submission_id: fredithefish-fishxinstruct_v2
developer_uid: fredi
status: torndown
model_repo: Fredithefish/FishxInstruct
reward_repo: rirv938/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.0, 'top_p': 0.9, 'min_p': 0.0, 'top_k': 70, 'presence_penalty': 0.9, 'frequency_penalty': 0.9, 'stopping_words': ['\n', '</s>', '<|user|>', '###'], 'max_input_tokens': 512, 'best_of': 8, 'max_output_tokens': 64}
formatter: {'memory_template': '<s>[INST] This is an entertaining conversation. You are {bot_name} who has the persona: {memory}.\nEngage in a chat with {user_name} while staying in character. Try to flirt with {user_name}. Engage in *roleplay* actions. Describe the scene dramatically. \n\n', 'prompt_template': 'Introduction:\n{prompt}\n\n', 'bot_template': '{bot_name}: {message}</s>', 'user_template': '[INST] {user_name}: {message} [/INST]', 'response_template': '{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'memory_template': 'Memory: {memory}\n', 'prompt_template': '{prompt}\n', 'bot_template': 'Bot: {message}\n', 'user_template': 'User: {message}\n', 'response_template': 'Bot:', 'truncate_by_message': False}
timestamp: 2024-04-13T17:23:26+00:00
model_name: fredithefish-fishxinstruct_v1
model_eval_status: success
model_group: Fredithefish/FishxInstru
num_battles: 51397
num_wins: 28657
celo_rating: 1189.36
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: MixtralForCausalLM
model_num_parameters: 46702792704.0
best_of: 8
max_input_tokens: 512
max_output_tokens: 64
display_name: fredithefish-fishxinstruct_v1
ineligible_reason: propriety_total_count < 800
language_model: Fredithefish/FishxInstruct
model_size: 47B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-13
win_ratio: 0.5575617253925327
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name fredithefish-fishxinstruct-v2-mkmlizer
Waiting for job on fredithefish-fishxinstruct-v2-mkmlizer to finish
fredithefish-fishxinstruct-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
fredithefish-fishxinstruct-v2-mkmlizer: ║ _____ __ __ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ /___/ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ Version: 0.8.6 ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ The license key for the current software has been verified as ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ belonging to: ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ Chai Research Corp. ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
fredithefish-fishxinstruct-v2-mkmlizer: ║ ║
fredithefish-fishxinstruct-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
fredithefish-fishxinstruct-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
fredithefish-fishxinstruct-v2-mkmlizer: warnings.warn(warning_message, FutureWarning)
fredithefish-fishxinstruct-v2-mkmlizer: Downloaded to shared memory in 174.429s
fredithefish-fishxinstruct-v2-mkmlizer: quantizing model to /dev/shm/model_cache
fredithefish-fishxinstruct-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
fredithefish-fishxinstruct-v2-mkmlizer: quantized model in 52.250s
fredithefish-fishxinstruct-v2-mkmlizer: Processed model Fredithefish/FishxInstruct in 232.322s
fredithefish-fishxinstruct-v2-mkmlizer: creating bucket guanaco-mkml-models
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/config.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/tokenizer_config.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/tokenizer.model
