INDEX

Explanations

terms related to data schemas and programming protocols

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

architecture and build parameters

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

ftagPool

-0.69

ReusableCell

-0.59

RegressionTest

-0.57

="@+

-0.57

UrlResolution

-0.54

reportWebVitals

-0.53

 <<<<<<<<<<<<<<

-0.53

principalColumn

-0.51

findpost

-0.51

saraba

-0.50

POSITIVE LOGITS

CppMethod

0.42

 Roskov

0.42

MigrationBuilder

0.41

MIDDLEWARE

0.41

 esternos

0.39

Nationalité

0.39

resizingMask

0.39

 виправивши

0.38

 ISNI

0.38

RotationOrder

0.38

Activations Density 0.061%

terms related to data schemas and programming protocols

architecture and build parameters

No Comments

No Known Activations