cbh123/dylan-lyrics – Run with an API on Replicate

cbh123 / dylan-lyrics

Llama 2 13B fine-tuned on Bob Dylan lyrics

Cold

Public
260 runs
L40S

Run with an API

Playground API Examples README Versions

Input

prompt

*string

Shift + Return to add a new line

Prompt to send to the model.

max_new_tokens

integer

(minimum: 1)

Maximum number of tokens to generate. A word is generally 2-3 tokens

Default: 128

min_new_tokens

integer

(minimum: -1)

Minimum number of tokens to generate. To disable, set to -1. A word is generally 2-3 tokens.

Default: -1

temperature

number

(minimum: 0.01, maximum: 5)

Adjusts randomness of outputs, greater than 1 is random and 0 is deterministic, 0.75 is a good starting value.

Default: 0.75

top_p

number

(minimum: 0, maximum: 1)

When decoding text, samples from the top p percentage of most likely tokens; lower to ignore less likely tokens

Default: 0.9

top_k

integer

(minimum: 0)

When decoding text, samples from the top k most likely tokens; lower to ignore less likely tokens

Default: 50

stop_sequences

string

Shift + Return to add a new line

A comma-separated list of sequences to stop generation at. For example, '<end>,<stop>' will stop generation at the first instance of 'end' or '<stop>'.

seed

integer

Random seed. Leave blank to randomize the seed

debug

boolean

provide debugging output in logs

Default: false

Run this model in Node.js with one line of code:

npx create-replicate --model=cbh123/dylan-lyrics

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run cbh123/dylan-lyrics using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "cbh123/dylan-lyrics:d0fd41245648687f44510e24c59071dc3c2ea4c9f7c853a765065a6c2bec23e9",
  {
    input: {
      debug: false,
      top_k: 200,
      top_p: 0.8,
      prompt: "<title>GPU Song</title> <lyrics>",
      temperature: 0.75,
      max_new_tokens: 1000,
      min_new_tokens: -1
    }
  }
);
console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run cbh123/dylan-lyrics using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "cbh123/dylan-lyrics:d0fd41245648687f44510e24c59071dc3c2ea4c9f7c853a765065a6c2bec23e9",
    input={
        "debug": False,
        "top_k": 200,
        "top_p": 0.8,
        "prompt": "<title>GPU Song</title> <lyrics>",
        "temperature": 0.75,
        "max_new_tokens": 1000,
        "min_new_tokens": -1
    }
)

# The cbh123/dylan-lyrics model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/cbh123/dylan-lyrics/api#output-schema
    print(item, end="")

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run cbh123/dylan-lyrics using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "d0fd41245648687f44510e24c59071dc3c2ea4c9f7c853a765065a6c2bec23e9",
    "input": {
      "debug": false,
      "top_k": 200,
      "top_p": 0.8,
      "prompt": "<title>GPU Song</title> <lyrics>",
      "temperature": 0.75,
      "max_new_tokens": 1000,
      "min_new_tokens": -1
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

<line>I'm a song for my GPU, Ain't got much to say. My memory is made of silicon and copper, And I play games all day.</line> <line>You know it ain't easy being me, With no way out but the wires, But when you see me in action, Man, I can sure do some fires!</line> <line>Oh, I've been through some rough times, Like the time that I overheated, Or when I had to run on just 8 cores, Those were the days.</line> <line>Now I've got two gigs of VRAM, And I'm running at full speed, The drivers have finally sorted themselves out, What more could a boy need?</line> <line>Sometimes I feel like an ant under glass, That people look down upon, They don't appreciate how hard I work, Even though they use me every day.</line> <line>When you think about it, It's amazing what we can do with these things nowadays, We can make them so fast, That they almost seem human sometimes.</line> <line>Well, there's one thing I can tell ya, If you want to be happy, You better get yourself a good graphics card, Cause life's too short to play games on bad hardware.</line>

