INDEX

Explanations

words ending with '-ness'

oai_token-act-pair · gpt-3.5-turbo

terms related to various types of "ness," indicating states or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chnology

-0.80

olan

-0.75

iov

-0.72

rament

-0.72

reon

-0.71

veyard

-0.71

zynski

-0.70

ctors

-0.69

inav

-0.65

ellar

-0.62

POSITIVE LOGITS

iness

0.90

ness

0.79

fulness

0.78

lessness

0.77

ledged

0.70

istence

0.68

 expectancy

0.68

ï¸

0.68

oft

0.65

Activations Density 0.025%

words ending with '-ness'

terms related to various types of "ness," indicating states or conditions

No Comments

No Known Activations