INDEX

Explanations

mentions of the name "Bon" with varying activations

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual named "Bon."

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

dfx

-0.83

 Ethics

-0.77

gamer

-0.68

æĸ¹

-0.68

 Marijuana

-0.68

INAL

-0.67

ELD

-0.65

 Editorial

-0.65

 Regulatory

-0.64

 Nicotine

-0.64

POSITIVE LOGITS

anza

1.11

iton

1.08

uses

1.04

Bon

0.99

gey

0.98

itors

0.97

illas

0.95

etooth

0.93

isson

0.92

vill

0.91

Activations Density 0.009%

mentions of the name "Bon" with varying activations

references to a specific individual named "Bon."

No Comments

No Known Activations