INDEX

Explanations

instances of the article "a" and phrases indicating quantity or specificity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

acker

-0.15

stÃ¥

-0.14

zeit

-0.14

åľ¨

-0.14

andas

-0.13

 shelves

-0.13

 spokeswoman

-0.13

_UNUSED

-0.13

lov

-0.13

 possibility

-0.13

POSITIVE LOGITS

 manner

0.53

 nutshell

0.39

way

0.39

 hurry

0.36

 fashion

0.33

 effort

0.29

 sposÃ³b

0.28

bid

0.26

 format

0.25

 context

0.25

Activations Density 0.149%

instances of the article "a" and phrases indicating quantity or specificity

No Comments

No Known Activations