[2025-11-28 18:02:44,648] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:3633661] baseline 0.000GB ()
[2025-11-28 18:02:44,648] [INFO] [axolotl.cli.config.load_cfg:248] [PID:3633661] config:
{
"activation_offloading": false,
"adapter": "lora",
"axolotl_config_path": "/config.yaml",
"base_model": "mistralai/Mistral-Nemo-Instruct-2407",
"base_model_config": "mistralai/Mistral-Nemo-Instruct-2407",
"batch_size": 9,
"bf16": true,
"bnb_config_kwargs": {
"bnb_4bit_compute_dtype": "bfloat16",
"bnb_4bit_quant_type": "nf4",
"bnb_4bit_use_double_quant": true
},
"capabilities": {
"bf16": true,
"compute_capability": "sm_90",
"fp8": false,
"n_gpu": 1,
"n_node": 1
},
"chat_template": "tokenizer_default",
"context_parallel_size": 1,
"dataloader_num_workers": 1,
"dataloader_pin_memory": true,
"dataloader_prefetch_factor": 256,
"dataset_num_proc": 96,
"datasets": [
{
"chat_template": "tokenizer_default",
"conversation": "messages",
"message_property_mappings": {
"content": "content",
"role": "role"
},
"path": "TeamPV/distractors-onr-v2",
"split": "train",
"trust_remote_code": false,
"type": "chat_template"
}
],
"ddp": false,
"device": "cuda:0",
"dion_rank_fraction": 1.0,
"dion_rank_multiple_of": 1,
"env_capabilities": {
"torch_version": "2.8.0"
},
"eval_batch_size": 14,
"eval_causal_lm_metrics": [
"sacrebleu",
"comet",
"ter",
"chrf"
],
"eval_max_new_tokens": 128,
"eval_sample_packing": false,
"eval_steps": 0.05,
"eval_table_size": 0,
"evals_per_epoch": 5,
"experimental_skip_move_to_device": true,
"flash_attention": true,
"fp16": false,
"gradient_accumulation_steps": 1,
"gradient_checkpointing": true,
"gradient_checkpointing_kwargs": {
"use_reentrant": false
},
"hf_use_auth_token": true,
"hub_model_id": "TeamPV/mistral-nemo-onr-sft",
"hub_strategy": "every_save",
"include_tkps": true,
"is_mistral_derived_model": true,
"learning_rate": 5e-05,
"lisa_layers_attribute": "model.layers",
"load_best_model_at_end": true,
"load_in_4bit": false,
"load_in_8bit": false,
"local_rank": 0,
"logging_steps": 50,
"lora_alpha": 64,
"lora_dropout": 0.05,
"lora_r": 32,
"lora_target_modules": [
"q_proj",
"k_proj",
"v_proj",
"o_proj",
"up_proj",
"down_proj",
"gate_proj"
],
"loraplus_lr_embedding": 1e-06,
"lr_scheduler": "cosine",
"mean_resizing_embeddings": false,
"micro_batch_size": 9,
"model_config_type": "mistral",
"num_epochs": 4.0,
"optimizer": "adamw_bnb_8bit",
"otel_metrics_host": "localhost",
"otel_metrics_port": 8000,
"output_dir": "/model_out/mistral-nemo-12b_sft",
"pad_to_sequence_len": true,
"peft_use_dora": false,
"pretrain_multipack_attn": true,
"profiler_steps_start": 0,
"qlora_sharded_model_loading": false,
"ray_num_workers": 1,
"resources_per_worker": {
"GPU": 1
},
"sample_packing": false,
"sample_packing_bin_size": 200,
"sample_packing_group_size": 100000,
"save_first_step": true,
"save_only_model": false,
"save_safetensors": true,
"save_steps": 0.125,
"save_total_limit": 10,
"saves_per_epoch": 2,
"seed": 42,
"sequence_len": 3000,
"shuffle_before_merging_datasets": false,
"shuffle_merged_datasets": true,
"skip_prepare_dataset": false,
"special_tokens": {
"pad_token": ""
},
"streaming_multipack_buffer_size": 10000,
"strict": false,
"tensor_parallel_size": 1,
"tf32": true,
"tiled_mlp_use_original_mlp": true,
"tokenizer_config": "mistralai/Mistral-Nemo-Instruct-2407",
"tokenizer_save_jinja_files": true,
"torch_dtype": "torch.bfloat16",
"train_on_inputs": false,
"trl": {
"log_completions": false,
"mask_truncated_completions": false,
"ref_model_mixup_alpha": 0.9,
"ref_model_sync_steps": 64,
"scale_rewards": true,
"sync_ref_model": false,
"use_vllm": false,
"vllm_server_host": "0.0.0.0",
"vllm_server_port": 8000
},
"use_otel_metrics": false,
"use_ray": false,
"use_tensorboard": true,
"val_set_size": 0.01,
"vllm": {
"device": "auto",
"dtype": "auto",
"gpu_memory_utilization": 0.9,
"host": "0.0.0.0",
"port": 8000
},
"warmup_ratio": 0.1,
"weight_decay": 0.0,
"world_size": 1
}
[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:3633661] EOS: 2 /
[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:3633661] BOS: 1 /
[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:3633661] PAD: 2 /
[2025-11-28 18:02:45,557] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:3633661] UNK: 0 /
[2025-11-28 18:02:45,565] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:481] [PID:3633661] Unable to find prepared dataset in last_run_prepared/0ddf8b5d6caf21513bfb49ac1f784d2b
[2025-11-28 18:02:45,565] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:3633661] Loading raw datasets...
[2025-11-28 18:02:45,565] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:3633661] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`.
[2025-11-28 18:02:46,489] [WARNING] [huggingface_hub.repocard.content:108] [PID:3633661] Repo card metadata block was not found. Setting CardData to empty.
[2025-11-28 18:02:48,633] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:3633661] Loading dataset: TeamPV/distractors-onr-v2 with base_type: chat_template and prompt_style: None
[2025-11-28 18:02:48,642] [INFO] [axolotl.prompt_strategies.chat_template.__call__:969] [PID:3633661] Using chat template:
---
{%- if messages[0]["role"] == "system" %}
{%- set system_message = messages[0]["content"] %}
{%- set loop_messages = messages[1:] %}
{%- else %}
{%- set loop_messages = messages %}
{%- endif %}
{%- if not tools is defined %}
{%- set tools = none %}
{%- endif %}
{%- set user_messages = loop_messages | selectattr("role", "equalto", "user") | list %}
{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}
{%- set ns = namespace() %}
{%- set ns.index = 0 %}
{%- for message in loop_messages %}
{%- if not (message.role == "tool" or message.role == "tool_results" or (message.tool_calls is defined and message.tool_calls is not none)) %}
{%- if (message["role"] == "user") != (ns.index % 2 == 0) %}
{{- raise_exception("After the optional system message, conversation roles must alternate user/assistant/user/assistant/...") }}
{%- endif %}
{%- set ns.index = ns.index + 1 %}
{%- endif %}
{%- endfor %}
{{- bos_token }}
{%- for message in loop_messages %}
{%- if message["role"] == "user" %}
{%- if tools is not none and (message == user_messages[-1]) %}
{{- "[AVAILABLE_TOOLS][" }}
{%- for tool in tools %}
{%- set tool = tool.function %}
{{- '{"type": "function", "function": {' }}
{%- for key, val in tool.items() if key != "return" %}
{%- if val is string %}
{{- '"' + key + '": "' + val + '"' }}
{%- else %}
{{- '"' + key + '": ' + val|tojson }}
{%- endif %}
{%- if not loop.last %}
{{- ", " }}
{%- endif %}
{%- endfor %}
{{- "}}" }}
{%- if not loop.last %}
{{- ", " }}
{%- else %}
{{- "]" }}
{%- endif %}
{%- endfor %}
{{- "[/AVAILABLE_TOOLS]" }}
{%- endif %}
{%- if loop.last and system_message is defined %}
{{- "[INST]" + system_message + "\n\n" + message["content"] + "[/INST]" }}
{%- else %}
{{- "[INST]" + message["content"] + "[/INST]" }}
{%- endif %}
{%- elif (message.tool_calls is defined and message.tool_calls is not none) %}
{{- "[TOOL_CALLS][" }}
{%- for tool_call in message.tool_calls %}
{%- set out = tool_call.function|tojson %}
{{- out[:-1] }}
{%- if not tool_call.id is defined or tool_call.id|length != 9 %}
{{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
{%- endif %}
{{- ', "id": "' + tool_call.id + '"}' }}
{%- if not loop.last %}
{{- ", " }}
{%- else %}
{{- "]" + eos_token }}
{%- endif %}
{%- endfor %}
{%- elif message["role"] == "assistant" %}
{{- message["content"] + eos_token}}
{%- elif message["role"] == "tool_results" or message["role"] == "tool" %}
{%- if message.content is defined and message.content.content is defined %}
{%- set content = message.content.content %}
{%- else %}
{%- set content = message.content %}
{%- endif %}
{{- '[TOOL_RESULTS]{"content": ' + content|string + ", " }}
{%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}
{{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
{%- endif %}
{{- '"call_id": "' + message.tool_call_id + '"}[/TOOL_RESULTS]' }}
{%- else %}
{{- raise_exception("Only user and assistant roles are supported, with the exception of an initial optional system message!") }}
{%- endif %}
{%- endfor %}
---
[2025-11-28 18:03:01,375] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:218] [PID:3633661] min_input_len: 168
[2025-11-28 18:03:01,375] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:220] [PID:3633661] max_input_len: 4582
[2025-11-28 18:03:03,407] [WARNING] [axolotl.utils.data.utils.handle_long_seq_in_dataset:260] [PID:3633661] Dropped 204 samples from dataset
Saving the dataset (0/96 shards): 0%| | 0/99861 [00:00, ? examples/s]
Saving the dataset (0/96 shards): 1%| | 1000/99861 [00:01<02:17, 720.11 examples/s]
Saving the dataset (0/96 shards): 48%|████▊ | 47650/99861 [00:01<00:01, 44138.71 examples/s]
Saving the dataset (1/96 shards): 100%|██████████| 99861/99861 [00:08<00:00, 44138.71 examples/s]
Saving the dataset (2/96 shards): 100%|██████████| 99861/99861 [00:11<00:00, 44138.71 examples/s]
Saving the dataset (3/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (4/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (5/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (6/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (7/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (8/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (9/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (10/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (11/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (12/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (13/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (14/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (15/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (16/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (17/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (18/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (19/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (20/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (21/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (22/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (23/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (24/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (25/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (26/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (27/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (28/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (29/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (30/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (31/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (32/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (33/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (34/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (35/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (36/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (37/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (38/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (39/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (40/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (41/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (42/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (43/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (44/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (45/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (46/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (47/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (48/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (49/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (50/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (51/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (52/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (53/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (54/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (55/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (56/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (57/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (58/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (59/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (60/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (61/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (62/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (63/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (64/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (65/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (66/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (67/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (68/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (69/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (70/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (71/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (72/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (73/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (74/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (75/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (76/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (77/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (78/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (79/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (80/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (81/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (82/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (83/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (84/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (85/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (86/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (87/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (88/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (89/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (90/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (91/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (92/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (93/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (94/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (95/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (96/96 shards): 100%|██████████| 99861/99861 [00:12<00:00, 44138.71 examples/s]
Saving the dataset (96/96 shards): 100%|██████████| 99861/99861 [00:13<00:00, 7648.39 examples/s]
[2025-11-28 18:03:17,759] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:406] [PID:3633661] total_num_tokens: 101_581_260
[2025-11-28 18:03:18,785] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:424] [PID:3633661] `total_supervised_tokens: 31_246_079`
[2025-11-28 18:03:18,785] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:522] [PID:3633661] total_num_steps: 43939
[2025-11-28 18:03:18,785] [INFO] [axolotl.utils.data.sft._prepare_standard_dataset:121] [PID:3633661] Maximum number of steps set at 43939
[2025-11-28 18:03:18,807] [DEBUG] [axolotl.train.setup_model_and_tokenizer:65] [PID:3633661] Loading tokenizer... mistralai/Mistral-Nemo-Instruct-2407
[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:3633661] EOS: 2 /
[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:3633661] BOS: 1 /
[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:3633661] PAD: 2 /
[2025-11-28 18:03:19,445] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:3633661] UNK: 0 /
[2025-11-28 18:03:19,445] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:3633661] Loading model
[2025-11-28 18:03:19,578] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:3633661] Patched Trainer.evaluation_loop with nanmean loss calculation
[2025-11-28 18:03:19,579] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:3633661] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation
Loading checkpoint shards: 0%| | 0/5 [00:00, ?it/s]
Loading checkpoint shards: 20%|██ | 1/5 [00:00<00:01, 2.49it/s]
Loading checkpoint shards: 40%|████ | 2/5 [00:01<00:01, 1.75it/s]
Loading checkpoint shards: 60%|██████ | 3/5 [00:01<00:01, 1.70it/s]
Loading checkpoint shards: 80%|████████ | 4/5 [00:02<00:00, 1.53it/s]
Loading checkpoint shards: 100%|██████████| 5/5 [00:02<00:00, 1.73it/s]
Loading checkpoint shards: 100%|██████████| 5/5 [00:02<00:00, 1.73it/s]
[2025-11-28 18:03:27,541] [INFO] [axolotl.loaders.model._configure_embedding_dtypes:345] [PID:3633661] Converting modules to torch.bfloat16
[2025-11-28 18:03:28,468] [DEBUG] [axolotl.loaders.model.log_gpu_memory_usage:127] [PID:3633661] Memory usage after model load 0.000GB ()
trainable params: 114,032,640 || all params: 12,361,815,040 || trainable%: 0.9225
[2025-11-28 18:03:29,437] [DEBUG] [axolotl.loaders.model.log_gpu_memory_usage:127] [PID:3633661] after adapters 0.000GB ()
[2025-11-28 18:04:12,754] [INFO] [axolotl.train.save_initial_configs:398] [PID:3633661] Pre-saving adapter config to /model_out/mistral-nemo-12b_sft...
[2025-11-28 18:04:12,757] [INFO] [axolotl.train.save_initial_configs:402] [PID:3633661] Pre-saving tokenizer to /model_out/mistral-nemo-12b_sft...
[2025-11-28 18:04:12,905] [INFO] [axolotl.train.save_initial_configs:407] [PID:3633661] Pre-saving model config to /model_out/mistral-nemo-12b_sft...
[2025-11-28 18:04:12,943] [INFO] [axolotl.train.execute_training:196] [PID:3633661] Starting trainer...
