INDEX

Explanations

terms and phrases indicating uncertainty or the presence of multiple options

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

quantifiers and specific quantities

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 partea

-0.50

 spôso

-0.47

combination

-0.47

 tentativo

-0.46

 combinação

-0.44

 combination

-0.44

 obdob

-0.43

 njia

-0.43

 possibilidade

-0.43

 Portion

-0.43

POSITIVE LOGITS

parsedMessage

0.59

قایناقلار

0.57

DockStyle

0.54

ViewInit

0.52

 MainAxisSize

0.52

 تانيه

0.50

RefNanny

0.49

tagHelper

0.48

__":

0.48

 Audiodateien

0.47

Activations Density 0.267%

terms and phrases indicating uncertainty or the presence of multiple options

quantifiers and specific quantities

No Comments

No Known Activations