INDEX

Explanations

specific historical references and details related to architectural developments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

iben

-0.17

iteli

-0.16

iset

-0.15

ois

-0.15

à¸ĵ

-0.15

 Benedict

-0.14

inality

-0.14

dre

-0.14

çĶ²

-0.14

ÙĪØ§Ø¡

-0.14

POSITIVE LOGITS

 tomb

0.23

 Jama

0.22

 Tomb

0.21

Hum

0.19

 Friday

0.19

Hum

0.19

mas

0.18

tom

0.18

min

0.18

 Sher

0.17

Activations Density 0.148%

specific historical references and details related to architectural developments

No Comments

No Known Activations