INDEX

Explanations

references to quantitative measurements and comparative descriptions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DockStyle

-0.63

gdx

-0.61

帖最后由

-0.61

 HasFactory

-0.61

uxxxx

-0.57

Билгалдахарш

-0.56

:+:

-0.54

dataclass

-0.54

Cyfarwyddwr

-0.53

 utafitiHapana

-0.53

POSITIVE LOGITS

 cuantos

0.33

 qualitative

0.29

yli

0.28

 enough

0.28

BitmapFactory

0.27

Voir

0.27

 Schä

0.27

 ciertamente

0.27

 Mittel

0.26

SMALL

0.26

Activations Density 0.126%

references to quantitative measurements and comparative descriptions

No Comments

No Known Activations