nateraw/samsum-llama-7b – Run with an API on Replicate

nateraw / samsum-llama-7b

llama-2-7b fine-tuned on the samsum dataset for dialogue summarization

Cold

Public
132 runs
L40S
GitHub
Paper
License

Run with an API

Playground API Examples README Versions

Input

prompt

*string

Shift + Return to add a new line

[INST] <<SYS>>
Use the Input to provide a summary of a conversation.
<</SYS>>

Input:
Harry: Who are you?
Hagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.
Harry: Sorry, no.
Hagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?
Harry: All what?
Hagrid: Yer a wizard, Harry.
Harry: I-- I'm a what?
Hagrid: A wizard! And a thumpin' good 'un, I'll wager, once you've been trained up a bit. [/INST]

Summary: [INST] <<SYS>>
Use the Input to provide a summary of a conversation.
<</SYS>>

Input:
Harry: Who are you?
Hagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.
Harry: Sorry, no.
Hagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?
Harry: All what?
Hagrid: Yer a wizard, Harry.
Harry: I-- I'm a what?
Hagrid: A wizard! And a thumpin' good 'un, I'll wager, once you've been trained up a bit. [/INST]

Summary: 

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=nateraw/samsum-llama-7b

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 nateraw/samsum-llama-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "nateraw/samsum-llama-7b:16665c1f00ad4d5d6c88393aa05390cf4d9f4e49c8abde4c58f2e1e71fd806f9",
  {
    input: {
      debug: false,
      top_k: 50,
      top_p: 0.95,
      prompt: "[INST] <<SYS>>\nUse the Input to provide a summary of a conversation.\n<</SYS>>\n\nInput:\nHarry: Who are you?\nHagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.\nHarry: Sorry, no.\nHagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?\nHarry: All what?\nHagrid: Yer a wizard, Harry.\nHarry: I-- I'm a what?\nHagrid: A wizard! And a thumpin' good 'un, I'll wager, once you've been trained up a bit. [/INST]\n\nSummary: ",
      temperature: 0.4,
      max_new_tokens: 256,
      min_new_tokens: -1,
      stop_sequences: "</s>"
    }
  }
);
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 nateraw/samsum-llama-7b using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "nateraw/samsum-llama-7b:16665c1f00ad4d5d6c88393aa05390cf4d9f4e49c8abde4c58f2e1e71fd806f9",
    input={
        "debug": False,
        "top_k": 50,
        "top_p": 0.95,
        "prompt": "[INST] <<SYS>>\nUse the Input to provide a summary of a conversation.\n<</SYS>>\n\nInput:\nHarry: Who are you?\nHagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.\nHarry: Sorry, no.\nHagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?\nHarry: All what?\nHagrid: Yer a wizard, Harry.\nHarry: I-- I'm a what?\nHagrid: A wizard! And a thumpin' good 'un, I'll wager, once you've been trained up a bit. [/INST]\n\nSummary: ",
        "temperature": 0.4,
        "max_new_tokens": 256,
        "min_new_tokens": -1,
        "stop_sequences": "</s>"
    }
)

# The nateraw/samsum-llama-7b 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/nateraw/samsum-llama-7b/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 nateraw/samsum-llama-7b 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": "16665c1f00ad4d5d6c88393aa05390cf4d9f4e49c8abde4c58f2e1e71fd806f9",
    "input": {
      "debug": false,
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "[INST] <<SYS>>\\nUse the Input to provide a summary of a conversation.\\n<</SYS>>\\n\\nInput:\\nHarry: Who are you?\\nHagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.\\nHarry: Sorry, no.\\nHagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?\\nHarry: All what?\\nHagrid: Yer a wizard, Harry.\\nHarry: I-- I\'m a what?\\nHagrid: A wizard! And a thumpin\' good \'un, I\'ll wager, once you\'ve been trained up a bit. [/INST]\\n\\nSummary: ",
      "temperature": 0.4,
      "max_new_tokens": 256,
      "min_new_tokens": -1,
      "stop_sequences": "</s>"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/nateraw/samsum-llama-7b@sha256:16665c1f00ad4d5d6c88393aa05390cf4d9f4e49c8abde4c58f2e1e71fd806f9 \
  -i 'debug=false' \
  -i 'top_k=50' \
  -i 'top_p=0.95' \
  -i $'prompt="[INST] <<SYS>>\\nUse the Input to provide a summary of a conversation.\\n<</SYS>>\\n\\nInput:\\nHarry: Who are you?\\nHagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.\\nHarry: Sorry, no.\\nHagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?\\nHarry: All what?\\nHagrid: Yer a wizard, Harry.\\nHarry: I-- I\'m a what?\\nHagrid: A wizard! And a thumpin\' good \'un, I\'ll wager, once you\'ve been trained up a bit. [/INST]\\n\\nSummary: "' \
  -i 'temperature=0.4' \
  -i 'max_new_tokens=256' \
  -i 'min_new_tokens=-1' \
  -i 'stop_sequences="</s>"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/nateraw/samsum-llama-7b@sha256:16665c1f00ad4d5d6c88393aa05390cf4d9f4e49c8abde4c58f2e1e71fd806f9
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "debug": false,
      "top_k": 50,
      "top_p": 0.95,
      "prompt": "[INST] <<SYS>>\\nUse the Input to provide a summary of a conversation.\\n<</SYS>>\\n\\nInput:\\nHarry: Who are you?\\nHagrid: Rubeus Hagrid, Keeper of Keys and Grounds at Hogwarts. Of course, you know all about Hogwarts.\\nHarry: Sorry, no.\\nHagrid: No? Blimey, Harry, did you never wonder where yer parents learned it all?\\nHarry: All what?\\nHagrid: Yer a wizard, Harry.\\nHarry: I-- I\'m a what?\\nHagrid: A wizard! And a thumpin\' good \'un, I\'ll wager, once you\'ve been trained up a bit. [/INST]\\n\\nSummary: ",
      "temperature": 0.4,
      "max_new_tokens": 256,
      "min_new_tokens": -1,
      "stop_sequences": "</s>"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

Harry is a wizard.

Generated in

0.7 seconds

Tweak it Report

Run time and cost

This model costs approximately $0.0013 to run on Replicate, or 769 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 2 seconds.

Readme

This model doesn't have a readme.