INDEX

Explanations

terms related to identification and personal verification

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

身分確認、督促、事情聴取

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

ValueStyle

-0.54

erialized

-0.51

 exitos

-0.50

 exitoso

-0.50

 héroe

-0.50

 successful

-0.49

 giant

-0.47

 destination

-0.47

 blockbuster

-0.47

 celebrity

-0.47

POSITIVE LOGITS

chism

0.38

logrus

0.35

ApiModelProperty

0.35

clusal

0.34

parian

0.33

бяза

0.33

 öğ

0.33

ContextCompat

0.33

reamble

0.33

 &___

0.32

Activations Density 0.081%

terms related to identification and personal verification

身分確認、督促、事情聴取

No Comments

No Known Activations