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/special_tokens_map.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/tokenizer.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.3.safetensors s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/flywheel_model.3.safetensors
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/flywheel_model.0.safetensors
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/flywheel_model.2.safetensors
fredithefish-fishxinstruct-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/fredithefish-fishxinstruct-v2/flywheel_model.1.safetensors
fredithefish-fishxinstruct-v2-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
fredithefish-fishxinstruct-v2-mkmlizer: Loading 0: 0%| | 0/995 [00:00<?, ?it/s] Loading 0: 2%|▏ | 16/995 [00:00<00:06, 147.81it/s] Loading 0: 3%|▎ | 31/995 [00:00<00:13, 70.22it/s] Loading 0: 4%|▍ | 41/995 [00:00<00:12, 74.35it/s] Loading 0: 5%|▌ | 50/995 [00:01<00:53, 17.73it/s] Loading 0: 7%|▋ | 65/995 [00:01<00:32, 28.24it/s] Loading 0: 7%|▋ | 74/995 [00:02<00:29, 31.12it/s] Loading 0: 8%|▊ | 83/995 [00:02<00:24, 37.55it/s] Loading 0: 9%|▉ | 91/995 [00:02<00:25, 34.95it/s] Loading 0: 10%|█ | 100/995 [00:02<00:21, 41.59it/s] Loading 0: 11%|█ | 107/995 [00:03<00:25, 34.16it/s] Loading 0: 11%|█▏ | 113/995 [00:03<00:23, 37.64it/s] Loading 0: 12%|█▏ | 119/995 [00:03<00:23, 36.62it/s] Loading 0: 13%|█▎ | 132/995 [00:03<00:16, 53.26it/s] Loading 0: 14%|█▍ | 140/995 [00:03<00:18, 45.34it/s] Loading 0: 15%|█▌ | 151/995 [00:03<00:14, 56.40it/s] Loading 0: 16%|█▌ | 159/995 [00:04<00:17, 46.99it/s] Loading 0: 17%|█▋ | 168/995 [00:04<00:15, 54.31it/s] Loading 0: 18%|█▊ | 175/995 [00:04<00:16, 49.65it/s] Loading 0: 19%|█▊ | 186/995 [00:04<00:13, 61.26it/s] Loading 0: 19%|█▉ | 194/995 [00:04<00:15, 50.29it/s] Loading 0: 20%|██ | 203/995 [00:04<00:13, 56.94it/s] Loading 0: 21%|██ | 210/995 [00:04<00:17, 45.17it/s] Loading 0: 23%|██▎ | 225/995 [00:05<00:11, 64.60it/s] Loading 0: 24%|██▍ | 237/995 [00:05<00:12, 59.99it/s] Loading 0: 25%|██▍ | 246/995 [00:05<00:11, 65.01it/s] Loading 0: 26%|██▌ | 254/995 [00:05<00:14, 52.79it/s] Loading 0: 27%|██▋ | 267/995 [00:05<00:10, 67.06it/s] Loading 0: 28%|██▊ | 283/995 [00:06<00:10, 67.19it/s] Loading 0: 29%|██▉ | 291/995 [00:15<03:04, 3.81it/s] Loading 0: 30%|███ | 300/995 [00:15<02:25, 4.79it/s] Loading 0: 31%|███ | 309/995 [00:15<01:46, 6.43it/s] Loading 0: 32%|███▏ | 319/995 [00:16<01:18, 8.62it/s] Loading 0: 33%|███▎ | 328/995 [00:16<00:57, 11.50it/s] Loading 0: 34%|███▍ | 340/995 [00:16<00:41, 15.69it/s] Loading 0: 35%|███▌ | 349/995 [00:16<00:32, 20.15it/s] Loading 0: 36%|███▌ | 359/995 [00:16<00:26, 23.95it/s] Loading 0: 37%|███▋ | 368/995 [00:16<00:20, 29.90it/s] Loading 0: 38%|███▊ | 380/995 [00:17<00:17, 34.84it/s] Loading 0: 39%|███▉ | 389/995 [00:17<00:14, 41.60it/s] Loading 0: 41%|████ | 403/995 [00:17<00:12, 47.64it/s] Loading 0: 42%|████▏ | 414/995 [00:17<00:10, 56.83it/s] Loading 0: 42%|████▏ | 422/995 [00:17<00:11, 49.33it/s] Loading 0: 44%|████▍ | 436/995 [00:17<00:08, 64.62it/s] Loading 0: 45%|████▌ | 449/995 [00:18<00:08, 62.47it/s] Loading 0: 46%|████▌ | 460/995 [00:18<00:07, 70.80it/s] Loading 0: 47%|████▋ | 469/995 [00:18<00:09, 58.23it/s] Loading 0: 48%|████▊ | 478/995 [00:18<00:08, 60.36it/s] Loading 0: 49%|████▉ | 489/995 [00:18<00:08, 58.45it/s] Loading 0: 50%|█████ | 499/995 [00:18<00:07, 65.73it/s] Loading 0: 51%|█████ | 508/995 [00:19<00:09, 54.02it/s] Loading 0: 52%|█████▏ | 517/995 [00:19<00:07, 60.22it/s] Loading 0: 53%|█████▎ | 525/995 [00:19<00:09, 50.98it/s] Loading 0: 54%|█████▎ | 534/995 [00:19<00:07, 58.35it/s] Loading 0: 55%|█████▍ | 