Spaces:
Running
Running
adds turkce_atasozleri dataset and results (#1)
Browse files- adds turkce_atasozleri dataset and results (5e62ec852c98968ce5a3779433ddc810c073f6f6)
- data.py +3 -2
- data/datasets.json +8 -0
- results/zero-shot/CerebrumTech__cere-llama-3-8b-tr.json +6 -0
- results/zero-shot/Llama-3.3-70B-Instruct.json +6 -0
- results/zero-shot/Ministral-8B-Instruct.json +6 -0
- results/zero-shot/Mistral-7B-Instruct-v0.3.json +6 -0
- results/zero-shot/Mistral-7B-v0.3.json +6 -0
- results/zero-shot/Mixtral-8x7B-Instruct-v0.1.json +6 -0
- results/zero-shot/Qwen2.5-0.5B-Instruct.json +6 -0
- results/zero-shot/Qwen2.5-0.5B.json +6 -0
- results/zero-shot/Qwen2.5-1.5B-Instruct.json +6 -0
- results/zero-shot/Qwen2.5-1.5B.json +6 -0
- results/zero-shot/Qwen2.5-14B-Instruct.json +6 -0
- results/zero-shot/Qwen2.5-14B.json +6 -0
- results/zero-shot/Qwen2.5-3B-Instruct.json +6 -0
- results/zero-shot/Qwen2.5-3B.json +6 -0
- results/zero-shot/Qwen2.5-7B-Instruct.json +6 -0
- results/zero-shot/Qwen2.5-7B.json +6 -0
- results/zero-shot/aya-23-35B.json +6 -0
- results/zero-shot/aya-23-8b.json +6 -0
- results/zero-shot/aya-expanse-32b.json +6 -0
- results/zero-shot/aya-expanse-8b.json +6 -0
- results/zero-shot/aya101.json +6 -0
- results/zero-shot/commencis-7b.json +6 -0
- results/zero-shot/kanarya-2b.json +6 -0
- results/zero-shot/llama-3-8b-instruct.json +6 -0
- results/zero-shot/llama-3-8b.json +6 -0
- results/zero-shot/llama-3.1-8b-instruct.json +6 -0
- results/zero-shot/llama-3.1-8b.json +6 -0
- results/zero-shot/llama-3.2-1b.json +6 -0
- results/zero-shot/llama-3.2-3b-instruct.json +6 -0
- results/zero-shot/llama-3.2-3b.json +6 -0
- results/zero-shot/mistral-7b.json +6 -0
- results/zero-shot/trendyol-7b.json +6 -0
- results/zero-shot/turna.json +6 -0
data.py
CHANGED
@@ -64,6 +64,7 @@ DATASET_TASK_DICT = {
|
|
64 |
# other generation
|
65 |
'wmt-tr-en-prompt': Tasks.MACHINE_TRANSLATION,
|
66 |
'gecturk_generation': Tasks.GRAMMATICAL_ERROR_CORRECTION,
|
|
|
67 |
}
|
68 |
|
69 |
|
@@ -90,8 +91,8 @@ DATASET_GROUPS = {
|
|
90 |
'description': 'Turkish splits of SQuAD-like datasets XQuAD and TQUAD.',
|
91 |
},
|
92 |
'MCQA': {
|
93 |
-
'datasets': ['xcopa_tr', 'exams_tr', 'belebele_tr'] + [x for x in DATASET_TASK_DICT.keys() if x.startswith('turkish_plu')],
|
94 |
-
'description': 'Multiple Choice Question Answering datasets: XCOPA, Exams, Belebele
|
95 |
},
|
96 |
'TC': {
|
97 |
'datasets': ['sts_tr', 'offenseval_tr', 'news_cat', 'ironytr', ],
|
|
|
64 |
# other generation
|
65 |
'wmt-tr-en-prompt': Tasks.MACHINE_TRANSLATION,
|
66 |
'gecturk_generation': Tasks.GRAMMATICAL_ERROR_CORRECTION,
|
67 |
+
'turkce_atasozleri': Tasks.MULTIPLE_CHOICE,
|
68 |
}
|
69 |
|
70 |
|
|
|
91 |
'description': 'Turkish splits of SQuAD-like datasets XQuAD and TQUAD.',
|
92 |
},
|
93 |
'MCQA': {
|
94 |
+
'datasets': ['xcopa_tr', 'exams_tr', 'belebele_tr', 'turkce_atasozleri'] + [x for x in DATASET_TASK_DICT.keys() if x.startswith('turkish_plu')],
|
95 |
+
'description': 'Multiple Choice Question Answering datasets: XCOPA, Exams, Belebele, Turkish PLU, and Turkce-Atasozleri.'