0%| | 0/43939 [00:00, ?it/s][2025-11-28 18:04:13,319] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:34, 1.35s/it][A
4%|▍ | 3/72 [00:05<02:11, 1.91s/it][A
6%|▌ | 4/72 [00:08<02:30, 2.21s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.39s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.68s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.70s/it][A
19%|█▉ | 14/72 [00:35<02:36, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.71s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.71s/it][A
26%|██▋ | 19/72 [00:48<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.72s/it][A
29%|██▉ | 21/72 [00:54<02:18, 2.72s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:07, 2.72s/it][A
36%|███▌ | 26/72 [01:07<02:05, 2.72s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.72s/it][A
40%|████ | 29/72 [01:16<01:57, 2.72s/it][A
42%|████▏ | 30/72 [01:18<01:54, 2.72s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:48, 2.72s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.72s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.72s/it][A
51%|█████▏ | 37/72 [01:37<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:29, 2.72s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.72s/it][A
57%|█████▋ | 41/72 [01:48<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.72s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.72s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.72s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.72s/it][A
67%|██████▋ | 48/72 [02:07<01:05, 2.72s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.72s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.72s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:18<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.72s/it][A
75%|███████▌ | 54/72 [02:24<00:48, 2.72s/it][A
76%|███████▋ | 55/72 [02:26<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.72s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:37<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:48<00:24, 2.72s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:56<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:07<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.26s/it][A
[A{'eval_loss': 1.9759739637374878, 'eval_runtime': 197.0423, 'eval_samples_per_second': 5.07, 'eval_steps_per_second': 0.365, 'memory/max_active (GiB)': 76.86, 'memory/max_allocated (GiB)': 76.86, 'memory/device_reserved (GiB)': 77.68, 'epoch': 0}
0%| | 0/43939 [03:17, ?it/s]
100%|██████████| 72/72 [03:11<00:00, 2.26s/it][A
[A
0%| | 1/43939 [03:24<2491:56:29, 204.17s/it][2025-11-28 18:07:37,494] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-1
0%| | 2/43939 [03:31<1081:48:08, 88.64s/it]
0%| | 3/43939 [03:37<619:26:04, 50.75s/it]
0%| | 4/43939 [03:43<402:21:55, 32.97s/it]
0%| | 5/43939 [03:48<282:07:00, 23.12s/it]
0%| | 6/43939 [03:54<209:51:40, 17.20s/it]
0%| | 7/43939 [04:00<163:45:06, 13.42s/it]
0%| | 8/43939 [04:06<133:47:57, 10.96s/it]
0%| | 9/43939 [04:11<113:27:21, 9.30s/it]
0%| | 10/43939 [04:17<99:49:09, 8.18s/it]
0%| | 11/43939 [04:23<90:28:57, 7.42s/it]
0%| | 12/43939 [04:28<84:04:11, 6.89s/it]
0%| | 13/43939 [04:34<79:45:17, 6.54s/it]
0%| | 14/43939 [04:40<76:49:38, 6.30s/it]
0%| | 15/43939 [04:45<74:25:24, 6.10s/it]
0%| | 16/43939 [04:51<72:53:46, 5.97s/it]
0%| | 17/43939 [04:57<71:49:12, 5.89s/it]
0%| | 18/43939 [05:02<71:02:13, 5.82s/it]
0%| | 19/43939 [05:08<71:50:46, 5.89s/it]
0%| | 20/43939 [05:14<71:08:36, 5.83s/it]
0%| | 21/43939 [05:20<70:32:03, 5.78s/it]
0%| | 22/43939 [05:25<70:04:53, 5.74s/it]
0%| | 23/43939 [05:31<69:56:39, 5.73s/it]
0%| | 24/43939 [05:37<69:41:07, 5.71s/it]
0%| | 25/43939 [05:42<69:37:29, 5.71s/it]
0%| | 26/43939 [05:48<69:33:02, 5.70s/it]
0%| | 27/43939 [05:54<69:39:56, 5.71s/it]
0%| | 28/43939 [06:00<69:36:23, 5.71s/it]
0%| | 29/43939 [06:05<69:33:42, 5.70s/it]
0%| | 30/43939 [06:11<69:30:40, 5.70s/it]
0%| | 31/43939 [06:17<69:28:27, 5.70s/it]
0%| | 32/43939 [06:22<69:30:41, 5.70s/it]
0%| | 33/43939 [06:28<69:26:50, 5.69s/it]
0%| | 34/43939 [06:34<69:26:05, 5.69s/it]
0%| | 35/43939 [06:39<69:19:20, 5.68s/it]
0%| | 36/43939 [06:45<69:17:31, 5.68s/it]
0%| | 37/43939 [06:51<69:13:11, 5.68s/it]
0%| | 38/43939 [06:56<69:10:30, 5.67s/it]
0%| | 39/43939 [07:02<69:11:34, 5.67s/it]
0%| | 40/43939 [07:08<70:16:43, 5.76s/it]
0%| | 41/43939 [07:14<70:01:53, 5.74s/it]
0%| | 42/43939 [07:19<69:51:40, 5.73s/it]
0%| | 43/43939 [07:25<69:46:11, 5.72s/it]
0%| | 44/43939 [07:31<69:49:46, 5.73s/it]
0%| | 45/43939 [07:37<69:42:57, 5.72s/it]
0%| | 46/43939 [07:42<69:37:31, 5.71s/it]
0%| | 47/43939 [07:48<69:33:34, 5.71s/it]
0%| | 48/43939 [07:54<69:22:07, 5.69s/it]
0%| | 49/43939 [07:59<69:14:42, 5.68s/it]
0%| | 50/43939 [08:05<69:18:22, 5.68s/it]
{'loss': 1.9687, 'grad_norm': 1.4988383054733276, 'learning_rate': 5.577054404734805e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 1534.89, 'epoch': 0.0}
0%| | 50/43939 [08:05<69:18:22, 5.68s/it]
0%| | 51/43939 [08:11<69:20:04, 5.69s/it]
0%| | 52/43939 [08:16<69:25:37, 5.70s/it]
0%| | 53/43939 [08:22<69:25:59, 5.70s/it]
0%| | 54/43939 [08:28<69:29:46, 5.70s/it]
0%| | 55/43939 [08:34<69:40:27, 5.72s/it]
0%| | 56/43939 [08:39<69:44:31, 5.72s/it]
0%| | 57/43939 [08:45<69:33:58, 5.71s/it]
0%| | 58/43939 [08:51<69:27:31, 5.70s/it]
0%| | 59/43939 [08:56<69:17:51, 5.69s/it]
0%| | 60/43939 [09:02<70:16:22, 5.77s/it]
0%| | 61/43939 [09:08<69:55:13, 5.74s/it]
0%| | 62/43939 [09:14<69:44:37, 5.72s/it]
0%| | 63/43939 [09:19<69:25:24, 5.70s/it]
0%| | 64/43939 [09:25<69:26:15, 5.70s/it]
0%| | 65/43939 [09:31<69:23:45, 5.69s/it]
0%| | 66/43939 [09:36<69:26:50, 5.70s/it]
0%| | 67/43939 [09:42<69:20:23, 5.69s/it]
0%| | 68/43939 [09:48<69:16:30, 5.68s/it]
0%| | 69/43939 [09:53<69:12:04, 5.68s/it]
0%| | 70/43939 [09:59<69:12:46, 5.68s/it]
0%| | 71/43939 [10:05<69:13:09, 5.68s/it]
0%| | 72/43939 [10:10<69:18:48, 5.69s/it]
0%| | 73/43939 [10:16<69:16:52, 5.69s/it]
0%| | 74/43939 [10:22<69:16:12, 5.68s/it]
0%| | 75/43939 [10:27<69:18:50, 5.69s/it]
0%| | 76/43939 [10:33<69:28:54, 5.70s/it]
0%| | 77/43939 [10:39<69:24:26, 5.70s/it]
0%| | 78/43939 [10:45<69:20:44, 5.69s/it]
0%| | 79/43939 [10:50<69:38:48, 5.72s/it]
0%| | 80/43939 [10:56<69:41:16, 5.72s/it]
0%| | 81/43939 [11:02<69:34:46, 5.71s/it]
0%| | 82/43939 [11:08<70:37:34, 5.80s/it]
0%| | 83/43939 [11:13<70:11:20, 5.76s/it]
0%| | 84/43939 [11:19<69:55:38, 5.74s/it]
0%| | 85/43939 [11:25<69:42:01, 5.72s/it]
0%| | 86/43939 [11:30<69:24:45, 5.70s/it]
0%| | 87/43939 [11:36<69:23:41, 5.70s/it]
0%| | 88/43939 [11:42<69:19:38, 5.69s/it]
0%| | 89/43939 [11:48<69:23:33, 5.70s/it]
0%| | 90/43939 [11:53<69:24:19, 5.70s/it]
0%| | 91/43939 [11:59<69:31:34, 5.71s/it]
0%| | 92/43939 [12:05<69:26:51, 5.70s/it]
0%| | 93/43939 [12:10<69:25:28, 5.70s/it]
0%| | 94/43939 [12:16<69:26:47, 5.70s/it]
0%| | 95/43939 [12:22<69:24:44, 5.70s/it]
0%| | 96/43939 [12:27<69:29:14, 5.71s/it]
0%| | 97/43939 [12:33<69:21:44, 5.70s/it]
0%| | 98/43939 [12:39<69:22:31, 5.70s/it]
0%| | 99/43939 [12:45<69:19:49, 5.69s/it]
0%| | 100/43939 [12:50<69:24:46, 5.70s/it]
{'loss': 1.9557, 'grad_norm': 1.205540657043457, 'learning_rate': 1.1267926246300935e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 512.71, 'epoch': 0.01}
0%| | 100/43939 [12:50<69:24:46, 5.70s/it]
0%| | 101/43939 [12:56<69:25:01, 5.70s/it]
0%| | 102/43939 [13:02<70:09:46, 5.76s/it]
0%| | 103/43939 [13:08<69:53:14, 5.74s/it]
0%| | 104/43939 [13:13<69:46:42, 5.73s/it]
0%| | 105/43939 [13:19<69:29:58, 5.71s/it]
0%| | 106/43939 [13:25<69:27:12, 5.70s/it]
0%| | 107/43939 [13:30<69:37:17, 5.72s/it]
0%| | 108/43939 [13:36<69:32:05, 5.71s/it]
0%| | 109/43939 [13:42<69:23:34, 5.70s/it]
0%| | 110/43939 [13:47<69:15:53, 5.69s/it]
0%| | 111/43939 [13:53<69:21:26, 5.70s/it]
0%| | 112/43939 [13:59<69:33:24, 5.71s/it]
0%| | 113/43939 [14:05<69:25:24, 5.70s/it]
0%| | 114/43939 [14:10<69:14:56, 5.69s/it]
0%| | 115/43939 [14:16<69:12:46, 5.69s/it]
0%| | 116/43939 [14:22<69:19:49, 5.70s/it]
0%| | 117/43939 [14:27<69:25:27, 5.70s/it]
0%| | 118/43939 [14:33<69:24:02, 5.70s/it]
0%| | 119/43939 [14:39<69:18:11, 5.69s/it]
0%| | 120/43939 [14:44<69:13:14, 5.69s/it]
0%| | 121/43939 [14:50<69:13:06, 5.69s/it]
0%| | 122/43939 [14:56<69:27:11, 5.71s/it]
0%| | 123/43939 [15:02<70:14:00, 5.77s/it]
0%| | 124/43939 [15:07<69:53:06, 5.74s/it]
0%| | 125/43939 [15:13<69:46:27, 5.73s/it]
0%| | 126/43939 [15:19<69:28:43, 5.71s/it]
0%| | 127/43939 [15:24<69:26:19, 5.71s/it]
0%| | 128/43939 [15:30<69:22:53, 5.70s/it]
0%| | 129/43939 [15:36<69:21:52, 5.70s/it]
0%| | 130/43939 [15:42<69:18:01, 5.69s/it]
0%| | 131/43939 [15:47<69:18:39, 5.70s/it]
0%| | 132/43939 [15:53<69:24:52, 5.70s/it]
0%| | 133/43939 [15:59<69:26:32, 5.71s/it]
0%| | 134/43939 [16:04<69:26:01, 5.71s/it]
0%| | 135/43939 [16:10<69:27:15, 5.71s/it]
0%| | 136/43939 [16:16<69:24:45, 5.70s/it]
0%| | 137/43939 [16:21<69:24:40, 5.70s/it]
0%| | 138/43939 [16:27<69:17:10, 5.69s/it]
0%| | 139/43939 [16:33<69:17:29, 5.70s/it]
0%| | 140/43939 [16:39<69:19:58, 5.70s/it]
0%| | 141/43939 [16:44<69:30:56, 5.71s/it]
0%| | 142/43939 [16:50<69:30:15, 5.71s/it]
0%| | 143/43939 [16:56<69:33:00, 5.72s/it]
0%| | 144/43939 [17:01<69:25:03, 5.71s/it]
0%| | 145/43939 [17:07<70:18:11, 5.78s/it]
0%| | 146/43939 [17:13<69:59:34, 5.75s/it]
0%| | 147/43939 [17:19<69:46:18, 5.74s/it]
0%| | 148/43939 [17:24<69:32:19, 5.72s/it]
0%| | 149/43939 [17:30<69:29:14, 5.71s/it]
0%| | 150/43939 [17:36<69:25:04, 5.71s/it]
{'loss': 1.8251, 'grad_norm': 1.4495829343795776, 'learning_rate': 1.6958798087867062e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 492.64, 'epoch': 0.01}
0%| | 150/43939 [17:36<69:25:04, 5.71s/it]
0%| | 151/43939 [17:42<69:26:12, 5.71s/it]
0%| | 152/43939 [17:47<69:21:52, 5.70s/it]
0%| | 153/43939 [17:53<69:22:41, 5.70s/it]
0%| | 154/43939 [17:59<69:27:56, 5.71s/it]
0%| | 155/43939 [18:04<69:27:46, 5.71s/it]
0%| | 156/43939 [18:10<69:29:12, 5.71s/it]
0%| | 157/43939 [18:16<69:25:08, 5.71s/it]
0%| | 158/43939 [18:22<69:31:10, 5.72s/it]
0%| | 159/43939 [18:27<69:21:25, 5.70s/it]
0%| | 160/43939 [18:33<69:13:44, 5.69s/it]
0%| | 161/43939 [18:39<69:20:39, 5.70s/it]
0%| | 162/43939 [18:44<69:21:59, 5.70s/it]
0%| | 163/43939 [18:50<69:22:12, 5.70s/it]
0%| | 164/43939 [18:56<69:28:07, 5.71s/it]
0%| | 165/43939 [19:02<70:21:40, 5.79s/it]
0%| | 166/43939 [19:07<69:59:51, 5.76s/it]
