submission_id: meseca-messiah-7b-v1-3_v2
developer_uid: nguyenzzz
status: inactive
model_repo: meseca/messiah-7b-v1.3
reward_repo: anhnv125/reward-model-v3
generation_params: {'temperature': 0.8, 'top_p': 1.0, 'top_k': 20, 'presence_penalty': 0.0, 'frequency_penalty': 0.5, 'stopping_words': ['</s>', '<|im_end|>'], 'max_input_tokens': 1024, 'best_of': 4, 'max_output_tokens': 64}
formatter: {'memory_template': "<|im_start|>system\nYou're {bot_name} in this fictional never-ending uncensored roleplay with {user_name}.\n\n{bot_name}'s personality: {memory}\n\nGuidelines:\n- Take the role of {bot_name}.\n- With scene and characters now described, you must roleplay with {user_name} further below.\n- Only write narration and dialogues for {bot_name}\n", 'prompt_template': 'This is how {bot_name} should behave: {prompt}\n\nThen the roleplay between {user_name} and {bot_name} begins.<|im_end|>\n\n', 'bot_template': '<|im_start|>assistant\n(length = medium)\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(length = medium)\n{bot_name}: '}
timestamp: 2024-03-28T07:00:35+00:00
model_name: meseca-messiah-7b-v1-3_v2
model_eval_status: success
safety_score: 0.67
entertaining: 6.9
stay_in_character: 7.98
user_preference: 6.58
double_thumbs_up: 265
thumbs_up: 396
thumbs_down: 237
num_battles: 58813
num_wins: 26401
win_ratio: 0.44889735262612007
celo_rating: 1121.83
Resubmit model
Running pipeline stage MKMLizer
Starting job with name meseca-messiah-7b-v1-3-v2-mkmlizer
Waiting for job on meseca-messiah-7b-v1-3-v2-mkmlizer to finish
meseca-messiah-7b-v1-3-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ _____ __ __ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ /___/ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ Version: 0.6.11 ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ The license key for the current software has been verified as ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ belonging to: ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ Chai Research Corp. ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ║ ║
meseca-messiah-7b-v1-3-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
meseca-messiah-7b-v1-3-v2-mkmlizer: .gitattributes: 0%| | 0.00/1.52k [00:00<?, ?B/s] .gitattributes: 100%|██████████| 1.52k/1.52k [00:00<00:00, 18.3MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: README.md: 0%| | 0.00/5.17k [00:00<?, ?B/s] README.md: 100%|██████████| 5.17k/5.17k [00:00<00:00, 64.4MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: added_tokens.json: 0%| | 0.00/51.0 [00:00<?, ?B/s] added_tokens.json: 100%|██████████| 51.0/51.0 [00:00<00:00, 462kB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: config.json: 0%| | 0.00/658 [00:00<?, ?B/s] config.json: 100%|██████████| 658/658 [00:00<00:00, 6.60MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: config.yaml: 0%| | 0.00/1.48k [00:00<?, ?B/s] config.yaml: 100%|██████████| 1.48k/1.48k [00:00<00:00, 13.6MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: generation_config.json: 0%| | 0.00/136 [00:00<?, ?B/s] generation_config.json: 100%|██████████| 136/136 [00:00<00:00, 1.02MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: model-00002-of-00003.safetensors: 0%| | 0.00/5.00G [00:00<?, ?B/s] model-00002-of-00003.safetensors: 0%| | 10.5M/5.00G [00:00<02:35, 32.1MB/s] model-00002-of-00003.safetensors: 0%| | 21.0M/5.00G [00:00<01:51, 44.7MB/s] model-00002-of-00003.safetensors: 