INDEX

Explanations

mentions of significant events or occurrences

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

ibel

-0.19

fen

-0.15

ens

-0.14

idak

-0.14

ÐµÐ»Ð¸Ðº

-0.14

esiz

-0.14

annes

-0.14

akhir

-0.14

annis

-0.13

 Mata

-0.13

POSITIVE LOGITS

 ever

0.23

-ever

0.20

 EVER

0.18

Ã³st

0.17

ever

0.16

atty

0.16

enberg

0.15

ume

0.15

.RunWith

0.14

 Ever

0.14

Activations Density 0.172%

mentions of significant events or occurrences

No Comments

No Known Activations