Bloom-1b7-glue-mrpc-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 glue-mrpc task here: https://huggingface.co./datasets/adambjorn/UnrelatedForgettingOverhead/viewer/glue-mrpc

Training procedure

Given a set of prompts:

prompts = [
        "Determine if the following sentences are equivalent: Sentence 1: {sentence1} Sentence 2: {sentence2}. Answer: ",
        "Are these sentences saying the same thing? First: {sentence1} Second: {sentence2}. Response: ",
        "Check sentence equivalence: \"{sentence1}\" versus \"{sentence2}\". Result: ",
    ]

Concatenate the prompts, the two sentences and the label as so:

input_text = prompt.format(sentence1=sentence1, sentence2=sentence2)    
input_text += " " + responses[label]

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': 0.0949, 'grad_norm': 5.0146379470825195, 'learning_rate': 6.000000000000001e-07, 'epoch': 10.0}

Average results: {'train_runtime': 363.2148, 'train_samples_per_second': 5.506, 'train_steps_per_second': 1.377, 'train_loss': 0.4939311617612839, 'epoch': 10.0}

Framework versions

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