mlabonne commited on
Commit
c615571
1 Parent(s): b20c742

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -10,7 +10,7 @@
10
  "hidden_size": 4096,
11
  "initializer_range": 0.02,
12
  "intermediate_size": 14336,
13
- "max_position_embeddings": 8192,
14
  "model_type": "mixtral",
15
  "num_attention_heads": 32,
16
  "num_experts_per_tok": 2,
@@ -24,7 +24,7 @@
24
  "sliding_window": null,
25
  "tie_word_embeddings": false,
26
  "torch_dtype": "bfloat16",
27
- "transformers_version": "4.36.2",
28
  "use_cache": true,
29
  "vocab_size": 32000
30
  }
 
10
  "hidden_size": 4096,
11
  "initializer_range": 0.02,
12
  "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
  "model_type": "mixtral",
15
  "num_attention_heads": 32,
16
  "num_experts_per_tok": 2,
 
24
  "sliding_window": null,
25
  "tie_word_embeddings": false,
26
  "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.37.1",
28
  "use_cache": true,
29
  "vocab_size": 32000
30
  }
mergekit_moe_config.yml ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ base_model: mlabonne/Marcoro14-7B-slerp
3
+ experts:
4
+ - source_model: openchat/openchat-3.5-1210
5
+ positive_prompts:
6
+ - "chat"
7
+ - "assistant"
8
+ - "tell me"
9
+ - "explain"
10
+ - source_model: beowolx/CodeNinja-1.0-OpenChat-7B
11
+ positive_prompts:
12
+ - "code"
13
+ - "python"
14
+ - "javascript"
15
+ - "programming"
16
+ - "algorithm"
17
+ - source_model: maywell/PiVoT-0.1-Starling-LM-RP
18
+ positive_prompts:
19
+ - "storywriting"
20
+ - "write"
21
+ - "scene"
22
+ - "story"
23
+ - "character"
24
+ - source_model: WizardLM/WizardMath-7B-V1.1
25
+ positive_prompts:
26
+ - "reason"
27
+ - "math"
28
+ - "mathematics"
29
+ - "solve"
30
+ - "count"
31
+ tokenizer_source: union
model-00005-of-00005.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79a84b6fdb3ece83c2703ab2de0a3333aeeafd213b872d82a8261de61e63429c
3
  size 8440279464
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1e3cc831230c664a8590c85633fccc26a93e51a0bcfb8ea3dd9039c22847632
3
  size 8440279464
model.safetensors.index.json CHANGED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json CHANGED
@@ -36,12 +36,11 @@
36
  "clean_up_tokenization_spaces": false,
37
  "eos_token": "</s>",
38
  "legacy": true,
39
- "model_max_length": 8192,
40
  "pad_token": "<s>",
41
  "sp_model_kwargs": {},
42
  "spaces_between_special_tokens": false,
43
  "tokenizer_class": "LlamaTokenizer",
44
  "unk_token": "<unk>",
45
- "chat_template": "{{ bos_token }}{% for message in messages %}{{ 'GPT4 Correct ' + message['role'].title() + ': ' + message['content'] + eos_token}}{% endfor %}{% if add_generation_prompt %}{{ 'GPT4 Correct Assistant:' }}{% endif %}",
46
  "use_default_system_prompt": true
47
- }
 
36
  "clean_up_tokenization_spaces": false,
37
  "eos_token": "</s>",
38
  "legacy": true,
39
+ "model_max_length": 1000000000000000019884624838656,
40
  "pad_token": "<s>",
41
  "sp_model_kwargs": {},
42
  "spaces_between_special_tokens": false,
43
  "tokenizer_class": "LlamaTokenizer",
44
  "unk_token": "<unk>",
 
45
  "use_default_system_prompt": true
46
+ }