Update README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,5 @@
|
|
1 |
---
|
2 |
-
license:
|
3 |
-
library_name: peft
|
4 |
tags:
|
5 |
- llama-factory
|
6 |
- lora
|
@@ -11,6 +10,10 @@ model-index:
|
|
11 |
---
|
12 |
# UNAversal-2x7B-v1
|
13 |
|
|
|
|
|
|
|
|
|
14 |
| Tasks |Version|Filter|n-shot| Metric |Value | |Stderr|
|
15 |
|--------------|-------|------|-----:|----------|-----:|---|-----:|
|
16 |
|arc_challenge |Yaml |none | 25|acc |0.7133|± |0.0132|
|
@@ -25,6 +28,4 @@ model-index:
|
|
25 |
|piqa |Yaml |none | 0|acc |0.8411|± |0.0085|
|
26 |
| | |none | 0|acc_norm |0.8526|± |0.0083|
|
27 |
|sciq |Yaml |none | 0|acc |0.9600|± |0.0062|
|
28 |
-
| | |none | 0|acc_norm |0.9370|± |0.0077|
|
29 |
-
|
30 |
-
|
|
|
1 |
---
|
2 |
+
license: apache-2.0
|
|
|
3 |
tags:
|
4 |
- llama-factory
|
5 |
- lora
|
|
|
10 |
---
|
11 |
# UNAversal-2x7B-v1
|
12 |
|
13 |
+
Merely Phase 1 UNA, only MLP's and its kinda of a beta. The goal was to produce a small but powerful MoE.
|
14 |
+
|
15 |
+
This is a 2 MoE model, of 7B each expert. Based on intel-neural series v3.
|
16 |
+
|
17 |
| Tasks |Version|Filter|n-shot| Metric |Value | |Stderr|
|
18 |
|--------------|-------|------|-----:|----------|-----:|---|-----:|
|
19 |
|arc_challenge |Yaml |none | 25|acc |0.7133|± |0.0132|
|
|
|
28 |
|piqa |Yaml |none | 0|acc |0.8411|± |0.0085|
|
29 |
| | |none | 0|acc_norm |0.8526|± |0.0083|
|
30 |
|sciq |Yaml |none | 0|acc |0.9600|± |0.0062|
|
31 |
+
| | |none | 0|acc_norm |0.9370|± |0.0077|
|
|
|
|