Codesota · Models · GPT-2-Medium (prefix-tuning)OpenAI5 results · 1 benchmarks
Model card

GPT-2-Medium (prefix-tuning).

OpenAIunknown355M paramsTransformer1 current SOTA

GPT-2 Medium with prefix-tuning (0.1% params) on E2E NLG. Reported in HTLM paper (arxiv:2107.06955).

§ 02 · Benchmarks

Every benchmark GPT-2-Medium (prefix-tuning) has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01e2eComputer Vision · Optical Character Recognitioncider2.5%#1/92021-07-14source ↗
02e2eComputer Vision · Optical Character Recognitionrouge-l71.4%#2/92021-07-14source ↗
03e2eComputer Vision · Optical Character Recognitionbleu69.7%#4/92021-07-14source ↗
04e2eComputer Vision · Optical Character Recognitionnist8.8%#4/92021-07-14source ↗
05e2eComputer Vision · Optical Character Recognitionmeteor46.1%#4/92021-07-14source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Strengths by area

Where GPT-2-Medium (prefix-tuning) actually performs.

Computer Vision
1
benchmark
avg rank #3.0 · 1 SOTA
§ 04 · Papers

1 paper with results for GPT-2-Medium (prefix-tuning).

  1. 2021-07-14· Computer Vision· 5 results

    HTLM: Hyper-Text Pre-Training and Prompting of Language Models

§ 05 · Related models

Other OpenAI models scored on Codesota.

GPT-4o
Undisclosed params · 57 results
o3
19 results
GPT-3.5-turbo
17 results
o4-mini
16 results
Whisper Large v2
1.5B params · 14 results
GPT-4
13 results
GPT-4 Turbo
Undisclosed params · 13 results
o1
12 results
§ 06 · Sources & freshness

Where these numbers come from.

papers-with-code
5
results
5 of 5 rows marked verified.