Quantization made by Richard Erkhov.

gpt2-context_generator - GGUF

Model creator: https://huggingface.co/Isotonic/
Original model: https://huggingface.co/Isotonic/gpt2-context_generator/

Name	Quant method	Size
gpt2-context_generator.Q2_K.gguf	Q2_K	0.08GB
gpt2-context_generator.IQ3_XS.gguf	IQ3_XS	0.08GB
gpt2-context_generator.IQ3_S.gguf	IQ3_S	0.08GB
gpt2-context_generator.Q3_K_S.gguf	Q3_K_S	0.08GB
gpt2-context_generator.IQ3_M.gguf	IQ3_M	0.09GB
gpt2-context_generator.Q3_K.gguf	Q3_K	0.09GB
gpt2-context_generator.Q3_K_M.gguf	Q3_K_M	0.09GB
gpt2-context_generator.Q3_K_L.gguf	Q3_K_L	0.1GB
gpt2-context_generator.IQ4_XS.gguf	IQ4_XS	0.1GB
gpt2-context_generator.Q4_0.gguf	Q4_0	0.1GB
gpt2-context_generator.IQ4_NL.gguf	IQ4_NL	0.1GB
gpt2-context_generator.Q4_K_S.gguf	Q4_K_S	0.1GB
gpt2-context_generator.Q4_K.gguf	Q4_K	0.11GB
gpt2-context_generator.Q4_K_M.gguf	Q4_K_M	0.11GB
gpt2-context_generator.Q4_1.gguf	Q4_1	0.11GB
gpt2-context_generator.Q5_0.gguf	Q5_0	0.11GB
gpt2-context_generator.Q5_K_S.gguf	Q5_K_S	0.11GB
gpt2-context_generator.Q5_K.gguf	Q5_K	0.12GB
gpt2-context_generator.Q5_K_M.gguf	Q5_K_M	0.12GB
gpt2-context_generator.Q5_1.gguf	Q5_1	0.12GB
gpt2-context_generator.Q6_K.gguf	Q6_K	0.13GB
gpt2-context_generator.Q8_0.gguf	Q8_0	0.17GB

Original model description:

language: - en license: cc-by-sa-4.0 tags: - generated_from_trainer - text-generation-inference datasets: - Non-Residual-Prompting/C2Gen pipeline_tag: text-generation base_model: gpt2 model-index: - name: gpt2-commongen-finetuned results: []

gpt2-context_generator

This model is a fine-tuned version of gpt2 on Non-Residual-Prompting/C2Gen dataset.

Model description

More information needed

Intended uses & limitations

Check config.json for prompt template and sampling strategy.

Dataset Summary

CommonGen Lin et al., 2020 is a dataset for the constrained text generation task of word inclusion. But the task does not allow to include context. Therefore, to complement CommonGen, we provide an extended test set C2Gen Carlsson et al., 2022 where an additional context is provided for each set of target words. The task is therefore reformulated to both generate commonsensical text which include the given words, and also have the generated text adhere to the given context.

Training procedure

Causal Language Modelling

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 9e-05
train_batch_size: 32
eval_batch_size: 32
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.2
num_epochs: 8

Framework versions

Transformers 4.27.3
Pytorch 1.13.1+cu116
Datasets 2.13.1
Tokenizers 0.13.2