INDEX

Explanations

terms related to specific technical attributes, classifications, or qualities in electronics or audio equipment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elstein

-0.50

════

-0.48

fxml

-0.47

flashdata

-0.46

Composable

-0.45

+:+

-0.45

AddHtmlAttribute

-0.45

 sessionFactory

-0.44

Consigli

-0.44

SPJ

-0.43

POSITIVE LOGITS

0.86

 uLocal

0.80

Ui

0.76

 Ulysses

0.75

 Utica

0.72

UI

0.71

UA

0.70

 UIKit

0.69

Ue

0.69

Activations Density 0.319%

terms related to specific technical attributes, classifications, or qualities in electronics or audio equipment

No Comments

No Known Activations