|
96 |
},
|
97 |
'TC': {
|
98 |
'datasets': ['sts_tr', 'offenseval_tr', 'news_cat', 'ironytr', ],
|
data/datasets.json
CHANGED
@@ -181,5 +181,13 @@
|
|
181 |
"url": "https://github.com/GGLAB-KU/gecturk",
|
182 |
"hf_name": "mcemilg/GECTurk-generation",
|
183 |
"generative": true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
184 |
}
|
185 |
}
|
|
|
181 |
"url": "https://github.com/GGLAB-KU/gecturk",
|
182 |
"hf_name": "mcemilg/GECTurk-generation",
|
183 |
"generative": true
|
184 |
+
},
|
185 |
+
"turkce_atasozleri": {
|
186 |
+
"name": "Proverbs",
|
187 |
+
"task": "multiple_choice",
|
188 |
+
"description": "A dataset for Turkish proverbs and their definitions.",
|
189 |
+
"url": "https://huggingface.co/datasets/furkanunluturk/turkce-atasozleri",
|
190 |
+
"hf_name": "abrek/turkce-atasozleri-lm-evaluation-harness",
|
191 |
+
"generative": false
|
192 |
}
|
193 |
}
|
results/zero-shot/CerebrumTech__cere-llama-3-8b-tr.json
CHANGED
@@ -168,6 +168,12 @@
|
|
168 |
"task": "multiple_choice",
|
169 |
"acc": 0.5144444444444445,
|
170 |
"acc_norm": 0.5144444444444445
|
|
|
|
|
|
|
|
|
|
|
|
|
171 |
}
|
172 |
]
|
173 |
}
|
|
|
168 |
"task": "multiple_choice",
|
169 |
"acc": 0.5144444444444445,
|
170 |
"acc_norm": 0.5144444444444445
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"name": "turkce_atasozleri",
|
174 |
+
"task": "multiple_choice",
|
175 |
+
"acc": 0.48092485549132946,
|
176 |
+
"acc_norm": 0.48092485549132946
|
177 |
}
|
178 |
]
|
179 |
}
|
results/zero-shot/Llama-3.3-70B-Instruct.json
CHANGED
@@ -170,6 +170,12 @@
|
|
170 |
"rouge1": 0.4242379216426208,
|
171 |
"rouge2": 0.2810704842229309,
|
172 |
"rougeL": 0.3500597320086401
|
|
|
|
|
|
|
|
|
|
|
|
|
173 |
}
|
174 |
]
|
175 |
}
|
|
|
170 |
"rouge1": 0.4242379216426208,
|
171 |
"rouge2": 0.2810704842229309,
|
172 |
"rougeL": 0.3500597320086401
|
173 |
+
},
|
174 |
+
{
|
175 |
+
"name": "turkce_atasozleri",
|
176 |
+
"task": "multiple_choice",
|
177 |
+
"acc": 0.9254335260115607,
|
178 |
+
"acc_norm": 0.9254335260115607
|
179 |
}
|
180 |
]
|
181 |
}
|
results/zero-shot/Ministral-8B-Instruct.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"name": "gecturk_generation",
|
168 |
"task": "grammatical_error_correction",
|
169 |
"exact_match": 0.009003803745967548
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"name": "gecturk_generation",
|
168 |
"task": "grammatical_error_correction",
|
169 |
"exact_match": 0.009003803745967548
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.4046242774566474,
|
175 |
+
"acc_norm": 0.4046242774566474
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Mistral-7B-Instruct-v0.3.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"name": "gecturk_generation",
|
168 |
"task": "grammatical_error_correction",
|
169 |
"exact_match": 0.009677885309836777
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"name": "gecturk_generation",
|
168 |
"task": "grammatical_error_correction",
|
169 |
"exact_match": 0.009677885309836777
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.3,
|
175 |
+
"acc_norm": 0.3
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Mistral-7B-v0.3.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.4111111111111111,
|
169 |
"acc_norm": 0.4111111111111111
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.4111111111111111,
|
169 |
"acc_norm": 0.4111111111111111
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.27572254335260116,
|
175 |
+
"acc_norm": 0.27572254335260116
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Mixtral-8x7B-Instruct-v0.1.json
CHANGED
@@ -169,6 +169,12 @@
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.036255958399537776
|
|
|
|
|
|
|
|
|
|
|
|
|
172 |
}
|
173 |
]
|
174 |
}
|
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.036255958399537776
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"name": "turkce_atasozleri",
|
175 |
+
"task": "multiple_choice",
|
176 |
+
"acc": 0.5150289017341041,
|
177 |
+
"acc_norm": 0.5150289017341041
|
178 |
}
|
179 |
]
|
180 |
}
|
results/zero-shot/Qwen2.5-0.5B-Instruct.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3,
|
169 |
"acc_norm": 0.3
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3,
|
169 |
"acc_norm": 0.3
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.2832369942196532,
|
175 |
+
"acc_norm": 0.2832369942196532
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-0.5B.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.29888888888888887,
|
169 |
"acc_norm": 0.29888888888888887
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.29888888888888887,
|
169 |
"acc_norm": 0.29888888888888887
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.20346820809248556,
|
175 |
+
"acc_norm": 0.20346820809248556
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-1.5B-Instruct.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.5344444444444445,
|
169 |
"acc_norm": 0.5344444444444445
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.5344444444444445,
|
169 |
"acc_norm": 0.5344444444444445
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.3468208092485549,
|
175 |
+
"acc_norm": 0.3468208092485549
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-1.5B.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.4666666666666667,
|
169 |
"acc_norm": 0.4666666666666667
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.4666666666666667,
|
169 |
"acc_norm": 0.4666666666666667
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.2300578034682081,
|
175 |
+
"acc_norm": 0.2300578034682081
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-14B-Instruct.json
CHANGED
@@ -169,6 +169,12 @@
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.00130001444460494
|
|
|
|
|
|
|
|
|
|
|
|
|
172 |
}
|
173 |
]
|
174 |
}
|
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.00130001444460494
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"name": "turkce_atasozleri",
|
175 |
+
"task": "multiple_choice",
|
176 |
+
"acc": 0.7832369942196532,
|
177 |
+
"acc_norm": 0.7832369942196532
|
178 |
}
|
179 |
]
|
180 |
}
|
results/zero-shot/Qwen2.5-14B.json
CHANGED
@@ -169,6 +169,12 @@
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.003707448601280755
|
|
|
|
|
|
|
|
|
|
|
|
|
172 |
}
|
173 |
]
|
174 |
}
|
|
|
169 |
"name": "gecturk_generation",
|
170 |
"task": "grammatical_error_correction",
|
171 |
"exact_match": 0.003707448601280755
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"name": "turkce_atasozleri",
|
175 |
+
"task": "multiple_choice",
|
176 |
+
"acc": 0.753757225433526,
|
177 |
+
"acc_norm": 0.753757225433526
|
178 |
}
|
179 |
]
|
180 |
}
|
results/zero-shot/Qwen2.5-3B-Instruct.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.6744444444444444,
|
169 |
"acc_norm": 0.6744444444444444
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.6744444444444444,
|
169 |
"acc_norm": 0.6744444444444444
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.6011560693641619,
|
175 |
+
"acc_norm": 0.6011560693641619
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-3B.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.6188888888888889,
