GitHub
186.8K
Download
Models
Docs
Pricing
Sign in
Download
Models
Docs
Pricing
Sign in
orcarouter
/
Ripple-65B-abliterated
:bf16
34
Downloads
Updated
yesterday
Cancel
Ripple-65B-abliterated:bf16
...
/
model
bf6878938010 · 45GB
Metadata
general.architecture
ricursive
ricursive
general.file_type
BF16
BF16
ricursive.attention.head_count
80
80
ricursive.attention.head_count_kv
8
8
ricursive.attention.key_length
128
128
ricursive.attention.layer_norm_rms_epsilon
1e-06
1e-06
ricursive.attention.sliding_window
1024
1024
ricursive.attention.sliding_window_pattern
[true, true, true, true, false, ...]
[true, true, true, true, false, ...]
ricursive.attention.value_length
128
128
ricursive.block_count
80
80
ricursive.context_length
2097152
2097152
ricursive.embedding_length
5120
5120
ricursive.embedding_scale
71.55418
71.55418
ricursive.feed_forward_length
40960
40960
ricursive.rope.dimension_count
128
128
ricursive.rope.freq_base
10000
10000
ricursive.swiglu_clamp
20
20
ricursive.vocab_size
100352
100352
tokenizer.ggml.bos_token_id
100257
100257
tokenizer.ggml.clean_spaces
false
false
tokenizer.ggml.eos_token_id
100265
100265
tokenizer.ggml.eot_token_id
100265
100265
tokenizer.ggml.merges
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
100277
100277
tokenizer.ggml.pre
dbrx
dbrx
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
tokenizer.ggml.unknown_token_id
100257
100257
split.count
3
3
split.no
0
0
split.tensors.count
1203
1203
Tensor
Name
Type
Shape
token_embd.weight
BF16
BF16
[5120, 100352]
blk.0
blk.0.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.0.attn_k.weight
BF16
BF16
[5120, 1024]
blk.0.attn_k_norm.weight
F32
F32
[128]
blk.0.attn_norm.weight
F32
F32
[5120]
blk.0.attn_output.weight
BF16
BF16
[10240, 5120]
blk.0.attn_q.weight
BF16
BF16
[5120, 10240]
blk.0.attn_q_norm.weight
F32
F32
[128]
blk.0.attn_v.weight
BF16
BF16
[5120, 1024]
blk.0.attn_vres_lambda.weight
F32
F32
[1]
blk.0.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.0.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.0.ffn_norm.weight
F32
F32
[5120]
blk.0.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.0.post_attention_norm.weight
F32
F32
[5120]
blk.0.post_ffw_norm.weight
F32
F32
[5120]
blk.1
blk.1.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.1.attn_k.weight
BF16
BF16
[5120, 1024]
blk.1.attn_k_norm.weight
F32
F32
[128]
blk.1.attn_norm.weight
F32
F32
[5120]
blk.1.attn_output.weight
BF16
BF16
[10240, 5120]
blk.1.attn_q.weight
BF16
BF16
[5120, 10240]
blk.1.attn_q_norm.weight
F32
F32
[128]
blk.1.attn_v.weight
BF16
BF16
[5120, 1024]
blk.1.attn_vres_lambda.weight
F32
F32
[1]
blk.1.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.1.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.1.ffn_norm.weight
F32
F32
[5120]
blk.1.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.1.post_attention_norm.weight
F32
F32
[5120]
blk.1.post_ffw_norm.weight
F32
F32
[5120]
blk.2
blk.2.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.2.attn_k.weight
BF16
BF16
[5120, 1024]
blk.2.attn_k_norm.weight
F32
F32
[128]
blk.2.attn_norm.weight
F32
F32
[5120]
blk.2.attn_output.weight
BF16
BF16
[10240, 5120]
blk.2.attn_q.weight
BF16
BF16
[5120, 10240]
blk.2.attn_q_norm.weight
F32
F32
[128]
blk.2.attn_v.weight
BF16
BF16
[5120, 1024]
blk.2.attn_vres_lambda.weight
F32
F32
[1]
blk.2.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.2.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.2.ffn_norm.weight
