INDEX

Explanations

punctuation marks, specifically periods

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

setVerticalGroup

-0.64

PDOException

-0.62

 препратки

-0.61

 oprot

-0.61

 мәкал

-0.61

InitVars

-0.59

AutoField

-0.59

Бахар

-0.59

abstractmethod

-0.58

UnusedPrivate

-0.58

POSITIVE LOGITS

 Occidente

0.31

excluded

0.31

 retrait

0.30

 vuestros

0.30

graphicx

0.30

 sienta

0.30

Erreferentziak

0.29

 commencer

0.28

 Gelb

0.28

ográficas

0.28

Activations Density 0.000%

punctuation marks, specifically periods

No Comments

No Known Activations