INDEX

Explanations

phrases related to branding and logos

oai_token-act-pair · gpt-3.5-turbo

mentions of the brand "Amazon."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

APS

-0.72

 Normandy

-0.67

 Bayer

-0.64

 Scha

-0.61

 tides

-0.60

 Gavin

-0.59

 Schn

-0.59

TON

-0.58

Fuk

-0.58

 Bedford

-0.58

POSITIVE LOGITS

azon

1.41

daq

0.91

ciating

0.87

icut

0.82

ian

0.81

iary

0.78

icon

0.78

ipedia

0.76

arily

0.75

 corrid

0.75

Activations Density 0.005%

phrases related to branding and logos

mentions of the brand "Amazon."

No Comments

No Known Activations