F32
F32
[5120]
blk.2.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.2.post_attention_norm.weight
F32
F32
[5120]
blk.2.post_ffw_norm.weight
F32
F32
[5120]
blk.3
blk.3.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.3.attn_k.weight
BF16
BF16
[5120, 1024]
blk.3.attn_k_norm.weight
F32
F32
[128]
blk.3.attn_norm.weight
F32
F32
[5120]
blk.3.attn_output.weight
BF16
BF16
[10240, 5120]
blk.3.attn_q.weight
BF16
BF16
[5120, 10240]
blk.3.attn_q_norm.weight
F32
F32
[128]
blk.3.attn_v.weight
BF16
BF16
[5120, 1024]
blk.3.attn_vres_lambda.weight
F32
F32
[1]
blk.3.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.3.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.3.ffn_norm.weight
F32
F32
[5120]
blk.3.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.3.post_attention_norm.weight
F32
F32
[5120]
blk.3.post_ffw_norm.weight
F32
F32
[5120]
blk.10
blk.10.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.10.attn_k.weight
BF16
BF16
[5120, 1024]
blk.10.attn_k_norm.weight
F32
F32
[128]
blk.10.attn_norm.weight
F32
F32
[5120]
blk.10.attn_output.weight
BF16
BF16
[10240, 5120]
blk.10.attn_q.weight
BF16
BF16
[5120, 10240]
blk.10.attn_q_norm.weight
F32
F32
[128]
blk.10.attn_v.weight
BF16
BF16
[5120, 1024]
blk.10.attn_vres_lambda.weight
F32
F32
[1]
blk.10.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.10.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.10.ffn_norm.weight
F32
F32
[5120]
blk.10.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.10.post_attention_norm.weight
F32
F32
[5120]
blk.10.post_ffw_norm.weight
F32
F32
[5120]
blk.11
blk.11.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.11.attn_k.weight
BF16
BF16
[5120, 1024]
blk.11.attn_k_norm.weight
F32
F32
[128]
blk.11.attn_norm.weight
F32
F32
[5120]
blk.11.attn_output.weight
BF16
BF16
[10240, 5120]
blk.11.attn_q.weight
BF16
BF16
[5120, 10240]
blk.11.attn_q_norm.weight
F32
F32
[128]
blk.11.attn_v.weight
BF16
BF16
[5120, 1024]
blk.11.attn_vres_lambda.weight
F32
F32
[1]
blk.11.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.11.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.11.ffn_norm.weight
F32
F32
[5120]
blk.11.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.11.post_attention_norm.weight
F32
F32
[5120]
blk.11.post_ffw_norm.weight
F32
F32
[5120]
blk.12
blk.12.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.12.attn_k.weight
BF16
BF16
[5120, 1024]
blk.12.attn_k_norm.weight
F32
F32
[128]
blk.12.attn_norm.weight
F32
F32
[5120]
blk.12.attn_output.weight
BF16
BF16
[10240, 5120]
blk.12.attn_q.weight
BF16
BF16
[5120, 10240]
blk.12.attn_q_norm.weight
F32
F32
[128]
blk.12.attn_v.weight
BF16
BF16
[5120, 1024]
blk.12.attn_vres_lambda.weight
F32
F32
[1]
blk.12.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.12.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.12.ffn_norm.weight
F32
F32
[5120]
blk.12.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.12.post_attention_norm.weight
F32
F32
[5120]
blk.12.post_ffw_norm.weight
F32
F32
[5120]
blk.13
blk.13.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.13.attn_k.weight
BF16
BF16
[5120, 1024]
blk.13.attn_k_norm.weight
F32
F32
[128]
blk.13.attn_norm.weight
F32
F32
[5120]
blk.13.attn_output.weight
BF16
BF16
[10240, 5120]
blk.13.attn_q.weight
BF16
BF16
[5120, 10240]
blk.13.attn_q_norm.weight
F32
F32
[128]
blk.13.attn_v.weight
BF16
BF16
[5120, 1024]
blk.13.attn_vres_lambda.weight
F32
F32
