INDEX

Explanations

elements related to knowledge and information retrieval

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

remembering or knowing

np_acts-logits-general · gemini-2.0-flash

recalling specific information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

ditor

-0.61

addContainerGap

-0.57

 OFDb

-0.57

isinstance

-0.56

waves

-0.55

anzeige

-0.55

Портал

-0.54

 dimaksud

-0.53

onsored

-0.53

 Lasso

-0.53

POSITIVE LOGITS

 memorized

1.34

 memorize

1.24

 memor

0.96

 memori

0.88

 Memor

0.87

Memor

0.82

memor

0.78

 knowledge

0.76

 familiarize

0.71

knowledge

0.71

Activations Density 0.204%

elements related to knowledge and information retrieval

remembering or knowing

recalling specific information

No Comments

No Known Activations