INDEX

Explanations

information extraction and communication processes

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

preceding "information"

np_acts-logits-general · gemini-2.0-flash

provides information

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InjectAttribute

-0.69

PhysRev

-0.59

ⓧ

-0.58

LookAnd

-0.55

SOUNDBITE

-0.55

 CreateTagHelper

-0.54

Scénario

-0.53

 IBOutlet

-0.49

fehlung

-0.48

expandindo

-0.48

POSITIVE LOGITS

 information

2.41

information

1.97

 info

1.82

 información

1.70

 informations

1.64

 Information

1.62

Information

1.60

 INFORMATION

1.60

 informatie

1.58

 informasi

1.52

Activations Density 0.695%

information extraction and communication processes

preceding "information"

provides information

No Comments

No Known Activations