Commit
·
98e2ff0
1
Parent(s):
57a349f
Delete README.md
Browse files
README.md
DELETED
|
@@ -1,29 +0,0 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: bigscience-openrail-m
|
| 3 |
-
datasets:
|
| 4 |
-
- manojpreveen/Instruction_Tuning
|
| 5 |
-
---
|
| 6 |
-
Instruction Tuned GPT-NeoXT-20B model on Instruction Tuning dataset as listed below (~560k data) using ***Colossal AI***
|
| 7 |
-
|
| 8 |
-
**Base Model:** togethercomputer/GPT-NeoXT-Chat-Base-20B (GPT-NeoXT-Chat-Base-20B-v0.16 - fine-tuned on feedback data)
|
| 9 |
-
|
| 10 |
-
**Training Details :**
|
| 11 |
-
* Epochs: 5
|
| 12 |
-
* Batch Size : 16 instantaneous per device x 1 gradient accumulation steps x 8 gpus = 128
|
| 13 |
-
* Max Length : 1024
|
| 14 |
-
* Weight Decay : 0
|
| 15 |
-
* Learning Rate : 2e-5
|
| 16 |
-
* Learning Rate Scheduler Type : Cosine
|
| 17 |
-
* Number of warmup steps : 240
|
| 18 |
-
* Machine : 8xA100 80GB
|
| 19 |
-
|
| 20 |
-
**Dataset Details :**
|
| 21 |
-
|
| 22 |
-
Dataset : manojpreveen/Instruction_Tuning
|
| 23 |
-
|
| 24 |
-
Files :
|
| 25 |
-
* stanford_alpaca_it_v2.csv
|
| 26 |
-
* ColossalChat.csv
|
| 27 |
-
* unified_chip2.csv
|
| 28 |
-
* iamai_summarization_v1.csv
|
| 29 |
-
* iamai_v1.csv
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|