INDEX

Explanations

references to structured lists or items in a professional context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Enumerations labeled with numbers and parentheses

np_acts-logits-general · gemini-2.0-flash

enumeration markers like (1) or (i)

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 AssemblyTitle

-0.67

 Theſe

-0.67

IndentedString

-0.64

Gent

-0.52

Autoritní

-0.52

 Drapeau

-0.51

Bhd

-0.51

Smarty

-0.49

ImageContext

-0.49

 excretion

-0.48

POSITIVE LOGITS

 either

0.60

AddTagHelper

0.58

0.56

reszcie

0.56

 where

0.55

;,

0.55

 whether

0.54

 ferner

0.53

 –,

0.53

²,

0.53

Activations Density 0.356%

references to structured lists or items in a professional context

Enumerations labeled with numbers and parentheses

enumeration markers like (1) or (i)

No Comments

No Known Activations