Hugging Face's logo

Datasets:
archit11
/
verl-code-corpus-track-a-file-split

Tasks:
Text Generation
Fill-Mask
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
< 1K
Tags:
code
python
verl
repo-specific-finetuning
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
1
archit11 commited on
Commit
7eedadc
·
verified ·
1 Parent(s): 1b65966

Upload split_manifest.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. split_manifest.json +223 -0
split_manifest.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": [
3
+ "verl__base_config.py",
4
+ "verl__checkpoint_engine__base.py",
5
+ "verl__checkpoint_engine__nixl_checkpoint_engine.py",
6
+ "verl__interactions__gsm8k_interaction.py",
7
+ "verl__interactions__utils__interaction_registry.py",
8
+ "verl__interactions__weather_interaction.py",
9
+ "verl__model_merger__base_model_merger.py",
10
+ "verl__model_merger__fsdp_model_merger.py",
11
+ "verl__model_merger__megatron_model_merger.py",
12
+ "verl__models__llama__megatron__checkpoint_utils__llama_loader.py",
13
+ "verl__models__llama__megatron__checkpoint_utils__llama_saver.py",
14
+ "verl__models__llama__megatron__layers__parallel_decoder.py",
15
+ "verl__models__llama__megatron__layers__parallel_linear.py",
16
+ "verl__models__llama__megatron__layers__parallel_mlp.py",
17
+ "verl__models__llama__megatron__layers__parallel_rmsnorm.py",
18
+ "verl__models__llama__megatron__modeling_llama_megatron.py",
19
+ "verl__models__mcore__config_converter.py",
20
+ "verl__models__mcore__loader.py",
21
+ "verl__models__mcore__model_forward.py",
22
+ "verl__models__mcore__model_forward_1f1b_overlap.py",
23
+ "verl__models__mcore__model_forward_fused.py",
24
+ "verl__models__mcore__model_initializer.py",
25
+ "verl__models__mcore__mtp_patch.py",
26
+ "verl__models__mcore__patch.py",
27
+ "verl__models__mcore__qwen2_5_vl__model.py",
28
+ "verl__models__mcore__qwen2_5_vl__rope_utils.py",
29
+ "verl__models__mcore__qwen2_5_vl__vision_config.py",
30
+ "verl__models__mcore__qwen2_5_vl__vision_model.py",
31
+ "verl__models__mcore__qwen2_5_vl__vision_transformer_block.py",
32
+ "verl__models__mcore__saver.py",
33
+ "verl__models__mcore__util.py",
34
+ "verl__models__mcore__weight_converter.py",
35
+ "verl__models__qwen2__megatron__checkpoint_utils__qwen2_loader.py",
36
+ "verl__models__qwen2__megatron__checkpoint_utils__qwen2_loader_depracated.py",
37
+ "verl__models__qwen2__megatron__layers__parallel_attention.py",
38
+ "verl__models__qwen2__megatron__layers__parallel_mlp.py",
39
+ "verl__models__qwen2__megatron__modeling_qwen2_megatron.py",
40
+ "verl__models__registry.py",
41
+ "verl__models__transformers__dense_common.py",
42
+ "verl__models__transformers__glm4v.py",
43
+ "verl__models__transformers__kimi_vl.py",
44
+ "verl__models__transformers__llama.py",
45
+ "verl__models__transformers__monkey_patch.py",
46
+ "verl__models__transformers__npu_patch.py",
47
+ "verl__models__transformers__qwen2.py",
48
+ "verl__models__transformers__qwen2_vl.py",
49
+ "verl__models__transformers__qwen3_vl.py",
50
+ "verl__models__transformers__tiled_mlp.py",
51
+ "verl__single_controller__base__decorator.py",
52
+ "verl__single_controller__base__worker.py",
53
+ "verl__single_controller__ray__base.py",
54
+ "verl__trainer__config__config.py",
55
+ "verl__trainer__fsdp_sft_trainer.py",
56
+ "verl__trainer__main_eval.py",
57
+ "verl__trainer__main_generation.py",
58
+ "verl__trainer__main_generation_server.py",
59
+ "verl__trainer__main_ppo.py",
60
+ "verl__trainer__ppo__metric_utils.py",
61
+ "verl__trainer__ppo__prefix_grouper_utils.py",
62
+ "verl__trainer__ppo__ray_trainer.py",
63
+ "verl__trainer__sft_trainer.py",
64
+ "verl__utils__activation_offload.py",
