INDEX

Explanations

quotations within text

oai_token-act-pair · gpt-3.5-turbo

quotations or dialogue in the text

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

 encl

-0.76

scan

-0.73

 Patriarch

-0.69

 appar

-0.67

 automation

-0.66

 popul

-0.66

 outsourcing

-0.64

govtrack

-0.64

å¥³

-0.64

 privat

-0.64

POSITIVE LOGITS

BuyableInstoreAndOnline

1.10

Yeah

1.01

Yes

0.92

Coach

0.92

Everybody

0.91

We

0.90

Players

0.89

Whoever

0.88

ussions

0.85

ESPN

0.81

Activations Density 0.103%

quotations within text

quotations or dialogue in the text

No Comments

No Known Activations