aixsatoshi commited on
Commit
47bb733
1 Parent(s): 930d320

Upload 5 files

Browse files
Files changed (5) hide show
  1. config.json +31 -0
  2. mergekit_config.yml +49 -0
  3. tokenizer.json +0 -0
  4. tokenizer.model +3 -0
  5. tokenizer_config.json +43 -0
config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "karakuri-ai/karakuri-lm-8x7b-chat-v0.1",
3
+ "architectures": [
4
+ "MixtralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
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,
17
+ "num_hidden_layers": 60,
18
+ "num_key_value_heads": 8,
19
+ "num_local_experts": 8,
20
+ "output_router_logits": false,
21
+ "rms_norm_eps": 1e-05,
22
+ "rope_theta": 1000000.0,
23
+ "router_aux_loss_coef": 0.02,
24
+ "router_jitter_noise": 0.0,
25
+ "sliding_window": null,
26
+ "tie_word_embeddings": false,
27
+ "torch_dtype": "bfloat16",
28
+ "transformers_version": "4.40.0.dev0",
29
+ "use_cache": true,
30
+ "vocab_size": 32000
31
+ }
mergekit_config.yml ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ slices:
2
+ - sources:
3
+ - layer_range: [0, 4]
4
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
5
+ - sources:
6
+ - layer_range: [2, 6]
7
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
8
+ - sources:
9
+ - layer_range: [4, 8]
10
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
11
+ - sources:
12
+ - layer_range: [6, 10]
13
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
14
+ - sources:
15
+ - layer_range: [8, 12]
16
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
17
+ - sources:
18
+ - layer_range: [10, 14]
19
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
20
+ - sources:
21
+ - layer_range: [12, 16]
22
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
23
+ - sources:
24
+ - layer_range: [14, 18]
25
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
26
+ - sources:
27
+ - layer_range: [16, 20]
28
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
29
+ - sources:
30
+ - layer_range: [18, 22]
31
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
32
+ - sources:
33
+ - layer_range: [20, 24]
34
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
35
+ - sources:
36
+ - layer_range: [22, 26]
37
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
38
+ - sources:
39
+ - layer_range: [24, 28]
40
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
41
+ - sources:
42
+ - layer_range: [26, 30]
43
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
44
+ - sources:
45
+ - layer_range: [28, 32]
46
+ model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
47
+ merge_method: passthrough
48
+ dtype: bfloat16
49
+
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{% set helpfulness = message['helpfulness']|string or '4' %}{% set correctness = message['correctness']|string or '4' %}{% set coherence = message['coherence']|string or '4' %}{% set complexity = message['complexity']|string or '4' %}{% set verbosity = message['verbosity']|string or '4' %}{% set quality = message['quality']|string or '4' %}{% set toxicity = message['toxicity']|string or '0' %}{% set humor = message['humor']|string or '0' %}{% set creativity = message['creativity']|string or '0' %}{{ '[INST] ' + content + ' [ATTR] helpfulness: ' + helpfulness + ' correctness: ' + correctness + ' coherence: ' + coherence + ' complexity: ' + complexity + ' verbosity: ' + verbosity + ' quality: ' + quality + ' toxicity: ' + toxicity + ' humor: ' + humor + ' creativity: ' + creativity + ' [/ATTR] [/INST]' }}{% elif message['role'] == 'assistant' %}{{ content + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": null,
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }