INDEX

Explanations

phrases indicating maintenance or updates of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 oprot

-0.72

новниш

-0.61

 <=",

-0.58

IContainer

-0.57

 appContext

-0.56

Geplaatst

-0.55

SequentialGroup

-0.54

 Buka

-0.53

 للمعارف

-0.52

strahlung

-0.52

POSITIVE LOGITS

 date

1.20

date

1.06

DATE

0.91

 DATE

0.89

Date

0.85

 Date

0.83

 dated

0.82

 speed

0.81

par

0.77

dated

0.70

Activations Density 0.119%

phrases indicating maintenance or updates of information

No Comments

No Known Activations