INDEX

Explanations

criticisms and assessments of design choices and layout issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 vegas

-0.14

 Millions

-0.14

ĥn

-0.13

 elsewhere

-0.13

elle

-0.13

 otherwise

-0.13

ichi

-0.13

ãĥ¼ãĥŃ

-0.13

 Either

-0.13

Ø¨Ø§Ø¯

-0.13

POSITIVE LOGITS

enor

0.15

orio

0.15

-Mart

0.14

alace

0.14

Ð´Ð°ÑĤ

0.14

inish

0.14

uien

0.14

 Ø§Ø³ØªØ®Ø¯Ø§Ùħ

0.14

afen

0.14

ebek

0.13

Activations Density 0.395%

criticisms and assessments of design choices and layout issues

No Comments

No Known Activations