INDEX

Explanations

references to locations and positional phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ÐµÑĢÐ°Ð»ÑĮ

-0.19

Ð½ÑĮÐ¾Ð¼Ñĥ

-0.17

kea

-0.17

 Ð½Ð°ÑĤÑĥ

-0.17

aggi

-0.17

lÃ©m

-0.16

ECH

-0.16

kÃ¡m

-0.16

Ã©m

-0.16

enum

-0.15

POSITIVE LOGITS

ÑīÐµÐ¹

0.27

ÐºÐ¾Ð¹

0.27

ÑĨÐ¸ÐµÐ¹

0.26

jÄħ

0.25

ÑĢÐ¾Ð¹

0.24

ÑĨÐµÐ¹

0.24

Ð»Ð¾Ð¹

0.23

ÐºÐ¾Ñİ

0.23

Ð¾Ð¹

0.23

ÑĢÐ¾Ñİ

0.23

Activations Density 0.017%

references to locations and positional phrases

No Comments

No Known Activations