INDEX

Explanations

links or references to Java methods and components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

󠁢

-0.59

 Munro

-0.57

 Zelensky

-0.57

 disambiguazione

-0.56

 Penrose

-0.56

 Niger

-0.55

 Voss

-0.52

 Melrose

-0.52

Gl

-0.51

URUS

-0.51

POSITIVE LOGITS

CloseOperation

0.80

Personendaten

0.80

 Cartagena

0.74

 Carlisle

0.71

 Chandler

0.69

リエ

0.68

Arial

0.68

Autoritní

0.68

 Ricard

0.68

 kart

0.68

Activations Density 0.841%

links or references to Java methods and components

No Comments

No Known Activations

links or references to Java methods and components

No Comments

No Known Activations