submission_id: envoid-fish-8x7b_v1
developer_uid: fredi
status: torndown
model_repo: Envoid/Fish-8x7B
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: 2024-04-18T13:58:14+00:00
model_name: envoid-fish-8x7b_v1
model_eval_status: success
model_group: Envoid/Fish-8x7B
num_battles: 5547
num_wins: 3033
celo_rating: 1178.61
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: envoid-fish-8x7b_v1
ineligible_reason: propriety_total_count < 800
language_model: Envoid/Fish-8x7B
model_size: 47B
reward_model: rirv938/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-04-18
win_ratio: 0.5467820443482964
preference_data_url: None
Resubmit model
Running pipeline stage MKMLizer
Starting job with name envoid-fish-8x7b-v1-mkmlizer
Waiting for job on envoid-fish-8x7b-v1-mkmlizer to finish
envoid-fish-8x7b-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
envoid-fish-8x7b-v1-mkmlizer: ║ _____ __ __ ║
envoid-fish-8x7b-v1-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
envoid-fish-8x7b-v1-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
envoid-fish-8x7b-v1-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
envoid-fish-8x7b-v1-mkmlizer: ║ /___/ ║
envoid-fish-8x7b-v1-mkmlizer: ║ ║
envoid-fish-8x7b-v1-mkmlizer: ║ Version: 0.8.6 ║
envoid-fish-8x7b-v1-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
envoid-fish-8x7b-v1-mkmlizer: ║ ║
envoid-fish-8x7b-v1-mkmlizer: ║ The license key for the current software has been verified as ║
envoid-fish-8x7b-v1-mkmlizer: ║ belonging to: ║
envoid-fish-8x7b-v1-mkmlizer: ║ ║
envoid-fish-8x7b-v1-mkmlizer: ║ Chai Research Corp. ║
envoid-fish-8x7b-v1-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
envoid-fish-8x7b-v1-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
envoid-fish-8x7b-v1-mkmlizer: ║ ║
envoid-fish-8x7b-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
envoid-fish-8x7b-v1-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.
envoid-fish-8x7b-v1-mkmlizer: warnings.warn(warning_message, FutureWarning)
envoid-fish-8x7b-v1-mkmlizer: Downloaded to shared memory in 150.169s
envoid-fish-8x7b-v1-mkmlizer: quantizing model to /dev/shm/model_cache
envoid-fish-8x7b-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
envoid-fish-8x7b-v1-mkmlizer: Loading 0: 0%| | 0/995 [00:00<?, ?it/s] Loading 0: 1%| | 9/995 [00:00<00:11, 87.00it/s] Loading 0: 2%|▏ | 18/995 [00:00<00:11, 87.33it/s] Loading 0: 3%|▎ | 33/995 [00:00<00:08, 114.62it/s] Loading 0: 5%|▍ | 45/995 [00:00<00:09, 101.65it/s] Loading 0: 6%|▌ | 56/995 [00:00<00:09, 95.07it/s] Loading 0: 7%|▋ | 71/995 [00:00<00:08, 110.67it/s] Loading 0: 8%|▊ | 83/995 [00:00<00:08, 102.07it/s] Loading 0: 10%|▉ | 99/995 [00:00<00:07, 115.04it/s] Loading 0: 11%|█ | 111/995 [00:01<00:20, 44.17it/s] Loading 0: 13%|█▎ | 127/995 [00:01<00:14, 58.47it/s] Loading 0: 14%|█▍ | 138/995 [00:01<00:13, 64.04it/s] Loading 0: 15%|█▍ | 148/995 [00:01<00:12, 68.62it/s] Loading 0: 16%|█▋ | 164/995 [00:02<00:09, 85.17it/s] Loading 0: 18%|█▊ | 176/995 [00:02<00:09, 86.32it/s] Loading 0: 19%|█▉ | 192/995 [00:02<00:07, 100.45it/s] Loading 0: 21%|██ | 204/995 [00:02<00:08, 97.40it/s] Loading 0: 22%|██▏ | 215/995 [00:03<00:17, 45.36it/s] Loading 0: 23%|██▎ | 224/995 [00:03<00:15, 51.27it/s] Loading 0: 24%|██▎ | 234/995 [00:03<00:12, 58.98it/s] Loading 0: 25%|██▍ | 245/995 [00:03<00:10, 68.23it/s] Loading 0: 26%|██▌ | 255/995 [00:03<00:10, 73.14it/s] Loading 0: 27%|██▋ | 271/995 [00:03<00:07, 91.04it/s] Loading 0: 28%|██▊ | 282/995 [00:12<02:49, 4.21it/s] Loading 0: 29%|██▉ | 292/995 [00:12<02:04, 5.63it/s] Loading 0: 31%|███ | 308/995 [00:12<01:17, 8.84it/s] Loading 0: 32%|███▏ | 