INDEX

Explanations

proper nouns related to locations, specifically boulevards

oai_token-act-pair · gpt-3.5-turbo

mentions of specific streets or boulevards

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

ERAL

-0.86

PLIED

-0.84

atorial

-0.73

glers

-0.73

sonian

-0.72

umar

-0.72

eared

-0.72

berman

-0.70

omsky

-0.69

conom

-0.69

POSITIVE LOGITS

 Boulevard

0.92

 Blvd

0.77

ards

0.75

naire

0.74

 Corridor

0.74

 Courier

0.72

 Yards

0.71

 Extension

0.69

 Racer

0.69

 Sector

0.69

Activations Density 0.013%

proper nouns related to locations, specifically boulevards

mentions of specific streets or boulevards

No Comments

No Known Activations