INDEX

Explanations

references to historical contexts or subjects

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

icz

-0.18

thon

-0.17

Rim

-0.16

er

-0.16

 chute

-0.16

ç©´

-0.15

ÐºÐ°Ð²

-0.15

IDS

-0.15

Î¹Î¸

-0.15

erp

-0.15

POSITIVE LOGITS

ories

0.25

amine

0.24

idine

0.23

inct

0.20

opath

0.19

eam

0.19

amines

0.19

 Hist

0.18

Ã³ria

0.17

 hist

0.17

Activations Density 0.020%

references to historical contexts or subjects

No Comments

No Known Activations