INDEX

Explanations

words related to application or use in a technical or physiological context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "applied" and its variations, indicating the action of using or implementing something

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

OUS

-0.72

omes

-0.68

arta

-0.67

keeping

-0.66

etsk

-0.64

 Proto

-0.64

aults

-0.63

\">

-0.63

bda

-0.63

quel

-0.63

POSITIVE LOGITS

aution

0.79

pse

0.78

 parach

0.72

 inconsist

0.72

 uniformly

0.72

 appl

0.71

 pressure

0.71

 mathemat

0.70

 applied

0.70

 satell

0.69

Activations Density 0.024%

words related to application or use in a technical or physiological context

instances of the word "applied" and its variations, indicating the action of using or implementing something

No Comments

No Known Activations