dreamgen/opus-v1.2-llama-3-8b

Llama 3 DreamGen Opus

🚨 WARNING 🚨

This model has issues, please use the following preview models instead:

• New train on top of Llama 3 8B Base
• New train on top of Llama 3 8B Instruct

Make sure to read this discussion if the model won't stop generating output.

Models for (steerable) story-writing and role-playing.
All Opus V1 models, including quants.

Resources

• Opus V1 prompting guide with many (interactive) examples and prompts that you can copy.
• Google Colab for interactive role-play using opus-v1.2-7b.
• Python code to format the prompt correctly.
• Join the community on Discord to get early access to new models.

Prompting

The models use an extended version of ChatML.

<|im_start|>system
(Story description in the right format here)
(Typically consists of plot description, style description and characters)<|im_end|>
<|im_start|>user
(Your instruction on how the story should continue)<|im_end|>
<|im_start|>text names= Alice
(Continuation of the story from the Alice character)<|im_end|>
<|im_start|>text
(Continuation of the story from no character in particular (pure narration))<|im_end|>
<|im_start|>user
(Your instruction on how the story should continue)<|im_end|>
<|im_start|>text names= Bob
(Continuation of the story from the Bob character)<|im_end|>

The Opus V1 extension is the addition of the text role, and the addition / modification of role names.

Pay attention to the following:

• The text messages can (but don't have to have) names, names are used to indicate the "active" character during role-play.
• There can be multiple subsequent message with a text role, especially if names are involved.
• There can be multiple names attached to a message.
• The format for names is names= {{name[0]}}; {{name[1]}}, beware of the spaces after names= and after the ;. This spacing leads to most natural tokenization for the names.

While the main goal for the models is great story-writing and role-playing performance, the models are also capable of several writing related tasks as well as general assistance.

Here's how you can prompt the model for the following tasks

Steerable Story-writing and Role-playing:

• Story-writing prompt guide
• Role-playing prompt guide
• Input:
  • System prompt: You provide story / role-play description, which consists of:
    • Plot description
    • Style description
    • Characters and their descriptions
  • Conversation turns:
    • Text / message turn: This represents part of the story or role play
    • Instruction: This tells the model what should happen next
• Output: Continuation of the story / role-play.

Story plot summarization

• Prompt guide
• Input: A story, or a few chapters of a story.
• Output: A description of the story or chapters.

Examples:

Summary of short fiction story in 50 words

Malygris, a powerful magician, summons his long-dead lover Nylissa from the grave using necromancy. However, upon seeing her, he realizes that she is not the same person he once knew, leading him to question the nature of love, memory, and the power of magic.

Summary of short fiction story in 200 words

The story is a dark fantasy set in the city of Susran, where an ancient and powerful magician named Malygris dwells in a tower atop a conical hill. Malygris, who has mastered all forms of sorcery and necromancy, is plagued by a deep sense of ennui and melancholy. He seeks solace in the memory of Nylissa, a girl he once loved in his youth.

One day, Malygris decides to summon Nylissa's spirit from the grave using his necromantic powers. He performs the necessary rituals and incantations, and Nylissa appears before him as a ghostly apparition. However, as Malygris gazes upon her, he becomes increasingly disillusioned. He notices subtle changes in her appearance and behavior that do not match his memories of her. He begins to question whether this is truly the same Nylissa he once knew, or if it is just another illusion conjured by his magic.

As Malygris grapples with this uncertainty, he becomes increasingly despondent and disillusioned with life itself. He realizes that even his magic cannot bring back the past or make him truly happy. In the end, he dismisses Nylissa's spirit and returns to his lonely tower, resigned to his fate as an old and weary sorcerer.

Summary of CNN News article in 150 words

The article discusses the recent human deaths from diseases transmitted by animals, specifically Alaskapox and bubonic plague. Dr. Leana Wen, an emergency physician and former Baltimore health commissioner, explains the various ways household pets can transmit diseases to humans. These include fecal-oral transmission, scratches and bites, and vector-borne diseases like ticks, fleas, and mosquitoes. Wen emphasizes the importance of vaccinating pets, maintaining cleanliness, and taking commonsense precautions to reduce disease transmission. She also advises that individuals who are immunocompromised, babies, and pregnant women should take additional care when interacting with animals.

Summary of CNN News article in 50 words

A doctor discusses the risks of diseases transmitted from animals to humans, focusing on household pets like cats and dogs. They emphasize the importance of vaccination, hygiene, and regular veterinary care to prevent these illnesses.

