Models
GitHub
Discord
Turbo
Sign in
Download
Models
Download
GitHub
Discord
Sign in
gabegoodhart
/
mamba2
:370m
2
Downloads
Updated
3 days ago
Cancel
370m
mamba2:370m
...
/
model
6ebabeeae00f · 225MB
Metadata
general.architecture
mamba2
mamba2
general.file_type
Q4_K_M
Q4_K_M
mamba2.attention.head_count
0
0
mamba2.attention.layer_norm_rms_epsilon
1e-05
1e-05
mamba2.block_count
48
48
mamba2.context_length
1048576
1048576
mamba2.embedding_length
1024
1024
mamba2.feed_forward_length
0
0
mamba2.ssm.conv_kernel
4
4
mamba2.ssm.group_count
1
1
mamba2.ssm.inner_size
2048
2048
mamba2.ssm.state_size
128
128
mamba2.ssm.time_step_rank
32
32
tokenizer.ggml.add_bos_token
false
false
tokenizer.ggml.add_eos_token
false
false
tokenizer.ggml.bos_token_id
0
0
tokenizer.ggml.eos_token_id
0
0
tokenizer.ggml.merges
[Ġ Ġ, Ġ t, Ġ a, h e, i n, ...]
[Ġ Ġ, Ġ t, Ġ a, h e, i n, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
0
0
tokenizer.ggml.pre
olmo
olmo
tokenizer.ggml.token_type
[3, 3, 1, 1, 1, ...]
[3, 3, 1, 1, 1, ...]
tokenizer.ggml.tokens
[<|endoftext|>, <|padding|>, !, ", #, ...]
[<|endoftext|>, <|padding|>, !, ", #, ...]
tokenizer.ggml.unknown_token_id
0
0
Tensor
Name
Type
Shape
token_embd.weight
Q6_K
Q6_K
[1024, 50288]
blk.0
blk.0.attn_norm.weight
F32
F32
[1024]
blk.0.ssm_a
F32
F32
[1, 32]
blk.0.ssm_conv1d.bias
F32
F32
[2304]
blk.0.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.0.ssm_d
F32
F32
[1, 32]
blk.0.ssm_dt.bias
F32
F32
[32]
blk.0.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.0.ssm_norm.weight
F32
F32
[2048]
blk.0.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.1
blk.1.attn_norm.weight
F32
F32
[1024]
blk.1.ssm_a
F32
F32
[1, 32]
blk.1.ssm_conv1d.bias
F32
F32
[2304]
blk.1.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.1.ssm_d
F32
F32
[1, 32]
blk.1.ssm_dt.bias
F32
F32
[32]
blk.1.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.1.ssm_norm.weight
F32
F32
[2048]
blk.1.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.2
blk.2.attn_norm.weight
F32
F32
[1024]
blk.2.ssm_a
F32
F32
[1, 32]
blk.2.ssm_conv1d.bias
F32
F32
[2304]
blk.2.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.2.ssm_d
F32
F32
[1, 32]
blk.2.ssm_dt.bias
F32
F32
[32]
blk.2.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.2.ssm_norm.weight
F32
F32
[2048]
blk.2.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.3
blk.3.attn_norm.weight
F32
F32
[1024]
blk.3.ssm_a
F32
F32
[1, 32]
blk.3.ssm_conv1d.bias
F32
F32
[2304]
blk.3.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.3.ssm_d
F32
F32
[1, 32]
blk.3.ssm_dt.bias
F32
F32
[32]
blk.3.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.3.ssm_norm.weight
F32
F32
[2048]
blk.3.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.4
blk.4.attn_norm.weight
F32
F32
[1024]
blk.4.ssm_a
F32
F32
[1, 32]
blk.4.ssm_conv1d.bias
F32
F32
[2304]
blk.4.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.4.ssm_d
F32
F32
[1, 32]
blk.4.ssm_dt.bias
F32
F32
[32]
blk.4.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.4.ssm_norm.weight
F32
F32
[2048]
blk.4.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.5
blk.5.attn_norm.weight
F32
F32
[1024]
blk.5.ssm_a
F32
F32
[1, 32]
blk.5.ssm_conv1d.bias
F32
F32
[2304]
blk.5.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.5.ssm_d
F32
F32
[1, 32]
blk.5.ssm_dt.bias
F32
F32
[32]
blk.5.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.5.ssm_norm.weight
F32
F32
[2048]
blk.5.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.6
blk.6.attn_norm.weight
F32
F32
[1024]
blk.6.ssm_a
F32
F32
[1, 32]
blk.6.ssm_conv1d.bias
F32
F32
[2304]
blk.6.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.6.ssm_d
F32
F32
[1, 32]
blk.6.ssm_dt.bias
F32
F32
[32]
blk.6.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.6.ssm_norm.weight
F32
F32
[2048]
blk.6.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.7
blk.7.attn_norm.weight
F32
F32
[1024]
blk.7.ssm_a
F32
F32