[1]
blk.13.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.13.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.13.ffn_norm.weight
F32
F32
[5120]
blk.13.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.13.post_attention_norm.weight
F32
F32
[5120]
blk.13.post_ffw_norm.weight
F32
F32
[5120]
blk.14
blk.14.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.14.attn_k.weight
BF16
BF16
[5120, 1024]
blk.14.attn_k_norm.weight
F32
F32
[128]
blk.14.attn_norm.weight
F32
F32
[5120]
blk.14.attn_output.weight
BF16
BF16
[10240, 5120]
blk.14.attn_q.weight
BF16
BF16
[5120, 10240]
blk.14.attn_q_norm.weight
F32
F32
[128]
blk.14.attn_v.weight
BF16
BF16
[5120, 1024]
blk.14.attn_vres_lambda.weight
F32
F32
[1]
blk.14.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.14.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.14.ffn_norm.weight
F32
F32
[5120]
blk.14.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.14.post_attention_norm.weight
F32
F32
[5120]
blk.14.post_ffw_norm.weight
F32
F32
[5120]
blk.15
blk.15.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.15.attn_k.weight
BF16
BF16
[5120, 1024]
blk.15.attn_k_norm.weight
F32
F32
[128]
blk.15.attn_norm.weight
F32
F32
[5120]
blk.15.attn_output.weight
BF16
BF16
[10240, 5120]
blk.15.attn_q.weight
BF16
BF16
[5120, 10240]
blk.15.attn_q_norm.weight
F32
F32
[128]
blk.15.attn_v.weight
BF16
BF16
[5120, 1024]
blk.15.attn_vres_lambda.weight
F32
F32
[1]
blk.15.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.15.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.15.ffn_norm.weight
F32
F32
[5120]
blk.15.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.15.post_attention_norm.weight
F32
F32
[5120]
blk.15.post_ffw_norm.weight
F32
F32
[5120]
blk.16
blk.16.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.16.attn_k.weight
BF16
BF16
[5120, 1024]
blk.16.attn_k_norm.weight
F32
F32
[128]
blk.16.attn_norm.weight
F32
F32
[5120]
blk.16.attn_output.weight
BF16
BF16
[10240, 5120]
blk.16.attn_q.weight
BF16
BF16
[5120, 10240]
blk.16.attn_q_norm.weight
F32
F32
[128]
blk.16.attn_v.weight
BF16
BF16
[5120, 1024]
blk.16.attn_vres_lambda.weight
F32
F32
[1]
blk.16.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.16.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.16.ffn_norm.weight
F32
F32
[5120]
blk.16.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.16.post_attention_norm.weight
F32
F32
[5120]
blk.16.post_ffw_norm.weight
F32
F32
[5120]
blk.17
blk.17.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.17.attn_k.weight
BF16
BF16
[5120, 1024]
blk.17.attn_k_norm.weight
F32
F32
[128]
blk.17.attn_norm.weight
F32
F32
[5120]
blk.17.attn_output.weight
BF16
BF16
[10240, 5120]
blk.17.attn_q.weight
BF16
BF16
[5120, 10240]
blk.17.attn_q_norm.weight
F32
F32
[128]
blk.17.attn_v.weight
BF16
BF16
[5120, 1024]
blk.17.attn_vres_lambda.weight
F32
F32
[1]
blk.17.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.17.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.17.ffn_norm.weight
F32
F32
[5120]
blk.17.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.17.post_attention_norm.weight
F32
F32
[5120]
blk.17.post_ffw_norm.weight
F32
F32
[5120]
blk.18
blk.18.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.18.attn_k.weight
BF16
BF16
[5120, 1024]
blk.18.attn_k_norm.weight
F32
F32
[128]
blk.18.attn_norm.weight
F32
F32
[5120]
blk.18.attn_output.weight
BF16
BF16
[10240, 5120]
blk.18.attn_q.weight
BF16
BF16
