INDEX

Explanations

mentions of contributors and editors in a publication context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ummer

-0.18

deen

-0.18

DNA

-0.14

imensional

-0.14

ension

-0.14

ebi

-0.14

ona

-0.14

arger

-0.14

gaard

-0.14

stell

-0.13

POSITIVE LOGITS

etc

0.15

etc

0.14

 ÙĪØ¬Ùĩ

0.14

Hub

0.14

 DEST

0.13

Ð°ÑģÑĤÐ¸

0.13

elin

0.13

ÏģÎ±Î½

0.13

ìŀ

0.13

 rotating

0.13

Activations Density 0.015%

mentions of contributors and editors in a publication context

No Comments

No Known Activations