INDEX

Explanations

technical terms related to attributes and structural components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 å¾Ĵ

-0.17

cÃ¼

-0.16

å¾Ĵ

-0.15

.intellij

-0.15

 muschi

-0.15

etooth

-0.14

ForMember

-0.14

atel

-0.14

.datatables

-0.14

éĿ

-0.14

POSITIVE LOGITS

emez

0.16

ies

0.15

 Awake

0.15

 Clara

0.14

less

0.14

iran

0.14

esse

0.14

ss

0.14

ecs

0.14

 Bieber

0.14

Activations Density 0.003%

technical terms related to attributes and structural components

No Comments

No Known Activations