INDEX

Explanations

code comments that describe parameters and functionality in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

echa

-0.15

stantiate

-0.15

 Dich

-0.15

 Ø§ÙĦÙħØ®Øª

-0.14

 fisse

-0.14

unte

-0.14

ich

-0.14

 phÃ¡i

-0.13

ÑĤÐ°Ð¿

-0.13

rap

-0.13

POSITIVE LOGITS

iffies

0.15

Lif

0.15

nop

0.15

ymoon

0.14

å¦

0.14

'{@

0.13

Rit

0.13

Ø¹Ùĩ

0.13

Īëĭ¤

0.13

@param

0.13

Activations Density 0.007%

code comments that describe parameters and functionality in a programming context

No Comments

No Known Activations