INDEX

Explanations

references to physical locations or properties of premises

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

premises and premise

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

tanleria

-0.56

SBATCH

-0.54

Vikt

-0.53

SHIPPING

-0.52

 ChromeDriver

-0.52

Svg

-0.52

 المعيارى

-0.51

vtk

-0.51

 lucky

-0.50

TargetException

-0.50

POSITIVE LOGITS

 Premises

0.98

 premises

0.94

Premise

0.83

 premise

0.82

premise

0.71

premises

0.69

mius

0.52

"]))

0.51

 Wikiseite

0.50

}}">

0.50

Activations Density 0.005%

references to physical locations or properties of premises

premises and premise

No Comments

No Known Activations