INDEX

Explanations

phrases that express comparisons or similarities

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

"As" followed by certain function words

np_acts-logits-general · gemini-2.0-flash

as well as

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

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

ような

-0.46

 prévoir

-0.45

 Schnitt

-0.45

jší

-0.45

minus

-0.45

 personnaliser

-0.42

ssysteme

-0.42

ftar

-0.42

wärts

-0.41

methe

-0.41

POSITIVE LOGITS

MigrationBuilder

1.11

 ExecuteAsync

0.96

脚注の使い方

0.87

 part

0.81

 متعلقه

0.80

 للاسماء

0.78

ंदीखरीदारी

0.77

 doInBackground

0.76

الحياه

0.73

 مشين

0.73

Activations Density 0.164%

phrases that express comparisons or similarities

"As" followed by certain function words

as well as

No Comments

No Known Activations