INDEX

Explanations

keywords related to programming tasks and actions

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

'@/

-0.49

çais

-0.49

omores

-0.47

ègre

-0.46

URLException

-0.45

vorous

-0.45

 Drapeau

-0.45

himovic

-0.44

nocześnie

-0.44

cerin

-0.44

POSITIVE LOGITS

 things

2.90

Things

2.45

 Things

2.38

things

2.35

 THINGS

2.29

 everything

2.13

everything

1.95

Everything

1.90

 Everything

1.90

 thing

1.80

Activations Density 0.449%

keywords related to programming tasks and actions

No Comments

No Known Activations