INDEX

Explanations

phrases related to political discussions and debates, specifically mentioning both sides of an argument or issue

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a division between opposing sides in a debate or discussion

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

Not in Any Lists

No Comments

Negative Logits

orable

-0.70

chy

-0.66

nces

-0.65

IAN

-0.64

vity

-0.63

uala

-0.62

ndra

-0.61

itious

-0.61

rian

-0.59

jew

-0.59

POSITIVE LOGITS

 aisle

0.93

 spectrum

0.92

 equation

0.74

 Atlantic

0.72

 ledger

0.71

Atlantic

0.70

 divide

0.69

ticket

0.69

 fence

0.68

strate

0.68

Activations Density 0.089%

phrases related to political discussions and debates, specifically mentioning both sides of an argument or issue

phrases indicating a division between opposing sides in a debate or discussion

No Comments

No Known Activations