INDEX

Explanations

references to archival content and collections

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

avec

-0.18

 tiÃªn

-0.15

InSection

-0.15

 Jarvis

-0.15

iverse

-0.15

amel

-0.14

ä¸ĸ

-0.14

osh

-0.14

 breed

-0.14

è°±

-0.14

POSITIVE LOGITS

itect

0.17

ipel

0.16

-breaking

0.16

tvrt

0.15

yard

0.15

lds

0.15

.scalablytyped

0.15

yards

0.14

ripper

0.14

ulated

0.14

Activations Density 0.025%

references to archival content and collections

No Comments

No Known Activations