INDEX

Explanations

keywords related to coding constructs and data management operations

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

code and punctuation boundaries

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

 CreateTagHelper

-0.78

Autoritní

-0.74

 AssemblyTitle

-0.69

queryInterface

-0.66

 betweenstory

-0.66

%")

-0.64

AnchorStyles

-0.62

principalColumn

-0.61

OGND

-0.60

orological

-0.60

POSITIVE LOGITS

 again

0.40

0.37

 Again

0.36

again

0.35

Again

0.30

 militaires

0.30

ニア

0.30

 blé

0.30

astéroïdes

0.29

):

0.29

Activations Density 0.166%

keywords related to coding constructs and data management operations

code and punctuation boundaries

No Comments

No Known Activations