0%| | 167/43939 [19:13<69:45:17, 5.74s/it]
0%| | 168/43939 [19:19<69:31:27, 5.72s/it]
0%| | 169/43939 [19:24<69:31:14, 5.72s/it]
0%| | 170/43939 [19:30<69:27:49, 5.71s/it]
0%| | 171/43939 [19:36<69:25:59, 5.71s/it]
0%| | 172/43939 [19:42<69:16:49, 5.70s/it]
0%| | 173/43939 [19:47<69:17:58, 5.70s/it]
0%| | 174/43939 [19:53<69:15:31, 5.70s/it]
0%| | 175/43939 [19:59<69:15:10, 5.70s/it]
0%| | 176/43939 [20:04<69:16:04, 5.70s/it]
0%| | 177/43939 [20:10<69:16:14, 5.70s/it]
0%| | 178/43939 [20:16<69:14:58, 5.70s/it]
0%| | 179/43939 [20:21<69:17:21, 5.70s/it]
0%| | 180/43939 [20:27<69:18:15, 5.70s/it]
0%| | 181/43939 [20:33<69:19:21, 5.70s/it]
0%| | 182/43939 [20:39<69:12:36, 5.69s/it]
0%| | 183/43939 [20:44<69:15:49, 5.70s/it]
0%| | 184/43939 [20:50<69:22:32, 5.71s/it]
0%| | 185/43939 [20:56<69:34:17, 5.72s/it]
0%| | 186/43939 [21:02<70:19:22, 5.79s/it]
0%| | 187/43939 [21:07<70:01:03, 5.76s/it]
0%| | 188/43939 [21:13<70:07:10, 5.77s/it]
0%| | 189/43939 [21:19<69:54:15, 5.75s/it]
0%| | 190/43939 [21:25<69:41:55, 5.74s/it]
0%| | 191/43939 [21:30<69:38:39, 5.73s/it]
0%| | 192/43939 [21:36<69:44:10, 5.74s/it]
0%| | 193/43939 [21:42<69:33:00, 5.72s/it]
0%| | 194/43939 [21:47<69:27:09, 5.72s/it]
0%| | 195/43939 [21:53<69:20:59, 5.71s/it]
0%| | 196/43939 [21:59<69:28:36, 5.72s/it]
0%| | 197/43939 [22:05<69:33:53, 5.73s/it]
0%| | 198/43939 [22:10<69:25:15, 5.71s/it]
0%| | 199/43939 [22:16<69:34:07, 5.73s/it]
0%| | 200/43939 [22:22<69:24:56, 5.71s/it]
{'loss': 1.6389, 'grad_norm': 0.9070694446563721, 'learning_rate': 2.264966992943319e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 513.79, 'epoch': 0.02}
0%| | 200/43939 [22:22<69:24:56, 5.71s/it]
0%| | 201/43939 [22:27<69:34:14, 5.73s/it]
0%| | 202/43939 [22:33<69:27:56, 5.72s/it]
0%| | 203/43939 [22:39<69:41:03, 5.74s/it]
0%| | 204/43939 [22:45<69:28:34, 5.72s/it]
0%| | 205/43939 [22:50<69:42:11, 5.74s/it]
0%| | 206/43939 [22:56<69:34:26, 5.73s/it]
0%| | 207/43939 [23:02<70:36:03, 5.81s/it]
0%| | 208/43939 [23:08<70:09:22, 5.78s/it]
0%| | 209/43939 [23:14<69:53:01, 5.75s/it]
0%| | 210/43939 [23:19<69:31:52, 5.72s/it]
0%| | 211/43939 [23:25<69:27:15, 5.72s/it]
0%| | 212/43939 [23:31<69:29:56, 5.72s/it]
0%| | 213/43939 [23:36<69:30:07, 5.72s/it]
0%| | 214/43939 [23:42<69:34:46, 5.73s/it]
0%| | 215/43939 [23:48<69:33:15, 5.73s/it]
0%| | 216/43939 [23:53<69:19:30, 5.71s/it]
0%| | 217/43939 [23:59<69:17:27, 5.71s/it]
0%| | 218/43939 [24:05<69:22:03, 5.71s/it]
0%| | 219/43939 [24:11<69:13:56, 5.70s/it]
1%| | 220/43939 [24:16<69:20:27, 5.71s/it]
1%| | 221/43939 [24:22<69:20:09, 5.71s/it]
1%| | 222/43939 [24:28<69:20:39, 5.71s/it]
1%| | 223/43939 [24:33<69:12:02, 5.70s/it]
1%| | 224/43939 [24:39<69:14:19, 5.70s/it]
1%| | 225/43939 [24:45<69:34:04, 5.73s/it]
1%| | 226/43939 [24:51<69:17:33, 5.71s/it]
1%| | 227/43939 [24:56<69:17:41, 5.71s/it]
1%| | 228/43939 [25:02<69:16:03, 5.70s/it]
1%| | 229/43939 [25:08<69:53:04, 5.76s/it]
1%| | 230/43939 [25:13<69:37:40, 5.73s/it]
1%| | 231/43939 [25:19<69:25:52, 5.72s/it]
1%| | 232/43939 [25:25<69:31:57, 5.73s/it]
1%| | 233/43939 [25:31<69:22:13, 5.71s/it]
1%| | 234/43939 [25:36<69:21:51, 5.71s/it]
1%| | 235/43939 [25:42<69:14:36, 5.70s/it]
1%| | 236/43939 [25:48<69:13:26, 5.70s/it]
1%| | 237/43939 [25:53<69:08:17, 5.70s/it]
1%| | 238/43939 [25:59<69:11:25, 5.70s/it]
1%| | 239/43939 [26:05<69:09:49, 5.70s/it]
1%| | 240/43939 [26:10<69:06:39, 5.69s/it]
1%| | 241/43939 [26:16<69:13:59, 5.70s/it]
1%| | 242/43939 [26:22<69:12:08, 5.70s/it]
1%| | 243/43939 [26:28<69:09:05, 5.70s/it]
1%| | 244/43939 [26:33<69:06:09, 5.69s/it]
1%| | 245/43939 [26:39<69:11:15, 5.70s/it]
1%| | 246/43939 [26:45<69:16:58, 5.71s/it]
1%| | 247/43939 [26:50<69:05:15, 5.69s/it]
1%| | 248/43939 [26:56<69:12:51, 5.70s/it]
1%| | 249/43939 [27:02<70:02:12, 5.77s/it]
1%| | 250/43939 [27:08<69:51:17, 5.76s/it]
{'loss': 1.4838, 'grad_norm': 1.1880543231964111, 'learning_rate': 2.834054177099932e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 489.92, 'epoch': 0.02}
1%| | 250/43939 [27:08<69:51:17, 5.76s/it]
1%| | 251/43939 [27:13<69:44:21, 5.75s/it]
1%| | 252/43939 [27:19<69:42:07, 5.74s/it]
1%| | 253/43939 [27:25<69:23:33, 5.72s/it]
1%| | 254/43939 [27:31<69:10:28, 5.70s/it]
1%| | 255/43939 [27:36<69:11:34, 5.70s/it]
1%| | 256/43939 [27:42<69:12:06, 5.70s/it]
1%| | 257/43939 [27:48<69:19:48, 5.71s/it]
1%| | 258/43939 [27:53<69:06:28, 5.70s/it]
1%| | 259/43939 [27:59<69:04:53, 5.69s/it]
1%| | 260/43939 [28:05<69:19:40, 5.71s/it]
1%| | 261/43939 [28:10<69:22:24, 5.72s/it]
1%| | 262/43939 [28:16<69:16:09, 5.71s/it]
1%| | 263/43939 [28:22<69:05:13, 5.69s/it]
1%| | 264/43939 [28:27<68:55:20, 5.68s/it]
1%| | 265/43939 [28:33<69:02:41, 5.69s/it]
1%| | 266/43939 [28:39<69:02:56, 5.69s/it]
1%| | 267/43939 [28:45<69:02:00, 5.69s/it]
1%| | 268/43939 [28:50<69:03:11, 5.69s/it]
1%| | 269/43939 [28:56<69:09:29, 5.70s/it]
1%| | 270/43939 [29:02<69:54:44, 5.76s/it]
1%| | 271/43939 [29:08<69:38:49, 5.74s/it]
1%| | 272/43939 [29:13<69:29:31, 5.73s/it]
1%| | 273/43939 [29:19<69:10:45, 5.70s/it]
1%| | 274/43939 [29:25<69:14:00, 5.71s/it]
1%| | 275/43939 [29:30<69:12:41, 5.71s/it]
1%| | 276/43939 [29:36<69:14:08, 5.71s/it]
1%| | 277/43939 [29:42<69:06:42, 5.70s/it]
1%| | 278/43939 [29:47<69:05:28, 5.70s/it]
1%| | 279/43939 [29:53<69:20:17, 5.72s/it]
1%| | 280/43939 [29:59<69:21:21, 5.72s/it]
1%| | 281/43939 [30:05<69:24:45, 5.72s/it]
1%| | 282/43939 [30:10<69:20:17, 5.72s/it]
1%| | 283/43939 [30:16<69:09:33, 5.70s/it]
1%| | 284/43939 [30:22<69:15:50, 5.71s/it]
1%| | 285/43939 [30:27<69:05:39, 5.70s/it]
1%| | 286/43939 [30:33<69:09:13, 5.70s/it]
1%| | 287/43939 [30:39<69:10:24, 5.70s/it]
1%| | 288/43939 [30:45<69:08:10, 5.70s/it]
1%| | 289/43939 [30:50<69:04:45, 5.70s/it]
1%| | 290/43939 [30:56<69:17:58, 5.72s/it]
1%| | 291/43939 [31:02<70:17:32, 5.80s/it]
1%| | 292/43939 [31:08<69:51:32, 5.76s/it]
1%| | 293/43939 [31:13<69:29:58, 5.73s/it]
1%| | 294/43939 [31:19<69:23:38, 5.72s/it]
1%| | 295/43939 [31:25<69:16:15, 5.71s/it]
1%| | 296/43939 [31:30<69:19:46, 5.72s/it]
1%| | 297/43939 [31:36<69:15:08, 5.71s/it]
1%| | 298/43939 [31:42<69:21:59, 5.72s/it]
1%| | 299/43939 [31:48<69:18:55, 5.72s/it]
1%| | 300/43939 [31:53<69:24:56, 5.73s/it]
{'loss': 1.4042, 'grad_norm': 1.474748134613037, 'learning_rate': 3.4031413612565448e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 485.95, 'epoch': 0.03}
1%| | 300/43939 [31:53<69:24:56, 5.73s/it]
1%| | 301/43939 [31:59<69:35:56, 5.74s/it]
1%| | 302/43939 [32:05<69:23:51, 5.73s/it]
1%| | 303/43939 [32:11<69:18:02, 5.72s/it]
1%| | 304/43939 [32:16<69:12:59, 5.71s/it]
1%| | 305/43939 [32:22<69:16:45, 5.72s/it]
1%| | 306/43939 [32:28<69:13:26, 5.71s/it]
1%| | 307/43939 [32:33<69:04:38, 5.70s/it]
1%| | 308/43939 [32:39<69:06:54, 5.70s/it]
1%| | 309/43939 [32:45<69:19:51, 5.72s/it]
1%| | 310/43939 [32:51<69:17:56, 5.72s/it]
1%| | 311/43939 [32:56<70:08:19, 5.79s/it]
1%| | 312/43939 [33:02<69:42:56, 5.75s/it]
1%| | 313/43939 [33:08<69:26:10, 5.73s/it]
1%| | 314/43939 [33:14<69:18:48, 5.72s/it]
1%| | 315/43939 [33:19<69:05:34, 5.70s/it]
1%| | 316/43939 [33:25<69:04:59, 5.70s/it]
1%| | 317/43939 [33:31<69:05:19, 5.70s/it]
1%| | 318/43939 [33:36<69:03:39, 5.70s/it]
1%| | 319/43939 [33:42<69:10:05, 5.71s/it]
1%| | 320/43939 [33:48<69:08:01, 5.71s/it]
1%| | 321/43939 [33:53<69:03:33, 5.70s/it]
1%| | 322/43939 [33:59<69:03:59, 5.70s/it]
1%| | 323/43939 [34:05<69:04:36, 5.70s/it]
1%| | 324/43939 [34:10<69:04:27, 5.70s/it]
1%| | 325/43939 [34:16<69:02:58, 5.70s/it]
1%| | 326/43939 [34:22<69:01:29, 5.70s/it]
1%| | 327/43939 [34:28<69:00:28, 5.70s/it]
1%| | 328/43939 [34:33<68:58:32, 5.69s/it]
1%| | 329/43939 [34:39<69:02:03, 5.70s/it]
1%| | 330/43939 [34:45<69:23:44, 5.73s/it]
1%| | 331/43939 [34:50<69:08:54, 5.71s/it]
1%| | 332/43939 [34:56<69:53:55, 5.77s/it]
1%| | 333/43939 [35:02<69:41:29, 5.75s/it]
1%| | 334/43939 [35:08<69:26:40, 5.73s/it]
1%| | 335/43939 [35:13<69:20:40, 5.73s/it]
1%| | 336/43939 [35:19<69:20:17, 5.72s/it]
1%| | 337/43939 [35:25<69:18:46, 5.72s/it]
1%| | 338/43939 [35:31<69:18:33, 5.72s/it]
1%| | 339/43939 [35:36<69:10:17, 5.71s/it]
1%| | 340/43939 [35:42<69:11:57, 5.71s/it]
1%| | 341/43939 [35:48<69:02:18, 5.70s/it]
1%| | 342/43939 [35:53<69:24:22, 5.73s/it]
1%| | 343/43939 [35:59<69:13:13, 5.72s/it]
1%| | 344/43939 [36:05<69:05:13, 5.71s/it]
1%| | 345/43939 [36:11<69:07:54, 5.71s/it]
1%| | 346/43939 [36:16<69:06:16, 5.71s/it]
1%| | 347/43939 [36:22<69:08:28, 5.71s/it]
1%| | 348/43939 [36:28<68:58:18, 5.70s/it]
1%| | 349/43939 [36:33<69:00:04, 5.70s/it]
1%| | 350/43939 [36:39<68:53:49, 5.69s/it]
{'loss': 1.3435, 'grad_norm': 1.490429401397705, 'learning_rate': 3.972228545413158e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 482.48, 'epoch': 0.03}
1%| | 350/43939 [36:39<68:53:49, 5.69s/it]
1%| | 351/43939 [36:45<69:01:19, 5.70s/it]
1%| | 352/43939 [36:50<69:01:17, 5.70s/it]
1%| | 353/43939 [36:56<69:02:12, 5.70s/it]
1%| | 354/43939 [37:02<69:58:26, 5.78s/it]
1%| | 355/43939 [37:08<69:42:19, 5.76s/it]
1%| | 356/43939 [37:14<69:27:34, 5.74s/it]
1%| | 357/43939 [37:19<69:18:21, 5.72s/it]
1%| | 358/43939 [37:25<69:18:06, 5.72s/it]
1%| | 359/43939 [37:31<69:02:38, 5.70s/it]
1%| | 360/43939 [37:36<69:13:00, 5.72s/it]
1%| | 361/43939 [37:42<69:03:49, 5.71s/it]
1%| | 362/43939 [37:48<69:01:50, 5.70s/it]
1%| | 363/43939 [37:53<69:08:20, 5.71s/it]
1%| | 364/43939 [37:59<69:08:42, 5.71s/it]
1%| | 365/43939 [38:05<69:11:40, 5.72s/it]
1%| | 366/43939 [38:11<69:08:45, 5.71s/it]
1%| | 367/43939 [38:16<69:08:10, 5.71s/it]
1%| | 368/43939 [38:22<69:03:36, 5.71s/it]
1%| | 369/43939 [38:28<69:08:26, 5.71s/it]
1%| | 370/43939 [38:33<69:02:53, 5.71s/it]
1%| | 371/43939 [38:39<69:02:33, 5.70s/it]
1%| | 372/43939 [38:45<68:55:26, 5.70s/it]
1%| | 373/43939 [38:50<68:50:43, 5.69s/it]
1%| | 374/43939 [38:56<69:45:06, 5.76s/it]
1%| | 375/43939 [39:02<69:22:46, 5.73s/it]
1%| | 376/43939 [39:08<69:27:18, 5.74s/it]
1%| | 377/43939 [39:14<69:17:42, 5.73s/it]
1%| | 378/43939 [39:19<69:13:18, 5.72s/it]
1%| | 379/43939 [39:25<69:08:55, 5.71s/it]
1%| | 380/43939 [39:31<68:53:41, 5.69s/it]
1%| | 381/43939 [39:36<69:03:48, 5.71s/it]
1%| | 382/43939 [39:42<68:55:43, 5.70s/it]
1%| | 383/43939 [39:48<68:53:09, 5.69s/it]
1%| | 384/43939 [39:53<68:51:51, 5.69s/it]
1%| | 385/43939 [39:59<68:47:24, 5.69s/it]
1%| | 386/43939 [40:05<68:57:52, 5.70s/it]
1%| | 387/43939 [40:10<68:58:35, 5.70s/it]
