INDEX

Explanations

references to instances or entities being discussed in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 snippetHide

-0.49

XmlAccessType

-0.49

Jeografia

-0.47

Jîn

-0.45

AutoScale

-0.45

 antworte

-0.44

HttpPut

-0.43

Odkazy

-0.43

 Rivas

-0.42

RegressionTest

-0.41

POSITIVE LOGITS

 вікісторінку

0.61

 houſe

0.52

 petition

0.50

 EconPapers

0.46

 fubject

0.45

 ſche

0.45

 workshop

0.44

ſelves

0.43

 Houſe

0.43

bibfield

0.42

Activations Density 1.621%

references to instances or entities being discussed in the text

No Comments

No Known Activations