INDEX

Explanations

phrases indicating personal information and life details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

DATED

-0.17

arat

-0.15

\grid

-0.15

AMESPACE

-0.15

esub

-0.14

InvalidOperationException

-0.14

jvu

-0.14

isay

-0.14

Pour

-0.14

ITT

-0.14

POSITIVE LOGITS

eger

0.18

 Maxim

0.16

istle

0.15

InBackground

0.15

opup

0.14

 Clash

0.14

 Clear

0.14

pec

0.14

 Branch

0.14

Activations Density 0.128%

phrases indicating personal information and life details

No Comments

No Known Activations