INDEX

Explanations

variations of the word "rever" (related to admiration or respect)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Ã§os

-0.15

_BATCH

-0.15

achment

-0.14

 ÑģÐ¾Ð¿

-0.14

ureau

-0.14

(Bytes

-0.14

UGHT

-0.14

reten

-0.14

Ø¯ÛĮÚ¯Ø±

-0.13

Ã§o

-0.13

POSITIVE LOGITS

ber

0.26

 rever

0.24

end

0.23

ential

0.23

ence

0.22

enced

0.20

ences

0.19

ently

0.18

ent

0.18

ENCE

0.18

Activations Density 0.006%

variations of the word "rever" (related to admiration or respect)

No Comments

No Known Activations