Edit model card

bart-large-instructiongen-w-inputs

Use this text2text model to find out what LLM instruction (and inputs if relevant) might have generated <arbitrary input text>!

This model is a fine-tuned version of facebook/bart-large on the pszemraj/fleece2instructions-inputs-alpaca-cleaned dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9302
  • Rouge1: 64.2236
  • Rouge2: 41.5632
  • Rougel: 60.5935
  • Rougelsum: 62.1285
  • Gen Len: 25.8938

example

api

Intended uses & limitations

This model is intended to be used to generate instructions from arbitrary text. You can then use these instructions + your data to fine-tune an LLM on instructions w.r.t. a specific domain. This model is primarily intended to enable low-resource domain adaptation, rather than "I want to generate even better prompts for the FLAN-V2 dataset!".

The fleece2instructions-inputs-alpaca-cleaned dataset, obtained from the alpaca-lora repo under the ODC-BY license, has been converted to a text2text format for use with language models. In this dataset, the original 'inputs' and 'instructions' columns are combined into a single 'instructions_inputs' column. To clearly separate the two types of content, each piece of text is prefixed with either an <instruction> or <inputs> token. These tokens not only facilitate model comprehension, but also allow for easy regex separation of model outputs during inference.

As such, users can expect the output of this model to be similarly structured with <instruction> and <inputs> tokens.

Training and evaluation data

Refer to the fleece2instructions-inputs-alpaca-cleaned dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 6e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • distributed_type: multi-GPU
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.03
  • num_epochs: 3.0

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum Gen Len
1.0145 1.0 1361 1.0460 62.8374 39.8538 59.2593 60.8095 25.2752
0.8796 2.0 2722 0.9289 63.7086 41.1315 60.1588 61.7145 25.7215
0.6943 3.0 4083 0.9302 64.2236 41.5632 60.5935 62.1285 25.8938
Downloads last month
4
Safetensors
Model size
406M params
Tensor type
F32
·

Dataset used to train pszemraj/bart-large-instructiongen-w-inputs