INDEX

Explanations

details regarding physical structures and their accessibility

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

after certain prepositions (to, in, of, as)

np_acts-logits-general · gemini-2.0-flash

use or position of things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 pomy

-0.32

 têtes

-0.31

 döv

-0.29

 Vorte

-0.29

 мест

-0.29

 đồng

-0.29

 encuent

-0.29

 kepentingan

-0.28

owe

-0.28

Type

-0.28

POSITIVE LOGITS

IntoConstraints

0.65

principalTable

0.62

 queſta

0.60

 betweenstory

0.59

ViewFeatures

0.59

SequentialGroup

0.58

 esternos

0.57

PMailer

0.57

RegressionTest

0.57

Життєпис

0.57

Activations Density 0.413%

details regarding physical structures and their accessibility

after certain prepositions (to, in, of, as)

use or position of things

No Comments

No Known Activations