|
169 |
"acc_norm": 0.6188888888888889
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.6188888888888889,
|
169 |
"acc_norm": 0.6188888888888889
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.4346820809248555,
|
175 |
+
"acc_norm": 0.4346820809248555
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-7B-Instruct.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.7344444444444445,
|
169 |
"acc_norm": 0.7344444444444445
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.7344444444444445,
|
169 |
"acc_norm": 0.7344444444444445
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.7121387283236994,
|
175 |
+
"acc_norm": 0.7121387283236994
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/Qwen2.5-7B.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.7388888888888889,
|
169 |
"acc_norm": 0.7388888888888889
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.7388888888888889,
|
169 |
"acc_norm": 0.7388888888888889
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.7352601156069364,
|
175 |
+
"acc_norm": 0.7352601156069364
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/aya-23-35B.json
CHANGED
@@ -169,6 +169,12 @@
|
|
169 |
"rouge1": 0.3920969453077054,
|
170 |
"rouge2": 0.25937196554017156,
|
171 |
"rougeL": 0.33144850765201345
|
|
|
|
|
|
|
|
|
|
|
|
|
172 |
}
|
173 |
]
|
174 |
}
|
|
|
169 |
"rouge1": 0.3920969453077054,
|
170 |
"rouge2": 0.25937196554017156,
|
171 |
"rougeL": 0.33144850765201345
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"name": "turkce_atasozleri",
|
175 |
+
"task": "multiple_choice",
|
176 |
+
"acc": 0.5687861271676301,
|
177 |
+
"acc_norm": 0.5687861271676301
|
178 |
}
|
179 |
]
|
180 |
}
|
results/zero-shot/aya-23-8b.json
CHANGED
@@ -163,6 +163,12 @@
|
|
163 |
"rouge1": 0.2855728817569547,
|
164 |
"rouge2": 0.14081555638864124,
|
165 |
"rougeL": 0.23467303626936886
|
|
|
|
|
|
|
|
|
|
|
|
|
166 |
}
|
167 |
]
|
168 |
}
|
|
|
163 |
"rouge1": 0.2855728817569547,
|
164 |
"rouge2": 0.14081555638864124,
|
165 |
"rougeL": 0.23467303626936886
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"name": "turkce_atasozleri",
|
169 |
+
"task": "multiple_choice",
|
170 |
+
"acc": 0.44971098265895953,
|
171 |
+
"acc_norm": 0.44971098265895953
|
172 |
}
|
173 |
]
|
174 |
}
|
results/zero-shot/aya-expanse-32b.json
CHANGED
@@ -168,6 +168,12 @@
|
|
168 |
"name": "gecturk_generation",
|
169 |
"task": "grammatical_error_correction",
|
170 |
"exact_match": 0
|
|
|
|
|
|
|
|
|
|
|
|
|
171 |
}
|
172 |
]
|
173 |
}
|
|
|
168 |
"name": "gecturk_generation",
|
169 |
"task": "grammatical_error_correction",
|
170 |
"exact_match": 0
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"name": "turkce_atasozleri",
|
174 |
+
"task": "multiple_choice",
|
175 |
+
"acc": 0.8236994219653179,
|
176 |
+
"acc_norm": 0.8236994219653179
|
177 |
}
|
178 |
]
|
179 |
}
|
results/zero-shot/aya-expanse-8b.json
CHANGED
@@ -154,6 +154,12 @@
|
|
154 |
"task": "extractive_question_answering",
|
155 |
"exact_match": 0.2495798319327731,
|
156 |
"f1": 0.4735125568867167
|
|
|
|
|
|
|
|
|
|
|
|
|
157 |
}
|
158 |
]
|
159 |
}
|
|
|
154 |
"task": "extractive_question_answering",
|
155 |
"exact_match": 0.2495798319327731,
|
156 |
"f1": 0.4735125568867167
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"name": "turkce_atasozleri",
|
160 |
+
"task": "multiple_choice",
|
161 |
+
"acc": 0.723121387283237,
|
162 |
+
"acc_norm": 0.723121387283237
|
163 |
}
|
164 |
]
|
165 |
}
|
