INDEX

Explanations

keywords and phrases related to lists and identification systems

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 transfieras

-0.81

sidemargin

-0.79

 Infórmanos

-0.78

Skocz

-0.74

 AssemblyCompany

-0.72

+#+#

-0.69

)*/

-0.68

 ujednoznacz

-0.65

verwijspagina

-0.65

 Exacts

-0.65

POSITIVE LOGITS

ėse

0.61

 hydrated

0.57

 massless

0.55

 helicopter

0.54

 efectivos

0.52

 automatiques

0.52

 riego

0.52

 kwiaty

0.51

́c

0.50

 hydrat

0.49

Activations Density 1.357%

keywords and phrases related to lists and identification systems

No Comments

No Known Activations