INDEX

Explanations

references to royal entities, institutions, or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

arness

-0.17

brig

-0.16

Ð¾Ð´ÐµÑĢÐ¶

-0.15

Ø±Ø§ÙĤ

-0.14

 Ð´Ð¾Ð¿

-0.14

ĤŃ

-0.14

 addslashes

-0.14

ÏĨÏħ

-0.14

NSMutable

-0.14

MainThread

-0.13

POSITIVE LOGITS

TY

0.19

Flush

0.18

izing

0.18

ton

0.18

bum

0.17

flush

0.17

 Flush

0.17

ising

0.17

 flush

0.17

-family

0.16

Activations Density 0.011%

references to royal entities, institutions, or titles

No Comments

No Known Activations