INDEX

Explanations

terms and phrases related to direct methods and interactions

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

directly followed by noun

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

UserScript

-0.59

toThrow

-0.56

ControllerAdvice

-0.53

ArgumentParser

-0.52

ngdoc

-0.51

PyExc

-0.50

 fubject

-0.50

yntaxException

-0.50

UnsafeEnabled

-0.49

addCriterion

-0.48

POSITIVE LOGITS

 direct

0.91

direct

0.86

直接

0.81

Direct

0.80

 Direct

0.79

 directly

0.76

 directa

0.75

directly

0.73

 直接

0.69

 Directly

0.68

Activations Density 0.033%

terms and phrases related to direct methods and interactions

directly followed by noun

No Comments

No Known Activations