submission_id: turboderp-cat-llama-3-7_8684_v21
developer_uid: chai_backend_admin
alignment_samples: 0
best_of: 4
celo_rating: 1207.63
display_name: turboderp-cat-llama-3k
formatter: {'memory_template': '<|im_start|>system\n{memory}<|im_end|>\n', 'prompt_template': '<|im_start|>user\n{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\n{bot_name}: {message}<|im_end|>\n', 'user_template': '<|im_start|>user\n{user_name}: {message}<|im_end|>\n', 'response_template': '<|im_start|>assistant\n{bot_name}:', 'truncate_by_message': False}
generation_params: {'temperature': 0.9, 'top_p': 0.8, 'min_p': 0.0, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 1500, 'best_of': 4, 'max_output_tokens': 64, 'reward_max_token_input': 256}
is_internal_developer: True
language_model: turboderp/Cat-Llama-3-70B-instruct
max_input_tokens: 1500
max_output_tokens: 64
model_architecture: LlamaForCausalLM
model_group: turboderp/Cat-Llama-3-70
model_name: turboderp-cat-llama-3k
model_num_parameters: 70553739264.0
model_repo: turboderp/Cat-Llama-3-70B-instruct
model_size: 71B
num_battles: 30873
num_wins: 16933
propriety_score: 0.7430875576036866
propriety_total_count: 2604.0
ranking_group: single
reward_formatter: {'bot_template': '{bot_name}: {message}\n', 'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'response_template': '{bot_name}:', 'truncate_by_message': False, 'user_template': '{user_name}: {message}\n'}
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
status: torndown
submission_type: basic
timestamp: 2024-07-24T01:44:50+00:00
us_pacific_date: 2024-07-23
win_ratio: 0.548472775564409
Download Preference Data
Resubmit model
Running pipeline stage MKMLizer
Starting job with name turboderp-cat-llama-3-7-8684-v21-mkmlizer
Waiting for job on turboderp-cat-llama-3-7-8684-v21-mkmlizer to finish
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ _____ __ __ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ /___/ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ Version: 0.9.5.post3 ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ https://mk1.ai ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ The license key for the current software has been verified as ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ belonging to: ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ Chai Research Corp. ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ Expiration: 2024-10-15 23:59:59 ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ║ ║
turboderp-cat-llama-3-7-8684-v21-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission blend_dunet_2024-07-19: ('http://undi95-meta-llama-3-70b-6209-v18-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"TypeError : SamplingParameters.__init__() got an unexpected keyword argument \'reward_max_tokens\'"}')
Failed to get response for submission blend_kobem_2024-07-23: ('http://neversleep-noromaid-v0-8068-v133-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:54774->127.0.0.1:8080: read: connection reset by peer\n')
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Downloaded to shared memory in 377.851s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpr34jcpog, device:0
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Saving flywheel model at /dev/shm/model_cache
