INDEX

Explanations

positive or notable aspects within sentences

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate the best or worst aspects of experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Klux

-0.86

ternity

-0.76

berus

-0.75

elsius

-0.73

 pione

-0.73

 practition

-0.73

ãĤī

-0.72

lished

-0.71

 cumbers

-0.69

arma

-0.69

POSITIVE LOGITS

ials

0.86

uring

0.72

ially

0.71

 Colbert

0.70

UID

0.69

ioned

0.69

 Whedon

0.65

nered

0.64

(>

0.63

Hoo

0.63

Activations Density 0.028%

positive or notable aspects within sentences

phrases that indicate the best or worst aspects of experiences

No Comments

No Known Activations