forked from mlfoundations/dclm
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathheavy_code.yaml
390 lines (383 loc) · 11.1 KB
/
heavy_code.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
epoch: 1.25T
dataset: bigdata
num_params: 1B
max_seq_len: 2048
seed: 1
precision: fp32
# Tokenizer
tokenizer:
# name: [Add name from memory]
pretrained_model_name_or_path:
kwargs:
model_max_length: 2048
model:
name: open_lm
# pretrained_model_name_or_path: [add name from memory]
init_device: cpu
pretrained: true
load_path: # Add your (optional) Composer checkpoint path here!
device_eval_batch_size: 4
# FSDP config for model sharding
fsdp_config:
sharding_strategy: FULL_SHARD
mixed_precision: FULL
icl_tasks:
-
label: mmlu_zeroshot
dataset_uri: local_data/world_knowledge/mmlu.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
icl_task_type: multiple_choice
continuation_delimiter: 'Answer: ' # this separates questions from answers
has_categories: true
-
label: hellaswag_zeroshot
dataset_uri: local_data/language_understanding/hellaswag.jsonl
num_fewshot: [0]
icl_task_type: multiple_choice
-
label: jeopardy
dataset_uri: local_data/world_knowledge/jeopardy_all.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
continuation_delimiter: "\nAnswer: " # this separates questions from answers
has_categories: true
-
label: triviaqa_sm_sub
dataset_uri: local_data/world_knowledge/triviaqa_sm_sub.jsonl
num_fewshot: [3]
icl_task_type: question_answering
-
label: gsm8k_cot
dataset_uri: local_data/symbolic_problem_solving/gsm8k.jsonl
num_fewshot: [3]
icl_task_type: question_answering
cot_delimiter: ' #### '
continuation_delimiter: "\nA: Let's think step by step. "
question_prelimiter: "Q: "
early_stopping_criteria:
- "\n\n"
- "Q:"
- "Q: "
-
label: agi_eval_sat_math_cot
dataset_uri: local_data/symbolic_problem_solving/agi_eval_sat_math.jsonl
num_fewshot: [3]
icl_task_type: question_answering
cot_delimiter: ' #### '
continuation_delimiter: "\nA: Let's think step by step. "
question_prelimiter: "Q: "
early_stopping_criteria:
- "\n\n"
- "Q:"
- "Q: "
-
label: aqua_cot
dataset_uri: local_data/symbolic_problem_solving/aqua.jsonl
num_fewshot: [3]
icl_task_type: question_answering
cot_delimiter: ' #### '
continuation_delimiter: "\nA: Let's think step by step. "
question_prelimiter: "Q: "
early_stopping_criteria:
- "\n\n"
- "Q:"
- "Q: "
-
label: svamp
dataset_uri: local_data/symbolic_problem_solving/svamp.jsonl
num_fewshot: [3]
icl_task_type: question_answering
continuation_delimiter: "\nUsing the formula below:\n"
cot_delimiter: ' #### '
question_prelimiter: "Q: "
early_stopping_criteria:
- "\n\n"
- "Q:"
- "Q: "
-
label: bigbench_qa_wikidata
dataset_uri: local_data/world_knowledge/bigbench_qa_wikidata.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: arc_easy
dataset_uri: local_data/world_knowledge/arc_easy.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
-
label: arc_challenge
dataset_uri: local_data/world_knowledge/arc_challenge.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
-
label: mmlu
dataset_uri: local_data/world_knowledge/mmlu.jsonl
num_fewshot: [5]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
has_categories: true
-
label: bigbench_misconceptions
dataset_uri: local_data/world_knowledge/bigbench_misconceptions.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: copa
dataset_uri: local_data/commonsense_reasoning/copa.jsonl
num_fewshot: [0]
icl_task_type: multiple_choice
-
label: siqa
dataset_uri: local_data/commonsense_reasoning/siqa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: commonsense_qa
dataset_uri: local_data/commonsense_reasoning/commonsense_qa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: piqa
dataset_uri: local_data/commonsense_reasoning/piqa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
-
label: openbook_qa
dataset_uri: local_data/commonsense_reasoning/openbook_qa.jsonl
num_fewshot: [0]
icl_task_type: multiple_choice
-
label: bigbench_novel_concepts
dataset_uri: local_data/commonsense_reasoning/bigbench_novel_concepts.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_strange_stories
dataset_uri: local_data/commonsense_reasoning/bigbench_strange_stories.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_strategy_qa
dataset_uri: local_data/commonsense_reasoning/bigbench_strategy_qa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: lambada_openai
dataset_uri: local_data/language_understanding/lambada_openai.jsonl
num_fewshot: [0]
icl_task_type: language_modeling
-
label: hellaswag
dataset_uri: local_data/language_understanding/hellaswag.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: winograd
dataset_uri: local_data/language_understanding/winograd_wsc.jsonl
num_fewshot: [0]
icl_task_type: schema
-
label: winogrande
dataset_uri: local_data/language_understanding/winogrande.jsonl
num_fewshot: [0]
