INDEX

Explanations

references to custom element APIs in programming documentation

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

 disambiguazione

-0.77

Datuak

-0.66

+#+

-0.64

ⓧ

-0.64

elemField

-0.63

 GenerationType

-0.60

 ujednoznacz

-0.59

出版年

-0.58

 Infórmanos

-0.56

oredCriteria

-0.56

POSITIVE LOGITS

uren

0.36

 Described

0.35

 Neurology

0.35

似乎

0.34

|[

0.34

لاق

0.34

ardom

0.34

getOne

0.33

っちり

0.33

ノリ

0.33

Activations Density 0.024%

references to custom element APIs in programming documentation

No Comments

No Known Activations