INDEX

Explanations

instances of procedural instructions or descriptions related to food preparation and dining

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

conclusions or endings

np_acts-logits-general · gemini-2.0-flash

finish concluding ending

np_acts-logits-general · gemini-2.5-flash-lite

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

wiſe

-0.57

 GenerationType

-0.55

arschijnlijk

-0.55

Билгалдахарш

-0.55

MENAFN

-0.52

 pleaſure

-0.51

RODUZIONE

-0.51

 odkazy

-0.50

CGRectMake

-0.50

\{\\

-0.49

POSITIVE LOGITS

 ending

2.41

 conclude

2.15

 concluding

2.14

 concludes

2.09

 finishing

2.08

 ended

2.03

 finish

2.03

 ends

2.02

 concluded

2.01

 finishes

1.99

Activations Density 0.265%

instances of procedural instructions or descriptions related to food preparation and dining

conclusions or endings

finish concluding ending

No Comments

No Known Activations