INDEX

Explanations

the word "manikin" or slight variations of it

oai_token-act-pair · gpt-3.5-turbo

tokens related to a specific repeated term or name

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 goose

-0.62

 common

-0.59

 poppy

-0.59

 equivalents

-0.58

âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ

-0.57

 fide

-0.56

 Breach

-0.56

ICE

-0.56

 Suff

-0.56

DMV

-0.55

POSITIVE LOGITS

achu

1.26

rish

1.23

ernel

1.16

ernels

1.12

ku

1.12

ileaks

1.10

hs

1.07

owski

1.06

yu

1.05

ki

1.02

Activations Density 0.043%

the word "manikin" or slight variations of it

tokens related to a specific repeated term or name

No Comments

No Known Activations