INDEX

Explanations

code elements related to struct or function definitions and their respective indicators in programming syntax

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

code generation context

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

nmgp

-0.49

rek

-0.46

}}}}

-0.46

dibles

-0.46

槻

-0.45

≦)

-0.45

codable

-0.44

iction

-0.44

 lên

-0.44

CodedInputStream

-0.44

POSITIVE LOGITS

 informée

0.62

GOTREF

0.55

ArgsConstructor

0.44

SourceChecksum

0.42

RTGC

0.42

 насељу

0.42

 renovables

0.41

 ***!

0.40

aarrggbb

0.39

 disponibilités

0.39

Activations Density 0.004%

code elements related to struct or function definitions and their respective indicators in programming syntax

code generation context

No Comments

No Known Activations