INDEX

Explanations

references to various articles and publications, particularly in a scholarly or academic context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

utter

-0.14

Į¨

-0.14

á»Ļc

-0.14

mani

-0.13

ê¸ī

-0.13

oley

-0.13

_CLOCK

-0.13

Fen

-0.13

ummy

-0.13

igr

-0.13

POSITIVE LOGITS

wiki

0.20

 sources

0.19

herits

0.19

etooth

0.19

Wik

0.18

Wiki

0.17

 hete

0.16

 baÄŁlantÄ±lar

0.16

 wiki

0.16

_Lean

0.16

Activations Density 0.181%

references to various articles and publications, particularly in a scholarly or academic context

No Comments

No Known Activations