huggingkot commited on
Commit ·
3ce4698
1
Parent(s): 9a7da16
add files
Browse files- .gitattributes +1 -0
- README.md +17 -0
- chat_template.jinja +25 -0
- config.json +3 -0
- generation_config.json +3 -0
- model-00001-of-00003.safetensors +3 -0
- model-00002-of-00003.safetensors +3 -0
- model-00003-of-00003.safetensors +3 -0
- model.safetensors.index.json +3 -0
- special_tokens_map.json +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ 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 |
+
*.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
base_model:
|
| 4 |
+
- TroyDoesAI/BlackSheep-24B
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
This is a converted weight from [BlackSheep-24B](https://huggingface.co/TroyDoesAI/BlackSheep-24B) model in [unsloth 4-bit dynamic quant](https://archive.is/EFz7P) using this [collab notebook](https://colab.research.google.com/drive/1P23C66j3ga49kBRnDNlmRce7R_l_-L5l?usp=sharing).
|
| 8 |
+
|
| 9 |
+
## About this Conversion
|
| 10 |
+
|
| 11 |
+
This conversion uses **Unsloth** to load the model in **4-bit** format and force-save it in the same **4-bit** format.
|
| 12 |
+
|
| 13 |
+
### How 4-bit Quantization Works
|
| 14 |
+
- The actual **4-bit quantization** is handled by **BitsAndBytes (bnb)**, which works under **Torch** via **AutoGPTQ** or **BitsAndBytes**.
|
| 15 |
+
- **Unsloth** acts as a wrapper, simplifying and optimizing the process for better efficiency.
|
| 16 |
+
|
| 17 |
+
This allows for reduced memory usage and faster inference while keeping the model compact.
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set today = strftime_now("%Y-%m-%d") %}
|
| 2 |
+
{%- set default_system_message = "Mistral Small 3, an Uncensored Large Language Model created by TroyDoesAI. The current date is " + today + ".\n\nWhen not sure of some information, say I don't have the information.\nIf the user's question is not clear, ambiguous, or does not provide enough context to accurately answer the question, do not try to answer it right away, ask the user to clarify their request (e.g. \"Where can I get some good ass tacos near me?\" => \"Where are you?\" or \"When's the next flight to Tokyo\" => \"Where you traveling from?\")" %}
|
| 3 |
+
|
| 4 |
+
{{- bos_token }}
|
| 5 |
+
|
| 6 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 7 |
+
{%- set system_message = messages[0]['content'] %}
|
| 8 |
+
{%- set loop_messages = messages[1:] %}
|
| 9 |
+
{%- else %}
|
| 10 |
+
{%- set system_message = default_system_message %}
|
| 11 |
+
{%- set loop_messages = messages %}
|
| 12 |
+
{%- endif %}
|
| 13 |
+
{{- '[SYSTEM_PROMPT]' + system_message + '[/SYSTEM_PROMPT]' }}
|
| 14 |
+
|
| 15 |
+
{%- for message in loop_messages %}
|
| 16 |
+
{%- if message['role'] == 'user' %}
|
| 17 |
+
{{- '[INST]' + message['content'] + '[/INST]' }}
|
| 18 |
+
{%- elif message['role'] == 'system' %}
|
| 19 |
+
{{- '[SYSTEM_PROMPT]' + message['content'] + '[/SYSTEM_PROMPT]' }}
|
| 20 |
+
{%- elif message['role'] == 'assistant' %}
|
| 21 |
+
{{- message['content'] + eos_token }}
|
| 22 |
+
{%- else %}
|
| 23 |
+
{{- raise_exception('Only user, system and assistant roles are supported!') }}
|
| 24 |
+
{%- endif %}
|
| 25 |
+
{%- endfor %}
|
config.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a80eea64e4f533495038d084a2dab236cb4aa04154f0488637b2ec40e0a8ab56
|
| 3 |
+
size 1079
|
generation_config.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:006fb887c5d0d33de21d25144ac1691f1680c7fd51b20c2cdd073c749baf3c95
|
| 3 |
+
size 199
|
model-00001-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80964144246947f480c9bfd4c72541d5e7259cfb432598a6efdb7ad0f58a20c2
|
| 3 |
+
size 4983049324
|
model-00002-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4af1cff658561d53bcf50060105e8a1837064c2793e1f245674995b0b3dd2936
|
| 3 |
+
size 4987974536
|
model-00003-of-00003.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc2a26afd0bb1ce28aae0979573aea492eca3e6a1ff14486a90fed1fa6f5ec10
|
| 3 |
+
size 4182397548
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3278c76dc96ef03b97bc99b1e2c8602b49b08dd994f533241434cf523e202284
|
| 3 |
+
size 165421
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e2723cf0c325652adb19cf4cadaa265054fc44c106e161bee36f13ce1b21a9b
|
| 3 |
+
size 21335
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b76085f9923309d873994d444989f7eb6ec074b06f25b58f1e8d7b7741070949
|
| 3 |
+
size 17078037
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e6d43894e22793841d73b4d16e3a2218edda4c4e18b0e625c58309a386695e9
|
| 3 |
+
size 198079
|