INDEX

Explanations

references to user interactions and page redirection in a programming context

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

verifyException

-0.65

Autoritní

-0.59

OGND

-0.57

 &___

-0.55

下载附件

-0.53

exitRule

-0.53

 שוליים

-0.52

 nakalista

-0.52

 mijne

-0.51

 utafitiHapana

-0.51

POSITIVE LOGITS

WebControls

0.45

 defaultstate

0.37

syste

0.36

 olet

0.35

sentinel

0.35

 Ponte

0.33

thums

0.32

 τά

0.32

nezeu

0.31

BSITE

0.31

Activations Density 0.010%

references to user interactions and page redirection in a programming context

No Comments

No Known Activations