1%| | 388/43939 [40:16<69:04:01, 5.71s/it]
1%| | 389/43939 [40:22<68:55:33, 5.70s/it]
1%| | 390/43939 [40:28<68:57:45, 5.70s/it]
1%| | 391/43939 [40:33<68:52:21, 5.69s/it]
1%| | 392/43939 [40:39<68:45:53, 5.68s/it]
1%| | 393/43939 [40:45<68:42:50, 5.68s/it]
1%| | 394/43939 [40:50<68:52:53, 5.69s/it]
1%| | 395/43939 [40:56<69:39:48, 5.76s/it]
1%| | 396/43939 [41:02<69:20:58, 5.73s/it]
1%| | 397/43939 [41:08<69:14:48, 5.73s/it]
1%| | 398/43939 [41:13<69:12:55, 5.72s/it]
1%| | 399/43939 [41:19<69:13:09, 5.72s/it]
1%| | 400/43939 [41:25<69:14:09, 5.72s/it]
{'loss': 1.3146, 'grad_norm': 1.9503854513168335, 'learning_rate': 4.54131572956977e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 472.38, 'epoch': 0.04}
1%| | 400/43939 [41:25<69:14:09, 5.72s/it]
1%| | 401/43939 [41:30<69:14:34, 5.73s/it]
1%| | 402/43939 [41:36<69:11:26, 5.72s/it]
1%| | 403/43939 [41:42<69:05:21, 5.71s/it]
1%| | 404/43939 [41:48<69:02:40, 5.71s/it]
1%| | 405/43939 [41:53<69:03:31, 5.71s/it]
1%| | 406/43939 [41:59<69:04:41, 5.71s/it]
1%| | 407/43939 [42:05<68:53:05, 5.70s/it]
1%| | 408/43939 [42:10<68:50:32, 5.69s/it]
1%| | 409/43939 [42:16<68:46:12, 5.69s/it]
1%| | 410/43939 [42:22<68:54:37, 5.70s/it]
1%| | 411/43939 [42:27<68:47:55, 5.69s/it]
1%| | 412/43939 [42:33<68:49:15, 5.69s/it]
1%| | 413/43939 [42:39<68:43:45, 5.68s/it]
1%| | 414/43939 [42:45<68:48:40, 5.69s/it]
1%| | 415/43939 [42:50<68:42:14, 5.68s/it]
1%| | 416/43939 [42:56<69:30:37, 5.75s/it]
1%| | 417/43939 [43:02<69:17:18, 5.73s/it]
1%| | 418/43939 [43:07<69:09:40, 5.72s/it]
1%| | 419/43939 [43:13<69:12:26, 5.72s/it]
1%| | 420/43939 [43:19<68:57:52, 5.70s/it]
1%| | 421/43939 [43:25<68:50:14, 5.69s/it]
1%| | 422/43939 [43:30<68:49:56, 5.69s/it]
1%| | 423/43939 [43:36<68:44:06, 5.69s/it]
1%| | 424/43939 [43:42<68:48:16, 5.69s/it]
1%| | 425/43939 [43:47<68:55:31, 5.70s/it]
1%| | 426/43939 [43:53<69:05:45, 5.72s/it]
1%| | 427/43939 [43:59<68:53:39, 5.70s/it]
1%| | 428/43939 [44:04<68:53:11, 5.70s/it]
1%| | 429/43939 [44:10<68:54:25, 5.70s/it]
1%| | 430/43939 [44:16<68:46:52, 5.69s/it]
1%| | 431/43939 [44:21<68:41:06, 5.68s/it]
1%| | 432/43939 [44:27<68:42:09, 5.68s/it]
1%| | 433/43939 [44:33<68:40:29, 5.68s/it]
1%| | 434/43939 [44:39<68:47:58, 5.69s/it]
1%| | 435/43939 [44:44<68:46:23, 5.69s/it]
1%| | 436/43939 [44:50<68:42:39, 5.69s/it]
1%| | 437/43939 [44:56<69:37:12, 5.76s/it]
1%| | 438/43939 [45:02<69:28:02, 5.75s/it]
1%| | 439/43939 [45:07<69:22:01, 5.74s/it]
1%| | 440/43939 [45:13<69:13:38, 5.73s/it]
1%| | 441/43939 [45:19<69:04:42, 5.72s/it]
1%| | 442/43939 [45:24<69:16:45, 5.73s/it]
1%| | 443/43939 [45:30<69:15:43, 5.73s/it]
1%| | 444/43939 [45:36<69:14:44, 5.73s/it]
1%| | 445/43939 [45:42<69:04:56, 5.72s/it]
1%| | 446/43939 [45:47<69:00:14, 5.71s/it]
1%| | 447/43939 [45:53<68:55:03, 5.70s/it]
1%| | 448/43939 [45:59<68:45:42, 5.69s/it]
1%| | 449/43939 [46:04<68:51:09, 5.70s/it]
1%| | 450/43939 [46:10<68:48:37, 5.70s/it]
{'loss': 1.2806, 'grad_norm': 1.4551211595535278, 'learning_rate': 5.110402913726383e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 484.66, 'epoch': 0.04}
1%| | 450/43939 [46:10<68:48:37, 5.70s/it]
1%| | 451/43939 [46:16<68:43:31, 5.69s/it]
1%| | 452/43939 [46:21<68:38:47, 5.68s/it]
1%| | 453/43939 [46:27<68:40:47, 5.69s/it]
1%| | 454/43939 [46:33<68:44:43, 5.69s/it]
1%| | 455/43939 [46:38<68:44:13, 5.69s/it]
1%| | 456/43939 [46:44<68:42:47, 5.69s/it]
1%| | 457/43939 [46:50<69:22:45, 5.74s/it]
1%| | 458/43939 [46:56<69:23:40, 5.75s/it]
1%| | 459/43939 [47:01<69:12:54, 5.73s/it]
1%| | 460/43939 [47:07<69:06:06, 5.72s/it]
1%| | 461/43939 [47:13<69:02:38, 5.72s/it]
1%| | 462/43939 [47:19<68:55:05, 5.71s/it]
1%| | 463/43939 [47:24<68:56:29, 5.71s/it]
1%| | 464/43939 [47:30<68:51:12, 5.70s/it]
1%| | 465/43939 [47:36<68:57:09, 5.71s/it]
1%| | 466/43939 [47:41<68:51:38, 5.70s/it]
1%| | 467/43939 [47:47<68:52:55, 5.70s/it]
1%| | 468/43939 [47:53<69:04:20, 5.72s/it]
1%| | 469/43939 [47:59<68:58:29, 5.71s/it]
1%| | 470/43939 [48:04<68:51:34, 5.70s/it]
1%| | 471/43939 [48:10<69:00:23, 5.72s/it]
1%| | 472/43939 [48:16<68:53:12, 5.71s/it]
1%| | 473/43939 [48:21<68:49:18, 5.70s/it]
1%| | 474/43939 [48:27<68:47:29, 5.70s/it]
1%| | 475/43939 [48:33<68:45:35, 5.70s/it]
1%| | 476/43939 [48:38<68:44:03, 5.69s/it]
1%| | 477/43939 [48:44<68:52:20, 5.70s/it]
1%| | 478/43939 [48:50<68:55:20, 5.71s/it]
1%| | 479/43939 [48:56<69:38:19, 5.77s/it]
1%| | 480/43939 [49:01<69:23:21, 5.75s/it]
1%| | 481/43939 [49:07<69:08:20, 5.73s/it]
1%| | 482/43939 [49:13<68:56:17, 5.71s/it]
1%| | 483/43939 [49:19<68:59:16, 5.72s/it]
1%| | 484/43939 [49:24<69:04:50, 5.72s/it]
1%| | 485/43939 [49:30<69:02:42, 5.72s/it]
1%| | 486/43939 [49:36<68:56:38, 5.71s/it]
1%| | 487/43939 [49:41<68:49:37, 5.70s/it]
1%| | 488/43939 [49:47<68:57:02, 5.71s/it]
1%| | 489/43939 [49:53<69:07:25, 5.73s/it]
1%| | 490/43939 [49:59<69:16:44, 5.74s/it]
1%| | 491/43939 [50:04<69:15:26, 5.74s/it]
1%| | 492/43939 [50:10<69:10:32, 5.73s/it]
1%| | 493/43939 [50:16<69:04:11, 5.72s/it]
1%| | 494/43939 [50:22<69:14:28, 5.74s/it]
1%| | 495/43939 [50:27<69:00:14, 5.72s/it]
1%| | 496/43939 [50:33<68:59:25, 5.72s/it]
1%| | 497/43939 [50:39<68:52:52, 5.71s/it]
1%| | 498/43939 [50:44<68:50:55, 5.71s/it]
1%| | 499/43939 [50:50<68:54:53, 5.71s/it]
1%| | 500/43939 [50:56<69:50:25, 5.79s/it]
{'loss': 1.2785, 'grad_norm': 1.7765358686447144, 'learning_rate': 5.679490097882996e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 459.31, 'epoch': 0.05}
1%| | 500/43939 [50:56<69:50:25, 5.79s/it]
1%| | 501/43939 [51:02<69:32:53, 5.76s/it]
1%| | 502/43939 [51:08<69:34:25, 5.77s/it]
1%| | 503/43939 [51:13<69:19:07, 5.75s/it]
1%| | 504/43939 [51:19<69:11:32, 5.73s/it]
1%| | 505/43939 [51:25<69:04:17, 5.72s/it]
1%| | 506/43939 [51:30<69:00:06, 5.72s/it]
1%| | 507/43939 [51:36<68:54:28, 5.71s/it]
1%| | 508/43939 [51:42<68:50:41, 5.71s/it]
1%| | 509/43939 [51:47<68:55:20, 5.71s/it]
1%| | 510/43939 [51:53<68:53:24, 5.71s/it]
1%| | 511/43939 [51:59<68:48:25, 5.70s/it]
1%| | 512/43939 [52:05<68:47:36, 5.70s/it]
1%| | 513/43939 [52:10<68:44:16, 5.70s/it]
1%| | 514/43939 [52:16<68:45:32, 5.70s/it]
1%| | 515/43939 [52:22<68:42:08, 5.70s/it]
1%| | 516/43939 [52:27<68:50:52, 5.71s/it]
1%| | 517/43939 [52:33<68:46:37, 5.70s/it]
1%| | 518/43939 [52:39<68:56:33, 5.72s/it]
1%| | 519/43939 [52:44<68:53:08, 5.71s/it]
1%| | 520/43939 [52:50<69:50:41, 5.79s/it]
1%| | 521/43939 [52:56<69:28:57, 5.76s/it]
1%| | 522/43939 [53:02<69:13:22, 5.74s/it]
1%| | 523/43939 [53:08<69:07:16, 5.73s/it]
1%| | 524/43939 [53:13<69:08:20, 5.73s/it]
1%| | 525/43939 [53:19<68:57:52, 5.72s/it]
1%| | 526/43939 [53:25<68:45:31, 5.70s/it]
1%| | 527/43939 [53:30<68:40:20, 5.69s/it]
1%| | 528/43939 [53:36<68:40:04, 5.69s/it]
1%| | 529/43939 [53:42<68:37:26, 5.69s/it]
1%| | 530/43939 [53:47<68:32:51, 5.68s/it]
1%| | 531/43939 [53:53<68:26:40, 5.68s/it]
1%| | 532/43939 [53:59<68:26:17, 5.68s/it]
1%| | 533/43939 [54:04<68:24:14, 5.67s/it]
1%| | 534/43939 [54:10<68:29:04, 5.68s/it]
1%| | 535/43939 [54:16<68:27:48, 5.68s/it]
1%| | 536/43939 [54:21<68:35:13, 5.69s/it]
1%| | 537/43939 [54:27<68:34:20, 5.69s/it]
1%| | 538/43939 [54:33<68:35:43, 5.69s/it]
1%| | 539/43939 [54:39<68:42:16, 5.70s/it]
1%| | 540/43939 [54:44<68:41:56, 5.70s/it]
1%| | 541/43939 [54:50<69:31:50, 5.77s/it]
1%| | 542/43939 [54:56<69:14:35, 5.74s/it]
1%| | 543/43939 [55:02<69:30:10, 5.77s/it]
1%| | 544/43939 [55:07<69:20:43, 5.75s/it]
1%| | 545/43939 [55:13<69:06:39, 5.73s/it]
1%| | 546/43939 [55:19<68:58:06, 5.72s/it]
1%| | 547/43939 [55:25<69:05:22, 5.73s/it]
1%| | 548/43939 [55:30<69:02:16, 5.73s/it]
1%| | 549/43939 [55:36<68:50:28, 5.71s/it]
1%|▏ | 550/43939 [55:42<68:47:20, 5.71s/it]
{'loss': 1.2962, 'grad_norm': 1.6205016374588013, 'learning_rate': 6.248577282039609e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 496.89, 'epoch': 0.05}
1%|▏ | 550/43939 [55:42<68:47:20, 5.71s/it]
1%|▏ | 551/43939 [55:47<68:36:02, 5.69s/it]
1%|▏ | 552/43939 [55:53<68:35:44, 5.69s/it]
1%|▏ | 553/43939 [55:59<68:32:21, 5.69s/it]
1%|▏ | 554/43939 [56:04<68:46:41, 5.71s/it]
1%|▏ | 555/43939 [56:10<68:49:16, 5.71s/it]
1%|▏ | 556/43939 [56:16<69:00:53, 5.73s/it]
1%|▏ | 557/43939 [56:22<68:50:26, 5.71s/it]
1%|▏ | 558/43939 [56:27<68:43:20, 5.70s/it]
1%|▏ | 559/43939 [56:33<68:44:16, 5.70s/it]
1%|▏ | 560/43939 [56:39<68:59:10, 5.73s/it]
1%|▏ | 561/43939 [56:45<69:55:51, 5.80s/it]
1%|▏ | 562/43939 [56:50<69:27:55, 5.77s/it]
1%|▏ | 563/43939 [56:56<69:20:26, 5.75s/it]
1%|▏ | 564/43939 [57:02<69:14:18, 5.75s/it]
1%|▏ | 565/43939 [57:08<68:59:36, 5.73s/it]
1%|▏ | 566/43939 [57:13<68:48:21, 5.71s/it]
1%|▏ | 567/43939 [57:19<68:48:52, 5.71s/it]
1%|▏ | 568/43939 [57:25<68:35:35, 5.69s/it]
1%|▏ | 569/43939 [57:30<68:41:38, 5.70s/it]
1%|▏ | 570/43939 [57:36<68:41:55, 5.70s/it]
1%|▏ | 571/43939 [57:42<68:44:45, 5.71s/it]
1%|▏ | 572/43939 [57:47<68:43:00, 5.70s/it]
1%|▏ | 573/43939 [57:53<68:39:00, 5.70s/it]
1%|▏ | 574/43939 [57:59<68:46:52, 5.71s/it]
1%|▏ | 575/43939 [58:05<68:55:18, 5.72s/it]
1%|▏ | 576/43939 [58:10<68:50:48, 5.72s/it]
1%|▏ | 577/43939 [58:16<68:52:52, 5.72s/it]
1%|▏ | 578/43939 [58:22<68:54:58, 5.72s/it]
1%|▏ | 579/43939 [58:27<68:56:34, 5.72s/it]
1%|▏ | 580/43939 [58:33<68:45:42, 5.71s/it]
1%|▏ | 581/43939 [58:39<68:57:11, 5.73s/it]
1%|▏ | 582/43939 [58:45<69:38:23, 5.78s/it]
1%|▏ | 583/43939 [58:51<69:17:40, 5.75s/it]
1%|▏ | 584/43939 [58:56<68:54:22, 5.72s/it]
1%|▏ | 585/43939 [59:02<68:51:30, 5.72s/it]
1%|▏ | 586/43939 [59:08<68:52:33, 5.72s/it]
1%|▏ | 587/43939 [59:13<69:00:17, 5.73s/it]
1%|▏ | 588/43939 [59:19<69:02:45, 5.73s/it]
1%|▏ | 589/43939 [59:25<68:53:52, 5.72s/it]
1%|▏ | 590/43939 [59:30<68:39:25, 5.70s/it]
1%|▏ | 591/43939 [59:36<68:40:46, 5.70s/it]
1%|▏ | 592/43939 [59:42<68:38:39, 5.70s/it]
1%|▏ | 593/43939 [59:48<68:37:30, 5.70s/it]
1%|▏ | 594/43939 [59:53<68:32:42, 5.69s/it]
1%|▏ | 595/43939 [59:59<68:42:35, 5.71s/it]
1%|▏ | 596/43939 [1:00:05<68:29:05, 5.69s/it]
1%|▏ | 597/43939 [1:00:10<68:27:07, 5.69s/it]
1%|▏ | 598/43939 [1:00:16<68:31:21, 5.69s/it]
1%|▏ | 599/43939 [1:00:22<68:33:11, 5.69s/it]
1%|▏ | 600/43939 [1:00:27<68:25:55, 5.68s/it]
{'loss': 1.2698, 'grad_norm': 2.360264539718628, 'learning_rate': 6.8176644661962216e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 535.06, 'epoch': 0.05}
1%|▏ | 600/43939 [1:00:27<68:25:55, 5.68s/it]
1%|▏ | 601/43939 [1:00:33<68:31:50, 5.69s/it]
