INDEX

Explanations

code and queries related to searching and retrieving data

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

search and search results

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

 publicação

-0.42

 prefeitura

-0.42

órias

-0.40

 sonno

-0.40

 pemer

-0.40

cshtml

-0.39

 folha

-0.37

 význam

-0.36

 acabamento

-0.36

 leyenda

-0.36

POSITIVE LOGITS

 Савезне

0.57

MessageTagHelper

0.55

 Infórmanos

0.54

epam

0.53

 utafitiHapana

0.53

⤹

0.50

LEncoder

0.49

 cong

0.48

johtaja

0.47

Spoljašnje

0.47

Activations Density 0.008%

code and queries related to searching and retrieving data

search and search results

No Comments

No Known Activations