[1, 32]
blk.7.ssm_conv1d.bias
F32
F32
[2304]
blk.7.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.7.ssm_d
F32
F32
[1, 32]
blk.7.ssm_dt.bias
F32
F32
[32]
blk.7.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.7.ssm_norm.weight
F32
F32
[2048]
blk.7.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.8
blk.8.attn_norm.weight
F32
F32
[1024]
blk.8.ssm_a
F32
F32
[1, 32]
blk.8.ssm_conv1d.bias
F32
F32
[2304]
blk.8.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.8.ssm_d
F32
F32
[1, 32]
blk.8.ssm_dt.bias
F32
F32
[32]
blk.8.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.8.ssm_norm.weight
F32
F32
[2048]
blk.8.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.9
blk.9.attn_norm.weight
F32
F32
[1024]
blk.9.ssm_a
F32
F32
[1, 32]
blk.9.ssm_conv1d.bias
F32
F32
[2304]
blk.9.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.9.ssm_d
F32
F32
[1, 32]
blk.9.ssm_dt.bias
F32
F32
[32]
blk.9.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.9.ssm_norm.weight
F32
F32
[2048]
blk.9.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.10
blk.10.attn_norm.weight
F32
F32
[1024]
blk.10.ssm_a
F32
F32
[1, 32]
blk.10.ssm_conv1d.bias
F32
F32
[2304]
blk.10.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.10.ssm_d
F32
F32
[1, 32]
blk.10.ssm_dt.bias
F32
F32
[32]
blk.10.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.10.ssm_norm.weight
F32
F32
[2048]
blk.10.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.11
blk.11.attn_norm.weight
F32
F32
[1024]
blk.11.ssm_a
F32
F32
[1, 32]
blk.11.ssm_conv1d.bias
F32
F32
[2304]
blk.11.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.11.ssm_d
F32
F32
[1, 32]
blk.11.ssm_dt.bias
F32
F32
[32]
blk.11.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.11.ssm_norm.weight
F32
F32
[2048]
blk.11.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.12
blk.12.attn_norm.weight
F32
F32
[1024]
blk.12.ssm_a
F32
F32
[1, 32]
blk.12.ssm_conv1d.bias
F32
F32
[2304]
blk.12.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.12.ssm_d
F32
F32
[1, 32]
blk.12.ssm_dt.bias
F32
F32
[32]
blk.12.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.12.ssm_norm.weight
F32
F32
[2048]
blk.12.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.13
blk.13.attn_norm.weight
F32
F32
[1024]
blk.13.ssm_a
F32
F32
[1, 32]
blk.13.ssm_conv1d.bias
F32
F32
[2304]
blk.13.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.13.ssm_d
F32
F32
[1, 32]
blk.13.ssm_dt.bias
F32
F32
[32]
blk.13.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.13.ssm_norm.weight
F32
F32
[2048]
blk.13.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.14
blk.14.attn_norm.weight
F32
F32
[1024]
blk.14.ssm_a
F32
F32
[1, 32]
blk.14.ssm_conv1d.bias
F32
F32
[2304]
blk.14.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.14.ssm_d
F32
F32
[1, 32]
blk.14.ssm_dt.bias
F32
F32
[32]
blk.14.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.14.ssm_norm.weight
F32
F32
[2048]
blk.14.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.15
blk.15.attn_norm.weight
F32
F32
[1024]
blk.15.ssm_a
F32
F32
[1, 32]
blk.15.ssm_conv1d.bias
F32
F32
[2304]
blk.15.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.15.ssm_d
F32
F32
[1, 32]
blk.15.ssm_dt.bias
F32
F32
[32]
blk.15.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.15.ssm_norm.weight
F32
F32
[2048]
blk.15.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.16
blk.16.attn_norm.weight
F32
F32
[1024]
blk.16.ssm_a
F32
F32
[1, 32]
blk.16.ssm_conv1d.bias
F32
F32
[2304]
blk.16.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.16.ssm_d
F32
F32
[1, 32]
blk.16.ssm_dt.bias
F32
F32
[32]
blk.16.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.16.ssm_norm.weight
F32
F32
[2048]
blk.16.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.17
blk.17.attn_norm.weight
F32
F32
[1024]
blk.17.ssm_a
F32
F32
[1, 32]
blk.17.ssm_conv1d.bias
F32
F32
[2304]
blk.17.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.17.ssm_d
