INDEX

Explanations

references to past experiences and repeated actions

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

gne

-0.15

brane

-0.15

codegen

-0.15

getReference

-0.14

anne

-0.14

(QL

-0.14

ildo

-0.14

oreach

-0.14

ioctl

-0.14

 Away

-0.14

POSITIVE LOGITS

 previously

0.31

 previous

0.28

Previous

0.24

 already

0.24

 Previously

0.24

 Previous

0.23

previous

0.22

 elsewhere

0.21

 before

0.21

 twice

0.21

Activations Density 0.175%

references to past experiences and repeated actions

No Comments

No Known Activations