INDEX

Explanations

items related to different parts within a transcript or document

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

berus

-0.68

acha

-0.63

GES

-0.57

âĤ¬

-0.57

 minded

-0.57

ãĤ¬

-0.56

ITH

-0.56

 POLITICO

-0.56

ELD

-0.55

 Keen

-0.55

POSITIVE LOGITS

icularly

1.42

icular

1.33

nered

1.23

icipated

1.22

ially

1.20

ners

1.17

ridge

1.15

isans

1.14

icle

1.02

isan

1.02

Activations Density 0.026%

items related to different parts within a transcript or document

No Comments

No Known Activations