Models
Docs
Pricing
Sign in
Download
Models
Download
Docs
Pricing
Sign in
heredos
/
rwkv7
:13.3b
462
Downloads
Updated
2 months ago
an rwkv repo that tries to be less confusing, more up to date and with some QOL features
an rwkv repo that tries to be less confusing, more up to date and with some QOL features
Cancel
tools
thinking
0.1b
0.4b
1.5b
2.9b
7.2b
13.3b
rwkv7:13.3b
...
/
model
b1e037047d02 · 8.4GB
Metadata
general.architecture
rwkv7
rwkv7
general.file_type
Q4_K_M
Q4_K_M
rwkv7.attention.decay_lora_rank
192
192
rwkv7.attention.gate_lora_rank
384
384
rwkv7.attention.head_count
0
0
rwkv7.attention.iclr_lora_rank
192
192
rwkv7.attention.layer_norm_epsilon
1e-05
1e-05
rwkv7.attention.value_residual_mix_lora_rank
128
128
rwkv7.block_count
61
61
rwkv7.context_length
1048576
1048576
rwkv7.embedding_length
4096
4096
rwkv7.feed_forward_length
16384
16384
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, ...]
quantize.imatrix.chunks_count
234
234
quantize.imatrix.dataset
calibration_data_v5_rc.txt
calibration_data_v5_rc.txt
quantize.imatrix.entries_count
852
852
quantize.imatrix.file
./imatrix.gguf
./imatrix.gguf
Tensor
Name
Type
Shape
token_embd.weight
Q4_K
Q4_K
[4096, 65536]
blk.0
blk.0.attn_norm.bias
F32
F32
[4096]
blk.0.attn_norm.weight
F32
F32
[4096]
blk.0.attn_norm_2.bias
F32
F32
[4096]
blk.0.attn_norm_2.weight
F32
F32
[4096]
blk.0.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.0.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.0.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.0.time_mix_a0.weight
F32
F32
[4096]
blk.0.time_mix_a1.weight
F16
F16
[4096, 192]
blk.0.time_mix_a2.weight
F16
F16
[192, 4096]
blk.0.time_mix_g1.weight
F16
F16
[4096, 384]
blk.0.time_mix_g2.weight
F16
F16
[384, 4096]
blk.0.time_mix_k_a.weight
F32
F32
[4096]
blk.0.time_mix_k_k.weight
F32
F32
[4096]
blk.0.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.0.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.0.time_mix_ln.bias
F32
F32
[4096]
blk.0.time_mix_ln.weight
F32
F32
[4096]
blk.0.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.0.time_mix_r_k.weight
F32
F32
[4096]
blk.0.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.0.time_mix_v0.weight
F32
F32
[4096]
blk.0.time_mix_v1.weight
F16
F16
[4096, 192]
blk.0.time_mix_v2.weight
F16
F16
[192, 4096]
blk.0.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.0.time_mix_w0.weight
F32
F32
[4096]
blk.0.time_mix_w1.weight
F32
F32
[4096, 192]
blk.0.time_mix_w2.weight
F32
F32
[192, 4096]
blk.1
blk.1.attn_norm.bias
F32
F32
[4096]
blk.1.attn_norm.weight
F32
F32
[4096]
blk.1.attn_norm_2.bias
F32
F32
[4096]
blk.1.attn_norm_2.weight
F32
F32
[4096]
blk.1.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.1.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.1.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.1.time_mix_a0.weight
F32
F32
[4096]
blk.1.time_mix_a1.weight
F16
F16
[4096, 192]
blk.1.time_mix_a2.weight
F16
F16
[192, 4096]
blk.1.time_mix_g1.weight
F16
F16
[4096, 384]
blk.1.time_mix_g2.weight
F16
F16
[384, 4096]
blk.1.time_mix_k_a.weight
F32
F32
[4096]
blk.1.time_mix_k_k.weight
F32
F32
[4096]
blk.1.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.1.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.1.time_mix_ln.bias
F32
F32
[4096]
blk.1.time_mix_ln.weight
F32
F32
[4096]
blk.1.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.1.time_mix_r_k.weight
F32
F32
[4096]
blk.1.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.1.time_mix_v0.weight
F32
F32
[4096]
blk.1.time_mix_v1.weight
F16
F16
[4096, 128]
blk.1.time_mix_v2.weight
F16
F16
[128, 4096]
blk.1.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.1.time_mix_w0.weight
F32
F32
[4096]
blk.1.time_mix_w1.weight
F32
F32
[4096, 192]
blk.1.time_mix_w2.weight
F32
F32
[192, 4096]
blk.2
blk.2.attn_norm.bias
F32
F32
[4096]
blk.2.attn_norm.weight
F32
F32
[4096]
blk.2.attn_norm_2.bias
F32
F32
[4096]
blk.2.attn_norm_2.weight
F32
F32
[4096]
blk.2.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.2.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.2.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.2.time_mix_a0.weight
F32
F32
[4096]
blk.2.time_mix_a1.weight
F16
F16
[4096, 192]
blk.2.time_mix_a2.weight
F16
F16
[192, 4096]
blk.2.time_mix_g1.weight
F16
F16
[4096, 384]
blk.2.time_mix_g2.weight
F16
F16
[384, 4096]
blk.2.time_mix_k_a.weight
F32
F32
[4096]
blk.2.time_mix_k_k.weight
F32
F32
[4096]
blk.2.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.2.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.2.time_mix_ln.bias
F32
F32
[4096]
blk.2.time_mix_ln.weight
F32
F32
[4096]
blk.2.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.2.time_mix_r_k.weight
F32
F32
[4096]
blk.2.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.2.time_mix_v0.weight
F32
F32
[4096]
blk.2.time_mix_v1.weight
F16
F16
[4096, 128]
blk.2.time_mix_v2.weight
F16
F16
[128, 4096]
blk.2.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.2.time_mix_w0.weight
F32
F32
[4096]
blk.2.time_mix_w1.weight
F32
F32
[4096, 192]
blk.2.time_mix_w2.weight
F32
F32
[192, 4096]
blk.3
blk.3.attn_norm.bias
F32
F32
[4096]
blk.3.attn_norm.weight
F32
F32
[4096]
blk.3.attn_norm_2.bias
F32
F32
[4096]
blk.3.attn_norm_2.weight
F32
F32
[4096]
blk.3.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.3.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.3.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.3.time_mix_a0.weight
F32
F32
[4096]
blk.3.time_mix_a1.weight
F16
F16
[4096, 192]
blk.3.time_mix_a2.weight
F16
F16
[192, 4096]
blk.3.time_mix_g1.weight
F16
F16
[4096, 384]
blk.3.time_mix_g2.weight
F16
F16
[384, 4096]
blk.3.time_mix_k_a.weight
F32
F32
[4096]
blk.3.time_mix_k_k.weight
F32
F32
[4096]
blk.3.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.3.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.3.time_mix_ln.bias
F32
F32
[4096]
blk.3.time_mix_ln.weight
F32
F32
[4096]
blk.3.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.3.time_mix_r_k.weight
F32
F32
[4096]
blk.3.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.3.time_mix_v0.weight
F32
F32
[4096]
blk.3.time_mix_v1.weight
F16
F16
[4096, 128]
blk.3.time_mix_v2.weight
F16
F16
[128, 4096]
blk.3.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.3.time_mix_w0.weight
F32
F32
[4096]
blk.3.time_mix_w1.weight
F32
F32
[4096, 192]
blk.3.time_mix_w2.weight
F32
F32
[192, 4096]
blk.4
blk.4.attn_norm.bias
F32
F32
[4096]
blk.4.attn_norm.weight
F32
F32
[4096]
blk.4.attn_norm_2.bias
F32
F32
[4096]
blk.4.attn_norm_2.weight
F32
F32
[4096]
blk.4.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.4.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.4.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.4.time_mix_a0.weight
F32
F32
[4096]
blk.4.time_mix_a1.weight
F16
F16
[4096, 192]
blk.4.time_mix_a2.weight
F16
F16
[192, 4096]
blk.4.time_mix_g1.weight
F16
F16
[4096, 384]
blk.4.time_mix_g2.weight
F16
F16
[384, 4096]
blk.4.time_mix_k_a.weight
F32
F32
[4096]
blk.4.time_mix_k_k.weight
F32
F32
[4096]
blk.4.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.4.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.4.time_mix_ln.bias
F32
F32
[4096]
blk.4.time_mix_ln.weight
F32
F32
[4096]
blk.4.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.4.time_mix_r_k.weight
F32
F32
[4096]
blk.4.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.4.time_mix_v0.weight
F32
F32
[4096]
blk.4.time_mix_v1.weight
F16
F16
[4096, 128]
blk.4.time_mix_v2.weight
F16
F16
[128, 4096]
blk.4.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.4.time_mix_w0.weight
F32
F32
[4096]
blk.4.time_mix_w1.weight
F32
F32
[4096, 192]
blk.4.time_mix_w2.weight
F32
F32
[192, 4096]
blk.5
blk.5.attn_norm.bias
F32
F32
[4096]
blk.5.attn_norm.weight
F32
F32
[4096]
blk.5.attn_norm_2.bias
F32
F32
[4096]
blk.5.attn_norm_2.weight
F32
F32
[4096]
blk.5.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.5.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.5.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.5.time_mix_a0.weight
F32
F32
[4096]
blk.5.time_mix_a1.weight
F16
F16
[4096, 192]
blk.5.time_mix_a2.weight
F16
F16
[192, 4096]
blk.5.time_mix_g1.weight
F16
F16
[4096, 384]
blk.5.time_mix_g2.weight
F16
F16
[384, 4096]
blk.5.time_mix_k_a.weight
F32
F32
[4096]
blk.5.time_mix_k_k.weight
F32
F32
[4096]
blk.5.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.5.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.5.time_mix_ln.bias
F32
F32
[4096]
blk.5.time_mix_ln.weight
F32
F32
[4096]
blk.5.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.5.time_mix_r_k.weight
F32
F32
[4096]
blk.5.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.5.time_mix_v0.weight
F32
F32
[4096]
blk.5.time_mix_v1.weight
F16
F16
[4096, 128]
blk.5.time_mix_v2.weight
F16
F16
[128, 4096]
blk.5.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.5.time_mix_w0.weight
F32
F32
[4096]
blk.5.time_mix_w1.weight
F32
F32
[4096, 192]
blk.5.time_mix_w2.weight
F32
F32
[192, 4096]
blk.6
blk.6.attn_norm.bias
F32
F32
[4096]
blk.6.attn_norm.weight
F32
F32
[4096]
blk.6.attn_norm_2.bias
F32
F32
[4096]
blk.6.attn_norm_2.weight
F32
F32
[4096]
blk.6.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.6.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.6.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.6.time_mix_a0.weight
F32
F32
[4096]
blk.6.time_mix_a1.weight
F16
F16
[4096, 192]
blk.6.time_mix_a2.weight
F16
F16
[192, 4096]
blk.6.time_mix_g1.weight
F16
F16
[4096, 384]
blk.6.time_mix_g2.weight
F16
F16
[384, 4096]
blk.6.time_mix_k_a.weight
F32
F32
[4096]
blk.6.time_mix_k_k.weight
F32
F32
[4096]
blk.6.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.6.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.6.time_mix_ln.bias
F32
F32
[4096]
blk.6.time_mix_ln.weight
F32
F32
[4096]
blk.6.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.6.time_mix_r_k.weight
F32
F32
[4096]
blk.6.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.6.time_mix_v0.weight
F32
F32
[4096]
blk.6.time_mix_v1.weight
F16
F16
[4096, 128]
blk.6.time_mix_v2.weight
F16
F16
[128, 4096]
blk.6.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.6.time_mix_w0.weight
F32
F32
[4096]
blk.6.time_mix_w1.weight
F32
F32
[4096, 192]
