INDEX

Explanations

references to the use of various methods or techniques in different contexts

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

NUT

-0.54

 plane

-0.48

 PLANE

-0.48

gun

-0.47

 Bris

-0.47

GUN

-0.45

Vill

-0.45

 Sonn

-0.44

 baum

-0.44

 ModelAndView

-0.44

POSITIVE LOGITS

 CreateTagHelper

0.59

 оригіналу

0.58

 nakalista

0.58

 autorytatywna

0.55

存于互联网档案馆

0.54

脚注の使い方

0.50

Espèce

0.48

]")]

0.47

 مرئيه

0.47

 GenerationType

0.47

Activations Density 0.092%

references to the use of various methods or techniques in different contexts

No Comments

No Known Activations