TimeMobius
commited on
Commit
•
deeb947
1
Parent(s):
6659471
Update README.md
Browse files
README.md
CHANGED
@@ -10,7 +10,7 @@ language:
|
|
10 |
inference: false
|
11 |
---
|
12 |
# Model Card for Mobius-12B-base-m1
|
13 |
-
The Mobius-12B-base-m1 Large Language Model (LLM) is a pretrained model based on RWKV v5 arch. We use
|
14 |
|
15 |
|
16 |
## Warning
|
@@ -56,9 +56,9 @@ The Mobius base m1 is the base model can be easily fine-tuned to achieve compell
|
|
56 |
| lambda ppl | 3.41 |
|
57 |
| lambda | 0.72 |
|
58 |
| piqa | 0.78 |
|
59 |
-
| hellaswag
|
60 |
| winogrande | 0.68 |
|
61 |
-
| arc_challenge
|
62 |
| arc_easy | 0.73 |
|
63 |
| openbookqa | 0.40 |
|
64 |
| sciq | 0.93 |
|
|
|
10 |
inference: false
|
11 |
---
|
12 |
# Model Card for Mobius-12B-base-m1
|
13 |
+
The Mobius-12B-base-m1 Large Language Model (LLM) is a pretrained model based on RWKV v5 arch. We use 0.01B tokens to post train this model.
|
14 |
|
15 |
|
16 |
## Warning
|
|
|
56 |
| lambda ppl | 3.41 |
|
57 |
| lambda | 0.72 |
|
58 |
| piqa | 0.78 |
|
59 |
+
| hellaswag 10 shots | 0.72 |
|
60 |
| winogrande | 0.68 |
|
61 |
+
| arc_challenge 25shots | 0.47 |
|
62 |
| arc_easy | 0.73 |
|
63 |
| openbookqa | 0.40 |
|
64 |
| sciq | 0.93 |
|