INDEX

Explanations

phrases related to legislative changes and cancellations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

azzo

-0.18

amax

-0.16

hausen

-0.15

apolis

-0.14

ovah

-0.14

 ê°ĢëĬ¥íķľ

-0.14

mÃ®

-0.14

ÇĲ

-0.13

ç¬¬ä¸Ģæ¬¡

-0.13

ahlen

-0.13

POSITIVE LOGITS

 altogether

0.28

 existing

0.26

 remaining

0.25

 entire

0.20

 entirely

0.19

 previously

0.19

 planned

0.19

 previous

0.19

ban

0.18

 requirement

0.18

Activations Density 0.205%

phrases related to legislative changes and cancellations

No Comments

No Known Activations