INDEX

Explanations

quantifiers referring to the amount or scale of something

oai_token-act-pair · gpt-3.5-turbo

terms indicating size or quantity

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

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

events

-0.98

hops

-0.97

speak

-0.93

guards

-0.84

 Statements

-0.84

iddles

-0.83

hots

-0.82

stories

-0.79

ometers

-0.77

aunts

-0.77

POSITIVE LOGITS

 amount

1.57

 swath

1.32

 degree

1.29

 number

1.26

 chunk

1.20

 portion

1.19

 percentage

1.18

 proportion

1.15

 variety

1.15

 subset

1.13

Activations Density 0.219%

quantifiers referring to the amount or scale of something

terms indicating size or quantity

No Comments

No Known Activations