INDEX

Explanations

technical terms and codes related to software functions and road abbreviations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

rung

-0.15

afen

-0.14

anon

-0.14

onn

-0.14

cle

-0.14

oller

-0.14

FT

-0.13

fdc

-0.13

enci

-0.13

asti

-0.13

POSITIVE LOGITS

##_

0.17

odash

0.15

íĭĢ

0.14

imdi

0.14

bsite

0.14

"('

0.14

cko

0.14

cade

0.14

ither

0.13

Activations Density 0.076%

technical terms and codes related to software functions and road abbreviations

No Comments

No Known Activations