INDEX

Explanations

discernments and evaluations of stories and performances

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

new

-0.20

 latest

-0.19

 newest

-0.19

 lately

-0.17

 Latest

-0.16

lef

-0.16

anut

-0.16

 recently

-0.15

 currently

-0.15

lei

-0.15

POSITIVE LOGITS

 ëĭ¹ìĭľ

0.17

 Rank

0.15

 iconic

0.14

 proto

0.14

Ð¾Ð´ÐµÑĢÐ¶

0.14

 Schwarz

0.14

 ÑĢÐ¸Ð·

0.14

 memor

0.14

0.13

ëĭ¹

0.13

Activations Density 0.035%

discernments and evaluations of stories and performances

No Comments

No Known Activations