INDEX

Explanations

well-known entities or individuals

oai_token-act-pair · gpt-3.5-turbo

references to well-known individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

plet

-1.09

ossession

-0.90

onew

-0.82

otos

-0.79

alos

-0.78

opes

-0.76

ikarp

-0.75

regate

-0.75

cair

-0.75

onds

-0.73

POSITIVE LOGITS

itarian

0.74

 comedic

0.73

ties

0.72

 landmarks

0.70

stood

0.67

 obscure

0.66

 celebrities

0.66

 progressive

0.65

 brands

0.64

tale

0.64

Activations Density 0.041%

well-known entities or individuals

references to well-known individuals or entities

No Comments

No Known Activations