INDEX

Explanations

phrases related to naming or designation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ems

-0.19

Ð»ÐµÑĩ

-0.17

ument

-0.17

grav

-0.15

 equivalents

-0.14

.kotlin

-0.14

ush

-0.14

_UNS

-0.14

 Sheldon

-0.14

aversal

-0.13

POSITIVE LOGITS

 named

0.29

named

0.29

 Named

0.27

 ÑĩÐµÑģÑĤÑĮ

0.26

Named

0.26

 after

0.22

 ÙĨØ³Ø¨Ø©

0.21

 honor

0.20

.Named

0.19

.named

0.18

Activations Density 0.028%

phrases related to naming or designation

No Comments

No Known Activations