INDEX

Explanations

specific scientific terms and related jargon

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

characterizations, alleles

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

styleable

-0.32

 Goed

-0.31

CheckedChanged

-0.27

主

-0.26

一

-0.26

 sampai

-0.25

PutMapping

-0.25

 Langkah

-0.25

 natur

-0.25

 Lozano

-0.24

POSITIVE LOGITS



0.75

Ӕ

0.69

thâu

0.67

хьтан

0.67

tonsoft

0.66

 ModelExpression

0.65

 ModelRenderer

0.64

 Administrativna

0.63

onedDateTime

0.63

EDEFAULT

0.63

Activations Density 0.196%

specific scientific terms and related jargon

characterizations, alleles

No Comments

No Known Activations