INDEX

Explanations

words and phrases indicating causal relationships and dependencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

aarrggbb

-0.87

/\.

-0.76

DoubleQuotes

-0.74

<bos>

-0.70

 TestBed

-0.61

はじめに

-0.61

хьтан

-0.59

norsk

-0.56

rsiniz

-0.56

 /\.(

-0.55

POSITIVE LOGITS

</caption>

0.85

 الرغم

0.76

ledem

0.75

請繼續往下閱讀

0.72

 ressemble

0.70

dientemente

0.69

 través

0.69

 Profitez

0.68

\{\\

0.68

quartered

0.68

Activations Density 1.338%

words and phrases indicating causal relationships and dependencies

No Comments

No Known Activations