jellywibble-qlora-90k-pr_7056

submission_id: jellywibble-qlora-90k-pr_7056_v2

developer_uid: Jellywibble

best_of: 16

celo_rating: 1199.44

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: True

language_model: Jellywibble/qlora_90k_pref_data_ep1

max_input_tokens: 512

max_output_tokens: 64

model_architecture: LlamaForCausalLM

model_group: Jellywibble/qlora_90k_pr

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

model_num_parameters: 8030261248.0

model_repo: Jellywibble/qlora_90k_pref_data_ep1

model_size: 8B

num_battles: 13583

num_wins: 6996

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/reward_gpt2_medium_preference_24m_e2

status: torndown

submission_type: basic

timestamp: 2024-07-03T22:37:48+00:00

us_pacific_date: 2024-07-03

win_ratio: 0.515055584186115

Resubmit model

Running pipeline stage MKMLizer
Starting job with name jellywibble-qlora-90k-pr-7056-v2-mkmlizer
Waiting for job on jellywibble-qlora-90k-pr-7056-v2-mkmlizer to finish
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║     _____            __           __                                ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║    / _/ /_ ___    __/ /  ___ ___ / /                                ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║   / _/ / // / |/|/ / _ \/ -_) -_) /                                 ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  /_//_/\_, /|__,__/_//_/\__/\__/_/                                  ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║       /___/                                                         ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║                                                                     ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  Version: 0.8.14                                                    ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  Copyright 2023 MK ONE TECHNOLOGIES Inc.                            ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  https://mk1.ai                                                     ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║                                                                     ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  The license key for the current software has been verified as      ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  belonging to:                                                      ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║                                                                     ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  Chai Research Corp.                                                ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f                   ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║  Expiration: 2024-07-15 23:59:59                                    ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ║                                                                     ║
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Downloaded to shared memory in 34.098s
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: quantizing model to /dev/shm/model_cache
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Saving flywheel model at /dev/shm/model_cache
Failed to get response for submission neversleep-noromaid-v0-_8068_v36: ('http://neversleep-noromaid-v0-8068-v36-predictor-default.tenant-chaiml-guanaco.knative.ord1.coreweave.cloud/v1/models/GPT-J-6B-lit-v2:predict', 'timed out dialing after 20.82s\n')
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: 
Loading 0:   0%|          | 0/291 [00:00<?, ?it/s]
Loading 0:   1%|          | 3/291 [00:00<00:11, 24.85it/s]
Loading 0:   4%|▍         | 13/291 [00:00<00:04, 60.81it/s]
Loading 0:   9%|▉         | 27/291 [00:00<00:02, 92.37it/s]
Loading 0:  14%|█▎        | 40/291 [00:00<00:02, 100.80it/s]
Loading 0:  18%|█▊        | 53/291 [00:00<00:02, 109.59it/s]
Loading 0:  23%|██▎       | 67/291 [00:00<00:01, 114.52it/s]
Loading 0:  28%|██▊       | 81/291 [00:00<00:01, 120.81it/s]
Loading 0:  32%|███▏      | 94/291 [00:01<00:02, 68.88it/s] 
Loading 0:  37%|███▋      | 109/291 [00:01<00:02, 84.38it/s]
Loading 0:  42%|████▏     | 121/291 [00:01<00:01, 90.48it/s]
Loading 0:  46%|████▋     | 135/291 [00:01<00:01, 101.41it/s]
Loading 0:  51%|█████     | 148/291 [00:01<00:01, 105.57it/s]
Loading 0:  56%|█████▌    | 162/291 [00:01<00:01, 114.33it/s]
Loading 0:  60%|██████    | 176/291 [00:01<00:00, 120.92it/s]
Loading 0:  65%|██████▍   | 189/291 [00:02<00:01, 75.86it/s] 
Loading 0:  69%|██████▉   | 202/291 [00:02<00:01, 85.08it/s]
Loading 0:  74%|███████▍  | 216/291 [00:02<00:00, 96.53it/s]
Loading 0:  79%|███████▊  | 229/291 [00:02<00:00, 102.05it/s]
Loading 0:  83%|████████▎ | 242/291 [00:02<00:00, 108.80it/s]
Loading 0:  88%|████████▊ | 256/291 [00:02<00:00, 113.68it/s]
Loading 0:  92%|█████████▏| 269/291 [00:02<00:00, 115.67it/s]
Loading 0:  97%|█████████▋| 282/291 [00:02<00:00, 117.11it/s]
                                                             
