INDEX

Explanations

symbols and formatting elements in text

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

punctuation and symbols

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

 للمعارف

-0.89

SequentialGroup

-0.78

sidemargin

-0.78

 AssemblyCompany

-0.77

 ویکی‌پدی

-0.74

migrationBuilder

-0.73

⟬

-0.72

 pinulongan

-0.72

IsMutable

-0.71

ftagPool

-0.71

POSITIVE LOGITS

 nostru

0.35

"}"

0.30

 обо

0.29

념

0.29

“

0.29

 Olympic

0.29

[toxicity=0]

0.29

 Affaires

0.29

 image

0.28

 reserves

0.28

Activations Density 0.274%

symbols and formatting elements in text

punctuation and symbols

No Comments

No Known Activations