INDEX

Explanations

mentions of inventions or innovative processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

ngt

-0.47

et

-0.46

 aspi

-0.46

ieuses

-0.46

ров

-0.45

克斯

-0.44

usi

-0.41

])))

-0.41

ActionPerformed

-0.41

 Процитовано

-0.41

POSITIVE LOGITS

 invention

1.36

 Invention

1.04

invention

1.03

 Erfindung

1.01

 inventions

1.00

RenderAtEndOf

0.91

ंदीखरीदारी

0.89

 invented

0.88

 invent

0.83

 inventing

0.82

Activations Density 0.118%

mentions of inventions or innovative processes

No Comments

No Known Activations