INDEX

Explanations

terms related to projectors and projections

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

projector and projection

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

Negative Logits

-0.44

 Tame

-0.43

 life

-0.43

Leben

-0.43

 stav

-0.41

 Cope

-0.41

etCode

-0.40

 Inte

-0.39

initComponents

-0.38

ContextCompat

-0.37

POSITIVE LOGITS

 projector

1.98

 Projector

1.79

 projectors

1.78

 Printer

0.74

printer

0.73

 printer

0.73

投影

0.73

 câmera

0.66

acuzzi

0.65

 geladeira

0.65

Activations Density 0.006%

terms related to projectors and projections

projector and projection

No Known Activations