If you haven’t yet trained a model on Replicate, we recommend you read one of the following guides.

• Fine-tune a language model
• Fine-tune an image model
• Fine-tune SDXL with your own images

Pricing

Trainings for this model run on 8x Nvidia A40 (Large) GPU hardware, which costs $0.0058 per second.

Create a training

pip install replicate

import replicate

training = replicate.trainings.create(
  version="meta/llama-2-7b:77dde5d6c56598691b9008f7d123a18d98f40e4b4978f8a72215ebfc2553ddd8",
  input={
    # set your inputs here
  },
  destination=f"{username}/<destination-model-name>"
)

print(training)

curl -s -X POST \
-d '{"destination": "{username}/<destination-model-name>", "input": {}}' \
  -H "Authorization: Token $REPLICATE_API_TOKEN" \
  https://api.replicate.com/v1/models/meta/llama-2-7b/versions/77dde5d6c56598691b9008f7d123a18d98f40e4b4978f8a72215ebfc2553ddd8/trainings

{
  "id": "zz4ibbonubfz7carwiefibzgga",
  "version": "77dde5d6c56598691b9008f7d123a18d98f40e4b4978f8a72215ebfc2553ddd8",
  "status": "starting",
  "input": {
    "data": "..."
  },
  "output": null,
  "error": null,
  "logs": null,
  "started_at": null,
  "created_at": "2023-03-28T21:47:58.566434Z",
  "completed_at": null
}

Note that before you can create a training, you’ll need to create a model and use its name as the value for the destination field.

Training inputs

• train_data (required): URL to a file of training data where each row is a JSON record in the format {"text": ...} or {"prompt": ..., "completion": ...}. Must be JSONL.

• num_train_epochs (optional, default=3): Number of epochs (iterations over the entire training dataset) to train for.

• train_batch_size (optional, default=4): Global batch size. This specifies the batch size that will be used to calculate gradients. Optimal batch size is data dependent; larger sizes train faster but may cause OOMs. 8 often works well for this configuration of llama-2-7B.

• micro_batch_size (optional, default=4): Micro batch size. This specifies the on-device batch size, if this is less than train_batch_size, gradient accumulation will be activated.

• num_validation_samples (optional, default=50): Number of samples to use for validation. If run_validation is True and validation_data is not specified, this number of samples will be selected from the tail of the training data. If validation_data is specified, this number of samples will be selected from the head of the validation data, up to the size of the validation data.

• validation_data (optional): URL to a file of eval data where each row is a JSON record in the format {"text": ...} or {"prompt": ..., "completion": ...} or {"prompt": ..., "completion": ...}. Must be JSONL.

• validation_batch_size (optional, default=1): Batch size for evaluation. For small validation sets, you should use the default batch size of 1.

• run_validation (optional, default=True): Whether to run validation during training.

• validation_prompt (optional, default=None): If provided, this prompt will be used to generate a model response during each validation step. Must be a string formatted prompt. Note: this is not implemented for QLoRA training.

• learning_rate (optional, default=1e-4): Learning rate!

• pack_sequences (optional, default=False): If ‘True’, sequences will be packed into a single sequences up to a given length of chunk_size. This improves computational efficiency.

• wrap_packed_sequences (optional, default=False): If ‘pack_sequences’ is ‘True’, this will wrap packed sequences across examples, ensuring a constant sequence length but breaking prompt formatting.

• chunk_size (optional, default=2048): If ‘pack_sequences’ is ‘True’, this will chunk sequences into chunks of this size.

• peft_method (optional, default=’lora’): Training method to use. Currently, only ‘lora’ and ‘qlora’ are supported.

• seed (optional, default=42): Random seed to use for training.

• lora_rank (optional, default=8): Rank of the LoRA matrices.

• lora_alpha (optional, default=16): Alpha parameter for scaling LoRA weights; weights are scaled by alpha/rank

• lora_dropout (optional, default=0.05): Dropout for LoRA training.

Please see ai.meta.com/llama for more information about the model, licensing, and acceptable use.