INDEX

Explanations

words with the substring "ose"

oai_token-act-pair · gpt-3.5-turbo

words that relate to the concept of a "pose" or "posing."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

DOC

-0.66

ersen

-0.62

angible

-0.61

INGTON

-0.61

enberg

-0.61

Tat

-0.60

DERR

-0.60

 ILCS

-0.58

England

-0.58

LOD

-0.58

POSITIVE LOGITS

velt

0.98

eker

0.90

idon

0.90

cond

0.88

ose

0.86

ppel

0.84

ppe

0.82

lect

0.81

itor

0.80

OPLE

0.80

Activations Density 0.010%

words with the substring "ose"

words that relate to the concept of a "pose" or "posing."

No Comments

No Known Activations