INDEX

Explanations

references to patent applications and legal terms associated with them

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

patent application identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 '\\;'

-0.47

 resilience

-0.38

 emotional

-0.38

DockStyle

-0.38

 budget

-0.37

WriteLiteral

-0.37

 PeEnEo

-0.36

 vibes

-0.35

FunctionFlags

-0.35

 illusion

-0.35

POSITIVE LOGITS

 PyTuple

0.56

 HasFactory

0.47

readdir

0.44

msgTypes

0.44

 disclosed

0.44

الدراسه

0.44

andExpect

0.43

gelopen

0.42

Demografía

0.42

Geografía

0.42

Activations Density 0.122%

references to patent applications and legal terms associated with them

patent application identifiers

No Comments

No Known Activations