TheBloke commited on
Commit
c8bd255
1 Parent(s): ed3156c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +58 -3
README.md CHANGED
@@ -1,6 +1,48 @@
1
  ---
2
  inference: false
3
- license: other
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4
  ---
5
 
6
  <!-- header start -->
@@ -29,6 +71,18 @@ It is the result of quantising to 4bit using [GPTQ-for-LLaMa](https://github.com
29
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/minotaur-15B-GGML)
30
  * [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/openaccess-ai-collective/minotaur-15b)
31
 
 
 
 
 
 
 
 
 
 
 
 
 
32
  ## How to easily download and use this model in text-generation-webui
33
 
34
  Please make sure you're using the latest version of text-generation-webui
@@ -74,8 +128,8 @@ model = AutoGPTQForCausalLM.from_quantized(model_name_or_path,
74
 
75
  # Note: check the prompt template is correct for this model.
76
  prompt = "Tell me about AI"
77
- prompt_template=f'''### Human: {prompt}
78
- ### Assistant:'''
79
 
80
  print("\n\n*** Generate:")
81
 
@@ -114,6 +168,7 @@ It was created with group_size 128 to increase inference accuracy, but without -
114
  * Works with AutoGPTQ in CUDA or Triton modes.
115
  * Works with GPTQ-for-LLaMa in CUDA mode. May have issues with GPTQ-for-LLaMa Triton mode.
116
  * Works with text-generation-webui, including one-click-installers.
 
117
  * Parameters: Groupsize = 128. Act Order / desc_act = False.
118
 
119
  <!-- footer start -->
 
1
  ---
2
  inference: false
3
+ pipeline_tag: text-generation
4
+ widget:
5
+ - text: 'def print_hello_world():'
6
+ example_title: Hello world
7
+ group: Python
8
+ - text: 'Gradient descent is'
9
+ example_title: Machine Learning
10
+ group: English
11
+ - license: bigcode-openrail-m
12
+ datasets:
13
+ - bigcode/the-stack-dedup
14
+ - tiiuae/falcon-refinedweb
15
+ - ehartford/WizardLM_alpaca_evol_instruct_70k_unfiltered
16
+ - QingyiSi/Alpaca-CoT
17
+ - teknium/GPTeacher-General-Instruct
18
+ - metaeval/ScienceQA_text_only
19
+ - hellaswag
20
+ - openai/summarize_from_feedback
21
+ - riddle_sense
22
+ - gsm8k
23
+ - camel-ai/math
24
+ - camel-ai/biology
25
+ - camel-ai/physics
26
+ - camel-ai/chemistry
27
+ - winglian/evals
28
+ metrics:
29
+ - code_eval
30
+ - mmlu
31
+ - arc
32
+ - hellaswag
33
+ - truthfulqa
34
+ library_name: transformers
35
+ tags:
36
+ - code
37
+ extra_gated_prompt: >-
38
+ ## Model License Agreement
39
+
40
+ Please read the BigCode [OpenRAIL-M
41
+ license](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement)
42
+ agreement before accepting it.
43
+
44
+ extra_gated_fields:
45
+ I accept the above license agreement, and will use the Model complying with the set of use restrictions and sharing requirements: checkbox
46
  ---
47
 
48
  <!-- header start -->
 
71
  * [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/minotaur-15B-GGML)
72
  * [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/openaccess-ai-collective/minotaur-15b)
73
 
74
+ ## Note about context length
75
+
76
+ It is currently untested as to whether the 8K context is compatible with available clients such as text-generation-webui.
77
+
78
+ If you have feedback on this, please let me know.
79
+
80
+ ## Prompt template
81
+ ```
82
+ USER: <prompt>
83
+ ASSISTANT:
84
+ ```
85
+
86
  ## How to easily download and use this model in text-generation-webui
87
 
88
  Please make sure you're using the latest version of text-generation-webui
 
128
 
129
  # Note: check the prompt template is correct for this model.
130
  prompt = "Tell me about AI"
131
+ prompt_template=f'''USER: {prompt}
132
+ ASSISTANT:'''
133
 
134
  print("\n\n*** Generate:")
135
 
 
168
  * Works with AutoGPTQ in CUDA or Triton modes.
169
  * Works with GPTQ-for-LLaMa in CUDA mode. May have issues with GPTQ-for-LLaMa Triton mode.
170
  * Works with text-generation-webui, including one-click-installers.
171
+ * Does not work with ExLlama, as it is not a Llama model.
172
  * Parameters: Groupsize = 128. Act Order / desc_act = False.
173
 
174
  <!-- footer start -->