GitHub
187.8K
Download
Models
Docs
Pricing
Sign in
Download
Models
Docs
Pricing
Sign in
jetelain
/
Ornith-1.5-122b
:latest
5
Downloads
Updated
3 days ago
Cancel
vision
Ornith-1.5-122b:latest
...
/
model
7159d31a6a6c · 11MB
Metadata
general.architecture
qwen35moe
qwen35moe
general.file_type
Q4_K_M
Q4_K_M
qwen35moe.attention.head_count
32
32
qwen35moe.attention.head_count_kv
2
2
qwen35moe.attention.key_length
256
256
qwen35moe.attention.layer_norm_rms_epsilon
1e-06
1e-06
qwen35moe.attention.recurrent_layers
[true, true, true, false, true, ...]
[true, true, true, false, true, ...]
qwen35moe.attention.value_length
256
256
qwen35moe.block_count
49
49
qwen35moe.context_length
262144
262144
qwen35moe.embedding_length
3072
3072
qwen35moe.expert_count
256
256
qwen35moe.expert_feed_forward_length
1024
1024
qwen35moe.expert_shared_feed_forward_length
1024
1024
qwen35moe.expert_used_count
8
8
qwen35moe.full_attention_interval
4
4
qwen35moe.nextn_predict_layers
1
1
qwen35moe.rope.dimension_count
64
64
qwen35moe.rope.dimension_sections
[11, 11, 10, 0]
[11, 11, 10, 0]
qwen35moe.rope.freq_base
1e+07
1e+07
qwen35moe.ssm.conv_kernel
4
4
qwen35moe.ssm.group_count
16
16
qwen35moe.ssm.inner_size
8192
8192
qwen35moe.ssm.state_size
128
128
qwen35moe.ssm.time_step_rank
64
64
tokenizer.ggml.add_bos_token
false
false
tokenizer.ggml.add_eos_token
false
false
tokenizer.ggml.eos_token_id
248046
248046
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.normalizer.nfc
true
true
tokenizer.ggml.padding_token_id
248044
248044
tokenizer.ggml.pre
qwen35
qwen35
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
split.count
3
3
split.no
0
0
split.tensors.count
899
899
Tensor
Name
Type
Shape