INDEX

Explanations

phrases indicating uncertainty or degrees of possibility

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

qualifiers and adverbs

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

ViewFeatures

-0.45

 vectorielles

-0.43

protoimpl

-0.43

 rents

-0.43

vias

-0.42

ensi

-0.41

 marrone

-0.41

ทีม

-0.41

Gön

-0.40

 Guadal

-0.40

POSITIVE LOGITS

Personendaten

0.46

cloudflare

0.45

UserScript

0.41

GIVEREF

0.39

licherweise

0.39

EDEFAULT

0.35

 betweenstory

0.35

 ✭✭

0.35

acular

0.35

 '*')

0.35

Activations Density 0.092%

phrases indicating uncertainty or degrees of possibility

qualifiers and adverbs

No Comments

No Known Activations