INDEX

Explanations

simple solutions or concepts that are effective and efficient

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ys

-0.17

ìĿ´ìĸ´

-0.15

andas

-0.15

YS

-0.14

asset

-0.14

declspec

-0.14

 Stephens

-0.14

enk

-0.14

gili

-0.14

.populate

-0.13

POSITIVE LOGITS

/simple

0.27

ton

0.25

tons

0.23

 simple

0.23

 simples

0.20

simple

0.19

.Simple

0.18

-simple

0.18

postalcode

0.17

TON

0.17

Activations Density 0.088%

simple solutions or concepts that are effective and efficient

No Comments

No Known Activations