INDEX

Explanations

mentions of the name "Joe"

oai_token-act-pair · gpt-3.5-turbo

instances of the substring "oe" in words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ifiers

-0.82

 glim

-0.81

ifier

-0.76

ifiable

-0.75

ãĥĥãĤ¯

-0.67

 Chandra

-0.65

 Sabha

-0.65

rons

-0.64

 à¨

-0.63

 Inquis

-0.62

POSITIVE LOGITS

cean

1.01

zzi

1.00

ppel

0.99

zie

0.98

oe

0.97

lect

0.95

ze

0.88

cker

0.88

hler

0.86

hl

0.85

Activations Density 0.012%

mentions of the name "Joe"

instances of the substring "oe" in words

No Comments

No Known Activations