Models
Docs
Pricing
Sign in
Download
Models
Download
Docs
Pricing
Sign in
heredos
/
rwkv7
:1.5b
151
Downloads
Updated
3 months ago
an rwkv repo that tries to be less confusing and more up to date with the latest GGUFs
an rwkv repo that tries to be less confusing and more up to date with the latest GGUFs
Cancel
thinking
0.1b
0.4b
1.5b
2.9b
7.2b
rwkv7:1.5b
...
/
model
3a6320f3d2ae · 1.3GB
Metadata
general.architecture
rwkv7
rwkv7
general.file_type
Q6_K
Q6_K
rwkv7.attention.decay_lora_rank
96
96
rwkv7.attention.gate_lora_rank
256
256
rwkv7.attention.head_count
0
0
rwkv7.attention.iclr_lora_rank
96
96
rwkv7.attention.layer_norm_epsilon
1e-05
1e-05
rwkv7.attention.value_residual_mix_lora_rank
64
64
rwkv7.block_count
24
24
rwkv7.context_length
1048576
1048576
rwkv7.embedding_length
2048
2048
rwkv7.feed_forward_length
8192
8192
rwkv7.wkv.head_size
64
64
tokenizer.ggml.bos_token_id
0
0
tokenizer.ggml.eos_token_id
0
0
tokenizer.ggml.eot_token_id
261
261
tokenizer.ggml.model
rwkv
rwkv
tokenizer.ggml.token_type
[3, 1, 1, 1, 1, ...]
[3, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[<s>, \x00, \x01, \x02, \x03, ...]
[<s>, \x00, \x01, \x02, \x03, ...]
Tensor
Name
Type
Shape
token_embd.weight
Q6_K
Q6_K
[2048, 65536]
blk.0
blk.0.attn_norm.bias
F32
F32
[2048]
blk.0.attn_norm.weight
F32
F32
[2048]
blk.0.attn_norm_2.bias
F32
F32
[2048]
blk.0.attn_norm_2.weight
F32
F32
[2048]
blk.0.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.0.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.0.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.0.time_mix_a0.weight
F32
F32
[2048]
blk.0.time_mix_a1.weight
F16
F16
[2048, 96]
blk.0.time_mix_a2.weight
F16
F16
[96, 2048]
blk.0.time_mix_g1.weight
F16
F16
[2048, 256]
blk.0.time_mix_g2.weight
F16
F16
[256, 2048]
blk.0.time_mix_k_a.weight
F32
F32
[2048]
blk.0.time_mix_k_k.weight
F32
F32
[2048]
blk.0.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.0.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.0.time_mix_ln.bias
F32
F32
[2048]
blk.0.time_mix_ln.weight
F32
F32
[2048]
blk.0.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.0.time_mix_r_k.weight
F32
F32
[2048]
blk.0.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.0.time_mix_v0.weight
F32
F32
[2048]
blk.0.time_mix_v1.weight
F16
F16
[2048, 96]
blk.0.time_mix_v2.weight
F16
F16
[96, 2048]
blk.0.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.0.time_mix_w0.weight
F32
F32
[2048]
blk.0.time_mix_w1.weight
F32
F32
[2048, 96]
blk.0.time_mix_w2.weight
F32
F32
[96, 2048]
blk.1
blk.1.attn_norm.bias
F32
F32
[2048]
blk.1.attn_norm.weight
F32
F32
[2048]
blk.1.attn_norm_2.bias
F32
F32
[2048]
blk.1.attn_norm_2.weight
F32
F32
[2048]
blk.1.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.1.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.1.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.1.time_mix_a0.weight
F32
F32
[2048]
blk.1.time_mix_a1.weight
F16
F16
[2048, 96]
blk.1.time_mix_a2.weight
F16
F16
[96, 2048]
blk.1.time_mix_g1.weight
F16
F16
[2048, 256]
blk.1.time_mix_g2.weight
F16
F16
[256, 2048]
blk.1.time_mix_k_a.weight
F32
F32
[2048]
blk.1.time_mix_k_k.weight
F32
F32
[2048]
blk.1.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.1.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.1.time_mix_ln.bias
F32
F32
[2048]
blk.1.time_mix_ln.weight
F32
F32
[2048]
blk.1.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.1.time_mix_r_k.weight
F32
F32
[2048]
blk.1.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.1.time_mix_v0.weight
F32
F32
[2048]
blk.1.time_mix_v1.weight
F16
F16
[2048, 64]
blk.1.time_mix_v2.weight
F16
F16
[64, 2048]
blk.1.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.1.time_mix_w0.weight
F32
F32
[2048]
blk.1.time_mix_w1.weight
F32
F32
[2048, 96]
blk.1.time_mix_w2.weight
F32
F32
[96, 2048]
blk.2
blk.2.attn_norm.bias
F32
F32
[2048]
blk.2.attn_norm.weight
F32
F32
[2048]
blk.2.attn_norm_2.bias
F32
F32
[2048]
blk.2.attn_norm_2.weight
F32
F32
[2048]
blk.2.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.2.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.2.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.2.time_mix_a0.weight
