INDEX

Explanations

words indicating frequency or predominance

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

mostly or mainly

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

HasForeignKey

-0.65

 Administrativna

-0.59

 Drog

-0.52

 JpaRepository

-0.51

TemporalType

-0.51

Produzione

-0.50

 NSCoder

-0.50

MessageTagHelper

-0.49

 resourceCulture

-0.49

Cock

-0.48

POSITIVE LOGITS

Mostly

1.08

 Mostly

1.03

Mainly

0.93

 mostly

0.91

mostly

0.88

 Mainly

0.87

 mainly

0.82

mainly

0.73

 chiefly

0.67

 Primarily

0.67

Activations Density 0.015%

words indicating frequency or predominance

mostly or mainly

No Comments

No Known Activations