rirv938-mistral-24b-dpo_18325

developer_uid: rirv938

submission_id: rirv938-mistral-24b-dpo_18325_v1

model_name: rirv938-mistral-24b-dpo_18325_v1

model_group: rirv938/mistral_24b_dpo_

status: torndown

timestamp: 2025-03-25T02:58:01+00:00

num_battles: 6337

num_wins: 3306

celo_rating: 1297.51

family_friendly_score: 0.5087999999999999

family_friendly_standard_error: 0.007069972560059904

submission_type: basic

model_repo: rirv938/mistral_24b_dpo_kl_40k_beta4_5000_v3

model_architecture: MistralForCausalLM

model_num_parameters: 24096691200.0

best_of: 8

max_input_tokens: 768

max_output_tokens: 64

latencies: [{'batch_size': 1, 'throughput': 0.3869426616266434, 'latency_mean': 2.5842950773239135, 'latency_p50': 2.580156683921814, 'latency_p90': 2.8585268020629884}, {'batch_size': 3, 'throughput': 0.7882557407510209, 'latency_mean': 3.791486920118332, 'latency_p50': 3.795741558074951, 'latency_p90': 4.129536032676697}, {'batch_size': 5, 'throughput': 1.049090051502511, 'latency_mean': 4.740434745550155, 'latency_p50': 4.746817946434021, 'latency_p90': 5.2898513078689575}, {'batch_size': 6, 'throughput': 1.1354550646672013, 'latency_mean': 5.254094160795212, 'latency_p50': 5.23393988609314, 'latency_p90': 5.866954612731933}, {'batch_size': 10, 'throughput': 1.290856871315856, 'latency_mean': 7.671727600097657, 'latency_p50': 7.7079468965530396, 'latency_p90': 8.622195816040039}]

gpu_counts: {'NVIDIA RTX A6000': 1}

display_name: rirv938-mistral-24b-dpo_18325_v1

is_internal_developer: True

language_model: rirv938/mistral_24b_dpo_kl_40k_beta4_5000_v3

model_size: 24B

ranking_group: single

throughput_3p7s: 0.77

us_pacific_date: 2025-03-24

win_ratio: 0.5216979643364368

generation_params: {'temperature': 0.9, 'top_p': 0.9, 'min_p': 0.6, 'top_k': 80, 'presence_penalty': 0.5, 'frequency_penalty': 0.5, 'stopping_words': ['\n', '</s>', '###', 'You:'], 'max_input_tokens': 768, 'best_of': 8, 'max_output_tokens': 64}

formatter: {'memory_template': '[INST]', 'prompt_template': '', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '[/INST]{bot_name}:', 'truncate_by_message': False}