F32
F32
[2048]
blk.2.time_mix_a1.weight
F16
F16
[2048, 96]
blk.2.time_mix_a2.weight
F16
F16
[96, 2048]
blk.2.time_mix_g1.weight
F16
F16
[2048, 256]
blk.2.time_mix_g2.weight
F16
F16
[256, 2048]
blk.2.time_mix_k_a.weight
F32
F32
[2048]
blk.2.time_mix_k_k.weight
F32
F32
[2048]
blk.2.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.2.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.2.time_mix_ln.bias
F32
F32
[2048]
blk.2.time_mix_ln.weight
F32
F32
[2048]
blk.2.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.2.time_mix_r_k.weight
F32
F32
[2048]
blk.2.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.2.time_mix_v0.weight
F32
F32
[2048]
blk.2.time_mix_v1.weight
F16
F16
[2048, 64]
blk.2.time_mix_v2.weight
F16
F16
[64, 2048]
blk.2.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.2.time_mix_w0.weight
F32
F32
[2048]
blk.2.time_mix_w1.weight
F32
F32
[2048, 96]
blk.2.time_mix_w2.weight
F32
F32
[96, 2048]
blk.3
blk.3.attn_norm.bias
F32
F32
[2048]
blk.3.attn_norm.weight
F32
F32
[2048]
blk.3.attn_norm_2.bias
F32
F32
[2048]
blk.3.attn_norm_2.weight
F32
F32
[2048]
blk.3.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.3.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.3.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.3.time_mix_a0.weight
F32
F32
[2048]
blk.3.time_mix_a1.weight
F16
F16
[2048, 96]
blk.3.time_mix_a2.weight
F16
F16
[96, 2048]
blk.3.time_mix_g1.weight
F16
F16
[2048, 256]
blk.3.time_mix_g2.weight
F16
F16
[256, 2048]
blk.3.time_mix_k_a.weight
F32
F32
[2048]
blk.3.time_mix_k_k.weight
F32
F32
[2048]
blk.3.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.3.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.3.time_mix_ln.bias
F32
F32
[2048]
blk.3.time_mix_ln.weight
F32
F32
[2048]
blk.3.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.3.time_mix_r_k.weight
F32
F32
[2048]
blk.3.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.3.time_mix_v0.weight
F32
F32
[2048]
blk.3.time_mix_v1.weight
F16
F16
[2048, 64]
blk.3.time_mix_v2.weight
F16
F16
[64, 2048]
blk.3.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.3.time_mix_w0.weight
F32
F32
[2048]
blk.3.time_mix_w1.weight
F32
F32
[2048, 96]
blk.3.time_mix_w2.weight
F32
F32
[96, 2048]
blk.4
blk.4.attn_norm.bias
F32
F32
[2048]
blk.4.attn_norm.weight
F32
F32
[2048]
blk.4.attn_norm_2.bias
F32
F32
[2048]
blk.4.attn_norm_2.weight
F32
F32
[2048]
blk.4.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.4.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.4.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.4.time_mix_a0.weight
F32
F32
[2048]
blk.4.time_mix_a1.weight
F16
F16
[2048, 96]
blk.4.time_mix_a2.weight
F16
F16
[96, 2048]
blk.4.time_mix_g1.weight
F16
F16
[2048, 256]
blk.4.time_mix_g2.weight
F16
F16
[256, 2048]
blk.4.time_mix_k_a.weight
F32
F32
[2048]
blk.4.time_mix_k_k.weight
F32
F32
[2048]
blk.4.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.4.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.4.time_mix_ln.bias
F32
F32
[2048]
blk.4.time_mix_ln.weight
F32
F32
[2048]
blk.4.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.4.time_mix_r_k.weight
F32
F32
[2048]
blk.4.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.4.time_mix_v0.weight
F32
F32
[2048]
blk.4.time_mix_v1.weight
F16
F16
[2048, 64]
blk.4.time_mix_v2.weight
F16
F16
[64, 2048]
blk.4.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.4.time_mix_w0.weight
F32
F32
[2048]
blk.4.time_mix_w1.weight
F32
F32
[2048, 96]
blk.4.time_mix_w2.weight
F32
F32
[96, 2048]
blk.5
blk.5.attn_norm.bias
F32
F32
[2048]
blk.5.attn_norm.weight
F32
F32
[2048]
blk.5.attn_norm_2.bias
F32
F32
[2048]
blk.5.attn_norm_2.weight
F32
F32
[2048]
blk.5.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.5.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.5.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.5.time_mix_a0.weight
F32
F32
[2048]
blk.5.time_mix_a1.weight
F16
F16
[2048, 96]
blk.5.time_mix_a2.weight
F16
F16
[96, 2048]
blk.5.time_mix_g1.weight
F16
F16
[2048, 256]
