INDEX

Explanations

numeric values or measurements that could indicate data or statistics

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

Referanser

-0.44

Selon

-0.43

 according

-0.43

 funktion

-0.40

 frequently

-0.40

Secondo

-0.39

According

-0.38

 fréquemment

-0.38

 Menurut

-0.38

 rather

-0.37

POSITIVE LOGITS

transQ

0.58

 joaat

0.55

expandindo

0.53

InputBorder

0.53

 Italijanski

0.52

 OMITBAD

0.49

toContain

0.49

mAh

0.49

enablog

0.48

kDa

0.48

Activations Density 0.360%

numeric values or measurements that could indicate data or statistics

No Comments

No Known Activations