Tristan commited on
Commit
0aea66e
1 Parent(s): c1b836a

Training in progress, epoch 0

Browse files
config.json CHANGED
@@ -23,7 +23,7 @@
23
  "rotary_pct": 0.25,
24
  "tie_word_embeddings": false,
25
  "torch_dtype": "float32",
26
- "transformers_version": "4.39.3",
27
  "use_cache": true,
28
  "use_parallel_residual": true,
29
  "vocab_size": 50304
 
23
  "rotary_pct": 0.25,
24
  "tie_word_embeddings": false,
25
  "torch_dtype": "float32",
26
+ "transformers_version": "4.40.2",
27
  "use_cache": true,
28
  "use_parallel_residual": true,
29
  "vocab_size": 50304
eval_job_output.txt CHANGED
@@ -1,4 +1,4 @@
1
- slurm submission log: 2024-05-08 15:15:19.808042
2
  created following sbatch script:
3
 
4
  ###############################
@@ -7,24 +7,24 @@ created following sbatch script:
7
 
8
  #SBATCH --account=nlp
9
  #SBATCH --cpus-per-task=16
10
- #SBATCH --dependency=afterok:7590696
11
  #SBATCH --gres=gpu:1
12
- #SBATCH --job-name=tthrush-job-478757
13
  #SBATCH --mem=60G
14
  #SBATCH --nodelist=sphinx2
15
  #SBATCH --open-mode=append
16
- #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/eval_job_output.txt
17
  #SBATCH --partition=sphinx
18
  #SBATCH --time=14-0
19
 
20
  # activate your desired anaconda environment
21
- . /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
22
 
23
  # cd to working directory
24
  cd .
25
 
26
  # launch commands
27
- srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/perf'
28
 
29
  ###############################
30
 
@@ -34,67 +34,265 @@ submission to slurm complete!
34
  ###############################
35
  slurm submission output
36
 
37
- Submitted batch job 7590697
38
 
39
 
40
 
41
  ###############################
42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
  ###############################
44
- start time: 2024-05-08 16:46:03.300046
45
- machine: sphinx2
46
- conda env: pretraining-coreset-selection
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
47
  ###############################
48
- running following processes
49
 
50
- lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/perf
51
 
52
 
53
  ###############################
54
- command outputs:
 
 
 
55
 
56
 
57
- 2024-05-08:16:46:16,301 INFO [utils.py:145] Note: detected 255 virtual cores but NumExpr set to maximum of 64, check "NUMEXPR_MAX_THREADS" environment variable.
58
- 2024-05-08:16:46:16,301 INFO [utils.py:148] Note: NumExpr detected 255 cores but "NUMEXPR_MAX_THREADS" not set, so enforcing safe limit of 8.
59
- 2024-05-08:16:46:16,301 INFO [utils.py:160] NumExpr defaulting to 8 threads.
60
- 2024-05-08:16:46:18,703 INFO [config.py:58] PyTorch version 2.2.2 available.
61
- 2024-05-08:16:46:38,285 INFO [__main__.py:156] Verbosity set to INFO
62
- 2024-05-08:16:46:52,372 WARNING [__init__.py:194] Some tasks could not be loaded due to missing dependencies. Run with `--verbosity DEBUG` for full details.
63
- /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/datasets/load.py:1429: FutureWarning: The repository for hails/mmlu_no_train contains custom code which must be executed to correctly load the dataset. You can inspect the repository content at https://hf.co/datasets/hails/mmlu_no_train
64
- You can avoid this message in future by passing the argument `trust_remote_code=True`.
65
- Passing `trust_remote_code=True` will be mandatory to load this dataset from the next major release of `datasets`.
66
- warnings.warn(
67
- 2024-05-08:16:48:12,611 WARNING [__init__.py:194] Some tasks could not be loaded due to missing dependencies. Run with `--verbosity DEBUG` for full details.
68
- 2024-05-08:16:48:12,617 INFO [__main__.py:229] Selected Tasks: ['arc_easy', 'lambada', 'piqa', 'sciq', 'xnli_en', 'xnli_fr']
69
- 2024-05-08:16:48:12,899 INFO [huggingface.py:148] Using device 'cuda'
70
- Traceback (most recent call last):
71
- File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/bin/lm_eval", line 8, in <module>
72
- sys.exit(cli_evaluate())
73
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/__main__.py", line 231, in cli_evaluate
74
- results = evaluator.simple_evaluate(
75
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/utils.py", line 415, in _wrapper
76
- return fn(*args, **kwargs)
77
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/evaluator.py", line 98, in simple_evaluate
78
- lm = lm_eval.api.registry.get_model(model).create_from_arg_string(
79
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/api/model.py", line 134, in create_from_arg_string
80
- return cls(**args, **args2)
81
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/models/huggingface.py", line 174, in __init__
82
- self._get_config(
83
- File "/sailhome/tthrush/lm-evaluation-harness/lm_eval/models/huggingface.py", line 420, in _get_config
84
- self._config = transformers.AutoConfig.from_pretrained(
85
- File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1138, in from_pretrained
86
- config_dict, unused_kwargs = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs)
87
- File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 631, in get_config_dict
88
- config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs)
89
- File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 686, in _get_config_dict
90
- resolved_config_file = cached_file(
91
- File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/utils/hub.py", line 369, in cached_file
92
- raise EnvironmentError(
93
- OSError: /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa does not appear to have a file named config.json. Checkout 'https://huggingface.co//juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/tree/main' for available files.
94
  ###############################
95
- end time: 2024-05-08 16:48:23.445722
96
- elapsed time: 0:02:20.145676
97
- slurm submission log: 2024-05-09 07:34:41.456517
98
  created following sbatch script:
99
 
