INDEX

Explanations

references to political or social entities and their actions

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

izr

-0.17

okol

-0.16

ylko

-0.15

enet

-0.15

Ã¼c

-0.14

çģ

-0.14

olley

-0.14

 commercially

-0.13

 mest

-0.13

eph

-0.13

POSITIVE LOGITS

aga

0.15

.Mail

0.15

 Hatch

0.15

Ecc

0.15

iaux

0.14

LS

0.14

íİ¸

0.14

fÃ¼g

0.14

Ð¾Ð±Ñĭ

0.14

æ¡£

0.14

Activations Density 0.561%

references to political or social entities and their actions

No Comments

No Known Activations