blk.5.time_mix_g2.weight
F16
F16
[256, 2048]
blk.5.time_mix_k_a.weight
F32
F32
[2048]
blk.5.time_mix_k_k.weight
F32
F32
[2048]
blk.5.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.5.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.5.time_mix_ln.bias
F32
F32
[2048]
blk.5.time_mix_ln.weight
F32
F32
[2048]
blk.5.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.5.time_mix_r_k.weight
F32
F32
[2048]
blk.5.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.5.time_mix_v0.weight
F32
F32
[2048]
blk.5.time_mix_v1.weight
F16
F16
[2048, 64]
blk.5.time_mix_v2.weight
F16
F16
[64, 2048]
blk.5.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.5.time_mix_w0.weight
F32
F32
[2048]
blk.5.time_mix_w1.weight
F32
F32
[2048, 96]
blk.5.time_mix_w2.weight
F32
F32
[96, 2048]
blk.6
blk.6.attn_norm.bias
F32
F32
[2048]
blk.6.attn_norm.weight
F32
F32
[2048]
blk.6.attn_norm_2.bias
F32
F32
[2048]
blk.6.attn_norm_2.weight
F32
F32
[2048]
blk.6.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.6.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.6.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.6.time_mix_a0.weight
F32
F32
[2048]
blk.6.time_mix_a1.weight
F16
F16
[2048, 96]
blk.6.time_mix_a2.weight
F16
F16
[96, 2048]
blk.6.time_mix_g1.weight
F16
F16
[2048, 256]
blk.6.time_mix_g2.weight
F16
F16
[256, 2048]
blk.6.time_mix_k_a.weight
F32
F32
[2048]
blk.6.time_mix_k_k.weight
F32
F32
[2048]
blk.6.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.6.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.6.time_mix_ln.bias
F32
F32
[2048]
blk.6.time_mix_ln.weight
F32
F32
[2048]
blk.6.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.6.time_mix_r_k.weight
F32
F32
[2048]
blk.6.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.6.time_mix_v0.weight
F32
F32
[2048]
blk.6.time_mix_v1.weight
F16
F16
[2048, 64]
blk.6.time_mix_v2.weight
F16
F16
[64, 2048]
blk.6.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.6.time_mix_w0.weight
F32
F32
[2048]
blk.6.time_mix_w1.weight
F32
F32
[2048, 96]
blk.6.time_mix_w2.weight
F32
F32
[96, 2048]
blk.7
blk.7.attn_norm.bias
F32
F32
[2048]
blk.7.attn_norm.weight
F32
F32
[2048]
blk.7.attn_norm_2.bias
F32
F32
[2048]
blk.7.attn_norm_2.weight
F32
F32
[2048]
blk.7.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.7.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.7.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.7.time_mix_a0.weight
F32
F32
[2048]
blk.7.time_mix_a1.weight
F16
F16
[2048, 96]
blk.7.time_mix_a2.weight
F16
F16
[96, 2048]
blk.7.time_mix_g1.weight
F16
F16
[2048, 256]
blk.7.time_mix_g2.weight
F16
F16
[256, 2048]
blk.7.time_mix_k_a.weight
F32
F32
[2048]
blk.7.time_mix_k_k.weight
F32
F32
[2048]
blk.7.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.7.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.7.time_mix_ln.bias
F32
F32
[2048]
blk.7.time_mix_ln.weight
F32
F32
[2048]
blk.7.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.7.time_mix_r_k.weight
F32
F32
[2048]
blk.7.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.7.time_mix_v0.weight
F32
F32
[2048]
blk.7.time_mix_v1.weight
F16
F16
[2048, 64]
blk.7.time_mix_v2.weight
F16
F16
[64, 2048]
blk.7.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.7.time_mix_w0.weight
F32
F32
[2048]
blk.7.time_mix_w1.weight
F32
F32
[2048, 96]
blk.7.time_mix_w2.weight
F32
F32
[96, 2048]
blk.8
blk.8.attn_norm.bias
F32
F32
[2048]
blk.8.attn_norm.weight
F32
F32
[2048]
blk.8.attn_norm_2.bias
F32
F32
[2048]
blk.8.attn_norm_2.weight
F32
F32
[2048]
blk.8.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.8.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.8.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.8.time_mix_a0.weight
F32
F32
[2048]
blk.8.time_mix_a1.weight
F16
F16
[2048, 96]
blk.8.time_mix_a2.weight
F16
F16
[96, 2048]
blk.8.time_mix_g1.weight
F16
F16
[2048, 256]
blk.8.time_mix_g2.weight
F16
F16
[256, 2048]
blk.8.time_mix_k_a.weight
F32
F32
[2048]
blk.8.time_mix_k_k.weight
F32
F32
[2048]
blk.8.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.8.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.8.time_mix_ln.bias
