INDEX

Explanations

elements related to recovery, restoration, and support

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

after "being" or "was" indicating a return to a prior state

np_acts-logits-general · gemini-2.0-flash

restored or regained

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

printLine

-0.44

спонд

-0.44

hearsed

-0.43

قيقة

-0.42

țion

-0.42

riba

-0.40

handleDelete

-0.40

reonine

-0.40

ựa

-0.38

ukup

-0.38

POSITIVE LOGITS

 restored

2.46

 regained

2.25

 restoration

2.24

 restoring

2.23

 restore

2.21

 regaining

2.19

 regain

2.17

 restores

2.16

 reinstatement

2.14

 reinstated

2.12

Activations Density 1.007%

elements related to recovery, restoration, and support

after "being" or "was" indicating a return to a prior state

restored or regained

No Comments

No Known Activations