INDEX

Explanations

references to various forms of classification and status

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

avana

-0.15

tut

-0.15

 Cummings

-0.14

ardin

-0.14

Na

-0.14

_alt

-0.14

enso

-0.14

cou

-0.14

ainen

-0.14

Na

-0.14

POSITIVE LOGITS

aight

0.19

ought

0.17

ught

0.16

__.__

0.15

amment

0.15

Ä°Y

0.15

 rencont

0.15

ektir

0.15

/Dk

0.15

inky

0.14

Activations Density 0.986%

references to various forms of classification and status

No Comments

No Known Activations