INDEX

Explanations

specific programming constructs and terminology related to query methods and filtering in code

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

type, flag, state, protocol

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

numerusform

-0.79

tagHelperRunner

-0.75

EndContext

-0.75

出版年

-0.71

StoryboardSegue

-0.71

ロウィン

-0.68

 <<<<<<<<<<<<<<

-0.68

principalColumn

-0.68

<unused8>

-0.67

<unused28>

-0.67

POSITIVE LOGITS

 poitrine

0.26

select

0.26

 select

0.25

au

0.25

 head

0.25

 begin

0.25

擇

0.25

head

0.24

<code>

0.24

知

0.24

Activations Density 0.019%

specific programming constructs and terminology related to query methods and filtering in code

type, flag, state, protocol

No Comments

No Known Activations