INDEX

Explanations

numerical references to rankings and standings in competitions

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

undry

-0.17

ÅĻeh

-0.17

azi

-0.17

orca

-0.16

chedules

-0.16

inks

-0.15

atron

-0.15

deaux

-0.14

dbcTemplate

-0.14

alles

-0.14

POSITIVE LOGITS

order

0.19

 order

0.18

 podium

0.18

frame

0.17

 finish

0.16

fram

0.16

posit

0.16

Order

0.16

Nos

0.15

 spots

0.15

Activations Density 0.044%

numerical references to rankings and standings in competitions

No Comments

No Known Activations