blk.6.time_mix_w2.weight
F32
F32
[192, 4096]
blk.7
blk.7.attn_norm.bias
F32
F32
[4096]
blk.7.attn_norm.weight
F32
F32
[4096]
blk.7.attn_norm_2.bias
F32
F32
[4096]
blk.7.attn_norm_2.weight
F32
F32
[4096]
blk.7.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.7.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.7.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.7.time_mix_a0.weight
F32
F32
[4096]
blk.7.time_mix_a1.weight
F16
F16
[4096, 192]
blk.7.time_mix_a2.weight
F16
F16
[192, 4096]
blk.7.time_mix_g1.weight
F16
F16
[4096, 384]
blk.7.time_mix_g2.weight
F16
F16
[384, 4096]
blk.7.time_mix_k_a.weight
F32
F32
[4096]
blk.7.time_mix_k_k.weight
F32
F32
[4096]
blk.7.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.7.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.7.time_mix_ln.bias
F32
F32
[4096]
blk.7.time_mix_ln.weight
F32
F32
[4096]
blk.7.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.7.time_mix_r_k.weight
F32
F32
[4096]
blk.7.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.7.time_mix_v0.weight
F32
F32
[4096]
blk.7.time_mix_v1.weight
F16
F16
[4096, 128]
blk.7.time_mix_v2.weight
F16
F16
[128, 4096]
blk.7.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.7.time_mix_w0.weight
F32
F32
[4096]
blk.7.time_mix_w1.weight
F32
F32
[4096, 192]
blk.7.time_mix_w2.weight
F32
F32
[192, 4096]
blk.8
blk.8.attn_norm.bias
F32
F32
[4096]
blk.8.attn_norm.weight
F32
F32
[4096]
blk.8.attn_norm_2.bias
F32
F32
[4096]
blk.8.attn_norm_2.weight
F32
F32
[4096]
blk.8.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.8.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.8.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.8.time_mix_a0.weight
F32
F32
[4096]
blk.8.time_mix_a1.weight
F16
F16
[4096, 192]
blk.8.time_mix_a2.weight
F16
F16
[192, 4096]
blk.8.time_mix_g1.weight
F16
F16
[4096, 384]
blk.8.time_mix_g2.weight
F16
F16
[384, 4096]
blk.8.time_mix_k_a.weight
F32
F32
[4096]
blk.8.time_mix_k_k.weight
F32
F32
[4096]
blk.8.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.8.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.8.time_mix_ln.bias
F32
F32
[4096]
blk.8.time_mix_ln.weight
F32
F32
[4096]
blk.8.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.8.time_mix_r_k.weight
F32
F32
[4096]
blk.8.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.8.time_mix_v0.weight
F32
F32
[4096]
blk.8.time_mix_v1.weight
F16
F16
[4096, 128]
blk.8.time_mix_v2.weight
F16
F16
[128, 4096]
blk.8.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.8.time_mix_w0.weight
F32
F32
[4096]
blk.8.time_mix_w1.weight
F32
F32
[4096, 192]
blk.8.time_mix_w2.weight
F32
F32
[192, 4096]
blk.9
blk.9.attn_norm.bias
F32
F32
[4096]
blk.9.attn_norm.weight
F32
F32
[4096]
blk.9.attn_norm_2.bias
F32
F32
[4096]
blk.9.attn_norm_2.weight
F32
F32
[4096]
blk.9.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.9.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.9.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.9.time_mix_a0.weight
F32
F32
[4096]
blk.9.time_mix_a1.weight
F16
F16
[4096, 192]
blk.9.time_mix_a2.weight
F16
F16
[192, 4096]
blk.9.time_mix_g1.weight
F16
F16
[4096, 384]
blk.9.time_mix_g2.weight
F16
F16
[384, 4096]
blk.9.time_mix_k_a.weight
F32
F32
[4096]
blk.9.time_mix_k_k.weight
F32
F32
[4096]
blk.9.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.9.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.9.time_mix_ln.bias
F32
F32
[4096]
blk.9.time_mix_ln.weight
F32
F32
[4096]
blk.9.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.9.time_mix_r_k.weight
F32
F32
[4096]
blk.9.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.9.time_mix_v0.weight
F32
F32
[4096]
blk.9.time_mix_v1.weight
F16
F16
[4096, 128]
blk.9.time_mix_v2.weight
F16
F16
[128, 4096]
blk.9.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.9.time_mix_w0.weight
F32
F32
[4096]
blk.9.time_mix_w1.weight
F32
F32
[4096, 192]
blk.9.time_mix_w2.weight
F32
F32
[192, 4096]
blk.10
blk.10.attn_norm.bias
F32
F32
[4096]
blk.10.attn_norm.weight
F32
F32
[4096]
blk.10.attn_norm_2.bias
F32
F32
[4096]
blk.10.attn_norm_2.weight
F32
F32
[4096]
blk.10.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.10.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.10.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.10.time_mix_a0.weight
F32
F32
[4096]
blk.10.time_mix_a1.weight
F16
F16
[4096, 192]
blk.10.time_mix_a2.weight
F16
F16
[192, 4096]
blk.10.time_mix_g1.weight
F16
F16
[4096, 384]
blk.10.time_mix_g2.weight
F16
F16
[384, 4096]
blk.10.time_mix_k_a.weight
F32
F32
[4096]
blk.10.time_mix_k_k.weight
F32
F32
[4096]
blk.10.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.10.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.10.time_mix_ln.bias
F32
F32
[4096]
blk.10.time_mix_ln.weight
F32
F32
[4096]
blk.10.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.10.time_mix_r_k.weight
F32
F32
[4096]
blk.10.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.10.time_mix_v0.weight
F32
F32
[4096]
blk.10.time_mix_v1.weight
F16
F16
[4096, 128]
blk.10.time_mix_v2.weight
F16
F16
[128, 4096]
blk.10.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.10.time_mix_w0.weight
F32
F32
[4096]
blk.10.time_mix_w1.weight
F32
F32
[4096, 192]
blk.10.time_mix_w2.weight
F32
F32
[192, 4096]
blk.11
blk.11.attn_norm.bias
F32
F32
[4096]
blk.11.attn_norm.weight
F32
F32
[4096]
blk.11.attn_norm_2.bias
F32
F32
[4096]
blk.11.attn_norm_2.weight
F32
F32
[4096]
blk.11.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.11.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.11.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.11.time_mix_a0.weight
F32
F32
[4096]
blk.11.time_mix_a1.weight
F16
F16
[4096, 192]
blk.11.time_mix_a2.weight
F16
F16
[192, 4096]
blk.11.time_mix_g1.weight
F16
F16
[4096, 384]
blk.11.time_mix_g2.weight
F16
F16
[384, 4096]
blk.11.time_mix_k_a.weight
F32
F32
[4096]
blk.11.time_mix_k_k.weight
F32
F32
[4096]
blk.11.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.11.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.11.time_mix_ln.bias
F32
F32
[4096]
blk.11.time_mix_ln.weight
F32
F32
[4096]
blk.11.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.11.time_mix_r_k.weight
F32
F32
[4096]
blk.11.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.11.time_mix_v0.weight
F32
F32
[4096]
blk.11.time_mix_v1.weight
F16
F16
[4096, 128]
blk.11.time_mix_v2.weight
F16
F16
[128, 4096]
blk.11.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.11.time_mix_w0.weight
F32
F32
[4096]
blk.11.time_mix_w1.weight
F32
F32
[4096, 192]
blk.11.time_mix_w2.weight
F32
F32
[192, 4096]
blk.12
blk.12.attn_norm.bias
F32
F32
[4096]
blk.12.attn_norm.weight
F32
F32
[4096]
blk.12.attn_norm_2.bias
F32
F32
[4096]
blk.12.attn_norm_2.weight
F32
F32
[4096]
blk.12.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.12.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.12.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.12.time_mix_a0.weight
F32
F32
[4096]
blk.12.time_mix_a1.weight
F16
F16
[4096, 192]
blk.12.time_mix_a2.weight
F16
F16
[192, 4096]
blk.12.time_mix_g1.weight
F16
F16
[4096, 384]
blk.12.time_mix_g2.weight
F16
F16
[384, 4096]
blk.12.time_mix_k_a.weight
F32
F32
[4096]
blk.12.time_mix_k_k.weight
F32
F32
[4096]
blk.12.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.12.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.12.time_mix_ln.bias
F32
F32
[4096]
blk.12.time_mix_ln.weight
F32
F32
[4096]
blk.12.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.12.time_mix_r_k.weight
F32
F32
[4096]
blk.12.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.12.time_mix_v0.weight
F32
F32
[4096]
blk.12.time_mix_v1.weight
F16
F16
[4096, 128]
blk.12.time_mix_v2.weight
F16
F16
[128, 4096]
blk.12.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.12.time_mix_w0.weight
F32
F32
[4096]
blk.12.time_mix_w1.weight
F32
F32
[4096, 192]
blk.12.time_mix_w2.weight
F32
F32
[192, 4096]
blk.13
blk.13.attn_norm.bias
F32
F32
[4096]
blk.13.attn_norm.weight
F32
F32
[4096]
blk.13.attn_norm_2.bias
F32
F32
[4096]
blk.13.attn_norm_2.weight
F32
F32
[4096]
blk.13.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.13.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.13.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.13.time_mix_a0.weight
F32
F32
[4096]
blk.13.time_mix_a1.weight
F16
F16
[4096, 192]
blk.13.time_mix_a2.weight
F16
F16
[192, 4096]
blk.13.time_mix_g1.weight
F16
F16
[4096, 384]
blk.13.time_mix_g2.weight
F16
F16
[384, 4096]
blk.13.time_mix_k_a.weight
F32
F32
[4096]
blk.13.time_mix_k_k.weight
F32
F32
[4096]
blk.13.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.13.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.13.time_mix_ln.bias
F32
F32
[4096]
blk.13.time_mix_ln.weight
F32
F32
[4096]
blk.13.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.13.time_mix_r_k.weight
F32
F32
[4096]
blk.13.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.13.time_mix_v0.weight
F32
F32
[4096]
blk.13.time_mix_v1.weight
F16
F16
[4096, 128]
blk.13.time_mix_v2.weight
F16
F16
[128, 4096]
blk.13.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.13.time_mix_w0.weight
F32
F32
[4096]
blk.13.time_mix_w1.weight
F32
F32
[4096, 192]
blk.13.time_mix_w2.weight
F32
F32
[192, 4096]
blk.14
blk.14.attn_norm.bias
F32
F32
[4096]
blk.14.attn_norm.weight
F32
F32
[4096]
blk.14.attn_norm_2.bias
F32
F32
[4096]
blk.14.attn_norm_2.weight
F32
F32
[4096]
blk.14.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.14.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.14.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.14.time_mix_a0.weight
F32
F32
[4096]
blk.14.time_mix_a1.weight
F16
F16
[4096, 192]
blk.14.time_mix_a2.weight
F16
F16
[192, 4096]
blk.14.time_mix_g1.weight
F16
F16
[4096, 384]
blk.14.time_mix_g2.weight
F16
F16
[384, 4096]
blk.14.time_mix_k_a.weight
F32
F32
[4096]
blk.14.time_mix_k_k.weight
F32
F32
[4096]
blk.14.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.14.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.14.time_mix_ln.bias
F32
F32
[4096]
blk.14.time_mix_ln.weight
F32
F32
[4096]
blk.14.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.14.time_mix_r_k.weight
F32
F32
[4096]
blk.14.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.14.time_mix_v0.weight
