Bloom-1b7-ropes-IT-baseline

This model is a fine-tuned version of bigscience/bloom-1b7 on an unknown dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

Instruction Tuned on the ropes task here: https://huggingface.co./datasets/adambjorn/UnrelatedForgettingOverhead/viewer/ropes

Training procedure

Given a set of prompts:

prompts = [
        "Given the following background and situation, answer the question: ",
        "Based on the background information and the current situation, what is the answer to the question? ",
        "Considering the background and the described situation, provide an answer to this question: ",
    ]

Each example is concatenated with the prompt, background, situation, question and answer:

input_text = f"{prompt}Background: {background} Situation: {situation} Question: {question} Answer: {answer_text}."

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Final results: {'loss': 0.024, 'grad_norm': 1.3331243991851807, 'learning_rate': 8.000000000000001e-07, 'epoch': 10.0}

Average results: {'train_runtime': 862.219, 'train_samples_per_second': 2.32, 'train_steps_per_second': 0.58, 'train_loss': 0.4160269268453121, 'epoch': 10.0}

Framework versions

• Transformers 4.38.1
• Pytorch 2.2.0+cu121
• Datasets 2.17.0
• Tokenizers 0.15.2