Quantization made by Richard Erkhov.

Bloom-1b7-dialogsum - GGUF

Model creator: https://huggingface.co/alonzogarbanzo/
Original model: https://huggingface.co/alonzogarbanzo/Bloom-1b7-dialogsum/

Name	Quant method	Size
Bloom-1b7-dialogsum.Q2_K.gguf	Q2_K	0.98GB
Bloom-1b7-dialogsum.Q3_K_S.gguf	Q3_K_S	1.1GB
Bloom-1b7-dialogsum.Q3_K.gguf	Q3_K	1.2GB
Bloom-1b7-dialogsum.Q3_K_M.gguf	Q3_K_M	1.2GB
Bloom-1b7-dialogsum.Q3_K_L.gguf	Q3_K_L	1.25GB
Bloom-1b7-dialogsum.IQ4_XS.gguf	IQ4_XS	1.27GB
Bloom-1b7-dialogsum.Q4_0.gguf	Q4_0	1.31GB
Bloom-1b7-dialogsum.IQ4_NL.gguf	IQ4_NL	1.31GB
Bloom-1b7-dialogsum.Q4_K_S.gguf	Q4_K_S	1.31GB
Bloom-1b7-dialogsum.Q4_K.gguf	Q4_K	1.39GB
Bloom-1b7-dialogsum.Q4_K_M.gguf	Q4_K_M	1.39GB
Bloom-1b7-dialogsum.Q4_1.gguf	Q4_1	1.41GB
Bloom-1b7-dialogsum.Q5_0.gguf	Q5_0	1.51GB
Bloom-1b7-dialogsum.Q5_K_S.gguf	Q5_K_S	1.51GB
Bloom-1b7-dialogsum.Q5_K.gguf	Q5_K	1.57GB
Bloom-1b7-dialogsum.Q5_K_M.gguf	Q5_K_M	1.57GB
Bloom-1b7-dialogsum.Q5_1.gguf	Q5_1	1.61GB
Bloom-1b7-dialogsum.Q6_K.gguf	Q6_K	1.72GB
Bloom-1b7-dialogsum.Q8_0.gguf	Q8_0	2.23GB

Original model description:

license: bigscience-bloom-rail-1.0 base_model: bigscience/bloom-1b7 tags: - generated_from_trainer model-index: - name: Bloom-1b7-dialogsum results: []

Bloom-1b7-dialogsum

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

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: 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: 5
mixed_precision_training: Native AMP

Training results

Final epoch results: {'loss': 0.024, 'learning_rate': 1.4000000000000001e-06, 'epoch': 5.0}

After finished: {'train_runtime': 582.2106, 'train_samples_per_second': 1.718, 'train_steps_per_second': 0.429, 'train_loss': 0.72078223118186, 'epoch': 5.0}

Framework versions

Transformers 4.37.2
Pytorch 2.2.0+cu121
Datasets 2.17.0
Tokenizers 0.15.2