azazelle-l3-hecate-8b-v1-2

developer_uid: Azazelle

submission_id: azazelle-l3-hecate-8b-v1-2_v1

model_name: Azazelle-L3-Hecate-8B-v1_2

model_group: Azazelle/L3-Hecate-8B-v1

status: torndown

timestamp: 2024-07-18T16:03:12+00:00

num_battles: 8835

num_wins: 4597

celo_rating: 1204.14

family_friendly_score: 0.0

submission_type: basic

model_repo: Azazelle/L3-Hecate-8B-v1.2

model_architecture: LlamaForCausalLM

reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2

model_num_parameters: 8030261248.0

best_of: 16

max_input_tokens: 512

max_output_tokens: 64

display_name: Azazelle-L3-Hecate-8B-v1_2

is_internal_developer: False

language_model: Azazelle/L3-Hecate-8B-v1.2

model_size: 8B

ranking_group: single

us_pacific_date: 2024-07-18

win_ratio: 0.520316921335597

generation_params: {'temperature': 1.6, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 80, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['<', '>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\\n\\n{bot_name}'s Persona: {memory}\\n\\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\\n\\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\\n\\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\\n\\n{bot_name}:', 'truncate_by_message': False}

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'}

Resubmit model

Running pipeline stage MKMLizer
Starting job with name azazelle-l3-hecate-8b-v1-2-v1-mkmlizer
Waiting for job on azazelle-l3-hecate-8b-v1-2-v1-mkmlizer to finish
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║     _____            __           __                                ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║       /___/                                                         ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║                                                                     ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  Version: 0.9.5.post3                                               ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  https://mk1.ai                                                     ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║                                                                     ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  belonging to:                                                      ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║                                                                     ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  Chai Research Corp.                                                ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ║                                                                     ║
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
Failed to get response for submission revile-halustheno-v0-2_v1: ('http://revile-halustheno-v0-2-v1-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'read tcp 127.0.0.1:59926->127.0.0.1:8080: read: connection reset by peer\n')
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Downloaded to shared memory in 43.841s
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmp02wd2jmp, device:0
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<10:52,  2.26s/it]
Loading 0:   3%|▎         | 9/291 [00:04<01:49,  2.57it/s]
Loading 0:   4%|▍         | 13/291 [00:05<01:24,  3.30it/s]
Loading 0:   8%|▊         | 22/291 [00:05<00:37,  7.23it/s]
Loading 0:  11%|█         | 31/291 [00:05<00:21, 12.24it/s]
Loading 0:  14%|█▎        | 40/291 [00:05<00:13, 18.37it/s]
Loading 0:  17%|█▋        | 49/291 [00:05<00:09, 25.52it/s]
Loading 0:  20%|█▉        | 58/291 [00:05<00:06, 33.44it/s]
Loading 0:  23%|██▎       | 66/291 [00:06<00:07, 29.24it/s]
Loading 0:  25%|██▌       | 73/291 [00:06<00:09, 21.82it/s]
Loading 0:  28%|██▊       | 81/291 [00:06<00:07, 28.11it/s]
Loading 0:  31%|███       | 90/291 [00:06<00:05, 36.02it/s]
Loading 0:  33%|███▎      | 97/291 [00:07<00:04, 41.36it/s]
Loading 0:  36%|███▌      | 105/291 [00:07<00:03, 47.38it/s]
Loading 0:  39%|███▉      | 113/291 [00:07<00:04, 36.01it/s]
Loading 0:  42%|████▏     | 122/291 [00:07<00:03, 44.18it/s]
Loading 0:  45%|████▌     | 131/291 [00:07<00:03, 51.90it/s]
Loading 0:  47%|████▋     | 138/291 [00:08<00:05, 29.16it/s]
Loading 0:  50%|████▉     | 145/291 [00:08<00:04, 34.44it/s]
Loading 0:  53%|█████▎    | 154/291 [00:08<00:03, 42.98it/s]
Loading 0:  56%|█████▌    | 163/291 [00:08<00:02, 51.33it/s]
Loading 0:  59%|█████▉    | 171/291 [00:08<00:03, 37.64it/s]
Loading 0:  62%|██████▏   | 180/291 [00:09<00:02, 45.84it/s]
Loading 0:  65%|██████▍   | 189/291 [00:09<00:01, 53.79it/s]
Loading 0:  68%|██████▊   | 198/291 [00:09<00:01, 60.55it/s]
