INDEX

Explanations

discrepancies in reporting or records related to incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

.scalablytyped

-0.17

á»ĩ

-0.15

ijken

-0.15

ÚĨØ§Ø±

-0.15

ÑĦÐ¾ÑĢÑĤ

-0.14

egal

-0.14

lag

-0.14

Ð¶ÐµÐ½ÑĮ

-0.14

urgeon

-0.14

ckpt

-0.14

POSITIVE LOGITS

 lockdown

0.36

 lock

0.34

 Lock

0.32

 LOCK

0.29

Lock

0.28

 locker

0.27

 locks

0.27

lock

0.27

 locking

0.27

-lock

0.27

Activations Density 0.035%

discrepancies in reporting or records related to incidents

No Comments

No Known Activations