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: quantized model in 20.746s
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Processed model Jellywibble/qlora_90k_pref_data_ep1 in 54.845s
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: creating bucket guanaco-mkml-models
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2/special_tokens_map.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2/config.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2/tokenizer_config.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2/tokenizer.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/jellywibble-qlora-90k-pr-7056-v2/flywheel_model.0.safetensors
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
jellywibble-qlora-90k-pr-7056-v2-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-qlora-90k-pr-7056-v2-mkmlizer:   warnings.warn(
jellywibble-qlora-90k-pr-7056-v2-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-qlora-90k-pr-7056-v2-mkmlizer:   warnings.warn(
jellywibble-qlora-90k-pr-7056-v2-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-qlora-90k-pr-7056-v2-mkmlizer:   warnings.warn(
jellywibble-qlora-90k-pr-7056-v2-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-qlora-90k-pr-7056-v2-mkmlizer:   warnings.warn(
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Saving duration: 0.295s
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.187s
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: creating bucket guanaco-reward-models
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: Bucket 's3://guanaco-reward-models/' created
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/special_tokens_map.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/config.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/tokenizer_config.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/merges.txt
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/vocab.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/tokenizer.json
jellywibble-qlora-90k-pr-7056-v2-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/jellywibble-qlora-90k-pr-7056-v2_reward/reward.tensors
Job jellywibble-qlora-90k-pr-7056-v2-mkmlizer completed after 93.64s with status: succeeded
Stopping job with name jellywibble-qlora-90k-pr-7056-v2-mkmlizer
Pipeline stage MKMLizer completed in 94.53s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.11s
Running pipeline stage ISVCDeployer
Creating inference service jellywibble-qlora-90k-pr-7056-v2
Waiting for inference service jellywibble-qlora-90k-pr-7056-v2 to be ready
Inference service jellywibble-qlora-90k-pr-7056-v2 ready after 40.275344371795654s
Pipeline stage ISVCDeployer completed in 47.26s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.0660572052001953s
Received healthy response to inference request in 1.4017839431762695s
Received healthy response to inference request in 1.3323185443878174s
Retrying (%r) after connection broken by '%r': %s
Received healthy response to inference request in 1.3022661209106445s
Received healthy response to inference request in 1.387484073638916s
5 requests
0 failed requests
5th percentile: 1.308276605606079
10th percentile: 1.3142870903015136
20th percentile: 1.3263080596923829
30th percentile: 1.3433516502380372
40th percentile: 1.3654178619384765
50th percentile: 1.387484073638916
60th percentile: 1.3932040214538575
70th percentile: 1.398923969268799
80th percentile: 1.5346385955810549
90th percentile: 1.800347900390625
95th percentile: 1.93320255279541
99th percentile: 2.039486274719238
mean time: 1.4979819774627685
Pipeline stage StressChecker completed in 8.25s
jellywibble-qlora-90k-pr_7056_v2 status is now deployed due to DeploymentManager action
jellywibble-qlora-90k-pr_7056_v2 status is now inactive due to auto deactivation removed underperforming models
admin requested tearing down of jellywibble-qlora-90k-pr_7056_v2
Running pipeline stage ISVCDeleter
Checking if service jellywibble-qlora-90k-pr-7056-v2 is running
Skipping teardown as no inference service was found
Pipeline stage ISVCDeleter completed in 3.89s
Running pipeline stage MKMLModelDeleter
Cleaning model data from S3
Cleaning model data from model cache
Deleting key jellywibble-qlora-90k-pr-7056-v2/config.json from bucket guanaco-mkml-models
Deleting key jellywibble-qlora-90k-pr-7056-v2/flywheel_model.0.safetensors from bucket guanaco-mkml-models
Deleting key jellywibble-qlora-90k-pr-7056-v2/special_tokens_map.json from bucket guanaco-mkml-models
Deleting key jellywibble-qlora-90k-pr-7056-v2/tokenizer.json from bucket guanaco-mkml-models
Deleting key jellywibble-qlora-90k-pr-7056-v2/tokenizer_config.json from bucket guanaco-mkml-models
Cleaning model data from model cache
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/config.json from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/merges.txt from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/reward.tensors from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/special_tokens_map.json from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/tokenizer.json from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/tokenizer_config.json from bucket guanaco-reward-models
Deleting key jellywibble-qlora-90k-pr-7056-v2_reward/vocab.json from bucket guanaco-reward-models
Pipeline stage MKMLModelDeleter completed in 5.56s
jellywibble-qlora-90k-pr_7056_v2 status is now torndown due to DeploymentManager action