INDEX

Explanations

terms associated with academic research and waterproofing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

ParallelGroup

-0.63

lossians

-0.57

ա

-0.53

 specchio

-0.52

RequestParam

-0.51

atott

-0.50

enumi

-0.50

 Dominicana

-0.49

abetes

-0.49

 Rosen

-0.48

POSITIVE LOGITS

 Thesis

0.74

 nakalista

0.73

InjectAttribute

0.70

 THESIS

0.70

 timestamps

0.68

 redraw

0.68

 thesis

0.68

 coales

0.66

NVIC

0.64

WER

0.64

Activations Density 0.036%

terms associated with academic research and waterproofing

No Comments

No Known Activations