[5120, 10240]
blk.18.attn_q_norm.weight
F32
F32
[128]
blk.18.attn_v.weight
BF16
BF16
[5120, 1024]
blk.18.attn_vres_lambda.weight
F32
F32
[1]
blk.18.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.18.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.18.ffn_norm.weight
F32
F32
[5120]
blk.18.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.18.post_attention_norm.weight
F32
F32
[5120]
blk.18.post_ffw_norm.weight
F32
F32
[5120]
blk.19
blk.19.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.19.attn_k.weight
BF16
BF16
[5120, 1024]
blk.19.attn_k_norm.weight
F32
F32
[128]
blk.19.attn_norm.weight
F32
F32
[5120]
blk.19.attn_output.weight
BF16
BF16
[10240, 5120]
blk.19.attn_q.weight
BF16
BF16
[5120, 10240]
blk.19.attn_q_norm.weight
F32
F32
[128]
blk.19.attn_v.weight
BF16
BF16
[5120, 1024]
blk.19.attn_vres_lambda.weight
F32
F32
[1]
blk.19.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.19.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.19.ffn_norm.weight
F32
F32
[5120]
blk.19.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.19.post_attention_norm.weight
F32
F32
[5120]
blk.19.post_ffw_norm.weight
F32
F32
[5120]
blk.20
blk.20.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.20.attn_k.weight
BF16
BF16
[5120, 1024]
blk.20.attn_k_norm.weight
F32
F32
[128]
blk.20.attn_norm.weight
F32
F32
[5120]
blk.20.attn_output.weight
BF16
BF16
[10240, 5120]
blk.20.attn_q.weight
BF16
BF16
[5120, 10240]
blk.20.attn_q_norm.weight
F32
F32
[128]
blk.20.attn_v.weight
BF16
BF16
[5120, 1024]
blk.20.attn_vres_lambda.weight
F32
F32
[1]
blk.20.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.20.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.20.ffn_norm.weight
F32
F32
[5120]
blk.20.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.20.post_attention_norm.weight
F32
F32
[5120]
blk.20.post_ffw_norm.weight
F32
F32
[5120]
blk.21
blk.21.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.21.attn_k.weight
BF16
BF16
[5120, 1024]
blk.21.attn_k_norm.weight
F32
F32
[128]
blk.21.attn_norm.weight
F32
F32
[5120]
blk.21.attn_output.weight
BF16
BF16
[10240, 5120]
blk.21.attn_q.weight
BF16
BF16
[5120, 10240]
blk.21.attn_q_norm.weight
F32
F32
[128]
blk.21.attn_v.weight
BF16
BF16
[5120, 1024]
blk.21.attn_vres_lambda.weight
F32
F32
[1]
blk.21.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.21.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.21.ffn_norm.weight
F32
F32
[5120]
blk.21.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.21.post_attention_norm.weight
F32
F32
[5120]
blk.21.post_ffw_norm.weight
F32
F32
[5120]
blk.22
blk.22.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.22.attn_k.weight
BF16
BF16
[5120, 1024]
blk.22.attn_k_norm.weight
F32
F32
[128]
blk.22.attn_norm.weight
F32
F32
[5120]
blk.22.attn_output.weight
BF16
BF16
[10240, 5120]
blk.22.attn_q.weight
BF16
BF16
[5120, 10240]
blk.22.attn_q_norm.weight
F32
F32
[128]
blk.22.attn_v.weight
BF16
BF16
[5120, 1024]
blk.22.attn_vres_lambda.weight
F32
F32
[1]
blk.22.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.22.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.22.ffn_norm.weight
F32
F32
[5120]
blk.22.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.22.post_attention_norm.weight
F32
F32
[5120]
blk.22.post_ffw_norm.weight
F32
F32
[5120]
blk.23
blk.23.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.23.attn_k.weight
BF16
BF16
[5120, 1024]
blk.23.attn_k_norm.weight
