INDEX

Explanations

phrases related to counting and inclusion in calculations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/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.16.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

Autoritní

-0.71

########.

-0.71

 ujednoznacz

-0.61

#+#

-0.56

Espèce

-0.54

WindowConstants

-0.53

 متعلقه

-0.52

homonymie

-0.51

Personendaten

-0.50

ultima

-0.48

POSITIVE LOGITS

 counted

1.18

 count

0.94

 counts

0.93

 counting

0.88

counted

0.81

 accounted

0.79

不算

0.79

 factored

0.79

ounted

0.77

 included

0.76

Activations Density 0.598%

phrases related to counting and inclusion in calculations

No Comments

No Known Activations