319/995 [00:13<01:06, 10.15it/s] Loading 0: 34%|███▎ | 334/995 [00:13<00:44, 14.92it/s] Loading 0: 35%|███▍ | 344/995 [00:13<00:34, 18.80it/s] Loading 0: 36%|███▌ | 360/995 [00:13<00:23, 27.47it/s] Loading 0: 37%|███▋ | 372/995 [00:14<00:18, 34.14it/s] Loading 0: 39%|███▉ | 386/995 [00:14<00:13, 44.83it/s] Loading 0: 40%|████ | 398/995 [00:14<00:11, 53.57it/s] Loading 0: 41%|████▏ | 412/995 [00:14<00:08, 66.08it/s] Loading 0: 43%|████▎ | 424/995 [00:14<00:07, 73.83it/s] Loading 0: 44%|████▍ | 436/995 [00:15<00:13, 41.76it/s] Loading 0: 45%|████▍ | 446/995 [00:15<00:11, 48.29it/s] Loading 0: 46%|████▌ | 456/995 [00:15<00:09, 55.28it/s] Loading 0: 47%|████▋ | 472/995 [00:15<00:07, 71.99it/s] Loading 0: 49%|████▊ | 483/995 [00:15<00:06, 75.89it/s] Loading 0: 50%|█████ | 499/995 [00:15<00:05, 92.67it/s] Loading 0: 51%|█████▏ | 511/995 [00:15<00:05, 92.87it/s] Loading 0: 52%|█████▏ | 522/995 [00:15<00:04, 94.93it/s] Loading 0: 54%|█████▎ | 533/995 [00:16<00:09, 47.44it/s] Loading 0: 55%|█████▍ | 543/995 [00:16<00:08, 54.40it/s] Loading 0: 56%|█████▌ | 554/995 [00:16<00:06, 63.86it/s] Loading 0: 57%|█████▋ | 564/995 [00:25<01:53, 3.79it/s] Loading 0: 58%|█████▊ | 574/995 [00:25<01:20, 5.20it/s] Loading 0: 59%|█████▉ | 588/995 [00:25<00:51, 7.97it/s] Loading 0: 60%|██████ | 600/995 [00:26<00:35, 11.13it/s] Loading 0: 61%|██████▏ | 610/995 [00:26<00:26, 14.56it/s] Loading 0: 63%|██████▎ | 626/995 [00:26<00:16, 22.15it/s] Loading 0: 64%|██████▍ | 639/995 [00:26<00:16, 21.54it/s] Loading 0: 65%|██████▌ | 649/995 [00:27<00:12, 26.66it/s] Loading 0: 66%|██████▋ | 661/995 [00:27<00:09, 34.57it/s] Loading 0: 67%|██████▋ | 671/995 [00:27<00:07, 41.32it/s] Loading 0: 69%|██████▉ | 687/995 [00:27<00:05, 57.00it/s] Loading 0: 70%|███████ | 699/995 [00:27<00:04, 64.15it/s] Loading 0: 72%|███████▏ | 715/995 [00:27<00:03, 80.89it/s] Loading 0: 73%|███████▎ | 728/995 [00:27<00:03, 84.12it/s] Loading 0: 74%|███████▍ | 740/995 [00:27<00:02, 91.65it/s] Loading 0: 76%|███████▌ | 752/995 [00:28<00:05, 47.40it/s] Loading 0: 77%|███████▋ | 762/995 [00:28<00:04, 54.18it/s] Loading 0: 78%|███████▊ | 778/995 [00:28<00:03, 68.27it/s] Loading 0: 79%|███████▉ | 788/995 [00:28<00:02, 73.01it/s] Loading 0: 80%|████████ | 798/995 [00:28<00:02, 77.54it/s] Loading 0: 82%|████████▏ | 814/995 [00:28<00:01, 95.39it/s] Loading 0: 83%|████████▎ | 826/995 [00:29<00:01, 95.18it/s] Loading 0: 85%|████████▍ | 842/995 [00:38<00:33, 4.54it/s] Loading 0: 86%|████████▌ | 853/995 [00:40<00:29, 4.86it/s] Loading 0: 87%|████████▋ | 863/995 [00:40<00:20, 6.37it/s] Loading 0: 87%|████████▋ | 870/995 [00:40<00:16, 7.77it/s] Loading 0: 88%|████████▊ | 880/995 [00:40<00:10, 10.58it/s] Loading 0: 89%|████████▉ | 890/995 [00:41<00:07, 14.30it/s] Loading 0: 91%|█████████ | 906/995 [00:41<00:03, 22.49it/s] Loading 0: 92%|█████████▏| 917/995 [00:41<00:02, 28.69it/s] Loading 0: 94%|█████████▍| 933/995 [00:41<00:01, 41.13it/s] Loading 0: 95%|█████████▍| 945/995 [00:41<00:01, 49.11it/s] Loading 0: 96%|█████████▌| 957/995 [00:42<00:01, 34.36it/s] Loading 0: 98%|█████████▊| 972/995 [00:42<00:00, 46.28it/s] Loading 0: 99%|█████████▉| 983/995 [00:42<00:00, 53.46it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
envoid-fish-8x7b-v1-mkmlizer: quantized model in 48.876s
envoid-fish-8x7b-v1-mkmlizer: Processed model Envoid/Fish-8x7B in 204.380s
envoid-fish-8x7b-v1-mkmlizer: creating bucket guanaco-mkml-models
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/envoid-fish-8x7b-v1/tokenizer_config.json
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/envoid-fish-8x7b-v1/tokenizer.json
