INDEX

Explanations

references to authors and their associated information in documents

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

tagHelperRunner

-0.88

parsedMessage

-0.79

Tembelea

-0.77

ロウィン

-0.75

RegressionTest

-0.73

ésultats

-0.68

 Infórmanos

-0.68

<unused14>

-0.67

<unused8>

-0.67

<pad>

-0.67

POSITIVE LOGITS

0.40

author

0.39

InjectMocks

0.36

0.34

ol

0.33

 author

0.33

 current

0.33

ep

0.32

vosti

0.32

de

0.32

Activations Density 0.001%

references to authors and their associated information in documents

No Comments

No Known Activations