INDEX

Explanations

references to document and file formats, particularly related to eBooks and proprietary technologies

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

pdf

-0.17

Pdf

-0.17

ç²Ĺ

-0.16

pdf

-0.15

Pdf

-0.15

enne

-0.15

 acrylic

-0.15

png

-0.14

PDF

-0.14

Dup

-0.14

POSITIVE LOGITS

 format

0.42

æł¼å¼ı

0.35

 Format

0.32

 formats

0.32

format

0.32

-format

0.31

	format

0.28

Format

0.28

 FORMAT

0.27

_format

0.27

Activations Density 0.078%

references to document and file formats, particularly related to eBooks and proprietary technologies

No Comments

No Known Activations