INDEX

Explanations

references to the number three or concepts related to triads

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

uet

-0.17

nemonic

-0.17

wers

-0.16

regor

-0.15

Ð¾Ð²Ð°ÑĢÐ¸

-0.15

regunta

-0.15

æľĽ

-0.14

è¿ħ

-0.14

mund

-0.14

EFI

-0.14

POSITIVE LOGITS

rd

0.28

-quarters

0.26

fold

0.20

 dozen

0.18

-dimensional

0.17

Ð²Ð¸ÑĤ

0.16

à¸±à¸į

0.15

rl

0.15

-four

0.15

 dint

0.15

Activations Density 0.258%

references to the number three or concepts related to triads

No Comments

No Known Activations