INDEX

Explanations

references to competitive outcomes and rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ften

-0.17

lico

-0.15

æħİ

-0.14

opper

-0.14

otlin

-0.14

amel

-0.14

eyen

-0.14

otal

-0.14

 DropIndex

-0.13

keh

-0.13

POSITIVE LOGITS

 beat

0.43

 overt

0.37

 beating

0.35

 beaten

0.34

 beats

0.33

beat

0.32

 Beat

0.32

Beat

0.32

 outs

0.30

sup

0.29

Activations Density 0.240%

references to competitive outcomes and rankings

No Comments

No Known Activations