INDEX

Explanations

mentions of people in politics or news

oai_token-act-pair · gpt-3.5-turbo

mentions of political figures and their statements or actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 metre

-0.61

 explorers

-0.61

 realised

-0.61

 Norn

-0.61

 endeavour

-0.58

 Layout

-0.57

 traveller

-0.57

 beginners

-0.57

 labou

-0.56

 kilomet

-0.56

POSITIVE LOGITS

 bipartisan

0.97

 Republican

0.95

 Hillary

0.92

 Politico

0.91

Democrats

0.89

Hillary

0.89

Republican

0.89

partisan

0.89

Republicans

0.88

Obama

0.88

Activations Density 4.180%

mentions of people in politics or news

mentions of political figures and their statements or actions

No Comments

No Known Activations