INDEX

Explanations

occurrences of the word "Wa" in various capitalizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

obuf

-0.16

emies

-0.15

Ð¸Ð½Ð¾Ð²

-0.15

æ»ĳ

-0.15

ogens

-0.14

pog

-0.14

unfold

-0.14

ERÄ°

-0.14

ensor

-0.14

æĸ¹

-0.14

POSITIVE LOGITS

fer

0.28

iving

0.27

iver

0.26

ivers

0.26

ivered

0.25

Wa

0.24

heed

0.24

IVER

0.23

Wa

0.22

agner

0.22

Activations Density 0.009%

occurrences of the word "Wa" in various capitalizations

No Comments

No Known Activations