INDEX

Explanations

phrases indicating processes, instructions, or conditions related to web content and assignments

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

Followed by personal pronouns

np_acts-logits-general · gemini-2.0-flash

maternal exposure to Cd

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WebVitals

-0.83

bootstrapcdn

-0.83

RenderAtEndOf

-0.82

LookAnd

-0.80

 ComVisible

-0.79

particular

-0.77

 resourceCulture

-0.76

SharedDtor

-0.75

 pinulongan

-0.72

 Signalez

-0.72

POSITIVE LOGITS

要

0.49

 order

0.48

以

0.44

 الدوري

0.43

PRWEB

0.43

由

0.41

 curé

0.40

 Order

0.39

ClientSize

0.38

从

0.37

Activations Density 0.047%

phrases indicating processes, instructions, or conditions related to web content and assignments

Followed by personal pronouns

maternal exposure to Cd

No Comments

No Known Activations