INDEX

Explanations

technical terms and identifiers related to media processing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

setViewportView

-0.41

bo

-0.38

romp

-0.38

pos

-0.35

rom

-0.34

Rom

-0.34

dal

-0.33

Store

-0.33

fung

-0.33

Rom

-0.32

POSITIVE LOGITS

RenderAtEndOf

0.88

 betweenstory

0.77

Personendaten

0.69

脚注の使い方

0.68

 ब्रेकडाउन

0.64

 Signalez

0.62

rungsseite

0.60

 BoxFit

0.60

FunctionFlags

0.60

 Италијани

0.59

Activations Density 5.958%

technical terms and identifiers related to media processing

No Comments

No Known Activations