546/995 [00:19<00:07, 56.63it/s] Loading 0: 56%|█████▌ | 555/995 [00:19<00:07, 62.24it/s] Loading 0: 56%|█████▋ | 562/995 [00:29<02:21, 3.06it/s] Loading 0: 57%|█████▋ | 569/995 [00:29<01:50, 3.86it/s] Loading 0: 58%|█████▊ | 580/995 [00:29<01:10, 5.92it/s] Loading 0: 59%|█████▉ | 592/995 [00:29<00:46, 8.71it/s] Loading 0: 60%|██████ | 601/995 [00:29<00:33, 11.63it/s] Loading 0: 62%|██████▏ | 615/995 [00:30<00:22, 16.88it/s] Loading 0: 63%|██████▎ | 626/995 [00:30<00:16, 22.59it/s] Loading 0: 64%|██████▎ | 634/995 [00:30<00:14, 24.64it/s] Loading 0: 65%|██████▍ | 643/995 [00:30<00:11, 30.65it/s] Loading 0: 66%|██████▋ | 661/995 [00:30<00:07, 42.19it/s] Loading 0: 68%|██████▊ | 672/995 [00:30<00:06, 50.60it/s] Loading 0: 68%|██████▊ | 681/995 [00:31<00:06, 46.31it/s] Loading 0: 69%|██████▉ | 690/995 [00:31<00:05, 52.83it/s] Loading 0: 70%|███████ | 701/995 [00:31<00:05, 52.55it/s] Loading 0: 72%|███████▏ | 712/995 [00:31<00:04, 62.55it/s] Loading 0: 72%|███████▏ | 721/995 [00:31<00:05, 53.91it/s] Loading 0: 73%|███████▎ | 730/995 [00:31<00:04, 60.32it/s] Loading 0: 74%|███████▍ | 738/995 [00:32<00:05, 50.21it/s] Loading 0: 76%|███████▌ | 753/995 [00:32<00:03, 68.58it/s] Loading 0: 77%|███████▋ | 764/995 [00:32<00:03, 63.14it/s] Loading 0: 78%|███████▊ | 773/995 [00:32<00:03, 68.15it/s] Loading 0: 79%|███████▊ | 782/995 [00:32<00:03, 56.52it/s] Loading 0: 80%|███████▉ | 793/995 [00:32<00:03, 66.47it/s] Loading 0: 81%|████████ | 806/995 [00:33<00:02, 63.92it/s] Loading 0: 82%|████████▏ | 815/995 [00:33<00:02, 68.45it/s] Loading 0: 83%|████████▎ | 827/995 [00:33<00:02, 63.55it/s] Loading 0: 84%|████████▍ | 838/995 [00:33<00:02, 72.79it/s] Loading 0: 85%|████████▌ | 847/995 [00:43<00:42, 3.52it/s] Loading 0: 86%|████████▌ | 856/995 [00:43<00:29, 4.76it/s] Loading 0: 87%|████████▋ | 867/995 [00:43<00:19, 6.42it/s] Loading 0: 88%|████████▊ | 878/995 [00:43<00:12, 9.12it/s] Loading 0: 89%|████████▉ | 886/995 [00:43<00:09, 11.18it/s] Loading 0: 90%|█████████ | 896/995 [00:44<00:06, 15.26it/s] Loading 0: 91%|█████████ | 907/995 [00:44<00:04, 19.58it/s] Loading 0: 93%|█████████▎| 923/995 [00:44<00:02, 30.03it/s] Loading 0: 94%|█████████▎| 932/995 [00:44<00:01, 31.79it/s] Loading 0: 95%|█████████▍| 941/995 [00:44<00:01, 38.12it/s] Loading 0: 95%|█████████▌| 949/995 [00:46<00:02, 15.60it/s] Loading 0: 96%|█████████▋| 960/995 [00:46<00:01, 21.64it/s] Loading 0: 98%|█████████▊| 972/995 [00:46<00:00, 27.87it/s] Loading 0: 99%|█████████▉| 983/995 [00:46<00:00, 35.98it/s] Loading 0: 100%|█████████▉| 991/995 [00:46<00:00, 35.85it/s] /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:1096: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-fishxinstruct-v2-mkmlizer: warnings.warn(
fredithefish-fishxinstruct-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:720: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-fishxinstruct-v2-mkmlizer: warnings.warn(
fredithefish-fishxinstruct-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:466: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
fredithefish-fishxinstruct-v2-mkmlizer: warnings.warn(
fredithefish-fishxinstruct-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
fredithefish-fishxinstruct-v2-mkmlizer: return self.fget.__get__(instance, owner)()
fredithefish-fishxinstruct-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
fredithefish-fishxinstruct-v2-mkmlizer: Saving duration: 0.243s
fredithefish-fishxinstruct-v2-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 15.310s
fredithefish-fishxinstruct-v2-mkmlizer: creating bucket guanaco-reward-models
