INDEX

Explanations

instances of the word "Read" and variations related to reading, knowledge, and updates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Cunning

-0.16

cone

-0.16

/form

-0.14

ju

-0.13

itness

-0.13

<context

-0.13

 Ð¿Ð¾ÑģÑĤÑĥÐ¿

-0.13

lav

-0.13

ona

-0.12

 juice

-0.12

POSITIVE LOGITS

&A

0.14

Ø§Ø¨Ø·

0.14

atorium

0.14

ptrdiff

0.14

 arsch

0.14

_PUR

0.14

ttp

0.13

íĭĢ

0.13

plementation

0.13

tg

0.13

Activations Density 0.080%

instances of the word "Read" and variations related to reading, knowledge, and updates

No Comments

No Known Activations