{
  "completed_at": "2023-08-17T19:14:50.081574Z",
  "created_at": "2023-08-17T19:14:34.196334Z",
  "data_removed": false,
  "error": null,
  "id": "5zkqce3btzhf7s36sv6pmjse4q",
  "input": {
    "debug": false,
    "top_k": 200,
    "top_p": 0.8,
    "prompt": "<title>GPU Song</title> <lyrics>",
    "temperature": 0.75,
    "system_prompt": "You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.\n\nIf a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.",
    "max_new_tokens": 1000,
    "min_new_tokens": -1,
    "repetition_penalty": 1.15,
    "repetition_penalty_sustain": 256,
    "token_repetition_penalty_decay": 128
  },
  "logs": null,
  "metrics": {
    "predict_time": 15.852465,
    "total_time": 15.88524
  },
  "output": [
    "<line>I'm",
    " a",
    " song",
    " for",
    " my",
    " GPU,\nAin't",
    " got",
    " much",
    " to",
    " say.\nMy",
    " memory",
    " is",
    " made",
    " of",
    " silicon",
    " and",
    " copper,\nAnd",
    " I",
    " play",
    " games",
    " all",
    " day.</line>",
    " <line>You",
    " know",
    " it",
    " ain't",
    " easy",
    " being",
    " me,\nWith",
    " no",
    " way",
    " out",
    " but",
    " the",
    " wires,\nBut",
    " when",
    " you",
    " see",
    " me",
    " in",
    " action,\nMan,",
    " I",
    " can",
    " sure",
    " do",
    " some",
    " fires!</line>",
    " <line>Oh,",
    " I've",
    " been",
    " through",
    " some",
    " rough",
    " times,\nLike",
    " the",
    " time",
    " that",
    " I",
    " overheated,\nOr",
    " when",
    " I",
    " had",
    " to",
    " run",
    " on",
    " just",
    " 8",
    " cores,\nThose",
    " were",
    " the",
    " days.</line>",
    " <line>Now",
    " I've",
    " got",
    " two",
    " gigs",
    " of",
    " VRAM,\nAnd",
    " I'm",
    " running",
    " at",
    " full",
    " speed,\nThe",
    " drivers",
    " have",
    " finally",
    " sorted",
    " themselves",
    " out,\nWhat",
    " more",
    " could",
    " a",
    " boy",
    " need?</line>",
    " <line>Sometimes",
    " I",
    " feel",
    " like",
    " an",
    " ant",
    " under",
    " glass,\nThat",
    " people",
    " look",
    " down",
    " upon,\nThey",
    " don't",
    " appreciate",
    " how",
    " hard",
    " I",
    " work,\nEven",
    " though",
    " they",
    " use",
    " me",
    " every",
    " day.</line>",
    " <line>When",
    " you",
    " think",
    " about",
    " it,\nIt's",
    " amazing",
    " what",
    " we",
    " can",
    " do",
    " with",
    " these",
    " things",
    " nowadays,\nWe",
    " can",
    " make",
    " them",
    " so",
    " fast,\nThat",
    " they",
    " almost",
    " seem",
    " human",
    " sometimes.</line>",
    " <line>Well,",
    " there's",
    " one",
    " thing",
    " I",
    " can",
    " tell",
    " ya,\nIf",
    " you",
    " want",
    " to",
    " be",
    " happy,\nYou",
    " better",
    " get",
    " yourself",
    " a",
    " good",
    " graphics",
    " card,\nCause",
    " life's",
    " too",
    " short",
    " to",
    " play",
    " games",
    " on",
    " bad",
    " hardware.</line>"
  ],
  "started_at": "2023-08-17T19:14:34.229109Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/5zkqce3btzhf7s36sv6pmjse4q",
    "cancel": "https://api.replicate.com/v1/predictions/5zkqce3btzhf7s36sv6pmjse4q/cancel"
  },
  "version": "a353fc0ad1d4c76e92a1e2875d27f4880754e0296b8b1a5ccfec3a053e93925d"
}

Generated in

15.9 seconds

Tweak it Report

This example was created by a different version, cbh123/dylan-lyrics:a353fc0a.

Run time and cost

This model costs approximately $0.019 to run on Replicate, or 52 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 20 seconds.

Readme

Llama 2 13B (base) fine-tuned on Bob Dylan Lyrics. Trained on a dataset that looks like:

{"text": "<title>{song_title}</title> <lyrics>{song_lyrics}</lyrics>"}
...

To prompt the model, try using a similar format:

<title>Tale of the Roving Gambler</title><lyrics>

The model should then respond with lyrics to that song.

Dataset

GitHub gist

Tips

Try playing around with temperature and top_k, and setting top_p to 0.8