INDEX

Explanations

numerical values and references related to significant company events and expansions

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

years following numbers

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

RotationOrder

-0.46

 nahilalakip

-0.42

 błę

-0.35

Географи

-0.35

 otra

-0.34

calciatore

-0.33

 dispuestos

-0.32

另一個

-0.32

 filós

-0.32

 spôso

-0.32

POSITIVE LOGITS

 CreateTagHelper

0.62

Cyfarwyddwr

0.60

 مشين

0.56

الدراسه

0.54

CppCodeGen

0.51

 otomatig

0.50

VYMaps

0.50

harapkan

0.47

 Préférences

0.47

 AssemblyVersion

0.46

Activations Density 0.138%

numerical values and references related to significant company events and expansions

years following numbers

No Comments

No Known Activations