INDEX

Explanations

abbreviations and specific scientific variables

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

variant instance, clones, iteration, B=0

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

Chham

-0.39

buttonBar

-0.35

"]));

-0.32

$_['

-0.32

).__

-0.31

"]),

-0.30

]));

-0.30

CJK

-0.29

 heilig

-0.29

 formally

-0.28

POSITIVE LOGITS

 instance

0.75

实例

0.68

 Instances

0.65

 variant

0.63

 instances

0.63

 Instance

0.63

 iteration

0.63

 INSTANCE

0.60

instance

0.59

 clones

0.59

Activations Density 0.367%

abbreviations and specific scientific variables

variant instance, clones, iteration, B=0

No Comments

No Known Activations