F32
F32
[4096]
blk.14.time_mix_v1.weight
F16
F16
[4096, 128]
blk.14.time_mix_v2.weight
F16
F16
[128, 4096]
blk.14.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.14.time_mix_w0.weight
F32
F32
[4096]
blk.14.time_mix_w1.weight
F32
F32
[4096, 192]
blk.14.time_mix_w2.weight
F32
F32
[192, 4096]
blk.15
blk.15.attn_norm.bias
F32
F32
[4096]
blk.15.attn_norm.weight
F32
F32
[4096]
blk.15.attn_norm_2.bias
F32
F32
[4096]
blk.15.attn_norm_2.weight
F32
F32
[4096]
blk.15.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.15.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.15.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.15.time_mix_a0.weight
F32
F32
[4096]
blk.15.time_mix_a1.weight
F16
F16
[4096, 192]
blk.15.time_mix_a2.weight
F16
F16
[192, 4096]
blk.15.time_mix_g1.weight
F16
F16
[4096, 384]
blk.15.time_mix_g2.weight
F16
F16
[384, 4096]
blk.15.time_mix_k_a.weight
F32
F32
[4096]
blk.15.time_mix_k_k.weight
F32
F32
[4096]
blk.15.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.15.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.15.time_mix_ln.bias
F32
F32
[4096]
blk.15.time_mix_ln.weight
F32
F32
[4096]
blk.15.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.15.time_mix_r_k.weight
F32
F32
[4096]
blk.15.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.15.time_mix_v0.weight
F32
F32
[4096]
blk.15.time_mix_v1.weight
F16
F16
[4096, 128]
blk.15.time_mix_v2.weight
F16
F16
[128, 4096]
blk.15.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.15.time_mix_w0.weight
F32
F32
[4096]
blk.15.time_mix_w1.weight
F32
F32
[4096, 192]
blk.15.time_mix_w2.weight
F32
F32
[192, 4096]
blk.16
blk.16.attn_norm.bias
F32
F32
[4096]
blk.16.attn_norm.weight
F32
F32
[4096]
blk.16.attn_norm_2.bias
F32
F32
[4096]
blk.16.attn_norm_2.weight
F32
F32
[4096]
blk.16.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.16.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.16.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.16.time_mix_a0.weight
F32
F32
[4096]
blk.16.time_mix_a1.weight
F16
F16
[4096, 192]
blk.16.time_mix_a2.weight
F16
F16
[192, 4096]
blk.16.time_mix_g1.weight
F16
F16
[4096, 384]
blk.16.time_mix_g2.weight
F16
F16
[384, 4096]
blk.16.time_mix_k_a.weight
F32
F32
[4096]
blk.16.time_mix_k_k.weight
F32
F32
[4096]
blk.16.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.16.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.16.time_mix_ln.bias
F32
F32
[4096]
blk.16.time_mix_ln.weight
F32
F32
[4096]
blk.16.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.16.time_mix_r_k.weight
F32
F32
[4096]
blk.16.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.16.time_mix_v0.weight
F32
F32
[4096]
blk.16.time_mix_v1.weight
F16
F16
[4096, 128]
blk.16.time_mix_v2.weight
F16
F16
[128, 4096]
blk.16.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.16.time_mix_w0.weight
F32
F32
[4096]
blk.16.time_mix_w1.weight
F32
F32
[4096, 192]
blk.16.time_mix_w2.weight
F32
F32
[192, 4096]
blk.17
blk.17.attn_norm.bias
F32
F32
[4096]
blk.17.attn_norm.weight
F32
F32
[4096]
blk.17.attn_norm_2.bias
F32
F32
[4096]
blk.17.attn_norm_2.weight
F32
F32
[4096]
blk.17.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.17.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.17.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.17.time_mix_a0.weight
F32
F32
[4096]
blk.17.time_mix_a1.weight
F16
F16
[4096, 192]
blk.17.time_mix_a2.weight
F16
F16
[192, 4096]
blk.17.time_mix_g1.weight
F16
F16
[4096, 384]
blk.17.time_mix_g2.weight
F16
F16
[384, 4096]
blk.17.time_mix_k_a.weight
F32
F32
[4096]
blk.17.time_mix_k_k.weight
F32
F32
[4096]
blk.17.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.17.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.17.time_mix_ln.bias
F32
F32
[4096]
blk.17.time_mix_ln.weight
F32
F32
[4096]
blk.17.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.17.time_mix_r_k.weight
F32
F32
[4096]
blk.17.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.17.time_mix_v0.weight
F32
F32
[4096]
blk.17.time_mix_v1.weight
F16
F16
[4096, 128]
blk.17.time_mix_v2.weight
F16
F16
[128, 4096]
blk.17.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.17.time_mix_w0.weight
F32
F32
[4096]
blk.17.time_mix_w1.weight
F32
F32
[4096, 192]
blk.17.time_mix_w2.weight
F32
F32
[192, 4096]
blk.18
blk.18.attn_norm.bias
F32
F32
[4096]
blk.18.attn_norm.weight
F32
F32
[4096]
blk.18.attn_norm_2.bias
F32
F32
[4096]
blk.18.attn_norm_2.weight
F32
F32
[4096]
blk.18.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.18.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.18.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.18.time_mix_a0.weight
F32
F32
[4096]
blk.18.time_mix_a1.weight
F16
F16
[4096, 192]
blk.18.time_mix_a2.weight
F16
F16
[192, 4096]
blk.18.time_mix_g1.weight
F16
F16
[4096, 384]
blk.18.time_mix_g2.weight
F16
F16
[384, 4096]
blk.18.time_mix_k_a.weight
F32
F32
[4096]
blk.18.time_mix_k_k.weight
F32
F32
[4096]
blk.18.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.18.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.18.time_mix_ln.bias
F32
F32
[4096]
blk.18.time_mix_ln.weight
F32
F32
[4096]
blk.18.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.18.time_mix_r_k.weight
F32
F32
[4096]
blk.18.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.18.time_mix_v0.weight
F32
F32
[4096]
blk.18.time_mix_v1.weight
F16
F16
[4096, 128]
blk.18.time_mix_v2.weight
F16
F16
[128, 4096]
blk.18.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.18.time_mix_w0.weight
F32
F32
[4096]
blk.18.time_mix_w1.weight
F32
F32
[4096, 192]
blk.18.time_mix_w2.weight
F32
F32
[192, 4096]
blk.19
blk.19.attn_norm.bias
F32
F32
[4096]
blk.19.attn_norm.weight
F32
F32
[4096]
blk.19.attn_norm_2.bias
F32
F32
[4096]
blk.19.attn_norm_2.weight
F32
F32
[4096]
blk.19.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.19.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.19.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.19.time_mix_a0.weight
F32
F32
[4096]
blk.19.time_mix_a1.weight
F16
F16
[4096, 192]
blk.19.time_mix_a2.weight
F16
F16
[192, 4096]
blk.19.time_mix_g1.weight
F16
F16
[4096, 384]
blk.19.time_mix_g2.weight
F16
F16
[384, 4096]
blk.19.time_mix_k_a.weight
F32
F32
[4096]
blk.19.time_mix_k_k.weight
F32
F32
[4096]
blk.19.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.19.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.19.time_mix_ln.bias
F32
F32
[4096]
blk.19.time_mix_ln.weight
F32
F32
[4096]
blk.19.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.19.time_mix_r_k.weight
F32
F32
[4096]
blk.19.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.19.time_mix_v0.weight
F32
F32
[4096]
blk.19.time_mix_v1.weight
F16
F16
[4096, 128]
blk.19.time_mix_v2.weight
F16
F16
[128, 4096]
blk.19.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.19.time_mix_w0.weight
F32
F32
[4096]
blk.19.time_mix_w1.weight
F32
F32
[4096, 192]
blk.19.time_mix_w2.weight
F32
F32
[192, 4096]
blk.20
blk.20.attn_norm.bias
F32
F32
[4096]
blk.20.attn_norm.weight
F32
F32
[4096]
blk.20.attn_norm_2.bias
F32
F32
[4096]
blk.20.attn_norm_2.weight
F32
F32
[4096]
blk.20.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.20.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.20.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.20.time_mix_a0.weight
F32
F32
[4096]
blk.20.time_mix_a1.weight
F16
F16
[4096, 192]
blk.20.time_mix_a2.weight
F16
F16
[192, 4096]
blk.20.time_mix_g1.weight
F16
F16
[4096, 384]
blk.20.time_mix_g2.weight
F16
F16
[384, 4096]
blk.20.time_mix_k_a.weight
F32
F32
[4096]
blk.20.time_mix_k_k.weight
F32
F32
[4096]
blk.20.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.20.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.20.time_mix_ln.bias
F32
F32
[4096]
blk.20.time_mix_ln.weight
F32
F32
[4096]
blk.20.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.20.time_mix_r_k.weight
F32
F32
[4096]
blk.20.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.20.time_mix_v0.weight
F32
F32
[4096]
blk.20.time_mix_v1.weight
F16
F16
[4096, 128]
blk.20.time_mix_v2.weight
F16
F16
[128, 4096]
blk.20.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.20.time_mix_w0.weight
F32
F32
[4096]
blk.20.time_mix_w1.weight
F32
F32
[4096, 192]
blk.20.time_mix_w2.weight
F32
F32
[192, 4096]
blk.21
blk.21.attn_norm.bias
F32
F32
[4096]
blk.21.attn_norm.weight
F32
F32
[4096]
blk.21.attn_norm_2.bias
F32
F32
[4096]
blk.21.attn_norm_2.weight
F32
F32
[4096]
blk.21.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.21.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.21.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.21.time_mix_a0.weight
F32
F32
[4096]
blk.21.time_mix_a1.weight
F16
F16
[4096, 192]
blk.21.time_mix_a2.weight
F16
F16
[192, 4096]
blk.21.time_mix_g1.weight
F16
F16
[4096, 384]
blk.21.time_mix_g2.weight
F16
F16
[384, 4096]
blk.21.time_mix_k_a.weight
F32
F32
[4096]
blk.21.time_mix_k_k.weight
F32
F32
[4096]
blk.21.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.21.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.21.time_mix_ln.bias
F32
F32
[4096]
blk.21.time_mix_ln.weight
F32
F32
[4096]
blk.21.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.21.time_mix_r_k.weight
F32
F32
[4096]
blk.21.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.21.time_mix_v0.weight
F32
F32
[4096]
blk.21.time_mix_v1.weight
F16
F16
[4096, 128]
blk.21.time_mix_v2.weight
F16
F16
[128, 4096]
blk.21.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.21.time_mix_w0.weight
F32
F32
[4096]
blk.21.time_mix_w1.weight
F32
F32
[4096, 192]
blk.21.time_mix_w2.weight
F32
F32
[192, 4096]
blk.22
blk.22.attn_norm.bias
F32
F32
[4096]
blk.22.attn_norm.weight
F32
F32
[4096]
blk.22.attn_norm_2.bias
F32
F32
[4096]
blk.22.attn_norm_2.weight
F32
F32
[4096]
blk.22.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.22.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.22.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.22.time_mix_a0.weight
F32
F32
[4096]
blk.22.time_mix_a1.weight
F16
F16
[4096, 192]
blk.22.time_mix_a2.weight
F16
F16
[192, 4096]
blk.22.time_mix_g1.weight
F16
F16
[4096, 384]
blk.22.time_mix_g2.weight
F16
F16
[384, 4096]
blk.22.time_mix_k_a.weight
F32
F32
[4096]
blk.22.time_mix_k_k.weight
F32
F32
[4096]
