INDEX

Explanations

phrases related to a specific tragic event or location

oai_token-act-pair · gpt-3.5-turbo

references to the location "Newtown."

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

oute

-0.67

 bruises

-0.62

 Majesty

-0.62

 Halls

-0.62

 noses

-0.62

 Chung

-0.61

 Vlad

-0.59

 Chou

-0.59

 personality

-0.59

loo

-0.59

POSITIVE LOGITS

ancial

0.90

cest

0.88

geon

0.82

geons

0.80

mares

0.79

enegger

0.78

acies

0.76

ifi

0.76

elling

0.75

nesday

0.74

Activations Density 0.044%

phrases related to a specific tragic event or location

references to the location "Newtown."

No Comments

No Known Activations