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Loading 0: 0%| | 0/723 [00:00<?, ?it/s] Loading 0: 1%| | 4/723 [00:00<00:28, 25.53it/s] Loading 0: 1%| | 9/723 [00:00<00:20, 35.43it/s] Loading 0: 2%|▏ | 14/723 [00:00<00:17, 39.72it/s] Loading 0: 3%|▎ | 19/723 [00:00<00:30, 22.91it/s] Loading 0: 3%|▎ | 23/723 [00:00<00:31, 21.99it/s] Loading 0: 4%|▍ | 30/723 [00:01<00:22, 31.00it/s] Loading 0: 5%|▍ | 34/723 [00:01<00:22, 30.63it/s] Loading 0: 6%|▌ | 42/723 [00:01<00:22, 30.63it/s] Loading 0: 6%|▋ | 46/723 [00:01<00:25, 27.04it/s] Loading 0: 7%|▋ | 49/723 [00:01<00:25, 26.13it/s] Loading 0: 7%|▋ | 54/723 [00:01<00:21, 30.46it/s] Loading 0: 8%|▊ | 58/723 [00:02<00:21, 30.46it/s] Loading 0: 9%|▊ | 63/723 [00:02<00:19, 34.42it/s] Loading 0: 9%|▉ | 68/723 [00:02<00:21, 30.07it/s] Loading 0: 10%|▉ | 72/723 [00:02<00:24, 26.83it/s] Loading 0: 11%|█ | 76/723 [00:02<00:23, 27.15it/s] Loading 0: 11%|█ | 81/723 [00:02<00:20, 31.65it/s] Loading 0: 12%|█▏ | 85/723 [00:02<00:20, 31.61it/s] Loading 0: 12%|█▏ | 90/723 [00:02<00:17, 35.66it/s] Loading 0: 13%|█▎ | 94/723 [00:03<00:26, 23.91it/s] Loading 0: 14%|█▎ | 99/723 [00:03<00:21, 28.62it/s] Loading 0: 14%|█▍ | 103/723 [00:03<00:21, 29.41it/s] Loading 0: 15%|█▍ | 108/723 [00:03<00:18, 33.70it/s] Loading 0: 15%|█▌ | 112/723 [00:03<00:18, 32.81it/s] Loading 0: 16%|█▌ | 116/723 [00:03<00:22, 27.36it/s] Loading 0: 17%|█▋ | 120/723 [00:04<00:21, 28.24it/s] Loading 0: 17%|█▋ | 124/723 [00:04<00:20, 28.94it/s] Loading 0: 18%|█▊ | 128/723 [00:18<10:45, 1.09s/it] Loading 0: 18%|█▊ | 130/723 [00:18<08:58, 1.10it/s] Loading 0: 19%|█▊ | 135/723 [00:18<05:34, 1.76it/s] Loading 0: 19%|█▉ | 140/723 [00:19<03:39, 2.66it/s] Loading 0: 20%|█▉ | 144/723 [00:19<02:48, 3.44it/s] Loading 0: 20%|██ | 148/723 [00:19<02:03, 4.64it/s] Loading 0: 21%|██ | 153/723 [00:19<01:24, 6.72it/s] Loading 0: 22%|██▏ | 157/723 [00:19<01:05, 8.63it/s] Loading 0: 22%|██▏ | 162/723 [00:19<00:47, 11.90it/s] Loading 0: 23%|██▎ | 168/723 [00:20<00:36, 15.07it/s] Loading 0: 24%|██▍ | 172/723 [00:20<00:33, 16.26it/s] Loading 0: 24%|██▍ | 176/723 [00:20<00:31, 17.36it/s] Loading 0: 25%|██▌ | 183/723 [00:20<00:21, 24.72it/s] Loading 0: 26%|██▌ | 187/723 [00:20<00:20, 25.89it/s] Loading 0: 27%|██▋ | 194/723 [00:20<00:19, 27.10it/s] Loading 0: 27%|██▋ | 198/723 [00:21<00:20, 25.32it/s] Loading 0: 28%|██▊ | 202/723 [00:21<00:20, 25.95it/s] Loading 0: 29%|██▊ | 207/723 [00:21<00:16, 30.38it/s] Loading 0: 29%|██▉ | 211/723 [00:21<00:16, 30.82it/s] Loading 0: 30%|██▉ | 216/723 [00:21<00:14, 34.88it/s] Loading 0: 30%|███ | 220/723 [00:21<00:20, 24.64it/s] Loading 0: 31%|███ | 225/723 [00:21<00:16, 29.45it/s] Loading 0: 32%|███▏ | 229/723 [00:22<00:16, 29.97it/s] Loading 0: 32%|███▏ | 234/723 [00:22<00:14, 34.07it/s] Loading 0: 33%|███▎ | 238/723 [00:22<00:14, 33.05it/s] Loading 0: 33%|███▎ | 242/723 [00:22<00:17, 28.17it/s] Loading 0: 34%|███▍ | 246/723 [00:22<00:16, 28.83it/s] Loading 0: 35%|███▍ | 250/723 [00:22<00:16, 29.34it/s] Loading 0: 35%|███▌ | 256/723 [00:22<00:14, 31.90it/s] Loading 0: 36%|███▌ | 261/723 [00:23<00:12, 35.65it/s] Loading 0: 37%|███▋ | 266/723 [00:23<00:11, 38.78it/s] Loading 0: 37%|███▋ | 271/723 [00:37<06:53, 1.09it/s] Loading 0: 38%|███▊ | 274/723 [00:37<05:28, 1.37it/s] Loading 0: 39%|███▊ | 279/723 [00:38<03:40, 2.01it/s] Loading 0: 39%|███▉ | 283/723 [00:38<02:42, 2.71it/s] Loading 0: 40%|███▉ | 288/723 [00:38<01:50, 3.93it/s] Loading 0: 41%|████ | 294/723 [00:38<01:15, 5.68it/s] Loading 0: 41%|████ | 298/723 [00:38<01:00, 6.98it/s] Loading 0: 42%|████▏ | 302/723 [00:38<00:49, 8.52it/s] Loading 0: 43%|████▎ | 309/723 [00:38<00:31, 13.10it/s] Loading 0: 43%|████▎ | 314/723 [00:39<00:25, 15.95it/s] Loading 0: 44%|████▍ | 320/723 [00:39<00:21, 18.62it/s] Loading 0: 45%|████▍ | 324/723 [00:39<00:20, 19.01it/s] Loading 0: 45%|████▌ | 328/723 [00:39<00:19, 20.78it/s] Loading 0: 46%|████▌ | 333/723 [00:39<00:15, 25.39it/s] Loading 0: 47%|████▋ | 337/723 [00:39<00:14, 26.76it/s] Loading 0: 47%|████▋ | 342/723 [00:39<00:12, 31.08it/s] Loading 0: 48%|████▊ | 346/723 [00:40<00:16, 23.20it/s] Loading 0: 49%|████▊ | 351/723 [00:40<00:13, 27.94it/s] Loading 0: 49%|████▉ | 355/723 [00:40<00:12, 28.88it/s] Loading 0: 50%|████▉ | 360/723 [00:40<00:10, 33.30it/s] Loading 0: 50%|█████ | 364/723 [00:40<00:11, 32.56it/s] Loading 0: 51%|█████ | 368/723 [00:40<00:12, 27.88it/s] Loading 0: 51%|█████▏ | 372/723 [00:41<00:12, 28.76it/s] Loading 0: 52%|█████▏ | 376/723 [00:41<00:11, 29.30it/s] Loading 0: 53%|█████▎ | 382/723 [00:41<00:10, 31.87it/s] Loading 0: 54%|█████▎ | 387/723 [00:41<00:09, 35.70it/s] Loading 0: 54%|█████▍ | 392/723 [00:41<00:08, 38.78it/s] Loading 0: 55%|█████▍ | 397/723 [00:41<00:11, 27.93it/s] Loading 0: 55%|█████▌ | 401/723 [00:56<05:06, 1.05it/s] Loading 0: 56%|█████▋ | 408/723 [00:56<03:03, 1.72it/s] Loading 0: 57%|█████▋ | 413/723 [00:56<02:11, 2.36it/s] Loading 0: 58%|█████▊ | 420/723 [00:56<01:25, 3.54it/s] Loading 0: 59%|█████▊ | 424/723 [00:56<01:08, 4.37it/s] Loading 0: 59%|█████▉ | 428/723 [00:56<00:54, 5.45it/s] Loading 0: 60%|██████ | 435/723 [00:57<00:34, 8.38it/s] Loading 0: 61%|██████ | 440/723 [00:57<00:26, 10.62it/s] Loading 0: 62%|██████▏ | 446/723 [00:57<00:20, 13.24it/s] Loading 0: 62%|██████▏ | 450/723 [00:57<00:18, 14.51it/s] Loading 0: 63%|██████▎ | 454/723 [00:57<00:16, 16.61it/s] Loading 0: 63%|██████▎ | 459/723 [00:57<00:12, 20.81it/s] Loading 0: 64%|██████▍ | 463/723 [00:58<00:11, 22.86it/s] Loading 0: 65%|██████▍ | 468/723 [00:58<00:09, 27.54it/s] Loading 0: 65%|██████▌ | 472/723 [00:58<00:11, 21.57it/s] Loading 0: 66%|██████▌ | 477/723 [00:58<00:09, 26.18it/s] Loading 0: 67%|██████▋ | 481/723 [00:58<00:08, 27.44it/s] Loading 0: 67%|██████▋ | 486/723 [00:58<00:07, 32.02it/s] Loading 0: 68%|██████▊ | 490/723 [00:58<00:07, 31.71it/s] Loading 0: 68%|██████▊ | 494/723 [00:59<00:08, 27.41it/s] Loading 0: 69%|██████▉ | 498/723 [00:59<00:07, 28.46it/s] Loading 0: 69%|██████▉ | 502/723 [00:59<00:07, 29.12it/s] Loading 0: 70%|███████ | 507/723 [00:59<00:06, 33.84it/s] Loading 0: 71%|███████ | 511/723 [00:59<00:06, 32.75it/s] Loading 0: 71%|███████ | 515/723 [00:59<00:06, 34.52it/s] Loading 0: 72%|███████▏ | 520/723 [00:59<00:06, 30.46it/s] Loading 0: 72%|███████▏ | 524/723 [00:59<00:06, 30.50it/s] Loading 0: 73%|███████▎ | 528/723 [01:00<00:07, 27.35it/s] Loading 0: 74%|███████▍ | 535/723 [01:00<00:05, 32.26it/s] Loading 0: 75%|███████▍ | 539/723 [01:14<02:47, 1.10it/s] Loading 0: 76%|███████▌ | 546/723 [01:14<01:41, 1.75it/s] Loading 0: 76%|███████▌ | 550/723 [01:14<01:17, 2.24it/s] Loading 0: 76%|███████▋ | 553/723 [01:14<01:01, 2.76it/s] Loading 0: 77%|███████▋ | 558/723 [01:15<00:41, 3.99it/s] Loading 0: 78%|███████▊ | 562/723 [01:15<00:30, 5.24it/s] Loading 0: 78%|███████▊ | 567/723 [01:15<00:21, 7.40it/s] Loading 0: 79%|███████▉ | 572/723 [01:15<00:15, 9.56it/s] Loading 0: 80%|███████▉ | 576/723 [01:15<00:13, 11.19it/s] Loading 0: 80%|████████ | 580/723 [01:15<00:10, 13.45it/s] Loading 0: 81%|████████ | 585/723 [01:15<00:07, 17.67it/s] Loading 0: 81%|████████▏ | 589/723 [01:16<00:06, 20.15it/s] Loading 0: 82%|████████▏ | 594/723 [01:16<00:05, 24.91it/s] Loading 0: 83%|████████▎ | 598/723 [01:16<00:06, 20.26it/s] Loading 0: 83%|████████▎ | 603/723 [01:16<00:04, 25.09it/s] Loading 0: 84%|████████▍ | 607/723 [01:16<00:04, 26.71it/s] Loading 0: 85%|████████▍ | 612/723 [01:16<00:03, 31.41it/s] Loading 0: 85%|████████▌ | 616/723 [01:16<00:03, 31.25it/s] Loading 0: 86%|████████▌ | 620/723 [01:17<00:03, 27.28it/s] Loading 0: 86%|████████▋ | 624/723 [01:17<00:03, 28.37it/s] Loading 0: 87%|████████▋ | 628/723 [01:17<00:03, 29.19it/s] Loading 0: 88%|████████▊ | 633/723 [01:17<00:02, 33.23it/s] Loading 0: 88%|████████▊ | 637/723 [01:17<00:02, 32.07it/s] Loading 0: 89%|████████▉ | 642/723 [01:17<00:02, 36.15it/s] Loading 0: 89%|████████▉ | 646/723 [01:17<00:02, 29.92it/s] Loading 0: 90%|████████▉ | 650/723 [01:17<00:02, 30.23it/s] Loading 0: 90%|█████████ | 654/723 [01:18<00:02, 27.25it/s] Loading 0: 91%|█████████▏| 661/723 [01:18<00:01, 32.12it/s] Loading 0: 92%|█████████▏| 666/723 [01:18<00:01, 35.67it/s] Loading 0: 93%|█████████▎| 672/723 [01:18<00:01, 33.31it/s] Loading 0: 93%|█████████▎| 674/723 [01:32<00:01, 33.31it/s] Loading 0: 93%|█████████▎| 675/723 [01:32<00:45, 1.05it/s] Loading 0: 94%|█████████▍| 679/723 [01:32<00:30, 1.43it/s] Loading 0: 95%|█████████▍| 684/723 [01:33<00:18, 2.09it/s] Loading 0: 95%|█████████▌| 688/723 [01:33<00:12, 2.80it/s] Loading 0: 96%|█████████▌| 693/723 [01:33<00:07, 4.04it/s] Loading 0: 97%|█████████▋| 698/723 [01:33<00:04, 5.49it/s] Loading 0: 97%|█████████▋| 702/723 [01:33<00:03, 6.82it/s] Loading 0: 98%|█████████▊| 706/723 [01:33<00:01, 8.63it/s] Loading 0: 98%|█████████▊| 711/723 [01:33<00:01, 11.84it/s] Loading 0: 99%|█████████▉| 715/723 [01:34<00:00, 14.31it/s] Loading 0: 100%|█████████▉| 720/723 [01:34<00:00, 18.61it/s] Loading 0: 100%|█████████▉| 722/723 [01:44<00:00, 18.61it/s] Loading 0: 100%|██████████| 723/723 [01:44<00:00, 1.25it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
turboderp-cat-llama-3-7-8684-v21-mkmlizer: quantized model in 121.023s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Processed model turboderp/Cat-Llama-3-70B-instruct in 498.874s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: creating bucket guanaco-mkml-models
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
turboderp-cat-llama-3-7-8684-v21-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/special_tokens_map.json
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/tokenizer_config.json
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/tokenizer.json
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.5.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.5.safetensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.0.safetensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.2.safetensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.1.safetensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.3.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.3.safetensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /dev/shm/model_cache/flywheel_model.4.safetensors s3://guanaco-mkml-models/turboderp-cat-llama-3-7-8684-v21/flywheel_model.4.safetensors
Connection pool is full, discarding connection: %s. Connection pool size: %s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
turboderp-cat-llama-3-7-8684-v21-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:950: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
turboderp-cat-llama-3-7-8684-v21-mkmlizer: warnings.warn(
turboderp-cat-llama-3-7-8684-v21-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:778: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
turboderp-cat-llama-3-7-8684-v21-mkmlizer: warnings.warn(
turboderp-cat-llama-3-7-8684-v21-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:469: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
turboderp-cat-llama-3-7-8684-v21-mkmlizer: warnings.warn(
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Saving duration: 0.328s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.364s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: creating bucket guanaco-reward-models
