metadata

license: bigscience-bloom-rail-1.0
tags:
  - generated_from_trainer
datasets:
  - allenai/common_gen
widget:
  - text: |-
      <s>Generate a sentence with the following words:
      ski, mountain, skier
      Sentence:
  - text: |-
      <s>Generate a sentence with the following words:
      dog, bed, play
      Sentence:
inference:
  parameters:
    eos_token_id: 2
    max_length: 200
model-index:
  - name: bloom-560m-finetuned-common_gen
    results: []

bloom-560m-finetuned-common_gen

This model is a fine-tuned version of bigscience/bloom-560m on the common_gen dataset. It achieves the following results on the evaluation set:

Loss: 1.3937

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 1
eval_batch_size: 1
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: 2
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.4819	0.38	100	1.4582
1.2451	0.76	200	1.4247
1.1016	1.14	300	1.4382
0.9724	1.53	400	1.4207
0.9396	1.91	500	1.3949

Framework versions

Transformers 4.22.1
Pytorch 1.12.1+cu113
Datasets 2.5.1
Tokenizers 0.12.1