INDEX

Explanations

phrases that describe characteristics or features of subjects, particularly in a scientific or technical context

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

characterized by

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

 Anſ

-0.52

 Reſ

-0.52

znamu

-0.48

 myſelf

-0.47

ECONDS

-0.47

spun

-0.47

igrette

-0.47

 bombe

-0.46

ſelves

-0.46

CreateMap

-0.46

POSITIVE LOGITS

 characteristic

0.83

 característica

0.77

 caracter

0.68

 caratter

0.66

特徴

0.66

 caracteriza

0.65

characteristic

0.65

 Characteristic

0.64

 caractéristique

0.63

 charakterysty

0.63

Activations Density 0.084%

phrases that describe characteristics or features of subjects, particularly in a scientific or technical context

characterized by

No Comments

No Known Activations