INDEX

Explanations

phrases indicating processes, findings, or claims made in research contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

and

-0.64

Життєпис

-0.60

ngdoc

-0.52

However

-0.47

ביוגרפיה

-0.46

atoms

-0.46

ⓘ

-0.46

but

-0.45

AxisAlignment

-0.45

sidemargin

-0.44

POSITIVE LOGITS

ografija

0.76

 Paglinawan

0.68

Personensuche

0.67

xase

0.65

 Wikimedijinoj

0.64

 дописавши

0.61

abestanden

0.60

(!__

0.59

hyrchwyd

0.57

SharedCtor

0.57

Activations Density 0.670%

phrases indicating processes, findings, or claims made in research contexts

No Comments

No Known Activations