INDEX

Explanations

words ending in 'ly'

oai_token-act-pair · gpt-3.5-turbo

adverbs that end with 'ly'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ilater

-1.00

irlf

-0.80

eers

-0.79

itive

-0.71

ifully

-0.70

arians

-0.70

ilogy

-0.70

icultural

-0.69

imilation

-0.68

ista

-0.68

POSITIVE LOGITS

tics

1.29

rics

1.12

puff

1.08

sis

1.00

mph

0.93

ffe

0.91

nda

0.90

cos

0.88

cemic

0.86

adv

0.86

Activations Density 0.055%

words ending in 'ly'

adverbs that end with 'ly'

No Comments

No Known Activations