INDEX

Explanations

indications of feedback or suggestions for clarification

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

questions and uncertainties

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

 queſta

-0.48

 TextAlign

-0.46

 ContentValues

-0.46

 handleMessage

-0.45

 TextAppearance

-0.45

iſen

-0.44

AnchorTagHelper

-0.44

 perfection

-0.43

basicConfig

-0.43

 GenerationType

-0.43

POSITIVE LOGITS

 resourceCulture

0.38

WriteTagHelper

0.36

nloa

0.35

OGND

0.34

setVerticalGroup

0.34

 judicious

0.34

rej

0.34

graphicx

0.33

Билгалдахарш

0.33

󠁴

0.33

Activations Density 0.053%

indications of feedback or suggestions for clarification

questions and uncertainties

No Comments

No Known Activations