blk.22.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.22.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.22.time_mix_ln.bias
F32
F32
[4096]
blk.22.time_mix_ln.weight
F32
F32
[4096]
blk.22.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.22.time_mix_r_k.weight
F32
F32
[4096]
blk.22.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.22.time_mix_v0.weight
F32
F32
[4096]
blk.22.time_mix_v1.weight
F16
F16
[4096, 128]
blk.22.time_mix_v2.weight
F16
F16
[128, 4096]
blk.22.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.22.time_mix_w0.weight
F32
F32
[4096]
blk.22.time_mix_w1.weight
F32
F32
[4096, 192]
blk.22.time_mix_w2.weight
F32
F32
[192, 4096]
blk.23
blk.23.attn_norm.bias
F32
F32
[4096]
blk.23.attn_norm.weight
F32
F32
[4096]
blk.23.attn_norm_2.bias
F32
F32
[4096]
blk.23.attn_norm_2.weight
F32
F32
[4096]
blk.23.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.23.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.23.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.23.time_mix_a0.weight
F32
F32
[4096]
blk.23.time_mix_a1.weight
F16
F16
[4096, 192]
blk.23.time_mix_a2.weight
F16
F16
[192, 4096]
blk.23.time_mix_g1.weight
F16
F16
[4096, 384]
blk.23.time_mix_g2.weight
F16
F16
[384, 4096]
blk.23.time_mix_k_a.weight
F32
F32
[4096]
blk.23.time_mix_k_k.weight
F32
F32
[4096]
blk.23.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.23.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.23.time_mix_ln.bias
F32
F32
[4096]
blk.23.time_mix_ln.weight
F32
F32
[4096]
blk.23.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.23.time_mix_r_k.weight
F32
F32
[4096]
blk.23.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.23.time_mix_v0.weight
F32
F32
[4096]
blk.23.time_mix_v1.weight
F16
F16
[4096, 128]
blk.23.time_mix_v2.weight
F16
F16
[128, 4096]
blk.23.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.23.time_mix_w0.weight
F32
F32
[4096]
blk.23.time_mix_w1.weight
F32
F32
[4096, 192]
blk.23.time_mix_w2.weight
F32
F32
[192, 4096]
blk.24
blk.24.attn_norm.bias
F32
F32
[4096]
blk.24.attn_norm.weight
F32
F32
[4096]
blk.24.attn_norm_2.bias
F32
F32
[4096]
blk.24.attn_norm_2.weight
F32
F32
[4096]
blk.24.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.24.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.24.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.24.time_mix_a0.weight
F32
F32
[4096]
blk.24.time_mix_a1.weight
F16
F16
[4096, 192]
blk.24.time_mix_a2.weight
F16
F16
[192, 4096]
blk.24.time_mix_g1.weight
F16
F16
[4096, 384]
blk.24.time_mix_g2.weight
F16
F16
[384, 4096]
blk.24.time_mix_k_a.weight
F32
F32
[4096]
blk.24.time_mix_k_k.weight
F32
F32
[4096]
blk.24.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.24.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.24.time_mix_ln.bias
F32
F32
[4096]
blk.24.time_mix_ln.weight
F32
F32
[4096]
blk.24.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.24.time_mix_r_k.weight
F32
F32
[4096]
blk.24.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.24.time_mix_v0.weight
F32
F32
[4096]
blk.24.time_mix_v1.weight
F16
F16
[4096, 128]
blk.24.time_mix_v2.weight
F16
F16
[128, 4096]
blk.24.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.24.time_mix_w0.weight
F32
F32
[4096]
blk.24.time_mix_w1.weight
F32
F32
[4096, 192]
blk.24.time_mix_w2.weight
F32
F32
[192, 4096]
blk.25
blk.25.attn_norm.bias
F32
F32
[4096]
blk.25.attn_norm.weight
F32
F32
[4096]
blk.25.attn_norm_2.bias
F32
F32
[4096]
blk.25.attn_norm_2.weight
F32
F32
[4096]
blk.25.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.25.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.25.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.25.time_mix_a0.weight
F32
F32
[4096]
blk.25.time_mix_a1.weight
F16
F16
[4096, 192]
blk.25.time_mix_a2.weight
F16
F16
[192, 4096]
blk.25.time_mix_g1.weight
F16
F16
[4096, 384]
blk.25.time_mix_g2.weight
F16
F16
[384, 4096]
blk.25.time_mix_k_a.weight
F32
F32
[4096]
blk.25.time_mix_k_k.weight
F32
F32
[4096]
blk.25.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.25.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.25.time_mix_ln.bias
F32
F32
[4096]
blk.25.time_mix_ln.weight
F32
F32
[4096]
blk.25.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.25.time_mix_r_k.weight
F32
F32
[4096]
blk.25.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.25.time_mix_v0.weight
F32
F32
[4096]
blk.25.time_mix_v1.weight
F16
F16
[4096, 128]
blk.25.time_mix_v2.weight
F16
F16
[128, 4096]
blk.25.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.25.time_mix_w0.weight
F32
F32
[4096]
blk.25.time_mix_w1.weight
F32
F32
[4096, 192]
blk.25.time_mix_w2.weight
F32
F32
[192, 4096]
blk.26
blk.26.attn_norm.bias
F32
F32
[4096]
blk.26.attn_norm.weight
F32
F32
[4096]
blk.26.attn_norm_2.bias
F32
F32
[4096]
blk.26.attn_norm_2.weight
F32
F32
[4096]
blk.26.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.26.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.26.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.26.time_mix_a0.weight
F32
F32
[4096]
blk.26.time_mix_a1.weight
F16
F16
[4096, 192]
blk.26.time_mix_a2.weight
F16
F16
[192, 4096]
blk.26.time_mix_g1.weight
F16
F16
[4096, 384]
blk.26.time_mix_g2.weight
F16
F16
[384, 4096]
blk.26.time_mix_k_a.weight
F32
F32
[4096]
blk.26.time_mix_k_k.weight
F32
F32
[4096]
blk.26.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.26.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.26.time_mix_ln.bias
F32
F32
[4096]
blk.26.time_mix_ln.weight
F32
F32
[4096]
blk.26.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.26.time_mix_r_k.weight
F32
F32
[4096]
blk.26.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.26.time_mix_v0.weight
F32
F32
[4096]
blk.26.time_mix_v1.weight
F16
F16
[4096, 128]
blk.26.time_mix_v2.weight
F16
F16
[128, 4096]
blk.26.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.26.time_mix_w0.weight
F32
F32
[4096]
blk.26.time_mix_w1.weight
F32
F32
[4096, 192]
blk.26.time_mix_w2.weight
F32
F32
[192, 4096]
blk.27
blk.27.attn_norm.bias
F32
F32
[4096]
blk.27.attn_norm.weight
F32
F32
[4096]
blk.27.attn_norm_2.bias
F32
F32
[4096]
blk.27.attn_norm_2.weight
F32
F32
[4096]
blk.27.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.27.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.27.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.27.time_mix_a0.weight
F32
F32
[4096]
blk.27.time_mix_a1.weight
F16
F16
[4096, 192]
blk.27.time_mix_a2.weight
F16
F16
[192, 4096]
blk.27.time_mix_g1.weight
F16
F16
[4096, 384]
blk.27.time_mix_g2.weight
F16
F16
[384, 4096]
blk.27.time_mix_k_a.weight
F32
F32
[4096]
blk.27.time_mix_k_k.weight
F32
F32
[4096]
blk.27.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.27.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.27.time_mix_ln.bias
F32
F32
[4096]
blk.27.time_mix_ln.weight
F32
F32
[4096]
blk.27.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.27.time_mix_r_k.weight
F32
F32
[4096]
blk.27.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.27.time_mix_v0.weight
F32
F32
[4096]
blk.27.time_mix_v1.weight
F16
F16
[4096, 128]
blk.27.time_mix_v2.weight
F16
F16
[128, 4096]
blk.27.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.27.time_mix_w0.weight
F32
F32
[4096]
blk.27.time_mix_w1.weight
F32
F32
[4096, 192]
blk.27.time_mix_w2.weight
F32
F32
[192, 4096]
blk.28
blk.28.attn_norm.bias
F32
F32
[4096]
blk.28.attn_norm.weight
F32
F32
[4096]
blk.28.attn_norm_2.bias
F32
F32
[4096]
blk.28.attn_norm_2.weight
F32
F32
[4096]
blk.28.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.28.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.28.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.28.time_mix_a0.weight
F32
F32
[4096]
blk.28.time_mix_a1.weight
F16
F16
[4096, 192]
blk.28.time_mix_a2.weight
F16
F16
[192, 4096]
blk.28.time_mix_g1.weight
F16
F16
[4096, 384]
blk.28.time_mix_g2.weight
F16
F16
[384, 4096]
blk.28.time_mix_k_a.weight
F32
F32
[4096]
blk.28.time_mix_k_k.weight
F32
F32
[4096]
blk.28.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.28.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.28.time_mix_ln.bias
F32
F32
[4096]
blk.28.time_mix_ln.weight
F32
F32
[4096]
blk.28.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.28.time_mix_r_k.weight
F32
F32
[4096]
blk.28.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.28.time_mix_v0.weight
F32
F32
[4096]
blk.28.time_mix_v1.weight
F16
F16
[4096, 128]
blk.28.time_mix_v2.weight
F16
F16
[128, 4096]
blk.28.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.28.time_mix_w0.weight
F32
F32
[4096]
blk.28.time_mix_w1.weight
F32
F32
[4096, 192]
blk.28.time_mix_w2.weight
F32
F32
[192, 4096]
blk.29
blk.29.attn_norm.bias
F32
F32
[4096]
blk.29.attn_norm.weight
F32
F32
[4096]
blk.29.attn_norm_2.bias
F32
F32
[4096]
blk.29.attn_norm_2.weight
F32
F32
[4096]
blk.29.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.29.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.29.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.29.time_mix_a0.weight
F32
F32
[4096]
blk.29.time_mix_a1.weight
F16
F16
[4096, 192]
blk.29.time_mix_a2.weight
F16
F16
[192, 4096]
blk.29.time_mix_g1.weight
F16
F16
[4096, 384]
blk.29.time_mix_g2.weight
F16
F16
[384, 4096]
blk.29.time_mix_k_a.weight
F32
F32
[4096]
blk.29.time_mix_k_k.weight
F32
F32
[4096]
blk.29.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.29.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.29.time_mix_ln.bias
F32
F32
[4096]
blk.29.time_mix_ln.weight
F32
F32
[4096]
blk.29.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.29.time_mix_r_k.weight
F32
F32
[4096]
blk.29.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.29.time_mix_v0.weight
F32
F32
[4096]
blk.29.time_mix_v1.weight
F16
F16
[4096, 128]
blk.29.time_mix_v2.weight
F16
F16
[128, 4096]
blk.29.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.29.time_mix_w0.weight
F32
F32
[4096]
blk.29.time_mix_w1.weight
F32
F32
[4096, 192]
blk.29.time_mix_w2.weight
F32
F32
[192, 4096]
blk.30
blk.30.attn_norm.bias
F32
F32
[4096]
blk.30.attn_norm.weight
F32
F32
[4096]
blk.30.attn_norm_2.bias
F32
F32
[4096]
blk.30.attn_norm_2.weight
F32
F32
[4096]
blk.30.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.30.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.30.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.30.time_mix_a0.weight
