Upload folder using huggingface_hub
Browse files- .gitattributes +17 -0
- README.md +143 -0
- config.json +24 -0
- mistral-7b-instruct-v0.1.Q2_K.gguf +3 -0
- mistral-7b-instruct-v0.1.Q3_K.gguf +3 -0
- mistral-7b-instruct-v0.1.Q3_K_L.gguf +3 -0
- mistral-7b-instruct-v0.1.Q3_K_M.gguf +3 -0
- mistral-7b-instruct-v0.1.Q3_K_S.gguf +3 -0
- mistral-7b-instruct-v0.1.Q4_0.gguf +3 -0
- mistral-7b-instruct-v0.1.Q4_1.gguf +3 -0
- mistral-7b-instruct-v0.1.Q4_K.gguf +3 -0
- mistral-7b-instruct-v0.1.Q4_K_M.gguf +3 -0
- mistral-7b-instruct-v0.1.Q4_K_S.gguf +3 -0
- mistral-7b-instruct-v0.1.Q5_0.gguf +3 -0
- mistral-7b-instruct-v0.1.Q5_1.gguf +3 -0
- mistral-7b-instruct-v0.1.Q5_K.gguf +3 -0
- mistral-7b-instruct-v0.1.Q5_K_M.gguf +3 -0
- mistral-7b-instruct-v0.1.Q5_K_S.gguf +3 -0
- mistral-7b-instruct-v0.1.Q6_K.gguf +3 -0
- mistral-7b-instruct-v0.1.Q8_0.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,20 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
mistral-7b-instruct-v0.1.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
37 |
+
mistral-7b-instruct-v0.1.Q3_K.gguf filter=lfs diff=lfs merge=lfs -text
|
38 |
+
mistral-7b-instruct-v0.1.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
39 |
+
mistral-7b-instruct-v0.1.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
40 |
+
mistral-7b-instruct-v0.1.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
41 |
+
mistral-7b-instruct-v0.1.Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
|
42 |
+
mistral-7b-instruct-v0.1.Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
|
43 |
+
mistral-7b-instruct-v0.1.Q4_K.gguf filter=lfs diff=lfs merge=lfs -text
|
44 |
+
mistral-7b-instruct-v0.1.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
45 |
+
mistral-7b-instruct-v0.1.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
46 |
+
mistral-7b-instruct-v0.1.Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
|
47 |
+
mistral-7b-instruct-v0.1.Q5_1.gguf filter=lfs diff=lfs merge=lfs -text
|
48 |
+
mistral-7b-instruct-v0.1.Q5_K.gguf filter=lfs diff=lfs merge=lfs -text
|
49 |
+
mistral-7b-instruct-v0.1.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
50 |
+
mistral-7b-instruct-v0.1.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
51 |
+
mistral-7b-instruct-v0.1.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
52 |
+
mistral-7b-instruct-v0.1.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,143 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- finetuned
|
5 |
+
base_model: mistralai/Mistral-7B-v0.1
|
6 |
+
pipeline_tag: text-generation
|
7 |
+
inference: true
|
8 |
+
widget:
|
9 |
+
- messages:
|
10 |
+
- role: user
|
11 |
+
content: What is your favorite condiment?
|
12 |
+
|
13 |
+
extra_gated_description: If you want to learn more about how we process your personal data, please read our <a href="https://mistral.ai/terms/">Privacy Policy</a>.
|
14 |
+
---
|
15 |
+
|
16 |
+
# Model Card for Mistral-7B-Instruct-v0.1
|
17 |
+
|
18 |
+
|
19 |
+
## Encode and Decode with `mistral_common`
|
20 |
+
|
21 |
+
```py
|
22 |
+
from mistral_common.tokens.tokenizers.mistral import MistralTokenizer
|
23 |
+
from mistral_common.protocol.instruct.messages import UserMessage
|
24 |
+
from mistral_common.protocol.instruct.request import ChatCompletionRequest
|
25 |
+
|
26 |
+
mistral_models_path = "MISTRAL_MODELS_PATH"
|
27 |
+
|
28 |
+
tokenizer = MistralTokenizer.v1()
|
29 |
+
|
30 |
+
completion_request = ChatCompletionRequest(messages=[UserMessage(content="Explain Machine Learning to me in a nutshell.")])
