INDEX

Explanations

references to actions and events involving people and their movements or encounters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

èµĸ

-0.14

Plot

-0.14

Ã¼ns

-0.14

queda

-0.13

è¥²

-0.13

äºŃ

-0.13

akedirs

-0.13

lopen

-0.13

POSITIVE LOGITS

 taken

0.28

 cart

0.28

 transported

0.27

 intern

0.27

 shipped

0.26

 whisk

0.26

 moved

0.26

 transferred

0.25

 transfer

0.24

taken

0.23

Activations Density 0.229%

references to actions and events involving people and their movements or encounters

No Comments

No Known Activations