INDEX

Explanations

references to media and entertainment news

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

iam

-0.16

 opak

-0.16

aggio

-0.15

ese

-0.14

urch

-0.14

Interior

-0.14

Rai

-0.14

ë©ĺ

-0.14

weeney

-0.14

iew

-0.14

POSITIVE LOGITS

#af

0.17

BOOLEAN

0.15

.createClass

0.15

 undef

0.15

CLOCKS

0.14

 Loved

0.14

ertz

0.13

vyk

0.13

ETF

0.13

pcodes

0.13

Activations Density 0.010%

references to media and entertainment news

No Comments

No Known Activations