INDEX

Explanations

the word "beach" with varying degrees of relevance

oai_token-act-pair · gpt-3.5-turbo

references to beaches

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

ilyn

-0.73

ioch

-0.71

cius

-0.69

ativity

-0.69

æł

-0.66

Interstitial

-0.66

 fetal

-0.64

ÑĮ

-0.62

encers

-0.62

atures

-0.61

POSITIVE LOGITS

goers

1.06

front

1.03

 beach

1.02

side

1.00

 volleyball

0.99

 Beach

0.93

 beaches

0.92

walk

0.90

tub

0.83

 Splash

0.83

Activations Density 0.010%

the word "beach" with varying degrees of relevance

references to beaches

No Comments

No Known Activations