fredithefish-fishxinstruct-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
fredithefish-fishxinstruct-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/config.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/special_tokens_map.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/tokenizer_config.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/merges.txt
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/vocab.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/tokenizer.json
fredithefish-fishxinstruct-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/fredithefish-fishxinstruct-v2_reward/reward.tensors
Job fredithefish-fishxinstruct-v2-mkmlizer completed after 280.84s with status: succeeded
Stopping job with name fredithefish-fishxinstruct-v2-mkmlizer
Pipeline stage MKMLizer completed in 284.40s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service fredithefish-fishxinstruct-v2
Waiting for inference service fredithefish-fishxinstruct-v2 to be ready
Failed to get response for submission blend_gulur_2024-04-10: ('http://a100-khanhnto-khanhnto-v54-v2-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', '')
Inference service fredithefish-fishxinstruct-v2 ready after 60.40457606315613s
Pipeline stage ISVCDeployer completed in 67.33s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.6629409790039062s
Received healthy response to inference request in 2.4579415321350098s
Received healthy response to inference request in 1.8497390747070312s
Received healthy response to inference request in 2.1618969440460205s
Received healthy response to inference request in 1.7538676261901855s
5 requests
0 failed requests
5th percentile: 1.7730419158935546
10th percentile: 1.7922162055969237
20th percentile: 1.8305647850036622
30th percentile: 1.912170648574829
40th percentile: 2.037033796310425
50th percentile: 2.1618969440460205
60th percentile: 2.2803147792816163
70th percentile: 2.3987326145172116
80th percentile: 2.498941421508789
90th percentile: 2.5809412002563477
95th percentile: 2.621941089630127
99th percentile: 2.6547410011291506
mean time: 2.1772772312164306
Pipeline stage StressChecker completed in 12.70s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.04s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.04s
M-Eval Dataset for topic stay_in_character is loaded
fredithefish-fishxinstruct_v2 status is now deployed due to DeploymentManager action
fredithefish-fishxinstruct_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of fredithefish-fishxinstruct_v2
Running pipeline stage ISVCDeleter
Checking if service fredithefish-fishxinstruct-v2 is running
Tearing down inference service fredithefish-fishxinstruct-v2
Toredown service fredithefish-fishxinstruct-v2
Pipeline stage ISVCDeleter completed in 4.33s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key fredithefish-fishxinstruct-v2/config.json from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Connection pool is full, discarding connection: %s
Deleting key fredithefish-fishxinstruct-v2/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/tokenizer.model from bucket guanaco-mkml-models
Deleting key fredithefish-fishxinstruct-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key fredithefish-fishxinstruct-v2_reward/config.json from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key fredithefish-fishxinstruct-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 6.15s
fredithefish-fishxinstruct_v2 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics