INDEX

Explanations

phrases emphasizing direct addresses to the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

iban

-0.18

ãĥ³ãĥĲ

-0.16

odos

-0.15

oshi

-0.15

Î¿ÏĤ

-0.14

ÐµÑĢÐ¸Ð½

-0.14

 VerfÃ¼g

-0.14

geois

-0.14

ather

-0.14

 sclerosis

-0.13

POSITIVE LOGITS

 must

0.20

 should

0.20

may

0.20

 will

0.20

 MUST

0.18

ustain

0.18

can

0.16

 mileage

0.15

.should

0.15

’ll

0.14

Activations Density 0.063%

phrases emphasizing direct addresses to the reader

No Comments

No Known Activations