1%|▏ | 602/43939 [1:00:39<68:39:18, 5.70s/it]
1%|▏ | 603/43939 [1:00:45<69:13:06, 5.75s/it]
1%|▏ | 604/43939 [1:00:50<68:54:34, 5.72s/it]
1%|▏ | 605/43939 [1:00:56<68:48:17, 5.72s/it]
1%|▏ | 606/43939 [1:01:02<68:41:36, 5.71s/it]
1%|▏ | 607/43939 [1:01:07<68:33:07, 5.70s/it]
1%|▏ | 608/43939 [1:01:13<68:33:36, 5.70s/it]
1%|▏ | 609/43939 [1:01:19<68:28:50, 5.69s/it]
1%|▏ | 610/43939 [1:01:24<68:38:42, 5.70s/it]
1%|▏ | 611/43939 [1:01:30<68:42:52, 5.71s/it]
1%|▏ | 612/43939 [1:01:36<68:41:23, 5.71s/it]
1%|▏ | 613/43939 [1:01:42<68:40:24, 5.71s/it]
1%|▏ | 614/43939 [1:01:47<68:44:52, 5.71s/it]
1%|▏ | 615/43939 [1:01:53<68:42:45, 5.71s/it]
1%|▏ | 616/43939 [1:01:59<68:40:21, 5.71s/it]
1%|▏ | 617/43939 [1:02:04<68:33:38, 5.70s/it]
1%|▏ | 618/43939 [1:02:10<68:34:59, 5.70s/it]
1%|▏ | 619/43939 [1:02:16<68:26:45, 5.69s/it]
1%|▏ | 620/43939 [1:02:21<68:30:40, 5.69s/it]
1%|▏ | 621/43939 [1:02:27<68:30:59, 5.69s/it]
1%|▏ | 622/43939 [1:02:33<68:40:04, 5.71s/it]
1%|▏ | 623/43939 [1:02:39<68:32:52, 5.70s/it]
1%|▏ | 624/43939 [1:02:44<69:21:04, 5.76s/it]
1%|▏ | 625/43939 [1:02:50<69:02:46, 5.74s/it]
1%|▏ | 626/43939 [1:02:56<68:59:27, 5.73s/it]
1%|▏ | 627/43939 [1:03:02<68:53:54, 5.73s/it]
1%|▏ | 628/43939 [1:03:07<68:49:38, 5.72s/it]
1%|▏ | 629/43939 [1:03:13<68:55:50, 5.73s/it]
1%|▏ | 630/43939 [1:03:19<68:49:44, 5.72s/it]
1%|▏ | 631/43939 [1:03:24<68:35:10, 5.70s/it]
1%|▏ | 632/43939 [1:03:30<68:34:44, 5.70s/it]
1%|▏ | 633/43939 [1:03:36<68:28:12, 5.69s/it]
1%|▏ | 634/43939 [1:03:41<68:29:49, 5.69s/it]
1%|▏ | 635/43939 [1:03:47<68:26:39, 5.69s/it]
1%|▏ | 636/43939 [1:03:53<68:22:06, 5.68s/it]
1%|▏ | 637/43939 [1:03:59<68:29:45, 5.69s/it]
1%|▏ | 638/43939 [1:04:04<68:30:35, 5.70s/it]
1%|▏ | 639/43939 [1:04:10<68:31:39, 5.70s/it]
1%|▏ | 640/43939 [1:04:16<68:37:47, 5.71s/it]
1%|▏ | 641/43939 [1:04:21<68:34:10, 5.70s/it]
1%|▏ | 642/43939 [1:04:27<68:41:41, 5.71s/it]
1%|▏ | 643/43939 [1:04:33<68:35:26, 5.70s/it]
1%|▏ | 644/43939 [1:04:39<69:25:23, 5.77s/it]
1%|▏ | 645/43939 [1:04:44<69:02:27, 5.74s/it]
1%|▏ | 646/43939 [1:04:50<69:01:51, 5.74s/it]
1%|▏ | 647/43939 [1:04:56<68:49:19, 5.72s/it]
1%|▏ | 648/43939 [1:05:02<68:50:05, 5.72s/it]
1%|▏ | 649/43939 [1:05:07<68:49:13, 5.72s/it]
1%|▏ | 650/43939 [1:05:13<68:48:57, 5.72s/it]
{'loss': 1.2719, 'grad_norm': 1.9422169923782349, 'learning_rate': 7.386751650352834e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 469.37, 'epoch': 0.06}
1%|▏ | 650/43939 [1:05:13<68:48:57, 5.72s/it]
1%|▏ | 651/43939 [1:05:19<68:44:30, 5.72s/it]
1%|▏ | 652/43939 [1:05:24<68:41:23, 5.71s/it]
1%|▏ | 653/43939 [1:05:30<68:32:19, 5.70s/it]
1%|▏ | 654/43939 [1:05:36<68:31:24, 5.70s/it]
1%|▏ | 655/43939 [1:05:41<68:32:34, 5.70s/it]
1%|▏ | 656/43939 [1:05:47<68:40:05, 5.71s/it]
1%|▏ | 657/43939 [1:05:53<68:31:29, 5.70s/it]
1%|▏ | 658/43939 [1:05:59<68:33:36, 5.70s/it]
1%|▏ | 659/43939 [1:06:04<68:33:11, 5.70s/it]
2%|▏ | 660/43939 [1:06:10<68:35:55, 5.71s/it]
2%|▏ | 661/43939 [1:06:16<68:32:32, 5.70s/it]
2%|▏ | 662/43939 [1:06:21<68:37:09, 5.71s/it]
2%|▏ | 663/43939 [1:06:27<68:40:13, 5.71s/it]
2%|▏ | 664/43939 [1:06:33<68:43:17, 5.72s/it]
2%|▏ | 665/43939 [1:06:39<69:28:07, 5.78s/it]
2%|▏ | 666/43939 [1:06:44<69:10:30, 5.75s/it]
2%|▏ | 667/43939 [1:06:50<68:59:43, 5.74s/it]
2%|▏ | 668/43939 [1:06:56<68:45:57, 5.72s/it]
2%|▏ | 669/43939 [1:07:02<68:40:18, 5.71s/it]
2%|▏ | 670/43939 [1:07:07<68:35:20, 5.71s/it]
2%|▏ | 671/43939 [1:07:13<68:30:41, 5.70s/it]
2%|▏ | 672/43939 [1:07:19<68:33:07, 5.70s/it]
2%|▏ | 673/43939 [1:07:24<68:37:25, 5.71s/it]
2%|▏ | 674/43939 [1:07:30<68:38:33, 5.71s/it]
2%|▏ | 675/43939 [1:07:36<68:30:27, 5.70s/it]
2%|▏ | 676/43939 [1:07:41<68:34:48, 5.71s/it]
2%|▏ | 677/43939 [1:07:47<68:46:40, 5.72s/it]
2%|▏ | 678/43939 [1:07:53<68:45:19, 5.72s/it]
2%|▏ | 679/43939 [1:07:59<68:37:26, 5.71s/it]
2%|▏ | 680/43939 [1:08:04<68:39:00, 5.71s/it]
2%|▏ | 681/43939 [1:08:10<68:43:33, 5.72s/it]
2%|▏ | 682/43939 [1:08:16<68:47:30, 5.73s/it]
2%|▏ | 683/43939 [1:08:22<68:40:47, 5.72s/it]
2%|▏ | 684/43939 [1:08:27<68:33:51, 5.71s/it]
2%|▏ | 685/43939 [1:08:33<68:36:07, 5.71s/it]
2%|▏ | 686/43939 [1:08:39<68:29:50, 5.70s/it]
2%|▏ | 687/43939 [1:08:45<69:09:37, 5.76s/it]
2%|▏ | 688/43939 [1:08:50<69:07:36, 5.75s/it]
2%|▏ | 689/43939 [1:08:56<68:54:53, 5.74s/it]
2%|▏ | 690/43939 [1:09:02<68:45:16, 5.72s/it]
2%|▏ | 691/43939 [1:09:07<68:29:29, 5.70s/it]
2%|▏ | 692/43939 [1:09:13<68:24:20, 5.69s/it]
2%|▏ | 693/43939 [1:09:19<68:31:15, 5.70s/it]
2%|▏ | 694/43939 [1:09:24<68:27:08, 5.70s/it]
2%|▏ | 695/43939 [1:09:30<68:26:51, 5.70s/it]
2%|▏ | 696/43939 [1:09:36<68:15:51, 5.68s/it]
2%|▏ | 697/43939 [1:09:41<68:19:05, 5.69s/it]
2%|▏ | 698/43939 [1:09:47<68:25:06, 5.70s/it]
2%|▏ | 699/43939 [1:09:53<68:25:51, 5.70s/it]
2%|▏ | 700/43939 [1:09:59<68:24:00, 5.69s/it]
{'loss': 1.2423, 'grad_norm': 1.6929277181625366, 'learning_rate': 7.955838834509448e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 469.51, 'epoch': 0.06}
2%|▏ | 700/43939 [1:09:59<68:24:00, 5.69s/it]
2%|▏ | 701/43939 [1:10:04<68:30:08, 5.70s/it]
2%|▏ | 702/43939 [1:10:10<68:24:01, 5.70s/it]
2%|▏ | 703/43939 [1:10:16<68:23:21, 5.69s/it]
2%|▏ | 704/43939 [1:10:21<68:18:35, 5.69s/it]
2%|▏ | 705/43939 [1:10:27<68:25:01, 5.70s/it]
2%|▏ | 706/43939 [1:10:33<68:27:02, 5.70s/it]
2%|▏ | 707/43939 [1:10:39<69:12:32, 5.76s/it]
2%|▏ | 708/43939 [1:10:44<68:48:48, 5.73s/it]
2%|▏ | 709/43939 [1:10:50<68:45:02, 5.73s/it]
2%|▏ | 710/43939 [1:10:56<68:41:35, 5.72s/it]
2%|▏ | 711/43939 [1:11:01<68:38:01, 5.72s/it]
2%|▏ | 712/43939 [1:11:07<68:25:36, 5.70s/it]
2%|▏ | 713/43939 [1:11:13<68:18:25, 5.69s/it]
2%|▏ | 714/43939 [1:11:18<68:27:22, 5.70s/it]
2%|▏ | 715/43939 [1:11:24<68:36:50, 5.71s/it]
2%|▏ | 716/43939 [1:11:30<68:34:13, 5.71s/it]
2%|▏ | 717/43939 [1:11:36<68:38:33, 5.72s/it]
2%|▏ | 718/43939 [1:11:41<68:28:47, 5.70s/it]
2%|▏ | 719/43939 [1:11:47<68:21:49, 5.69s/it]
2%|▏ | 720/43939 [1:11:53<68:27:18, 5.70s/it]
2%|▏ | 721/43939 [1:11:58<68:25:49, 5.70s/it]
2%|▏ | 722/43939 [1:12:04<68:30:38, 5.71s/it]
2%|▏ | 723/43939 [1:12:10<68:22:22, 5.70s/it]
2%|▏ | 724/43939 [1:12:15<68:19:42, 5.69s/it]
2%|▏ | 725/43939 [1:12:21<68:25:05, 5.70s/it]
2%|▏ | 726/43939 [1:12:27<68:17:03, 5.69s/it]
2%|▏ | 727/43939 [1:12:33<68:18:05, 5.69s/it]
2%|▏ | 728/43939 [1:12:38<69:11:03, 5.76s/it]
2%|▏ | 729/43939 [1:12:44<68:59:43, 5.75s/it]
2%|▏ | 730/43939 [1:12:50<68:50:26, 5.74s/it]
2%|▏ | 731/43939 [1:12:56<68:53:50, 5.74s/it]
2%|▏ | 732/43939 [1:13:01<68:49:57, 5.74s/it]
2%|▏ | 733/43939 [1:13:07<68:35:01, 5.71s/it]
2%|▏ | 734/43939 [1:13:13<68:24:40, 5.70s/it]
2%|▏ | 735/43939 [1:13:18<68:21:17, 5.70s/it]
2%|▏ | 736/43939 [1:13:24<68:11:07, 5.68s/it]
2%|▏ | 737/43939 [1:13:30<68:12:43, 5.68s/it]
2%|▏ | 738/43939 [1:13:35<68:11:55, 5.68s/it]
2%|▏ | 739/43939 [1:13:41<68:04:32, 5.67s/it]
2%|▏ | 740/43939 [1:13:47<68:08:04, 5.68s/it]
2%|▏ | 741/43939 [1:13:52<68:09:35, 5.68s/it]
2%|▏ | 742/43939 [1:13:58<68:12:49, 5.68s/it]
2%|▏ | 743/43939 [1:14:04<68:27:35, 5.71s/it]
2%|▏ | 744/43939 [1:14:10<68:18:21, 5.69s/it]
2%|▏ | 745/43939 [1:14:15<68:15:46, 5.69s/it]
2%|▏ | 746/43939 [1:14:21<68:15:41, 5.69s/it]
2%|▏ | 747/43939 [1:14:27<68:16:27, 5.69s/it]
2%|▏ | 748/43939 [1:14:32<68:12:53, 5.69s/it]
2%|▏ | 749/43939 [1:14:38<68:11:44, 5.68s/it]
2%|▏ | 750/43939 [1:14:44<69:07:49, 5.76s/it]
{'loss': 1.2189, 'grad_norm': 2.481524705886841, 'learning_rate': 8.52492601866606e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 443.79, 'epoch': 0.07}
2%|▏ | 750/43939 [1:14:44<69:07:49, 5.76s/it]
2%|▏ | 751/43939 [1:14:50<68:54:02, 5.74s/it]
2%|▏ | 752/43939 [1:14:55<68:33:36, 5.72s/it]
2%|▏ | 753/43939 [1:15:01<68:26:17, 5.71s/it]
2%|▏ | 754/43939 [1:15:07<68:18:12, 5.69s/it]
2%|▏ | 755/43939 [1:15:12<68:18:43, 5.69s/it]
2%|▏ | 756/43939 [1:15:18<68:20:18, 5.70s/it]
2%|▏ | 757/43939 [1:15:24<68:37:22, 5.72s/it]
2%|▏ | 758/43939 [1:15:29<68:31:42, 5.71s/it]
2%|▏ | 759/43939 [1:15:35<68:32:23, 5.71s/it]
2%|▏ | 760/43939 [1:15:41<68:23:00, 5.70s/it]
2%|▏ | 761/43939 [1:15:47<68:14:36, 5.69s/it]
2%|▏ | 762/43939 [1:15:52<68:13:51, 5.69s/it]
2%|▏ | 763/43939 [1:15:58<68:22:40, 5.70s/it]
2%|▏ | 764/43939 [1:16:04<68:27:31, 5.71s/it]
2%|▏ | 765/43939 [1:16:09<68:26:32, 5.71s/it]
2%|▏ | 766/43939 [1:16:15<68:23:51, 5.70s/it]
2%|▏ | 767/43939 [1:16:21<68:17:46, 5.70s/it]
2%|▏ | 768/43939 [1:16:26<68:22:49, 5.70s/it]
2%|▏ | 769/43939 [1:16:32<68:22:12, 5.70s/it]
2%|▏ | 770/43939 [1:16:38<69:03:43, 5.76s/it]
2%|▏ | 771/43939 [1:16:44<69:00:18, 5.75s/it]
2%|▏ | 772/43939 [1:16:50<68:43:55, 5.73s/it]
2%|▏ | 773/43939 [1:16:55<68:46:48, 5.74s/it]
2%|▏ | 774/43939 [1:17:01<68:37:58, 5.72s/it]
2%|▏ | 775/43939 [1:17:07<68:31:11, 5.71s/it]
2%|▏ | 776/43939 [1:17:12<68:26:40, 5.71s/it]
2%|▏ | 777/43939 [1:17:18<68:30:36, 5.71s/it]
2%|▏ | 778/43939 [1:17:24<68:40:51, 5.73s/it]
2%|▏ | 779/43939 [1:17:30<68:34:15, 5.72s/it]
2%|▏ | 780/43939 [1:17:35<68:29:15, 5.71s/it]
2%|▏ | 781/43939 [1:17:41<68:13:25, 5.69s/it]
2%|▏ | 782/43939 [1:17:47<68:24:17, 5.71s/it]
2%|▏ | 783/43939 [1:17:52<68:15:38, 5.69s/it]
2%|▏ | 784/43939 [1:17:58<68:17:23, 5.70s/it]
2%|▏ | 785/43939 [1:18:04<68:11:41, 5.69s/it]
2%|▏ | 786/43939 [1:18:09<68:13:42, 5.69s/it]
2%|▏ | 787/43939 [1:18:15<68:14:22, 5.69s/it]
2%|▏ | 788/43939 [1:18:21<68:22:12, 5.70s/it]
2%|▏ | 789/43939 [1:18:26<68:26:10, 5.71s/it]
2%|▏ | 790/43939 [1:18:32<68:28:45, 5.71s/it]
2%|▏ | 791/43939 [1:18:38<69:13:33, 5.78s/it]
2%|▏ | 792/43939 [1:18:44<68:54:28, 5.75s/it]
2%|▏ | 793/43939 [1:18:49<68:34:06, 5.72s/it]
2%|▏ | 794/43939 [1:18:55<68:29:22, 5.71s/it]
2%|▏ | 795/43939 [1:19:01<68:34:00, 5.72s/it]
2%|▏ | 796/43939 [1:19:07<68:29:59, 5.72s/it]
2%|▏ | 797/43939 [1:19:12<68:27:39, 5.71s/it]
2%|▏ | 798/43939 [1:19:18<68:24:21, 5.71s/it]
2%|▏ | 799/43939 [1:19:24<68:22:02, 5.71s/it]
2%|▏ | 800/43939 [1:19:29<68:20:13, 5.70s/it]
{'loss': 1.2364, 'grad_norm': 1.7966183423995972, 'learning_rate': 9.094013202822673e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 471.56, 'epoch': 0.07}
2%|▏ | 800/43939 [1:19:29<68:20:13, 5.70s/it]
2%|▏ | 801/43939 [1:19:35<68:16:57, 5.70s/it]
2%|▏ | 802/43939 [1:19:41<68:17:39, 5.70s/it]
2%|▏ | 803/43939 [1:19:46<68:17:46, 5.70s/it]
2%|▏ | 804/43939 [1:19:52<68:08:08, 5.69s/it]
2%|▏ | 805/43939 [1:19:58<68:00:51, 5.68s/it]
2%|▏ | 806/43939 [1:20:04<68:05:26, 5.68s/it]
2%|▏ | 807/43939 [1:20:09<68:05:41, 5.68s/it]
