INDEX

Explanations

repeated elements or structures in a message format

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

piece

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 NavController

-0.58

Wald

-0.53

 Darlington

-0.53

 Toxicology

-0.52

 Wald

-0.52

 Lakeland

-0.50

 Toxicological

-0.49

 Brentwood

-0.49

 Malvern

-0.49

DRS

-0.48

POSITIVE LOGITS

 piece

2.22

piece

2.13

 Piece

2.11

Piece

2.09

 PIECE

1.92

PIECE

1.79

 pieces

1.77

 Pieces

1.68

pieces

1.63

Pieces

1.61

Activations Density 0.008%

repeated elements or structures in a message format

piece

No Comments

No Known Activations