INDEX

Explanations

occurrences of the word "we."

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

"We" followed by action

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 houſe

-0.49

“

-0.47

المناصب

-0.46

ientôt

-0.46

 Италијани

-0.44

UFC

-0.43

《

-0.43

 pleaſure

-0.42

 Perſ

-0.42

democracy

-0.42

POSITIVE LOGITS

use

0.58

 consider

0.53

DockStyle

0.51

 hierzu

0.51

 hierfür

0.50

 must

0.49

を用

0.49

ธ์

0.48

 using

0.48

 celu

0.48

Activations Density 0.093%

occurrences of the word "we."

"We" followed by action

No Comments

No Known Activations