INDEX

Explanations

comparative phrases related to averages and levels, particularly those that indicate being below normal or expected thresholds

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

after "below" or "above"

np_acts-logits-general · gemini-2.0-flash

above or below normal

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 fubject

-0.35

CreateMap

-0.35

.*")]

-0.35

értelmű

-0.34

 estekak

-0.31

 Roskov

-0.31

 cụ

-0.30

Specific

-0.30

vertx

-0.30

RTLR

-0.29

POSITIVE LOGITS

 normal

0.67

 StatelessWidget

0.63

 optimal

0.60

 normalen

0.59

normal

0.57

 optimum

0.57

 usual

0.56

 expected

0.56

 üblichen

0.53

genodigd

0.52

Activations Density 0.524%

comparative phrases related to averages and levels, particularly those that indicate being below normal or expected thresholds

after "below" or "above"

above or below normal

No Comments

No Known Activations