INDEX

Explanations

names or identifiers of individuals or entities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "names."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OPLE

-0.75

Bed

-0.67

UTERS

-0.66

yrinth

-0.65

 Bulletin

-0.65

irth

-0.64

 Springs

-0.63

UGE

-0.62

IENT

-0.61

Returns

-0.60

POSITIVE LOGITS

paces

1.66

pace

1.20

paced

1.09

plates

1.06

hips

1.03

mith

0.97

peed

0.96

erver

0.94

ames

0.94

 aliases

0.89

Activations Density 0.027%

names or identifiers of individuals or entities

occurrences of the word "names."

No Comments

No Known Activations