INDEX

Explanations

digital or technical structures and programming terminologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Äįer

-0.15

odi

-0.14

subtype

-0.14

Ã¼m

-0.14

kowski

-0.14

avan

-0.13

 Mustafa

-0.13

zee

-0.13

uchar

-0.13

POSITIVE LOGITS

bu

0.15

UnderTest

0.14

Î¯Î¸

0.14

eka

0.14

uale

0.14

/items

0.14

hev

0.14

Ì£

0.14

ural

0.13

BU

0.13

Activations Density 0.019%

digital or technical structures and programming terminologies

No Comments

No Known Activations