INDEX

Explanations

words related to specific names, particularly variations of the name "Lew"

oai_token-act-pair · gpt-3.5-turbo

references to the individual named Lew

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

ACTED

-0.93

 à¨

-0.78

CHR

-0.70

OPLE

-0.68

expr

-0.67

thood

-0.65

ĸļ

-0.65

NetMessage

-0.65

¥µ

-0.64

============

-0.63

POSITIVE LOGITS

andowski

1.46

insky

1.07

enstein

0.94

Lew

0.90

inski

0.89

etz

0.88

ington

0.88

lem

0.86

sey

0.86

Lew

0.85

Activations Density 0.021%

words related to specific names, particularly variations of the name "Lew"

references to the individual named Lew

No Comments

No Known Activations