INDEX

Explanations

occurrences of the term "US" and related variations

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

"US" or "us"

np_acts-logits-general · gemini-2.0-flash

USH, USH1, usury

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

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

DataAnnotations

-0.59

tagHelperRunner

-0.58

endregion

-0.57

jspb

-0.54

はじめに

-0.51

 Inſ

-0.50

حياته

-0.50

transQ

-0.50

ChildScrollView

-0.50

tır

-0.50

POSITIVE LOGITS

su

0.54

Us

0.49

RUS

0.46

TUS

0.46

ses

0.46

ser

0.45

suri

0.42

Gus

0.42

Aus

0.41

Cus

0.41

Activations Density 0.302%

occurrences of the term "US" and related variations

"US" or "us"

USH, USH1, usury

No Comments

No Known Activations