INDEX

Explanations

mentions of specific groups or entities and their interactions

oai_token-act-pair · gpt-3.5-turbo

references to the American public and political context

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

tackle

-0.75

blem

-0.74

escription

-0.73

thood

-0.72

atto

-0.68

vari

-0.66

nect

-0.65

endix

-0.65

lia

-0.65

Episode

-0.64

POSITIVE LOGITS

 populace

1.18

 fanbase

1.12

 likes

1.09

 masses

1.08

 locals

0.99

 mainstream

0.98

 detractors

0.98

 tabl

0.98

 electorate

0.93

 skeptics

0.91

Activations Density 0.285%

mentions of specific groups or entities and their interactions

references to the American public and political context

No Comments

No Known Activations