INDEX

Explanations

connections and relationships between concepts or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Ð¶ÐµÐ½ÑĮ

-0.16

 persons

-0.15

zdy

-0.14

.fhir

-0.14

zin

-0.14

contres

-0.14

aryawan

-0.14

umar

-0.13

ongyang

-0.13

 zdarma

-0.13

POSITIVE LOGITS

 clients

0.52

 Clients

0.50

 client

0.49

clients

0.46

Clients

0.44

 Client

0.43

 klient

0.42

 CLIENT

0.41

client

0.41

-client

0.39

Activations Density 0.011%

connections and relationships between concepts or entities

No Comments

No Known Activations