turboderp-cat-llama-3-7-8684-v21-mkmlizer: Bucket 's3://guanaco-reward-models/' created
turboderp-cat-llama-3-7-8684-v21-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/turboderp-cat-llama-3-7-8684-v21_reward
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
Connection pool is full, discarding connection: %s. Connection pool size: %s
turboderp-cat-llama-3-7-8684-v21-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/turboderp-cat-llama-3-7-8684-v21_reward/reward.tensors
Job turboderp-cat-llama-3-7-8684-v21-mkmlizer completed after 560.33s with status: succeeded
Stopping job with name turboderp-cat-llama-3-7-8684-v21-mkmlizer
Pipeline stage MKMLizer completed in 561.41s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.09s
Running pipeline stage ISVCDeployer
Creating inference service turboderp-cat-llama-3-7-8684-v21
Waiting for inference service turboderp-cat-llama-3-7-8684-v21 to be ready
Inference service turboderp-cat-llama-3-7-8684-v21 ready after 90.87875127792358s
Pipeline stage ISVCDeployer completed in 92.90s
Running pipeline stage StressChecker
Failed to get response for submission blend_dunet_2024-07-19: ('http://undi95-meta-llama-3-70b-6209-v18-predictor.tenant-chaiml-guanaco.k.chaiverse.com/v1/models/GPT-J-6B-lit-v2:predict', '{"error":"TypeError : SamplingParameters.__init__() got an unexpected keyword argument \'reward_max_tokens\'"}')
Received healthy response to inference request in 5.358253240585327s
Received healthy response to inference request in 4.266397476196289s
Received healthy response to inference request in 4.293575763702393s
Received healthy response to inference request in 4.314587593078613s
Received healthy response to inference request in 4.244871377944946s
5 requests
0 failed requests
5th percentile: 4.2491765975952145
10th percentile: 4.253481817245484
20th percentile: 4.262092256546021
30th percentile: 4.27183313369751
40th percentile: 4.282704448699951
50th percentile: 4.293575763702393
60th percentile: 4.301980495452881
70th percentile: 4.310385227203369
80th percentile: 4.523320722579956
90th percentile: 4.940786981582642
95th percentile: 5.1495201110839846
99th percentile: 5.316506614685059
mean time: 4.4955370903015135
Pipeline stage StressChecker completed in 23.22s
turboderp-cat-llama-3-7_8684_v21 status is now deployed due to DeploymentManager action
admin requested tearing down of turboderp-cat-llama-3-7_8684_v21
Running pipeline stage ISVCDeleter
Checking if service turboderp-cat-llama-3-7-8684-v21 is running
Tearing down inference service turboderp-cat-llama-3-7-8684-v21
Service turboderp-cat-llama-3-7-8684-v21 has been torndown
Pipeline stage ISVCDeleter completed in 5.26s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key turboderp-cat-llama-3-7-8684-v21/config.json from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.4.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/flywheel_model.5.safetensors from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/tokenizer.json from bucket guanaco-mkml-models
Deleting key turboderp-cat-llama-3-7-8684-v21/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/config.json from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/merges.txt from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/reward.tensors from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key turboderp-cat-llama-3-7-8684-v21_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 12.18s
turboderp-cat-llama-3-7_8684_v21 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics