INDEX

Explanations

code-related terms and constructs, particularly in programming and software development

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

angiotensin

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

 שוליים

-0.56

Hentet

-0.56

 esboço

-0.52

Lähteet

-0.51

 الاطلاع

-0.48

__*/

-0.48

RectangleBorder

-0.47

脚注の使い方

-0.46

 menga

-0.45

sólo

-0.45

POSITIVE LOGITS

MeasureSpec

0.91

ejs

0.75

ButterKnife

0.75

spyOn

0.68

 getItemId

0.66

 onFailure

0.63

iotensin

0.61

 localObject

0.55

OpenHelper

0.55

bazel

0.51

Activations Density 0.001%

code-related terms and constructs, particularly in programming and software development

angiotensin

No Comments

No Known Activations

code-related terms and constructs, particularly in programming and software development

angiotensin

No Comments

No Known Activations