|
31 |
+
|
32 |
+
tokens = tokenizer.encode_chat_completion(completion_request).tokens
|
33 |
+
```
|
34 |
+
|
35 |
+
## Inference with `mistral_inference`
|
36 |
+
|
37 |
+
```py
|
38 |
+
from mistral_inference.transformer import Transformer
|
39 |
+
from mistral_inference.generate import generate
|
40 |
+
|
41 |
+
model = Transformer.from_folder(mistral_models_path)
|
42 |
+
out_tokens, _ = generate([tokens], model, max_tokens=64, temperature=0.0, eos_id=tokenizer.instruct_tokenizer.tokenizer.eos_id)
|
43 |
+
|
44 |
+
result = tokenizer.decode(out_tokens[0])
|
45 |
+
|
46 |
+
print(result)
|
47 |
+
```
|
48 |
+
|
49 |
+
## Inference with hugging face `transformers`
|
50 |
+
|
51 |
+
```py
|
52 |
+
from transformers import AutoModelForCausalLM
|
53 |
+
|
54 |
+
model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
|
55 |
+
model.to("cuda")
|
56 |
+
|
57 |
+
generated_ids = model.generate(tokens, max_new_tokens=1000, do_sample=True)
|
58 |
+
|
59 |
+
# decode with mistral tokenizer
|
60 |
+
result = tokenizer.decode(generated_ids[0].tolist())
|
61 |
+
print(result)
|
62 |
+
```
|
63 |
+
|
64 |
+
> [!TIP]
|
65 |
+
> PRs to correct the `transformers` tokenizer so that it gives 1-to-1 the same results as the `mistral_common` reference implementation are very welcome!
|
66 |
+
|
67 |
+
---
|
68 |
+
|
69 |
+
The Mistral-7B-Instruct-v0.1 Large Language Model (LLM) is a instruct fine-tuned version of the [Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) generative text model using a variety of publicly available conversation datasets.
|
70 |
+
|
71 |
+
For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/announcing-mistral-7b/).
|
72 |
+
|
73 |
+
## Instruction format
|
74 |
+
|
75 |
+
In order to leverage instruction fine-tuning, your prompt should be surrounded by `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin of sentence id. The next instructions should not. The assistant generation will be ended by the end-of-sentence token id.
|
76 |
+
|
77 |
+
E.g.
|
78 |
+
```
|
79 |
+
text = "<s>[INST] What is your favourite condiment? [/INST]"
|
80 |
+
"Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
|
81 |
+
"[INST] Do you have mayonnaise recipes? [/INST]"
|
82 |
+
```
|
83 |
+
|
84 |
+
This format is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating) via the `apply_chat_template()` method:
|
85 |
+
|
86 |
+
```python
|
87 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
88 |
+
|
89 |
+
device = "cuda" # the device to load the model onto
|
90 |
+
|
91 |
+
model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
|
92 |
+
tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.1")
|
93 |
+
|
94 |
+
messages = [
|
95 |
+
{"role": "user", "content": "What is your favourite condiment?"},
|
96 |
+
{"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
|
97 |
+
{"role": "user", "content": "Do you have mayonnaise recipes?"}
|
98 |
+
]
|
99 |
+
|
100 |
+
encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
|
101 |
+
|
102 |
+
model_inputs = encodeds.to(device)
|
103 |
+
model.to(device)
|
104 |
+
|
105 |
+
generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
|
106 |
+
decoded = tokenizer.batch_decode(generated_ids)
|
107 |
+
print(decoded[0])
|
108 |
+
```
|
109 |
+
|
110 |
+
## Model Architecture
|
111 |
+
This instruction model is based on Mistral-7B-v0.1, a transformer model with the following architecture choices:
|
112 |
+
- Grouped-Query Attention
|
113 |
+
- Sliding-Window Attention
|
114 |
+
- Byte-fallback BPE tokenizer
|
115 |
+
|
116 |
+
## Troubleshooting
|
117 |
+
- If you see the following error:
|
118 |
+
```
|
119 |
+
Traceback (most recent call last):
|
120 |
+
File "", line 1, in
|
121 |
+
File "/transformers/models/auto/auto_factory.py", line 482, in from_pretrained
|
122 |
+
config, kwargs = AutoConfig.from_pretrained(
|
123 |
+
File "/transformers/models/auto/configuration_auto.py", line 1022, in from_pretrained
|
124 |
+
config_class = CONFIG_MAPPING[config_dict["model_type"]]
|
125 |
+
File "/transformers/models/auto/configuration_auto.py", line 723, in getitem
|
126 |
+
raise KeyError(key)
|
127 |
+
KeyError: 'mistral'
|
128 |
+
```
|
129 |
+
|
130 |
+
Installing transformers from source should solve the issue
|
131 |
+
pip install git+https://github.com/huggingface/transformers
|
132 |
+
|
133 |
+
This should not be required after transformers-v4.33.4.
|
134 |
+
|
135 |
+
## Limitations
|
136 |
+
|
137 |
+
The Mistral 7B Instruct model is a quick demonstration that the base model can be easily fine-tuned to achieve compelling performance.
|
138 |
+
It does not have any moderation mechanisms. We're looking forward to engaging with the community on ways to
|
139 |
+
make the model finely respect guardrails, allowing for deployment in environments requiring moderated outputs.
|
140 |
+
|
141 |
+
## The Mistral AI Team
|
142 |
+
|
143 |
+
Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|
config.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"MistralForCausalLM"
|
4 |
+
],
|
5 |
+
"bos_token_id": 1,
|
6 |
+
"eos_token_id": 2,
|
7 |
+
"hidden_act": "silu",
|
8 |
+
"hidden_size": 4096,
|
9 |
+
"initializer_range": 0.02,
|
10 |
+
"intermediate_size": 14336,
|
11 |
+
"max_position_embeddings": 32768,
|
12 |
+
"model_type": "mistral",
|
13 |
+
"num_attention_heads": 32,
|
14 |
+
"num_hidden_layers": 32,
|
15 |
+
"num_key_value_heads": 8,
|
16 |
+
"rms_norm_eps": 1e-05,
|
17 |
+
"rope_theta": 10000.0,
|
18 |
+
"sliding_window": 4096,
|
19 |
+
"tie_word_embeddings": false,
|
20 |
+
"torch_dtype": "bfloat16",
|
21 |
+
"transformers_version": "4.34.0.dev0",
|
22 |
+
"use_cache": true,
|
23 |
+
"vocab_size": 32000
|
24 |
+
}
|
mistral-7b-instruct-v0.1.Q2_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aa26dc10c3739e2a68f327876344bbb238396bc81677a1e8b7672e1be80b1261
|
3 |
+
size 2719243808
|
mistral-7b-instruct-v0.1.Q3_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c3d1926c9ab518c5c2debf9dad208a4c20c2f49bf34e06cbcb79f4b4d9e36564
|
3 |
+
size 3518987808
|
mistral-7b-instruct-v0.1.Q3_K_L.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ad7675be40c656c8f4f1cd8ad1f4aa04fc145e468645402255d17d91ce7d7125
|
3 |
+
size 3822026272
|
mistral-7b-instruct-v0.1.Q3_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c3d1926c9ab518c5c2debf9dad208a4c20c2f49bf34e06cbcb79f4b4d9e36564
|
3 |
+
size 3518987808
|
mistral-7b-instruct-v0.1.Q3_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a10a9f3a42271df1ed117f6cbba589f8a2c36dc4c8fee6e08d73bc653f78b81
|
3 |
+
size 3164569120
|
mistral-7b-instruct-v0.1.Q4_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5a7d4f43933ccf111c0a586ba027698e1d111173a595dd8e4bf28d00ae6cbeb
|
3 |
+
size 4108918304
|
mistral-7b-instruct-v0.1.Q4_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9cf789fd37e33e9f01e71cae50e1a814dd44569ddedee3ec6192d439d0365a61
|
3 |
+
size 4553317920
|
mistral-7b-instruct-v0.1.Q4_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afe690a8d8dd5021f0b3abc1aa25e3379197c8fd7b1c712b4dc64d4bf3106ecb
|
3 |
+
size 4368440864
|
mistral-7b-instruct-v0.1.Q4_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afe690a8d8dd5021f0b3abc1aa25e3379197c8fd7b1c712b4dc64d4bf3106ecb
|
3 |
+
size 4368440864
|
mistral-7b-instruct-v0.1.Q4_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3bc7ba03de9880339b9d9a8d5152ccfe9328ad1d49ed3441e003088dac49b3e6
|
3 |
+
size 4140375584
|
mistral-7b-instruct-v0.1.Q5_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:84a763fa1eae58959d60bb2f91180279540001e7cce22facfe991bf073d49ef0
|
3 |
+
size 4997717536
|
mistral-7b-instruct-v0.1.Q5_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ba782381352b7a518f9d0b76bc7c9010a5428dbd135bca2c7dd19a734296a621
|
3 |
+
size 5442117152
|
mistral-7b-instruct-v0.1.Q5_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f8df439e21b9680cb7668e4483eb165ff31d2f54b11a44f6206006877c80ee00
|
3 |
+
size 5131410976
|
mistral-7b-instruct-v0.1.Q5_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f8df439e21b9680cb7668e4483eb165ff31d2f54b11a44f6206006877c80ee00
|
3 |
+
size 5131410976
|
mistral-7b-instruct-v0.1.Q5_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b44de61c1b7876a57269586ff79a387a88074a8100d9ff513b95a0222ceb4ac4
|
3 |
+
size 4997717536
|
mistral-7b-instruct-v0.1.Q6_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b833423c65b573fd0803ce2d47c315a344dd6bf2ce679692dbd3129583e38a1c
|
3 |
+
size 5942066720
|
mistral-7b-instruct-v0.1.Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fc3ee6009901673dedbd30430dc599495b55407f6c1ccaf0b434fa016640456e
|
3 |
+
size 7695859232
|