INDEX

Explanations

phrases that suggest recommendations or advice

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

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

 المعيارى

-0.71

Hauptartikel

-0.71

 StatefulWidget

-0.69

Искәрмәләр

-0.68

发表于

-0.63

 Audiodateien

-0.59

 VIAF

-0.58

anglès

-0.57

âu

-0.57

urator

-0.57

POSITIVE LOGITS

 vivement

0.90

 strongly

0.75

 enca

0.71

atory

0.67

 fortemente

0.66

 fortement

0.65

 dringend

0.63

strongly

0.60

 AGAINST

0.60

 against

0.59

Activations Density 0.115%

phrases that suggest recommendations or advice

No Comments

No Known Activations