INDEX

Explanations

specific references to technical terms and educational contexts

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

before technical terms

np_acts-logits-general · gemini-2.0-flash

technical and programming terms

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

يميديا

-0.40

CloseOperation

-0.37

verwijspagina

-0.37

LLocation

-0.36

oa̍t

-0.35

Ārējās

-0.34

回报

-0.33

 ligiloj

-0.32

BagConstraints

-0.31

sof

-0.31

POSITIVE LOGITS

帖最后由

0.60

 CreateTagHelper

0.59

󠁴

0.57

 оригіналу

0.56

fromnode

0.48

audiovisuel

0.47

scryfall

0.46

Ӕ

0.46

Rujuakan

0.46

 equivalently

0.46

Activations Density 1.534%

specific references to technical terms and educational contexts

before technical terms

technical and programming terms

No Comments

No Known Activations