Models
GitHub
Discord
Turbo
Sign in
Download
Models
Download
GitHub
Discord
Sign in
ertghiu256
/
DeepQuestioner-phi4
:latest
44
Downloads
Updated
6 months ago
From microsoft Phi 4, using system prompt to think ask question to it self
From microsoft Phi 4, using system prompt to think ask question to it self
Cancel
DeepQuestioner-phi4:latest
...
/
model
f6968ef012aa · 5.6GB
Metadata
general.architecture
llama
llama
general.file_type
Q2_K
Q2_K
llama.attention.head_count
40
40
llama.attention.head_count_kv
10
10
llama.attention.key_length
128
128
llama.attention.layer_norm_rms_epsilon
1e-05
1e-05
llama.attention.value_length
128
128
llama.block_count
40
40
llama.context_length
16384
16384
llama.embedding_length
5120
5120
llama.feed_forward_length
17920
17920
llama.rope.dimension_count
128
128
llama.rope.freq_base
250000
250000
llama.vocab_size
100352
100352
tokenizer.ggml.add_space_prefix
false
false
tokenizer.ggml.bos_token_id
100257
100257
tokenizer.ggml.eos_token_id
100265
100265
tokenizer.ggml.merges
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
100351
100351
tokenizer.ggml.pre
dbrx
dbrx
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
tokenizer.ggml.unknown_token_id
5809
5809
mradermacher.convert_type
hf
hf
mradermacher.quantize_version
2
2
mradermacher.quantized_at
2025-01-25T18:07:02+01:00
2025-01-25T18:07:02+01:00
mradermacher.quantized_by
mradermacher
mradermacher
mradermacher.quantized_on
leia
leia
Tensor
Name
Type
Shape
token_embd.weight
Q2_K
Q2_K
[5120, 100352]
blk.0
blk.0.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.0.attn_norm.weight
F32
F32
[5120]
blk.0.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.0.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.0.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.0.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.0.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.0.ffn_norm.weight
F32
F32
[5120]
blk.0.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.1
blk.1.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.1.attn_norm.weight
F32
F32
[5120]
blk.1.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.1.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.1.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.1.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.1.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.1.ffn_norm.weight
F32
F32
[5120]
blk.1.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.2
blk.2.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.2.attn_norm.weight
F32
F32
[5120]
blk.2.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.2.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.2.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.2.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.2.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.2.ffn_norm.weight
F32
F32
[5120]
blk.2.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.3
blk.3.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.3.attn_norm.weight
F32
F32
[5120]
blk.3.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.3.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.3.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.3.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.3.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.3.ffn_norm.weight
F32
F32
[5120]
blk.3.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.4
blk.4.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.4.attn_norm.weight
F32
F32
[5120]
blk.4.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.4.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.4.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.4.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.4.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.4.ffn_norm.weight
F32
F32
[5120]
blk.4.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.5
blk.5.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.5.attn_norm.weight
F32
F32
[5120]
blk.5.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.5.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.5.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.5.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.5.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.5.ffn_norm.weight
F32
F32
[5120]
blk.5.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.6
blk.6.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.6.attn_norm.weight
F32
F32
[5120]
blk.6.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.6.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.6.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.6.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.6.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.6.ffn_norm.weight
F32
F32
[5120]
blk.6.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.7
blk.7.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.7.attn_norm.weight
F32
F32
[5120]
blk.7.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.7.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.7.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.7.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.7.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.7.ffn_norm.weight
F32
F32
[5120]
blk.7.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.8
blk.8.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.8.attn_norm.weight
F32
F32
[5120]
blk.8.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.8.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.8.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.8.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.8.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.8.ffn_norm.weight
F32
F32
[5120]
blk.8.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.9
blk.9.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.9.attn_norm.weight
F32
F32
[5120]
blk.9.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.9.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.9.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.9.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.9.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.9.ffn_norm.weight
F32
F32
[5120]
blk.9.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.10
blk.10.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.10.attn_norm.weight
F32
F32
[5120]
blk.10.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.10.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.10.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.10.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.10.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.10.ffn_norm.weight
F32
F32
[5120]
blk.10.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.11
blk.11.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.11.attn_norm.weight
F32
F32
[5120]
blk.11.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.11.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.11.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.11.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.11.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.11.ffn_norm.weight
F32
F32
[5120]
blk.11.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.12
blk.12.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.12.attn_norm.weight
F32
F32
[5120]
blk.12.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.12.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.12.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.12.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.12.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.12.ffn_norm.weight
F32
F32
[5120]
blk.12.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.13
blk.13.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.13.attn_norm.weight
F32
F32
[5120]
blk.13.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.13.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.13.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.13.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.13.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.13.ffn_norm.weight
F32
F32
[5120]
blk.13.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.14
blk.14.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.14.attn_norm.weight
F32
F32
[5120]
blk.14.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.14.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.14.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.14.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.14.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.14.ffn_norm.weight
F32
F32
[5120]
blk.14.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.15
blk.15.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.15.attn_norm.weight
F32
F32
[5120]
blk.15.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.15.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.15.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.15.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.15.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.15.ffn_norm.weight
F32
F32
[5120]
blk.15.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.16
blk.16.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.16.attn_norm.weight
F32
F32
[5120]
blk.16.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.16.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.16.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.16.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.16.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.16.ffn_norm.weight
F32
F32
[5120]
blk.16.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.17
blk.17.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.17.attn_norm.weight
F32
F32
[5120]
blk.17.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.17.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.17.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.17.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.17.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.17.ffn_norm.weight
F32
F32
[5120]
blk.17.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.18
blk.18.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.18.attn_norm.weight
F32
F32
[5120]
blk.18.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.18.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.18.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.18.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.18.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.18.ffn_norm.weight
F32
F32
[5120]
blk.18.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.19
blk.19.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.19.attn_norm.weight
F32
F32
[5120]
blk.19.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.19.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.19.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.19.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.19.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.19.ffn_norm.weight
F32
F32
[5120]
blk.19.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.20
blk.20.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.20.attn_norm.weight
F32
F32
[5120]
blk.20.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.20.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.20.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.20.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.20.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.20.ffn_norm.weight
F32
F32
[5120]
blk.20.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.21
blk.21.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.21.attn_norm.weight
F32
F32
[5120]
blk.21.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.21.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.21.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.21.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.21.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.21.ffn_norm.weight
F32
F32
[5120]
blk.21.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.22
blk.22.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.22.attn_norm.weight
F32
F32
[5120]
blk.22.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.22.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.22.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.22.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.22.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.22.ffn_norm.weight
F32
F32
[5120]
blk.22.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.23
blk.23.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.23.attn_norm.weight
F32
F32
[5120]
blk.23.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.23.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.23.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.23.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.23.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.23.ffn_norm.weight
F32
F32
[5120]
blk.23.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.24
blk.24.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.24.attn_norm.weight
F32
F32
[5120]
blk.24.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.24.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.24.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.24.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.24.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.24.ffn_norm.weight
F32
F32
[5120]
blk.24.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.25
blk.25.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.25.attn_norm.weight
F32
F32
[5120]
blk.25.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.25.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.25.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.25.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.25.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.25.ffn_norm.weight
F32
F32
[5120]
blk.25.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.26
blk.26.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.26.attn_norm.weight
F32
F32
[5120]
blk.26.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.26.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.26.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.26.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.26.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.26.ffn_norm.weight
F32
F32
[5120]
blk.26.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.27
blk.27.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.27.attn_norm.weight
F32
F32
[5120]
blk.27.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.27.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.27.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.27.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.27.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.27.ffn_norm.weight
F32
F32
[5120]
blk.27.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.28
blk.28.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.28.attn_norm.weight
F32
F32
[5120]
blk.28.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.28.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.28.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.28.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.28.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.28.ffn_norm.weight
F32
F32
[5120]
blk.28.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.29
blk.29.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.29.attn_norm.weight
F32
F32
[5120]
blk.29.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.29.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.29.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.29.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.29.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.29.ffn_norm.weight
F32
F32
[5120]
blk.29.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.30
blk.30.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.30.attn_norm.weight
F32
F32
[5120]
blk.30.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.30.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.30.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.30.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.30.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.30.ffn_norm.weight
F32
F32
[5120]
blk.30.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.31
blk.31.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.31.attn_norm.weight
F32
F32
[5120]
blk.31.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.31.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.31.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.31.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.31.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.31.ffn_norm.weight
F32
F32
[5120]
blk.31.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.32
blk.32.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.32.attn_norm.weight
F32
F32
[5120]
blk.32.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.32.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.32.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.32.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.32.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.32.ffn_norm.weight
F32
F32
[5120]
blk.32.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.33
blk.33.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.33.attn_norm.weight
F32
F32
[5120]
blk.33.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.33.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.33.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.33.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.33.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.33.ffn_norm.weight
F32
F32
[5120]
blk.33.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.34
blk.34.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.34.attn_norm.weight
F32
F32
[5120]
blk.34.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.34.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.34.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.34.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.34.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.34.ffn_norm.weight
F32
F32
[5120]
blk.34.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.35
blk.35.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.35.attn_norm.weight
F32
F32
[5120]
blk.35.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.35.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.35.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.35.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.35.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.35.ffn_norm.weight
F32
F32
[5120]
blk.35.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.36
blk.36.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.36.attn_norm.weight
F32
F32
[5120]
blk.36.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.36.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.36.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.36.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.36.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.36.ffn_norm.weight
F32
F32
[5120]
blk.36.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.37
blk.37.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.37.attn_norm.weight
F32
F32
[5120]
blk.37.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.37.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.37.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.37.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.37.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.37.ffn_norm.weight
F32
F32
[5120]
blk.37.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.38
blk.38.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.38.attn_norm.weight
F32
F32
[5120]
blk.38.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.38.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.38.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.38.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.38.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.38.ffn_norm.weight
F32
F32
[5120]
blk.38.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
blk.39
blk.39.attn_k.weight
Q2_K
Q2_K
[5120, 1280]
blk.39.attn_norm.weight
F32
F32
[5120]
blk.39.attn_output.weight
Q3_K
Q3_K
[5120, 5120]
blk.39.attn_q.weight
Q2_K
Q2_K
[5120, 5120]
blk.39.attn_v.weight
Q4_K
Q4_K
[5120, 1280]
blk.39.ffn_down.weight
Q3_K
Q3_K
[17920, 5120]
blk.39.ffn_gate.weight
Q2_K
Q2_K
[5120, 17920]
blk.39.ffn_norm.weight
F32
F32
[5120]
blk.39.ffn_up.weight
Q2_K
Q2_K
[5120, 17920]
output.weight
Q6_K
Q6_K
[5120, 100352]
output_norm.weight
F32
F32
[5120]