffn_norm.bias
F32
F32
[6144]
blk.70.ffn_norm.weight
F32
F32
[6144]
blk.70.ffn_up.bias
F32
F32
[24576]
blk.70.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.71
blk.71.attn_norm.bias
F32
F32
[6144]
blk.71.attn_norm.weight
F32
F32
[6144]
blk.71.attn_output.bias
F32
F32
[6144]
blk.71.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.71.attn_qkv.bias
F32
F32
[6400]
blk.71.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.71.ffn_down.bias
F32
F32
[6144]
blk.71.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.71.ffn_norm.bias
F32
F32
[6144]
blk.71.ffn_norm.weight
F32
F32
[6144]
blk.71.ffn_up.bias
F32
F32
[24576]
blk.71.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.72
blk.72.attn_norm.bias
F32
F32
[6144]
blk.72.attn_norm.weight
F32
F32
[6144]
blk.72.attn_output.bias
F32
F32
[6144]
blk.72.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.72.attn_qkv.bias
F32
F32
[6400]
blk.72.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.72.ffn_down.bias
F32
F32
[6144]
blk.72.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.72.ffn_norm.bias
F32
F32
[6144]
blk.72.ffn_norm.weight
F32
F32
[6144]
blk.72.ffn_up.bias
F32
F32
[24576]
blk.72.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.73
blk.73.attn_norm.bias
F32
F32
[6144]
blk.73.attn_norm.weight
F32
F32
[6144]
blk.73.attn_output.bias
F32
F32
[6144]
blk.73.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.73.attn_qkv.bias
F32
F32
[6400]
blk.73.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.73.ffn_down.bias
F32
F32
[6144]
blk.73.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.73.ffn_norm.bias
F32
F32
[6144]
blk.73.ffn_norm.weight
F32
F32
[6144]
blk.73.ffn_up.bias
F32
F32
[24576]
blk.73.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.74
blk.74.attn_norm.bias
F32
F32
[6144]
blk.74.attn_norm.weight
F32
F32
[6144]
blk.74.attn_output.bias
F32
F32
[6144]
blk.74.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.74.attn_qkv.bias
F32
F32
[6400]
blk.74.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.74.ffn_down.bias
F32
F32
[6144]
blk.74.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.74.ffn_norm.bias
F32
F32
[6144]
blk.74.ffn_norm.weight
F32
F32
[6144]
blk.74.ffn_up.bias
F32
F32
[24576]
blk.74.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.75
blk.75.attn_norm.bias
F32
F32
[6144]
blk.75.attn_norm.weight
F32
F32
[6144]
blk.75.attn_output.bias
F32
F32
[6144]
blk.75.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.75.attn_qkv.bias
F32
F32
[6400]
blk.75.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.75.ffn_down.bias
F32
F32
[6144]
blk.75.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.75.ffn_norm.bias
F32
F32
[6144]
blk.75.ffn_norm.weight
F32
F32
[6144]
blk.75.ffn_up.bias
F32
F32
[24576]
blk.75.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.76
blk.76.attn_norm.bias
F32
F32
[6144]
blk.76.attn_norm.weight
F32
F32
[6144]
blk.76.attn_output.bias
F32
F32
[6144]
blk.76.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.76.attn_qkv.bias
F32
F32
[6400]
blk.76.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.76.ffn_down.bias
F32
F32
[6144]
blk.76.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.76.ffn_norm.bias
F32
F32
[6144]
blk.76.ffn_norm.weight
F32
F32
[6144]
blk.76.ffn_up.bias
F32
F32
[24576]
blk.76.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.77
blk.77.attn_norm.bias
F32
F32
[6144]
blk.77.attn_norm.weight
F32
F32
[6144]
blk.77.attn_output.bias
F32
F32
[6144]
blk.77.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.77.attn_qkv.bias
F32
F32
[6400]
blk.77.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.77.ffn_down.bias
F32
F32
[6144]
blk.77.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.77.ffn_norm.bias
F32
F32
[6144]
blk.77.ffn_norm.weight
F32
F32
[6144]
blk.77.ffn_up.bias
F32
F32
[24576]
blk.77.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.78
blk.78.attn_norm.bias
F32
F32
[6144]
blk.78.attn_norm.weight
F32
F32
[6144]
blk.78.attn_output.bias
F32
F32
[6144]
blk.78.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.78.attn_qkv.bias
F32
F32
[6400]
blk.78.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.78.ffn_down.bias
F32
F32
[6144]
blk.78.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.78.ffn_norm.bias
F32
F32
[6144]
blk.78.ffn_norm.weight
F32
F32
[6144]
blk.78.ffn_up.bias
F32
F32
[24576]
blk.78.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.79
blk.79.attn_norm.bias
F32
F32
[6144]
blk.79.attn_norm.weight
F32
F32
[6144]
blk.79.attn_output.bias
F32
F32
[6144]
blk.79.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.79.attn_qkv.bias
F32
F32
[6400]