F32
F32
[128]
blk.23.attn_norm.weight
F32
F32
[5120]
blk.23.attn_output.weight
BF16
BF16
[10240, 5120]
blk.23.attn_q.weight
BF16
BF16
[5120, 10240]
blk.23.attn_q_norm.weight
F32
F32
[128]
blk.23.attn_v.weight
BF16
BF16
[5120, 1024]
blk.23.attn_vres_lambda.weight
F32
F32
[1]
blk.23.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.23.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.23.ffn_norm.weight
F32
F32
[5120]
blk.23.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.23.post_attention_norm.weight
F32
F32
[5120]
blk.23.post_ffw_norm.weight
F32
F32
[5120]
blk.24
blk.24.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.24.attn_k.weight
BF16
BF16
[5120, 1024]
blk.24.attn_k_norm.weight
F32
F32
[128]
blk.24.attn_norm.weight
F32
F32
[5120]
blk.24.attn_output.weight
BF16
BF16
[10240, 5120]
blk.24.attn_q.weight
BF16
BF16
[5120, 10240]
blk.24.attn_q_norm.weight
F32
F32
[128]
blk.24.attn_v.weight
BF16
BF16
[5120, 1024]
blk.24.attn_vres_lambda.weight
F32
F32
[1]
blk.24.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.24.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.24.ffn_norm.weight
F32
F32
[5120]
blk.24.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.24.post_attention_norm.weight
F32
F32
[5120]
blk.24.post_ffw_norm.weight
F32
F32
[5120]
blk.25
blk.25.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.25.attn_k.weight
BF16
BF16
[5120, 1024]
blk.25.attn_k_norm.weight
F32
F32
[128]
blk.25.attn_norm.weight
F32
F32
[5120]
blk.25.attn_output.weight
BF16
BF16
[10240, 5120]
blk.25.attn_q.weight
BF16
BF16
[5120, 10240]
blk.25.attn_q_norm.weight
F32
F32
[128]
blk.25.attn_v.weight
BF16
BF16
[5120, 1024]
blk.25.attn_vres_lambda.weight
F32
F32
[1]
blk.25.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.25.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.25.ffn_norm.weight
F32
F32
[5120]
blk.25.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.25.post_attention_norm.weight
F32
F32
[5120]
blk.25.post_ffw_norm.weight
F32
F32
[5120]
blk.26
blk.26.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.26.attn_k.weight
BF16
BF16
[5120, 1024]
blk.26.attn_k_norm.weight
F32
F32
[128]
blk.26.attn_norm.weight
F32
F32
[5120]
blk.26.attn_output.weight
BF16
BF16
[10240, 5120]
blk.26.attn_q.weight
BF16
BF16
[5120, 10240]
blk.26.attn_q_norm.weight
F32
F32
[128]
blk.26.attn_v.weight
BF16
BF16
[5120, 1024]
blk.26.attn_vres_lambda.weight
F32
F32
[1]
blk.26.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.26.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.26.ffn_norm.weight
F32
F32
[5120]
blk.26.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.26.post_attention_norm.weight
F32
F32
[5120]
blk.26.post_ffw_norm.weight
F32
F32
[5120]
blk.27
blk.27.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.27.attn_k.weight
BF16
BF16
[5120, 1024]
blk.27.attn_k_norm.weight
F32
F32
[128]
blk.27.attn_norm.weight
F32
F32
[5120]
blk.27.attn_output.weight
BF16
BF16
[10240, 5120]
blk.27.attn_q.weight
BF16
BF16
[5120, 10240]
blk.27.attn_q_norm.weight
F32
F32
[128]
blk.27.attn_v.weight
BF16
BF16
[5120, 1024]
blk.27.attn_vres_lambda.weight
F32
F32
[1]
blk.27.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.27.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.27.ffn_norm.weight
F32
F32
[5120]
blk.27.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.27.post_attention_norm.weight
F32
F32
[5120]
blk.27.post_ffw_norm.weight
F32
F32
[5120]
