Fine-tuned-AWD-LSTM-DOC (fin)

Training compute

5.2×10¹⁶ FLOP

Parameters

46M

Published

Nov 12, 2018

Fine-tuned-AWD-LSTM-DOC (fin) is an AI model developed by Samsung R&D Institute Russia (Russia), first published in November 2018. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 5.2×10¹⁶ FLOP of compute (estimation method: operation counting). The model has 46,000,000 parameters. It was trained on roughly 1M datapoints.

Access: Unreleased. Its weights are not openly released. It is built on top of AWD-LSTM-DOC (fin) (23M). The reference paper has 2 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Samsung R&D Institute Russia
Country of organization: Russia
Domain: Language
Task: Language modeling
Training compute: 5.2×10¹⁶ FLOP
Compute estimation method: Operation counting
Parameters: 46,000,000
Dataset size: 1M
Model accessibility: Unreleased
Open weights: No
Base model: AWD-LSTM-DOC (fin) (23M)
Citations: 2
Epoch confidence: Confident