INDEX

Explanations

a mix of monetary references, dates, and common prefixes/suffixes in words.

oai_token-act-pair · gemini-2.0-flash

French words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

#![

-0.37

pym

-0.36

lwz

-0.35

Marius

-0.35

+:+

-0.35

 mips

-0.35

 USART

-0.35

 nero

-0.34

gql

-0.34

 communis

-0.33

POSITIVE LOGITS

 hâte

0.84

 démission

0.84

 dépens

0.83

 dégâts

0.79

 rêves

0.78

 écl

0.78

 égard

0.77

 deuil

0.74

 autorité

0.74

 larmes

0.73

Activations Density 19.347%

a mix of monetary references, dates, and common prefixes/suffixes in words.

French words

No Comments

No Known Activations