INDEX

Explanations

terms related to software assembly and systematic descriptions

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

Code, abbreviations, or uncommon words

np_acts-logits-general · gemini-2.0-flash

code, definitions, and concepts

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-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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<eos>

-0.34

↵

-0.34

jimo

-0.34

 Defensa

-0.34

rade

-0.33

țiunea

-0.33

↵↵

-0.33

 geben

-0.33

 LIABLE

-0.32

POSITIVE LOGITS

 disambiguazione

0.90

 AssemblyCompany

0.81

DIPSETTING

0.78

RTLR

0.73

 للاسماء

0.70

 nahilalakip

0.68

weightedMode

0.67

SharedDtor

0.65

 ISNI

0.64

delwed

0.64

Activations Density 0.278%

terms related to software assembly and systematic descriptions

Code, abbreviations, or uncommon words

code, definitions, and concepts

No Comments

No Known Activations