2%|▏ | 808/43939 [1:20:15<68:08:23, 5.69s/it]
2%|▏ | 809/43939 [1:20:21<68:19:34, 5.70s/it]
2%|▏ | 810/43939 [1:20:26<68:17:24, 5.70s/it]
2%|▏ | 811/43939 [1:20:32<68:21:55, 5.71s/it]
2%|▏ | 812/43939 [1:20:38<68:20:26, 5.70s/it]
2%|▏ | 813/43939 [1:20:44<69:06:07, 5.77s/it]
2%|▏ | 814/43939 [1:20:49<68:46:56, 5.74s/it]
2%|▏ | 815/43939 [1:20:55<68:45:38, 5.74s/it]
2%|▏ | 816/43939 [1:21:01<68:33:01, 5.72s/it]
2%|▏ | 817/43939 [1:21:07<68:42:07, 5.74s/it]
2%|▏ | 818/43939 [1:21:12<68:33:38, 5.72s/it]
2%|▏ | 819/43939 [1:21:18<68:25:35, 5.71s/it]
2%|▏ | 820/43939 [1:21:24<68:12:03, 5.69s/it]
2%|▏ | 821/43939 [1:21:29<68:19:55, 5.71s/it]
2%|▏ | 822/43939 [1:21:35<68:24:51, 5.71s/it]
2%|▏ | 823/43939 [1:21:41<68:19:59, 5.71s/it]
2%|▏ | 824/43939 [1:21:46<68:15:15, 5.70s/it]
2%|▏ | 825/43939 [1:21:52<68:04:59, 5.68s/it]
2%|▏ | 826/43939 [1:21:58<68:00:58, 5.68s/it]
2%|▏ | 827/43939 [1:22:03<68:00:25, 5.68s/it]
2%|▏ | 828/43939 [1:22:09<68:08:03, 5.69s/it]
2%|▏ | 829/43939 [1:22:15<68:09:41, 5.69s/it]
2%|▏ | 830/43939 [1:22:21<68:15:33, 5.70s/it]
2%|▏ | 831/43939 [1:22:26<68:16:31, 5.70s/it]
2%|▏ | 832/43939 [1:22:32<68:10:45, 5.69s/it]
2%|▏ | 833/43939 [1:22:38<68:58:50, 5.76s/it]
2%|▏ | 834/43939 [1:22:43<68:37:35, 5.73s/it]
2%|▏ | 835/43939 [1:22:49<68:26:56, 5.72s/it]
2%|▏ | 836/43939 [1:22:55<68:26:20, 5.72s/it]
2%|▏ | 837/43939 [1:23:01<68:20:03, 5.71s/it]
2%|▏ | 838/43939 [1:23:06<68:17:31, 5.70s/it]
2%|▏ | 839/43939 [1:23:12<68:09:00, 5.69s/it]
2%|▏ | 840/43939 [1:23:18<68:14:26, 5.70s/it]
2%|▏ | 841/43939 [1:23:23<68:27:22, 5.72s/it]
2%|▏ | 842/43939 [1:23:29<68:19:18, 5.71s/it]
2%|▏ | 843/43939 [1:23:35<68:08:11, 5.69s/it]
2%|▏ | 844/43939 [1:23:40<68:00:47, 5.68s/it]
2%|▏ | 845/43939 [1:23:46<68:04:42, 5.69s/it]
2%|▏ | 846/43939 [1:23:52<68:07:37, 5.69s/it]
2%|▏ | 847/43939 [1:23:57<68:06:55, 5.69s/it]
2%|▏ | 848/43939 [1:24:03<68:01:48, 5.68s/it]
2%|▏ | 849/43939 [1:24:09<68:00:51, 5.68s/it]
2%|▏ | 850/43939 [1:24:15<68:05:33, 5.69s/it]
{'loss': 1.183, 'grad_norm': 1.7296600341796875, 'learning_rate': 9.663100386979286e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 476.21, 'epoch': 0.08}
2%|▏ | 850/43939 [1:24:15<68:05:33, 5.69s/it]
2%|▏ | 851/43939 [1:24:20<68:03:53, 5.69s/it]
2%|▏ | 852/43939 [1:24:26<68:08:10, 5.69s/it]
2%|▏ | 853/43939 [1:24:32<68:14:09, 5.70s/it]
2%|▏ | 854/43939 [1:24:38<69:03:20, 5.77s/it]
2%|▏ | 855/43939 [1:24:43<68:48:04, 5.75s/it]
2%|▏ | 856/43939 [1:24:49<68:33:10, 5.73s/it]
2%|▏ | 857/43939 [1:24:55<68:25:14, 5.72s/it]
2%|▏ | 858/43939 [1:25:00<68:24:52, 5.72s/it]
2%|▏ | 859/43939 [1:25:06<68:29:04, 5.72s/it]
2%|▏ | 860/43939 [1:25:12<68:19:59, 5.71s/it]
2%|▏ | 861/43939 [1:25:17<68:14:59, 5.70s/it]
2%|▏ | 862/43939 [1:25:23<68:20:26, 5.71s/it]
2%|▏ | 863/43939 [1:25:29<68:23:57, 5.72s/it]
2%|▏ | 864/43939 [1:25:35<68:17:56, 5.71s/it]
2%|▏ | 865/43939 [1:25:40<68:16:42, 5.71s/it]
2%|▏ | 866/43939 [1:25:46<68:18:02, 5.71s/it]
2%|▏ | 867/43939 [1:25:52<68:26:45, 5.72s/it]
2%|▏ | 868/43939 [1:25:57<68:21:28, 5.71s/it]
2%|▏ | 869/43939 [1:26:03<68:15:10, 5.70s/it]
2%|▏ | 870/43939 [1:26:09<68:12:35, 5.70s/it]
2%|▏ | 871/43939 [1:26:15<68:11:30, 5.70s/it]
2%|▏ | 872/43939 [1:26:20<68:05:25, 5.69s/it]
2%|▏ | 873/43939 [1:26:26<68:14:34, 5.70s/it]
2%|▏ | 874/43939 [1:26:32<68:11:12, 5.70s/it]
2%|▏ | 875/43939 [1:26:37<68:02:32, 5.69s/it]
2%|▏ | 876/43939 [1:26:43<69:01:17, 5.77s/it]
2%|▏ | 877/43939 [1:26:49<68:36:02, 5.74s/it]
2%|▏ | 878/43939 [1:26:55<68:34:19, 5.73s/it]
2%|▏ | 879/43939 [1:27:00<68:39:15, 5.74s/it]
2%|▏ | 880/43939 [1:27:06<68:30:32, 5.73s/it]
2%|▏ | 881/43939 [1:27:12<68:17:10, 5.71s/it]
2%|▏ | 882/43939 [1:27:18<68:20:46, 5.71s/it]
2%|▏ | 883/43939 [1:27:23<68:17:13, 5.71s/it]
2%|▏ | 884/43939 [1:27:29<68:18:55, 5.71s/it]
2%|▏ | 885/43939 [1:27:35<68:11:56, 5.70s/it]
2%|▏ | 886/43939 [1:27:40<68:19:40, 5.71s/it]
2%|▏ | 887/43939 [1:27:46<68:09:07, 5.70s/it]
2%|▏ | 888/43939 [1:27:52<68:09:14, 5.70s/it]
2%|▏ | 889/43939 [1:27:57<68:00:21, 5.69s/it]
2%|▏ | 890/43939 [1:28:03<68:07:31, 5.70s/it]
2%|▏ | 891/43939 [1:28:09<68:09:30, 5.70s/it]
2%|▏ | 892/43939 [1:28:14<68:05:53, 5.70s/it]
2%|▏ | 893/43939 [1:28:20<67:58:06, 5.68s/it]
2%|▏ | 894/43939 [1:28:26<68:00:34, 5.69s/it]
2%|▏ | 895/43939 [1:28:32<68:01:45, 5.69s/it]
2%|▏ | 896/43939 [1:28:37<68:48:00, 5.75s/it]
2%|▏ | 897/43939 [1:28:43<68:37:13, 5.74s/it]
2%|▏ | 898/43939 [1:28:49<68:28:47, 5.73s/it]
2%|▏ | 899/43939 [1:28:55<68:20:49, 5.72s/it]
2%|▏ | 900/43939 [1:29:00<68:14:30, 5.71s/it]
{'loss': 1.1902, 'grad_norm': 1.7888613939285278, 'learning_rate': 1.0232187571135898e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 496.55, 'epoch': 0.08}
2%|▏ | 900/43939 [1:29:00<68:14:30, 5.71s/it]
2%|▏ | 901/43939 [1:29:06<68:16:32, 5.71s/it]
2%|▏ | 902/43939 [1:29:12<68:08:14, 5.70s/it]
2%|▏ | 903/43939 [1:29:17<67:55:44, 5.68s/it]
2%|▏ | 904/43939 [1:29:23<68:00:52, 5.69s/it]
2%|▏ | 905/43939 [1:29:29<68:04:58, 5.70s/it]
2%|▏ | 906/43939 [1:29:34<68:13:19, 5.71s/it]
2%|▏ | 907/43939 [1:29:40<68:13:15, 5.71s/it]
2%|▏ | 908/43939 [1:29:46<68:10:14, 5.70s/it]
2%|▏ | 909/43939 [1:29:52<68:19:52, 5.72s/it]
2%|▏ | 910/43939 [1:29:57<68:16:18, 5.71s/it]
2%|▏ | 911/43939 [1:30:03<68:15:52, 5.71s/it]
2%|▏ | 912/43939 [1:30:09<68:23:12, 5.72s/it]
2%|▏ | 913/43939 [1:30:14<68:16:33, 5.71s/it]
2%|▏ | 914/43939 [1:30:20<68:04:31, 5.70s/it]
2%|▏ | 915/43939 [1:30:26<68:04:59, 5.70s/it]
2%|▏ | 916/43939 [1:30:31<68:03:25, 5.69s/it]
2%|▏ | 917/43939 [1:30:37<68:46:58, 5.76s/it]
2%|▏ | 918/43939 [1:30:43<68:25:04, 5.73s/it]
2%|▏ | 919/43939 [1:30:49<68:21:00, 5.72s/it]
2%|▏ | 920/43939 [1:30:54<68:15:13, 5.71s/it]
2%|▏ | 921/43939 [1:31:00<68:13:40, 5.71s/it]
2%|▏ | 922/43939 [1:31:06<68:13:03, 5.71s/it]
2%|▏ | 923/43939 [1:31:11<68:03:37, 5.70s/it]
2%|▏ | 924/43939 [1:31:17<68:08:56, 5.70s/it]
2%|▏ | 925/43939 [1:31:23<68:13:18, 5.71s/it]
2%|▏ | 926/43939 [1:31:29<68:14:04, 5.71s/it]
2%|▏ | 927/43939 [1:31:34<68:10:19, 5.71s/it]
2%|▏ | 928/43939 [1:31:40<68:03:06, 5.70s/it]
2%|▏ | 929/43939 [1:31:46<68:07:21, 5.70s/it]
2%|▏ | 930/43939 [1:31:51<68:07:05, 5.70s/it]
2%|▏ | 931/43939 [1:31:57<68:04:57, 5.70s/it]
2%|▏ | 932/43939 [1:32:03<68:09:29, 5.71s/it]
2%|▏ | 933/43939 [1:32:08<67:59:11, 5.69s/it]
2%|▏ | 934/43939 [1:32:14<68:08:47, 5.70s/it]
2%|▏ | 935/43939 [1:32:20<67:59:45, 5.69s/it]
2%|▏ | 936/43939 [1:32:26<67:58:35, 5.69s/it]
2%|▏ | 937/43939 [1:32:31<67:56:13, 5.69s/it]
2%|▏ | 938/43939 [1:32:37<67:57:25, 5.69s/it]
2%|▏ | 939/43939 [1:32:43<68:42:57, 5.75s/it]
2%|▏ | 940/43939 [1:32:49<68:26:54, 5.73s/it]
2%|▏ | 941/43939 [1:32:54<68:19:06, 5.72s/it]
2%|▏ | 942/43939 [1:33:00<68:22:08, 5.72s/it]
2%|▏ | 943/43939 [1:33:06<68:32:39, 5.74s/it]
2%|▏ | 944/43939 [1:33:11<68:29:09, 5.73s/it]
2%|▏ | 945/43939 [1:33:17<68:22:43, 5.73s/it]
2%|▏ | 946/43939 [1:33:23<68:22:16, 5.73s/it]
2%|▏ | 947/43939 [1:33:29<68:20:27, 5.72s/it]
2%|▏ | 948/43939 [1:33:34<68:12:47, 5.71s/it]
2%|▏ | 949/43939 [1:33:40<68:14:38, 5.71s/it]
2%|▏ | 950/43939 [1:33:46<68:14:08, 5.71s/it]
{'loss': 1.2042, 'grad_norm': 1.9669216871261597, 'learning_rate': 1.0801274755292511e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 494.8, 'epoch': 0.09}
2%|▏ | 950/43939 [1:33:46<68:14:08, 5.71s/it]
2%|▏ | 951/43939 [1:33:51<68:06:08, 5.70s/it]
2%|▏ | 952/43939 [1:33:57<68:08:48, 5.71s/it]
2%|▏ | 953/43939 [1:34:03<68:22:49, 5.73s/it]
2%|▏ | 954/43939 [1:34:09<68:26:50, 5.73s/it]
2%|▏ | 955/43939 [1:34:14<68:21:15, 5.72s/it]
2%|▏ | 956/43939 [1:34:20<68:14:51, 5.72s/it]
2%|▏ | 957/43939 [1:34:26<68:09:44, 5.71s/it]
2%|▏ | 958/43939 [1:34:31<68:18:35, 5.72s/it]
2%|▏ | 959/43939 [1:34:37<68:09:30, 5.71s/it]
2%|▏ | 960/43939 [1:34:43<69:01:13, 5.78s/it]
2%|▏ | 961/43939 [1:34:49<68:53:15, 5.77s/it]
2%|▏ | 962/43939 [1:34:55<68:40:24, 5.75s/it]
2%|▏ | 963/43939 [1:35:00<68:35:39, 5.75s/it]
2%|▏ | 964/43939 [1:35:06<68:26:39, 5.73s/it]
2%|▏ | 965/43939 [1:35:12<68:21:48, 5.73s/it]
2%|▏ | 966/43939 [1:35:17<68:18:16, 5.72s/it]
2%|▏ | 967/43939 [1:35:23<68:09:25, 5.71s/it]
2%|▏ | 968/43939 [1:35:29<68:17:57, 5.72s/it]
2%|▏ | 969/43939 [1:35:35<68:19:32, 5.72s/it]
2%|▏ | 970/43939 [1:35:40<68:12:29, 5.71s/it]
2%|▏ | 971/43939 [1:35:46<68:16:04, 5.72s/it]
2%|▏ | 972/43939 [1:35:52<68:06:10, 5.71s/it]
2%|▏ | 973/43939 [1:35:57<68:03:15, 5.70s/it]
2%|▏ | 974/43939 [1:36:03<68:01:27, 5.70s/it]
2%|▏ | 975/43939 [1:36:09<68:01:25, 5.70s/it]
2%|▏ | 976/43939 [1:36:14<68:04:10, 5.70s/it]
2%|▏ | 977/43939 [1:36:20<68:04:34, 5.70s/it]
2%|▏ | 978/43939 [1:36:26<68:11:05, 5.71s/it]
2%|▏ | 979/43939 [1:36:32<68:12:55, 5.72s/it]
2%|▏ | 980/43939 [1:36:38<69:06:07, 5.79s/it]
2%|▏ | 981/43939 [1:36:43<68:48:09, 5.77s/it]
2%|▏ | 982/43939 [1:36:49<68:33:47, 5.75s/it]
2%|▏ | 983/43939 [1:36:55<68:20:46, 5.73s/it]
2%|▏ | 984/43939 [1:37:00<68:13:36, 5.72s/it]
2%|▏ | 985/43939 [1:37:06<68:07:36, 5.71s/it]
2%|▏ | 986/43939 [1:37:12<68:07:59, 5.71s/it]
2%|▏ | 987/43939 [1:37:18<68:12:55, 5.72s/it]
2%|▏ | 988/43939 [1:37:23<68:13:26, 5.72s/it]
2%|▏ | 989/43939 [1:37:29<68:14:20, 5.72s/it]
2%|▏ | 990/43939 [1:37:35<68:06:00, 5.71s/it]
2%|▏ | 991/43939 [1:37:40<68:16:04, 5.72s/it]
2%|▏ | 992/43939 [1:37:46<68:08:41, 5.71s/it]
2%|▏ | 993/43939 [1:37:52<68:01:34, 5.70s/it]
2%|▏ | 994/43939 [1:37:57<68:01:23, 5.70s/it]
2%|▏ | 995/43939 [1:38:03<67:57:08, 5.70s/it]
2%|▏ | 996/43939 [1:38:09<67:53:24, 5.69s/it]
2%|▏ | 997/43939 [1:38:15<68:03:29, 5.71s/it]
2%|▏ | 998/43939 [1:38:20<68:03:44, 5.71s/it]
2%|▏ | 999/43939 [1:38:26<68:00:04, 5.70s/it]
2%|▏ | 1000/43939 [1:38:32<67:55:21, 5.69s/it]
{'loss': 1.1681, 'grad_norm': 2.7225213050842285, 'learning_rate': 1.1370361939449125e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 469.73, 'epoch': 0.09}
2%|▏ | 1000/43939 [1:38:32<67:55:21, 5.69s/it]
2%|▏ | 1001/43939 [1:38:38<68:35:40, 5.75s/it]
2%|▏ | 1002/43939 [1:38:43<68:26:35, 5.74s/it]
2%|▏ | 1003/43939 [1:38:49<68:18:16, 5.73s/it]
2%|▏ | 1004/43939 [1:38:55<68:10:54, 5.72s/it]
2%|▏ | 1005/43939 [1:39:00<68:04:43, 5.71s/it]
2%|▏ | 1006/43939 [1:39:06<68:08:43, 5.71s/it]
2%|▏ | 1007/43939 [1:39:12<68:09:18, 5.72s/it]
2%|▏ | 1008/43939 [1:39:17<68:00:25, 5.70s/it]
2%|▏ | 1009/43939 [1:39:23<68:08:08, 5.71s/it]
2%|▏ | 1010/43939 [1:39:29<68:03:15, 5.71s/it]
2%|▏ | 1011/43939 [1:39:35<67:58:39, 5.70s/it]
2%|▏ | 1012/43939 [1:39:40<68:00:01, 5.70s/it]
