INDEX

Explanations

references to documents and their status within a text

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

lewati

-0.56

 zasi

-0.56

 Turquía

-0.56

 Konkurrenz

-0.53

 popularidad

-0.52

 espoir

-0.51

 pitié

-0.50

 mesmas

-0.50

 incendie

-0.50

 jälkeen

-0.49

POSITIVE LOGITS

 Document

1.26

Document

1.16

 document

1.16

document

1.12

getDocument

0.96

 documents

0.95

 DOCUMENT

0.94

 Documents

0.93

DOCUMENT

0.90

ocument

0.87

Activations Density 0.084%

references to documents and their status within a text

No Comments

No Known Activations