INDEX

Explanations

references to notable individuals or achievements in a sports or competitive context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

elian

-0.17

IFORM

-0.17

lessly

-0.17

ãģĦãģ¦

-0.17

elijke

-0.17

raquo

-0.16

holes

-0.16

.VK

-0.16

hole

-0.16

fully

-0.15

POSITIVE LOGITS

mente

0.41

ities

0.30

ity

0.28

ness

0.27

ized

0.25

-minded

0.23

most

0.23

ised

0.22

zeitig

0.20

-speaking

0.20

Activations Density 1.002%

references to notable individuals or achievements in a sports or competitive context

No Comments

No Known Activations

references to notable individuals or achievements in a sports or competitive context

No Comments

No Known Activations