INDEX

Explanations

names of cities or sports teams

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to sports teams and cities

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

schild

-0.91

redd

-0.88

enegger

-0.86

wcsstore

-0.72

hig

-0.72

nces

-0.71

eport

-0.70

milo

-0.70

BOOK

-0.70

clinton

-0.69

POSITIVE LOGITS

 Eagles

1.16

 teammate

1.14

 Chiefs

1.12

 Jets

1.09

 Falcons

1.05

 Giants

1.05

 Wolves

1.04

 Cardinals

1.03

 Royals

1.03

 rookie

1.00

Activations Density 0.107%

names of cities or sports teams

proper nouns related to sports teams and cities

No Comments

No Known Activations