INDEX

Explanations

phrases indicating past experiences or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 whenever

-0.22

 Whenever

-0.18

When

-0.17

Whenever

-0.16

 increasingly

-0.15

 when

-0.15

Ã§e

-0.15

emek

-0.15

 quando

-0.15

_iff

-0.15

POSITIVE LOGITS

had

0.39

 kdyby

0.38

had

0.36

 given

0.30

Had

0.28

 hÃ¤tte

0.28

 hadn

0.28

 hÃ¤t

0.27

Had

0.26

 would

0.25

Activations Density 0.322%

phrases indicating past experiences or conditions

No Comments

No Known Activations