INDEX

Explanations

references to legal terminology and conditions related to liability and loss

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

Code, criteria, or restrictions related to database queries

np_acts-logits-general · gemini-2.0-flash

add criterion

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

ChildIndex

-0.86

+#+#

-0.86

ſelf

-0.85

 BorderRadius

-0.84

 Monfieur

-0.82

HideFlags

-0.81

 raiſ

-0.78

دانشنامهٔ

-0.78

makeText

-0.77

AndEndTag

-0.77

POSITIVE LOGITS

addCriterion

0.89

Criteria

0.57

 Criteria

0.53

criterion

0.52

 criteria

0.49

')}}"

0.48

 Criterion

0.48

 criterion

0.48

)}-

0.47

Columns

0.46

Activations Density 0.005%

references to legal terminology and conditions related to liability and loss

Code, criteria, or restrictions related to database queries

add criterion

No Comments

No Known Activations