INDEX

Explanations

terms related to legal and procedural actions, particularly those involving restrictions or formalities

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

"prevent" and "impairment"

np_acts-logits-general · gemini-2.0-flash

preventing actions

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

AndEndTag

-0.49

MemoryWarning

-0.48

transQ

-0.47

MessageWindow

-0.46

WebServlet

-0.44

ViewFeatures

-0.44

BeginContext

-0.43

IUrlHelper

-0.43

 Wiktionnaire

-0.43

 useContext

-0.42

POSITIVE LOGITS

 prevent

2.02

 preventing

2.02

 prevented

1.93

 prevents

1.90

prevent

1.88

 Prevent

1.74

Prevent

1.73

 Preventing

1.70

Preventing

1.64

 prevention

1.61

Activations Density 0.838%

terms related to legal and procedural actions, particularly those involving restrictions or formalities

"prevent" and "impairment"

preventing actions

No Comments

No Known Activations