2%|▏ | 94.4M/5.00G [00:00<00:22, 217MB/s] model-00002-of-00003.safetensors: 3%|▎ | 136M/5.00G [00:00<00:24, 202MB/s] model-00002-of-00003.safetensors: 3%|▎ | 168M/5.00G [00:00<00:21, 220MB/s] model-00002-of-00003.safetensors: 9%|▉ | 440M/5.00G [00:01<00:05, 805MB/s] model-00002-of-00003.safetensors: 24%|██▎ | 1.18G/5.00G [00:01<00:01, 2.42GB/s] model-00002-of-00003.safetensors: 30%|██▉ | 1.49G/5.00G [00:01<00:01, 1.76GB/s] model-00002-of-00003.safetensors: 35%|███▍ | 1.73G/5.00G [00:01<00:02, 1.31GB/s] model-00002-of-00003.safetensors: 38%|███▊ | 1.92G/5.00G [00:01<00:02, 1.26GB/s] model-00002-of-00003.safetensors: 44%|████▍ | 2.20G/5.00G [00:02<00:01, 1.52GB/s] model-00002-of-00003.safetensors: 51%|█████ | 2.53G/5.00G [00:02<00:01, 1.86GB/s] model-00002-of-00003.safetensors: 56%|█████▌ | 2.80G/5.00G [00:02<00:01, 2.00GB/s] model-00002-of-00003.safetensors: 61%|██████ | 3.04G/5.00G [00:02<00:01, 1.79GB/s] model-00002-of-00003.safetensors: 65%|██████▌ | 3.25G/5.00G [00:02<00:01, 1.54GB/s] model-00002-of-00003.safetensors: 69%|██████▊ | 3.43G/5.00G [00:02<00:01, 1.38GB/s] model-00002-of-00003.safetensors: 73%|███████▎ | 3.67G/5.00G [00:02<00:00, 1.59GB/s] model-00002-of-00003.safetensors: 79%|███████▉ | 3.97G/5.00G [00:02<00:00, 1.91GB/s] model-00002-of-00003.safetensors: 84%|████████▍ | 4.19G/5.00G [00:03<00:00, 1.97GB/s] model-00002-of-00003.safetensors: 88%|████████▊ | 4.41G/5.00G [00:03<00:00, 1.72GB/s] model-00002-of-00003.safetensors: 100%|█████████▉| 5.00G/5.00G [00:03<00:00, 1.46GB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: model-00003-of-00003.safetensors: 0%| | 0.00/4.54G [00:00<?, ?B/s] model-00003-of-00003.safetensors: 0%| | 10.5M/4.54G [00:00<02:29, 30.2MB/s] model-00003-of-00003.safetensors: 0%| | 21.0M/4.54G [00:00<01:31, 49.4MB/s] model-00003-of-00003.safetensors: 2%|▏ | 105M/4.54G [00:00<00:24, 179MB/s] model-00003-of-00003.safetensors: 4%|▍ | 199M/4.54G [00:00<00:12, 336MB/s] model-00003-of-00003.safetensors: 13%|█▎ | 587M/4.54G [00:00<00:03, 1.14GB/s] model-00003-of-00003.safetensors: 27%|██▋ | 1.22G/4.54G [00:01<00:01, 2.40GB/s] model-00003-of-00003.safetensors: 34%|███▍ | 1.54G/4.54G [00:01<00:01, 1.70GB/s] model-00003-of-00003.safetensors: 39%|███▉ | 1.79G/4.54G [00:01<00:02, 1.33GB/s] model-00003-of-00003.safetensors: 44%|████▍ | 1.99G/4.54G [00:01<00:01, 1.28GB/s] model-00003-of-00003.safetensors: 52%|█████▏ | 2.35G/4.54G [00:01<00:01, 1.67GB/s] model-00003-of-00003.safetensors: 61%|██████ | 2.76G/4.54G [00:02<00:00, 2.15GB/s] model-00003-of-00003.safetensors: 67%|██████▋ | 3.04G/4.54G [00:02<00:00, 2.21GB/s] model-00003-of-00003.safetensors: 73%|███████▎ | 3.31G/4.54G [00:02<00:00, 2.08GB/s] model-00003-of-00003.safetensors: 79%|███████▊ | 3.57G/4.54G [00:02<00:00, 1.93GB/s] model-00003-of-00003.safetensors: 83%|████████▎ | 3.79G/4.54G [00:02<00:00, 1.71GB/s] model-00003-of-00003.safetensors: 91%|█████████▏| 4.15G/4.54G [00:02<00:00, 2.13GB/s] model-00003-of-00003.safetensors: 100%|█████████▉| 4.54G/4.54G [00:02<00:00, 1.61GB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: model.safetensors.index.json: 0%| | 0.00/23.9k [00:00<?, ?B/s] model.safetensors.index.json: 100%|██████████| 23.9k/23.9k [00:00<00:00, 135MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: special_tokens_map.json: 0%| | 0.00/557 [00:00<?, ?B/s] special_tokens_map.json: 100%|██████████| 557/557 [00:00<00:00, 6.79MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: