Upload folder using huggingface_hub
Browse files- config.json +60 -57
- model.safetensors +2 -2
config.json
CHANGED
@@ -1,82 +1,85 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
"add_bias_linear": false,
|
4 |
-
"architectures": [
|
5 |
-
"Zamba2ForCausalLM"
|
6 |
-
],
|
7 |
"attention_dropout": 0.0,
|
|
|
|
|
8 |
"bos_token_id": 1,
|
9 |
-
"
|
10 |
"eos_token_id": 2,
|
11 |
-
"expansion_factor": 2,
|
12 |
"ffn_hidden_size": 8192,
|
13 |
-
"
|
14 |
-
"gated_linear_unit": true,
|
15 |
"hidden_size": 2048,
|
16 |
"initializer_range": 0.02,
|
|
|
17 |
"kv_channels": 64,
|
18 |
"layers_block_type": [
|
19 |
-
"
|
20 |
-
"
|
21 |
-
"
|
22 |
-
"
|
23 |
-
"
|
24 |
-
"
|
25 |
-
"
|
26 |
-
"
|
27 |
-
"
|
28 |
-
"
|
29 |
-
"
|
30 |
-
"
|
31 |
-
"
|
32 |
-
"
|
33 |
-
"
|
34 |
-
"
|
35 |
-
"
|
36 |
-
"
|
37 |
-
"
|
38 |
-
"
|
39 |
-
"
|
40 |
-
"
|
41 |
-
"
|
42 |
-
"
|
43 |
-
"
|
44 |
-
"
|
45 |
-
"
|
46 |
-
"
|
47 |
-
"
|
48 |
-
"
|
49 |
-
"
|
50 |
-
"
|
51 |
-
"
|
52 |
-
"
|
53 |
-
"
|
54 |
-
"
|
55 |
-
"
|
56 |
-
"
|
57 |
],
|
58 |
-
"
|
59 |
-
"
|
|
|
60 |
"mamba_headdim": 64,
|
|
|
61 |
"max_position_embeddings": 4096,
|
62 |
"model_type": "zamba2",
|
|
|
63 |
"num_attention_heads": 32,
|
64 |
"num_hidden_layers": 38,
|
65 |
"num_key_value_heads": 32,
|
66 |
"num_logits_to_keep": 1,
|
67 |
"num_mem_blocks": 1,
|
68 |
"num_query_groups": 32,
|
69 |
-
"pad_token_id":
|
70 |
"rms_norm_eps": 1e-05,
|
71 |
"rope_theta": 10000,
|
72 |
-
"
|
73 |
-
"
|
74 |
-
"
|
75 |
-
"
|
76 |
-
"
|
77 |
-
"
|
|
|
|
|
78 |
"use_mem_rope": true,
|
79 |
-
"
|
80 |
-
"
|
81 |
"vocab_size": 32000
|
82 |
}
|
|
|
1 |
{
|
2 |
+
"adapter_rank": 128,
|
3 |
"add_bias_linear": false,
|
|
|
|
|
|
|
4 |
"attention_dropout": 0.0,
|
5 |
+
"attention_head_dim": 128,
|
6 |
+
"attention_hidden_size": 4096,
|
7 |
"bos_token_id": 1,
|
8 |
+
"chunk_size": 256,
|
9 |
"eos_token_id": 2,
|
|
|
10 |
"ffn_hidden_size": 8192,
|
11 |
+
"hidden_act": "gelu",
|
|
|
12 |
"hidden_size": 2048,
|
13 |
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 8192,
|
15 |
"kv_channels": 64,
|
16 |
"layers_block_type": [
|
17 |
+
"mamba",
|
18 |
+
"mamba",
|
19 |
+
"mamba",
|
20 |
+
"mamba",
|
21 |
+
"mamba",
|
22 |
+
"hybrid",
|
23 |
+
"mamba",
|
24 |
+
"mamba",
|
25 |
+
"mamba",
|
26 |
+
"mamba",
|
27 |
+
"mamba",
|
28 |
+
"hybrid",
|
29 |
+
"mamba",
|
30 |
+
"mamba",
|
31 |
+
"mamba",
|
32 |
+
"mamba",
|
33 |
+
"mamba",
|
34 |
+
"hybrid",
|
35 |
+
"mamba",
|
36 |
+
"mamba",
|
37 |
+
"mamba",
|
38 |
+
"mamba",
|
39 |
+
"mamba",
|
40 |
+
"hybrid",
|
41 |
+
"mamba",
|
42 |
+
"mamba",
|
43 |
+
"mamba",
|
44 |
+
"mamba",
|
45 |
+
"mamba",
|
46 |
+
"hybrid",
|
47 |
+
"mamba",
|
48 |
+
"mamba",
|
49 |
+
"mamba",
|
50 |
+
"mamba",
|
51 |
+
"mamba",
|
52 |
+
"hybrid",
|
53 |
+
"mamba",
|
54 |
+
"mamba"
|
55 |
],
|
56 |
+
"mamba_d_conv": 4,
|
57 |
+
"mamba_d_state": 128,
|
58 |
+
"mamba_expand": 2,
|
59 |
"mamba_headdim": 64,
|
60 |
+
"mamba_ngroups": 1,
|
61 |
"max_position_embeddings": 4096,
|
62 |
"model_type": "zamba2",
|
63 |
+
"n_mamba_heads": 64,
|
64 |
"num_attention_heads": 32,
|
65 |
"num_hidden_layers": 38,
|
66 |
"num_key_value_heads": 32,
|
67 |
"num_logits_to_keep": 1,
|
68 |
"num_mem_blocks": 1,
|
69 |
"num_query_groups": 32,
|
70 |
+
"pad_token_id": 0,
|
71 |
"rms_norm_eps": 1e-05,
|
72 |
"rope_theta": 10000,
|
73 |
+
"time_step_floor": 0.0001,
|
74 |
+
"time_step_limit": null,
|
75 |
+
"time_step_max": 0.1,
|
76 |
+
"time_step_min": 0.001,
|
77 |
+
"transformers_version": "4.49.0.dev0",
|
78 |
+
"use_cache": true,
|
79 |
+
"use_conv_bias": true,
|
80 |
+
"use_long_context": false,
|
81 |
"use_mem_rope": true,
|
82 |
+
"use_shared_attention_adapter": true,
|
83 |
+
"use_shared_mlp_adapter": true,
|
84 |
"vocab_size": 32000
|
85 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b442e3dd5477526c396465a1342c23ce7468d4c60e973224691e79af81d65328
|
3 |
+
size 4860300264
|