F32
F32
[2048]
blk.8.time_mix_ln.weight
F32
F32
[2048]
blk.8.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.8.time_mix_r_k.weight
F32
F32
[2048]
blk.8.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.8.time_mix_v0.weight
F32
F32
[2048]
blk.8.time_mix_v1.weight
F16
F16
[2048, 64]
blk.8.time_mix_v2.weight
F16
F16
[64, 2048]
blk.8.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.8.time_mix_w0.weight
F32
F32
[2048]
blk.8.time_mix_w1.weight
F32
F32
[2048, 96]
blk.8.time_mix_w2.weight
F32
F32
[96, 2048]
blk.9
blk.9.attn_norm.bias
F32
F32
[2048]
blk.9.attn_norm.weight
F32
F32
[2048]
blk.9.attn_norm_2.bias
F32
F32
[2048]
blk.9.attn_norm_2.weight
F32
F32
[2048]
blk.9.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.9.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.9.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.9.time_mix_a0.weight
F32
F32
[2048]
blk.9.time_mix_a1.weight
F16
F16
[2048, 96]
blk.9.time_mix_a2.weight
F16
F16
[96, 2048]
blk.9.time_mix_g1.weight
F16
F16
[2048, 256]
blk.9.time_mix_g2.weight
F16
F16
[256, 2048]
blk.9.time_mix_k_a.weight
F32
F32
[2048]
blk.9.time_mix_k_k.weight
F32
F32
[2048]
blk.9.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.9.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.9.time_mix_ln.bias
F32
F32
[2048]
blk.9.time_mix_ln.weight
F32
F32
[2048]
blk.9.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.9.time_mix_r_k.weight
F32
F32
[2048]
blk.9.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.9.time_mix_v0.weight
F32
F32
[2048]
blk.9.time_mix_v1.weight
F16
F16
[2048, 64]
blk.9.time_mix_v2.weight
F16
F16
[64, 2048]
blk.9.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.9.time_mix_w0.weight
F32
F32
[2048]
blk.9.time_mix_w1.weight
F32
F32
[2048, 96]
blk.9.time_mix_w2.weight
F32
F32
[96, 2048]
blk.10
blk.10.attn_norm.bias
F32
F32
[2048]
blk.10.attn_norm.weight
F32
F32
[2048]
blk.10.attn_norm_2.bias
F32
F32
[2048]
blk.10.attn_norm_2.weight
F32
F32
[2048]
blk.10.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.10.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.10.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.10.time_mix_a0.weight
F32
F32
[2048]
blk.10.time_mix_a1.weight
F16
F16
[2048, 96]
blk.10.time_mix_a2.weight
F16
F16
[96, 2048]
blk.10.time_mix_g1.weight
F16
F16
[2048, 256]
blk.10.time_mix_g2.weight
F16
F16
[256, 2048]
blk.10.time_mix_k_a.weight
F32
F32
[2048]
blk.10.time_mix_k_k.weight
F32
F32
[2048]
blk.10.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.10.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.10.time_mix_ln.bias
F32
F32
[2048]
blk.10.time_mix_ln.weight
F32
F32
[2048]
blk.10.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.10.time_mix_r_k.weight
F32
F32
[2048]
blk.10.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.10.time_mix_v0.weight
F32
F32
[2048]
blk.10.time_mix_v1.weight
F16
F16
[2048, 64]
blk.10.time_mix_v2.weight
F16
F16
[64, 2048]
blk.10.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.10.time_mix_w0.weight
F32
F32
[2048]
blk.10.time_mix_w1.weight
F32
F32
[2048, 96]
blk.10.time_mix_w2.weight
F32
F32
[96, 2048]
blk.11
blk.11.attn_norm.bias
F32
F32
[2048]
blk.11.attn_norm.weight
F32
F32
[2048]
blk.11.attn_norm_2.bias
F32
F32
[2048]
blk.11.attn_norm_2.weight
F32
F32
[2048]
blk.11.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.11.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.11.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.11.time_mix_a0.weight
F32
F32
[2048]
blk.11.time_mix_a1.weight
F16
F16
[2048, 96]
blk.11.time_mix_a2.weight
F16
F16
[96, 2048]
blk.11.time_mix_g1.weight
F16
F16
[2048, 256]
blk.11.time_mix_g2.weight
F16
F16
[256, 2048]
blk.11.time_mix_k_a.weight
F32
F32
[2048]
blk.11.time_mix_k_k.weight
F32
F32
[2048]
blk.11.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.11.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.11.time_mix_ln.bias
F32
F32
[2048]
blk.11.time_mix_ln.weight
F32
F32
[2048]
blk.11.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.11.time_mix_r_k.weight
