jellywibble-lora-120k-pr_2827

submission_id: jellywibble-lora-120k-pr_2827_v5

developer_uid: kay5555

best_of: 16

celo_rating: 1266.06

display_name: nitral-ai-hathor-l3-8b-v-01_v1

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': 0.95, 'top_p': 1.0, 'min_p': 0.08, 'top_k': 40, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n', '<|eot_id|>'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}

is_internal_developer: False

language_model: Jellywibble/lora_120k_pref_data_ep2

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: Jellywibble/lora_120k_pr

model_name: nitral-ai-hathor-l3-8b-v-01_v1

model_num_parameters: 8030261248.0

model_repo: Jellywibble/lora_120k_pref_data_ep2

model_size: 8B

num_battles: 42607

num_wins: 25121

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: ChaiML/gpt2_xl_pairwise_89m_step_347634

status: torndown

submission_type: basic

timestamp: 2024-07-05T04:01:24+00:00

us_pacific_date: 2024-07-04

win_ratio: 0.589597953387941

Resubmit model

Running pipeline stage MKMLizer
Starting job with name jellywibble-lora-120k-pr-2827-v5-mkmlizer
Waiting for job on jellywibble-lora-120k-pr-2827-v5-mkmlizer to finish
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║     _____            __           __                                ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║       /___/                                                         ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  Version: 0.8.14                                                    ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  https://mk1.ai                                                     ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  The license key for the current software has been verified as      ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  belonging to:                                                      ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  Chai Research Corp.                                                ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ║                                                                     ║
jellywibble-lora-120k-pr-2827-v5-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Downloaded to shared memory in 32.578s
jellywibble-lora-120k-pr-2827-v5-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Saving flywheel model at /dev/shm/model_cache
jellywibble-lora-120k-pr-2827-v5-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/291 [00:00<00:38,  7.52it/s]
Loading 0:   1%|▏         | 4/291 [00:00<01:03,  4.49it/s]
Loading 0:   2%|▏         | 5/291 [00:01<01:22,  3.48it/s]
Loading 0:   3%|▎         | 8/291 [00:01<00:42,  6.69it/s]
Loading 0:   3%|▎         | 10/291 [00:01<00:33,  8.30it/s]
Loading 0:   4%|▍         | 12/291 [00:01<00:38,  7.16it/s]
Loading 0:   4%|▍         | 13/291 [00:02<00:41,  6.77it/s]
Loading 0:   7%|▋         | 19/291 [00:02<00:17, 15.11it/s]
Loading 0:   8%|▊         | 23/291 [00:02<00:13, 19.27it/s]
Loading 0:  11%|█▏        | 33/291 [00:02<00:10, 25.70it/s]
Loading 0:  12%|█▏        | 36/291 [00:02<00:10, 25.48it/s]
Loading 0:  14%|█▍        | 41/291 [00:02<00:08, 29.52it/s]
Loading 0:  17%|█▋        | 49/291 [00:02<00:06, 39.48it/s]
Loading 0:  20%|█▉        | 57/291 [00:03<00:04, 47.92it/s]
Loading 0:  22%|██▏       | 63/291 [00:03<00:04, 50.16it/s]
Loading 0:  24%|██▎       | 69/291 [00:03<00:04, 50.03it/s]
Loading 0:  26%|██▌       | 75/291 [00:03<00:04, 51.77it/s]
Loading 0:  28%|██▊       | 81/291 [00:03<00:05, 37.77it/s]
Loading 0:  30%|██▉       | 87/291 [00:03<00:04, 42.24it/s]
Loading 0:  32%|███▏      | 92/291 [00:03<00:04, 42.59it/s]
Loading 0:  34%|███▍      | 100/291 [00:03<00:03, 51.22it/s]
Loading 0:  37%|███▋      | 109/291 [00:04<00:03, 55.80it/s]
Loading 0:  40%|████      | 117/291 [00:04<00:02, 61.51it/s]
Loading 0:  43%|████▎     | 126/291 [00:04<00:02, 62.58it/s]
Loading 0:  46%|████▌     | 133/291 [00:04<00:03, 44.35it/s]
Loading 0:  48%|████▊     | 139/291 [00:04<00:03, 42.89it/s]
Loading 0:  51%|█████     | 147/291 [00:04<00:02, 49.92it/s]
Loading 0:  53%|█████▎    | 154/291 [00:04<00:02, 54.22it/s]
Loading 0:  55%|█████▌    | 161/291 [00:05<00:02, 54.85it/s]
Loading 0:  57%|█████▋    | 167/291 [00:05<00:02, 52.29it/s]
Loading 0:  60%|██████    | 175/291 [00:05<00:01, 58.76it/s]
Loading 0:  64%|██████▍   | 186/291 [00:05<00:02, 45.95it/s]
Loading 0:  66%|██████▌   | 192/291 [00:05<00:02, 43.91it/s]
Loading 0:  68%|██████▊   | 198/291 [00:05<00:01, 46.74it/s]
Loading 0:  70%|███████   | 204/291 [00:05<00:01, 47.67it/s]
Loading 0:  73%|███████▎  | 212/291 [00:06<00:01, 50.77it/s]
Loading 0:  76%|███████▌  | 220/291 [00:06<00:01, 56.73it/s]
Loading 0:  78%|███████▊  | 227/291 [00:06<00:01, 59.04it/s]
Loading 0:  80%|████████  | 234/291 [00:06<00:01, 40.38it/s]
Loading 0:  82%|████████▏ | 240/291 [00:06<00:01, 42.85it/s]
Loading 0:  85%|████████▍ | 247/291 [00:06<00:00, 46.57it/s]
Loading 0:  88%|████████▊ | 255/291 [00:06<00:00, 53.35it/s]
Loading 0:  90%|████████▉ | 261/291 [00:07<00:00, 54.56it/s]
Loading 0:  92%|█████████▏| 267/291 [00:07<00:00, 53.44it/s]
Loading 0:  94%|█████████▍| 274/291 [00:07<00:00, 57.24it/s]
Loading 0:  97%|█████████▋| 282/291 [00:07<00:00, 62.90it/s]
Loading 0:  99%|█████████▉| 289/291 [00:13<00:00,  3.80it/s]
                                                            
