INDEX

Explanations

negations and expressions of uncertainty

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

numbers and quantities

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

 Störungen

-0.40

urably

-0.35

scryfall

-0.34

 aank

-0.34

vaného

-0.33

 voort

-0.32

 Signalez

-0.32

otoro

-0.32

 zost

-0.30

 připo

-0.29

POSITIVE LOGITS

 EconPapers

0.68

MessageOf

0.63

 ProtoMessage

0.62

MigrationBuilder

0.56

$_['

0.56

usermodel

0.55

 misspelled

0.54

uxxxx

0.53

 pinulongan

0.53

kháu

0.52

Activations Density 0.344%

negations and expressions of uncertainty

numbers and quantities

No Comments

No Known Activations