INDEX

Explanations

references to conventional methods and systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

abestanden

-0.64

-0.59

 nakalista

-0.56

ScopeManager

-0.53

 surla

-0.52

ValueStyle

-0.51

Hochspringen

-0.51

 refusé

-0.49

enschappe

-0.49

POSITIVE LOGITS

GEBURTS

0.65

だけでは

0.58

Tazama

0.56

 nemmeno

0.56

 quelconque

0.55

databinding

0.55

enderror

0.55

SpringBootTest

0.52

cure

0.52

sete

0.51

Activations Density 0.785%

references to conventional methods and systems

No Comments

No Known Activations