INDEX

Explanations

phrases relating to scientific publications and their authors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 clip

-0.15

Clip

-0.14

istant

-0.14

IE

-0.14

ennes

-0.14

 Clip

-0.13

caff

-0.13

 surv

-0.13

RoleId

-0.13

clip

-0.13

POSITIVE LOGITS

 relev

0.15

uras

0.15

inventory

0.15

usan

0.14

arcy

0.14

veh

0.14

gle

0.14

ncy

0.14

psc

0.14

 benim

0.14

Activations Density 0.004%

phrases relating to scientific publications and their authors

No Comments

No Known Activations