INDEX

Explanations

elements related to procedural instructions and interactions in a formal context

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

SBATCH

-0.56

 intptr

-0.54

GOTREF

-0.52

 continuant

-0.49

TabIndex

-0.49

ezers

-0.49

uoco

-0.48

postmedia

-0.47

chillo

-0.46

脚注の使い方

-0.46

POSITIVE LOGITS

 visit

2.68

 visits

2.62

 visited

2.48

visit

2.42

 visiting

2.39

 Visits

2.32

 visita

2.20

 Visit

2.19

visits

2.19

 visite

2.13

Activations Density 0.338%

elements related to procedural instructions and interactions in a formal context

No Comments

No Known Activations