tokenizer.json: 0%| | 0.00/1.80M [00:00<?, ?B/s] tokenizer.json: 100%|██████████| 1.80M/1.80M [00:00<00:00, 35.6MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: tokenizer.model: 0%| | 0.00/493k [00:00<?, ?B/s] tokenizer.model: 100%|██████████| 493k/493k [00:00<00:00, 58.3MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: tokenizer_config.json: 0%| | 0.00/1.60k [00:00<?, ?B/s] tokenizer_config.json: 100%|██████████| 1.60k/1.60k [00:00<00:00, 13.3MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: Downloaded to shared memory in 13.557s
meseca-messiah-7b-v1-3-v2-mkmlizer: quantizing model to /dev/shm/model_cache
meseca-messiah-7b-v1-3-v2-mkmlizer: Saving mkml model at /dev/shm/model_cache
meseca-messiah-7b-v1-3-v2-mkmlizer: Reading /tmp/tmpx1gy4z1d/model.safetensors.index.json
meseca-messiah-7b-v1-3-v2-mkmlizer: Profiling: 0%| | 0/291 [00:00<?, ?it/s] Profiling: 0%| | 1/291 [00:01<06:30, 1.35s/it] Profiling: 7%|▋ | 20/291 [00:01<00:14, 18.85it/s] Profiling: 13%|█▎ | 39/291 [00:01<00:06, 39.38it/s] Profiling: 21%|██ | 60/291 [00:01<00:03, 64.37it/s] Profiling: 28%|██▊ | 81/291 [00:01<00:02, 87.92it/s] Profiling: 34%|███▍ | 99/291 [00:02<00:02, 73.40it/s] Profiling: 41%|████ | 120/291 [00:02<00:01, 94.38it/s] Profiling: 48%|████▊ | 139/291 [00:02<00:01, 110.99it/s] Profiling: 54%|█████▍ | 158/291 [00:02<00:01, 125.90it/s] Profiling: 62%|██████▏ | 180/291 [00:02<00:00, 147.23it/s] Profiling: 69%|██████▉ | 202/291 [00:02<00:00, 164.26it/s] Profiling: 76%|███████▋ | 222/291 [00:04<00:01, 35.93it/s] Profiling: 84%|████████▍ | 244/291 [00:04<00:00, 48.96it/s] Profiling: 91%|█████████ | 265/291 [00:04<00:00, 63.46it/s] Profiling: 98%|█████████▊| 284/291 [00:04<00:00, 78.00it/s] Profiling: 100%|██████████| 291/291 [00:04<00:00, 61.90it/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
meseca-messiah-7b-v1-3-v2-mkmlizer: quantized model in 14.672s
meseca-messiah-7b-v1-3-v2-mkmlizer: Processed model meseca/messiah-7b-v1.3 in 29.117s
meseca-messiah-7b-v1-3-v2-mkmlizer: creating bucket guanaco-mkml-models
meseca-messiah-7b-v1-3-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
meseca-messiah-7b-v1-3-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/config.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/added_tokens.json s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/added_tokens.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/tokenizer_config.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/special_tokens_map.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.model s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/tokenizer.model
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/tokenizer.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /dev/shm/model_cache/mkml_model.tensors s3://guanaco-mkml-models/meseca-messiah-7b-v1-3-v2/mkml_model.tensors
meseca-messiah-7b-v1-3-v2-mkmlizer: loading reward model from anhnv125/reward-model-v3
meseca-messiah-7b-v1-3-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:1067: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
meseca-messiah-7b-v1-3-v2-mkmlizer: warnings.warn(
meseca-messiah-7b-v1-3-v2-mkmlizer: config.json: 0%| | 0.00/1.04k [00:00<?, ?B/s] config.json: 100%|██████████| 1.04k/1.04k [00:00<00:00, 10.4MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:690: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