F32
F32
[1, 32]
blk.17.ssm_dt.bias
F32
F32
[32]
blk.17.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.17.ssm_norm.weight
F32
F32
[2048]
blk.17.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.18
blk.18.attn_norm.weight
F32
F32
[1024]
blk.18.ssm_a
F32
F32
[1, 32]
blk.18.ssm_conv1d.bias
F32
F32
[2304]
blk.18.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.18.ssm_d
F32
F32
[1, 32]
blk.18.ssm_dt.bias
F32
F32
[32]
blk.18.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.18.ssm_norm.weight
F32
F32
[2048]
blk.18.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.19
blk.19.attn_norm.weight
F32
F32
[1024]
blk.19.ssm_a
F32
F32
[1, 32]
blk.19.ssm_conv1d.bias
F32
F32
[2304]
blk.19.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.19.ssm_d
F32
F32
[1, 32]
blk.19.ssm_dt.bias
F32
F32
[32]
blk.19.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.19.ssm_norm.weight
F32
F32
[2048]
blk.19.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.20
blk.20.attn_norm.weight
F32
F32
[1024]
blk.20.ssm_a
F32
F32
[1, 32]
blk.20.ssm_conv1d.bias
F32
F32
[2304]
blk.20.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.20.ssm_d
F32
F32
[1, 32]
blk.20.ssm_dt.bias
F32
F32
[32]
blk.20.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.20.ssm_norm.weight
F32
F32
[2048]
blk.20.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.21
blk.21.attn_norm.weight
F32
F32
[1024]
blk.21.ssm_a
F32
F32
[1, 32]
blk.21.ssm_conv1d.bias
F32
F32
[2304]
blk.21.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.21.ssm_d
F32
F32
[1, 32]
blk.21.ssm_dt.bias
F32
F32
[32]
blk.21.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.21.ssm_norm.weight
F32
F32
[2048]
blk.21.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.22
blk.22.attn_norm.weight
F32
F32
[1024]
blk.22.ssm_a
F32
F32
[1, 32]
blk.22.ssm_conv1d.bias
F32
F32
[2304]
blk.22.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.22.ssm_d
F32
F32
[1, 32]
blk.22.ssm_dt.bias
F32
F32
[32]
blk.22.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.22.ssm_norm.weight
F32
F32
[2048]
blk.22.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.23
blk.23.attn_norm.weight
F32
F32
[1024]
blk.23.ssm_a
F32
F32
[1, 32]
blk.23.ssm_conv1d.bias
F32
F32
[2304]
blk.23.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.23.ssm_d
F32
F32
[1, 32]
blk.23.ssm_dt.bias
F32
F32
[32]
blk.23.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.23.ssm_norm.weight
F32
F32
[2048]
blk.23.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.24
blk.24.attn_norm.weight
F32
F32
[1024]
blk.24.ssm_a
F32
F32
[1, 32]
blk.24.ssm_conv1d.bias
F32
F32
[2304]
blk.24.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.24.ssm_d
F32
F32
[1, 32]
blk.24.ssm_dt.bias
F32
F32
[32]
blk.24.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.24.ssm_norm.weight
F32
F32
[2048]
blk.24.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.25
blk.25.attn_norm.weight
F32
F32
[1024]
blk.25.ssm_a
F32
F32
[1, 32]
blk.25.ssm_conv1d.bias
F32
F32
[2304]
blk.25.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.25.ssm_d
F32
F32
[1, 32]
blk.25.ssm_dt.bias
F32
F32
[32]
blk.25.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.25.ssm_norm.weight
F32
F32
[2048]
blk.25.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.26
blk.26.attn_norm.weight
F32
F32
[1024]
blk.26.ssm_a
F32
F32
[1, 32]
blk.26.ssm_conv1d.bias
F32
F32
[2304]
blk.26.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.26.ssm_d
F32
F32
[1, 32]
blk.26.ssm_dt.bias
F32
F32
[32]
blk.26.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.26.ssm_norm.weight
F32
F32
[2048]
blk.26.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.27
blk.27.attn_norm.weight
F32
F32
[1024]
blk.27.ssm_a
F32
F32
[1, 32]
blk.27.ssm_conv1d.bias
F32
F32
[2304]
blk.27.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.27.ssm_d
F32
F32
[1, 32]
blk.27.ssm_dt.bias
F32
F32
[32]
