INDEX

Explanations

technical terms related to software or web application components

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ãĥĥãĤ«ãĥ¼

-0.18

rex

-0.15

 ÐºÐ¸ÑĢ

-0.15

ìĦ

-0.15

arrass

-0.14

á»ĳn

-0.14

edn

-0.14

rices

-0.14

geg

-0.14

ACES

-0.13

POSITIVE LOGITS

WND

0.17

issy

0.16

ÑĢÐ°ÑĤÐ¸

0.16

 Compound

0.15

ies

0.15

compound

0.15

Wat

0.14

éĳ

0.14

/Gate

0.14

LastError

0.14

Activations Density 0.006%

technical terms related to software or web application components

No Comments

No Known Activations