INDEX

Explanations

references to memory addresses in a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

iwa

-0.75

Redacción

-0.73

 Leland

-0.69

Kjelder

-0.68

 Fland

-0.67

kuper

-0.67

 disponibilités

-0.67

 ziren

-0.65

 Pitman

-0.65

mények

-0.64

POSITIVE LOGITS

 ADDRESS

1.62

 Addresses

1.57

 address

1.57

 addresses

1.54

 Address

1.53

address

1.51

Addresses

1.49

Address

1.46

 getAddress

1.43

addresses

1.42

Activations Density 0.064%

references to memory addresses in a technical context

No Comments

No Known Activations