INDEX

Explanations

references to books and their publication details

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

 modeling

-0.17

 honors

-0.17

 fiber

-0.17

 Harbor

-0.17

 favored

-0.16

 fibers

-0.16

pto

-0.16

 favors

-0.15

 favorable

-0.15

 artifact

-0.15

POSITIVE LOGITS

 Guardian

0.24

extract

0.24

BBC

0.23

BBC

0.22

 Extract

0.22

 extracts

0.21

Extract

0.21

bbc

0.21

 Booker

0.21

 extract

0.20

Activations Density 0.247%

references to books and their publication details

No Comments

No Known Activations