INDEX

Explanations

mentions of NPR and related entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

-0.16

-0.15

opi

-0.15

 Lifestyle

-0.14

inha

-0.14

ahun

-0.14

lege

-0.14

 Lange

-0.14

hong

-0.14

more

-0.13

POSITIVE LOGITS

anium

0.15

$MESS

0.15

oulos

0.15

ÎµÎ¯

0.14

Ã©c

0.14

utzer

0.14

 Ð°Ð´Ð¶Ðµ

0.14

.parallel

0.13

NullException

0.13

Ã©es

0.13

Activations Density 0.003%

mentions of NPR and related entities

No Comments

No Known Activations