INDEX

Explanations

references to specific decades in relation to cultural or historical content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uter

-0.16

issen

-0.15

ultz

-0.15

utz

-0.14

Ð¸Ð¼Ð²

-0.14

iosa

-0.14

ieren

-0.14

Ø¹ÙĬ

-0.14

@testable

-0.13

abilit

-0.13

POSITIVE LOGITS

orex

0.15

 Orchard

0.14

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.14

 odds

0.14

 endregion

0.14

ÙĬØ§

0.14

0.13

Ð¼Ð¸

0.13

slaught

0.13

 incom

0.13

Activations Density 0.018%

references to specific decades in relation to cultural or historical content

No Comments

No Known Activations