INDEX

Explanations

proper nouns related to people, places, and brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 kernels

-0.17

azon

-0.17

ãĤ¾

-0.16

kernel

-0.15

odable

-0.14

quirrel

-0.14

ÑĢÐµÐ±

-0.14

keyword

-0.14

quito

-0.14

chef

-0.14

POSITIVE LOGITS

(K

0.25

SK

0.23

CK

0.22

SK

0.22

WK

0.21

IK

0.21

0.20

[K

0.20

CK

0.20

,K

0.19

Activations Density 0.154%

proper nouns related to people, places, and brands

No Comments

No Known Activations