upload model weights files
Browse files- config.json +43 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +19 -0
config.json
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_dropout": 0.0,
|
3 |
+
"attention_dropout": 0.0,
|
4 |
+
"attn_cfg": {
|
5 |
+
"ffn_expand_ratio": 1.3333333333333333,
|
6 |
+
"num_heads": 16,
|
7 |
+
"num_heads_k": 1,
|
8 |
+
"num_heads_v": 16,
|
9 |
+
"rotary_emb_dim": -1,
|
10 |
+
"rotary_emb_scale_base": null,
|
11 |
+
"window_size": 4096
|
12 |
+
},
|
13 |
+
"d_model": 2048,
|
14 |
+
"dropout": 0.0,
|
15 |
+
"initializer_range": 0.02,
|
16 |
+
"max_position_embeddings": 4096,
|
17 |
+
"mixer_cfg": {
|
18 |
+
"expand_ratio": 2.0,
|
19 |
+
"input_gate_low_rank": "auto",
|
20 |
+
"mem_size": 64,
|
21 |
+
"mode": "fused_chunk",
|
22 |
+
"normalize_epsilon": 1e-05
|
23 |
+
},
|
24 |
+
"model_type": "rodimus",
|
25 |
+
"n_layer": 24,
|
26 |
+
"norm_epsilon": 1e-05,
|
27 |
+
"residual_in_fp32": true,
|
28 |
+
"tie_word_embeddings": true,
|
29 |
+
"transformers_version": "4.37.2",
|
30 |
+
"use_cache": true,
|
31 |
+
"use_fast_path": true,
|
32 |
+
"vocab_size": 126464,
|
33 |
+
"eos_token_id": 126081,
|
34 |
+
"bos_token_id": 126080,
|
35 |
+
"pad_token_id": null,
|
36 |
+
"block_type": "rodimus_plus",
|
37 |
+
"auto_map": {
|
38 |
+
"AutoConfig": "configuration_rodimus.RodimusConfig",
|
39 |
+
"AutoModel": "modeling_rodimus.RodimusForCausalLM",
|
40 |
+
"AutoModelForCausalLM": "modeling_rodimus.RodimusForCausalLM"
|
41 |
+
},
|
42 |
+
"torch_dtype": "float16"
|
43 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:297ef59a0ed5c33e8fb81db81a72414aa77159841847e2c44fa4edd415af82c9
|
3 |
+
size 7464659434
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|startoftext|>",
|
3 |
+
"cls_token": "[CLS]",
|
4 |
+
"eos_token": "<|endoftext|>",
|
5 |
+
"gmask_token": "[gMASK]"
|
6 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": false,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"bos_token": "<|startoftext|>",
|
5 |
+
"clean_up_tokenization_spaces": false,
|
6 |
+
"cls_token": "[CLS]",
|
7 |
+
"eos_token": "<|endoftext|>",
|
8 |
+
"gmask_token": "[gMASK]",
|
9 |
+
"merges_file": null,
|
10 |
+
"model_max_length": 1000000000000000019884624838656,
|
11 |
+
"tokenizer_class": "RodimusTokenizer",
|
12 |
+
"vocab_file": null,
|
13 |
+
"auto_map": {
|
14 |
+
"AutoTokenizer": [
|
15 |
+
null,
|
16 |
+
"tokenization_rodimus_fast.RodimusTokenizer"
|
17 |
+
]
|
18 |
+
}
|
19 |
+
}
|