65
+ "verl__utils__attention_utils.py",
66
+ "verl__utils__chat_template.py",
67
+ "verl__utils__checkpoint__checkpoint_handler.py",
68
+ "verl__utils__checkpoint__checkpoint_manager.py",
69
+ "verl__utils__checkpoint__fsdp_checkpoint_manager.py",
70
+ "verl__utils__checkpoint__megatron_checkpoint_manager.py",
71
+ "verl__utils__dataset__dataset_utils.py",
72
+ "verl__utils__dataset__multiturn_sft_dataset.py",
73
+ "verl__utils__dataset__rl_dataset.py",
74
+ "verl__utils__dataset__rm_dataset.py",
75
+ "verl__utils__dataset__sft_dataset.py",
76
+ "verl__utils__dataset__vision_utils.py",
77
+ "verl__utils__debug__metrics.py",
78
+ "verl__utils__debug__trajectory_tracker.py",
79
+ "verl__utils__device.py",
80
+ "verl__utils__flops_counter.py",
81
+ "verl__utils__fsdp_utils.py",
82
+ "verl__utils__groupwise.py",
83
+ "verl__utils__hdfs_io.py",
84
+ "verl__utils__kernel__fp8_kernel.py",
85
+ "verl__utils__kernel__kernels.py",
86
+ "verl__utils__kernel__linear_cross_entropy.py",
87
+ "verl__utils__logger__aggregate_logger.py",
88
+ "verl__utils__megatron__memory.py",
89
+ "verl__utils__megatron__pipeline_parallel.py",
90
+ "verl__utils__megatron__router_replay_patch.py",
91
+ "verl__utils__megatron__router_replay_utils.py",
92
+ "verl__utils__megatron__sequence_parallel.py",
93
+ "verl__utils__megatron__tensor_parallel.py",
94
+ "verl__utils__megatron_peft_utils.py",
95
+ "verl__utils__megatron_utils.py",
96
+ "verl__utils__metric__utils.py",
97
+ "verl__utils__model.py",
98
+ "verl__utils__npu_flash_attn_utils.py",
99
+ "verl__utils__profiler__empty_annotations.py",
100
+ "verl__utils__profiler__mstx_profile.py",
101
+ "verl__utils__profiler__nvtx_profile.py",
102
+ "verl__utils__profiler__performance.py",
103
+ "verl__utils__profiler__profile.py",
104
+ "verl__utils__profiler__torch_profile.py",
105
+ "verl__utils__qat__core.py",
106
+ "verl__utils__qat__linear.py",
107
+ "verl__utils__qat__quantizer.py",
108
+ "verl__utils__qat__vllm_patch.py",
109
+ "verl__utils__ray_utils.py",
110
+ "verl__utils__rendezvous__ray_backend.py",
111
+ "verl__utils__reward_score____init__.py",
112
+ "verl__utils__reward_score__geo3k.py",
113
+ "verl__utils__reward_score__gsm8k.py",
114
+ "verl__utils__reward_score__math_dapo.py",
115
+ "verl__utils__reward_score__math_reward.py",
116
+ "verl__utils__reward_score__prime_code__utils.py",
117
+ "verl__utils__reward_score__prime_math__grader.py",
118
+ "verl__utils__reward_score__prime_math__math_normalize.py",
119
+ "verl__utils__reward_score__sandbox_fusion__utils.py",
120
+ "verl__utils__reward_score__search_r1_like_qa_em.py",
121
+ "verl__utils__rollout_skip.py",
122
+ "verl__utils__rollout_trace.py",
123
+ "verl__utils__seqlen_balancing.py",
124
+ "verl__utils__sglang__sglang_fp8_utils.py",
125
+ "verl__utils__tensordict_utils.py",
126
+ "verl__utils__tokenizer.py",
127
+ "verl__utils__torch_dtypes.py",
128
+ "verl__utils__torch_functional.py",
129
+ "verl__utils__transferqueue_utils.py",
130
+ "verl__utils__ulysses.py",
131
+ "verl__utils__vllm__utils.py",
132
+ "verl__utils__vllm__vllm_fp8_utils.py",
133
+ "verl__workers__actor__base.py",
134
+ "verl__workers__actor__megatron_actor.py",
135
+ "verl__workers__config__critic.py",
136
+ "verl__workers__config__engine.py",
137
+ "verl__workers__config__optimizer.py",
138
+ "verl__workers__config__reward.py",
139
+ "verl__workers__config__rollout.py",
140
+ "verl__workers__critic__base.py",
141
+ "verl__workers__critic__megatron_critic.py",
142
+ "verl__workers__engine__base.py",
143
+ "verl__workers__engine__fsdp__transformer_impl.py",
144
+ "verl__workers__engine__fsdp__utils.py",
145
+ "verl__workers__engine__megatron__transformer_impl.py",
146
