INDEX

Explanations

numerical values or identifiers

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

Code or markup snippets

np_acts-logits-general · gemini-2.0-flash

foreign language computer code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

Datuak

-0.96

 estekak

-0.93

CodedInputStream

-0.89

TagMode

-0.83

 betweenstory

-0.81

 unknownFields

-0.81

 >=",

-0.78

={()=>

-0.74

ChildScrollView

-0.74

 ProtoMessage

-0.73

POSITIVE LOGITS

0.46

↵↵

0.44

 présentes

0.41

 françaises

0.40

 XNUMX

0.39

ständige

0.37

 tích

0.37

 milit

0.36

/>}

0.35

 kanyang

0.35

Activations Density 0.077%

numerical values or identifiers

Code or markup snippets

foreign language computer code

No Comments

No Known Activations