This model is based on the GPT-2 architecture and was further pre-trained (domain-adapted) using the text in Ohsumed 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	97	epoch-1	step-97
5	489	epoch-5	step-489
10	978	epoch-10	step-978
20	1956	epoch-20	step-1956
40	3913	epoch-40	step-3913
60	5870	epoch-60	step-5870
80	7826	epoch-80	step-7826
100	9783	epoch-100	step-9783
120	11740	epoch-120	step-11740
140	13696	epoch-140	step-13696
160	15653	epoch-160	step-15653
180	17610	epoch-180	step-17610
198	19400	epoch-198	step-19400

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

python

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 Ohsumed 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: 35 h
Cluster Provider: Artemisa
Compute Region: EU
Carbon Emitted: 5.42 kg CO2 eq.

Citation

BibTeX:

[More Information Needed]

cglez/gpt2-dapt-ohsumed

Model Card: GPT-2-DAPT-Ohsumed

Model Details

Description