INDEX

Explanations

terms related to openings and passages in mechanical or structural contexts

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

entry into openings

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

 hunne

-0.40

 HttpNotFound

-0.40

verläs

-0.39

IsPostBack

-0.38

BaseController

-0.38

 australis

-0.37

CascadeType

-0.37

 BaseActivity

-0.36

 Cartes

-0.36

saraba

-0.35

POSITIVE LOGITS

 hole

0.69

 Holes

0.65

hole

0.64

httphttps

0.64

 Hole

0.61

 holes

0.60

Hole

0.58

 отверсти

0.57

 HOLE

0.57

 loophole

0.57

Activations Density 0.074%

terms related to openings and passages in mechanical or structural contexts

entry into openings

No Comments

No Known Activations