INDEX

Explanations

numerical values associated with power measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Datuak

-0.91

تقاوى

-0.90

 Gill

-0.83

FieldNumber

-0.81

tagHelperRunner

-0.81

Архівовано

-0.80

 onPostExecute

-0.76

曖昧さ回避

-0.76

 CYTHON

-0.75

 ویکی‌پدیا

-0.75

POSITIVE LOGITS

tabPage

0.62

Jop

0.57

 Ades

0.57

ıç

0.55

vad

0.54

ECA

0.54

 Artem

0.54

yves

0.53

quic

0.53

 Fleisch

0.52

Activations Density 0.126%

numerical values associated with power measurements

No Comments

No Known Activations