INDEX

Explanations

mentions of the word "ON", potentially associated with technical contexts or instructions

oai_token-act-pair · gpt-3.5-turbo

the word "ON" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ãĤ±

-0.73

ãĥ¼ãĥ³

-0.71

ãĤ§

-0.68

ãĤ¶

-0.68

Pwr

-0.66

utenberg

-0.66

ÄŁ

-0.66

lished

-0.65

utters

-0.65

ãĥĥ

-0.64

POSITIVE LOGITS

LY

1.20

etheless

1.05

LINE

1.03

ON

1.02

AUT

0.97

IUM

0.90

CE

0.88

ISH

0.87

GS

0.87

GO

0.87

Activations Density 0.009%

mentions of the word "ON", potentially associated with technical contexts or instructions

the word "ON" in different contexts

No Comments

No Known Activations