INDEX

Explanations

qualifiers and intensifiers that modify the degree of an adjective or adverb

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

menor

-0.63

 soumis

-0.59

 ligiloj

-0.57

nější

-0.57

jší

-0.55

 adjustable

-0.54

 mínimos

-0.54

üht

-0.53

 erhältlich

-0.53

jahtera

-0.53

POSITIVE LOGITS

 efficiently

1.02

 aggressively

1.01

 differently

0.99

 creatively

0.96

 smoothly

0.95

 comfortably

0.94

successfully

0.93

 effortlessly

0.91

 peacefully

0.91

 successfully

0.91

Activations Density 0.308%

qualifiers and intensifiers that modify the degree of an adjective or adverb

No Comments

No Known Activations