F32
F32
[2048]
blk.11.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.11.time_mix_v0.weight
F32
F32
[2048]
blk.11.time_mix_v1.weight
F16
F16
[2048, 64]
blk.11.time_mix_v2.weight
F16
F16
[64, 2048]
blk.11.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.11.time_mix_w0.weight
F32
F32
[2048]
blk.11.time_mix_w1.weight
F32
F32
[2048, 96]
blk.11.time_mix_w2.weight
F32
F32
[96, 2048]
blk.12
blk.12.attn_norm.bias
F32
F32
[2048]
blk.12.attn_norm.weight
F32
F32
[2048]
blk.12.attn_norm_2.bias
F32
F32
[2048]
blk.12.attn_norm_2.weight
F32
F32
[2048]
blk.12.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.12.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.12.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.12.time_mix_a0.weight
F32
F32
[2048]
blk.12.time_mix_a1.weight
F16
F16
[2048, 96]
blk.12.time_mix_a2.weight
F16
F16
[96, 2048]
blk.12.time_mix_g1.weight
F16
F16
[2048, 256]
blk.12.time_mix_g2.weight
F16
F16
[256, 2048]
blk.12.time_mix_k_a.weight
F32
F32
[2048]
blk.12.time_mix_k_k.weight
F32
F32
[2048]
blk.12.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.12.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.12.time_mix_ln.bias
F32
F32
[2048]
blk.12.time_mix_ln.weight
F32
F32
[2048]
blk.12.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.12.time_mix_r_k.weight
F32
F32
[2048]
blk.12.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.12.time_mix_v0.weight
F32
F32
[2048]
blk.12.time_mix_v1.weight
F16
F16
[2048, 64]
blk.12.time_mix_v2.weight
F16
F16
[64, 2048]
blk.12.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.12.time_mix_w0.weight
F32
F32
[2048]
blk.12.time_mix_w1.weight
F32
F32
[2048, 96]
blk.12.time_mix_w2.weight
F32
F32
[96, 2048]
blk.13
blk.13.attn_norm.bias
F32
F32
[2048]
blk.13.attn_norm.weight
F32
F32
[2048]
blk.13.attn_norm_2.bias
F32
F32
[2048]
blk.13.attn_norm_2.weight
F32
F32
[2048]
blk.13.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.13.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.13.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.13.time_mix_a0.weight
F32
F32
[2048]
blk.13.time_mix_a1.weight
F16
F16
[2048, 96]
blk.13.time_mix_a2.weight
F16
F16
[96, 2048]
blk.13.time_mix_g1.weight
F16
F16
[2048, 256]
blk.13.time_mix_g2.weight
F16
F16
[256, 2048]
blk.13.time_mix_k_a.weight
F32
F32
[2048]
blk.13.time_mix_k_k.weight
F32
F32
[2048]
blk.13.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.13.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.13.time_mix_ln.bias
F32
F32
[2048]
blk.13.time_mix_ln.weight
F32
F32
[2048]
blk.13.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.13.time_mix_r_k.weight
F32
F32
[2048]
blk.13.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.13.time_mix_v0.weight
F32
F32
[2048]
blk.13.time_mix_v1.weight
F16
F16
[2048, 64]
blk.13.time_mix_v2.weight
F16
F16
[64, 2048]
blk.13.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.13.time_mix_w0.weight
F32
F32
[2048]
blk.13.time_mix_w1.weight
F32
F32
[2048, 96]
blk.13.time_mix_w2.weight
F32
F32
[96, 2048]
blk.14
blk.14.attn_norm.bias
F32
F32
[2048]
blk.14.attn_norm.weight
F32
F32
[2048]
blk.14.attn_norm_2.bias
F32
F32
[2048]
blk.14.attn_norm_2.weight
F32
F32
[2048]
blk.14.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.14.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.14.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.14.time_mix_a0.weight
F32
F32
[2048]
blk.14.time_mix_a1.weight
F16
F16
[2048, 96]
blk.14.time_mix_a2.weight
F16
F16
[96, 2048]
blk.14.time_mix_g1.weight
F16
F16
[2048, 256]
blk.14.time_mix_g2.weight
F16
F16
[256, 2048]
blk.14.time_mix_k_a.weight
F32
F32
[2048]
blk.14.time_mix_k_k.weight
F32
F32
[2048]
blk.14.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.14.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.14.time_mix_ln.bias
F32
F32
[2048]
blk.14.time_mix_ln.weight
F32
F32
[2048]
blk.14.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.14.time_mix_r_k.weight
F32
F32
[2048]
blk.14.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.14.time_mix_v0.weight
F32
F32
[2048]
blk.14.time_mix_v1.weight
