INDEX

Explanations

sequences and patterns in numerical data representation

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

peat

-0.14

 Leonard

-0.14

cae

-0.14

ventional

-0.14

Ori

-0.14

 Frauen

-0.14

 addCriterion

-0.14

à¸Ļà¸§

-0.13

ariant

-0.13

POSITIVE LOGITS

0.25

0.24

0.23

0.22

0.21

ads

0.20

0.19

Activations Density 0.065%

sequences and patterns in numerical data representation

No Comments

No Known Activations