INDEX

Explanations

imperative sentences instructing to try something

oai_token-act-pair · gpt-3.5-turbo

repeated prompts or suggestions to take action

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĺħ

-0.80

ļé

-0.62

å¥³

-0.60

independence

-0.60

fashion

-0.59

liber

-0.59

\-

-0.59

conservancy

-0.58

 Creed

-0.58

reported

-0.58

POSITIVE LOGITS

Try

0.99

Try

0.93

ters

0.73

ipers

0.72

tery

0.69

ctory

0.68

try

0.68

nir

0.67

iban

0.67

 Mayo

0.66

Activations Density 0.014%

imperative sentences instructing to try something

repeated prompts or suggestions to take action

No Comments

No Known Activations