INDEX

Explanations

technical terms related to mathematical equations and models

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

Ð»ÑĥÑĩ

-0.17

nila

-0.17

.gc

-0.16

esel

-0.16

.managed

-0.15

IFS

-0.15

suz

-0.15

(æľĪ

-0.14

elman

-0.14

adera

-0.14

POSITIVE LOGITS

 Bolt

0.28

 Chapman

0.26

 Collision

0.24

 collision

0.24

Collision

0.23

pdf

0.22

Rare

0.22

 rare

0.22

 Rare

0.22

collision

0.22

Activations Density 0.064%

technical terms related to mathematical equations and models

No Comments

No Known Activations