INDEX

Explanations

references to educational recognition and achievements

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

èģĮä¸ļ

-0.15

],[-

-0.14

.Syntax

-0.14

igner

-0.14

nip

-0.14

Ð²Ð°ÑĤ

-0.14

Locker

-0.13

avo

-0.13

 åĽ

-0.13

UME

-0.13

POSITIVE LOGITS

 Gold

0.28

 Bronze

0.26

 Platinum

0.26

 seal

0.25

 bronze

0.25

 gold

0.25

 Seal

0.25

 platinum

0.25

 silver

0.24

 Silver

0.23

Activations Density 0.083%

references to educational recognition and achievements

No Comments

No Known Activations