Resubmit model

Shutdown handler not registered because Python interpreter is not running in the main thread
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLizer
Starting job with name rirv938-mistral-24b-dpo-18325-v1-mkmlizer
Waiting for job on rirv938-mistral-24b-dpo-18325-v1-mkmlizer to finish
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: Downloaded to shared memory in 168.417s
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpuh5ll5df, device:0
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: quantized model in 69.065s
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: Processed model rirv938/mistral_24b_dpo_kl_40k_beta4_5000_v3 in 237.487s
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: creating bucket guanaco-mkml-models
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/config.json
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/special_tokens_map.json
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/tokenizer_config.json
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/tokenizer.json
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/flywheel_model.1.safetensors
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/rirv938-mistral-24b-dpo-18325-v1/flywheel_model.0.safetensors
rirv938-mistral-24b-dpo-18325-v1-mkmlizer: 
Loading 0:   0%|          | 0/363 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/363 [00:00<00:13, 27.48it/s]
Loading 0:   2%|▏         | 6/363 [00:00<00:27, 13.14it/s]
Loading 0:   3%|▎         | 10/363 [00:00<00:17, 20.16it/s]
Loading 0:   4%|▎         | 13/363 [00:00<00:25, 13.58it/s]
Loading 0:   4%|▍         | 15/363 [00:01<00:31, 10.97it/s]
Loading 0:   5%|▌         | 19/363 [00:01<00:21, 15.70it/s]
Loading 0:   6%|▌         | 22/363 [00:01<00:23, 14.82it/s]
Loading 0:   7%|▋         | 24/363 [00:01<00:24, 13.66it/s]
Loading 0:   8%|▊         | 28/363 [00:01<00:18, 17.93it/s]
Loading 0:   9%|▉         | 32/363 [00:01<00:15, 21.33it/s]
Loading 0:  10%|▉         | 35/363 [00:02<00:22, 14.45it/s]
Loading 0:  10%|█         | 38/363 [00:02<00:21, 14.83it/s]
Loading 0:  11%|█         | 40/363 [00:02<00:24, 13.03it/s]
Loading 0:  12%|█▏        | 42/363 [00:02<00:26, 12.25it/s]
Loading 0:  13%|█▎        | 46/363 [00:03<00:19, 16.04it/s]
Loading 0:  14%|█▍        | 50/363 [00:03<00:16, 19.48it/s]
Loading 0:  15%|█▍        | 53/363 [00:03<00:22, 13.80it/s]
Loading 0:  15%|█▌        | 55/363 [00:03<00:23, 12.94it/s]
Loading 0:  16%|█▌        | 57/363 [00:03<00:22, 13.73it/s]
Loading 0:  16%|█▋        | 59/363 [00:04<00:27, 10.91it/s]
Loading 0:  18%|█▊        | 64/363 [00:04<00:18, 16.45it/s]
Loading 0:  19%|█▊        | 68/363 [00:04<00:14, 19.75it/s]
Loading 0:  20%|█▉        | 71/363 [00:04<00:21, 13.88it/s]
Loading 0:  20%|██        | 73/363 [00:04<00:22, 13.12it/s]
Loading 0:  21%|██        | 75/363 [00:05<00:20, 13.80it/s]
Loading 0:  21%|██        | 77/363 [00:05<00:25, 11.21it/s]
Loading 0:  23%|██▎       | 82/363 [00:05<00:16, 16.75it/s]
Loading 0:  24%|██▎       | 86/363 [00:05<00:13, 20.14it/s]
Loading 0:  25%|██▍       | 89/363 [00:06<00:19, 14.10it/s]
Loading 0:  25%|██▌       | 91/363 [00:06<00:20, 13.30it/s]
Loading 0:  26%|██▌       | 95/363 [00:06<00:15, 17.10it/s]
Loading 0:  27%|██▋       | 99/363 [00:06<00:13, 20.26it/s]
Loading 0:  28%|██▊       | 102/363 [00:06<00:14, 18.30it/s]
Loading 0:  29%|██▉       | 105/363 [00:07<00:18, 13.67it/s]
Loading 0:  29%|██▉       | 107/363 [00:07<00:20, 12.52it/s]
Loading 0:  30%|███       | 109/363 [00:07<00:20, 12.10it/s]
Loading 0:  31%|███       | 111/363 [00:07<00:19, 12.98it/s]
Loading 0:  31%|███       | 113/363 [00:07<00:23, 10.57it/s]
Loading 0:  33%|███▎      | 118/363 [00:07<00:14, 16.35it/s]
Loading 0:  34%|███▎      | 122/363 [00:08<00:12, 19.67it/s]
Loading 0:  34%|███▍      | 125/363 [00:08<00:17, 13.76it/s]
Loading 0:  35%|███▍      | 127/363 [00:08<00:18, 12.99it/s]
Loading 0:  36%|███▌      | 129/363 [00:08<00:17, 13.76it/s]
Loading 0:  36%|███▌      | 131/363 [00:09<00:20, 11.12it/s]
Loading 0:  37%|███▋      | 136/363 [00:09<00:13, 16.64it/s]
Loading 0:  39%|███▊      | 140/363 [00:09<00:11, 19.92it/s]
Loading 0:  39%|███▉      | 143/363 [00:09<00:15, 13.82it/s]
Loading 0:  40%|███▉      | 145/363 [00:09<00:16, 13.08it/s]
Loading 0:  40%|████      | 147/363 [00:10<00:15, 13.83it/s]
Loading 0:  41%|████      | 149/363 [00:10<00:19, 11.14it/s]
