INDEX

Explanations

instances of explanatory phrases regarding testing scenarios or configurations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_19/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.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

ValueStyle

-0.73

Jereo

-0.71

 发表于

-0.70

Tikang

-0.69

 protoimpl

-0.67

存于互联网档案馆

-0.67

 shitty

-0.65

 FUCKING

-0.64

AccessorTable

-0.64

fucking

-0.63

POSITIVE LOGITS

 becauſe

0.65

 houſe

0.65

 pleaſure

0.64

 neceſſ

0.63

ſy

0.61

 whoſe

0.60

 ſtand

0.59

 uſe

0.58

 ſtate

0.58

 ſon

0.58

Activations Density 0.529%

instances of explanatory phrases regarding testing scenarios or configurations

No Comments

No Known Activations