hastagaras-quantmerge

submission_id: hastagaras-quantmerge_v1

developer_uid: Hastagaras

best_of: 16

celo_rating: 1195.54

display_name: hastagaras-quantmerge_v1

family_friendly_score: 0.0

formatter: {'memory_template': '<|im_start|>system\n{memory}\n\n', 'prompt_template': '{prompt}<|im_end|>\n', 'bot_template': '<|im_start|>assistant\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{bot_name}:', 'truncate_by_message': False}

generation_params: {'temperature': 0.85, 'top_p': 0.95, 'min_p': 0.075, 'top_k': 45, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['<|im_end|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64, 'reward_max_token_input': 256}

is_internal_developer: False

language_model: Hastagaras/quantmerge

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: Hastagaras/quantmerge

model_name: hastagaras-quantmerge_v1

model_num_parameters: 8030261248.0

model_repo: Hastagaras/quantmerge

model_size: 8B

num_battles: 13182

num_wins: 6420

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: 2024-07-26T09:25:02+00:00

us_pacific_date: 2024-07-26

win_ratio: 0.48702776513427404

Download Preference Data

Resubmit model

Running pipeline stage MKMLizer
Starting job with name hastagaras-quantmerge-v1-mkmlizer
Waiting for job on hastagaras-quantmerge-v1-mkmlizer to finish
hastagaras-quantmerge-v1-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-quantmerge-v1-mkmlizer: ║     _____            __           __                                ║
hastagaras-quantmerge-v1-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
hastagaras-quantmerge-v1-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
hastagaras-quantmerge-v1-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
hastagaras-quantmerge-v1-mkmlizer: ║       /___/                                                         ║
hastagaras-quantmerge-v1-mkmlizer: ║                                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ║  Version: 0.9.7                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
hastagaras-quantmerge-v1-mkmlizer: ║  https://mk1.ai                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ║                                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ║  The license key for the current software has been verified as      ║
hastagaras-quantmerge-v1-mkmlizer: ║  belonging to:                                                      ║
hastagaras-quantmerge-v1-mkmlizer: ║                                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ║  Chai Research Corp.                                                ║
hastagaras-quantmerge-v1-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
hastagaras-quantmerge-v1-mkmlizer: ║  Expiration: 2024-10-15 23:59:59                                    ║
hastagaras-quantmerge-v1-mkmlizer: ║                                                                     ║
hastagaras-quantmerge-v1-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-quantmerge-v1-mkmlizer: Downloaded to shared memory in 33.249s
hastagaras-quantmerge-v1-mkmlizer: quantizing model to /dev/shm/model_cache, profile:s0, folder:/tmp/tmpr6rlrh5i, device:0
hastagaras-quantmerge-v1-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-quantmerge-v1-mkmlizer: quantized model in 25.519s
hastagaras-quantmerge-v1-mkmlizer: Processed model Hastagaras/quantmerge in 58.769s
hastagaras-quantmerge-v1-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-quantmerge-v1-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-quantmerge-v1-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-quantmerge-v1
hastagaras-quantmerge-v1-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-quantmerge-v1/config.json
hastagaras-quantmerge-v1-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-quantmerge-v1/special_tokens_map.json
hastagaras-quantmerge-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-quantmerge-v1/tokenizer_config.json
hastagaras-quantmerge-v1-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-quantmerge-v1/tokenizer.json
hastagaras-quantmerge-v1-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-quantmerge-v1/flywheel_model.0.safetensors
hastagaras-quantmerge-v1-mkmlizer: loading reward model from Jellywibble/gpt2_xl_pairwise_89m_step_347634
hastagaras-quantmerge-v1-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   2%|▏         | 5/291 [00:00<00:08, 32.94it/s]
Loading 0:   5%|▍         | 14/291 [00:00<00:05, 47.12it/s]
Loading 0:   8%|▊         | 23/291 [00:00<00:05, 49.19it/s]
Loading 0:  11%|█         | 31/291 [00:00<00:04, 56.87it/s]
Loading 0:  13%|█▎        | 37/291 [00:00<00:04, 53.54it/s]
Loading 0:  15%|█▍        | 43/291 [00:00<00:04, 54.87it/s]
Loading 0:  17%|█▋        | 50/291 [00:00<00:04, 51.45it/s]
Loading 0:  20%|██        | 59/291 [00:01<00:04, 53.33it/s]
Loading 0:  23%|██▎       | 67/291 [00:01<00:03, 57.78it/s]
Loading 0:  25%|██▌       | 73/291 [00:01<00:04, 53.65it/s]
Loading 0:  27%|██▋       | 79/291 [00:01<00:03, 53.26it/s]
Loading 0:  29%|██▉       | 85/291 [00:01<00:05, 37.64it/s]
Loading 0:  31%|███▏      | 91/291 [00:01<00:05, 39.13it/s]
Loading 0:  33%|███▎      | 96/291 [00:02<00:04, 40.59it/s]
Loading 0:  35%|███▌      | 103/291 [00:02<00:04, 46.83it/s]
Loading 0:  37%|███▋      | 109/291 [00:02<00:03, 45.54it/s]
Loading 0:  39%|███▉      | 114/291 [00:02<00:03, 45.83it/s]
Loading 0:  42%|████▏     | 122/291 [00:02<00:03, 47.05it/s]
