INDEX

Explanations

references to academic publications and their structural components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

/downloads

-0.15

-0.14

ail

-0.14

jug

-0.14

 Anglo

-0.13

íĴį

-0.13

lio

-0.13

pun

-0.13

POSITIVE LOGITS

ayload

0.16

ummings

0.15

 chapters

0.15

chapter

0.15

 chapter

0.14

ved

0.14

ascade

0.14

vÄĽt

0.14

ÙħØ¯

0.14

erval

0.14

Activations Density 0.023%

references to academic publications and their structural components

No Comments

No Known Activations