INDEX

Explanations

phrases related to being ranked or rated

oai_token-act-pair · gpt-3.5-turbo

instances of ranking or scoring systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Tanz

-0.64

Materials

-0.58

endish

-0.57

ETHOD

-0.56

 Disabled

-0.55

imar

-0.55

endon

-0.55

REM

-0.54

 Howell

-0.53

IPS

-0.53

POSITIVE LOGITS

plates

0.99

 crunch

0.92

one

0.91

one

0.89

 Ones

0.89

 thirteen

0.88

 eleven

0.87

less

0.87

two

0.87

eight

0.86

Activations Density 0.053%

phrases related to being ranked or rated

instances of ranking or scoring systems

No Comments

No Known Activations