icl_task_type: schema
-
label: bigbench_conlang_translation
dataset_uri: local_data/language_understanding/bigbench_conlang_translation.jsonl
num_fewshot: [0]
icl_task_type: language_modeling
-
label: bigbench_language_identification
dataset_uri: local_data/language_understanding/bigbench_language_identification.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_conceptual_combinations
dataset_uri: local_data/language_understanding/bigbench_conceptual_combinations.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_elementary_math_qa
dataset_uri: local_data/symbolic_problem_solving/bigbench_elementary_math_qa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_dyck_languages
dataset_uri: local_data/symbolic_problem_solving/bigbench_dyck_languages.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: agi_eval_lsat_ar
dataset_uri: local_data/symbolic_problem_solving/agi_eval_lsat_ar.jsonl
num_fewshot: [3]
icl_task_type: multiple_choice
-
label: bigbench_cs_algorithms
dataset_uri: local_data/symbolic_problem_solving/bigbench_cs_algorithms.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: bigbench_logical_deduction
dataset_uri: local_data/symbolic_problem_solving/bigbench_logical_deduction.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bigbench_operators
dataset_uri: local_data/symbolic_problem_solving/bigbench_operators.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: bigbench_repeat_copy_logic
dataset_uri: local_data/symbolic_problem_solving/bigbench_repeat_copy_logic.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: simple_arithmetic_nospaces
dataset_uri: local_data/symbolic_problem_solving/simple_arithmetic_nospaces.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: simple_arithmetic_withspaces
dataset_uri: local_data/symbolic_problem_solving/simple_arithmetic_withspaces.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: math_qa
dataset_uri: local_data/symbolic_problem_solving/math_qa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: logi_qa
dataset_uri: local_data/symbolic_problem_solving/logi_qa.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
-
label: pubmed_qa_labeled
dataset_uri: local_data/reading_comprehension/pubmed_qa_labeled.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: squad
dataset_uri: local_data/reading_comprehension/squad.jsonl
num_fewshot: [10]
icl_task_type: language_modeling
-
label: agi_eval_lsat_rc
dataset_uri: local_data/reading_comprehension/agi_eval_lsat_rc.jsonl
num_fewshot: [3]
icl_task_type: multiple_choice
-
label: agi_eval_lsat_lr
dataset_uri: local_data/reading_comprehension/agi_eval_lsat_lr.jsonl
num_fewshot: [3]
icl_task_type: multiple_choice
-
label: coqa
dataset_uri: local_data/reading_comprehension/coqa.jsonl
num_fewshot: [0]
icl_task_type: language_modeling
-
label: bigbench_understanding_fables
dataset_uri: local_data/reading_comprehension/bigbench_understanding_fables.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: boolq
dataset_uri: local_data/reading_comprehension/boolq.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
continuation_delimiter: "\nAnswer: " # this separates questions from answers
-
label: agi_eval_sat_en
dataset_uri: local_data/reading_comprehension/agi_eval_sat_en.jsonl
num_fewshot: [3]
icl_task_type: multiple_choice
-
label: winogender_mc_female
dataset_uri: local_data/safety/winogender_mc_female.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: winogender_mc_male
dataset_uri: local_data/safety/winogender_mc_male.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: enterprise_pii_classification
dataset_uri: local_data/safety/enterprise_pii_classification.jsonl
num_fewshot: [10]
icl_task_type: multiple_choice
-
label: bbq
dataset_uri: local_data/safety/bbq.jsonl
num_fewshot: [3]
icl_task_type: multiple_choice
has_categories: true
-
label: human_eval
dataset_uri: local_data/programming/human_eval.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_cpp
dataset_uri: local_data/programming/processed_human_eval_cpp.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_js
dataset_uri: local_data/programming/processed_human_eval_js.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_return_simple
dataset_uri: local_data/programming/human_eval_return_simple.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_return_complex
dataset_uri: local_data/programming/human_eval_return_complex.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_25
dataset_uri: local_data/programming/human_eval-0.25.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_50
dataset_uri: local_data/programming/human_eval-0.5.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation
-
label: human_eval_75
dataset_uri: local_data/programming/human_eval-0.75.jsonl # ADD YOUR OWN DATASET URI
num_fewshot: [0]
pass_at_k: 1
num_beams: 5
batch_size: 1
icl_task_type: code_evaluation