INDEX

Explanations

themes and topics related to discussions about experiences and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

lotte

-0.15

-BEGIN

-0.14

ourg

-0.14

argas

-0.14

 commute

-0.14

itol

-0.14

ÐµÐ¼Ð¾Ð½

-0.13

EFR

-0.13

apor

-0.13

ema

-0.13

POSITIVE LOGITS

 touch

0.50

 covered

0.49

 cover

0.47

 covers

0.45

 covering

0.45

 touched

0.44

touch

0.42

 address

0.41

 discuss

0.41

covered

0.41

Activations Density 0.233%

themes and topics related to discussions about experiences and recommendations

No Comments

No Known Activations