INDEX

Explanations

HTML list elements

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

/Table

-0.20

 tables

-0.18

 Tables

-0.17

 table

-0.17

 Table

-0.16

etable

-0.16

 TABLE

-0.15

.='

-0.15

(table

-0.14

 @"↵

-0.14

POSITIVE LOGITS

li

0.38

li

0.34

<li

0.31

.li

0.28

Li

0.28

/li

0.28

_li

0.26

Li

0.25

LI

0.25

-li

0.24

Activations Density 0.051%

HTML list elements

No Comments

No Known Activations

HTML list elements

No Comments

No Known Activations