INDEX

Explanations

terms related to safety and biocompatibility in materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

WithMany

-0.58

ConstraintMaker

-0.57

eradish

-0.55

bibinfo

-0.54

icep

-0.54

 مرئيه

-0.53

 Exactos

-0.52

avajillas

-0.52

 >=",

-0.52

CodeAttribute

-0.52

POSITIVE LOGITS

 harmless

0.62

 ramah

0.60

 harmful

0.60

vironment

0.57

 safe

0.56

otoro

0.55

 environmentally

0.53

harm

0.52

\{\\

0.52

Espèce

0.52

Activations Density 0.018%

terms related to safety and biocompatibility in materials

No Comments

No Known Activations