INDEX

Explanations

references to requests or demands for assistance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

eroon

-0.18

ellen

-0.17

asant

-0.17

uju

-0.16

altar

-0.16

imbus

-0.16

>{!!

-0.15

zung

-0.14

irie

-0.14

arity

-0.14

POSITIVE LOGITS

Ali

0.27

Ab

0.25

Ab

0.25

Ali

0.24

 Said

0.23

ali

0.23

Mus

0.21

Abd

0.20

Yun

0.19

Isa

0.19

Activations Density 0.308%

references to requests or demands for assistance

No Comments

No Known Activations