INDEX

Explanations

the word "big" with varying levels of emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

the word "big" in various contexts

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

manship

-0.75

llor

-0.73

anwhile

-0.73

DragonMagazine

-0.72

yrinth

-0.69

Reviewed

-0.69

chron

-0.68

 Reincarnated

-0.67

Dialogue

-0.67

Configuration

-0.67

POSITIVE LOGITS

oted

1.11

ol

1.06

gest

1.04

 chunk

0.98

gie

0.93

GER

0.92

wig

0.91

ger

0.90

 daddy

0.87

 bang

0.85

Activations Density 0.036%

the word "big" with varying levels of emphasis or importance

the word "big" in various contexts

No Comments

No Known Activations