INDEX

Explanations

the word "Not" indicating negation

oai_token-act-pair · gpt-3.5-turbo

the phrase "Not" followed by various contexts indicating exceptions or disclaimers

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

éĥ

-0.76

ç·

-0.76

éģ

-0.72

stakes

-0.71

kamp

-0.71

ãģ®ç

-0.70

æ©

-0.69

creen

-0.69

å¤§

-0.69

 Mehran

-0.67

POSITIVE LOGITS

epad

1.21

withstanding

1.16

orious

1.10

icably

1.07

 necessarily

1.04

eworthy

0.98

ices

0.93

ific

0.90

ifications

0.89

icia

0.86

Activations Density 0.063%

the word "Not" indicating negation

the phrase "Not" followed by various contexts indicating exceptions or disclaimers

No Comments

No Known Activations