INDEX

Explanations

programming-related terminology and coding syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ury

-0.17

abs

-0.15

abs

-0.14

rim

-0.14

anni

-0.14

 comma

-0.14

 Ø§ÙĦØªÙĤ

-0.14

chet

-0.13

ãĥ¼ãĤ¸

-0.13

POSITIVE LOGITS

 capital

0.39

 capitals

0.33

 Capital

0.33

 upper

0.32

capital

0.32

 lower

0.31

Capital

0.30

 caps

0.30

 CAPITAL

0.30

 case

0.29

Activations Density 0.132%

programming-related terminology and coding syntax

No Comments

No Known Activations