INDEX

Explanations

references to the core library in programming context

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

core module or path component

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

SerializedName

-0.61

endphp

-0.61

CodedInputStream

-0.60

лтемелер

-0.59

 ModelExpression

-0.58

 дописавши

-0.56

AntiForgeryToken

-0.56

postIndex

-0.56

BufferException

-0.56

IBarButtonItem

-0.56

POSITIVE LOGITS

core

1.70

Core

1.23

 core

1.23

 Core

1.09

CORE

1.05

 CORE

0.96

核心

0.90

コア

0.79

cores

0.77

 cores

0.71

Activations Density 0.004%

references to the core library in programming context

core module or path component

No Comments

No Known Activations