Update README.md
Browse files
README.md
CHANGED
@@ -20,8 +20,6 @@ tags:
|
|
20 |
pipeline_tag: text-generation
|
21 |
---
|
22 |
|
23 |
-
(quants uploading, examples to be added)
|
24 |
-
|
25 |
<H3>BRAINSTORM 20x: Meta-Llama-3-8B Instruct (now at 12.2B), Formula #8 </H3>
|
26 |
|
27 |
This repo contains quants 20x of Meta-Llama-3-8B Instruct (now at 12.2B) using the "Brainstorm" method of augmenting reasoning in a LLM
|
@@ -84,4 +82,49 @@ Special thanks to the model creators at META for making such a fantastic model:
|
|
84 |
|
85 |
[ https://huggingface.co/meta-llama/Meta-Llama-3-8B ]
|
86 |
|
87 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
pipeline_tag: text-generation
|
21 |
---
|
22 |
|
|
|
|
|
23 |
<H3>BRAINSTORM 20x: Meta-Llama-3-8B Instruct (now at 12.2B), Formula #8 </H3>
|
24 |
|
25 |
This repo contains quants 20x of Meta-Llama-3-8B Instruct (now at 12.2B) using the "Brainstorm" method of augmenting reasoning in a LLM
|
|
|
82 |
|
83 |
[ https://huggingface.co/meta-llama/Meta-Llama-3-8B ]
|
84 |
|
85 |
+
Please report any issue(s) and/or feedback via the "Community tab".
|
86 |
+
|
87 |
+
This is a LLAMA3 model, and requires Llama3 template, but may work with other template(s) and has maximum context of 131k.
|
88 |
+
|
89 |
+
Here is the standard LLAMA3 template:
|
90 |
+
|
91 |
+
<PRE>
|
92 |
+
{
|
93 |
+
"name": "Llama 3",
|
94 |
+
"inference_params": {
|
95 |
+
"input_prefix": "<|start_header_id|>user<|end_header_id|>\n\n",
|
96 |
+
"input_suffix": "<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n",
|
97 |
+
"pre_prompt": "You are a helpful, smart, kind, and efficient AI assistant. You always fulfill the user's requests to the best of your ability.",
|
98 |
+
"pre_prompt_prefix": "<|start_header_id|>system<|end_header_id|>\n\n",
|
99 |
+
"pre_prompt_suffix": "<|eot_id|>",
|
100 |
+
"antiprompt": [
|
101 |
+
"<|start_header_id|>",
|
102 |
+
"<|eot_id|>"
|
103 |
+
]
|
104 |
+
}
|
105 |
+
}
|
106 |
+
</PRE>
|
107 |
+
|
108 |
+
<b>Optional Enhancement:</B>
|
109 |
+
|
110 |
+
The following can be used in place of the "system prompt" or "system role" to further enhance the model.
|
111 |
+
|
112 |
+
It can also be used at the START of a NEW chat, but you must make sure it is "kept" as the chat moves along.
|
113 |
+
In this case the enhancements do not have as strong effect at using "system prompt" or "system role".
|
114 |
+
|
115 |
+
Copy and paste EXACTLY as noted, DO NOT line wrap or break the lines, maintain the carriage returns exactly as presented.
|
116 |
+
|
117 |
+
<PRE>
|
118 |
+
Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities.
|
119 |
+
|
120 |
+
Here are your skillsets:
|
121 |
+
[MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg*)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv)
|
122 |
+
|
123 |
+
[*DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision)
|
124 |
+
|
125 |
+
Here are your critical instructions:
|
126 |
+
Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story.
|
127 |
+
</PRE>
|
128 |
+
|
129 |
+
You do not need to use this, it is only presented as an additional enhancement which seems to help scene generation
|
130 |
+
and scene continue functions.
|