INDEX

Explanations

numerical values and references to time or sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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§ak

-0.16

-0.15

bersome

-0.14

ROKE

-0.14

.publish

-0.14

Neutral

-0.14

iyel

-0.14

orden

-0.14

Î¾Î¹

-0.14

icap

-0.13

POSITIVE LOGITS

 transfer

0.29

 Transfer

0.27

 transfers

0.24

transfer

0.24

 loan

0.23

 signed

0.22

 transferred

0.22

Transfer

0.22

-transfer

0.22

 unsigned

0.21

Activations Density 0.036%

numerical values and references to time or sequences

No Comments

No Known Activations