INDEX

Explanations

tokens preceding numbers

oai_token-act-pair · gemini-2.0-flash

uncommon, formal usage

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

mas

-0.48

ég

-0.40

 ре

-0.40

 AppCompat

-0.40

\{\\

-0.39

DateTimeField

-0.39

ConstraintMaker

-0.38

mt

-0.38

дно

-0.38

ixe

-0.37

POSITIVE LOGITS

存于互联网档案馆

0.91

 raiſ

0.82

SharedDtor

0.77

 виправивши

0.74

مصادر

0.73

 doInBackground

0.72

 poffible

0.72

 themſelves

0.71

 myſelf

0.71

 חיצוניים

0.71

Activations Density 4.326%

tokens preceding numbers

uncommon, formal usage

No Comments

No Known Activations