INDEX

Explanations

adverbs ending in 'ly'

oai_token-act-pair · gpt-3.5-turbo

adverbs that convey intensity or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

-0.84

afety

-0.73

 GOODMAN

-0.71

 respectively

-0.71

irlf

-0.68

 canvas

-0.67

ERA

-0.66

 hemor

-0.66

erto

-0.64

ËĪ

-0.63

POSITIVE LOGITS

puff

1.04

ly

0.90

forth

0.88

wash

0.82

lihood

0.80

upe

0.80

rics

0.79

brate

0.78

aqu

0.77

clad

0.77

Activations Density 0.026%

adverbs ending in 'ly'

adverbs that convey intensity or degree

No Comments

No Known Activations