INDEX

Explanations

numerical values and data types relevant to programming and computational processes

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

aura

-0.18

OPY

-0.16

caret

-0.15

 dÃ¡n

-0.15

á»ĳng

-0.15

.gwt

-0.15

orb

-0.15

ÙĪØ§Ø¡

-0.14

oyer

-0.14

POSITIVE LOGITS

0.18

yz

0.18

 Trad

0.15

 Keller

0.15

.Alpha

0.15

 Gross

0.15

 Scha

0.14

Te

0.14

Activations Density 0.156%

numerical values and data types relevant to programming and computational processes

No Comments

No Known Activations