INDEX

Explanations

elements and attributes associated with programmatic or structural hierarchy in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

and

-0.63

-0.52

 free

-0.52

in

-0.51

-0.50

on

-0.50

of

-0.49

 used

-0.49

しており

-0.49

POSITIVE LOGITS

 متعلقه

0.94

NUMX

0.76

\{\\

0.72

 appelez

0.72

 تانيه

0.70

Билгалдахарш

0.70

存于互联网档案馆

0.69

 purpoſe

0.68

UnifiedTopology

0.67

 esternos

0.67

Activations Density 0.114%

elements and attributes associated with programmatic or structural hierarchy in code

No Comments

No Known Activations