INDEX

Explanations

places or entities with the abbreviation "WI" (likely referring to Wisconsin)

oai_token-act-pair · gpt-3.5-turbo

references to Wisconsin and other related abbreviations or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oons

-0.71

mast

-0.71

izers

-0.71

hill

-0.70

apy

-0.69

aney

-0.69

aic

-0.69

antine

-0.68

nets

-0.67

cake

-0.67

POSITIVE LOGITS

REL

1.22

ELD

1.10

WI

1.07

RECT

1.03

pered

0.94

RE

0.91

pering

0.89

merce

0.89

REP

0.88

REC

0.88

Activations Density 0.032%

places or entities with the abbreviation "WI" (likely referring to Wisconsin)

references to Wisconsin and other related abbreviations or identifiers

No Comments

No Known Activations