INDEX

Explanations

phrases related to professional services and business operations

oai_token-act-pair · gpt-3.5-turbo

phrases or elements related to authority or reverence

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

Ļ½

-0.84

ļé

-0.80

ģĸ

-0.75

ĨĴ

-0.66

TAIN

-0.66

©¶æ

-0.65

ury

-0.64

³

-0.64

yon

-0.64

®

-0.63

POSITIVE LOGITS

iest

0.64

 logo

0.64

 whence

0.63

liest

0.59

runners

0.59

 hierarchy

0.59

 which

0.59

 moniker

0.58

hest

0.58

 itself

0.58

Activations Density 0.897%

phrases related to professional services and business operations

phrases or elements related to authority or reverence

No Comments

No Known Activations