INDEX

Explanations

phrases that indicate near or total completeness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

cektir

-0.72

 Guen

-0.57

Algérie

-0.57

either

-0.56

esség

-0.55

 Leicester

-0.55

ceği

-0.54

 lugs

-0.54

RefreshLayout

-0.54

GenerationType

-0.53

POSITIVE LOGITS

 nearly

1.15

 almost

1.08

Almost

1.04

 Almost

1.04

Casi

1.02

 Nearly

0.99

nearly

0.97

 Casi

0.95

almost

0.90

Nearly

0.90

Activations Density 0.097%

phrases that indicate near or total completeness

No Comments

No Known Activations