Story character description

• Prompt guide
• Input: A story, or a few chapters of a story, set of characters.
• Output: A description of the characters.

Examples:

Description of the Malygris character from this story in 50 words

Malygris is an ancient and powerful magician, characterized by his long white beard, cold green eyes, and a violet robe. He is a master of necromancy and possesses a vast knowledge of forbidden magic. Despite his immense power, he is plagued by a deep sense of melancholy and ennui.

Description of the prince character from this story in 50 words

The prince is a devoted and determined individual, driven by his love for his late queen. He is described as a graceful youth in the beginning, but as he grows older, his dedication to the Pearl of Love transforms him into a grave and intent man. His sense of color and design evolves over time, reflecting his growing maturity and artistic vision.

Story style description

• Prompt guide
• Input: A story, or a few chapters of a story.
• Output: A description the style of the story.

Style description of this story

The story is written in a descriptive and evocative style, with a focus on the atmosphere and setting. The author uses vivid and detailed language to create a sense of place and time, and employs a variety of literary devices such as similes, metaphors, and allusions to enhance the mood and tone of the narrative. The sentence structure is varied, with a mix of short and long sentences that create a rhythmic flow to the prose.

Style description of this story

The writing style is descriptive and evocative, with a focus on the beauty and grandeur of the Pearl of Love. The author uses vivid imagery and sensory details to create a rich and immersive atmosphere. The tone is reverential and contemplative, reflecting the prince's deep love for his queen and his dedication to creating a lasting monument to her memory.

Story description to chapters

• Prompt guide
• Input: A brief plot description and the desired number of chapters.
• Output: A description for each chapter.

And more...

Sampling params

For story-writing and role-play, I recommend "Min P" based sampling with min_p in the range [0.01, 0.1] and with temperature in the range [0.5, 1.5], depending on your preferences. A good starting point would be min_p=0.1; temperature=0.8.

You may also benefit from setting presence, frequency and repetition penalties, especially at lower temperatures.

Dataset

The fine-tuning dataset consisted of ~100M tokens of steerable story-writing, role-playing, writing-assistant and general-assistant examples. Each example was up to 31000 tokens long.

All story-writing and role-playing examples were based on human-written text.

Running the model

The model is should be compatible with any software that supports the base model, but beware of prompting and tokenization.

I recommend using these model versions:

• 7B: no quant (opus-v1.2-7b)
• 34B: no quant (opus-v1-34b) or awq (opus-v1-34b-awq)
• 34B: no quant (opus-v1.2-70b) or awq (opus-v1.2-70b-awq)

Running on DreamGen.com (free)

You can run the models on dreamgen.com for free — you can use the built-in UI for story-writing & role-playing, or use the API.

Running Locally

• Make sure your prompt is as close as possible to the Opus V1
  • Regardless of which backend you use, it's important that you format your prompt well and that the tokenization works correctly.
  • Read the prompt guide
  • Read the prompt formatting code
  • Make sure <|im_start|> and <|im_end|> are tokenized correctly
• vLLM
  • Google Colab: This is a simple interactive Google Colab to do role-play with the 7B model, it should fit on the T4 GPU.
  • Code: This is simple script for interactive chat for one hard-coded scenario.
• SillyTavern
  • Official SillyTavern documentation for DreamGen -- applies to both the API an local models
  • SillyTavern (staging) comes with built-in DreamGen preset for RP
    • Other presets can be found here, v2 kindly provided by @MarinaraSpaghetti
    • Make sure to unselect Skip special tokens, otherwise it won't work
    • This is just an attempt at approximating the Opus V1 prompt, it won't be perfect
  • Character cards specifically rewritten for the built-in DreamGen preset:
    • Seraphina (based on the default Seraphina card)
    • Lara Lightland (based on the card by Deffcolony)
• LM Studio
  • Config
  • Just like ChatML, just changed "assistant" to "text" role.
  • There's a bug in LM Studio if you delete a message or click "Continue", see here for details.
• HuggingFace
  • Chat template
  • Just like ChatML, just changed "assistant" to "text" role.

Known Issues

• 34B repetition:
  • The 34B sometimes gets stuck repeating the same word, or synonyms. This seems to be a common problem across various Yi 34B fine-tunes.
• GGUF:
  • The tokenization might be messed up. Some users reported that <|im_start|> and <|im_end|> are tokenized as multiple tokens. Also llama.cpp may not tokenize correctly (the Yi tokenizer is subtly different from the Llama 2 tokenizer).

License

• This model is intended for personal use only, other use is not permitted.