INDEX

Explanations

copyright information at the bottom of images

oai_token-act-pair · gpt-3.5-turbo

references to copyright

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

oven

-0.84

rency

-0.80

nesia

-0.78

ĪĴ

-0.78

cius

-0.69

OHN

-0.67

aru

-0.67

eday

-0.65

atos

-0.65

APTER

-0.64

POSITIVE LOGITS

 Getty

0.87

SPL

0.85

AFP

0.84

 Provided

0.77

EPA

0.77

 Â©

0.76

 Courtesy

0.74

 Image

0.73

 IMAGES

0.73

 Misty

0.70

Activations Density 0.009%

copyright information at the bottom of images

references to copyright

No Comments

No Known Activations