INDEX

Explanations

words related to absence or removal

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "abs" or absence in various contexts

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

eco

-0.65

 Gaia

-0.64

 bees

-0.61

ning

-0.60

 Inspector

-0.59

cf

-0.59

 BIOS

-0.59

 autos

-0.59

 eminent

-0.58

 Sass

-0.58

POSITIVE LOGITS

olute

1.42

olution

1.30

ences

1.26

cess

1.25

urd

1.23

olve

1.20

ence

1.17

olved

1.16

orb

1.13

ol

1.11

Activations Density 0.045%

words related to absence or removal

references to the concept of "abs" or absence in various contexts

No Comments

No Known Activations