INDEX

Explanations

patterns in structured data, possibly related to identifiers or numerical sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 ligiloj

-0.62

 متعلقه

-0.55

TestUtil

-0.53

TestBed

-0.51

Viitteet

-0.51

Mis

-0.48

مصادر

-0.48

Földrajzportál

-0.47

iocene

-0.47

postmedia

-0.47

POSITIVE LOGITS

 okay

2.17

ok

2.03

OK

2.01

 OKAY

1.78

 alright

1.70

okay

1.64

 Okay

1.62

Ok

1.57

Ok

1.55

OK

1.55

Activations Density 0.075%

patterns in structured data, possibly related to identifiers or numerical sequences

No Comments

No Known Activations