INDEX

Explanations

references to Hollywood

oai_token-act-pair · gpt-3.5-turbo

mentions of Hollywood and its related figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

İĭ

-0.85

arity

-0.84

tnc

-0.78

avez

-0.77

cific

-0.76

uter

-0.74

unin

-0.74

rack

-0.73

tf

-0.70

gs

-0.69

POSITIVE LOGITS

 Reporter

1.20

 movies

1.03

 blockbuster

1.01

 studios

1.01

 Studios

1.01

 Hollywood

0.99

 Hills

0.98

 celebrities

0.98

 films

0.96

 Pictures

0.92

Activations Density 0.037%

references to Hollywood

mentions of Hollywood and its related figures

No Comments

No Known Activations