INDEX

Explanations

variations of the word "sort."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

hip

-0.19

ous

-0.17

uria

-0.17

imals

-0.15

IDES

-0.15

aters

-0.15

aves

-0.15

ized

-0.15

ummer

-0.15

allet

-0.14

POSITIVE LOGITS

ilege

0.18

iment

0.18

.Sort

0.18

ileges

0.17

alim

0.16

 sobie

0.16

-lived

0.16

red

0.15

iller

0.15

/filter

0.15

Activations Density 0.013%

variations of the word "sort."

No Comments

No Known Activations