INDEX

Explanations

phrases that express agreement or similar sentiments

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

informal or emphatic 'like'

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

 arşivlendi

-0.63

ьажоргаш

-0.51

 przyczyn

-0.49

 soluções

-0.48

 vertrou

-0.48

BagConstraints

-0.48

 tradizionale

-0.48

StoryboardSegue

-0.48

 gebouw

-0.47

//

-0.47

POSITIVE LOGITS

 weird

0.68

 dude

0.68

 crazy

0.66

 LIKE

0.63

 like

0.60

like

0.59

 Like

0.59

 hella

0.57

weird

0.57

Like

0.57

Activations Density 0.017%

phrases that express agreement or similar sentiments

informal or emphatic 'like'

No Comments

No Known Activations