+ "verl__workers__engine__mindspeed__transformer_impl.py",
147
+ "verl__workers__engine__utils.py",
148
+ "verl__workers__engine__veomni__transformer_impl.py",
149
+ "verl__workers__engine__veomni__utils.py",
150
+ "verl__workers__engine_workers.py",
151
+ "verl__workers__fsdp_workers.py",
152
+ "verl__workers__megatron_workers.py",
153
+ "verl__workers__reward_manager__abstract.py",
154
+ "verl__workers__reward_manager__batch.py",
155
+ "verl__workers__reward_manager__dapo.py",
156
+ "verl__workers__reward_manager__naive.py",
157
+ "verl__workers__reward_manager__prime.py",
158
+ "verl__workers__reward_manager__registry.py",
159
+ "verl__workers__rollout__base.py",
160
+ "verl__workers__rollout__hf_rollout.py",
161
+ "verl__workers__rollout__naive__naive_rollout.py",
162
+ "verl__workers__rollout__replica.py",
163
+ "verl__workers__rollout__schemas.py",
164
+ "verl__workers__rollout__sglang_rollout__async_sglang_server.py",
165
+ "verl__workers__rollout__sglang_rollout__sglang_rollout.py",
166
+ "verl__workers__rollout__sglang_rollout__utils.py",
167
+ "verl__workers__rollout__trtllm_rollout__trtllm_async_server.py",
168
+ "verl__workers__rollout__trtllm_rollout__trtllm_rollout.py",
169
+ "verl__workers__rollout__vllm_rollout__vllm_async_server.py",
170
+ "verl__workers__rollout__vllm_rollout__vllm_rollout.py",
171
+ "verl__workers__sharding_manager__base.py",
172
+ "verl__workers__sharding_manager__fsdp_ulysses.py",
173
+ "verl__workers__utils__losses.py",
174
+ "verl__workers__utils__padding.py"
175
+ ],
176
+ "validation": [
177
+ "verl__checkpoint_engine__hccl_checkpoint_engine.py",
178
+ "verl__models__llama__megatron__layers__parallel_attention.py",
179
+ "verl__models__mcore__registry.py",
180
+ "verl__models__weight_loader_registry.py",
181
+ "verl__trainer__config__algorithm.py",
182
+ "verl__trainer__ppo__core_algos.py",
183
+ "verl__trainer__ppo__rollout_corr_helper.py",
184
+ "verl__trainer__ppo__utils.py",
185
+ "verl__trainer__sft_trainer_ray.py",
186
+ "verl__utils__config.py",
187
+ "verl__utils__fs.py",
188
+ "verl__utils__memory_buffer.py",
189
+ "verl__utils__net_utils.py",
190
+ "verl__utils__profiler__config.py",
191
+ "verl__utils__py_functional.py",
192
+ "verl__utils__reward_score__prime_math____init__.py",
193
+ "verl__utils__tracking.py",
194
+ "verl__workers__actor__dp_actor.py",
195
+ "verl__workers__config__model.py",
196
+ "verl__workers__rollout__utils.py",
197
+ "verl__workers__rollout__vllm_rollout__utils.py"
198
+ ],
199
+ "test": [
200
+ "verl__checkpoint_engine__nccl_checkpoint_engine.py",
201
+ "verl__interactions__base.py",
202
+ "verl__models__llama__megatron__checkpoint_utils__llama_loader_depracated.py",
203
+ "verl__models__mcore__bridge.py",
204
+ "verl__models__mcore__qwen2_5_vl__attention.py",
205
+ "verl__models__qwen2__megatron__checkpoint_utils__qwen2_saver.py",
206
+ "verl__models__qwen2__megatron__layers__parallel_decoder.py",
207
+ "verl__models__qwen2__megatron__layers__parallel_linear.py",
208
+ "verl__models__qwen2__megatron__layers__parallel_rmsnorm.py",
209
+ "verl__protocol.py",
210
+ "verl__single_controller__base__worker_group.py",
211
+ "verl__trainer__ppo__reward.py",
212
+ "verl__utils__import_utils.py",
213
+ "verl__utils__logging_utils.py",
214
+ "verl__utils__megatron__optimizer.py",
215
+ "verl__utils__memory_utils.py",
216
+ "verl__utils__transformers_compat.py",
217
+ "verl__workers__config__actor.py",
218
+ "verl__workers__critic__dp_critic.py",
219
+ "verl__workers__rollout__sglang_rollout__http_server_engine.py",
220
+ "verl__workers__rollout__tokenizer.py"
221
+ ],
222
+ "metrics_file": "results/track_a_verl_metrics_lr1e4_fileholdout_bs8_clip05.json"
223
+ }