100
  ###############################
@@ -103,24 +301,108 @@ created following sbatch script:
103
 
104
  #SBATCH --account=nlp
105
  #SBATCH --cpus-per-task=16
106
- #SBATCH --dependency=afterok:7591656
107
  #SBATCH --gres=gpu:1
108
- #SBATCH --job-name=tthrush-job-942141
109
  #SBATCH --mem=60G
110
  #SBATCH --nodelist=sphinx2
111
  #SBATCH --open-mode=append
112
- #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/eval_job_output.txt
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
113
  #SBATCH --partition=sphinx
114
  #SBATCH --time=14-0
115
 
116
  # activate your desired anaconda environment
117
- . /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
118
 
119
  # cd to working directory
120
  cd .
121
 
122
  # launch commands
123
- srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/perf'
124
 
125
  ###############################
126
 
@@ -130,7 +412,7 @@ submission to slurm complete!
130
  ###############################
131
  slurm submission output
132
 
133
- Submitted batch job 7591657
134
 
135
 
136
 
 
1
+ slurm submission log: 2024-05-19 09:14:41.870170
2
  created following sbatch script:
3
 
4
  ###############################
 
7
 
8
  #SBATCH --account=nlp
9
  #SBATCH --cpus-per-task=16
10
+ #SBATCH --dependency=afterok:7631089
11
  #SBATCH --gres=gpu:1
12
+ #SBATCH --job-name=tthrush-job-4203243
13
  #SBATCH --mem=60G
14
  #SBATCH --nodelist=sphinx2
15
  #SBATCH --open-mode=append
16
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
17
  #SBATCH --partition=sphinx
18
  #SBATCH --time=14-0
19
 
20
  # activate your desired anaconda environment
21
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
22
 
23
  # cd to working directory
24
  cd .
25
 
26
  # launch commands
27
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
28
 
