INDEX

Explanations

references to environmental features and geographical characteristics

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

 ropes

-0.15

êµ´

-0.14

 cables

-0.14

 Heap

-0.14

-0.13

æī§

-0.13

perse

-0.13

sku

-0.13

å²©

-0.13

 Pipeline

-0.12

POSITIVE LOGITS

 plain

0.31

 rolling

0.30

 flat

0.29

 fertile

0.27

 Plain

0.26

plain

0.25

 plains

0.25

 Ð¿Ð»Ð¾Ð´

0.23

rolling

0.22

å¹³

0.22

Activations Density 0.119%

references to environmental features and geographical characteristics

No Comments

No Known Activations