blk.28
blk.28.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.28.attn_k.weight
BF16
BF16
[5120, 1024]
blk.28.attn_k_norm.weight
F32
F32
[128]
blk.28.attn_norm.weight
F32
F32
[5120]
blk.28.attn_output.weight
BF16
BF16
[10240, 5120]
blk.28.attn_q.weight
BF16
BF16
[5120, 10240]
blk.28.attn_q_norm.weight
F32
F32
[128]
blk.28.attn_v.weight
BF16
BF16
[5120, 1024]
blk.28.attn_vres_lambda.weight
F32
F32
[1]
blk.28.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.28.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.28.ffn_norm.weight
F32
F32
[5120]
blk.28.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.28.post_attention_norm.weight
F32
F32
[5120]
blk.28.post_ffw_norm.weight
F32
F32
[5120]
blk.29
blk.29.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.29.attn_k.weight
BF16
BF16
[5120, 1024]
blk.29.attn_k_norm.weight
F32
F32
[128]
blk.29.attn_norm.weight
F32
F32
[5120]
blk.29.attn_output.weight
BF16
BF16
[10240, 5120]
blk.29.attn_q.weight
BF16
BF16
[5120, 10240]
blk.29.attn_q_norm.weight
F32
F32
[128]
blk.29.attn_v.weight
BF16
BF16
[5120, 1024]
blk.29.attn_vres_lambda.weight
F32
F32
[1]
blk.29.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.29.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.29.ffn_norm.weight
F32
F32
[5120]
blk.29.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.29.post_attention_norm.weight
F32
F32
[5120]
blk.29.post_ffw_norm.weight
F32
F32
[5120]
blk.30
blk.30.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.30.attn_k.weight
BF16
BF16
[5120, 1024]
blk.30.attn_k_norm.weight
F32
F32
[128]
blk.30.attn_norm.weight
F32
F32
[5120]
blk.30.attn_output.weight
BF16
BF16
[10240, 5120]
blk.30.attn_q.weight
BF16
BF16
[5120, 10240]
blk.30.attn_q_norm.weight
F32
F32
[128]
blk.30.attn_v.weight
BF16
BF16
[5120, 1024]
blk.30.attn_vres_lambda.weight
F32
F32
[1]
blk.30.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.30.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.30.ffn_norm.weight
F32
F32
[5120]
blk.30.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.30.post_attention_norm.weight
F32
F32
[5120]
blk.30.post_ffw_norm.weight
F32
F32
[5120]
blk.31
blk.31.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.31.attn_k.weight
BF16
BF16
[5120, 1024]
blk.31.attn_k_norm.weight
F32
F32
[128]
blk.31.attn_norm.weight
F32
F32
[5120]
blk.31.attn_output.weight
BF16
BF16
[10240, 5120]
blk.31.attn_q.weight
BF16
BF16
[5120, 10240]
blk.31.attn_q_norm.weight
F32
F32
[128]
blk.31.attn_v.weight
BF16
BF16
[5120, 1024]
blk.31.attn_vres_lambda.weight
F32
F32
[1]
blk.31.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.31.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.31.ffn_norm.weight
F32
F32
[5120]
blk.31.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.31.post_attention_norm.weight
F32
F32
[5120]
blk.31.post_ffw_norm.weight
F32
F32
[5120]
blk.32
blk.32.attn_gate.weight
BF16
BF16
[5120, 10240]
blk.32.attn_k.weight
BF16
BF16
[5120, 1024]
blk.32.attn_k_norm.weight
F32
F32
[128]
blk.32.attn_norm.weight
F32
F32
[5120]
blk.32.attn_output.weight
BF16
BF16
[10240, 5120]
blk.32.attn_q_norm.weight
F32
F32
[128]
blk.32.ffn_down.weight
BF16
BF16
[40960, 5120]
blk.32.ffn_gate.weight
BF16
BF16
[5120, 40960]
blk.32.ffn_norm.weight
F32
F32
[5120]
blk.32.ffn_up.weight
BF16
BF16
[5120, 40960]
blk.32.post_attention_norm.weight
F32
F32
[5120]
blk.32.post_ffw_norm.weight
F32
F32
[5120]
output.weight
BF16
BF16
[5120, 100352]