INDEX

Explanations

mentions of the word "nose"

oai_token-act-pair · gpt-3.5-turbo

words related to conditions or attributes associated with excess

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

ersen

-0.79

icum

-0.72

angible

-0.71

ufact

-0.71

azon

-0.71

agna

-0.69

rators

-0.69

oola

-0.69

cdn

-0.67

enberg

-0.66

POSITIVE LOGITS

velt

1.05

cond

1.03

idon

1.01

lihood

1.00

lect

0.94

LECT

0.85

eker

0.82

ï¸

0.79

OPLE

0.78

ols

0.77

Activations Density 0.022%

mentions of the word "nose"

words related to conditions or attributes associated with excess

No Comments

No Known Activations