F16
F16
[2048, 64]
blk.14.time_mix_v2.weight
F16
F16
[64, 2048]
blk.14.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.14.time_mix_w0.weight
F32
F32
[2048]
blk.14.time_mix_w1.weight
F32
F32
[2048, 96]
blk.14.time_mix_w2.weight
F32
F32
[96, 2048]
blk.15
blk.15.attn_norm.bias
F32
F32
[2048]
blk.15.attn_norm.weight
F32
F32
[2048]
blk.15.attn_norm_2.bias
F32
F32
[2048]
blk.15.attn_norm_2.weight
F32
F32
[2048]
blk.15.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.15.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.15.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.15.time_mix_a0.weight
F32
F32
[2048]
blk.15.time_mix_a1.weight
F16
F16
[2048, 96]
blk.15.time_mix_a2.weight
F16
F16
[96, 2048]
blk.15.time_mix_g1.weight
F16
F16
[2048, 256]
blk.15.time_mix_g2.weight
F16
F16
[256, 2048]
blk.15.time_mix_k_a.weight
F32
F32
[2048]
blk.15.time_mix_k_k.weight
F32
F32
[2048]
blk.15.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.15.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.15.time_mix_ln.bias
F32
F32
[2048]
blk.15.time_mix_ln.weight
F32
F32
[2048]
blk.15.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.15.time_mix_r_k.weight
F32
F32
[2048]
blk.15.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.15.time_mix_v0.weight
F32
F32
[2048]
blk.15.time_mix_v1.weight
F16
F16
[2048, 64]
blk.15.time_mix_v2.weight
F16
F16
[64, 2048]
blk.15.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.15.time_mix_w0.weight
F32
F32
[2048]
blk.15.time_mix_w1.weight
F32
F32
[2048, 96]
blk.15.time_mix_w2.weight
F32
F32
[96, 2048]
blk.16
blk.16.attn_norm.bias
F32
F32
[2048]
blk.16.attn_norm.weight
F32
F32
[2048]
blk.16.attn_norm_2.bias
F32
F32
[2048]
blk.16.attn_norm_2.weight
F32
F32
[2048]
blk.16.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.16.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.16.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.16.time_mix_a0.weight
F32
F32
[2048]
blk.16.time_mix_a1.weight
F16
F16
[2048, 96]
blk.16.time_mix_a2.weight
F16
F16
[96, 2048]
blk.16.time_mix_g1.weight
F16
F16
[2048, 256]
blk.16.time_mix_g2.weight
F16
F16
[256, 2048]
blk.16.time_mix_k_a.weight
F32
F32
[2048]
blk.16.time_mix_k_k.weight
F32
F32
[2048]
blk.16.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.16.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.16.time_mix_ln.bias
F32
F32
[2048]
blk.16.time_mix_ln.weight
F32
F32
[2048]
blk.16.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.16.time_mix_r_k.weight
F32
F32
[2048]
blk.16.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.16.time_mix_v0.weight
F32
F32
[2048]
blk.16.time_mix_v1.weight
F16
F16
[2048, 64]
blk.16.time_mix_v2.weight
F16
F16
[64, 2048]
blk.16.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.16.time_mix_w0.weight
F32
F32
[2048]
blk.16.time_mix_w1.weight
F32
F32
[2048, 96]
blk.16.time_mix_w2.weight
F32
F32
[96, 2048]
blk.17
blk.17.attn_norm.bias
F32
F32
[2048]
blk.17.attn_norm.weight
F32
F32
[2048]
blk.17.attn_norm_2.bias
F32
F32
[2048]
blk.17.attn_norm_2.weight
F32
F32
[2048]
blk.17.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.17.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.17.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.17.time_mix_a0.weight
F32
F32
[2048]
blk.17.time_mix_a1.weight
F16
F16
[2048, 96]
blk.17.time_mix_a2.weight
F16
F16
[96, 2048]
blk.17.time_mix_g1.weight
F16
F16
[2048, 256]
blk.17.time_mix_g2.weight
F16
F16
[256, 2048]
blk.17.time_mix_k_a.weight
F32
F32
[2048]
blk.17.time_mix_k_k.weight
F32
F32
[2048]
blk.17.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.17.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.17.time_mix_ln.bias
F32
F32
[2048]
blk.17.time_mix_ln.weight
F32
F32
[2048]
blk.17.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.17.time_mix_r_k.weight
F32
F32
[2048]
blk.17.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.17.time_mix_v0.weight
F32
F32
[2048]
blk.17.time_mix_v1.weight
F16
F16
[2048, 64]
blk.17.time_mix_v2.weight
F16
F16
[64, 2048]
blk.17.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.17.time_mix_w0.weight
