INDEX

Explanations

repeated instances of the word "we."

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

Sentences starting with "We"

np_acts-logits-general · gemini-2.0-flash

We followed by verb

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 JADX

-0.41

-0.40

ly

-0.38

 other

-0.37

 another

-0.37

덟

-0.36

 foreign

-0.35

 Autre

-0.33

ization

-0.32

 Hyun

-0.32

POSITIVE LOGITS

We

1.21

we

1.03

We

1.02

Mereka

0.98

Мы

0.93

they

0.90

 เรา

0.89

mereka

0.88

 Mereka

0.88

we

0.87

Activations Density 0.179%

repeated instances of the word "we."

Sentences starting with "We"

We followed by verb

No Comments

No Known Activations