INDEX

Explanations

references to film or media content

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

ese

-0.17

ses

-0.15

 Wander

-0.15

ish

-0.14

rus

-0.14

co

-0.14

ÙĬÙĩ

-0.14

lou

-0.14

esi

-0.14

POSITIVE LOGITS

unma

0.15

typings

0.15

umlu

0.15

krom

0.15

arcer

0.14

mlink

0.14

ammo

0.14

omik

0.14

Ã¡bado

0.14

endir

0.14

Activations Density 0.014%

references to film or media content

No Comments

No Known Activations