Models
GitHub
Discord
Turbo
Sign in
Download
Models
Download
GitHub
Discord
Sign in
alibayram
/
hunyuan
:4b
2,158
Downloads
Updated
1 month ago
Hunyuan is Tencent’s open-source efficient large language model series, designed for versatile deployment across diverse computational environments.
Hunyuan is Tencent’s open-source efficient large language model series, designed for versatile deployment across diverse computational environments.
Cancel
0.5b
1.8b
4b
7b
hunyuan:4b
...
/
model
640d2e889061 · 2.6GB
Metadata
general.architecture
hunyuan-dense
hunyuan-dense
general.file_type
Q4_K_M
Q4_K_M
hunyuan-dense.attention.head_count
32
32
hunyuan-dense.attention.head_count_kv
8
8
hunyuan-dense.attention.key_length
128
128
hunyuan-dense.attention.layer_norm_rms_epsilon
1e-05
1e-05
hunyuan-dense.attention.value_length
128
128
hunyuan-dense.block_count
36
36
hunyuan-dense.context_length
262144
262144
hunyuan-dense.embedding_length
3072
3072
hunyuan-dense.feed_forward_length
8192
8192
hunyuan-dense.rope.freq_base
1.115884e+07
1.115884e+07
hunyuan-dense.rope.scaling.factor
1
1
hunyuan-dense.rope.scaling.original_context_length
262144
262144
hunyuan-dense.rope.scaling.type
none
none
tokenizer.ggml.bos_token_id
120000
120000
tokenizer.ggml.eos_token_id
120020
120020
tokenizer.ggml.merges
[Ġ Ġ, Ġ t, Ġ a, i n, h e, ...]
[Ġ Ġ, Ġ t, Ġ a, i n, h e, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
120002
120002
tokenizer.ggml.pre
hunyuan-dense
hunyuan-dense
tokenizer.ggml.seperator_token_id
120007
120007
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
quantize.imatrix.chunks_count
125
125
quantize.imatrix.dataset
calibration_datav3.txt
calibration_datav3.txt
quantize.imatrix.entries_count
252
252
quantize.imatrix.file
Hunyuan-4B-Instruct/Hunyuan-4B-Instruct.dat
Hunyuan-4B-Instruct/Hunyuan-4B-Instruct.dat
Tensor
Name
Type
Shape
token_embd.weight
Q6_K
Q6_K
[3072, 120818]
blk.0
blk.0.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.0.attn_k_norm.weight
F32
F32
[128]
blk.0.attn_norm.weight
F32
F32
[3072]
blk.0.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.0.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.0.attn_q_norm.weight
F32
F32
[128]
blk.0.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.0.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.0.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.0.ffn_norm.weight
F32
F32
[3072]
blk.0.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.1
blk.1.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.1.attn_k_norm.weight
F32
F32
[128]
blk.1.attn_norm.weight
F32
F32
[3072]
blk.1.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.1.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.1.attn_q_norm.weight
F32
F32
[128]
blk.1.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.1.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.1.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.1.ffn_norm.weight
F32
F32
[3072]
blk.1.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.2
blk.2.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.2.attn_k_norm.weight
F32
F32
[128]
blk.2.attn_norm.weight
F32
F32
[3072]
blk.2.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.2.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.2.attn_q_norm.weight
F32
F32
[128]
blk.2.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.2.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.2.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.2.ffn_norm.weight
F32
F32
[3072]
blk.2.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.3
blk.3.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.3.attn_k_norm.weight
F32
F32
[128]
blk.3.attn_norm.weight
F32
F32
[3072]
blk.3.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.3.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.3.attn_q_norm.weight
F32
F32
[128]
blk.3.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.3.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.3.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.3.ffn_norm.weight
F32
F32
[3072]
blk.3.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.4
blk.4.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.4.attn_k_norm.weight
F32
F32
[128]
blk.4.attn_norm.weight
F32
F32
[3072]
blk.4.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.4.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.4.attn_q_norm.weight
F32
F32
[128]
blk.4.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.4.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.4.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.4.ffn_norm.weight
F32
F32
[3072]
blk.4.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.5
blk.5.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.5.attn_k_norm.weight
F32
F32
[128]
blk.5.attn_norm.weight
F32
F32
[3072]
blk.5.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.5.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.5.attn_q_norm.weight
F32
F32
[128]
blk.5.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.5.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.5.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.5.ffn_norm.weight
F32
F32
[3072]
blk.5.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.6
blk.6.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.6.attn_k_norm.weight
F32
F32
[128]
blk.6.attn_norm.weight
F32
F32
[3072]
blk.6.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.6.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.6.attn_q_norm.weight
F32
F32
[128]
blk.6.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.6.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.6.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.6.ffn_norm.weight
F32
F32
[3072]
blk.6.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.7
blk.7.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.7.attn_k_norm.weight
F32
F32
[128]
blk.7.attn_norm.weight
F32
F32
[3072]
blk.7.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.7.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.7.attn_q_norm.weight
F32
F32
[128]
blk.7.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.7.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.7.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.7.ffn_norm.weight
F32
F32
[3072]
blk.7.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.8
blk.8.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.8.attn_k_norm.weight
F32
F32
[128]
blk.8.attn_norm.weight
F32
F32
[3072]
blk.8.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.8.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.8.attn_q_norm.weight
F32
F32
[128]
blk.8.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.8.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.8.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.8.ffn_norm.weight
F32
F32
[3072]
blk.8.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.9
blk.9.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.9.attn_k_norm.weight
F32
F32
[128]
blk.9.attn_norm.weight
F32
F32
[3072]
blk.9.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.9.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.9.attn_q_norm.weight
F32
F32
[128]
blk.9.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.9.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.9.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.9.ffn_norm.weight
