sao10k-l3-8b-niitama-v1

submission_id: sao10k-l3-8b-niitama-v1_v2

developer_uid: sao10k

best_of: 16

celo_rating: 1256.78

display_name: niitama2

family_friendly_score: 0.0

formatter: {'memory_template': "<|begin_of_text|><|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}

generation_params: {'temperature': 1.2, 'top_p': 1.0, 'min_p': 0.2, 'top_k': 50, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|end_header_id|>,', '<|eot_id|>,', '\n\n{user_name}', 'You:', '\n\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}

is_internal_developer: False

language_model: Sao10K/L3-8B-Niitama-v1

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: Sao10K/L3-8B-Niitama-v1

model_name: niitama2

model_num_parameters: 8030261248.0

model_repo: Sao10K/L3-8B-Niitama-v1

model_size: 8B

num_battles: 10714

num_wins: 6145

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: Jellywibble/gpt2_xl_pairwise_89m_step_347634

status: torndown

submission_type: basic

timestamp: 2024-07-27T00:24:24+00:00

us_pacific_date: 2024-07-26

win_ratio: 0.5735486279634123

Download Preference Data

Resubmit model

Running pipeline stage MKMLizer
Starting job with name sao10k-l3-8b-niitama-v1-v2-mkmlizer
Waiting for job on sao10k-l3-8b-niitama-v1-v2-mkmlizer to finish
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║     _____            __           __                                ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║       /___/                                                         ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║                                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  Version: 0.9.7                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  https://mk1.ai                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║                                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  belonging to:                                                      ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║                                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  Chai Research Corp.                                                ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ║                                                                     ║
sao10k-l3-8b-niitama-v1-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Downloaded to shared memory in 23.746s
sao10k-l3-8b-niitama-v1-v2-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmppatjvt20, device:0
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
sao10k-l3-8b-niitama-v1-v2-mkmlizer: quantized model in 26.770s
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Processed model Sao10K/L3-8B-Niitama-v1 in 50.516s
sao10k-l3-8b-niitama-v1-v2-mkmlizer: creating bucket guanaco-mkml-models
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
sao10k-l3-8b-niitama-v1-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2/special_tokens_map.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2/config.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2/tokenizer_config.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2/tokenizer.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/sao10k-l3-8b-niitama-v1-v2/flywheel_model.0.safetensors
sao10k-l3-8b-niitama-v1-v2-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
sao10k-l3-8b-niitama-v1-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 2/291 [00:04<11:33,  2.40s/it]
Loading 0:   2%|▏         | 6/291 [00:04<03:05,  1.54it/s]
Loading 0:   4%|▍         | 13/291 [00:05<01:06,  4.20it/s]
Loading 0:   6%|▌         | 18/291 [00:05<00:41,  6.60it/s]
Loading 0:   8%|▊         | 24/291 [00:05<00:27,  9.82it/s]
Loading 0:  11%|█         | 32/291 [00:05<00:16, 15.80it/s]
Loading 0:  13%|█▎        | 38/291 [00:05<00:12, 19.68it/s]
Loading 0:  15%|█▍        | 43/291 [00:05<00:10, 23.17it/s]
Loading 0:  17%|█▋        | 50/291 [00:05<00:08, 29.90it/s]
Loading 0:  19%|█▉        | 56/291 [00:05<00:07, 32.44it/s]
Loading 0:  21%|██        | 61/291 [00:06<00:08, 25.59it/s]
Loading 0:  23%|██▎       | 68/291 [00:06<00:06, 32.05it/s]
Loading 0:  25%|██▌       | 74/291 [00:06<00:06, 33.82it/s]
Loading 0:  27%|██▋       | 79/291 [00:06<00:05, 35.66it/s]
Loading 0:  29%|██▉       | 85/291 [00:06<00:05, 39.85it/s]
Loading 0:  31%|███       | 90/291 [00:06<00:05, 40.07it/s]
Loading 0:  33%|███▎      | 95/291 [00:07<00:04, 42.01it/s]
Loading 0:  35%|███▍      | 101/291 [00:07<00:04, 39.59it/s]
Loading 0:  36%|███▋      | 106/291 [00:07<00:04, 38.96it/s]
Loading 0:  39%|███▉      | 113/291 [00:07<00:03, 45.00it/s]
Loading 0:  41%|████      | 119/291 [00:07<00:03, 43.31it/s]
Loading 0:  43%|████▎     | 124/291 [00:07<00:03, 43.49it/s]
Loading 0:  45%|████▌     | 131/291 [00:07<00:03, 48.71it/s]
Loading 0:  47%|████▋     | 137/291 [00:07<00:03, 45.54it/s]
Loading 0:  49%|████▉     | 142/291 [00:08<00:03, 44.01it/s]
Loading 0:  51%|█████     | 149/291 [00:08<00:02, 48.39it/s]
Loading 0:  53%|█████▎    | 154/291 [00:08<00:02, 47.56it/s]
Loading 0:  55%|█████▍    | 159/291 [00:08<00:03, 38.30it/s]
Loading 0:  57%|█████▋    | 166/291 [00:08<00:03, 32.02it/s]
Loading 0:  58%|█████▊    | 170/291 [00:08<00:03, 33.06it/s]
Loading 0:  60%|██████    | 176/291 [00:08<00:03, 38.03it/s]
Loading 0:  62%|██████▏   | 181/291 [00:09<00:02, 40.62it/s]
Loading 0:  64%|██████▍   | 186/291 [00:09<00:03, 34.53it/s]
Loading 0:  66%|██████▋   | 193/291 [00:09<00:02, 42.14it/s]
Loading 0:  68%|██████▊   | 198/291 [00:09<00:02, 43.05it/s]
Loading 0:  70%|██████▉   | 203/291 [00:09<00:01, 44.02it/s]
Loading 0:  71%|███████▏  | 208/291 [00:09<00:01, 44.55it/s]
Loading 0:  73%|███████▎  | 213/291 [00:09<00:02, 38.23it/s]
Loading 0:  76%|███████▌  | 221/291 [00:10<00:01, 46.71it/s]
Loading 0:  78%|███████▊  | 227/291 [00:10<00:01, 44.00it/s]
Loading 0:  80%|███████▉  | 232/291 [00:10<00:01, 43.07it/s]
Loading 0:  82%|████████▏ | 239/291 [00:10<00:01, 48.07it/s]
Loading 0:  84%|████████▍ | 245/291 [00:10<00:01, 44.77it/s]
Loading 0:  86%|████████▌ | 250/291 [00:10<00:00, 43.70it/s]
Loading 0:  88%|████████▊ | 257/291 [00:10<00:00, 49.50it/s]
Loading 0:  90%|█████████ | 263/291 [00:10<00:00, 46.27it/s]
Loading 0:  92%|█████████▏| 268/291 [00:11<00:00, 31.22it/s]
Loading 0:  95%|█████████▍| 275/291 [00:11<00:00, 37.30it/s]
Loading 0:  96%|█████████▌| 280/291 [00:11<00:00, 39.63it/s]
Loading 0:  98%|█████████▊| 285/291 [00:11<00:00, 34.37it/s]
                                                            
/opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:957: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
sao10k-l3-8b-niitama-v1-v2-mkmlizer:   warnings.warn(
sao10k-l3-8b-niitama-v1-v2-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:785: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
sao10k-l3-8b-niitama-v1-v2-mkmlizer:   warnings.warn(
sao10k-l3-8b-niitama-v1-v2-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.
sao10k-l3-8b-niitama-v1-v2-mkmlizer:   warnings.warn(
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Saving duration: 1.423s
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 10.424s
sao10k-l3-8b-niitama-v1-v2-mkmlizer: creating bucket guanaco-reward-models
sao10k-l3-8b-niitama-v1-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
sao10k-l3-8b-niitama-v1-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/config.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/special_tokens_map.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/tokenizer_config.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/merges.txt
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/vocab.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/tokenizer.json
sao10k-l3-8b-niitama-v1-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/sao10k-l3-8b-niitama-v1-v2_reward/reward.tensors
Job sao10k-l3-8b-niitama-v1-v2-mkmlizer completed after 94.82s with status: succeeded
Stopping job with name sao10k-l3-8b-niitama-v1-v2-mkmlizer
Pipeline stage MKMLizer completed in 95.84s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.10s
Running pipeline stage ISVCDeployer
Creating inference service sao10k-l3-8b-niitama-v1-v2
Waiting for inference service sao10k-l3-8b-niitama-v1-v2 to be ready
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
Inference service sao10k-l3-8b-niitama-v1-v2 ready after 91.1263108253479s
Pipeline stage ISVCDeployer completed in 93.09s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3494958877563477s
Received healthy response to inference request in 1.4390602111816406s
Received healthy response to inference request in 1.4426274299621582s
Received healthy response to inference request in 1.3870782852172852s
Received healthy response to inference request in 1.4886376857757568s
5 requests
0 failed requests
5th percentile: 1.3974746704101562
10th percentile: 1.4078710556030274
20th percentile: 1.4286638259887696
30th percentile: 1.4397736549377442
40th percentile: 1.4412005424499512
50th percentile: 1.4426274299621582
60th percentile: 1.4610315322875977
70th percentile: 1.479435634613037
80th percentile: 1.6608093261718753
90th percentile: 2.0051526069641112
95th percentile: 2.1773242473602292
99th percentile: 2.315061559677124
mean time: 1.6213798999786377
Pipeline stage StressChecker completed in 8.84s
sao10k-l3-8b-niitama-v1_v2 status is now deployed due to DeploymentManager action
sao10k-l3-8b-niitama-v1_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of sao10k-l3-8b-niitama-v1_v2
Running pipeline stage ISVCDeleter
Checking if service sao10k-l3-8b-niitama-v1-v2 is running
Tearing down inference service sao10k-l3-8b-niitama-v1-v2
Service sao10k-l3-8b-niitama-v1-v2 has been torndown
Pipeline stage ISVCDeleter completed in 5.10s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key sao10k-l3-8b-niitama-v1-v2/config.json from bucket guanaco-mkml-models
Deleting key sao10k-l3-8b-niitama-v1-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key sao10k-l3-8b-niitama-v1-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key sao10k-l3-8b-niitama-v1-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key sao10k-l3-8b-niitama-v1-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/config.json from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key sao10k-l3-8b-niitama-v1-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.73s
sao10k-l3-8b-niitama-v1_v2 status is now torndown due to DeploymentManager action