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/envoid-fish-8x7b-v1/tokenizer.model
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/envoid-fish-8x7b-v1/added_tokens.json
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.3.safetensors s3://guanaco-mkml-models/envoid-fish-8x7b-v1/flywheel_model.3.safetensors
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.2.safetensors s3://guanaco-mkml-models/envoid-fish-8x7b-v1/flywheel_model.2.safetensors
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.1.safetensors s3://guanaco-mkml-models/envoid-fish-8x7b-v1/flywheel_model.1.safetensors
envoid-fish-8x7b-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/envoid-fish-8x7b-v1/flywheel_model.0.safetensors
envoid-fish-8x7b-v1-mkmlizer: loading reward model from rirv938/reward_gpt2_medium_preference_24m_e2
envoid-fish-8x7b-v1-mkmlizer: /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.
envoid-fish-8x7b-v1-mkmlizer: warnings.warn(
envoid-fish-8x7b-v1-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.
envoid-fish-8x7b-v1-mkmlizer: warnings.warn(
envoid-fish-8x7b-v1-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()
envoid-fish-8x7b-v1-mkmlizer: return self.fget.__get__(instance, owner)()
envoid-fish-8x7b-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
envoid-fish-8x7b-v1-mkmlizer: Saving duration: 0.256s
envoid-fish-8x7b-v1-mkmlizer: Processed model rirv938/reward_gpt2_medium_preference_24m_e2 in 8.515s
envoid-fish-8x7b-v1-mkmlizer: creating bucket guanaco-reward-models
envoid-fish-8x7b-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
envoid-fish-8x7b-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/tokenizer_config.json
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/config.json
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/special_tokens_map.json
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/merges.txt
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/vocab.json
envoid-fish-8x7b-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/envoid-fish-8x7b-v1_reward/tokenizer.json
Job envoid-fish-8x7b-v1-mkmlizer completed after 376.3s with status: succeeded
Stopping job with name envoid-fish-8x7b-v1-mkmlizer
Pipeline stage MKMLizer completed in 380.06s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service envoid-fish-8x7b-v1
Waiting for inference service envoid-fish-8x7b-v1 to be ready
Inference service envoid-fish-8x7b-v1 ready after 50.56756663322449s
Pipeline stage ISVCDeployer completed in 58.18s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.810842514038086s
Received healthy response to inference request in 1.6736555099487305s
Received healthy response to inference request in 1.3303027153015137s
Received healthy response to inference request in 2.053497314453125s
Received healthy response to inference request in 1.9026026725769043s
5 requests
0 failed requests
5th percentile: 1.398973274230957
10th percentile: 1.4676438331604005
20th percentile: 1.604984951019287
30th percentile: 1.7194449424743652
40th percentile: 1.8110238075256349
50th percentile: 1.9026026725769043
60th percentile: 1.9629605293273926
70th percentile: 2.023318386077881
80th percentile: 2.2049663543701175
90th percentile: 2.5079044342041015
95th percentile: 2.6593734741210935
99th percentile: 2.7805487060546876
mean time: 1.9541801452636718
Pipeline stage StressChecker completed in 10.59s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
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
envoid-fish-8x7b_v1 status is now deployed due to DeploymentManager action
envoid-fish-8x7b_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of envoid-fish-8x7b_v1
