INDEX

Explanations

elements related to analysis and critique of evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

-0.15

 commercial

-0.14

gang

-0.14

aine

-0.14

 Commercial

-0.13

goo

-0.13

 simplest

-0.13

 frac

-0.13

dd

-0.13

POSITIVE LOGITS

 chapters

0.18

 chapter

0.18

 Chapters

0.17

 Chapter

0.17

amber

0.17

Reviewer

0.17

 reviewer

0.17

anlÄ±

0.16

chapter

0.16

 reader

0.16

Activations Density 0.034%

elements related to analysis and critique of evidence

No Comments

No Known Activations