INDEX

Explanations

structured data or numerical representations related to statistics or code formats

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

Followed by opening parenthesis or "U.S."

np_acts-logits-general · gemini-2.0-flash

What or List

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

hésite

-0.77

 RIPRODUZIONE

-0.73

horabuena

-0.68

espère

-0.66

 hâte

-0.64

 Conſ

-0.64

åne

-0.63

 ―――――

-0.63

 parís

-0.62

 Reſ

-0.62

POSITIVE LOGITS

0.77

Autoritní

0.71

</b>

0.65

0.56

</i>

0.56

</td>

0.53

 propOrder

0.53

The

0.52

«

0.52

Activations Density 1.871%

structured data or numerical representations related to statistics or code formats

Followed by opening parenthesis or "U.S."

What or List

No Comments

No Known Activations