new version (?)
Browse files
README.md
CHANGED
|
@@ -30,7 +30,8 @@ model-index:
|
|
| 30 |
value: 73.44
|
| 31 |
name: strict accuracy
|
| 32 |
source:
|
| 33 |
-
url:
|
|
|
|
| 34 |
name: Open LLM Leaderboard
|
| 35 |
- task:
|
| 36 |
type: text-generation
|
|
@@ -45,7 +46,8 @@ model-index:
|
|
| 45 |
value: 22.55
|
| 46 |
name: normalized accuracy
|
| 47 |
source:
|
| 48 |
-
url:
|
|
|
|
| 49 |
name: Open LLM Leaderboard
|
| 50 |
- task:
|
| 51 |
type: text-generation
|
|
@@ -60,7 +62,8 @@ model-index:
|
|
| 60 |
value: 16.31
|
| 61 |
name: exact match
|
| 62 |
source:
|
| 63 |
-
url:
|
|
|
|
| 64 |
name: Open LLM Leaderboard
|
| 65 |
- task:
|
| 66 |
type: text-generation
|
|
@@ -75,7 +78,8 @@ model-index:
|
|
| 75 |
value: 2.35
|
| 76 |
name: acc_norm
|
| 77 |
source:
|
| 78 |
-
url:
|
|
|
|
| 79 |
name: Open LLM Leaderboard
|
| 80 |
- task:
|
| 81 |
type: text-generation
|
|
@@ -90,7 +94,8 @@ model-index:
|
|
| 90 |
value: 3.57
|
| 91 |
name: acc_norm
|
| 92 |
source:
|
| 93 |
-
url:
|
|
|
|
| 94 |
name: Open LLM Leaderboard
|
| 95 |
- task:
|
| 96 |
type: text-generation
|
|
@@ -107,8 +112,10 @@ model-index:
|
|
| 107 |
value: 24.25
|
| 108 |
name: accuracy
|
| 109 |
source:
|
| 110 |
-
url:
|
|
|
|
| 111 |
name: Open LLM Leaderboard
|
|
|
|
| 112 |
---
|
| 113 |
|
| 114 |
# Model Description
|
|
@@ -158,4 +165,4 @@ print("ANSWER: " + response_output)
|
|
| 158 |
|
| 159 |
This Llama model was trained faster than [Unsloth](https://github.com/unslothai/unsloth) using [custom training code](https://www.kaggle.com/code/piotr25691/distributed-llama-training-with-2xt4).
|
| 160 |
|
| 161 |
-
Visit https://www.kaggle.com/code/piotr25691/distributed-llama-training-with-2xt4 to find out how you can finetune your models using BOTH of the Kaggle provided GPUs.
|
|
|
|
| 30 |
value: 73.44
|
| 31 |
name: strict accuracy
|
| 32 |
source:
|
| 33 |
+
url: >-
|
| 34 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 35 |
name: Open LLM Leaderboard
|
| 36 |
- task:
|
| 37 |
type: text-generation
|
|
|
|
| 46 |
value: 22.55
|
| 47 |
name: normalized accuracy
|
| 48 |
source:
|
| 49 |
+
url: >-
|
| 50 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 51 |
name: Open LLM Leaderboard
|
| 52 |
- task:
|
| 53 |
type: text-generation
|
|
|
|
| 62 |
value: 16.31
|
| 63 |
name: exact match
|
| 64 |
source:
|
| 65 |
+
url: >-
|
| 66 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 67 |
name: Open LLM Leaderboard
|
| 68 |
- task:
|
| 69 |
type: text-generation
|
|
|
|
| 78 |
value: 2.35
|
| 79 |
name: acc_norm
|
| 80 |
source:
|
| 81 |
+
url: >-
|
| 82 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 83 |
name: Open LLM Leaderboard
|
| 84 |
- task:
|
| 85 |
type: text-generation
|
|
|
|
| 94 |
value: 3.57
|
| 95 |
name: acc_norm
|
| 96 |
source:
|
| 97 |
+
url: >-
|
| 98 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 99 |
name: Open LLM Leaderboard
|
| 100 |
- task:
|
| 101 |
type: text-generation
|
|
|
|
| 112 |
value: 24.25
|
| 113 |
name: accuracy
|
| 114 |
source:
|
| 115 |
+
url: >-
|
| 116 |
+
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=lunahr/thea-3b-25r
|
| 117 |
name: Open LLM Leaderboard
|
| 118 |
+
new_version: lunahr/thea-3b-50r-u1
|
| 119 |
---
|
| 120 |
|
| 121 |
# Model Description
|
|
|
|
| 165 |
|
| 166 |
This Llama model was trained faster than [Unsloth](https://github.com/unslothai/unsloth) using [custom training code](https://www.kaggle.com/code/piotr25691/distributed-llama-training-with-2xt4).
|
| 167 |
|
| 168 |
+
Visit https://www.kaggle.com/code/piotr25691/distributed-llama-training-with-2xt4 to find out how you can finetune your models using BOTH of the Kaggle provided GPUs.
|