INDEX

Explanations

data or references related to percentage values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 StringTokenizer

-0.48

而出

-0.47

 ModelExpression

-0.46

 travaillons

-0.46

 konfer

-0.45

Diweddarwch

-0.45

urrent

-0.44

 bahay

-0.43

writeHead

-0.43

ModelSerializer

-0.43

POSITIVE LOGITS

AndroidJUnit

0.75

%}

0.66

出版年

0.63

FTFY

0.60

lüğ

0.59

NUMX

0.57

 numberWith

0.56

Composable

0.56

pyx

0.56

醐

0.55

Activations Density 0.255%

data or references related to percentage values

No Comments

No Known Activations