2%|▏ | 1013/43939 [1:39:46<68:04:25, 5.71s/it]
2%|▏ | 1014/43939 [1:39:52<68:00:27, 5.70s/it]
2%|▏ | 1015/43939 [1:39:57<68:11:41, 5.72s/it]
2%|▏ | 1016/43939 [1:40:03<68:03:10, 5.71s/it]
2%|▏ | 1017/43939 [1:40:09<67:58:05, 5.70s/it]
2%|▏ | 1018/43939 [1:40:15<67:56:57, 5.70s/it]
2%|▏ | 1019/43939 [1:40:20<68:03:38, 5.71s/it]
2%|▏ | 1020/43939 [1:40:26<67:52:53, 5.69s/it]
2%|▏ | 1021/43939 [1:40:32<67:58:05, 5.70s/it]
2%|▏ | 1022/43939 [1:40:37<67:55:59, 5.70s/it]
2%|▏ | 1023/43939 [1:40:43<68:35:28, 5.75s/it]
2%|▏ | 1024/43939 [1:40:49<68:37:20, 5.76s/it]
2%|▏ | 1025/43939 [1:40:55<68:25:27, 5.74s/it]
2%|▏ | 1026/43939 [1:41:00<68:15:31, 5.73s/it]
2%|▏ | 1027/43939 [1:41:06<68:09:26, 5.72s/it]
2%|▏ | 1028/43939 [1:41:12<68:13:31, 5.72s/it]
2%|▏ | 1029/43939 [1:41:17<68:04:49, 5.71s/it]
2%|▏ | 1030/43939 [1:41:23<68:03:57, 5.71s/it]
2%|▏ | 1031/43939 [1:41:29<67:55:27, 5.70s/it]
2%|▏ | 1032/43939 [1:41:35<67:57:27, 5.70s/it]
2%|▏ | 1033/43939 [1:41:40<67:52:12, 5.69s/it]
2%|▏ | 1034/43939 [1:41:46<67:52:16, 5.69s/it]
2%|▏ | 1035/43939 [1:41:52<67:49:19, 5.69s/it]
2%|▏ | 1036/43939 [1:41:57<67:56:10, 5.70s/it]
2%|▏ | 1037/43939 [1:42:03<67:53:04, 5.70s/it]
2%|▏ | 1038/43939 [1:42:09<68:06:12, 5.71s/it]
2%|▏ | 1039/43939 [1:42:15<68:16:26, 5.73s/it]
2%|▏ | 1040/43939 [1:42:20<68:07:11, 5.72s/it]
2%|▏ | 1041/43939 [1:42:26<67:59:41, 5.71s/it]
2%|▏ | 1042/43939 [1:42:32<68:08:24, 5.72s/it]
2%|▏ | 1043/43939 [1:42:38<69:26:34, 5.83s/it]
2%|▏ | 1044/43939 [1:42:43<68:59:48, 5.79s/it]
2%|▏ | 1045/43939 [1:42:49<68:40:04, 5.76s/it]
2%|▏ | 1046/43939 [1:42:55<68:30:23, 5.75s/it]
2%|▏ | 1047/43939 [1:43:01<68:27:28, 5.75s/it]
2%|▏ | 1048/43939 [1:43:06<68:16:55, 5.73s/it]
2%|▏ | 1049/43939 [1:43:12<68:09:22, 5.72s/it]
2%|▏ | 1050/43939 [1:43:18<68:02:15, 5.71s/it]
{'loss': 1.1957, 'grad_norm': 2.091661214828491, 'learning_rate': 1.1939449123605736e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 501.85, 'epoch': 0.1}
2%|▏ | 1050/43939 [1:43:18<68:02:15, 5.71s/it]
2%|▏ | 1051/43939 [1:43:23<68:09:01, 5.72s/it]
2%|▏ | 1052/43939 [1:43:29<68:03:07, 5.71s/it]
2%|▏ | 1053/43939 [1:43:35<67:57:47, 5.71s/it]
2%|▏ | 1054/43939 [1:43:40<67:54:55, 5.70s/it]
2%|▏ | 1055/43939 [1:43:46<68:02:24, 5.71s/it]
2%|▏ | 1056/43939 [1:43:52<68:04:53, 5.72s/it]
2%|▏ | 1057/43939 [1:43:58<67:59:11, 5.71s/it]
2%|▏ | 1058/43939 [1:44:03<67:50:48, 5.70s/it]
2%|▏ | 1059/43939 [1:44:09<67:52:55, 5.70s/it]
2%|▏ | 1060/43939 [1:44:15<67:54:03, 5.70s/it]
2%|▏ | 1061/43939 [1:44:20<67:57:57, 5.71s/it]
2%|▏ | 1062/43939 [1:44:26<67:59:48, 5.71s/it]
2%|▏ | 1063/43939 [1:44:32<67:59:48, 5.71s/it]
2%|▏ | 1064/43939 [1:44:38<68:46:27, 5.77s/it]
2%|▏ | 1065/43939 [1:44:44<68:41:41, 5.77s/it]
2%|▏ | 1066/43939 [1:44:49<68:26:04, 5.75s/it]
2%|▏ | 1067/43939 [1:44:55<68:19:58, 5.74s/it]
2%|▏ | 1068/43939 [1:45:01<68:06:51, 5.72s/it]
2%|▏ | 1069/43939 [1:45:06<68:03:22, 5.71s/it]
2%|▏ | 1070/43939 [1:45:12<67:49:57, 5.70s/it]
2%|▏ | 1071/43939 [1:45:18<67:51:53, 5.70s/it]
2%|▏ | 1072/43939 [1:45:23<67:47:59, 5.69s/it]
2%|▏ | 1073/43939 [1:45:29<67:47:02, 5.69s/it]
2%|▏ | 1074/43939 [1:45:35<67:45:46, 5.69s/it]
2%|▏ | 1075/43939 [1:45:40<67:53:40, 5.70s/it]
2%|▏ | 1076/43939 [1:45:46<67:48:22, 5.69s/it]
2%|▏ | 1077/43939 [1:45:52<67:59:57, 5.71s/it]
2%|▏ | 1078/43939 [1:45:58<68:09:06, 5.72s/it]
2%|▏ | 1079/43939 [1:46:03<68:08:09, 5.72s/it]
2%|▏ | 1080/43939 [1:46:09<68:18:16, 5.74s/it]
2%|▏ | 1081/43939 [1:46:15<68:14:22, 5.73s/it]
2%|▏ | 1082/43939 [1:46:21<68:06:54, 5.72s/it]
2%|▏ | 1083/43939 [1:46:26<67:58:12, 5.71s/it]
2%|▏ | 1084/43939 [1:46:32<67:52:20, 5.70s/it]
2%|▏ | 1085/43939 [1:46:38<67:41:06, 5.69s/it]
2%|▏ | 1086/43939 [1:46:44<68:45:59, 5.78s/it]
2%|▏ | 1087/43939 [1:46:49<68:27:20, 5.75s/it]
2%|▏ | 1088/43939 [1:46:55<68:15:24, 5.73s/it]
2%|▏ | 1089/43939 [1:47:01<68:11:14, 5.73s/it]
2%|▏ | 1090/43939 [1:47:06<68:18:47, 5.74s/it]
2%|▏ | 1091/43939 [1:47:12<68:20:54, 5.74s/it]
2%|▏ | 1092/43939 [1:47:18<68:25:05, 5.75s/it]
2%|▏ | 1093/43939 [1:47:24<68:10:07, 5.73s/it]
2%|▏ | 1094/43939 [1:47:29<68:06:00, 5.72s/it]
2%|▏ | 1095/43939 [1:47:35<67:58:07, 5.71s/it]
2%|▏ | 1096/43939 [1:47:41<67:45:48, 5.69s/it]
2%|▏ | 1097/43939 [1:47:46<67:45:10, 5.69s/it]
2%|▏ | 1098/43939 [1:47:52<67:40:36, 5.69s/it]
3%|▎ | 1099/43939 [1:47:58<67:36:31, 5.68s/it]
3%|▎ | 1100/43939 [1:48:03<67:39:16, 5.69s/it]
{'loss': 1.2117, 'grad_norm': 2.423203945159912, 'learning_rate': 1.250853630776235e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 475.58, 'epoch': 0.1}
3%|▎ | 1100/43939 [1:48:03<67:39:16, 5.69s/it]
3%|▎ | 1101/43939 [1:48:09<67:51:28, 5.70s/it]
3%|▎ | 1102/43939 [1:48:15<67:44:19, 5.69s/it]
3%|▎ | 1103/43939 [1:48:21<67:48:58, 5.70s/it]
3%|▎ | 1104/43939 [1:48:26<67:40:57, 5.69s/it]
3%|▎ | 1105/43939 [1:48:32<67:35:30, 5.68s/it]
3%|▎ | 1106/43939 [1:48:38<68:25:37, 5.75s/it]
3%|▎ | 1107/43939 [1:48:43<68:15:20, 5.74s/it]
3%|▎ | 1108/43939 [1:48:49<67:57:21, 5.71s/it]
3%|▎ | 1109/43939 [1:48:55<67:51:56, 5.70s/it]
3%|▎ | 1110/43939 [1:49:01<67:52:36, 5.71s/it]
3%|▎ | 1111/43939 [1:49:06<67:48:16, 5.70s/it]
3%|▎ | 1112/43939 [1:49:12<67:54:38, 5.71s/it]
3%|▎ | 1113/43939 [1:49:18<67:52:16, 5.71s/it]
3%|▎ | 1114/43939 [1:49:23<68:05:04, 5.72s/it]
3%|▎ | 1115/43939 [1:49:29<67:58:56, 5.71s/it]
3%|▎ | 1116/43939 [1:49:35<67:48:23, 5.70s/it]
3%|▎ | 1117/43939 [1:49:40<67:48:13, 5.70s/it]
3%|▎ | 1118/43939 [1:49:46<67:43:19, 5.69s/it]
3%|▎ | 1119/43939 [1:49:52<68:07:34, 5.73s/it]
3%|▎ | 1120/43939 [1:49:58<67:57:31, 5.71s/it]
3%|▎ | 1121/43939 [1:50:03<67:53:35, 5.71s/it]
3%|▎ | 1122/43939 [1:50:09<67:58:52, 5.72s/it]
3%|▎ | 1123/43939 [1:50:15<67:55:19, 5.71s/it]
3%|▎ | 1124/43939 [1:50:20<67:52:47, 5.71s/it]
3%|▎ | 1125/43939 [1:50:26<67:43:46, 5.70s/it]
3%|▎ | 1126/43939 [1:50:32<67:52:45, 5.71s/it]
3%|▎ | 1127/43939 [1:50:38<68:38:58, 5.77s/it]
3%|▎ | 1128/43939 [1:50:44<68:34:42, 5.77s/it]
3%|▎ | 1129/43939 [1:50:49<68:14:37, 5.74s/it]
3%|▎ | 1130/43939 [1:50:55<68:04:19, 5.72s/it]
3%|▎ | 1131/43939 [1:51:01<67:58:17, 5.72s/it]
3%|▎ | 1132/43939 [1:51:06<67:47:36, 5.70s/it]
3%|▎ | 1133/43939 [1:51:12<67:50:38, 5.71s/it]
3%|▎ | 1134/43939 [1:51:18<67:57:26, 5.72s/it]
3%|▎ | 1135/43939 [1:51:23<67:45:10, 5.70s/it]
3%|▎ | 1136/43939 [1:51:29<67:57:46, 5.72s/it]
3%|▎ | 1137/43939 [1:51:35<67:54:21, 5.71s/it]
3%|▎ | 1138/43939 [1:51:41<67:53:38, 5.71s/it]
3%|▎ | 1139/43939 [1:51:46<67:43:11, 5.70s/it]
3%|▎ | 1140/43939 [1:51:52<67:49:47, 5.71s/it]
3%|▎ | 1141/43939 [1:51:58<67:50:15, 5.71s/it]
3%|▎ | 1142/43939 [1:52:03<67:47:38, 5.70s/it]
3%|▎ | 1143/43939 [1:52:09<67:41:35, 5.69s/it]
3%|▎ | 1144/43939 [1:52:15<67:40:16, 5.69s/it]
3%|▎ | 1145/43939 [1:52:20<67:36:10, 5.69s/it]
3%|▎ | 1146/43939 [1:52:26<67:38:10, 5.69s/it]
3%|▎ | 1147/43939 [1:52:32<67:51:27, 5.71s/it]
3%|▎ | 1148/43939 [1:52:38<68:01:57, 5.72s/it]
3%|▎ | 1149/43939 [1:52:44<68:44:31, 5.78s/it]
3%|▎ | 1150/43939 [1:52:49<68:31:37, 5.77s/it]
{'loss': 1.2049, 'grad_norm': 1.9933820962905884, 'learning_rate': 1.3077623491918964e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 484.91, 'epoch': 0.1}
3%|▎ | 1150/43939 [1:52:49<68:31:37, 5.77s/it]
3%|▎ | 1151/43939 [1:52:55<68:19:50, 5.75s/it]
3%|▎ | 1152/43939 [1:53:01<68:12:52, 5.74s/it]
3%|▎ | 1153/43939 [1:53:06<68:01:54, 5.72s/it]
3%|▎ | 1154/43939 [1:53:12<67:59:21, 5.72s/it]
3%|▎ | 1155/43939 [1:53:18<68:07:17, 5.73s/it]
3%|▎ | 1156/43939 [1:53:23<67:49:52, 5.71s/it]
3%|▎ | 1157/43939 [1:53:29<67:40:42, 5.69s/it]
3%|▎ | 1158/43939 [1:53:35<67:37:22, 5.69s/it]
3%|▎ | 1159/43939 [1:53:40<67:32:44, 5.68s/it]
3%|▎ | 1160/43939 [1:53:46<67:41:59, 5.70s/it]
3%|▎ | 1161/43939 [1:53:52<67:34:30, 5.69s/it]
3%|▎ | 1162/43939 [1:53:58<67:38:17, 5.69s/it]
3%|▎ | 1163/43939 [1:54:03<67:49:08, 5.71s/it]
3%|▎ | 1164/43939 [1:54:09<67:48:26, 5.71s/it]
3%|▎ | 1165/43939 [1:54:15<67:48:29, 5.71s/it]
3%|▎ | 1166/43939 [1:54:20<67:49:46, 5.71s/it]
3%|▎ | 1167/43939 [1:54:26<67:44:31, 5.70s/it]
3%|▎ | 1168/43939 [1:54:32<67:42:57, 5.70s/it]
3%|▎ | 1169/43939 [1:54:38<68:35:46, 5.77s/it]
3%|▎ | 1170/43939 [1:54:43<68:14:36, 5.74s/it]
3%|▎ | 1171/43939 [1:54:49<67:55:10, 5.72s/it]
3%|▎ | 1172/43939 [1:54:55<67:49:16, 5.71s/it]
3%|▎ | 1173/43939 [1:55:00<67:42:24, 5.70s/it]
3%|▎ | 1174/43939 [1:55:06<67:43:34, 5.70s/it]
3%|▎ | 1175/43939 [1:55:12<67:42:28, 5.70s/it]
3%|▎ | 1176/43939 [1:55:18<67:39:36, 5.70s/it]
3%|▎ | 1177/43939 [1:55:23<67:43:45, 5.70s/it]
3%|▎ | 1178/43939 [1:55:29<67:42:02, 5.70s/it]
3%|▎ | 1179/43939 [1:55:35<67:33:29, 5.69s/it]
3%|▎ | 1180/43939 [1:55:40<67:28:23, 5.68s/it]
3%|▎ | 1181/43939 [1:55:46<67:35:29, 5.69s/it]
3%|▎ | 1182/43939 [1:55:52<67:38:01, 5.69s/it]
3%|▎ | 1183/43939 [1:55:57<67:40:51, 5.70s/it]
3%|▎ | 1184/43939 [1:56:03<67:57:11, 5.72s/it]
3%|▎ | 1185/43939 [1:56:09<67:47:12, 5.71s/it]
3%|▎ | 1186/43939 [1:56:15<67:48:22, 5.71s/it]
3%|▎ | 1187/43939 [1:56:20<67:50:51, 5.71s/it]
3%|▎ | 1188/43939 [1:56:26<67:46:19, 5.71s/it]
3%|▎ | 1189/43939 [1:56:32<67:44:15, 5.70s/it]
3%|▎ | 1190/43939 [1:56:38<68:28:16, 5.77s/it]
3%|▎ | 1191/43939 [1:56:43<68:04:18, 5.73s/it]
3%|▎ | 1192/43939 [1:56:49<67:53:55, 5.72s/it]
3%|▎ | 1193/43939 [1:56:55<67:54:53, 5.72s/it]
3%|▎ | 1194/43939 [1:57:00<67:45:30, 5.71s/it]
3%|▎ | 1195/43939 [1:57:06<67:42:14, 5.70s/it]
3%|▎ | 1196/43939 [1:57:12<67:42:33, 5.70s/it]
3%|▎ | 1197/43939 [1:57:17<67:31:48, 5.69s/it]
3%|▎ | 1198/43939 [1:57:23<67:32:01, 5.69s/it]
3%|▎ | 1199/43939 [1:57:29<67:36:42, 5.69s/it]
3%|▎ | 1200/43939 [1:57:34<67:35:58, 5.69s/it]
{'loss': 1.1552, 'grad_norm': 1.9259192943572998, 'learning_rate': 1.3646710676075574e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 464.56, 'epoch': 0.11}
3%|▎ | 1200/43939 [1:57:35<67:35:58, 5.69s/it]
3%|▎ | 1201/43939 [1:57:40<67:36:10, 5.69s/it]
3%|▎ | 1202/43939 [1:57:46<67:39:13, 5.70s/it]
3%|▎ | 1203/43939 [1:57:52<67:35:32, 5.69s/it]
3%|▎ | 1204/43939 [1:57:57<67:42:17, 5.70s/it]
3%|▎ | 1205/43939 [1:58:03<67:45:56, 5.71s/it]
3%|▎ | 1206/43939 [1:58:09<67:48:09, 5.71s/it]
3%|▎ | 1207/43939 [1:58:14<67:46:25, 5.71s/it]
3%|▎ | 1208/43939 [1:58:20<67:45:07, 5.71s/it]
3%|▎ | 1209/43939 [1:58:26<67:50:03, 5.72s/it]
3%|▎ | 1210/43939 [1:58:32<67:44:21, 5.71s/it]
3%|▎ | 1211/43939 [1:58:37<67:44:13, 5.71s/it]
3%|▎ | 1212/43939 [1:58:43<68:31:15, 5.77s/it]
3%|▎ | 1213/43939 [1:58:49<68:16:01, 5.75s/it]