Loading 0:  42%|████▏     | 154/363 [00:10<00:12, 16.59it/s]
Loading 0:  44%|████▎     | 158/363 [00:10<00:10, 19.79it/s]
Loading 0:  44%|████▍     | 161/363 [00:10<00:14, 13.83it/s]
Loading 0:  45%|████▍     | 163/363 [00:11<00:15, 13.04it/s]
Loading 0:  45%|████▌     | 165/363 [00:11<00:14, 13.82it/s]
Loading 0:  46%|████▌     | 167/363 [00:11<00:17, 11.24it/s]
Loading 0:  47%|████▋     | 172/363 [00:11<00:11, 16.76it/s]
Loading 0:  48%|████▊     | 176/363 [00:11<00:09, 20.14it/s]
Loading 0:  49%|████▉     | 179/363 [00:12<00:13, 13.95it/s]
Loading 0:  50%|████▉     | 181/363 [00:12<00:13, 13.10it/s]
Loading 0:  50%|█████     | 183/363 [00:12<00:12, 13.86it/s]
Loading 0:  51%|█████     | 185/363 [00:12<00:15, 11.24it/s]
Loading 0:  52%|█████▏    | 190/363 [00:12<00:10, 16.83it/s]
Loading 0:  53%|█████▎    | 194/363 [00:13<00:08, 20.09it/s]
Loading 0:  54%|█████▍    | 197/363 [00:13<00:12, 13.76it/s]
Loading 0:  55%|█████▍    | 199/363 [00:13<00:12, 13.02it/s]
Loading 0:  55%|█████▌    | 201/363 [00:29<05:05,  1.88s/it]
Loading 0:  56%|█████▌    | 203/363 [00:30<03:54,  1.46s/it]
Loading 0:  57%|█████▋    | 207/363 [00:30<02:16,  1.15it/s]
Loading 0:  58%|█████▊    | 210/363 [00:30<01:34,  1.61it/s]
Loading 0:  59%|█████▊    | 213/363 [00:30<01:08,  2.18it/s]
Loading 0:  59%|█████▉    | 215/363 [00:30<00:56,  2.63it/s]
Loading 0:  60%|█████▉    | 217/363 [00:30<00:45,  3.22it/s]
Loading 0:  60%|██████    | 219/363 [00:31<00:35,  4.06it/s]
Loading 0:  61%|██████    | 221/363 [00:31<00:31,  4.55it/s]
Loading 0:  62%|██████▏   | 226/363 [00:31<00:16,  8.12it/s]
Loading 0:  63%|██████▎   | 230/363 [00:31<00:11, 11.24it/s]
Loading 0:  64%|██████▍   | 233/363 [00:31<00:13,  9.88it/s]
Loading 0:  65%|██████▌   | 236/363 [00:32<00:11, 11.10it/s]
Loading 0:  66%|██████▌   | 238/363 [00:32<00:11, 10.52it/s]
Loading 0:  66%|██████▌   | 240/363 [00:32<00:11, 10.75it/s]
Loading 0:  67%|██████▋   | 244/363 [00:32<00:08, 14.79it/s]
Loading 0:  68%|██████▊   | 248/363 [00:32<00:06, 18.66it/s]
Loading 0:  69%|██████▉   | 251/363 [00:33<00:08, 13.08it/s]
Loading 0:  70%|██████▉   | 253/363 [00:33<00:09, 12.18it/s]
Loading 0:  70%|███████   | 255/363 [00:33<00:08, 12.70it/s]
Loading 0:  71%|███████   | 257/363 [00:33<00:10, 10.43it/s]
Loading 0:  72%|███████▏  | 262/363 [00:33<00:06, 15.93it/s]
Loading 0:  73%|███████▎  | 266/363 [00:34<00:04, 19.41it/s]
Loading 0:  74%|███████▍  | 269/363 [00:34<00:06, 13.51it/s]
Loading 0:  75%|███████▍  | 271/363 [00:34<00:07, 12.81it/s]
Loading 0:  75%|███████▌  | 273/363 [00:34<00:06, 13.34it/s]
Loading 0:  76%|███████▌  | 275/363 [00:35<00:08, 10.94it/s]
Loading 0:  77%|███████▋  | 280/363 [00:35<00:05, 16.40it/s]
Loading 0:  78%|███████▊  | 284/363 [00:35<00:04, 19.57it/s]
Loading 0:  79%|███████▉  | 287/363 [00:35<00:05, 13.45it/s]
Loading 0:  80%|███████▉  | 289/363 [00:35<00:05, 12.38it/s]
Loading 0:  80%|████████  | 291/363 [00:36<00:05, 13.05it/s]
Loading 0:  81%|████████  | 293/363 [00:36<00:06, 10.64it/s]
Loading 0:  82%|████████▏ | 298/363 [00:36<00:04, 16.15it/s]
Loading 0:  83%|████████▎ | 302/363 [00:36<00:03, 19.29it/s]
Loading 0:  84%|████████▍ | 305/363 [00:37<00:04, 13.00it/s]
Loading 0:  85%|████████▍ | 307/363 [00:37<00:04, 12.12it/s]
Loading 0:  85%|████████▌ | 309/363 [00:37<00:04, 12.79it/s]
Loading 0:  86%|████████▌ | 311/363 [00:37<00:04, 10.52it/s]
Loading 0:  87%|████████▋ | 316/363 [00:37<00:02, 16.01it/s]
Loading 0:  88%|████████▊ | 320/363 [00:37<00:02, 19.28it/s]
Loading 0:  89%|████████▉ | 323/363 [00:38<00:02, 13.52it/s]
Loading 0:  90%|████████▉ | 325/363 [00:38<00:03, 12.65it/s]
Loading 0:  90%|█████████ | 327/363 [00:38<00:02, 13.15it/s]
Loading 0:  91%|█████████ | 329/363 [00:39<00:03, 10.41it/s]
Loading 0:  92%|█████████▏| 334/363 [00:39<00:01, 15.89it/s]
Loading 0:  93%|█████████▎| 338/363 [00:39<00:01, 19.14it/s]
Loading 0:  94%|█████████▍| 341/363 [00:39<00:01, 13.26it/s]
Loading 0:  94%|█████████▍| 343/363 [00:39<00:01, 12.48it/s]
Loading 0:  95%|█████████▌| 345/363 [00:40<00:01, 13.29it/s]
Loading 0:  96%|█████████▌| 347/363 [00:40<00:01, 10.89it/s]
Loading 0:  97%|█████████▋| 352/363 [00:40<00:00, 16.55it/s]
Loading 0:  98%|█████████▊| 356/363 [00:40<00:00, 19.90it/s]
Loading 0:  99%|█████████▉| 359/363 [00:48<00:02,  1.37it/s]
Loading 0:  99%|█████████▉| 361/363 [00:48<00:01,  1.64it/s]
                                                            
