INDEX

Explanations

words and phrases with specific references to names, brands, and unique identifiers

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

OMPI

-0.17

empor

-0.16

erland

-0.15

 Ø§ÙĦØ¹Ø±Ø¨

-0.15

arten

-0.15

arine

-0.15

olib

-0.15

 Deposit

-0.15

jar

-0.15

 Cube

-0.15

POSITIVE LOGITS

ansas

0.21

ipel

0.18

/inet

0.17

itecture

0.17

adia

0.16

itect

0.15

-ar

0.15

Î³Îµ

0.15

.ArgumentParser

0.15

ULD

0.15

Activations Density 0.080%

words and phrases with specific references to names, brands, and unique identifiers

No Comments

No Known Activations