INDEX

Explanations

technical and scholarly terms ending in 'ive', 'able', 'atory', 'itive', 'ical', 'matic', 'ing', 'uary', 'ging', and other similar suffixes

oai_token-act-pair · gpt-3.5-turbo

words that suggest characteristics or attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Pak

-0.66

nir

-0.64

DEV

-0.63

away

-0.61

ibal

-0.58

igslist

-0.58

erves

-0.58

jab

-0.57

KC

-0.57

tab

-0.56

POSITIVE LOGITS

rium

0.77

 behav

0.77

*/(

0.73

rious

0.70

uese

0.68

theless

0.66

 indu

0.63

tenance

0.63

Background

0.62

rons

0.60

Activations Density 0.242%

technical and scholarly terms ending in 'ive', 'able', 'atory', 'itive', 'ical', 'matic', 'ing', 'uary', 'ging', and other similar suffixes

words that suggest characteristics or attributes

No Comments

No Known Activations