INDEX

Explanations

sequences of non-standard formatting or unusual textual patterns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

á»³

-0.15

ÑĥÐ½Ðº

-0.13

Lev

-0.13

Unused

-0.13

 Wise

-0.13

itarian

-0.13

Tut

-0.13

finity

-0.12

(Op

-0.12

 Levine

-0.12

POSITIVE LOGITS

0.49

0.28

âĤĢ

0.28

 zero

0.27

ï¼Ĳ

0.25

Û°

0.24

à¥¦

0.24

0.23

éĽ¶

0.22

Ùł

0.22

Activations Density 0.196%

sequences of non-standard formatting or unusual textual patterns

No Comments

No Known Activations