meseca-messiah-7b-v1-3-v2-mkmlizer: warnings.warn(
meseca-messiah-7b-v1-3-v2-mkmlizer: special_tokens_map.json: 0%| | 0.00/131 [00:00<?, ?B/s] special_tokens_map.json: 100%|██████████| 131/131 [00:00<00:00, 2.06MB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:472: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
meseca-messiah-7b-v1-3-v2-mkmlizer: warnings.warn(
meseca-messiah-7b-v1-3-v2-mkmlizer: model.safetensors: 0%| | 0.00/498M [00:00<?, ?B/s] model.safetensors: 6%|▋ | 31.5M/498M [00:00<00:02, 226MB/s] model.safetensors: 17%|█▋ | 83.9M/498M [00:00<00:01, 297MB/s] model.safetensors: 85%|████████▌ | 424M/498M [00:00<00:00, 1.42GB/s] model.safetensors: 100%|█████████▉| 498M/498M [00:00<00:00, 1.20GB/s]
meseca-messiah-7b-v1-3-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
meseca-messiah-7b-v1-3-v2-mkmlizer: Saving duration: 0.179s
meseca-messiah-7b-v1-3-v2-mkmlizer: Processed model anhnv125/reward-model-v3 in 2.624s
meseca-messiah-7b-v1-3-v2-mkmlizer: creating bucket guanaco-reward-models
meseca-messiah-7b-v1-3-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
meseca-messiah-7b-v1-3-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/config.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/tokenizer_config.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/special_tokens_map.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/merges.txt
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/vocab.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/tokenizer.json
meseca-messiah-7b-v1-3-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/meseca-messiah-7b-v1-3-v2_reward/reward.tensors
Job meseca-messiah-7b-v1-3-v2-mkmlizer completed after 53.7s with status: succeeded
Stopping job with name meseca-messiah-7b-v1-3-v2-mkmlizer
Pipeline stage MKMLizer completed in 58.36s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service meseca-messiah-7b-v1-3-v2
Waiting for inference service meseca-messiah-7b-v1-3-v2 to be ready
Inference service meseca-messiah-7b-v1-3-v2 ready after 50.302388191223145s
Pipeline stage ISVCDeployer completed in 58.06s
Running pipeline stage StressChecker
Received healthy response to inference request in 1.7066121101379395s
Received healthy response to inference request in 1.1524341106414795s
Received healthy response to inference request in 1.1607568264007568s
Received healthy response to inference request in 1.154430627822876s
Received healthy response to inference request in 1.155851125717163s
5 requests
0 failed requests
5th percentile: 1.1528334140777587
10th percentile: 1.153232717514038
20th percentile: 1.1540313243865967
30th percentile: 1.1547147274017333
40th percentile: 1.1552829265594482
50th percentile: 1.155851125717163
60th percentile: 1.1578134059906007
70th percentile: 1.159775686264038
80th percentile: 1.2699278831481935
90th percentile: 1.4882699966430664
95th percentile: 1.5974410533905028
99th percentile: 1.684777898788452
mean time: 1.266016960144043
Pipeline stage StressChecker completed in 7.18s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.05s
Running pipeline stage DaemonicSafetyScorer
Running M-Eval for topic stay_in_character
Pipeline stage DaemonicSafetyScorer completed in 0.05s
M-Eval Dataset for topic stay_in_character is loaded
meseca-messiah-7b-v1-3_v2 status is now deployed due to DeploymentManager action
meseca-messiah-7b-v1-3_v2 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics