INDEX

Explanations

references to programming elements, particularly related to testing and system calls

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

code identifiers and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 Urbano

-0.35

 conjunta

-0.33

 jointly

-0.33

 pessoais

-0.33

Autoritní

-0.33

 joint

-0.32

 Romeo

-0.32

dataTables

-0.32

 AssemblyProduct

-0.31

Rendez

-0.31

POSITIVE LOGITS

 насељу

0.92

AppSettings

0.92

NoSuchAlgorithm

0.87

NSNotification

0.78

parametrize

0.76

ItemLayout

0.73

Vidite

0.67

InlineData

0.63

DebuggerNonUser

0.62

CppMethod

0.61

Activations Density 0.004%

references to programming elements, particularly related to testing and system calls

code identifiers and constructs

No Comments

No Known Activations