F32
F32
[4096]
blk.30.time_mix_a1.weight
F16
F16
[4096, 192]
blk.30.time_mix_a2.weight
F16
F16
[192, 4096]
blk.30.time_mix_g1.weight
F16
F16
[4096, 384]
blk.30.time_mix_g2.weight
F16
F16
[384, 4096]
blk.30.time_mix_k_a.weight
F32
F32
[4096]
blk.30.time_mix_k_k.weight
F32
F32
[4096]
blk.30.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.30.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.30.time_mix_ln.bias
F32
F32
[4096]
blk.30.time_mix_ln.weight
F32
F32
[4096]
blk.30.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.30.time_mix_r_k.weight
F32
F32
[4096]
blk.30.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.30.time_mix_v0.weight
F32
F32
[4096]
blk.30.time_mix_v1.weight
F16
F16
[4096, 128]
blk.30.time_mix_v2.weight
F16
F16
[128, 4096]
blk.30.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.30.time_mix_w0.weight
F32
F32
[4096]
blk.30.time_mix_w1.weight
F32
F32
[4096, 192]
blk.30.time_mix_w2.weight
F32
F32
[192, 4096]
blk.31
blk.31.attn_norm.bias
F32
F32
[4096]
blk.31.attn_norm.weight
F32
F32
[4096]
blk.31.attn_norm_2.bias
F32
F32
[4096]
blk.31.attn_norm_2.weight
F32
F32
[4096]
blk.31.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.31.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.31.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.31.time_mix_a0.weight
F32
F32
[4096]
blk.31.time_mix_a1.weight
F16
F16
[4096, 192]
blk.31.time_mix_a2.weight
F16
F16
[192, 4096]
blk.31.time_mix_g1.weight
F16
F16
[4096, 384]
blk.31.time_mix_g2.weight
F16
F16
[384, 4096]
blk.31.time_mix_k_a.weight
F32
F32
[4096]
blk.31.time_mix_k_k.weight
F32
F32
[4096]
blk.31.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.31.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.31.time_mix_ln.bias
F32
F32
[4096]
blk.31.time_mix_ln.weight
F32
F32
[4096]
blk.31.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.31.time_mix_r_k.weight
F32
F32
[4096]
blk.31.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.31.time_mix_v0.weight
F32
F32
[4096]
blk.31.time_mix_v1.weight
F16
F16
[4096, 128]
blk.31.time_mix_v2.weight
F16
F16
[128, 4096]
blk.31.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.31.time_mix_w0.weight
F32
F32
[4096]
blk.31.time_mix_w1.weight
F32
F32
[4096, 192]
blk.31.time_mix_w2.weight
F32
F32
[192, 4096]
blk.32
blk.32.attn_norm.bias
F32
F32
[4096]
blk.32.attn_norm.weight
F32
F32
[4096]
blk.32.attn_norm_2.bias
F32
F32
[4096]
blk.32.attn_norm_2.weight
F32
F32
[4096]
blk.32.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.32.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.32.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.32.time_mix_a0.weight
F32
F32
[4096]
blk.32.time_mix_a1.weight
F16
F16
[4096, 192]
blk.32.time_mix_a2.weight
F16
F16
[192, 4096]
blk.32.time_mix_g1.weight
F16
F16
[4096, 384]
blk.32.time_mix_g2.weight
F16
F16
[384, 4096]
blk.32.time_mix_k_a.weight
F32
F32
[4096]
blk.32.time_mix_k_k.weight
F32
F32
[4096]
blk.32.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.32.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.32.time_mix_ln.bias
F32
F32
[4096]
blk.32.time_mix_ln.weight
F32
F32
[4096]
blk.32.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.32.time_mix_r_k.weight
F32
F32
[4096]
blk.32.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.32.time_mix_v0.weight
F32
F32
[4096]
blk.32.time_mix_v1.weight
F16
F16
[4096, 128]
blk.32.time_mix_v2.weight
F16
F16
[128, 4096]
blk.32.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.32.time_mix_w0.weight
F32
F32
[4096]
blk.32.time_mix_w1.weight
F32
F32
[4096, 192]
blk.32.time_mix_w2.weight
F32
F32
[192, 4096]
blk.33
blk.33.attn_norm.bias
F32
F32
[4096]
blk.33.attn_norm.weight
F32
F32
[4096]
blk.33.attn_norm_2.bias
F32
F32
[4096]
blk.33.attn_norm_2.weight
F32
F32
[4096]
blk.33.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.33.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.33.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.33.time_mix_a0.weight
F32
F32
[4096]
blk.33.time_mix_a1.weight
F16
F16
[4096, 192]
blk.33.time_mix_a2.weight
F16
F16
[192, 4096]
blk.33.time_mix_g1.weight
F16
F16
[4096, 384]
blk.33.time_mix_g2.weight
F16
F16
[384, 4096]
blk.33.time_mix_k_a.weight
F32
F32
[4096]
blk.33.time_mix_k_k.weight
F32
F32
[4096]
blk.33.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.33.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.33.time_mix_ln.bias
F32
F32
[4096]
blk.33.time_mix_ln.weight
F32
F32
[4096]
blk.33.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.33.time_mix_r_k.weight
F32
F32
[4096]
blk.33.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.33.time_mix_v0.weight
F32
F32
[4096]
blk.33.time_mix_v1.weight
F16
F16
[4096, 128]
blk.33.time_mix_v2.weight
F16
F16
[128, 4096]
blk.33.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.33.time_mix_w0.weight
F32
F32
[4096]
blk.33.time_mix_w1.weight
F32
F32
[4096, 192]
blk.33.time_mix_w2.weight
F32
F32
[192, 4096]
blk.34
blk.34.attn_norm.bias
F32
F32
[4096]
blk.34.attn_norm.weight
F32
F32
[4096]
blk.34.attn_norm_2.bias
F32
F32
[4096]
blk.34.attn_norm_2.weight
F32
F32
[4096]
blk.34.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.34.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.34.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.34.time_mix_a0.weight
F32
F32
[4096]
blk.34.time_mix_a1.weight
F16
F16
[4096, 192]
blk.34.time_mix_a2.weight
F16
F16
[192, 4096]
blk.34.time_mix_g1.weight
F16
F16
[4096, 384]
blk.34.time_mix_g2.weight
F16
F16
[384, 4096]
blk.34.time_mix_k_a.weight
F32
F32
[4096]
blk.34.time_mix_k_k.weight
F32
F32
[4096]
blk.34.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.34.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.34.time_mix_ln.bias
F32
F32
[4096]
blk.34.time_mix_ln.weight
F32
F32
[4096]
blk.34.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.34.time_mix_r_k.weight
F32
F32
[4096]
blk.34.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.34.time_mix_v0.weight
F32
F32
[4096]
blk.34.time_mix_v1.weight
F16
F16
[4096, 128]
blk.34.time_mix_v2.weight
F16
F16
[128, 4096]
blk.34.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.34.time_mix_w0.weight
F32
F32
[4096]
blk.34.time_mix_w1.weight
F32
F32
[4096, 192]
blk.34.time_mix_w2.weight
F32
F32
[192, 4096]
blk.35
blk.35.attn_norm.bias
F32
F32
[4096]
blk.35.attn_norm.weight
F32
F32
[4096]
blk.35.attn_norm_2.bias
F32
F32
[4096]
blk.35.attn_norm_2.weight
F32
F32
[4096]
blk.35.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.35.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.35.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.35.time_mix_a0.weight
F32
F32
[4096]
blk.35.time_mix_a1.weight
F16
F16
[4096, 192]
blk.35.time_mix_a2.weight
F16
F16
[192, 4096]
blk.35.time_mix_g1.weight
F16
F16
[4096, 384]
blk.35.time_mix_g2.weight
F16
F16
[384, 4096]
blk.35.time_mix_k_a.weight
F32
F32
[4096]
blk.35.time_mix_k_k.weight
F32
F32
[4096]
blk.35.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.35.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.35.time_mix_ln.bias
F32
F32
[4096]
blk.35.time_mix_ln.weight
F32
F32
[4096]
blk.35.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.35.time_mix_r_k.weight
F32
F32
[4096]
blk.35.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.35.time_mix_v0.weight
F32
F32
[4096]
blk.35.time_mix_v1.weight
F16
F16
[4096, 128]
blk.35.time_mix_v2.weight
F16
F16
[128, 4096]
blk.35.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.35.time_mix_w0.weight
F32
F32
[4096]
blk.35.time_mix_w1.weight
F32
F32
[4096, 192]
blk.35.time_mix_w2.weight
F32
F32
[192, 4096]
blk.36
blk.36.attn_norm.bias
F32
F32
[4096]
blk.36.attn_norm.weight
F32
F32
[4096]
blk.36.attn_norm_2.bias
F32
F32
[4096]
blk.36.attn_norm_2.weight
F32
F32
[4096]
blk.36.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.36.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.36.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.36.time_mix_a0.weight
F32
F32
[4096]
blk.36.time_mix_a1.weight
F16
F16
[4096, 192]
blk.36.time_mix_a2.weight
F16
F16
[192, 4096]
blk.36.time_mix_g1.weight
F16
F16
[4096, 384]
blk.36.time_mix_g2.weight
F16
F16
[384, 4096]
blk.36.time_mix_k_a.weight
F32
F32
[4096]
blk.36.time_mix_k_k.weight
F32
F32
[4096]
blk.36.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.36.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.36.time_mix_ln.bias
F32
F32
[4096]
blk.36.time_mix_ln.weight
F32
F32
[4096]
blk.36.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.36.time_mix_r_k.weight
F32
F32
[4096]
blk.36.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.36.time_mix_v0.weight
F32
F32
[4096]
blk.36.time_mix_v1.weight
F16
F16
[4096, 128]
blk.36.time_mix_v2.weight
F16
F16
[128, 4096]
blk.36.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.36.time_mix_w0.weight
F32
F32
[4096]
blk.36.time_mix_w1.weight
F32
F32
[4096, 192]
blk.36.time_mix_w2.weight
F32
F32
[192, 4096]
blk.37
blk.37.attn_norm.bias
F32
F32
[4096]
blk.37.attn_norm.weight
F32
F32
[4096]
blk.37.attn_norm_2.bias
F32
F32
[4096]
blk.37.attn_norm_2.weight
F32
F32
[4096]
blk.37.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.37.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.37.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.37.time_mix_a0.weight
F32
F32
[4096]
blk.37.time_mix_a1.weight
F16
F16
[4096, 192]
blk.37.time_mix_a2.weight
F16
F16
[192, 4096]
blk.37.time_mix_g1.weight
F16
F16
[4096, 384]
blk.37.time_mix_g2.weight
F16
F16
[384, 4096]
blk.37.time_mix_k_a.weight
F32
F32
[4096]
blk.37.time_mix_k_k.weight
F32
F32
[4096]
blk.37.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.37.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.37.time_mix_ln.bias
F32
F32
[4096]
blk.37.time_mix_ln.weight
F32
F32
[4096]
blk.37.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.37.time_mix_r_k.weight
F32
F32
[4096]
blk.37.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.37.time_mix_v0.weight
F32
F32
[4096]
blk.37.time_mix_v1.weight
F16
F16
[4096, 128]
blk.37.time_mix_v2.weight
F16
F16
[128, 4096]
blk.37.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.37.time_mix_w0.weight
F32
F32
[4096]
