INDEX

Explanations

programming-related terminology and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÑĤÑı

-0.16

undle

-0.15

bil

-0.15

 CascadeType

-0.14

 horns

-0.14

ãĥ¼ãĥĨãĤ£

-0.14

Muk

-0.14

hots

-0.14

ccion

-0.13

quette

-0.13

POSITIVE LOGITS

 series

0.33

 Series

0.31

series

0.31

.series

0.29

Series

0.29

-series

0.28

 serie

0.27

 SERIES

0.26

 chart

0.26

(series

0.25

Activations Density 0.008%

programming-related terminology and constructs

No Comments

No Known Activations