INDEX

Explanations

references to geographical locations and nationalities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

-0.16

al

-0.15

colo

-0.14

pad

-0.14

aint

-0.14

 Temper

-0.14

Ĩ

-0.14

etz

-0.14

.Forms

-0.14

zsche

-0.14

POSITIVE LOGITS

opia

0.17

isson

0.16

 domestic

0.16

/Framework

0.15

/mol

0.14

ABL

0.14

 nationalist

0.14

 native

0.14

à¸µà¸Ńà¸¢

0.14

'=>"

0.14

Activations Density 0.202%

references to geographical locations and nationalities

No Comments

No Known Activations