blk.37.time_mix_w1.weight
F32
F32
[4096, 192]
blk.37.time_mix_w2.weight
F32
F32
[192, 4096]
blk.38
blk.38.attn_norm.bias
F32
F32
[4096]
blk.38.attn_norm.weight
F32
F32
[4096]
blk.38.attn_norm_2.bias
F32
F32
[4096]
blk.38.attn_norm_2.weight
F32
F32
[4096]
blk.38.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.38.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.38.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.38.time_mix_a0.weight
F32
F32
[4096]
blk.38.time_mix_a1.weight
F16
F16
[4096, 192]
blk.38.time_mix_a2.weight
F16
F16
[192, 4096]
blk.38.time_mix_g1.weight
F16
F16
[4096, 384]
blk.38.time_mix_g2.weight
F16
F16
[384, 4096]
blk.38.time_mix_k_a.weight
F32
F32
[4096]
blk.38.time_mix_k_k.weight
F32
F32
[4096]
blk.38.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.38.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.38.time_mix_ln.bias
F32
F32
[4096]
blk.38.time_mix_ln.weight
F32
F32
[4096]
blk.38.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.38.time_mix_r_k.weight
F32
F32
[4096]
blk.38.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.38.time_mix_v0.weight
F32
F32
[4096]
blk.38.time_mix_v1.weight
F16
F16
[4096, 128]
blk.38.time_mix_v2.weight
F16
F16
[128, 4096]
blk.38.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.38.time_mix_w0.weight
F32
F32
[4096]
blk.38.time_mix_w1.weight
F32
F32
[4096, 192]
blk.38.time_mix_w2.weight
F32
F32
[192, 4096]
blk.39
blk.39.attn_norm.bias
F32
F32
[4096]
blk.39.attn_norm.weight
F32
F32
[4096]
blk.39.attn_norm_2.bias
F32
F32
[4096]
blk.39.attn_norm_2.weight
F32
F32
[4096]
blk.39.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.39.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.39.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.39.time_mix_a0.weight
F32
F32
[4096]
blk.39.time_mix_a1.weight
F16
F16
[4096, 192]
blk.39.time_mix_a2.weight
F16
F16
[192, 4096]
blk.39.time_mix_g1.weight
F16
F16
[4096, 384]
blk.39.time_mix_g2.weight
F16
F16
[384, 4096]
blk.39.time_mix_k_a.weight
F32
F32
[4096]
blk.39.time_mix_k_k.weight
F32
F32
[4096]
blk.39.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.39.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.39.time_mix_ln.bias
F32
F32
[4096]
blk.39.time_mix_ln.weight
F32
F32
[4096]
blk.39.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.39.time_mix_r_k.weight
F32
F32
[4096]
blk.39.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.39.time_mix_v0.weight
F32
F32
[4096]
blk.39.time_mix_v1.weight
F16
F16
[4096, 128]
blk.39.time_mix_v2.weight
F16
F16
[128, 4096]
blk.39.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.39.time_mix_w0.weight
F32
F32
[4096]
blk.39.time_mix_w1.weight
F32
F32
[4096, 192]
blk.39.time_mix_w2.weight
F32
F32
[192, 4096]
blk.40
blk.40.attn_norm.bias
F32
F32
[4096]
blk.40.attn_norm.weight
F32
F32
[4096]
blk.40.attn_norm_2.bias
F32
F32
[4096]
blk.40.attn_norm_2.weight
F32
F32
[4096]
blk.40.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.40.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.40.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.40.time_mix_a0.weight
F32
F32
[4096]
blk.40.time_mix_a1.weight
F16
F16
[4096, 192]
blk.40.time_mix_a2.weight
F16
F16
[192, 4096]
blk.40.time_mix_g1.weight
F16
F16
[4096, 384]
blk.40.time_mix_g2.weight
F16
F16
[384, 4096]
blk.40.time_mix_k_a.weight
F32
F32
[4096]
blk.40.time_mix_k_k.weight
F32
F32
[4096]
blk.40.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.40.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.40.time_mix_ln.bias
F32
F32
[4096]
blk.40.time_mix_ln.weight
F32
F32
[4096]
blk.40.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.40.time_mix_r_k.weight
F32
F32
[4096]
blk.40.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.40.time_mix_v0.weight
F32
F32
[4096]
blk.40.time_mix_v1.weight
F16
F16
[4096, 128]
blk.40.time_mix_v2.weight
F16
F16
[128, 4096]
blk.40.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.40.time_mix_w0.weight
F32
F32
[4096]
blk.40.time_mix_w1.weight
F32
F32
[4096, 192]
blk.40.time_mix_w2.weight
F32
F32
[192, 4096]
blk.41
blk.41.attn_norm.bias
F32
F32
[4096]
blk.41.attn_norm.weight
F32
F32
[4096]
blk.41.attn_norm_2.bias
F32
F32
[4096]
blk.41.attn_norm_2.weight
F32
F32
[4096]
blk.41.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.41.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.41.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.41.time_mix_a0.weight
F32
F32
[4096]
blk.41.time_mix_a1.weight
F16
F16
[4096, 192]
blk.41.time_mix_a2.weight
F16
F16
[192, 4096]
blk.41.time_mix_g1.weight
F16
F16
[4096, 384]
blk.41.time_mix_g2.weight
F16
F16
[384, 4096]
blk.41.time_mix_k_a.weight
F32
F32
[4096]
blk.41.time_mix_k_k.weight
F32
F32
[4096]
blk.41.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.41.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.41.time_mix_ln.bias
F32
F32
[4096]
blk.41.time_mix_ln.weight
F32
F32
[4096]
blk.41.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.41.time_mix_r_k.weight
F32
F32
[4096]
blk.41.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.41.time_mix_v0.weight
F32
F32
[4096]
blk.41.time_mix_v1.weight
F16
F16
[4096, 128]
blk.41.time_mix_v2.weight
F16
F16
[128, 4096]
blk.41.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.41.time_mix_w0.weight
F32
F32
[4096]
blk.41.time_mix_w1.weight
F32
F32
[4096, 192]
blk.41.time_mix_w2.weight
F32
F32
[192, 4096]
blk.42
blk.42.attn_norm.bias
F32
F32
[4096]
blk.42.attn_norm.weight
F32
F32
[4096]
blk.42.attn_norm_2.bias
F32
F32
[4096]
blk.42.attn_norm_2.weight
F32
F32
[4096]
blk.42.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.42.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.42.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.42.time_mix_a0.weight
F32
F32
[4096]
blk.42.time_mix_a1.weight
F16
F16
[4096, 192]
blk.42.time_mix_a2.weight
F16
F16
[192, 4096]
blk.42.time_mix_g1.weight
F16
F16
[4096, 384]
blk.42.time_mix_g2.weight
F16
F16
[384, 4096]
blk.42.time_mix_k_a.weight
F32
F32
[4096]
blk.42.time_mix_k_k.weight
F32
F32
[4096]
blk.42.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.42.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.42.time_mix_ln.bias
F32
F32
[4096]
blk.42.time_mix_ln.weight
F32
F32
[4096]
blk.42.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.42.time_mix_r_k.weight
F32
F32
[4096]
blk.42.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.42.time_mix_v0.weight
F32
F32
[4096]
blk.42.time_mix_v1.weight
F16
F16
[4096, 128]
blk.42.time_mix_v2.weight
F16
F16
[128, 4096]
blk.42.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.42.time_mix_w0.weight
F32
F32
[4096]
blk.42.time_mix_w1.weight
F32
F32
[4096, 192]
blk.42.time_mix_w2.weight
F32
F32
[192, 4096]
blk.43
blk.43.attn_norm.bias
F32
F32
[4096]
blk.43.attn_norm.weight
F32
F32
[4096]
blk.43.attn_norm_2.bias
F32
F32
[4096]
blk.43.attn_norm_2.weight
F32
F32
[4096]
blk.43.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.43.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.43.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.43.time_mix_a0.weight
F32
F32
[4096]
blk.43.time_mix_a1.weight
F16
F16
[4096, 192]
blk.43.time_mix_a2.weight
F16
F16
[192, 4096]
blk.43.time_mix_g1.weight
F16
F16
[4096, 384]
blk.43.time_mix_g2.weight
F16
F16
[384, 4096]
blk.43.time_mix_k_a.weight
F32
F32
[4096]
blk.43.time_mix_k_k.weight
F32
F32
[4096]
blk.43.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.43.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.43.time_mix_ln.bias
F32
F32
[4096]
blk.43.time_mix_ln.weight
F32
F32
[4096]
blk.43.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.43.time_mix_r_k.weight
F32
F32
[4096]
blk.43.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.43.time_mix_v0.weight
F32
F32
[4096]
blk.43.time_mix_v1.weight
F16
F16
[4096, 128]
blk.43.time_mix_v2.weight
F16
F16
[128, 4096]
blk.43.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.43.time_mix_w0.weight
F32
F32
[4096]
blk.43.time_mix_w1.weight
F32
F32
[4096, 192]
blk.43.time_mix_w2.weight
F32
F32
[192, 4096]
blk.44
blk.44.attn_norm.bias
F32
F32
[4096]
blk.44.attn_norm.weight
F32
F32
[4096]
blk.44.attn_norm_2.bias
F32
F32
[4096]
blk.44.attn_norm_2.weight
F32
F32
[4096]
blk.44.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.44.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.44.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.44.time_mix_a0.weight
F32
F32
[4096]
blk.44.time_mix_a1.weight
F16
F16
[4096, 192]
blk.44.time_mix_a2.weight
F16
F16
[192, 4096]
blk.44.time_mix_g1.weight
F16
F16
[4096, 384]
blk.44.time_mix_g2.weight
F16
F16
[384, 4096]
blk.44.time_mix_k_a.weight
F32
F32
[4096]
blk.44.time_mix_k_k.weight
F32
F32
[4096]
blk.44.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.44.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.44.time_mix_ln.bias
F32
F32
[4096]
blk.44.time_mix_ln.weight
F32
F32
[4096]
blk.44.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.44.time_mix_r_k.weight
F32
F32
[4096]
blk.44.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.44.time_mix_v0.weight
F32
F32
[4096]
blk.44.time_mix_v1.weight
F16
F16
[4096, 128]
blk.44.time_mix_v2.weight
F16
F16
[128, 4096]
blk.44.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.44.time_mix_w0.weight
F32
F32
[4096]
blk.44.time_mix_w1.weight
F32
F32
[4096, 192]
blk.44.time_mix_w2.weight
F32
F32
[192, 4096]
blk.45
blk.45.attn_norm.bias
F32
F32
[4096]
blk.45.attn_norm.weight
F32
F32
[4096]
blk.45.attn_norm_2.bias
F32
F32
[4096]
blk.45.attn_norm_2.weight
F32
F32
[4096]
blk.45.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.45.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.45.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.45.time_mix_a0.weight
F32
F32
[4096]
blk.45.time_mix_a1.weight
F16
F16
[4096, 192]
blk.45.time_mix_a2.weight
F16
F16
[192, 4096]
blk.45.time_mix_g1.weight
F16
F16
[4096, 384]
blk.45.time_mix_g2.weight
F16
F16
[384, 4096]
blk.45.time_mix_k_a.weight
F32
F32
[4096]
blk.45.time_mix_k_k.weight
F32
F32
[4096]
blk.45.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.45.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.45.time_mix_ln.bias
F32
F32
[4096]
blk.45.time_mix_ln.weight
F32
F32
[4096]
