INDEX

Explanations

technical terms and checks related to file systems and programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

lio

-0.16

essler

-0.15

unas

-0.14

zell

-0.14

iete

-0.14

ghi

-0.14

lox

-0.14

eh

-0.14

lik

-0.14

POSITIVE LOGITS

ãĥ³ãĥģ

0.15

subst

0.15

 ê·Ģ

0.14

compress

0.14

Orc

0.14

 compress

0.14

bpp

0.14

uppen

0.13

compressed

0.13

 puck

0.13

Activations Density 0.108%

technical terms and checks related to file systems and programming constructs

No Comments

No Known Activations