F32
F32
[3072]
blk.9.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.10
blk.10.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.10.attn_k_norm.weight
F32
F32
[128]
blk.10.attn_norm.weight
F32
F32
[3072]
blk.10.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.10.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.10.attn_q_norm.weight
F32
F32
[128]
blk.10.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.10.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.10.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.10.ffn_norm.weight
F32
F32
[3072]
blk.10.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.11
blk.11.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.11.attn_k_norm.weight
F32
F32
[128]
blk.11.attn_norm.weight
F32
F32
[3072]
blk.11.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.11.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.11.attn_q_norm.weight
F32
F32
[128]
blk.11.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.11.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.11.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.11.ffn_norm.weight
F32
F32
[3072]
blk.11.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.12
blk.12.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.12.attn_k_norm.weight
F32
F32
[128]
blk.12.attn_norm.weight
F32
F32
[3072]
blk.12.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.12.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.12.attn_q_norm.weight
F32
F32
[128]
blk.12.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.12.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.12.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.12.ffn_norm.weight
F32
F32
[3072]
blk.12.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.13
blk.13.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.13.attn_k_norm.weight
F32
F32
[128]
blk.13.attn_norm.weight
F32
F32
[3072]
blk.13.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.13.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.13.attn_q_norm.weight
F32
F32
[128]
blk.13.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.13.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.13.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.13.ffn_norm.weight
F32
F32
[3072]
blk.13.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.14
blk.14.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.14.attn_k_norm.weight
F32
F32
[128]
blk.14.attn_norm.weight
F32
F32
[3072]
blk.14.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.14.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.14.attn_q_norm.weight
F32
F32
[128]
blk.14.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.14.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.14.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.14.ffn_norm.weight
F32
F32
[3072]
blk.14.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.15
blk.15.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.15.attn_k_norm.weight
F32
F32
[128]
blk.15.attn_norm.weight
F32
F32
[3072]
blk.15.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.15.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.15.attn_q_norm.weight
F32
F32
[128]
blk.15.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.15.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.15.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.15.ffn_norm.weight
F32
F32
[3072]
blk.15.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.16
blk.16.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.16.attn_k_norm.weight
F32
F32
[128]
blk.16.attn_norm.weight
F32
F32
[3072]
blk.16.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.16.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.16.attn_q_norm.weight
F32
F32
[128]
blk.16.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.16.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.16.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.16.ffn_norm.weight
F32
F32
[3072]
blk.16.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.17
blk.17.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.17.attn_k_norm.weight
F32
F32
[128]
blk.17.attn_norm.weight
F32
F32
[3072]
blk.17.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.17.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.17.attn_q_norm.weight
F32
F32
[128]
blk.17.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.17.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.17.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.17.ffn_norm.weight
F32
F32
[3072]
blk.17.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.18
blk.18.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.18.attn_k_norm.weight
F32
F32
[128]
blk.18.attn_norm.weight
F32
F32
[3072]
blk.18.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.18.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.18.attn_q_norm.weight
F32
F32
[128]
blk.18.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.18.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.18.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.18.ffn_norm.weight
F32
F32
[3072]
blk.18.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.19
blk.19.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.19.attn_k_norm.weight
F32
F32
[128]
blk.19.attn_norm.weight
F32
F32
[3072]
blk.19.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.19.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.19.attn_q_norm.weight
F32
F32
[128]
blk.19.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.19.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.19.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.19.ffn_norm.weight
F32
F32
[3072]
blk.19.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.20
blk.20.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.20.attn_k_norm.weight
F32
F32
[128]
blk.20.attn_norm.weight
F32
F32
[3072]
blk.20.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.20.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.20.attn_q_norm.weight
F32
F32
[128]
blk.20.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.20.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.20.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.20.ffn_norm.weight
F32
F32
[3072]
blk.20.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.21
blk.21.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.21.attn_k_norm.weight
F32
F32
[128]
blk.21.attn_norm.weight
F32
F32
[3072]
blk.21.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.21.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.21.attn_q_norm.weight
F32
F32
[128]
blk.21.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.21.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.21.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.21.ffn_norm.weight
F32
F32
[3072]
blk.21.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.22
blk.22.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.22.attn_k_norm.weight
F32
F32
[128]
blk.22.attn_norm.weight
F32
F32
[3072]
blk.22.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.22.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.22.attn_q_norm.weight
F32
F32
[128]
blk.22.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.22.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.22.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.22.ffn_norm.weight
