INDEX

Explanations

words and phrases related to location and infrastructure

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

 kolo

-0.17

iland

-0.15

ama

-0.15

Ð²Ð¸ÑĩÐ°Ð¹

-0.15

jezd

-0.15

lette

-0.14

å°¼äºļ

-0.14

esch

-0.14

eger

-0.14

onth

-0.14

POSITIVE LOGITS

 Duch

0.20

 Peak

0.20

 patch

0.20

 Potter

0.19

patch

0.18

ATCH

0.18

atched

0.18

 Penn

0.18

 constitu

0.17

 Midlands

0.17

Activations Density 0.065%

words and phrases related to location and infrastructure

No Comments

No Known Activations