INDEX

Explanations

terms related to dominance and hierarchy within structures or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 للاسماء

-0.90

 дописавши

-0.73

Vidite

-0.72

LEncoder

-0.72

yntaxException

-0.71

 initComponents

-0.71

 utafitiHapana

-0.71

tvguidetime

-0.70



-0.69

reportWebVitals

-0.68

POSITIVE LOGITS

 facie

0.57

равда

0.56

 minde

0.56

CEC

0.56

 Dede

0.54

 pleaſure

0.54

 promis

0.54

Портали

0.53

CLEAR

0.53

WithMany

0.53

Activations Density 0.744%

terms related to dominance and hierarchy within structures or systems

No Comments

No Known Activations