INDEX

Explanations

calls to action or prompts for user engagement

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

ÐµÑĢÑĪ

-0.15

illary

-0.14

antic

-0.14

ptune

-0.14

.idea

-0.13

è¾Ľ

-0.13

prit

-0.13

vala

-0.13

bye

-0.13

 originally

-0.13

POSITIVE LOGITS

ernen

0.17

ÑĢÑı

0.16

/Edit

0.16

Ã¡zev

0.15

FREE

0.15

eny

0.14

 DbSet

0.14

elles

0.14

fÃ¼hr

0.14

orget

0.13

Activations Density 0.244%

calls to action or prompts for user engagement

No Comments

No Known Activations