INDEX

Explanations

instances where certain elements or features are prominently highlighted or emphasized, possibly in technical or detailed contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

nep

-0.58

 constitutions

-0.58

subplots

-0.57

ifiquement

-0.57

ulgação

-0.55

 MonoBehaviour

-0.52

arXiv

-0.51

findpost

-0.51

amboo

-0.51

 PackageManager

-0.51

POSITIVE LOGITS

 مشين

0.90

tagHelperRunner

0.73

giphy

0.60

 חיצוניים

0.59

ArrowToggle

0.59

 ویکی‌پدیا

0.57

 دیکھیے

0.57

 nahilalakip

0.57

tanleria

0.56

 Вікі

0.55

Activations Density 0.212%

instances where certain elements or features are prominently highlighted or emphasized, possibly in technical or detailed contexts

No Comments

No Known Activations