INDEX

Explanations

keywords related to JSON data structures and processes

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

government domain names

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

 utafitiHapana

-0.68

 betweenstory

-0.67

 MainAxisSize

-0.66

DIPSETTING

-0.64

 المعيارى

-0.61

AndEndTag

-0.59

 CreateTagHelper

-0.59

+#+#

-0.56

 <<<<<<<<<<<<<<

-0.56

RenderAtEndOf

-0.56

POSITIVE LOGITS

 nahilalakip

0.44

json

0.43

исленность

0.39

haj

0.37

MessageTagHelper

0.37

PJ

0.35

 عص

0.34

 Mara

0.34

rø

0.34

audiovisuel

0.34

Activations Density 0.000%

keywords related to JSON data structures and processes

government domain names

No Comments

No Known Activations