INDEX

Explanations

phrases involving specific names and controversial topics

oai_token-act-pair · gpt-3.5-turbo

references to significant societal issues or controversies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 eleph

-1.06

oun

-0.96

aditional

-0.94

ò

-0.93

 tremend

-0.91

 undermin

-0.85

 practition

-0.83

Þ

-0.83

 exting

-0.79

 pione

-0.79

POSITIVE LOGITS

Merit

0.75

FL

0.74

↵

0.72

FOX

0.67

sonian

0.67

Breaking

0.66

 VIDEOS

0.65

 Franch

0.65

Study

0.65

Activations Density 0.183%

phrases involving specific names and controversial topics

references to significant societal issues or controversies

No Comments

No Known Activations