INDEX

Explanations

references to news content and related links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ves

-0.15

rooms

-0.14

 Village

-0.14

ERR

-0.14

CAPE

-0.14

luvÃŃ

-0.13

 Mech

-0.13

achts

-0.13

 ëĶ°ë¥¸

-0.13

licit

-0.13

POSITIVE LOGITS

Ð°Ð´ÐºÐ¸

0.15

Palette

0.15

åº§

0.14

.setViewport

0.14

imin

0.14

²

0.14

gratis

0.14

èĸĦ

0.14

weapon

0.14

asar

0.14

Activations Density 0.085%

references to news content and related links

No Comments

No Known Activations