29
  ###############################
30
 
 
34
  ###############################
35
  slurm submission output
36
 
37
+ Submitted batch job 7631090
38
 
39
 
40
 
41
  ###############################
42
 
43
+ slurm submission log: 2024-05-19 09:16:21.073119
44
+ created following sbatch script:
45
+
46
+ ###############################
47
+
48
+ #!/bin/bash
49
+
50
+ #SBATCH --account=nlp
51
+ #SBATCH --cpus-per-task=16
52
+ #SBATCH --dependency=afterok:7631150
53
+ #SBATCH --gres=gpu:1
54
+ #SBATCH --job-name=tthrush-job-3010961
55
+ #SBATCH --mem=60G
56
+ #SBATCH --nodelist=sphinx2
57
+ #SBATCH --open-mode=append
58
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
59
+ #SBATCH --partition=sphinx
60
+ #SBATCH --time=14-0
61
+
62
+ # activate your desired anaconda environment
63
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
64
+
65
+ # cd to working directory
66
+ cd .
67
+
68
+ # launch commands
69
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
70
+
71
+ ###############################
72
+
73
+ submission to slurm complete!
74
+
75
+
76
+ ###############################
77
+ slurm submission output
78
+
79
+ Submitted batch job 7631151
80
+
81
+
82
+
83
+ ###############################
84
+
85
+ slurm submission log: 2024-05-19 09:25:08.396129
86
+ created following sbatch script:
87
+
88
+ ###############################
89
+
90
+ #!/bin/bash
91
+
92
+ #SBATCH --account=nlp
93
+ #SBATCH --cpus-per-task=16
94
+ #SBATCH --dependency=afterok:7631221
95
+ #SBATCH --gres=gpu:1
96
+ #SBATCH --job-name=tthrush-job-4399372
97
+ #SBATCH --mem=60G
98
+ #SBATCH --nodelist=sphinx2
99
+ #SBATCH --open-mode=append
100
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
101
+ #SBATCH --partition=sphinx
102
+ #SBATCH --time=14-0
103
+
104
+ # activate your desired anaconda environment
105
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
106
+
107
+ # cd to working directory
108
+ cd .
109
+
110
+ # launch commands
111
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
112
+
113
+ ###############################
114
+
115
+ submission to slurm complete!
116
+
117
+
118
+ ###############################
119
+ slurm submission output
120
+
121
+ Submitted batch job 7631222
122
+
123
+
124
+
125
+ ###############################
126
+
127
+ slurm submission log: 2024-05-19 09:27:21.404018
128
+ created following sbatch script:
129
+
130
+ ###############################
131
+
132
+ #!/bin/bash
133
+
134
+ #SBATCH --account=nlp
135
+ #SBATCH --cpus-per-task=16
136
+ #SBATCH --dependency=afterok:7631284
137
+ #SBATCH --gres=gpu:1
138
+ #SBATCH --job-name=tthrush-job-1884627
139
+ #SBATCH --mem=60G
140
+ #SBATCH --nodelist=sphinx2
141
+ #SBATCH --open-mode=append
142
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
143
+ #SBATCH --partition=sphinx
144
+ #SBATCH --time=14-0
145
+
146
+ # activate your desired anaconda environment
147
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
148
+
149
+ # cd to working directory
150
+ cd .
151
+
152
+ # launch commands
153
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
154
+
155
+ ###############################
156
+
157
+ submission to slurm complete!
158
+
159
+
160
+ ###############################
161
+ slurm submission output
162
+
163
+ Submitted batch job 7631285
164
+
165
+
166
+
167
+ ###############################
168
+
169
+ slurm submission log: 2024-05-19 09:28:18.004148
170
+ created following sbatch script:
171
+
172
+ ###############################
173
+
174
+ #!/bin/bash
175
+
176
+ #SBATCH --account=nlp
177
+ #SBATCH --cpus-per-task=16
178
+ #SBATCH --dependency=afterok:7631348
179
+ #SBATCH --gres=gpu:1
180
+ #SBATCH --job-name=tthrush-job-1636654
181
+ #SBATCH --mem=60G
182
+ #SBATCH --nodelist=sphinx2
183
+ #SBATCH --open-mode=append
184
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
185
+ #SBATCH --partition=sphinx
186
+ #SBATCH --time=14-0
187
+
188
+ # activate your desired anaconda environment
189
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
190
+
191
+ # cd to working directory
192
+ cd .
193
+
194
+ # launch commands
195
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
196
+
197
+ ###############################
198
+
199
+ submission to slurm complete!
200
+
201
+
202
+ ###############################
203
+ slurm submission output
204
+
205
+ Submitted batch job 7631349
206
+
207
+
208
+
209
+ ###############################
210
+
211
+ slurm submission log: 2024-05-19 09:29:20.681949
212
+ created following sbatch script:
213
+
214
+ ###############################
215
+
216
+ #!/bin/bash
217
+
218
+ #SBATCH --account=nlp
219
+ #SBATCH --cpus-per-task=16
220
+ #SBATCH --dependency=afterok:7631409
221
+ #SBATCH --gres=gpu:1
222
+ #SBATCH --job-name=tthrush-job-2759412
223
+ #SBATCH --mem=60G
224
+ #SBATCH --nodelist=sphinx2
225
+ #SBATCH --open-mode=append
226
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
227
+ #SBATCH --partition=sphinx
228
+ #SBATCH --time=14-0
229
+
230
+ # activate your desired anaconda environment
231
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
232
+
233
+ # cd to working directory
234
+ cd .
235
+
236
+ # launch commands
237
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
238
+
239
+ ###############################
240
+
241
+ submission to slurm complete!
242
+
243
+
244
+ ###############################
245
+ slurm submission output
246
+
247
+ Submitted batch job 7631410
248
+
249
+
250
+
251
  ###############################
252
+
253
+ slurm submission log: 2024-05-19 09:30:34.292755
254
+ created following sbatch script:
255
+
256
+ ###############################
257
+
258
+ #!/bin/bash
259
+
260
+ #SBATCH --account=nlp
261
+ #SBATCH --cpus-per-task=16
262
+ #SBATCH --dependency=afterok:7631469
263
+ #SBATCH --gres=gpu:1
264
+ #SBATCH --job-name=tthrush-job-4781467
265
+ #SBATCH --mem=60G
266
+ #SBATCH --nodelist=sphinx2
267
+ #SBATCH --open-mode=append
268
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
269
+ #SBATCH --partition=sphinx
270
+ #SBATCH --time=14-0
271
+
272
+ # activate your desired anaconda environment
273
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
274
+
275
+ # cd to working directory
276
+ cd .
277
+
278
+ # launch commands
279
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
280
+
281
  ###############################
 
