INDEX

Explanations

references to groups or collectives of people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

 èı²å¾ĭå®¾

-0.17

ogs

-0.15

fet

-0.14

Ã¶s

-0.14

ãĥ¼ãĥĪ

-0.14

agn

-0.14

iland

-0.14

aan

-0.14

=YES

-0.13

cks

-0.13

POSITIVE LOGITS

 staff

0.21

 audience

0.18

 Congress

0.18

 crew

0.17

 families

0.17

Î¹ÎºÎ¿

0.16

elves

0.16

 team

0.16

 crews

0.16

 workforce

0.15

Activations Density 0.057%

references to groups or collectives of people

No Comments

No Known Activations