INDEX

Explanations

references to figures, tables, and related scientific notations

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

figure and table references

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

 незавершена

-0.59

RectangleBorder

-0.58

AsUp

-0.57

ſcher

-0.57

ſchen

-0.57

iſen

-0.54

Portály

-0.54

DataPropertyName

-0.54

DIPSETTING

-0.53

niſſe

-0.52

POSITIVE LOGITS

 gyhoeddwyd

0.43

Viited

0.41

kannte

0.38

DockStyle

0.35

rosine

0.33

Grand

0.33

jenost

0.32

ostante

0.31

 Савезне

0.31

 Tomaten

0.29

Activations Density 0.031%

references to figures, tables, and related scientific notations

figure and table references

No Comments

No Known Activations