INDEX

Explanations

references to concepts and their definitions or discussions in various contexts

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

following "concept"

np_acts-logits-general · gemini-2.0-flash

concept development

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

IBarButtonItem

-0.54

 automática

-0.51

 tavo

-0.51

 personalizada

-0.50

InputBorder

-0.50

 Conducted

-0.49

 electrónico

-0.49

üyü

-0.49

omla

-0.48

Schedules

-0.48

POSITIVE LOGITS

 concepts

1.11

 concept

0.94

 CONCEPT

0.85

 Concepts

0.84

Concepts

0.84

onAttach

0.77

XmlType

0.76

Concept

0.75

 idea

0.75

tagHelperRunner

0.74

Activations Density 0.060%

references to concepts and their definitions or discussions in various contexts

following "concept"

concept development

No Comments

No Known Activations