INDEX

Explanations

references to statements or reports related to community issues

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

oki

-0.14

 dolar

-0.14

 Ø¯Ø¹

-0.14

steller

-0.13

erten

-0.13

oftware

-0.13

rof

-0.13

ns

-0.13

inish

-0.13

POSITIVE LOGITS

how

0.32

why

0.28

å¦Ĥä½ķ

0.25

how

0.24

 behalf

0.24

 whether

0.22

 cÃ³mo

0.22

 matters

0.22

 topics

0.22

 subjects

0.22

Activations Density 0.164%

references to statements or reports related to community issues

No Comments

No Known Activations