INDEX

Explanations

names of individuals, especially Laurence

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals with the name "Laurence" and variations, as well as mentions of laureates and awards

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

icago

-0.79

DEM

-0.71

INESS

-0.70

WD

-0.69

ORT

-0.67

imation

-0.67

ractical

-0.65

 rink

-0.63

 interfering

-0.63

 Panic

-0.62

POSITIVE LOGITS

nce

1.29

nces

1.08

 Laure

0.92

 laure

0.89

ls

0.88

ance

0.85

nda

0.83

¨

0.83

ira

0.81

ates

0.79

Activations Density 0.038%

names of individuals, especially Laurence

references to specific individuals with the name "Laurence" and variations, as well as mentions of laureates and awards

No Comments

No Known Activations