INDEX

Explanations

specific evaluative metrics and assessments relating to comparisons and measurements in various contexts

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

gwin

-0.42

rigen

-0.39

IDAY

-0.38

erol

-0.38

oby

-0.38

DIPSETTING

-0.38

ringen

-0.36

CHOICES

-0.36

Observers

-0.35

OGND

-0.35

POSITIVE LOGITS

 وتسجيلات

0.62

новништво

0.55

 Italijanski

0.51

RTHOOK

0.49

تقاوى

0.49

 gynhyrchwyd

0.49

StructEnd

0.47

Билгалдахарш

0.46

ViewFeatures

0.46

 saites

0.45

Activations Density 0.118%

specific evaluative metrics and assessments relating to comparisons and measurements in various contexts

No Comments

No Known Activations