INDEX

Explanations

disclaimers and liability statements in the text

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

disclaimer

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

 المعيارى

-0.66

tagHelperRunner

-0.65

SequentialGroup

-0.61

➊

-0.61

InSection

-0.60

 kasarigan

-0.60

 CreateTagHelper

-0.57

-0.56

 electrolux

-0.56

copg

-0.56

POSITIVE LOGITS

 Disclaimer

0.95

Disclaimer

0.89

 disclaimer

0.88

CLAIMER

0.82

DISCLAIMER

0.76

claimers

0.63

claimer

0.61

 Haftung

0.42

 embarazada

0.39

 Leinwand

0.39

Activations Density 0.008%

disclaimers and liability statements in the text

disclaimer

No Comments

No Known Activations