INDEX

Explanations

abbreviations and codes, likely related to technical or scientific contexts

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

Prefixed character combinations

np_acts-logits-general · gemini-2.0-flash

DNA bases and related codes

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

ingar

-0.46

-0.45

nextInt

-0.44

LW

-0.42

mbggenerated

-0.42

simas

-0.41

SD

-0.41

SRP

-0.41

MENT

-0.41

Facades

-0.41

POSITIVE LOGITS

 EconPapers

0.91

ftagPool

0.82

 مشين

0.80

GEBURTSDATUM

0.79

脚注の使い方

0.79

ngdoc

0.77

 Paglinawan

0.75

 kasarigan

0.75

GraphicsUnit

0.75

AndEndTag

0.74

Activations Density 0.590%

abbreviations and codes, likely related to technical or scientific contexts

Prefixed character combinations

DNA bases and related codes

No Comments

No Known Activations