INDEX

Explanations

a variety of words or brands related to online media or entertainment

oai_token-act-pair · gemini-2.0-flash

Proper nouns/entities

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nakalista

-0.70

 chrétienne

-0.69

 chrétiens

-0.59

 marié

-0.59

 credere

-0.58

 autorytatywna

-0.58

 brancas

-0.58

 précédents

-0.58

 africains

-0.58

 stessi

-0.57

POSITIVE LOGITS

sm

0.55

ownie

0.54

 CURIAM

0.52

 estimés

0.51

 Computer

0.51

XtraGrid

0.50

enthal

0.49

 Magazine

0.49

 Cement

0.49

def

0.48

Activations Density 7.374%

a variety of words or brands related to online media or entertainment

Proper nouns/entities

No Comments

No Known Activations