INDEX

Explanations

programming constructs and definitions used in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

pe

-0.14

edException

-0.14

Hel

-0.14

osity

-0.14

bab

-0.14

ãģ»ãģĨ

-0.14

ness

-0.14

\(

-0.14

hips

-0.13

 rever

-0.13

POSITIVE LOGITS

ritch

0.15

 forKey

0.14

ÑĻ

0.14

 Ð¡ÐµÑĢÐµÐ´

0.14

embro

0.14

erras

0.14

ardi

0.14

maj

0.13

adio

0.13

æĥ

0.13

Activations Density 0.048%

programming constructs and definitions used in code

No Comments

No Known Activations