INDEX

Explanations

references to automotive or ignition system components and related terms

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

providedIn

-0.57

 doInBackground

-0.52

⟬

-0.51

 pitié

-0.50

 institucional

-0.49

 urbanos

-0.47

 liberación

-0.47

 věci

-0.47

 Rohr

-0.47

sumi

-0.47

POSITIVE LOGITS

 TMPro

0.41

 eggs

0.39

 spark

0.37

getActiveSheet

0.35

 fils

0.35

keb

0.35

 drain

0.34

 Alfa

0.34

jspx

0.33

okovic

0.33

Activations Density 0.001%

references to automotive or ignition system components and related terms

No Comments

No Known Activations