INDEX

Explanations

programming-related terms and structures

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

Rust code and code-related tokens

np_acts-logits-general · gemini-2.0-flash

code keywords `pub` `def` `impl` `fn`

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DotNetBar

-0.37

//

-0.36

amssymb

-0.34

 sociala

-0.31

xxl

-0.31

/*

-0.30

fromnode

-0.30

 preto

-0.30

 Catering

-0.29

 trefoil

-0.29

POSITIVE LOGITS

 AssemblyVersion

0.76

DockStyle

0.67

SharedDtor

0.60

setVerticalGroup

0.59

 rhestr

0.59

SharedCtor

0.58

arangay

0.57

 ProtoMessage

0.56

zzleHttp

0.56

๣

0.56

Activations Density 0.211%

programming-related terms and structures

Rust code and code-related tokens

code keywords `pub` `def` `impl` `fn`

No Comments

No Known Activations