INDEX

Explanations

references to specificity and detailed descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

à¸£

-0.18

our

-0.16

alim

-0.15

ony

-0.15

Ø§ÙĨÙĩ

-0.15

een

-0.15

oric

-0.15

awn

-0.15

work

-0.14

istrovstvÃŃ

-0.14

POSITIVE LOGITS

ities

0.30

ally

0.29

ially

0.26

ity

0.26

-purpose

0.25

ALLY

0.24

 gravity

0.19

/general

0.18

atio

0.18

 Gravity

0.18

Activations Density 0.034%

references to specificity and detailed descriptions

No Comments

No Known Activations