blk.27.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.27.ssm_norm.weight
F32
F32
[2048]
blk.27.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.28
blk.28.attn_norm.weight
F32
F32
[1024]
blk.28.ssm_a
F32
F32
[1, 32]
blk.28.ssm_conv1d.bias
F32
F32
[2304]
blk.28.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.28.ssm_d
F32
F32
[1, 32]
blk.28.ssm_dt.bias
F32
F32
[32]
blk.28.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.28.ssm_norm.weight
F32
F32
[2048]
blk.28.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.29
blk.29.attn_norm.weight
F32
F32
[1024]
blk.29.ssm_a
F32
F32
[1, 32]
blk.29.ssm_conv1d.bias
F32
F32
[2304]
blk.29.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.29.ssm_d
F32
F32
[1, 32]
blk.29.ssm_dt.bias
F32
F32
[32]
blk.29.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.29.ssm_norm.weight
F32
F32
[2048]
blk.29.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.30
blk.30.attn_norm.weight
F32
F32
[1024]
blk.30.ssm_a
F32
F32
[1, 32]
blk.30.ssm_conv1d.bias
F32
F32
[2304]
blk.30.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.30.ssm_d
F32
F32
[1, 32]
blk.30.ssm_dt.bias
F32
F32
[32]
blk.30.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.30.ssm_norm.weight
F32
F32
[2048]
blk.30.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.31
blk.31.attn_norm.weight
F32
F32
[1024]
blk.31.ssm_a
F32
F32
[1, 32]
blk.31.ssm_conv1d.bias
F32
F32
[2304]
blk.31.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.31.ssm_d
F32
F32
[1, 32]
blk.31.ssm_dt.bias
F32
F32
[32]
blk.31.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.31.ssm_norm.weight
F32
F32
[2048]
blk.31.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.32
blk.32.attn_norm.weight
F32
F32
[1024]
blk.32.ssm_a
F32
F32
[1, 32]
blk.32.ssm_conv1d.bias
F32
F32
[2304]
blk.32.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.32.ssm_d
F32
F32
[1, 32]
blk.32.ssm_dt.bias
F32
F32
[32]
blk.32.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.32.ssm_norm.weight
F32
F32
[2048]
blk.32.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.33
blk.33.attn_norm.weight
F32
F32
[1024]
blk.33.ssm_a
F32
F32
[1, 32]
blk.33.ssm_conv1d.bias
F32
F32
[2304]
blk.33.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.33.ssm_d
F32
F32
[1, 32]
blk.33.ssm_dt.bias
F32
F32
[32]
blk.33.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.33.ssm_norm.weight
F32
F32
[2048]
blk.33.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.34
blk.34.attn_norm.weight
F32
F32
[1024]
blk.34.ssm_a
F32
F32
[1, 32]
blk.34.ssm_conv1d.bias
F32
F32
[2304]
blk.34.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.34.ssm_d
F32
F32
[1, 32]
blk.34.ssm_dt.bias
F32
F32
[32]
blk.34.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.34.ssm_norm.weight
F32
F32
[2048]
blk.34.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.35
blk.35.attn_norm.weight
F32
F32
[1024]
blk.35.ssm_a
F32
F32
[1, 32]
blk.35.ssm_conv1d.bias
F32
F32
[2304]
blk.35.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.35.ssm_d
F32
F32
[1, 32]
blk.35.ssm_dt.bias
F32
F32
[32]
blk.35.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.35.ssm_norm.weight
F32
F32
[2048]
blk.35.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.36
blk.36.attn_norm.weight
F32
F32
[1024]
blk.36.ssm_a
F32
F32
[1, 32]
blk.36.ssm_conv1d.bias
F32
F32
[2304]
blk.36.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.36.ssm_d
F32
F32
[1, 32]
blk.36.ssm_dt.bias
F32
F32
[32]
blk.36.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.36.ssm_norm.weight
F32
F32
[2048]
blk.36.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.37
blk.37.attn_norm.weight
F32
F32
[1024]
blk.37.ssm_a
F32
F32
[1, 32]
blk.37.ssm_conv1d.bias
F32
F32
[2304]
blk.37.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.37.ssm_d
F32
F32
[1, 32]
blk.37.ssm_dt.bias
F32
F32
[32]
blk.37.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.37.ssm_norm.weight
