INDEX

Explanations

sentences containing numbers, measurements, or references to dates

oai_token-act-pair · gemini-2.0-flash

exactly

np_token-act-pair-logits · gemini-2.0-flash

fine adjustments

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 bezeichneter

-0.85

 actionMode

-0.75

ValueStyle

-0.75

 виправивши

-0.70

 ProtoMessage

-0.69

Geplaatst

-0.68

IsContent

-0.67

Portail

-0.67

+#+

-0.66

 AssemblyTitle

-0.66

POSITIVE LOGITS

Сред

0.47

ⓘ

0.47

 Original

0.46

 dentes

0.46

Сере

0.46

Original

0.46

℗

0.45

FunctionFlags

0.43

length

0.43

"$

0.43

Activations Density 1.965%

sentences containing numbers, measurements, or references to dates

exactly

fine adjustments

No Comments

No Known Activations