282
 
283
+ submission to slurm complete!
284
 
285
 
286
  ###############################
287
+ slurm submission output
288
+
289
+ Submitted batch job 7631470
290
+
291
 
292
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
293
  ###############################
294
+
295
+ slurm submission log: 2024-05-19 09:31:39.869529
 
296
  created following sbatch script:
297
 
298
  ###############################
 
301
 
302
  #SBATCH --account=nlp
303
  #SBATCH --cpus-per-task=16
304
+ #SBATCH --dependency=afterok:7631529
305
  #SBATCH --gres=gpu:1
306
+ #SBATCH --job-name=tthrush-job-2913641
307
  #SBATCH --mem=60G
308
  #SBATCH --nodelist=sphinx2
309
  #SBATCH --open-mode=append
310
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
311
+ #SBATCH --partition=sphinx
312
+ #SBATCH --time=14-0
313
+
314
+ # activate your desired anaconda environment
315
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
316
+
317
+ # cd to working directory
318
+ cd .
319
+
320
+ # launch commands
321
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
322
+
323
+ ###############################
324
+
325
+ submission to slurm complete!
326
+
327
+
328
+ ###############################
329
+ slurm submission output
330
+
331
+ Submitted batch job 7631530
332
+
333
+
334
+
335
+ ###############################
336
+
337
+ slurm submission log: 2024-05-19 09:34:33.468022
338
+ created following sbatch script:
339
+
340
+ ###############################
341
+
342
+ #!/bin/bash
343
+
344
+ #SBATCH --account=nlp
345
+ #SBATCH --cpus-per-task=16
346
+ #SBATCH --dependency=afterok:7631593
347
+ #SBATCH --gres=gpu:1
348
+ #SBATCH --job-name=tthrush-job-2085637
349
+ #SBATCH --mem=60G
350
+ #SBATCH --nodelist=sphinx2
351
+ #SBATCH --open-mode=append
352
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
353
+ #SBATCH --partition=sphinx
354
+ #SBATCH --time=14-0
355
+
356
+ # activate your desired anaconda environment
357
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
358
+
359
+ # cd to working directory
360
+ cd .
361
+
362
+ # launch commands
363
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
364
+
365
+ ###############################
366
+
367
+ submission to slurm complete!
368
+
369
+
370
+ ###############################
371
+ slurm submission output
372
+
373
+ Submitted batch job 7631594
374
+
375
+
376
+
377
+ ###############################
378
+
379
+ slurm submission log: 2024-05-19 09:45:20.755632
380
+ created following sbatch script:
381
+
382
+ ###############################
383
+
384
+ #!/bin/bash
385
+
386
+ #SBATCH --account=nlp
387
+ #SBATCH --cpus-per-task=16
388
+ #SBATCH --dependency=afterok:7631663
389
+ #SBATCH --gres=gpu:1
390
+ #SBATCH --job-name=tthrush-job-1519057
391
+ #SBATCH --mem=60G
392
+ #SBATCH --nodelist=sphinx1
393
+ #SBATCH --open-mode=append
394
+ #SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/eval_job_output.txt
395
  #SBATCH --partition=sphinx
396
  #SBATCH --time=14-0
397
 
398
  # activate your desired anaconda environment
399
+ . /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
400
 
401
  # cd to working directory
402
  cd .
403
 
404
  # launch commands
405
+ srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/perf'
406
 
407
  ###############################
408
 
 
412
  ###############################
413
  slurm submission output
414
 
415
+ Submitted batch job 7631664
416
 
417
 
418
 
logs/events.out.tfevents.1716139274.sphinx2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7205004b6380aaabe1bc990ab9d6ac1255680e7e4e7c874d56b4feff4207f3b8
3
+ size 62915
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fb7ac611e2cfac4a1797cf2eb619c6f4234c50633edc03c4f04aba834a682447
3
  size 56279344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3afd047f9e5e4ca187b5da65a2eb8e83ffe8cd36e546c2a2d02bb834f72e0369
3
  size 56279344
train_job_output.txt CHANGED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:44b10589ffdbb717e36eb18a5c618b90ac401425ea6ab7d2f777359d5bf576c7
3
- size 5048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2bb7b3dce46a3ed23526d06f496a5d2feb3176cdf51b3c78e887057bddf0e2e
3
+ size 5176