F32
F32
[3072]
blk.22.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.23
blk.23.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.23.attn_k_norm.weight
F32
F32
[128]
blk.23.attn_norm.weight
F32
F32
[3072]
blk.23.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.23.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.23.attn_q_norm.weight
F32
F32
[128]
blk.23.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.23.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.23.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.23.ffn_norm.weight
F32
F32
[3072]
blk.23.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.24
blk.24.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.24.attn_k_norm.weight
F32
F32
[128]
blk.24.attn_norm.weight
F32
F32
[3072]
blk.24.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.24.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.24.attn_q_norm.weight
F32
F32
[128]
blk.24.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.24.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.24.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.24.ffn_norm.weight
F32
F32
[3072]
blk.24.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.25
blk.25.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.25.attn_k_norm.weight
F32
F32
[128]
blk.25.attn_norm.weight
F32
F32
[3072]
blk.25.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.25.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.25.attn_q_norm.weight
F32
F32
[128]
blk.25.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.25.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.25.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.25.ffn_norm.weight
F32
F32
[3072]
blk.25.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.26
blk.26.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.26.attn_k_norm.weight
F32
F32
[128]
blk.26.attn_norm.weight
F32
F32
[3072]
blk.26.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.26.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.26.attn_q_norm.weight
F32
F32
[128]
blk.26.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.26.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.26.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.26.ffn_norm.weight
F32
F32
[3072]
blk.26.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.27
blk.27.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.27.attn_k_norm.weight
F32
F32
[128]
blk.27.attn_norm.weight
F32
F32
[3072]
blk.27.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.27.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.27.attn_q_norm.weight
F32
F32
[128]
blk.27.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.27.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.27.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.27.ffn_norm.weight
F32
F32
[3072]
blk.27.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.28
blk.28.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.28.attn_k_norm.weight
F32
F32
[128]
blk.28.attn_norm.weight
F32
F32
[3072]
blk.28.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.28.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.28.attn_q_norm.weight
F32
F32
[128]
blk.28.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.28.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.28.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.28.ffn_norm.weight
F32
F32
[3072]
blk.28.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.29
blk.29.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.29.attn_k_norm.weight
F32
F32
[128]
blk.29.attn_norm.weight
F32
F32
[3072]
blk.29.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.29.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.29.attn_q_norm.weight
F32
F32
[128]
blk.29.attn_v.weight
Q4_K
Q4_K
[3072, 1024]
blk.29.ffn_down.weight
Q4_K
Q4_K
[8192, 3072]
blk.29.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.29.ffn_norm.weight
F32
F32
[3072]
blk.29.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.30
blk.30.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.30.attn_k_norm.weight
F32
F32
[128]
blk.30.attn_norm.weight
F32
F32
[3072]
blk.30.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.30.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.30.attn_q_norm.weight
F32
F32
[128]
blk.30.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.30.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.30.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.30.ffn_norm.weight
F32
F32
[3072]
blk.30.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.31
blk.31.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.31.attn_k_norm.weight
F32
F32
[128]
blk.31.attn_norm.weight
F32
F32
[3072]
blk.31.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.31.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.31.attn_q_norm.weight
F32
F32
[128]
blk.31.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.31.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.31.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.31.ffn_norm.weight
F32
F32
[3072]
blk.31.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.32
blk.32.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.32.attn_k_norm.weight
F32
F32
[128]
blk.32.attn_norm.weight
F32
F32
[3072]
blk.32.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.32.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.32.attn_q_norm.weight
F32
F32
[128]
blk.32.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.32.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.32.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.32.ffn_norm.weight
F32
F32
[3072]
blk.32.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.33
blk.33.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.33.attn_k_norm.weight
F32
F32
[128]
blk.33.attn_norm.weight
F32
F32
[3072]
blk.33.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.33.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.33.attn_q_norm.weight
F32
F32
[128]
blk.33.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.33.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.33.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.33.ffn_norm.weight
F32
F32
[3072]
blk.33.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.34
blk.34.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.34.attn_k_norm.weight
F32
F32
[128]
blk.34.attn_norm.weight
F32
F32
[3072]
blk.34.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.34.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.34.attn_q_norm.weight
F32
F32
[128]
blk.34.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.34.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.34.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.34.ffn_norm.weight
F32
F32
[3072]
blk.34.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
blk.35
blk.35.attn_k.weight
Q4_K
Q4_K
[3072, 1024]
blk.35.attn_k_norm.weight
F32
F32
[128]
blk.35.attn_norm.weight
F32
F32
[3072]
blk.35.attn_output.weight
Q4_K
Q4_K
[4096, 3072]
blk.35.attn_q.weight
Q4_K
Q4_K
[3072, 4096]
blk.35.attn_q_norm.weight
F32
F32
[128]
blk.35.attn_v.weight
Q6_K
Q6_K
[3072, 1024]
blk.35.ffn_down.weight
Q6_K
Q6_K
[8192, 3072]
blk.35.ffn_gate.weight
Q4_K
Q4_K
[3072, 8192]
blk.35.ffn_norm.weight
F32
F32
[3072]
blk.35.ffn_up.weight
Q4_K
Q4_K
[3072, 8192]
output_norm.weight
F32
F32
[3072]