blk.45.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.45.time_mix_r_k.weight
F32
F32
[4096]
blk.45.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.45.time_mix_v0.weight
F32
F32
[4096]
blk.45.time_mix_v1.weight
F16
F16
[4096, 128]
blk.45.time_mix_v2.weight
F16
F16
[128, 4096]
blk.45.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.45.time_mix_w0.weight
F32
F32
[4096]
blk.45.time_mix_w1.weight
F32
F32
[4096, 192]
blk.45.time_mix_w2.weight
F32
F32
[192, 4096]
blk.46
blk.46.attn_norm.bias
F32
F32
[4096]
blk.46.attn_norm.weight
F32
F32
[4096]
blk.46.attn_norm_2.bias
F32
F32
[4096]
blk.46.attn_norm_2.weight
F32
F32
[4096]
blk.46.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.46.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.46.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.46.time_mix_a0.weight
F32
F32
[4096]
blk.46.time_mix_a1.weight
F16
F16
[4096, 192]
blk.46.time_mix_a2.weight
F16
F16
[192, 4096]
blk.46.time_mix_g1.weight
F16
F16
[4096, 384]
blk.46.time_mix_g2.weight
F16
F16
[384, 4096]
blk.46.time_mix_k_a.weight
F32
F32
[4096]
blk.46.time_mix_k_k.weight
F32
F32
[4096]
blk.46.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.46.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.46.time_mix_ln.bias
F32
F32
[4096]
blk.46.time_mix_ln.weight
F32
F32
[4096]
blk.46.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.46.time_mix_r_k.weight
F32
F32
[4096]
blk.46.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.46.time_mix_v0.weight
F32
F32
[4096]
blk.46.time_mix_v1.weight
F16
F16
[4096, 128]
blk.46.time_mix_v2.weight
F16
F16
[128, 4096]
blk.46.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.46.time_mix_w0.weight
F32
F32
[4096]
blk.46.time_mix_w1.weight
F32
F32
[4096, 192]
blk.46.time_mix_w2.weight
F32
F32
[192, 4096]
blk.47
blk.47.attn_norm.bias
F32
F32
[4096]
blk.47.attn_norm.weight
F32
F32
[4096]
blk.47.attn_norm_2.bias
F32
F32
[4096]
blk.47.attn_norm_2.weight
F32
F32
[4096]
blk.47.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.47.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.47.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.47.time_mix_a0.weight
F32
F32
[4096]
blk.47.time_mix_a1.weight
F16
F16
[4096, 192]
blk.47.time_mix_a2.weight
F16
F16
[192, 4096]
blk.47.time_mix_g1.weight
F16
F16
[4096, 384]
blk.47.time_mix_g2.weight
F16
F16
[384, 4096]
blk.47.time_mix_k_a.weight
F32
F32
[4096]
blk.47.time_mix_k_k.weight
F32
F32
[4096]
blk.47.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.47.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.47.time_mix_ln.bias
F32
F32
[4096]
blk.47.time_mix_ln.weight
F32
F32
[4096]
blk.47.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.47.time_mix_r_k.weight
F32
F32
[4096]
blk.47.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.47.time_mix_v0.weight
F32
F32
[4096]
blk.47.time_mix_v1.weight
F16
F16
[4096, 128]
blk.47.time_mix_v2.weight
F16
F16
[128, 4096]
blk.47.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.47.time_mix_w0.weight
F32
F32
[4096]
blk.47.time_mix_w1.weight
F32
F32
[4096, 192]
blk.47.time_mix_w2.weight
F32
F32
[192, 4096]
blk.48
blk.48.attn_norm.bias
F32
F32
[4096]
blk.48.attn_norm.weight
F32
F32
[4096]
blk.48.attn_norm_2.bias
F32
F32
[4096]
blk.48.attn_norm_2.weight
F32
F32
[4096]
blk.48.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.48.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.48.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.48.time_mix_a0.weight
F32
F32
[4096]
blk.48.time_mix_a1.weight
F16
F16
[4096, 192]
blk.48.time_mix_a2.weight
F16
F16
[192, 4096]
blk.48.time_mix_g1.weight
F16
F16
[4096, 384]
blk.48.time_mix_g2.weight
F16
F16
[384, 4096]
blk.48.time_mix_k_a.weight
F32
F32
[4096]
blk.48.time_mix_k_k.weight
F32
F32
[4096]
blk.48.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.48.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.48.time_mix_ln.bias
F32
F32
[4096]
blk.48.time_mix_ln.weight
F32
F32
[4096]
blk.48.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.48.time_mix_r_k.weight
F32
F32
[4096]
blk.48.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.48.time_mix_v0.weight
F32
F32
[4096]
blk.48.time_mix_v1.weight
F16
F16
[4096, 128]
blk.48.time_mix_v2.weight
F16
F16
[128, 4096]
blk.48.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.48.time_mix_w0.weight
F32
F32
[4096]
blk.48.time_mix_w1.weight
F32
F32
[4096, 192]
blk.48.time_mix_w2.weight
F32
F32
[192, 4096]
blk.49
blk.49.attn_norm.bias
F32
F32
[4096]
blk.49.attn_norm.weight
F32
F32
[4096]
blk.49.attn_norm_2.bias
F32
F32
[4096]
blk.49.attn_norm_2.weight
F32
F32
[4096]
blk.49.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.49.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.49.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.49.time_mix_a0.weight
F32
F32
[4096]
blk.49.time_mix_a1.weight
F16
F16
[4096, 192]
blk.49.time_mix_a2.weight
F16
F16
[192, 4096]
blk.49.time_mix_g1.weight
F16
F16
[4096, 384]
blk.49.time_mix_g2.weight
F16
F16
[384, 4096]
blk.49.time_mix_k_a.weight
F32
F32
[4096]
blk.49.time_mix_k_k.weight
F32
F32
[4096]
blk.49.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.49.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.49.time_mix_ln.bias
F32
F32
[4096]
blk.49.time_mix_ln.weight
F32
F32
[4096]
blk.49.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.49.time_mix_r_k.weight
F32
F32
[4096]
blk.49.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.49.time_mix_v0.weight
F32
F32
[4096]
blk.49.time_mix_v1.weight
F16
F16
[4096, 128]
blk.49.time_mix_v2.weight
F16
F16
[128, 4096]
blk.49.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.49.time_mix_w0.weight
F32
F32
[4096]
blk.49.time_mix_w1.weight
F32
F32
[4096, 192]
blk.49.time_mix_w2.weight
F32
F32
[192, 4096]
blk.50
blk.50.attn_norm.bias
F32
F32
[4096]
blk.50.attn_norm.weight
F32
F32
[4096]
blk.50.attn_norm_2.bias
F32
F32
[4096]
blk.50.attn_norm_2.weight
F32
F32
[4096]
blk.50.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.50.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.50.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.50.time_mix_a0.weight
F32
F32
[4096]
blk.50.time_mix_a1.weight
F16
F16
[4096, 192]
blk.50.time_mix_a2.weight
F16
F16
[192, 4096]
blk.50.time_mix_g1.weight
F16
F16
[4096, 384]
blk.50.time_mix_g2.weight
F16
F16
[384, 4096]
blk.50.time_mix_k_a.weight
F32
F32
[4096]
blk.50.time_mix_k_k.weight
F32
F32
[4096]
blk.50.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.50.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.50.time_mix_ln.bias
F32
F32
[4096]
blk.50.time_mix_ln.weight
F32
F32
[4096]
blk.50.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.50.time_mix_r_k.weight
F32
F32
[4096]
blk.50.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.50.time_mix_v0.weight
F32
F32
[4096]
blk.50.time_mix_v1.weight
F16
F16
[4096, 128]
blk.50.time_mix_v2.weight
F16
F16
[128, 4096]
blk.50.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.50.time_mix_w0.weight
F32
F32
[4096]
blk.50.time_mix_w1.weight
F32
F32
[4096, 192]
blk.50.time_mix_w2.weight
F32
F32
[192, 4096]
blk.51
blk.51.attn_norm.bias
F32
F32
[4096]
blk.51.attn_norm.weight
F32
F32
[4096]
blk.51.attn_norm_2.bias
F32
F32
[4096]
blk.51.attn_norm_2.weight
F32
F32
[4096]
blk.51.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.51.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.51.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.51.time_mix_a0.weight
F32
F32
[4096]
blk.51.time_mix_a1.weight
F16
F16
[4096, 192]
blk.51.time_mix_a2.weight
F16
F16
[192, 4096]
blk.51.time_mix_g1.weight
F16
F16
[4096, 384]
blk.51.time_mix_g2.weight
F16
F16
[384, 4096]
blk.51.time_mix_k_a.weight
F32
F32
[4096]
blk.51.time_mix_k_k.weight
F32
F32
[4096]
blk.51.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.51.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.51.time_mix_ln.bias
F32
F32
[4096]
blk.51.time_mix_ln.weight
F32
F32
[4096]
blk.51.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.51.time_mix_r_k.weight
F32
F32
[4096]
blk.51.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.51.time_mix_v0.weight
F32
F32
[4096]
blk.51.time_mix_v1.weight
F16
F16
[4096, 128]
blk.51.time_mix_v2.weight
F16
F16
[128, 4096]
blk.51.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.51.time_mix_w0.weight
F32
F32
[4096]
blk.51.time_mix_w1.weight
F32
F32
[4096, 192]
blk.51.time_mix_w2.weight
F32
F32
[192, 4096]
blk.52
blk.52.attn_norm.bias
F32
F32
[4096]
blk.52.attn_norm.weight
F32
F32
[4096]
blk.52.attn_norm_2.bias
F32
F32
[4096]
blk.52.attn_norm_2.weight
F32
F32
[4096]
blk.52.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.52.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.52.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.52.time_mix_a0.weight
F32
F32
[4096]
blk.52.time_mix_a1.weight
F16
F16
[4096, 192]
blk.52.time_mix_a2.weight
F16
F16
[192, 4096]
blk.52.time_mix_g1.weight
F16
F16
[4096, 384]
blk.52.time_mix_g2.weight
F16
F16
[384, 4096]
blk.52.time_mix_k_a.weight
F32
F32
[4096]
blk.52.time_mix_k_k.weight
F32
F32
[4096]
blk.52.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.52.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.52.time_mix_ln.bias
F32
F32
[4096]
blk.52.time_mix_ln.weight
F32
F32
[4096]
blk.52.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.52.time_mix_r_k.weight
F32
F32
[4096]
blk.52.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.52.time_mix_v0.weight
F32
F32
[4096]
blk.52.time_mix_v1.weight
F16
F16
[4096, 128]
blk.52.time_mix_v2.weight
F16
F16
[128, 4096]
blk.52.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.52.time_mix_w0.weight
F32
F32
[4096]
blk.52.time_mix_w1.weight
F32
F32
[4096, 192]
blk.52.time_mix_w2.weight
F32
F32
[192, 4096]
blk.53
blk.53.attn_norm.bias
F32
F32
[4096]
blk.53.attn_norm.weight
F32
F32
[4096]
blk.53.attn_norm_2.bias
F32
F32
[4096]
blk.53.attn_norm_2.weight
F32
F32
[4096]
blk.53.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.53.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.53.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.53.time_mix_a0.weight
F32
F32
[4096]
blk.53.time_mix_a1.weight
F16
F16
[4096, 192]
blk.53.time_mix_a2.weight
F16
F16
[192, 4096]
blk.53.time_mix_g1.weight
F16
F16
[4096, 384]
