INDEX

Explanations

references to the term "upper" in various contexts, particularly related to grades or classes

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

aho

-0.14

 pÃ¡n

-0.14

antal

-0.14

Ø¨ÛĮÙĨ

-0.14

pais

-0.13

AccessType

-0.13

lbl

-0.13

sky

-0.13

ismic

-0.13

 unbiased

-0.13

POSITIVE LOGITS

most

0.43

MOST

0.27

/l

0.23

-most

0.22

-middle

0.22

 bound

0.21

 bounds

0.21

cased

0.21

 reaches

0.20

-level

0.20

Activations Density 0.023%

references to the term "upper" in various contexts, particularly related to grades or classes

No Comments

No Known Activations