INDEX

Explanations

terms related to health, wellness, and educational accomplishments

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

IRON

-0.17

 Everett

-0.16

 Peyton

-0.15

ÙĲÙħ

-0.14

rons

-0.14

Iron

-0.14

Ø´Ùħ

-0.14

Peer

-0.14

EG

-0.14

GE

-0.13

POSITIVE LOGITS

(cps

0.15

adesh

0.15

akter

0.14

:;↵

0.14

ULK

0.14

ogan

0.14

agues

0.14

ertext

0.14

Ã¸d

0.13

ungi

0.13

Activations Density 0.006%

terms related to health, wellness, and educational accomplishments

No Comments

No Known Activations