INDEX

Explanations

the word "nothing" followed by phrases indicating a lack of specific qualities or characteristics

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the absence or lack of something

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

ebin

-0.76

anos

-0.76

odan

-0.73

ahime

-0.71

eda

-0.71

ATIONS

-0.70

vale

-0.70

agna

-0.69

 Pastebin

-0.69

querade

-0.68

POSITIVE LOGITS

 wrong

1.21

 inherently

1.13

 stopping

1.11

 intrinsically

1.01

 else

0.98

 objectionable

0.94

 nefarious

0.92

 preventing

0.91

 shameful

0.91

 unusual

0.86

Activations Density 0.050%

the word "nothing" followed by phrases indicating a lack of specific qualities or characteristics

phrases that emphasize the absence or lack of something

No Comments

No Known Activations