INDEX

Explanations

terms related to martial arts and credentials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

é»

-0.18

.hadoop

-0.15

.flink

-0.15

END

-0.14

dog

-0.14

áº«n

-0.13

uslim

-0.13

uestos

-0.13

uml

-0.13

ingly

-0.13

POSITIVE LOGITS

 Erotik

0.15

Nit

0.15

.UnitTesting

0.14

 Prostit

0.14

.SizeType

0.14

ponent

0.14

ordum

0.14

printStats

0.14

anik

0.14

_unused

0.13

Activations Density 0.025%

terms related to martial arts and credentials

No Comments

No Known Activations