INDEX

Explanations

numerical or statistical elements in text

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

хьтан

-0.67

IsMutable

-0.65

 дописавши

-0.65

 <>",

-0.65

 Italijanski

-0.62

hoeddwyd

-0.56

 BorderSide

-0.54

rungsseite

-0.54

Климаты

-0.54

 resourceCulture

-0.53

POSITIVE LOGITS

evos

0.35

triangleq

0.34

ACLE

0.33

şe

0.32

ANGEL

0.32

 tiro

0.31

Tir

0.31

nement

0.31

 cortina

0.31

Jessica

0.30

Activations Density 0.031%

numerical or statistical elements in text

No Comments

No Known Activations