INDEX

Explanations

phrases related to specific projects, possibly involving research, initiatives, or programs

oai_token-act-pair · gpt-3.5-turbo

references to specific projects or initiatives

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

aws

-0.73

ãģĦ

-0.72

bane

-0.72

sbm

-0.72

^^^^

-0.72

taboola

-0.71

mson

-0.70

ãĥ©ãĥ³

-0.70

combe

-0.69

ously

-0.66

POSITIVE LOGITS

 Gutenberg

1.38

ions

1.05

ion

1.02

ioned

0.89

ually

0.78

Ara

0.78

ors

0.78

JECT

0.75

ional

0.74

 Project

0.74

Activations Density 0.016%

phrases related to specific projects, possibly involving research, initiatives, or programs

references to specific projects or initiatives

No Comments

No Known Activations