INDEX

Explanations

controversial topics and statements

oai_token-act-pair · gpt-3.5-turbo

references to controversial topics and issues

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

elsen

-0.92

á

-0.87

©¶æ

-0.81

abetic

-0.79

vation

-0.79

ĺħ

-0.71

abetes

-0.70

abet

-0.69

strings

-0.68

ruary

-0.67

POSITIVE LOGITS

ity

0.91

 aspects

0.89

 topics

0.87

 fringe

0.81

 controversial

0.81

 topic

0.80

 viewpoints

0.77

 views

0.76

 opinions

0.76

iating

0.76

Activations Density 0.074%

controversial topics and statements

references to controversial topics and issues

No Known Activations