INDEX

Explanations

structured data in a specific format, likely related to programming or configuration syntax

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

HTML and code elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 виправивши

-0.82

 utafitiHapana

-0.79

 wikipagina

-0.78

ьаж

-0.74

 Bewußt

-0.73

 Geſch

-0.71

Билгалдахарш

-0.69

 queſta

-0.68

parsedMessage

-0.67

PreferredItem

-0.66

POSITIVE LOGITS

0.41

Mas

0.40

Bo

0.40

das

0.40

0.39

<tr>

0.39

bo

0.38

Ver

0.38

Activations Density 0.015%

structured data in a specific format, likely related to programming or configuration syntax

HTML and code elements

No Comments

No Known Activations