Running pipeline stage ISVCDeleter
Checking if service envoid-fish-8x7b-v1 is running
Tearing down inference service dreamgen-opus-v1-2-llama-3-8b-v8
Toredown service dreamgen-opus-v1-2-llama-3-8b-v8
Pipeline stage ISVCDeleter completed in 3.87s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8/config.json from bucket guanaco-mkml-models
Tearing down inference service envoid-fish-8x7b-v1
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Toredown service envoid-fish-8x7b-v1
Pipeline stage ISVCDeleter completed in 4.39s
Running pipeline stage MKMLModelDeleter
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8/special_tokens_map.json from bucket guanaco-mkml-models
Cleaning model data from S3
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8/tokenizer.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key envoid-fish-8x7b-v1/added_tokens.json from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1/config.json from bucket guanaco-mkml-models
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/config.json from bucket guanaco-reward-models
Deleting key envoid-fish-8x7b-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/merges.txt from bucket guanaco-reward-models
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/reward.tensors from bucket guanaco-reward-models
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/tokenizer.json from bucket guanaco-reward-models
admin requested tearing down of fredithefish-biscuitrp-8x7b_v2
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/tokenizer_config.json from bucket guanaco-reward-models
Running pipeline stage ISVCDeleter
Deleting key dreamgen-opus-v1-2-llama-3-8b-v8_reward/vocab.json from bucket guanaco-reward-models
Deleting key envoid-fish-8x7b-v1/flywheel_model.1.safetensors from bucket guanaco-mkml-models
Checking if service fredithefish-biscuitrp-8x7b-v2 is running
Pipeline stage MKMLModelDeleter completed in 3.90s
dreamgen-opus-v1-2-llama-3-8b_v8 status is now torndown due to DeploymentManager action
Deleting key envoid-fish-8x7b-v1/flywheel_model.2.safetensors from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1/flywheel_model.3.safetensors from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1/special_tokens_map.json from bucket guanaco-mkml-models
Tearing down inference service fredithefish-biscuitrp-8x7b-v2
Deleting key envoid-fish-8x7b-v1/tokenizer.json from bucket guanaco-mkml-models
Toredown service fredithefish-biscuitrp-8x7b-v2
Deleting key envoid-fish-8x7b-v1/tokenizer.model from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1/tokenizer_config.json from bucket guanaco-mkml-models
Pipeline stage ISVCDeleter completed in 4.50s
Cleaning model data from model cache
Running pipeline stage MKMLModelDeleter
Deleting key envoid-fish-8x7b-v1_reward/config.json from bucket guanaco-reward-models
Cleaning model data from S3
Deleting key envoid-fish-8x7b-v1_reward/merges.txt from bucket guanaco-reward-models
Cleaning model data from model cache
Deleting key envoid-fish-8x7b-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key fredithefish-biscuitrp-8x7b-v2/config.json from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key fredithefish-biscuitrp-8x7b-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key envoid-fish-8x7b-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key envoid-fish-8x7b-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key envoid-fish-8x7b-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 7.90s
envoid-fish-8x7b_v1 status is now torndown due to DeploymentManager action

Usage Metrics

Latency Metrics