Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
<div align="center">
|
| 2 |
|
| 3 |
# TinyLlama-1.1B
|
|
@@ -55,5 +63,4 @@ Below are some details of our training setup:
|
|
| 55 |
| Combined Dataset Size | 1 trillion tokens |
|
| 56 |
| Total Tokens During Training | 3 trillion (3 epochs/1430k steps) |
|
| 57 |
| Natural Language to Code Ratio | 7:3 |
|
| 58 |
-
| Hardware | 16 A100-40G GPUs |
|
| 59 |
-
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- cerebras/SlimPajama-627B
|
| 5 |
+
- bigcode/starcoderdata
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
---
|
| 9 |
<div align="center">
|
| 10 |
|
| 11 |
# TinyLlama-1.1B
|
|
|
|
| 63 |
| Combined Dataset Size | 1 trillion tokens |
|
| 64 |
| Total Tokens During Training | 3 trillion (3 epochs/1430k steps) |
|
| 65 |
| Natural Language to Code Ratio | 7:3 |
|
| 66 |
+
| Hardware | 16 A100-40G GPUs |
|
|
|