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
jellywibble-lora-120k-pr-2827-v5-mkmlizer: quantized model in 79.867s
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Processed model Jellywibble/lora_120k_pref_data_ep2 in 112.446s
jellywibble-lora-120k-pr-2827-v5-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-lora-120k-pr-2827-v5-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5/special_tokens_map.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5/config.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5/tokenizer_config.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5/tokenizer.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-lora-120k-pr-2827-v5/flywheel_model.0.safetensors
jellywibble-lora-120k-pr-2827-v5-mkmlizer: loading reward model from ChaiML/gpt2_xl_pairwise_89m_step_347634
jellywibble-lora-120k-pr-2827-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:919: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-lora-120k-pr-2827-v5-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
jellywibble-lora-120k-pr-2827-v5-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:769: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-lora-120k-pr-2827-v5-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v5-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:468: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
jellywibble-lora-120k-pr-2827-v5-mkmlizer:   warnings.warn(
jellywibble-lora-120k-pr-2827-v5-mkmlizer: 
Downloading shards:   0%|          | 0/2 [00:00<?, ?it/s]
Downloading shards:  50%|█████     | 1/2 [00:05<00:05,  5.24s/it]
Downloading shards: 100%|██████████| 2/2 [00:06<00:00,  2.97s/it]
Downloading shards: 100%|██████████| 2/2 [00:06<00:00,  3.31s/it]
jellywibble-lora-120k-pr-2827-v5-mkmlizer: 
Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]
Loading checkpoint shards:  50%|█████     | 1/2 [00:00<00:00,  2.03it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.40it/s]
Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00,  3.08it/s]
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Saving duration: 1.280s
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Processed model ChaiML/gpt2_xl_pairwise_89m_step_347634 in 10.223s
jellywibble-lora-120k-pr-2827-v5-mkmlizer: creating bucket guanaco-reward-models
jellywibble-lora-120k-pr-2827-v5-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jellywibble-lora-120k-pr-2827-v5-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/config.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/special_tokens_map.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/tokenizer_config.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/vocab.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/merges.txt
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/tokenizer.json
jellywibble-lora-120k-pr-2827-v5-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jellywibble-lora-120k-pr-2827-v5_reward/reward.tensors
Job jellywibble-lora-120k-pr-2827-v5-mkmlizer completed after 226.75s with status: succeeded
Stopping job with name jellywibble-lora-120k-pr-2827-v5-mkmlizer
Pipeline stage MKMLizer completed in 227.46s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.12s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-lora-120k-pr-2827-v5
Waiting for inference service jellywibble-lora-120k-pr-2827-v5 to be ready
Inference service jellywibble-lora-120k-pr-2827-v5 ready after 52.010111570358276s
Pipeline stage ISVCDeployer completed in 58.65s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.2732341289520264s
Received healthy response to inference request in 1.5519213676452637s
Received healthy response to inference request in 1.543630599975586s
Received healthy response to inference request in 1.5136523246765137s
Received healthy response to inference request in 1.5772511959075928s
5 requests
0 failed requests
5th percentile: 1.519647979736328
10th percentile: 1.5256436347961426
20th percentile: 1.5376349449157716
30th percentile: 1.5452887535095214
40th percentile: 1.5486050605773927
50th percentile: 1.5519213676452637
60th percentile: 1.5620532989501954
70th percentile: 1.572185230255127
80th percentile: 1.7164477825164797
90th percentile: 1.994840955734253
95th percentile: 2.1340375423431395
99th percentile: 2.245394811630249
mean time: 1.6919379234313965
Pipeline stage StressChecker completed in 9.34s
jellywibble-lora-120k-pr_2827_v5 status is now deployed due to DeploymentManager action
jellywibble-lora-120k-pr_2827_v5 status is now inactive due to auto deactivation removed underperforming models
jellywibble-lora-120k-pr_2827_v5 status is now deployed due to admin request
jellywibble-lora-120k-pr_2827_v5 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jellywibble-lora-120k-pr_2827_v5
Running pipeline stage ISVCDeleter
Checking if service jellywibble-lora-120k-pr-2827-v5 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 5.18s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2827-v5/config.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v5/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v5/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v5/tokenizer.json from bucket guanaco-mkml-models
Deleting key jellywibble-lora-120k-pr-2827-v5/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/merges.txt from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jellywibble-lora-120k-pr-2827-v5_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 6.10s
jellywibble-lora-120k-pr_2827_v5 status is now torndown due to DeploymentManager action