INDEX

Explanations

phrases that express uncertainty or seek clarification

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

exactness and precision

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

 pinulongan

-0.61

 Muscat

-0.45

 another

-0.43

almo

-0.42

Phone

-0.42

anair

-0.42

rifice

-0.41

 econó

-0.41

drawal

-0.40

 demurrer

-0.40

POSITIVE LOGITS

precise

0.65

 exactly

0.60

Precise

0.57

particular

0.56

 precisely

0.56

 Exactly

0.55

Exactly

0.54

 دقیق

0.54

exact

0.54

exactly

0.54

Activations Density 0.018%

phrases that express uncertainty or seek clarification

exactness and precision

No Comments

No Known Activations