Loading 0:  45%|████▍     | 130/291 [00:02<00:02, 54.07it/s]
Loading 0:  47%|████▋     | 136/291 [00:02<00:03, 50.54it/s]
Loading 0:  49%|████▉     | 142/291 [00:02<00:02, 50.04it/s]
Loading 0:  51%|█████     | 148/291 [00:03<00:02, 52.04it/s]
Loading 0:  53%|█████▎    | 154/291 [00:03<00:02, 49.76it/s]
Loading 0:  55%|█████▍    | 160/291 [00:03<00:02, 51.92it/s]
Loading 0:  57%|█████▋    | 167/291 [00:03<00:02, 48.46it/s]
Loading 0:  61%|██████    | 177/291 [00:03<00:02, 55.47it/s]
Loading 0:  63%|██████▎   | 183/291 [00:03<00:01, 54.30it/s]
Loading 0:  65%|██████▍   | 189/291 [00:04<00:02, 34.93it/s]
Loading 0:  67%|██████▋   | 194/291 [00:04<00:02, 36.71it/s]
Loading 0:  69%|██████▉   | 202/291 [00:04<00:02, 44.14it/s]
Loading 0:  71%|███████▏  | 208/291 [00:04<00:01, 42.00it/s]
Loading 0:  73%|███████▎  | 213/291 [00:04<00:01, 41.50it/s]
Loading 0:  76%|███████▌  | 220/291 [00:04<00:01, 46.14it/s]
Loading 0:  77%|███████▋  | 225/291 [00:04<00:01, 46.72it/s]
Loading 0:  79%|███████▉  | 230/291 [00:04<00:01, 39.83it/s]
Loading 0:  82%|████████▏ | 239/291 [00:05<00:01, 44.25it/s]
Loading 0:  85%|████████▍ | 247/291 [00:05<00:00, 51.13it/s]
Loading 0:  87%|████████▋ | 253/291 [00:05<00:00, 46.76it/s]
Loading 0:  89%|████████▊ | 258/291 [00:05<00:00, 46.05it/s]
Loading 0:  91%|█████████ | 265/291 [00:05<00:00, 50.94it/s]
Loading 0:  93%|█████████▎| 271/291 [00:05<00:00, 48.93it/s]
Loading 0:  95%|█████████▌| 277/291 [00:05<00:00, 50.84it/s]
Loading 0:  97%|█████████▋| 283/291 [00:05<00:00, 46.75it/s]
Loading 0:  99%|█████████▉| 288/291 [00:11<00:00,  3.47it/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.
hastagaras-quantmerge-v1-mkmlizer:   warnings.warn(
hastagaras-quantmerge-v1-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.
hastagaras-quantmerge-v1-mkmlizer:   warnings.warn(
hastagaras-quantmerge-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.
hastagaras-quantmerge-v1-mkmlizer:   warnings.warn(
hastagaras-quantmerge-v1-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-quantmerge-v1-mkmlizer: Saving duration: 1.408s
hastagaras-quantmerge-v1-mkmlizer: Processed model Jellywibble/gpt2_xl_pairwise_89m_step_347634 in 11.462s
hastagaras-quantmerge-v1-mkmlizer: creating bucket guanaco-reward-models
hastagaras-quantmerge-v1-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-quantmerge-v1-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/config.json
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/special_tokens_map.json
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/tokenizer_config.json
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/merges.txt
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/vocab.json
hastagaras-quantmerge-v1-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-quantmerge-v1_reward/tokenizer.json
Job hastagaras-quantmerge-v1-mkmlizer completed after 105.91s with status: succeeded
Stopping job with name hastagaras-quantmerge-v1-mkmlizer
Pipeline stage MKMLizer completed in 106.85s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-quantmerge-v1
Waiting for inference service hastagaras-quantmerge-v1 to be ready
Inference service hastagaras-quantmerge-v1 ready after 80.63103699684143s
Pipeline stage ISVCDeployer completed in 82.19s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.3773741722106934s
Received healthy response to inference request in 1.419588565826416s
Received healthy response to inference request in 1.399359941482544s
Received healthy response to inference request in 1.4244964122772217s
Received healthy response to inference request in 1.4134986400604248s
5 requests
0 failed requests
5th percentile: 1.4021876811981202
10th percentile: 1.4050154209136962
20th percentile: 1.4106709003448485
30th percentile: 1.414716625213623
40th percentile: 1.4171525955200195
50th percentile: 1.419588565826416
60th percentile: 1.4215517044067383
70th percentile: 1.4235148429870605
80th percentile: 1.6150719642639162
90th percentile: 1.9962230682373048
95th percentile: 2.186798620223999
99th percentile: 2.3392590618133546
mean time: 1.60686354637146
Pipeline stage StressChecker completed in 8.72s
hastagaras-quantmerge_v1 status is now deployed due to DeploymentManager action
hastagaras-quantmerge_v1 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of hastagaras-quantmerge_v1
Running pipeline stage ISVCDeleter
Checking if service hastagaras-quantmerge-v1 is running
Tearing down inference service hastagaras-quantmerge-v1
Service hastagaras-quantmerge-v1 has been torndown
Pipeline stage ISVCDeleter completed in 4.23s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key hastagaras-quantmerge-v1/config.json from bucket guanaco-mkml-models
Deleting key hastagaras-quantmerge-v1/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key hastagaras-quantmerge-v1/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key hastagaras-quantmerge-v1/tokenizer.json from bucket guanaco-mkml-models
Deleting key hastagaras-quantmerge-v1/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key hastagaras-quantmerge-v1_reward/config.json from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/merges.txt from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/reward.tensors from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key hastagaras-quantmerge-v1_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.62s
hastagaras-quantmerge_v1 status is now torndown due to DeploymentManager action