INDEX

Explanations

references to properties and methods in code-related contexts

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

function calls to get

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

NgModule

-0.45

langkah

-0.44

=$?

-0.43

 betweenstory

-0.42

Batis

-0.39

 Parties

-0.39

 Jahn

-0.38

着

-0.38

AISSEE

-0.38

irvana

-0.38

POSITIVE LOGITS

SharedCtor

0.57

">//

0.51

 <>",

0.47

PerformLayout

0.44

 Foire

0.40

XmlElement

0.39

ChildScrollView

0.39

PreInfinity

0.39

lovakia

0.38

ElementException

0.38

Activations Density 0.038%

references to properties and methods in code-related contexts

function calls to get

No Comments

No Known Activations