INDEX

Explanations

words and phrases related to randomness and programming constructs

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CloseOperation

-0.83

Geplaatst

-0.73

 GenerationType

-0.69

 invokingState

-0.68

\{\\

-0.66

 ویکی‌پدی

-0.65

Tikang

-0.65

parsedMessage

-0.62

 мәкал

-0.62

 snippetHide

-0.61

POSITIVE LOGITS

lorem

0.37

</blockquote>

0.35

 machine

0.35

 random

0.35

 lact

0.35

┛

0.34

 filtered

0.34

 software

0.34

 varia

0.34

 cognitive

0.33

Activations Density 0.002%

words and phrases related to randomness and programming constructs

No Comments

No Known Activations