INDEX

Explanations

phrases indicating obligations, agreements, and conditions in discussions or transactions

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

agreed to, admitted that, decided whether

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

DataAnnotations

-0.59

EDEFAULT

-0.53

MigrationBuilder

-0.52

kwds

-0.50

noscript

-0.50

HandlerContext

-0.49

AccessorTable

-0.47

❹

-0.47

obé

-0.47

parsedMessage

-0.47

POSITIVE LOGITS

WriteTagHelper

0.43

 HttpNotFound

0.32

SuspendLayout

0.31

 exame

0.31

 tarifas

0.29

記事

0.29

DNEY

0.28

Save

0.28

rawdę

0.28

 AssemblyCulture

0.27

Activations Density 0.058%

phrases indicating obligations, agreements, and conditions in discussions or transactions

agreed to, admitted that, decided whether

No Comments

No Known Activations