INDEX

Explanations

temporal concepts and durations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

Sho

-0.14

 Lowe

-0.14

oyo

-0.13

 ÑĤÐ¾Ð²Ð°ÑĢÐ¸

-0.13

bri

-0.13

 veloc

-0.13

lotte

-0.13

 spor

-0.13

Lev

-0.13

-level

-0.13

POSITIVE LOGITS

 longer

0.33

 length

0.28

 duration

0.27

 Longer

0.26

 longest

0.25

 Length

0.23

 Duration

0.23

 lengths

0.23

 time

0.22

 lenght

0.21

Activations Density 0.233%

temporal concepts and durations

No Comments

No Known Activations