INDEX

Explanations

instances of structured data notations or formatting elements

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

IFI

-0.16

@section

-0.16

rh

-0.15

 æ¯

-0.15

.repository

-0.14

åįĵ

-0.13

_arrays

-0.13

Ã©ru

-0.13

Ã²ng

-0.13

ActiveSheet

-0.13

POSITIVE LOGITS

item

0.45

 item

0.36

-item

0.29

 Item

0.28

.item

0.28

Item

0.27

_item

0.25

items

0.24

ITEM

0.24

 items

0.24

Activations Density 0.022%

instances of structured data notations or formatting elements

No Comments

No Known Activations