INDEX

Explanations

mentions of studios

oai_token-act-pair · gpt-3.5-turbo

mentions of specific studios, particularly Studio 9 and Studio 8

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

pend

-0.68

avez

-0.68

isms

-0.66

abad

-0.65

bh

-0.64

 refere

-0.63

 Adin

-0.63

Fas

-0.61

vati

-0.61

MpServer

-0.61

POSITIVE LOGITS

 studio

0.94

 studios

0.84

orpor

0.79

ctor

0.75

rats

0.74

venant

0.73

floor

0.72

hesda

0.71

mates

0.71

smanship

0.71

Activations Density 0.043%

mentions of studios

mentions of specific studios, particularly Studio 9 and Studio 8

No Comments

No Known Activations