INDEX

Explanations

numeric and non-roman characters in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

fromCharCode

-0.43

 thither

-0.41

 محفوظة

-0.39

Rohy

-0.38

 downvote

-0.38

เหร

-0.37

الدراسه

-0.37

orianCalendar

-0.36

 hamburgers

-0.36

 resourceCulture

-0.36

POSITIVE LOGITS

étape

0.51

฿

0.47

曖昧さ回避

0.45

Välislingid

0.44

thai

0.44

swire

0.43

 vios

0.42

sphase

0.42

échelle

0.42

 Absicht

0.42

Activations Density 0.002%

numeric and non-roman characters in a text

No Comments

No Known Activations