INDEX

Explanations

references to studies or data analyses in a scientific context

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

citation markers

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

 AssemblyVersion

-0.81

 CURIAM

-0.81

IsMutable

-0.73

ConstraintMaker

-0.73

 utafitiHapana

-0.71

 EconPapers

-0.69

AddTagHelper

-0.68

DockStyle

-0.66

ForRow

-0.65

 transfieras

-0.65

POSITIVE LOGITS

 warszaw

0.33

 recently

0.32

 recientemente

0.31

źródło

0.31

Recently

0.31

Czytaj

0.30

 Warszawie

0.30

tryk

0.30

 Recently

0.30

def

0.30

Activations Density 0.229%

references to studies or data analyses in a scientific context

citation markers

No Comments

No Known Activations