INDEX

Explanations

terms related to scientific measurements and classifications

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

M initials and two-letter tags

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

short uppercase alphabetic tokens — acronyms or initials (e.g., two‑letter/abbreviated scientific or name initials).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

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

setAccessible

-0.51

 proyec

-0.50

 proyección

-0.49

 tinto

-0.47

Quiénes

-0.47

getLogin

-0.45

 calcetines

-0.45

 você

-0.45

 cerah

-0.45

 getLogin

-0.45

POSITIVE LOGITS

0.91

0.76

mybatisplus

0.76

MH

0.76

0.73

 Mayhem

0.71

ConstraintMaker

0.71

MB

0.70

MC

0.70

MU

0.69

Activations Density 3.777%

terms related to scientific measurements and classifications

M initials and two-letter tags

short uppercase alphabetic tokens — acronyms or initials (e.g., two‑letter/abbreviated scientific or name initials).

No Comments

No Known Activations

terms related to scientific measurements and classifications

M initials and two-letter tags

short uppercase alphabetic tokens — acronyms or initials (e.g., two‑letter/abbreviated scientific or name initials).

No Comments

No Known Activations