Loading 0:  71%|███████   | 207/291 [00:09<00:01, 66.31it/s]
Loading 0:  74%|███████▍  | 215/291 [00:10<00:02, 26.79it/s]
Loading 0:  76%|███████▋  | 222/291 [00:10<00:02, 31.74it/s]
Loading 0:  79%|███████▉  | 231/291 [00:10<00:01, 39.94it/s]
Loading 0:  82%|████████▏ | 240/291 [00:10<00:01, 48.21it/s]
Loading 0:  86%|████████▌ | 249/291 [00:10<00:00, 55.98it/s]
Loading 0:  89%|████████▊ | 258/291 [00:10<00:00, 62.41it/s]
Loading 0:  92%|█████████▏| 267/291 [00:11<00:00, 28.23it/s]
Loading 0:  95%|█████████▍| 276/291 [00:11<00:00, 35.36it/s]
Loading 0:  98%|█████████▊| 285/291 [00:11<00:00, 43.11it/s]
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: quantized model in 27.958s
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Processed model Azazelle/L3-Hecate-8B-v1.2 in 71.799s
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: creating bucket guanaco-mkml-models
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1/config.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1/special_tokens_map.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1/tokenizer_config.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1/tokenizer.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/azazelle-l3-hecate-8b-v1-2-v1/flywheel_model.0.safetensors
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
azazelle-l3-hecate-8b-v1-2-v1-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.
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer:   warnings.warn(
azazelle-l3-hecate-8b-v1-2-v1-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.
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer:   warnings.warn(
azazelle-l3-hecate-8b-v1-2-v1-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.
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer:   warnings.warn(
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Saving duration: 0.408s
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 4.659s
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: creating bucket guanaco-reward-models
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/config.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/special_tokens_map.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/tokenizer_config.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/merges.txt
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/vocab.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/tokenizer.json
azazelle-l3-hecate-8b-v1-2-v1-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/azazelle-l3-hecate-8b-v1-2-v1_reward/reward.tensors
Job azazelle-l3-hecate-8b-v1-2-v1-mkmlizer completed after 111.5s with status: succeeded
Stopping job with name azazelle-l3-hecate-8b-v1-2-v1-mkmlizer
Pipeline stage MKMLizer completed in 113.32s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service azazelle-l3-hecate-8b-v1-2-v1
Waiting for inference service azazelle-l3-hecate-8b-v1-2-v1 to be ready
Inference service azazelle-l3-hecate-8b-v1-2-v1 ready after 193.7188081741333s
Pipeline stage ISVCDeployer completed in 196.04s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.24678111076355s
Received healthy response to inference request in 1.2556993961334229s
Received healthy response to inference request in 1.3090260028839111s
Received healthy response to inference request in 1.2583250999450684s
Received healthy response to inference request in 1.215055227279663s
5 requests
0 failed requests
5th percentile: 1.223184061050415
10th percentile: 1.231312894821167
20th percentile: 1.247570562362671
30th percentile: 1.2562245368957519
40th percentile: 1.2572748184204101
50th percentile: 1.2583250999450684
60th percentile: 1.2786054611206055
70th percentile: 1.2988858222961426
80th percentile: 1.496577024459839
90th percentile: 1.8716790676116943
95th percentile: 2.059230089187622
99th percentile: 2.2092709064483644
mean time: 1.456977367401123
Pipeline stage StressChecker completed in 8.16s
azazelle-l3-hecate-8b-v1-2_v1 status is now deployed due to DeploymentManager action
azazelle-l3-hecate-8b-v1-2_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of azazelle-l3-hecate-8b-v1-2_v1
Running pipeline stage ISVCDeleter
Checking if service azazelle-l3-hecate-8b-v1-2-v1 is running
Tearing down inference service azazelle-l3-hecate-8b-v1-2-v1
Service azazelle-l3-hecate-8b-v1-2-v1 has been torndown
Pipeline stage ISVCDeleter completed in 5.19s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key azazelle-l3-hecate-8b-v1-2-v1/config.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key azazelle-l3-hecate-8b-v1-2-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.36s
azazelle-l3-hecate-8b-v1-2_v1 status is now torndown due to DeploymentManager action