INDEX

Explanations

the name "Ke" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ke" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Palestin

-0.86

displayText

-0.72

anamo

-0.71

ategory

-0.65

acebook

-0.65

ilogy

-0.64

iotics

-0.63

 tremend

-0.63

ó

-0.63

ãĤ¢ãĥ«

-0.62

POSITIVE LOGITS

llers

1.16

ller

1.12

ptic

1.10

eling

0.99

nda

0.97

ffect

0.95

ls

0.95

jriwal

0.93

leys

0.92

lev

0.91

Activations Density 0.011%

the name "Ke" in various contexts

instances of the word "ke" and its variations

No Comments

No Known Activations