F32
F32
[2048]
blk.37.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.38
blk.38.attn_norm.weight
F32
F32
[1024]
blk.38.ssm_a
F32
F32
[1, 32]
blk.38.ssm_conv1d.bias
F32
F32
[2304]
blk.38.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.38.ssm_d
F32
F32
[1, 32]
blk.38.ssm_dt.bias
F32
F32
[32]
blk.38.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.38.ssm_norm.weight
F32
F32
[2048]
blk.38.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.39
blk.39.attn_norm.weight
F32
F32
[1024]
blk.39.ssm_a
F32
F32
[1, 32]
blk.39.ssm_conv1d.bias
F32
F32
[2304]
blk.39.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.39.ssm_d
F32
F32
[1, 32]
blk.39.ssm_dt.bias
F32
F32
[32]
blk.39.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.39.ssm_norm.weight
F32
F32
[2048]
blk.39.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.40
blk.40.attn_norm.weight
F32
F32
[1024]
blk.40.ssm_a
F32
F32
[1, 32]
blk.40.ssm_conv1d.bias
F32
F32
[2304]
blk.40.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.40.ssm_d
F32
F32
[1, 32]
blk.40.ssm_dt.bias
F32
F32
[32]
blk.40.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.40.ssm_norm.weight
F32
F32
[2048]
blk.40.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.41
blk.41.attn_norm.weight
F32
F32
[1024]
blk.41.ssm_a
F32
F32
[1, 32]
blk.41.ssm_conv1d.bias
F32
F32
[2304]
blk.41.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.41.ssm_d
F32
F32
[1, 32]
blk.41.ssm_dt.bias
F32
F32
[32]
blk.41.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.41.ssm_norm.weight
F32
F32
[2048]
blk.41.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.42
blk.42.attn_norm.weight
F32
F32
[1024]
blk.42.ssm_a
F32
F32
[1, 32]
blk.42.ssm_conv1d.bias
F32
F32
[2304]
blk.42.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.42.ssm_d
F32
F32
[1, 32]
blk.42.ssm_dt.bias
F32
F32
[32]
blk.42.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.42.ssm_norm.weight
F32
F32
[2048]
blk.42.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.43
blk.43.attn_norm.weight
F32
F32
[1024]
blk.43.ssm_a
F32
F32
[1, 32]
blk.43.ssm_conv1d.bias
F32
F32
[2304]
blk.43.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.43.ssm_d
F32
F32
[1, 32]
blk.43.ssm_dt.bias
F32
F32
[32]
blk.43.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.43.ssm_norm.weight
F32
F32
[2048]
blk.43.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.44
blk.44.attn_norm.weight
F32
F32
[1024]
blk.44.ssm_a
F32
F32
[1, 32]
blk.44.ssm_conv1d.bias
F32
F32
[2304]
blk.44.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.44.ssm_d
F32
F32
[1, 32]
blk.44.ssm_dt.bias
F32
F32
[32]
blk.44.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.44.ssm_norm.weight
F32
F32
[2048]
blk.44.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.45
blk.45.attn_norm.weight
F32
F32
[1024]
blk.45.ssm_a
F32
F32
[1, 32]
blk.45.ssm_conv1d.bias
F32
F32
[2304]
blk.45.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.45.ssm_d
F32
F32
[1, 32]
blk.45.ssm_dt.bias
F32
F32
[32]
blk.45.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.45.ssm_norm.weight
F32
F32
[2048]
blk.45.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.46
blk.46.attn_norm.weight
F32
F32
[1024]
blk.46.ssm_a
F32
F32
[1, 32]
blk.46.ssm_conv1d.bias
F32
F32
[2304]
blk.46.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.46.ssm_d
F32
F32
[1, 32]
blk.46.ssm_dt.bias
F32
F32
[32]
blk.46.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.46.ssm_norm.weight
F32
F32
[2048]
blk.46.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
blk.47
blk.47.attn_norm.weight
F32
F32
[1024]
blk.47.ssm_a
F32
F32
[1, 32]
blk.47.ssm_conv1d.bias
F32
F32
[2304]
blk.47.ssm_conv1d.weight
F32
F32
[4, 2304]
blk.47.ssm_d
F32
F32
[1, 32]
blk.47.ssm_dt.bias
F32
F32
[32]
blk.47.ssm_in.weight
Q4_K
Q4_K
[1024, 4384]
blk.47.ssm_norm.weight
F32
F32
[2048]
blk.47.ssm_out.weight
Q4_K
Q4_K
[2048, 1024]
output_norm.weight
F32
F32
[1024]