INDEX

Explanations

phrases that express complexity and depth, often referencing systems or constructs with strong ideological or emotional underpinnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

oris

-0.15

 Fancy

-0.15

vertisement

-0.14

ìĿ´íĦ°

-0.14

TERM

-0.14

aned

-0.14

-0.13

Ŀ

-0.13

ErrorException

-0.13

å±¥

-0.13

POSITIVE LOGITS

avra

0.17

à¸¸à¸¢

0.15

apo

0.15

 AltÄ±n

0.14

++↵

0.14

-alist

0.14

Bak

0.13

 alphabet

0.13

 series

0.13

oint

0.13

Activations Density 0.225%

phrases that express complexity and depth, often referencing systems or constructs with strong ideological or emotional underpinnings

No Comments

No Known Activations