INDEX

Explanations

phrases related to historical events and legislative actions

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

abwe

-0.18

tae

-0.17

 Ø§Ø°

-0.17

ksiyon

-0.15

idal

-0.15

late

-0.15

egra

-0.14

 novÃ©ho

-0.14

lds

-0.14

wing

-0.13

POSITIVE LOGITS

 almost

0.20

 nearly

0.20

 century

0.19

 Cent

0.19

 cent

0.18

 years

0.18

 decades

0.17

Cent

0.17

almost

0.17

 Almost

0.16

Activations Density 0.063%

phrases related to historical events and legislative actions

No Comments

No Known Activations