INDEX

Explanations

structured data format elements

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

Strings in JSON-like structures

np_acts-logits-general · gemini-2.0-flash

technical documentation or specific knowledge

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

AndEndTag

-0.86

 kasarigan

-0.81

UnitTesting

-0.78

Архівовано

-0.78

 EconPapers

-0.77

Tembelea

-0.75

tvguidetime

-0.75

原始内容存档于

-0.73

AddTagHelper

-0.70

EndContext

-0.70

POSITIVE LOGITS

Son

0.45

Son

0.43

 components

0.41

KHR

0.40

 properties

0.40

 asim

0.39

 loto

0.38

ents

0.38

 sections

0.38

bar

0.37

Activations Density 0.508%

structured data format elements

Strings in JSON-like structures

technical documentation or specific knowledge

No Comments

No Known Activations