INDEX

Explanations

programming-related terms and functions

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

openqa

-0.44

 "..\..\..\

-0.44

 toolStrip

-0.44

wear

-0.43

 UserDao

-0.41

opher

-0.40

orea

-0.40

 Roper

-0.40

 ReactDOM

-0.39

MathML

-0.39

POSITIVE LOGITS

 propOrder

0.90

compact

0.74

Personendaten

0.73

 compact

0.73

QMetaType

0.71

Compact

0.69

 AssemblyTitle

0.67

 nahilalakip

0.66

 Compact

0.65

コンパクト

0.64

Activations Density 0.000%

programming-related terms and functions

No Comments

No Known Activations