cglez/gpt2-trec

Model Card: GPT-2-TREC

An in-domain GPT-2, pre-trained from scratch on the TREC dataset text.

Model Details

Description

This model is based on the GPT-2 architecture and was pre-trained from scratch (in-domain) using the text in TREC dataset, excluding its test split.

• Developed by: Cesar Gonzalez-Gutierrez
• Funded by: ERC
• Architecture: GPT-2
• Language: English
• License: MIT
• Base model: GPT-2

Checkpoints

Intermediate checkpoints from the pre-training process are available and can be accessed using specific tags, which correspond to training epochs and steps:

Epoch	Step	Tags
1	51	epoch-1	step-51
5	255	epoch-5	step-255
10	511	epoch-10	step-511
20	1023	epoch-20	step-1023
40	2046	epoch-40	step-2046
60	3070	epoch-60	step-3070
80	4093	epoch-80	step-4093
100	5116	epoch-100	step-5116
120	6140	epoch-120	step-6140
140	7163	epoch-140	step-7163
160	8186	epoch-160	step-8186
180	9210	epoch-180	step-9210
199	10200	epoch-199	step-10200

To load a model from a specific intermediate checkpoint, use the revision parameter with the corresponding tag:

from transformers import AutoModelForCausalLM

model = AutoModelForMaskedLM.from_pretrained("<model-name>", revision="<checkpoint-tag>")

Sources

• Paper: [Information pending]

Training Details

For more details on the training procedure, please refer to the base model's documentation: Training procedure.

Training Data

All texts from TREC dataset, excluding the test partition.

Training Hyperparameters

• Precision: fp16
• Batch size: 8
• Gradient accumulation steps: 12

Uses

For typical use cases and limitations, please refer to the base model's guidance: Inteded uses & limitations.

Bias, Risks, and Limitations

This model inherits potential risks and limitations from the base model. Refer to: Limitations and bias.

Environmental Impact

• Hardware Type: NVIDIA A100 PCIE 40GB
• Runtime: 28.5
• Cluster Provider: Artemisa
• Compute Region: EU
• Carbon Emitted: 4.42 kg CO2 eq.

Citation

BibTeX:

[More Information Needed]