INDEX

Explanations

references to legal or regulatory actions and processes involving organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

umlu

-0.15

cion

-0.15

 operative

-0.14

.oper

-0.14

ÑĢÐ°ÑīÐµÐ½Ð¸Ñı

-0.13

iations

-0.13

 Ramos

-0.13

EG

-0.13

ILD

-0.13

arsity

-0.13

POSITIVE LOGITS

use

0.18

 manufacture

0.16

 sale

0.16

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.15

oldown

0.15

 collapse

0.15

_likelihood

0.14

 asses

0.14

ape

0.14

 existence

0.14

Activations Density 0.211%

references to legal or regulatory actions and processes involving organizations

No Comments

No Known Activations