results/zero-shot/aya101.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.2288888888888889,
|
169 |
"acc_norm": 0.2288888888888889
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.2288888888888889,
|
169 |
"acc_norm": 0.2288888888888889
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.009826589595375723,
|
175 |
+
"acc_norm": 0.009826589595375723
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/commencis-7b.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3233333333333333,
|
169 |
"acc_norm": 0.3233333333333333
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3233333333333333,
|
169 |
"acc_norm": 0.3233333333333333
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.22658959537572254,
|
175 |
+
"acc_norm": 0.22658959537572254
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/kanarya-2b.json
CHANGED
@@ -166,6 +166,12 @@
|
|
166 |
"task": "extractive_question_answering",
|
167 |
"exact_match": 0.008403361344537815,
|
168 |
"f1": 0.027799180278171867
|
|
|
|
|
|
|
|
|
|
|
|
|
169 |
}
|
170 |
]
|
171 |
}
|
|
|
166 |
"task": "extractive_question_answering",
|
167 |
"exact_match": 0.008403361344537815,
|
168 |
"f1": 0.027799180278171867
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"name": "turkce_atasozleri",
|
172 |
+
"task": "multiple_choice",
|
173 |
+
"acc": 0.0,
|
174 |
+
"acc_norm": 0.0
|
175 |
}
|
176 |
]
|
177 |
}
|
results/zero-shot/llama-3-8b-instruct.json
CHANGED
@@ -162,6 +162,12 @@
|
|
162 |
"rouge1": 0.29619456321037296,
|
163 |
"rouge2": 0.13520487191226377,
|
164 |
"rougeL": 0.220446635816053
|
|
|
|
|
|
|
|
|
|
|
|
|
165 |
}
|
166 |
]
|
167 |
}
|
|
|
162 |
"rouge1": 0.29619456321037296,
|
163 |
"rouge2": 0.13520487191226377,
|
164 |
"rougeL": 0.220446635816053
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"name": "turkce_atasozleri",
|
168 |
+
"task": "multiple_choice",
|
169 |
+
"acc": 0.6947976878612717,
|
170 |
+
"acc_norm": 0.6947976878612717
|
171 |
}
|
172 |
]
|
173 |
}
|
results/zero-shot/llama-3-8b.json
CHANGED
@@ -161,6 +161,12 @@
|
|
161 |
"rouge1": 0.2615001361521869,
|
162 |
"rouge2": 0.11093149007661907,
|
163 |
"rougeL": 0.20321693263972507
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
}
|
165 |
]
|
166 |
}
|
|
|
161 |
"rouge1": 0.2615001361521869,
|
162 |
"rouge2": 0.11093149007661907,
|
163 |
"rougeL": 0.20321693263972507
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"name": "turkce_atasozleri",
|
167 |
+
"task": "multiple_choice",
|
168 |
+
"acc": 0.44046242774566474,
|
169 |
+
"acc_norm": 0.44046242774566474
|
170 |
}
|
171 |
]
|
172 |
}
|
results/zero-shot/llama-3.1-8b-instruct.json
CHANGED
@@ -161,6 +161,12 @@
|
|
161 |
"rouge1": 0.2805962791068744,
|
162 |
"rouge2": 0.12421139697660691,
|
163 |
"rougeL": 0.21080710839195932
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
}
|
165 |
]
|
166 |
}
|
|
|
161 |
"rouge1": 0.2805962791068744,
|
162 |
"rouge2": 0.12421139697660691,
|
163 |
"rougeL": 0.21080710839195932
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"name": "turkce_atasozleri",
|
167 |
+
"task": "multiple_choice",
|
168 |
+
"acc": 0.7549132947976879,
|
169 |
+
"acc_norm": 0.7549132947976879
|
170 |
}
|
171 |
]
|
172 |
}
|
results/zero-shot/llama-3.1-8b.json
CHANGED
@@ -161,6 +161,12 @@
|
|
161 |
"rouge1": 0.2805962791068744,
|
162 |
"rouge2": 0.12421139697660691,
|
163 |
"rougeL": 0.21080710839195932
|
|
|
|
|
|
|
|
|
|
|
|
|
164 |
}
|
165 |
]
|
166 |
}
|
|
|
161 |
"rouge1": 0.2805962791068744,
|
162 |
"rouge2": 0.12421139697660691,
|
163 |
"rougeL": 0.21080710839195932