F32
F32
[2048]
blk.17.time_mix_w1.weight
F32
F32
[2048, 96]
blk.17.time_mix_w2.weight
F32
F32
[96, 2048]
blk.18
blk.18.attn_norm.bias
F32
F32
[2048]
blk.18.attn_norm.weight
F32
F32
[2048]
blk.18.attn_norm_2.bias
F32
F32
[2048]
blk.18.attn_norm_2.weight
F32
F32
[2048]
blk.18.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.18.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.18.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.18.time_mix_a0.weight
F32
F32
[2048]
blk.18.time_mix_a1.weight
F16
F16
[2048, 96]
blk.18.time_mix_a2.weight
F16
F16
[96, 2048]
blk.18.time_mix_g1.weight
F16
F16
[2048, 256]
blk.18.time_mix_g2.weight
F16
F16
[256, 2048]
blk.18.time_mix_k_a.weight
F32
F32
[2048]
blk.18.time_mix_k_k.weight
F32
F32
[2048]
blk.18.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.18.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.18.time_mix_ln.bias
F32
F32
[2048]
blk.18.time_mix_ln.weight
F32
F32
[2048]
blk.18.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.18.time_mix_r_k.weight
F32
F32
[2048]
blk.18.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.18.time_mix_v0.weight
F32
F32
[2048]
blk.18.time_mix_v1.weight
F16
F16
[2048, 64]
blk.18.time_mix_v2.weight
F16
F16
[64, 2048]
blk.18.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.18.time_mix_w0.weight
F32
F32
[2048]
blk.18.time_mix_w1.weight
F32
F32
[2048, 96]
blk.18.time_mix_w2.weight
F32
F32
[96, 2048]
blk.19
blk.19.attn_norm.bias
F32
F32
[2048]
blk.19.attn_norm.weight
F32
F32
[2048]
blk.19.attn_norm_2.bias
F32
F32
[2048]
blk.19.attn_norm_2.weight
F32
F32
[2048]
blk.19.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.19.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.19.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.19.time_mix_a0.weight
F32
F32
[2048]
blk.19.time_mix_a1.weight
F16
F16
[2048, 96]
blk.19.time_mix_a2.weight
F16
F16
[96, 2048]
blk.19.time_mix_g1.weight
F16
F16
[2048, 256]
blk.19.time_mix_g2.weight
F16
F16
[256, 2048]
blk.19.time_mix_k_a.weight
F32
F32
[2048]
blk.19.time_mix_k_k.weight
F32
F32
[2048]
blk.19.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.19.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.19.time_mix_ln.bias
F32
F32
[2048]
blk.19.time_mix_ln.weight
F32
F32
[2048]
blk.19.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.19.time_mix_r_k.weight
F32
F32
[2048]
blk.19.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.19.time_mix_v0.weight
F32
F32
[2048]
blk.19.time_mix_v1.weight
F16
F16
[2048, 64]
blk.19.time_mix_v2.weight
F16
F16
[64, 2048]
blk.19.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.19.time_mix_w0.weight
F32
F32
[2048]
blk.19.time_mix_w1.weight
F32
F32
[2048, 96]
blk.19.time_mix_w2.weight
F32
F32
[96, 2048]
blk.20
blk.20.attn_norm.bias
F32
F32
[2048]
blk.20.attn_norm.weight
F32
F32
[2048]
blk.20.attn_norm_2.bias
F32
F32
[2048]
blk.20.attn_norm_2.weight
F32
F32
[2048]
blk.20.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.20.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.20.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.20.time_mix_a0.weight
F32
F32
[2048]
blk.20.time_mix_a1.weight
F16
F16
[2048, 96]
blk.20.time_mix_a2.weight
F16
F16
[96, 2048]
blk.20.time_mix_g1.weight
F16
F16
[2048, 256]
blk.20.time_mix_g2.weight
F16
F16
[256, 2048]
blk.20.time_mix_k_a.weight
F32
F32
[2048]
blk.20.time_mix_k_k.weight
F32
F32
[2048]
blk.20.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.20.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.20.time_mix_ln.bias
F32
F32
[2048]
blk.20.time_mix_ln.weight
F32
F32
[2048]
blk.20.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.20.time_mix_r_k.weight
F32
F32
[2048]
blk.20.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.20.time_mix_v0.weight
F32
F32
[2048]
blk.20.time_mix_v1.weight
F16
F16
[2048, 64]
blk.20.time_mix_v2.weight
F16
F16
[64, 2048]
blk.20.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.20.time_mix_w0.weight
F32
F32
[2048]
blk.20.time_mix_w1.weight
F32
F32
[2048, 96]
blk.20.time_mix_w2.weight
F32
F32
[96, 2048]
blk.21
