INDEX

Explanations

phrases that emphasize maximizing efficiency or utility

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

Å¡tÄĽ

-0.15

nten

-0.14

ideographic

-0.14

usercontent

-0.14

Å¼e

-0.14

AINS

-0.14

omain

-0.13

uraa

-0.13

peon

-0.13

POSITIVE LOGITS

 resources

0.23

Limited

0.22

 limited

0.22

resources

0.21

limited

0.21

 Limited

0.20

 scarce

0.20

æľīéĻĲ

0.20

 Resources

0.19

imited

0.19

Activations Density 0.063%

phrases that emphasize maximizing efficiency or utility

No Comments

No Known Activations