INDEX

Explanations

words related to geopolitics and specific geographical locations, specifically "Donbas."

oai_token-act-pair · gpt-3.5-turbo

repeated occurrences of the term "bas."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Oaks

-0.75

 hopeful

-0.62

 Destroyer

-0.62

raught

-0.61

 spect

-0.59

Aut

-0.57

 Loving

-0.57

 reson

-0.56

 capture

-0.56

 husband

-0.55

POSITIVE LOGITS

sembly

1.07

bas

1.06

ciating

0.94

etooth

0.93

thood

0.91

cius

0.89

ibi

0.87

chet

0.87

MpServer

0.87

cham

0.85

Activations Density 0.005%

words related to geopolitics and specific geographical locations, specifically "Donbas."

repeated occurrences of the term "bas."

No Comments

No Known Activations