INDEX

Explanations

references to parents and parental figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ramework

-0.17

.BorderFactory

-0.16

rd

-0.15

rious

-0.15

zier

-0.14

artin

-0.14

fall

-0.14

ty

-0.14

ards

-0.14

curity

-0.14

POSITIVE LOGITS

-child

0.24

è¦ª

0.22

age

0.21

preneur

0.20

eral

0.19

aced

0.17

prene

0.17

 Superior

0.16

ationally

0.16

joy

0.15

Activations Density 0.035%

references to parents and parental figures

No Comments

No Known Activations