INDEX

Explanations

mentions of political entities or figures

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "the" or related determiners in a political context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.80

smanship

-0.78

ciation

-0.78

ufact

-0.78

vity

-0.75

thood

-0.74

nil

-0.73

pection

-0.73

ocket

-0.72

brainer

-0.71

POSITIVE LOGITS

 Clintons

1.43

 likes

1.28

 neocons

1.14

ocrats

1.13

GOP

1.13

 Democratic

1.11

 Kremlin

1.09

 Conservatives

1.05

 Republican

1.05

Tea

1.05

Activations Density 0.295%

mentions of political entities or figures

references to the concept of "the" or related determiners in a political context

No Comments

No Known Activations