INDEX

Explanations

instances of the word "Baja" along with variations

oai_token-act-pair · gpt-3.5-turbo

references to the word "Baja."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

rigan

-0.80

ndra

-0.78

ted

-0.76

changes

-0.75

uers

-0.74

ters

-0.74

ername

-0.74

auld

-0.73

utenberg

-0.73

urgy

-0.72

POSITIVE LOGITS

aja

1.03

isoft

0.80

Ã±

0.75

 Dhabi

0.74

BILITY

0.71

ournal

0.70

arta

0.70

olla

0.68

 Airlines

0.66

Ã±a

0.66

Activations Density 0.009%

instances of the word "Baja" along with variations

references to the word "Baja."

No Comments

No Known Activations