INDEX

Explanations

names of locations or entities related to California

oai_token-act-pair · gpt-3.5-turbo

instances of the word "California."

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

htar

-0.97

mble

-0.79

chel

-0.77

Ò

-0.76

minded

-0.75

hern

-0.75

gey

-0.75

ahs

-0.74

ebin

-0.73

umbn

-0.73

POSITIVE LOGITS

 Edison

0.92

Oro

0.90

 Calif

0.86

 Springs

0.83

ISO

0.81

 Penal

0.76

 Superior

0.75

 ILCS

0.75

 California

0.74

 State

0.74

Activations Density 0.026%

names of locations or entities related to California

instances of the word "California."

No Comments

No Known Activations