INDEX

Explanations

phrases indicating uncertainty or speculation

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

apparently factual statements

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

 confronti

-0.56

ษา

-0.54

ScopeManager

-0.52

eradish

-0.50

psack

-0.50

Renewal

-0.48

 Drink

-0.48

 HasFactory

-0.47

Dog

-0.47

Knife

-0.47

POSITIVE LOGITS

 apparently

0.84

apparently

0.82

Apparently

0.80

 Apparently

0.77

cheinend

0.56

 aparentemente

0.52

 presumably

0.51

umably

0.51

AppData

0.51

 evidently

0.50

Activations Density 0.009%

phrases indicating uncertainty or speculation

apparently factual statements

No Comments

No Known Activations