INDEX

Explanations

references to numerical data, specifically in the context of academic or scientific citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 &___

-0.78

 ſever

-0.77

 виправивши

-0.75

 pleaſure

-0.73

 ſeveral

-0.73

 fhew

-0.72

 laſt

-0.68

httphttps

-0.68

Хьажоргаш

-0.67

цездатний

-0.66

POSITIVE LOGITS

ParallelGroup

0.63

)|^{

0.55

новништво

0.53

0.51

İstinadlar

0.47

gdx

0.47

Rptr

0.47

FileChooser

0.46

ioutil

0.46

parseDouble

0.45

Activations Density 0.309%

references to numerical data, specifically in the context of academic or scientific citations

No Comments

No Known Activations