INDEX

Explanations

references to military-related topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 ſind

-0.70

majánló

-0.68

bewerken

-0.66

تقاوى

-0.64

ſelves

-0.62

ſelf

-0.62

 typelib

-0.61

 ―――――

-0.58

 NSCoder

-0.58

 pleaſure

-0.58

POSITIVE LOGITS

des

0.62

de

0.59

 nahilalakip

0.52

0.51

Gruß

0.50

0.48

MfG

0.42

Grüße

0.42

 TimeUnit

0.41

ⓧ

0.40

Activations Density 0.336%

references to military-related topics

No Comments

No Known Activations