INDEX

Explanations

references to Osama bin Laden

oai_token-act-pair · gpt-3.5-turbo

references to Osama bin Laden

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hire

-0.73

WD

-0.71

eries

-0.70

âĢ¢âĢ¢

-0.68

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.68

 Drawn

-0.67

CAP

-0.66

place

-0.65

 Concord

-0.65

haw

-0.65

POSITIVE LOGITS

bin

1.22

 Osama

1.12

Bin

1.04

 rall

1.01

 Laden

0.96

 Hussein

0.88

abad

0.82

 Mubarak

0.81

atis

0.81

 Tayyip

0.80

Activations Density 0.004%

references to Osama bin Laden

references to Osama bin Laden

No Comments

No Known Activations