INDEX

Explanations

details related to historical education and perception

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

oubles

-0.15

_almost

-0.15

=title

-0.15

ura

-0.15

ãĤıãģĽ

-0.14

AndWait

-0.14

ymm

-0.14

ind

-0.13

inery

-0.13

endl

-0.12

POSITIVE LOGITS

 ignorance

0.38

ä¸įçŁ¥éģĵ

0.36

 descon

0.35

 ignorant

0.33

 unaware

0.31

unknown

0.30

 unknown

0.29

 unfamiliar

0.28

 hadn

0.28

 ignor

0.27

Activations Density 0.322%

details related to historical education and perception

No Comments

No Known Activations