INDEX

Explanations

questions and requests for information or assistance

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

➯

-0.61

 CreateTagHelper

-0.57

AndEndTag

-0.56



-0.56

 Winaray

-0.52

bootstrapcdn

-0.52

wahati

-0.51

 zijne

-0.50

 arşivlendi

-0.50

RTLD

-0.50

POSITIVE LOGITS

ref

0.41

 cref

0.36

mens

0.36

bre

0.35

gu

0.35

AKES

0.35

xtype

0.34

口

0.34

Tun

0.33

bes

0.33

Activations Density 0.357%

questions and requests for information or assistance

No Comments

No Known Activations