INDEX

Explanations

phrases containing the word "so"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

theless

-0.77

NTS

-0.68

 Mens

-0.62

 royalty

-0.57

 crim

-0.56

Mog

-0.56

FAA

-0.56

 Mock

-0.55

 Customs

-0.55

 Alpine

-0.55

POSITIVE LOGITS

apy

1.35

oths

1.31

iled

1.18

aps

1.12

bered

1.11

aring

1.05

vi

1.04

pping

1.03

bs

1.01

othe

0.99

Activations Density 0.040%

phrases containing the word "so"

No Comments

No Known Activations