INDEX

Explanations

phrases like "they never", "almost never", and "it's", along with several seemingly random words

oai_token-act-pair · gemini-2.0-flash

never

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

PreferredItem

-0.48

новништво

-0.47

 Einzelnen

-0.43

 nationaux

-0.41

 AssemblyCulture

-0.40

 CURIAM

-0.40

 gerekiyor

-0.39

 rospy

-0.39

зя

-0.39

AdapterView

-0.38

POSITIVE LOGITS

 never

2.86

 Never

2.67

never

2.64

Never

2.53

 NEVER

2.45

NEVER

2.30

 nigdy

2.20

 nunca

2.19

 ever

2.11

 jamais

2.09

Activations Density 2.315%

phrases like "they never", "almost never", and "it's", along with several seemingly random words

never

No Comments

No Known Activations