INDEX

Explanations

comparative phrases indicating magnitude or extent

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

Ð°ÑĢÐ°ÑĤ

-0.15

 Milan

-0.15

lob

-0.15

UNET

-0.14

 milan

-0.14

tem

-0.14

Rog

-0.13

anc

-0.13

rego

-0.13

elmet

-0.13

POSITIVE LOGITS

.liferay

0.16

unden

0.15

ocos

0.15

Ð¾ÑģÐ½Ð¾Ð²

0.14

esin

0.14

ienes

0.14

.final

0.14

Ð»Ðµ

0.13

andles

0.13

Activations Density 0.020%

comparative phrases indicating magnitude or extent

No Comments

No Known Activations