INDEX

Explanations

references to locations and events in a specific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

odont

-0.16

illus

-0.15

caret

-0.15

ikat

-0.15

WV

-0.15

íĥģ

-0.15

 Chern

-0.14

Ãłu

-0.14

 Injectable

-0.14

arak

-0.14

POSITIVE LOGITS

SD

0.40

SD

0.40

SDS

0.39

San

0.38

 Padres

0.36

 Qualcomm

0.33

sd

0.32

 Chargers

0.32

_SD

0.31

San

0.30

Activations Density 0.029%

references to locations and events in a specific context

No Comments

No Known Activations