blk.79.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.79.ffn_down.bias
F32
F32
[6144]
blk.79.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.79.ffn_norm.bias
F32
F32
[6144]
blk.79.ffn_norm.weight
F32
F32
[6144]
blk.79.ffn_up.bias
F32
F32
[24576]
blk.79.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.80
blk.80.attn_norm.bias
F32
F32
[6144]
blk.80.attn_norm.weight
F32
F32
[6144]
blk.80.attn_output.bias
F32
F32
[6144]
blk.80.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.80.attn_qkv.bias
F32
F32
[6400]
blk.80.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.80.ffn_down.bias
F32
F32
[6144]
blk.80.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.80.ffn_norm.bias
F32
F32
[6144]
blk.80.ffn_norm.weight
F32
F32
[6144]
blk.80.ffn_up.bias
F32
F32
[24576]
blk.80.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.81
blk.81.attn_norm.bias
F32
F32
[6144]
blk.81.attn_norm.weight
F32
F32
[6144]
blk.81.attn_output.bias
F32
F32
[6144]
blk.81.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.81.attn_qkv.bias
F32
F32
[6400]
blk.81.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.81.ffn_down.bias
F32
F32
[6144]
blk.81.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.81.ffn_norm.bias
F32
F32
[6144]
blk.81.ffn_norm.weight
F32
F32
[6144]
blk.81.ffn_up.bias
F32
F32
[24576]
blk.81.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.82
blk.82.attn_norm.bias
F32
F32
[6144]
blk.82.attn_norm.weight
F32
F32
[6144]
blk.82.attn_output.bias
F32
F32
[6144]
blk.82.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.82.attn_qkv.bias
F32
F32
[6400]
blk.82.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.82.ffn_down.bias
F32
F32
[6144]
blk.82.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.82.ffn_norm.bias
F32
F32
[6144]
blk.82.ffn_norm.weight
F32
F32
[6144]
blk.82.ffn_up.bias
F32
F32
[24576]
blk.82.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.83
blk.83.attn_norm.bias
F32
F32
[6144]
blk.83.attn_norm.weight
F32
F32
[6144]
blk.83.attn_output.bias
F32
F32
[6144]
blk.83.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.83.attn_qkv.bias
F32
F32
[6400]
blk.83.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.83.ffn_down.bias
F32
F32
[6144]
blk.83.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.83.ffn_norm.bias
F32
F32
[6144]
blk.83.ffn_norm.weight
F32
F32
[6144]
blk.83.ffn_up.bias
F32
F32
[24576]
blk.83.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.84
blk.84.attn_norm.bias
F32
F32
[6144]
blk.84.attn_norm.weight
F32
F32
[6144]
blk.84.attn_output.bias
F32
F32
[6144]
blk.84.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.84.attn_qkv.bias
F32
F32
[6400]
blk.84.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.84.ffn_down.bias
F32
F32
[6144]
blk.84.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.84.ffn_norm.bias
F32
F32
[6144]
blk.84.ffn_norm.weight
F32
F32
[6144]
blk.84.ffn_up.bias
F32
F32
[24576]
blk.84.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.85
blk.85.attn_norm.bias
F32
F32
[6144]
blk.85.attn_norm.weight
F32
F32
[6144]
blk.85.attn_output.bias
F32
F32
[6144]
blk.85.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.85.attn_qkv.bias
F32
F32
[6400]
blk.85.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.85.ffn_down.bias
F32
F32
[6144]
blk.85.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.85.ffn_norm.bias
F32
F32
[6144]
blk.85.ffn_norm.weight
F32
F32
[6144]
blk.85.ffn_up.bias
F32
F32
[24576]
blk.85.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.86
blk.86.attn_norm.bias
F32
F32
[6144]
blk.86.attn_norm.weight
F32
F32
[6144]
blk.86.attn_output.bias
F32
F32
[6144]
blk.86.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.86.attn_qkv.bias
F32
F32
[6400]
blk.86.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.86.ffn_down.bias
F32
F32
[6144]
blk.86.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.86.ffn_norm.bias
F32
F32
[6144]
blk.86.ffn_norm.weight
F32
F32
[6144]
blk.86.ffn_up.bias
F32
F32
[24576]
blk.86.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
blk.87
blk.87.attn_norm.bias
F32
F32
[6144]
blk.87.attn_norm.weight
F32
F32
[6144]
blk.87.attn_output.bias
F32
F32
[6144]
blk.87.attn_output.weight
Q4_0
Q4_0
[6144, 6144]
blk.87.attn_qkv.bias
F32
F32
[6400]
blk.87.attn_qkv.weight
Q4_0
Q4_0
[6144, 6400]
blk.87.ffn_down.bias
F32
F32
[6144]
blk.87.ffn_down.weight
Q4_0
Q4_0
[24576, 6144]
blk.87.ffn_norm.bias
F32
F32
[6144]
blk.87.ffn_norm.weight
F32
F32
[6144]
blk.87.ffn_up.bias
F32
F32
[24576]
blk.87.ffn_up.weight
Q4_0
Q4_0
[6144, 24576]
output_norm.bias
F32
F32
[6144]
position_embd.weight
F32
F32
[6144, 8192]
output_norm.weight
F32
F32
[6144]