blk.21.attn_norm.bias
F32
F32
[2048]
blk.21.attn_norm.weight
F32
F32
[2048]
blk.21.attn_norm_2.bias
F32
F32
[2048]
blk.21.attn_norm_2.weight
F32
F32
[2048]
blk.21.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.21.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.21.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.21.time_mix_a0.weight
F32
F32
[2048]
blk.21.time_mix_a1.weight
F16
F16
[2048, 96]
blk.21.time_mix_a2.weight
F16
F16
[96, 2048]
blk.21.time_mix_g1.weight
F16
F16
[2048, 256]
blk.21.time_mix_g2.weight
F16
F16
[256, 2048]
blk.21.time_mix_k_a.weight
F32
F32
[2048]
blk.21.time_mix_k_k.weight
F32
F32
[2048]
blk.21.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.21.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.21.time_mix_ln.bias
F32
F32
[2048]
blk.21.time_mix_ln.weight
F32
F32
[2048]
blk.21.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.21.time_mix_r_k.weight
F32
F32
[2048]
blk.21.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.21.time_mix_v0.weight
F32
F32
[2048]
blk.21.time_mix_v1.weight
F16
F16
[2048, 64]
blk.21.time_mix_v2.weight
F16
F16
[64, 2048]
blk.21.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.21.time_mix_w0.weight
F32
F32
[2048]
blk.21.time_mix_w1.weight
F32
F32
[2048, 96]
blk.21.time_mix_w2.weight
F32
F32
[96, 2048]
blk.22
blk.22.attn_norm.bias
F32
F32
[2048]
blk.22.attn_norm.weight
F32
F32
[2048]
blk.22.attn_norm_2.bias
F32
F32
[2048]
blk.22.attn_norm_2.weight
F32
F32
[2048]
blk.22.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.22.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.22.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.22.time_mix_a0.weight
F32
F32
[2048]
blk.22.time_mix_a1.weight
F16
F16
[2048, 96]
blk.22.time_mix_a2.weight
F16
F16
[96, 2048]
blk.22.time_mix_g1.weight
F16
F16
[2048, 256]
blk.22.time_mix_g2.weight
F16
F16
[256, 2048]
blk.22.time_mix_k_a.weight
F32
F32
[2048]
blk.22.time_mix_k_k.weight
F32
F32
[2048]
blk.22.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.22.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.22.time_mix_ln.bias
F32
F32
[2048]
blk.22.time_mix_ln.weight
F32
F32
[2048]
blk.22.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.22.time_mix_r_k.weight
F32
F32
[2048]
blk.22.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.22.time_mix_v0.weight
F32
F32
[2048]
blk.22.time_mix_v1.weight
F16
F16
[2048, 64]
blk.22.time_mix_v2.weight
F16
F16
[64, 2048]
blk.22.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.22.time_mix_w0.weight
F32
F32
[2048]
blk.22.time_mix_w1.weight
F32
F32
[2048, 96]
blk.22.time_mix_w2.weight
F32
F32
[96, 2048]
blk.23
blk.23.attn_norm.bias
F32
F32
[2048]
blk.23.attn_norm.weight
F32
F32
[2048]
blk.23.attn_norm_2.bias
F32
F32
[2048]
blk.23.attn_norm_2.weight
F32
F32
[2048]
blk.23.channel_mix_key.weight
Q6_K
Q6_K
[2048, 8192]
blk.23.channel_mix_lerp_k.weight
F32
F32
[2048]
blk.23.channel_mix_value.weight
Q6_K
Q6_K
[8192, 2048]
blk.23.time_mix_a0.weight
F32
F32
[2048]
blk.23.time_mix_a1.weight
F16
F16
[2048, 96]
blk.23.time_mix_a2.weight
F16
F16
[96, 2048]
blk.23.time_mix_g1.weight
F16
F16
[2048, 256]
blk.23.time_mix_g2.weight
F16
F16
[256, 2048]
blk.23.time_mix_k_a.weight
F32
F32
[2048]
blk.23.time_mix_k_k.weight
F32
F32
[2048]
blk.23.time_mix_key.weight
Q6_K
Q6_K
[2048, 2048]
blk.23.time_mix_lerp_fused.weight
F32
F32
[2048, 1, 1, 6]
blk.23.time_mix_ln.bias
F32
F32
[2048]
blk.23.time_mix_ln.weight
F32
F32
[2048]
blk.23.time_mix_output.weight
Q6_K
Q6_K
[2048, 2048]
blk.23.time_mix_r_k.weight
F32
F32
[2048]
blk.23.time_mix_receptance.weight
Q6_K
Q6_K
[2048, 2048]
blk.23.time_mix_v0.weight
F32
F32
[2048]
blk.23.time_mix_v1.weight
F16
F16
[2048, 64]
blk.23.time_mix_v2.weight
F16
F16
[64, 2048]
blk.23.time_mix_value.weight
Q6_K
Q6_K
[2048, 2048]
blk.23.time_mix_w0.weight
F32
F32
[2048]
blk.23.time_mix_w1.weight
F32
F32
[2048, 96]
blk.23.time_mix_w2.weight
F32
F32
[96, 2048]
output.weight
Q6_K
Q6_K
[2048, 65536]
output_norm.bias
F32
F32
[2048]
token_embd_norm.bias
F32
F32
[2048]
token_embd_norm.weight
F32
F32
[2048]
output_norm.weight
F32
F32
[2048]