INDEX

Explanations

statements that express uncertainty or declare something unexpected

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

this is

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

 enfans

-0.57

 Italij

-0.55

 Königin

-0.50

 ſtand

-0.49

 siinä

-0.48

 récents

-0.48

jenigen

-0.48

 käytt

-0.47

 Flasche

-0.47

 esprits

-0.46

POSITIVE LOGITS

 للمعارف

0.51

 ['./

0.51

########.

0.46

 useAppContext

0.45

 Bruno

0.44

 announce

0.43

 happening

0.43

 embarrassing

0.40

 THIS

0.40

Gai

0.40

Activations Density 0.041%

statements that express uncertainty or declare something unexpected

this is

No Comments

No Known Activations