INDEX

Explanations

the word "star" in the text

oai_token-act-pair · gpt-3.5-turbo

ratings associated with quality or excellence

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.87

 Canaver

-0.85

pse

-0.80

odcast

-0.72

£ı

-0.72

hower

-0.71

Cue

-0.69

 timelines

-0.68

nown

-0.67

utenberg

-0.67

POSITIVE LOGITS

vation

1.06

ved

0.99

bucks

0.85

ving

0.85

star

0.84

rating

0.84

ring

0.82

red

0.81

rated

0.79

iated

0.79

Activations Density 0.016%

the word "star" in the text

ratings associated with quality or excellence

No Comments

No Known Activations