INDEX

Explanations

adjectives indicating positive qualities or attributes

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

something followed by an adjective

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

 disambiguazione

-0.50

Ārējās

-0.48

 Pozna

-0.47

 térm

-0.47

haikusbot

-0.46

 invokingState

-0.45

 rând

-0.44

enderror

-0.44

multicolumn

-0.44

 parís

-0.43

POSITIVE LOGITS

 Something

0.80

something

0.79

Something

0.74

 SOMETHING

0.70

ETHING

0.68

anything

0.64

 something

0.63

 THING

0.61

thing

0.60

 Anything

0.60

Activations Density 0.043%

adjectives indicating positive qualities or attributes

something followed by an adjective

No Comments

No Known Activations