INDEX

Explanations

discussions related to homework and academic workload

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

Å¡ak

-0.17

loff

-0.16

agger

-0.15

à¸ļà¸£à¸£

-0.15

å°ĳå¹´

-0.15

 totals

-0.15

paralle

-0.14

ponible

-0.14

 Affero

-0.14

 Frankie

-0.14

POSITIVE LOGITS

 average

0.55

 Average

0.49

average

0.48

å¹³åĿĩ

0.46

Average

0.45

avg

0.44

-average

0.42

avg

0.41

 median

0.39

median

0.38

Activations Density 0.129%

discussions related to homework and academic workload

No Comments

No Known Activations