blk.53.time_mix_g2.weight
F16
F16
[384, 4096]
blk.53.time_mix_k_a.weight
F32
F32
[4096]
blk.53.time_mix_k_k.weight
F32
F32
[4096]
blk.53.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.53.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.53.time_mix_ln.bias
F32
F32
[4096]
blk.53.time_mix_ln.weight
F32
F32
[4096]
blk.53.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.53.time_mix_r_k.weight
F32
F32
[4096]
blk.53.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.53.time_mix_v0.weight
F32
F32
[4096]
blk.53.time_mix_v1.weight
F16
F16
[4096, 128]
blk.53.time_mix_v2.weight
F16
F16
[128, 4096]
blk.53.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.53.time_mix_w0.weight
F32
F32
[4096]
blk.53.time_mix_w1.weight
F32
F32
[4096, 192]
blk.53.time_mix_w2.weight
F32
F32
[192, 4096]
blk.54
blk.54.attn_norm.bias
F32
F32
[4096]
blk.54.attn_norm.weight
F32
F32
[4096]
blk.54.attn_norm_2.bias
F32
F32
[4096]
blk.54.attn_norm_2.weight
F32
F32
[4096]
blk.54.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.54.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.54.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.54.time_mix_a0.weight
F32
F32
[4096]
blk.54.time_mix_a1.weight
F16
F16
[4096, 192]
blk.54.time_mix_a2.weight
F16
F16
[192, 4096]
blk.54.time_mix_g1.weight
F16
F16
[4096, 384]
blk.54.time_mix_g2.weight
F16
F16
[384, 4096]
blk.54.time_mix_k_a.weight
F32
F32
[4096]
blk.54.time_mix_k_k.weight
F32
F32
[4096]
blk.54.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.54.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.54.time_mix_ln.bias
F32
F32
[4096]
blk.54.time_mix_ln.weight
F32
F32
[4096]
blk.54.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.54.time_mix_r_k.weight
F32
F32
[4096]
blk.54.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.54.time_mix_v0.weight
F32
F32
[4096]
blk.54.time_mix_v1.weight
F16
F16
[4096, 128]
blk.54.time_mix_v2.weight
F16
F16
[128, 4096]
blk.54.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.54.time_mix_w0.weight
F32
F32
[4096]
blk.54.time_mix_w1.weight
F32
F32
[4096, 192]
blk.54.time_mix_w2.weight
F32
F32
[192, 4096]
blk.55
blk.55.attn_norm.bias
F32
F32
[4096]
blk.55.attn_norm.weight
F32
F32
[4096]
blk.55.attn_norm_2.bias
F32
F32
[4096]
blk.55.attn_norm_2.weight
F32
F32
[4096]
blk.55.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.55.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.55.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.55.time_mix_a0.weight
F32
F32
[4096]
blk.55.time_mix_a1.weight
F16
F16
[4096, 192]
blk.55.time_mix_a2.weight
F16
F16
[192, 4096]
blk.55.time_mix_g1.weight
F16
F16
[4096, 384]
blk.55.time_mix_g2.weight
F16
F16
[384, 4096]
blk.55.time_mix_k_a.weight
F32
F32
[4096]
blk.55.time_mix_k_k.weight
F32
F32
[4096]
blk.55.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.55.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.55.time_mix_ln.bias
F32
F32
[4096]
blk.55.time_mix_ln.weight
F32
F32
[4096]
blk.55.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.55.time_mix_r_k.weight
F32
F32
[4096]
blk.55.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.55.time_mix_v0.weight
F32
F32
[4096]
blk.55.time_mix_v1.weight
F16
F16
[4096, 128]
blk.55.time_mix_v2.weight
F16
F16
[128, 4096]
blk.55.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.55.time_mix_w0.weight
F32
F32
[4096]
blk.55.time_mix_w1.weight
F32
F32
[4096, 192]
blk.55.time_mix_w2.weight
F32
F32
[192, 4096]
blk.56
blk.56.attn_norm.bias
F32
F32
[4096]
blk.56.attn_norm.weight
F32
F32
[4096]
blk.56.attn_norm_2.bias
F32
F32
[4096]
blk.56.attn_norm_2.weight
F32
F32
[4096]
blk.56.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.56.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.56.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.56.time_mix_a0.weight
F32
F32
[4096]
blk.56.time_mix_a1.weight
F16
F16
[4096, 192]
blk.56.time_mix_a2.weight
F16
F16
[192, 4096]
blk.56.time_mix_g1.weight
F16
F16
[4096, 384]
blk.56.time_mix_g2.weight
F16
F16
[384, 4096]
blk.56.time_mix_k_a.weight
F32
F32
[4096]
blk.56.time_mix_k_k.weight
F32
F32
[4096]
blk.56.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.56.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.56.time_mix_ln.bias
F32
F32
[4096]
blk.56.time_mix_ln.weight
F32
F32
[4096]
blk.56.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.56.time_mix_r_k.weight
F32
F32
[4096]
blk.56.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.56.time_mix_v0.weight
F32
F32
[4096]
blk.56.time_mix_v1.weight
F16
F16
[4096, 128]
blk.56.time_mix_v2.weight
F16
F16
[128, 4096]
blk.56.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.56.time_mix_w0.weight
F32
F32
[4096]
blk.56.time_mix_w1.weight
F32
F32
[4096, 192]
blk.56.time_mix_w2.weight
F32
F32
[192, 4096]
blk.57
blk.57.attn_norm.bias
F32
F32
[4096]
blk.57.attn_norm.weight
F32
F32
[4096]
blk.57.attn_norm_2.bias
F32
F32
[4096]
blk.57.attn_norm_2.weight
F32
F32
[4096]
blk.57.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.57.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.57.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.57.time_mix_a0.weight
F32
F32
[4096]
blk.57.time_mix_a1.weight
F16
F16
[4096, 192]
blk.57.time_mix_a2.weight
F16
F16
[192, 4096]
blk.57.time_mix_g1.weight
F16
F16
[4096, 384]
blk.57.time_mix_g2.weight
F16
F16
[384, 4096]
blk.57.time_mix_k_a.weight
F32
F32
[4096]
blk.57.time_mix_k_k.weight
F32
F32
[4096]
blk.57.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.57.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.57.time_mix_ln.bias
F32
F32
[4096]
blk.57.time_mix_ln.weight
F32
F32
[4096]
blk.57.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.57.time_mix_r_k.weight
F32
F32
[4096]
blk.57.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.57.time_mix_v0.weight
F32
F32
[4096]
blk.57.time_mix_v1.weight
F16
F16
[4096, 128]
blk.57.time_mix_v2.weight
F16
F16
[128, 4096]
blk.57.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.57.time_mix_w0.weight
F32
F32
[4096]
blk.57.time_mix_w1.weight
F32
F32
[4096, 192]
blk.57.time_mix_w2.weight
F32
F32
[192, 4096]
blk.58
blk.58.attn_norm.bias
F32
F32
[4096]
blk.58.attn_norm.weight
F32
F32
[4096]
blk.58.attn_norm_2.bias
F32
F32
[4096]
blk.58.attn_norm_2.weight
F32
F32
[4096]
blk.58.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.58.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.58.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.58.time_mix_a0.weight
F32
F32
[4096]
blk.58.time_mix_a1.weight
F16
F16
[4096, 192]
blk.58.time_mix_a2.weight
F16
F16
[192, 4096]
blk.58.time_mix_g1.weight
F16
F16
[4096, 384]
blk.58.time_mix_g2.weight
F16
F16
[384, 4096]
blk.58.time_mix_k_a.weight
F32
F32
[4096]
blk.58.time_mix_k_k.weight
F32
F32
[4096]
blk.58.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.58.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.58.time_mix_ln.bias
F32
F32
[4096]
blk.58.time_mix_ln.weight
F32
F32
[4096]
blk.58.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.58.time_mix_r_k.weight
F32
F32
[4096]
blk.58.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.58.time_mix_v0.weight
F32
F32
[4096]
blk.58.time_mix_v1.weight
F16
F16
[4096, 128]
blk.58.time_mix_v2.weight
F16
F16
[128, 4096]
blk.58.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.58.time_mix_w0.weight
F32
F32
[4096]
blk.58.time_mix_w1.weight
F32
F32
[4096, 192]
blk.58.time_mix_w2.weight
F32
F32
[192, 4096]
blk.59
blk.59.attn_norm.bias
F32
F32
[4096]
blk.59.attn_norm.weight
F32
F32
[4096]
blk.59.attn_norm_2.bias
F32
F32
[4096]
blk.59.attn_norm_2.weight
F32
F32
[4096]
blk.59.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.59.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.59.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.59.time_mix_a0.weight
F32
F32
[4096]
blk.59.time_mix_a1.weight
F16
F16
[4096, 192]
blk.59.time_mix_a2.weight
F16
F16
[192, 4096]
blk.59.time_mix_g1.weight
F16
F16
[4096, 384]
blk.59.time_mix_g2.weight
F16
F16
[384, 4096]
blk.59.time_mix_k_a.weight
F32
F32
[4096]
blk.59.time_mix_k_k.weight
F32
F32
[4096]
blk.59.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.59.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.59.time_mix_ln.bias
F32
F32
[4096]
blk.59.time_mix_ln.weight
F32
F32
[4096]
blk.59.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.59.time_mix_r_k.weight
F32
F32
[4096]
blk.59.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.59.time_mix_v0.weight
F32
F32
[4096]
blk.59.time_mix_v1.weight
F16
F16
[4096, 128]
blk.59.time_mix_v2.weight
F16
F16
[128, 4096]
blk.59.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.59.time_mix_w0.weight
F32
F32
[4096]
blk.59.time_mix_w1.weight
F32
F32
[4096, 192]
blk.59.time_mix_w2.weight
F32
F32
[192, 4096]
blk.60
blk.60.attn_norm.bias
F32
F32
[4096]
blk.60.attn_norm.weight
F32
F32
[4096]
blk.60.attn_norm_2.bias
F32
F32
[4096]
blk.60.attn_norm_2.weight
F32
F32
[4096]
blk.60.channel_mix_key.weight
Q4_K
Q4_K
[4096, 16384]
blk.60.channel_mix_lerp_k.weight
F32
F32
[4096]
blk.60.channel_mix_value.weight
Q4_K
Q4_K
[16384, 4096]
blk.60.time_mix_a0.weight
F32
F32
[4096]
blk.60.time_mix_a1.weight
F16
F16
[4096, 192]
blk.60.time_mix_a2.weight
F16
F16
[192, 4096]
blk.60.time_mix_g1.weight
F16
F16
[4096, 384]
blk.60.time_mix_g2.weight
F16
F16
[384, 4096]
blk.60.time_mix_k_a.weight
F32
F32
[4096]
blk.60.time_mix_k_k.weight
F32
F32
[4096]
blk.60.time_mix_key.weight
Q4_K
Q4_K
[4096, 4096]
blk.60.time_mix_lerp_fused.weight
F32
F32
[4096, 1, 1, 6]
blk.60.time_mix_ln.bias
F32
F32
[4096]
blk.60.time_mix_ln.weight
F32
F32
[4096]
blk.60.time_mix_output.weight
Q4_K
Q4_K
[4096, 4096]
blk.60.time_mix_r_k.weight
F32
F32
[4096]
blk.60.time_mix_receptance.weight
Q4_K
Q4_K
[4096, 4096]
blk.60.time_mix_v0.weight
F32
F32
[4096]
blk.60.time_mix_v1.weight
F16
F16
[4096, 128]
blk.60.time_mix_v2.weight
F16
F16
[128, 4096]
blk.60.time_mix_value.weight
Q4_K
Q4_K
[4096, 4096]
blk.60.time_mix_w0.weight
F32
F32
[4096]
blk.60.time_mix_w1.weight
F32
F32
[4096, 192]
blk.60.time_mix_w2.weight
F32
F32
[192, 4096]
output.weight
Q6_K
Q6_K
[4096, 65536]
output_norm.bias
F32
F32
[4096]
token_embd_norm.bias
F32
F32
[4096]
token_embd_norm.weight
F32
F32
[4096]
output_norm.weight
F32
F32
[4096]