INDEX

Explanations

references to technical specifications or codes related to software or hardware components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/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.15.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

GraphicsUnit

-0.73

 invokingState

-0.66

__':

-0.61

 كومونز

-0.60

__':

-0.58

UnsafeEnabled

-0.57

__":

-0.55

SpringRunner

-0.55

 gespannt

-0.54

setVerticalGroup

-0.53

POSITIVE LOGITS

 disambiguazione

0.59

 kneeling

0.56

batore

0.51

>{"

0.51

ServiceTest

0.49

UserRepository

0.49

 ibid

0.48

сылкі

0.48

Hero

0.47

 antelope

0.47

Activations Density 0.579%

references to technical specifications or codes related to software or hardware components

No Comments

No Known Activations