INDEX

Explanations

words that convey high praise or quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ویکی‌پدیا

-0.72

ScopeManager

-0.72

UnifiedTopology

-0.65

 soyez

-0.64

mars

-0.60

mMap

-0.58

undaki

-0.57

TableBody

-0.57

 تعدى

-0.57

 getItemCount

-0.57

POSITIVE LOGITS

 wonderful

1.15

 excellent

1.15

 bagus

1.07

excellent

1.07

 terrific

1.06

 good

1.06

 fantastic

1.05

wonderful

1.01

fantastic

1.00

 marvelous

0.99

Activations Density 0.084%

words that convey high praise or quality

No Comments

No Known Activations