Job rirv938-mistral-24b-dpo-18325-v1-mkmlizer completed after 286.5s with status: succeeded
Stopping job with name rirv938-mistral-24b-dpo-18325-v1-mkmlizer
Pipeline stage MKMLizer completed in 286.95s
run pipeline stage %s
Running pipeline stage MKMLTemplater
Pipeline stage MKMLTemplater completed in 0.14s
run pipeline stage %s
Running pipeline stage MKMLDeployer
Creating inference service rirv938-mistral-24b-dpo-18325-v1
Waiting for inference service rirv938-mistral-24b-dpo-18325-v1 to be ready
Inference service rirv938-mistral-24b-dpo-18325-v1 ready after 100.3808810710907s
Pipeline stage MKMLDeployer completed in 100.88s
run pipeline stage %s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.958799123764038s
Received healthy response to inference request in 2.452544927597046s
Received healthy response to inference request in 2.510925769805908s
Received healthy response to inference request in 2.377990961074829s
Received healthy response to inference request in 2.473418712615967s
5 requests
0 failed requests
5th percentile: 2.3929017543792725
10th percentile: 2.407812547683716
20th percentile: 2.4376341342926025
30th percentile: 2.45671968460083
40th percentile: 2.4650691986083983
50th percentile: 2.473418712615967
60th percentile: 2.4884215354919434
70th percentile: 2.50342435836792
80th percentile: 2.6005004405975343
90th percentile: 2.779649782180786
95th percentile: 2.869224452972412
99th percentile: 2.940884189605713
mean time: 2.5547358989715576
Pipeline stage StressChecker completed in 14.12s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyTriggerPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage OfflineFamilyFriendlyTriggerPipeline completed in 0.62s
run pipeline stage %s
Running pipeline stage TriggerMKMLProfilingPipeline
run_pipeline:run_in_cloud %s
starting trigger_guanaco_pipeline args=%s
triggered trigger_guanaco_pipeline args=%s
Pipeline stage TriggerMKMLProfilingPipeline completed in 0.70s
Shutdown handler de-registered
rirv938-mistral-24b-dpo_18325_v1 status is now deployed due to DeploymentManager action
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeleter
Skipping teardown as no inference service was successfully deployed
Pipeline stage MKMLProfilerDeleter completed in 0.08s
run pipeline stage %s
Running pipeline stage MKMLProfilerTemplater
Pipeline stage MKMLProfilerTemplater completed in 0.07s
run pipeline stage %s
Running pipeline stage MKMLProfilerDeployer
Creating inference service rirv938-mistral-24b-dpo-18325-v1-profiler
Waiting for inference service rirv938-mistral-24b-dpo-18325-v1-profiler to be ready
Shutdown handler registered
run pipeline %s
run pipeline stage %s
Running pipeline stage OfflineFamilyFriendlyScorer
Evaluating %s Family Friendly Score with %s threads
Pipeline stage OfflineFamilyFriendlyScorer completed in 2931.34s
Shutdown handler de-registered
rirv938-mistral-24b-dpo_18325_v1 status is now inactive due to auto deactivation removed underperforming models
rirv938-mistral-24b-dpo_18325_v1 status is now torndown due to DeploymentManager action

ChatRequest

Bot Name

Generation Params

Prompt Formatter

ChatMessage 1

Sender

Message

User Message