INDEX

Explanations

numerical values related to memory and cache specifications in a technical context

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

Within HTML or code comments

np_acts-logits-general · gemini-2.0-flash

memory_ports, cache statistics

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 estekak

-0.97

 CreateTagHelper

-0.90

 Roskov

-0.80

Spoljašnje

-0.77

utives

-0.76

SBATCH

-0.76

 resourceCulture

-0.75

醐

-0.75

Datuak

-0.73

----+

-0.73

POSITIVE LOGITS

nie

0.53

atrième

0.51

<b>

0.50

 acer

0.48

ل

0.48

0.47

 upor

0.46

//

0.45

hawatir

0.44

Crom

0.43

Activations Density 0.325%

numerical values related to memory and cache specifications in a technical context

Within HTML or code comments

memory_ports, cache statistics

No Comments

No Known Activations