victormiller
commited on
Commit
•
072a2d4
1
Parent(s):
a496f8f
Update README.md
Browse files
README.md
CHANGED
@@ -13,12 +13,12 @@ tags:
|
|
13 |
LLM360 demystifies the training recipe used for Llama 2 - 70B with K2.
|
14 |
|
15 |
About K2:
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
|
23 |
K2 was developed as a collaboration between [MBZUAI](https://mbzuai.ac.ae/institute-of-foundation-models/), [Petuum](petuum.com), and [LLM360](llm360.ai).
|
24 |
|
|
|
13 |
LLM360 demystifies the training recipe used for Llama 2 - 70B with K2.
|
14 |
|
15 |
About K2:
|
16 |
+
- 65 billion parameter LLM
|
17 |
+
- Tokens: 1.4T
|
18 |
+
* Languages: English
|
19 |
+
* Models Released: base, chat model
|
20 |
+
* Trained in 2 stages
|
21 |
+
* License: Apache 2.0
|
22 |
|
23 |
K2 was developed as a collaboration between [MBZUAI](https://mbzuai.ac.ae/institute-of-foundation-models/), [Petuum](petuum.com), and [LLM360](llm360.ai).
|
24 |
|