INDEX

Explanations

code-related elements, specifically those involving hash codes and session management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ë§ī

-0.18

ALAR

-0.16

adel

-0.15

InnerText

-0.15

 Cipher

-0.15

oner

-0.15

jabi

-0.15

 decryption

-0.14

ëħ

-0.14

ToWorld

-0.14

POSITIVE LOGITS

Collision

0.22

 Collision

0.21

 collisions

0.21

 Digest

0.19

 collision

0.19

 seed

0.18

hexdigest

0.18

 digest

0.17

Digest

0.17

 signature

0.17

Activations Density 0.038%

code-related elements, specifically those involving hash codes and session management

No Comments

No Known Activations