INDEX

Explanations

words related to ineffectiveness or inefficiency

oai_token-act-pair · gpt-3.5-turbo

variations of the suffix "ine" and "ne" in words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Spa

-0.82

 Brands

-0.71

Yor

-0.68

ACY

-0.65

EVA

-0.64

 Shepherd

-0.62

 Lauder

-0.62

 Kara

-0.62

 Spears

-0.62

 Brach

-0.61

POSITIVE LOGITS

fficient

1.20

getic

1.11

nerg

1.07

fficiency

1.07

ased

1.03

velop

1.03

ffect

1.02

ither

1.02

farious

1.01

gment

1.01

Activations Density 0.112%

words related to ineffectiveness or inefficiency

variations of the suffix "ine" and "ne" in words

No Comments

No Known Activations