INDEX

Explanations

references to judges, legal proceedings, and courtroom terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 ujednoznacz

-0.54

deaths

-0.46

 '\\;'

-0.41

 Privacidad

-0.41

ppuden

-0.40

 réaliste

-0.39

 suerte

-0.39

 držav

-0.38

balenciaga

-0.37

 garantías

-0.37

POSITIVE LOGITS

évaluateur

0.53

ApiOperation

0.53

AccessorTable

0.50

 iprot

0.48

WithIOException

0.47

MessageOf

0.47

RTLR

0.45

pektor

0.44

 espont

0.43

reich

0.43

Activations Density 0.049%

references to judges, legal proceedings, and courtroom terminology

No Comments

No Known Activations