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"name": "turkce_atasozleri",
|
167 |
+
"task": "multiple_choice",
|
168 |
+
"acc": 0.5410404624277456,
|
169 |
+
"acc_norm": 0.5410404624277456
|
170 |
}
|
171 |
]
|
172 |
}
|
results/zero-shot/llama-3.2-1b.json
CHANGED
@@ -193,6 +193,12 @@
|
|
193 |
"rouge1": 0.2429304790539497,
|
194 |
"rouge2": 0.09668008744707143,
|
195 |
"rougeL": 0.18327092913535944
|
|
|
|
|
|
|
|
|
|
|
|
|
196 |
}
|
197 |
]
|
198 |
}
|
|
|
193 |
"rouge1": 0.2429304790539497,
|
194 |
"rouge2": 0.09668008744707143,
|
195 |
"rougeL": 0.18327092913535944
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"name": "turkce_atasozleri",
|
199 |
+
"task": "multiple_choice",
|
200 |
+
"acc": 0.21676300578034682,
|
201 |
+
"acc_norm": 0.21676300578034682
|
202 |
}
|
203 |
]
|
204 |
}
|
results/zero-shot/llama-3.2-3b-instruct.json
CHANGED
@@ -186,6 +186,12 @@
|
|
186 |
"rouge1": 0.2616423061938248,
|
187 |
"rouge2": 0.11064039063859936,
|
188 |
"rougeL": 0.19686955120787036
|
|
|
|
|
|
|
|
|
|
|
|
|
189 |
}
|
190 |
]
|
191 |
}
|
|
|
186 |
"rouge1": 0.2616423061938248,
|
187 |
"rouge2": 0.11064039063859936,
|
188 |
"rougeL": 0.19686955120787036
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"name": "turkce_atasozleri",
|
192 |
+
"task": "multiple_choice",
|
193 |
+
"acc": 0.010982658959537572,
|
194 |
+
"acc_norm": 0.010982658959537572
|
195 |
}
|
196 |
]
|
197 |
}
|
results/zero-shot/llama-3.2-3b.json
CHANGED
@@ -155,6 +155,12 @@
|
|
155 |
"rouge1": 0.2718989612690207,
|
156 |
"rouge2": 0.12031087767355472,
|
157 |
"rougeL": 0.21504760362022454
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
}
|
159 |
]
|
160 |
}
|
|
|
155 |
"rouge1": 0.2718989612690207,
|
156 |
"rouge2": 0.12031087767355472,
|
157 |
"rougeL": 0.21504760362022454
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"name": "turkce_atasozleri",
|
161 |
+
"task": "multiple_choice",
|
162 |
+
"acc": 0.1994219653179191,
|
163 |
+
"acc_norm": 0.1994219653179191
|
164 |
}
|
165 |
]
|
166 |
}
|
results/zero-shot/mistral-7b.json
CHANGED
@@ -160,6 +160,12 @@
|
|
160 |
"rouge1": 0.02720399421152351,
|
161 |
"rouge2": 0.012032606076011431,
|
162 |
"rougeL": 0.02311080687545987
|
|
|
|
|
|
|
|
|
|
|
|
|
163 |
}
|
164 |
]
|
165 |
}
|
|
|
160 |
"rouge1": 0.02720399421152351,
|
161 |
"rouge2": 0.012032606076011431,
|
162 |
"rougeL": 0.02311080687545987
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"name": "turkce_atasozleri",
|
166 |
+
"task": "multiple_choice",
|
167 |
+
"acc": 0.30809248554913293,
|
168 |
+
"acc_norm": 0.30809248554913293
|
169 |
}
|
170 |
]
|
171 |
}
|
results/zero-shot/trendyol-7b.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3622222222222222,
|
169 |
"acc_norm": 0.3622222222222222
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.3622222222222222,
|
169 |
"acc_norm": 0.3622222222222222
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.0,
|
175 |
+
"acc_norm": 0.0
|
176 |
}
|
177 |
]
|
178 |
}
|
results/zero-shot/turna.json
CHANGED
@@ -167,6 +167,12 @@
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.22555555555555556,
|
169 |
"acc_norm": 0.22555555555555556
|
|
|
|
|
|
|
|
|
|
|
|
|
170 |
}
|
171 |
]
|
172 |
}
|
|
|
167 |
"task": "multiple_choice",
|
168 |
"acc": 0.22555555555555556,
|
169 |
"acc_norm": 0.22555555555555556
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"name": "turkce_atasozleri",
|
173 |
+
"task": "multiple_choice",
|
174 |
+
"acc": 0.19248554913294796,
|
175 |
+
"acc_norm": 0.19248554913294796
|
176 |
}
|
177 |
]
|
178 |
}
|