INDEX

Explanations

properties or attributes related to type definitions in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

affer

-0.16

 ÑĢÐ¾Ñģ

-0.15

anni

-0.15

ead

-0.14

aci

-0.14

.prototype

-0.14

éĺ

-0.14

iddi

-0.14

uyu

-0.14

 fÃ©

-0.14

POSITIVE LOGITS

Erl

0.28

OTP

0.27

 Phoenix

0.26

lixir

0.24

OTP

0.23

nesia

0.23

elix

0.22

.er

0.22

erts

0.22

Phoenix

0.22

Activations Density 0.003%

properties or attributes related to type definitions in code

No Comments

No Known Activations