INDEX

Explanations

instances of the word "li" and related lexical forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 AspNetCore

-0.67

 aikaa

-0.56

acimiento

-0.52

haustible

-0.52

 floors

-0.50

 oluşan

-0.49

 bibliothèque

-0.49

trand

-0.49

 magasiner

-0.49

runOnUiThread

-0.48

POSITIVE LOGITS

ⓧ

0.66

 Winaray

0.60

@}

0.56

بوابة

0.53

PhysRev

0.53

BufferException

0.52

ագրություններ

0.52

op

0.51

Portale

0.51

ValueStyle

0.49

Activations Density 0.157%

instances of the word "li" and related lexical forms

No Comments

No Known Activations