INDEX

Explanations

names or parts of names related to individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

sburg

-0.82

wagen

-0.78

-0.77

sylvania

-0.75

=-=-=-=-

-0.75

oÄŁ

-0.67

eric

-0.66

edly

-0.65

eur

-0.63

er

-0.63

POSITIVE LOGITS

plin

1.03

unta

0.94

ihara

0.82

uchi

0.80

ihad

0.79

rahim

0.78

wered

0.77

linger

0.75

umen

0.75

combe

0.74

Activations Density 0.094%

names or parts of names related to individuals

proper nouns, particularly names

No Comments

No Known Activations