@josephani's meta/llama-2-70b-chat

prompt: Can you write a poem about open source machine learning? Let's make it in the style of Martin Luther King.
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. If 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: 500
min_new_tokens: -1
temperature: 0.5
top_p: 1
top_k: 250
repetition_penalty: 1
repetition_penalty_sustain: 256
token_repetition_penalty_decay: 128
debug: false

{ "debug": false, "max_new_tokens": 500, "min_new_tokens": -1, "prompt": "Can you write a poem about open source machine learning? Let's make it in the style of Martin Luther King.", "repetition_penalty": 1, "repetition_penalty_sustain": 256, "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.", "temperature": 0.5, "token_repetition_penalty_decay": 128, "top_k": 250, "top_p": 1 }

const output = await replicate.run(
  "meta/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1",
  {
    input: {
      debug: false,
      max_new_tokens: 500,
      min_new_tokens: -1,
      prompt: "Can you write a poem about open source machine learning? Let's make it in the style of Martin Luther King.",
      repetition_penalty: 1,
      repetition_penalty_sustain: 256,
      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.",
      temperature: 0.5,
      token_repetition_penalty_decay: 128,
      top_k: 250,
      top_p: 1
    }
  }
);

console.log(output);

output = replicate.run(
    "meta/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1",
    input={
        "debug": False,
        "max_new_tokens": 500,
        "min_new_tokens": -1,
        "prompt": "Can you write a poem about open source machine learning? Let's make it in the style of Martin Luther King.",
        "repetition_penalty": 1,
        "repetition_penalty_sustain": 256,
        "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.",
        "temperature": 0.5,
        "token_repetition_penalty_decay": 128,
        "top_k": 250,
        "top_p": 1
    }
)

# The meta/llama-2-70b-chat 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/meta/llama-2-70b-chat/api#output-schema
    print(item, end="")

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "meta/llama-2-70b-chat:2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1",
    "input": {
      "debug": false,
      "max_new_tokens": 500,
      "min_new_tokens": -1,
      "prompt": "Can you write a poem about open source machine learning? Let\'s make it in the style of Martin Luther King.",
      "repetition_penalty": 1,
      "repetition_penalty_sustain": 256,
      "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.",
      "temperature": 0.5,
      "token_repetition_penalty_decay": 128,
      "top_k": 250,
      "top_p": 1
    }
  }' \
  https://api.replicate.com/v1/predictions

Sure, here's a poem about open source machine learning in the style of Martin Luther King: In the land of technology, a movement brews, A call to open source, for all to use, Machine learning, once a closed-off art, Now free for all, a new depart. The code, once locked away, now free, A gift to all, for all to see, The power of AI, for everyone, To learn, to grow, to be the sun. The community, diverse and strong, Works together, right where they belong, Sharing knowledge, skills and time, A bond that's strong, a bond that's fine. Together they strive, for a world that's fair, Where AI is used, to show they care, For justice, equality and rights, A future bright, with open sight. The open source way, a path they choose, To make a difference, they don't refuse, For the greater good, they toil and strive, A movement strong, they will survive. With each line of code, a door is opened, A chance for all, to be empowered, To learn, to grow, to be the change, A world where AI, is a human range. So let us join, this open source band, And work together, hand in hand, For a brighter future, we'll stand, With open source machine learning, we'll take a stand.

{ "id": "7omksadbmcnqkj6gtsyczhjhqq", "model": "meta/llama-2-70b-chat", "version": "2c1608e18606fad2812020dc541930f2d0495ce32eee50074220b87300bc16e1", "input": { "debug": false, "max_new_tokens": 500, "min_new_tokens": -1, "prompt": "Can you write a poem about open source machine learning? Let's make it in the style of Martin Luther King.", "repetition_penalty": 1, "repetition_penalty_sustain": 256, "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.", "temperature": 0.5, "token_repetition_penalty_decay": 128, "top_k": 250, "top_p": 1 }, "logs": "** Speed: 11.55 tokens/second", "output": [ "", " Sure", ",", " here", "'", "s", " a", " poem", " about", " open", " source", " machine", " learning", " in", " the", " style", " of", " Martin", " Luther", " King", ":", "\n", "\n", "In", " the", " land", " of", " technology", ",", " a", " movement", " bre", "ws", ",", "\n", "A", " call", " to", " open", " source", ",", " for", " all", " to", " use", ",", "\n", "Machine", " learning", ",", " once", " a", " closed", "-", "off", " art", ",", "\n", "Now", " free", " for", " all", ",", " a", " new", " depart", ".", "\n", "\n", "The", " code", ",", " once", " locked", " away", ",", " now", " free", ",", "\n", "A", " gift", " to", " all", ",", " for", " all", " to", " see", ",", "\n", "The", " power", " of", " A", "I", ",", " for", " everyone", ",", "\n", "To", " learn", ",", " to", " grow", ",", " to", " be", " the", " sun", ".", "\n", "\n", "The", " community", ",", " diverse", " and", " strong", ",", "\n", "Work", "s", " together", ",", " right", " where", " they", " belong", ",", "\n", "Sh", "aring", " knowledge", ",", " skills", " and", " time", ",", "\n", "A", " bond", " that", "'", "s", " strong", ",", " a", " bond", " that", "'", "s", " fine", ".", "\n", "\n", "T", "ogether", " they", " st", "rive", ",", " for", " a", " world", " that", "'", "s", " fair", ",", "\n", "Where", " A", "I", " is", " used", ",", " to", " show", " they", " care", ",", "\n", "For", " justice", ",", " equality", " and", " rights", ",", "\n", "A", " future", " bright", ",", " with", " open", " sight", ".", "\n", "\n", "The", " open", " source", " way", ",", " a", " path", " they", " choose", ",", "\n", "To", " make", " a", " difference", ",", " they", " don", "'", "t", " refuse", ",", "\n", "For", " the", " greater", " good", ",", " they", " to", "il", " and", " st", "rive", ",", "\n", "A", " movement", " strong", ",", " they", " will", " surv", "ive", ".", "\n", "\n", "With", " each", " line", " of", " code", ",", " a", " door", " is", " opened", ",", "\n", "A", " chance", " for", " all", ",", " to", " be", " emp", "ower", "ed", ",", "\n", "To", " learn", ",", " to", " grow", ",", " to", " be", " the", " change", ",", "\n", "A", " world", " where", " A", "I", ",", " is", " a", " human", " range", ".", "\n", "\n", "So", " let", " us", " join", ",", " this", " open", " source", " band", ",", "\n", "And", " work", " together", ",", " hand", " in", " hand", ",", "\n", "For", " a", " br", "ighter", " future", ",", " we", "'", "ll", " stand", ",", "\n", "With", " open", " source", " machine", " learning", ",", " we", "'", "ll", " take", " a", " stand", "." ], "data_removed": false, "error": null, "source": "web", "status": "succeeded", "created_at": "2023-08-11T15:25:09.514281Z", "started_at": "2023-08-11T15:25:16.774265Z", "completed_at": "2023-08-11T15:25:46.486576Z", "urls": { "cancel": "https://api.replicate.com/v1/predictions/7omksadbmcnqkj6gtsyczhjhqq/cancel", "get": "https://api.replicate.com/v1/predictions/7omksadbmcnqkj6gtsyczhjhqq" }, "metrics": { "predict_time": 29.712311, "total_time": 36.972295 } }

Prediction

Input

Output