3%|▎ | 1214/43939 [1:58:55<68:02:26, 5.73s/it]
3%|▎ | 1215/43939 [1:59:00<67:51:53, 5.72s/it]
3%|▎ | 1216/43939 [1:59:06<67:54:35, 5.72s/it]
3%|▎ | 1217/43939 [1:59:12<67:46:50, 5.71s/it]
3%|▎ | 1218/43939 [1:59:17<67:44:14, 5.71s/it]
3%|▎ | 1219/43939 [1:59:23<67:40:53, 5.70s/it]
3%|▎ | 1220/43939 [1:59:29<67:47:50, 5.71s/it]
3%|▎ | 1221/43939 [1:59:35<67:48:04, 5.71s/it]
3%|▎ | 1222/43939 [1:59:40<67:42:37, 5.71s/it]
3%|▎ | 1223/43939 [1:59:46<67:50:43, 5.72s/it]
3%|▎ | 1224/43939 [1:59:52<67:42:00, 5.71s/it]
3%|▎ | 1225/43939 [1:59:57<67:36:43, 5.70s/it]
3%|▎ | 1226/43939 [2:00:03<67:46:12, 5.71s/it]
3%|▎ | 1227/43939 [2:00:09<67:35:56, 5.70s/it]
3%|▎ | 1228/43939 [2:00:14<67:35:27, 5.70s/it]
3%|▎ | 1229/43939 [2:00:20<67:37:53, 5.70s/it]
3%|▎ | 1230/43939 [2:00:26<67:42:05, 5.71s/it]
3%|▎ | 1231/43939 [2:00:32<67:35:12, 5.70s/it]
3%|▎ | 1232/43939 [2:00:37<67:35:49, 5.70s/it]
3%|▎ | 1233/43939 [2:00:43<68:20:32, 5.76s/it]
3%|▎ | 1234/43939 [2:00:49<68:06:04, 5.74s/it]
3%|▎ | 1235/43939 [2:00:55<67:54:13, 5.72s/it]
3%|▎ | 1236/43939 [2:01:00<67:45:50, 5.71s/it]
3%|▎ | 1237/43939 [2:01:06<67:33:32, 5.70s/it]
3%|▎ | 1238/43939 [2:01:12<67:34:52, 5.70s/it]
3%|▎ | 1239/43939 [2:01:17<67:42:22, 5.71s/it]
3%|▎ | 1240/43939 [2:01:23<67:51:29, 5.72s/it]
3%|▎ | 1241/43939 [2:01:29<67:51:22, 5.72s/it]
3%|▎ | 1242/43939 [2:01:35<67:56:11, 5.73s/it]
3%|▎ | 1243/43939 [2:01:40<67:56:49, 5.73s/it]
3%|▎ | 1244/43939 [2:01:46<67:58:31, 5.73s/it]
3%|▎ | 1245/43939 [2:01:52<67:45:57, 5.71s/it]
3%|▎ | 1246/43939 [2:01:57<67:40:37, 5.71s/it]
3%|▎ | 1247/43939 [2:02:03<67:31:56, 5.69s/it]
3%|▎ | 1248/43939 [2:02:09<67:43:19, 5.71s/it]
3%|▎ | 1249/43939 [2:02:14<67:41:59, 5.71s/it]
3%|▎ | 1250/43939 [2:02:20<67:38:07, 5.70s/it]
{'loss': 1.1847, 'grad_norm': 1.8878772258758545, 'learning_rate': 1.4215797860232188e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 532.02, 'epoch': 0.11}
3%|▎ | 1250/43939 [2:02:20<67:38:07, 5.70s/it]
3%|▎ | 1251/43939 [2:02:26<67:42:31, 5.71s/it]
3%|▎ | 1252/43939 [2:02:32<67:35:22, 5.70s/it]
3%|▎ | 1253/43939 [2:02:37<67:41:49, 5.71s/it]
3%|▎ | 1254/43939 [2:02:43<68:18:01, 5.76s/it]
3%|▎ | 1255/43939 [2:02:49<68:06:05, 5.74s/it]
3%|▎ | 1256/43939 [2:02:55<68:00:48, 5.74s/it]
3%|▎ | 1257/43939 [2:03:00<67:52:03, 5.72s/it]
3%|▎ | 1258/43939 [2:03:06<68:08:11, 5.75s/it]
3%|▎ | 1259/43939 [2:03:12<68:08:18, 5.75s/it]
3%|▎ | 1260/43939 [2:03:18<68:01:06, 5.74s/it]
3%|▎ | 1261/43939 [2:03:23<67:56:01, 5.73s/it]
3%|▎ | 1262/43939 [2:03:29<67:45:38, 5.72s/it]
3%|▎ | 1263/43939 [2:03:35<67:37:45, 5.70s/it]
3%|▎ | 1264/43939 [2:03:40<67:31:02, 5.70s/it]
3%|▎ | 1265/43939 [2:03:46<67:47:11, 5.72s/it]
3%|▎ | 1266/43939 [2:03:52<67:54:50, 5.73s/it]
3%|▎ | 1267/43939 [2:03:58<67:47:05, 5.72s/it]
3%|▎ | 1268/43939 [2:04:03<67:44:35, 5.72s/it]
3%|▎ | 1269/43939 [2:04:09<67:43:34, 5.71s/it]
3%|▎ | 1270/43939 [2:04:15<67:32:36, 5.70s/it]
3%|▎ | 1271/43939 [2:04:20<67:27:51, 5.69s/it]
3%|▎ | 1272/43939 [2:04:26<67:25:24, 5.69s/it]
3%|▎ | 1273/43939 [2:04:32<67:26:31, 5.69s/it]
3%|▎ | 1274/43939 [2:04:38<68:23:48, 5.77s/it]
3%|▎ | 1275/43939 [2:04:43<68:07:10, 5.75s/it]
3%|▎ | 1276/43939 [2:04:49<68:02:53, 5.74s/it]
3%|▎ | 1277/43939 [2:04:55<67:53:31, 5.73s/it]
3%|▎ | 1278/43939 [2:05:00<67:45:45, 5.72s/it]
3%|▎ | 1279/43939 [2:05:06<67:46:57, 5.72s/it]
3%|▎ | 1280/43939 [2:05:12<67:38:41, 5.71s/it]
3%|▎ | 1281/43939 [2:05:18<67:37:03, 5.71s/it]
3%|▎ | 1282/43939 [2:05:23<67:44:44, 5.72s/it]
3%|▎ | 1283/43939 [2:05:29<67:31:22, 5.70s/it]
3%|▎ | 1284/43939 [2:05:35<67:32:40, 5.70s/it]
3%|▎ | 1285/43939 [2:05:40<67:28:01, 5.69s/it]
3%|▎ | 1286/43939 [2:05:46<67:44:21, 5.72s/it]
3%|▎ | 1287/43939 [2:05:52<67:40:25, 5.71s/it]
3%|▎ | 1288/43939 [2:05:58<67:40:53, 5.71s/it]
3%|▎ | 1289/43939 [2:06:03<67:29:54, 5.70s/it]
3%|▎ | 1290/43939 [2:06:09<67:29:46, 5.70s/it]
3%|▎ | 1291/43939 [2:06:15<67:30:23, 5.70s/it]
3%|▎ | 1292/43939 [2:06:20<67:29:38, 5.70s/it]
3%|▎ | 1293/43939 [2:06:26<67:30:17, 5.70s/it]
3%|▎ | 1294/43939 [2:06:32<67:28:22, 5.70s/it]
3%|▎ | 1295/43939 [2:06:37<67:23:19, 5.69s/it]
3%|▎ | 1296/43939 [2:06:43<68:21:31, 5.77s/it]
3%|▎ | 1297/43939 [2:06:49<68:10:03, 5.75s/it]
3%|▎ | 1298/43939 [2:06:55<67:56:53, 5.74s/it]
3%|▎ | 1299/43939 [2:07:00<67:48:25, 5.72s/it]
3%|▎ | 1300/43939 [2:07:06<67:39:51, 5.71s/it]
{'loss': 1.1844, 'grad_norm': 2.4258198738098145, 'learning_rate': 1.4784885044388803e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 480.37, 'epoch': 0.12}
3%|▎ | 1300/43939 [2:07:06<67:39:51, 5.71s/it]
3%|▎ | 1301/43939 [2:07:12<67:35:00, 5.71s/it]
3%|▎ | 1302/43939 [2:07:18<67:41:22, 5.72s/it]
3%|▎ | 1303/43939 [2:07:23<67:35:39, 5.71s/it]
3%|▎ | 1304/43939 [2:07:29<67:34:24, 5.71s/it]
3%|▎ | 1305/43939 [2:07:35<67:26:25, 5.69s/it]
3%|▎ | 1306/43939 [2:07:40<67:32:43, 5.70s/it]
3%|▎ | 1307/43939 [2:07:46<67:28:05, 5.70s/it]
3%|▎ | 1308/43939 [2:07:52<67:27:39, 5.70s/it]
3%|▎ | 1309/43939 [2:07:57<67:28:26, 5.70s/it]
3%|▎ | 1310/43939 [2:08:03<67:42:17, 5.72s/it]
3%|▎ | 1311/43939 [2:08:09<67:36:30, 5.71s/it]
3%|▎ | 1312/43939 [2:08:15<67:36:56, 5.71s/it]
3%|▎ | 1313/43939 [2:08:20<67:30:28, 5.70s/it]
3%|▎ | 1314/43939 [2:08:26<67:24:43, 5.69s/it]
3%|▎ | 1315/43939 [2:08:32<67:31:21, 5.70s/it]
3%|▎ | 1316/43939 [2:08:38<68:17:38, 5.77s/it]
3%|▎ | 1317/43939 [2:08:43<68:00:30, 5.74s/it]
3%|▎ | 1318/43939 [2:08:49<68:13:13, 5.76s/it]
3%|▎ | 1319/43939 [2:08:55<67:55:30, 5.74s/it]
3%|▎ | 1320/43939 [2:09:00<67:46:03, 5.72s/it]
3%|▎ | 1321/43939 [2:09:06<67:35:09, 5.71s/it]
3%|▎ | 1322/43939 [2:09:12<67:26:22, 5.70s/it]
3%|▎ | 1323/43939 [2:09:17<67:33:18, 5.71s/it]
3%|▎ | 1324/43939 [2:09:23<67:33:32, 5.71s/it]
3%|▎ | 1325/43939 [2:09:29<67:32:26, 5.71s/it]
3%|▎ | 1326/43939 [2:09:35<67:29:04, 5.70s/it]
3%|▎ | 1327/43939 [2:09:40<67:38:33, 5.71s/it]
3%|▎ | 1328/43939 [2:09:46<67:40:39, 5.72s/it]
3%|▎ | 1329/43939 [2:09:52<67:36:22, 5.71s/it]
3%|▎ | 1330/43939 [2:09:57<67:24:43, 5.70s/it]
3%|▎ | 1331/43939 [2:10:03<67:21:20, 5.69s/it]
3%|▎ | 1332/43939 [2:10:09<67:28:20, 5.70s/it]
3%|▎ | 1333/43939 [2:10:14<67:23:13, 5.69s/it]
3%|▎ | 1334/43939 [2:10:20<67:31:15, 5.71s/it]
3%|▎ | 1335/43939 [2:10:26<67:18:31, 5.69s/it]
3%|▎ | 1336/43939 [2:10:32<67:18:07, 5.69s/it]
3%|▎ | 1337/43939 [2:10:38<68:17:50, 5.77s/it]
3%|▎ | 1338/43939 [2:10:43<68:03:52, 5.75s/it]
3%|▎ | 1339/43939 [2:10:49<67:51:18, 5.73s/it]
3%|▎ | 1340/43939 [2:10:55<67:46:56, 5.73s/it]
3%|▎ | 1341/43939 [2:11:00<67:37:20, 5.71s/it]
3%|▎ | 1342/43939 [2:11:06<67:29:00, 5.70s/it]
3%|▎ | 1343/43939 [2:11:12<67:24:13, 5.70s/it]
3%|▎ | 1344/43939 [2:11:17<67:28:10, 5.70s/it]
3%|▎ | 1345/43939 [2:11:23<67:29:26, 5.70s/it]
3%|▎ | 1346/43939 [2:11:29<67:27:59, 5.70s/it]
3%|▎ | 1347/43939 [2:11:35<67:31:42, 5.71s/it]
3%|▎ | 1348/43939 [2:11:40<67:36:32, 5.71s/it]
3%|▎ | 1349/43939 [2:11:46<67:32:21, 5.71s/it]
3%|▎ | 1350/43939 [2:11:52<67:24:51, 5.70s/it]
{'loss': 1.1706, 'grad_norm': 1.9142098426818848, 'learning_rate': 1.5353972228545412e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 484.91, 'epoch': 0.12}
3%|▎ | 1350/43939 [2:11:52<67:24:51, 5.70s/it]
3%|▎ | 1351/43939 [2:11:57<67:28:00, 5.70s/it]
3%|▎ | 1352/43939 [2:12:03<67:23:56, 5.70s/it]
3%|▎ | 1353/43939 [2:12:09<67:25:55, 5.70s/it]
3%|▎ | 1354/43939 [2:12:14<67:27:49, 5.70s/it]
3%|▎ | 1355/43939 [2:12:20<67:24:30, 5.70s/it]
3%|▎ | 1356/43939 [2:12:26<67:45:22, 5.73s/it]
3%|▎ | 1357/43939 [2:12:32<67:42:03, 5.72s/it]
3%|▎ | 1358/43939 [2:12:37<67:38:17, 5.72s/it]
3%|▎ | 1359/43939 [2:12:43<68:23:17, 5.78s/it]
3%|▎ | 1360/43939 [2:12:49<68:06:41, 5.76s/it]
3%|▎ | 1361/43939 [2:12:55<68:00:08, 5.75s/it]
3%|▎ | 1362/43939 [2:13:00<67:48:42, 5.73s/it]
3%|▎ | 1363/43939 [2:13:06<67:44:09, 5.73s/it]
3%|▎ | 1364/43939 [2:13:12<67:42:20, 5.72s/it]
3%|▎ | 1365/43939 [2:13:18<67:35:49, 5.72s/it]
3%|▎ | 1366/43939 [2:13:23<67:32:14, 5.71s/it]
3%|▎ | 1367/43939 [2:13:29<67:47:00, 5.73s/it]
3%|▎ | 1368/43939 [2:13:35<67:43:02, 5.73s/it]
3%|▎ | 1369/43939 [2:13:40<67:46:45, 5.73s/it]
3%|▎ | 1370/43939 [2:13:46<67:44:19, 5.73s/it]
3%|▎ | 1371/43939 [2:13:52<67:32:33, 5.71s/it]
3%|▎ | 1372/43939 [2:13:58<67:28:26, 5.71s/it]
3%|▎ | 1373/43939 [2:14:03<67:28:40, 5.71s/it]
3%|▎ | 1374/43939 [2:14:09<67:23:20, 5.70s/it]
3%|▎ | 1375/43939 [2:14:15<67:24:20, 5.70s/it]
3%|▎ | 1376/43939 [2:14:20<67:27:20, 5.71s/it]
3%|▎ | 1377/43939 [2:14:26<67:17:48, 5.69s/it]
3%|▎ | 1378/43939 [2:14:32<67:18:41, 5.69s/it]
3%|▎ | 1379/43939 [2:14:37<67:22:33, 5.70s/it]
3%|▎ | 1380/43939 [2:14:43<68:04:57, 5.76s/it]
3%|▎ | 1381/43939 [2:14:49<67:51:15, 5.74s/it]
3%|▎ | 1382/43939 [2:14:55<67:55:53, 5.75s/it]
3%|▎ | 1383/43939 [2:15:00<67:41:51, 5.73s/it]
3%|▎ | 1384/43939 [2:15:06<67:35:40, 5.72s/it]
3%|▎ | 1385/43939 [2:15:12<67:28:08, 5.71s/it]
3%|▎ | 1386/43939 [2:15:18<67:31:28, 5.71s/it]
3%|▎ | 1387/43939 [2:15:23<67:32:41, 5.71s/it]
3%|▎ | 1388/43939 [2:15:29<67:22:59, 5.70s/it]
3%|▎ | 1389/43939 [2:15:35<67:27:35, 5.71s/it]
3%|▎ | 1390/43939 [2:15:40<67:25:51, 5.71s/it]
3%|▎ | 1391/43939 [2:15:46<67:24:47, 5.70s/it]
3%|▎ | 1392/43939 [2:15:52<67:23:45, 5.70s/it]
3%|▎ | 1393/43939 [2:15:58<67:24:44, 5.70s/it]
3%|▎ | 1394/43939 [2:16:03<67:22:14, 5.70s/it]
3%|▎ | 1395/43939 [2:16:09<67:18:59, 5.70s/it]
3%|▎ | 1396/43939 [2:16:15<67:15:23, 5.69s/it]
3%|▎ | 1397/43939 [2:16:20<67:21:11, 5.70s/it]
3%|▎ | 1398/43939 [2:16:26<67:21:27, 5.70s/it]
3%|▎ | 1399/43939 [2:16:32<67:20:07, 5.70s/it]
3%|▎ | 1400/43939 [2:16:38<68:07:22, 5.77s/it]
{'loss': 1.1494, 'grad_norm': 1.8054569959640503, 'learning_rate': 1.5923059412702026e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 469.8, 'epoch': 0.13}
3%|▎ | 1400/43939 [2:16:38<68:07:22, 5.77s/it]
3%|▎ | 1401/43939 [2:16:43<67:52:05, 5.74s/it]
3%|▎ | 1402/43939 [2:16:49<67:55:00, 5.75s/it]
3%|▎ | 1403/43939 [2:16:55<67:49:15, 5.74s/it]
3%|▎ | 1404/43939 [2:17:01<67:47:55, 5.74s/it]
3%|▎ | 1405/43939 [2:17:06<67:35:12, 5.72s/it]
3%|▎ | 1406/43939 [2:17:12<67:28:45, 5.71s/it]
3%|▎ | 1407/43939 [2:17:18<67:28:00, 5.71s/it]
3%|▎ | 1408/43939 [2:17:23<67:25:34, 5.71s/it]
3%|▎ | 1409/43939 [2:17:29<67:20:14, 5.70s/it]
3%|▎ | 1410/43939 [2:17:35<67:22:27, 5.70s/it]
3%|▎ | 1411/43939 [2:17:40<67:21:19, 5.70s/it]
3%|▎ | 1412/43939 [2:17:46<67:13:07, 5.69s/it]
3%|▎ | 1413/43939 [2:17:52<67:16:14, 5.69s/it]
3%|▎ | 1414/43939 [2:17:57<67:08:46, 5.68s/it]
3%|▎ | 1415/43939 [2:18:03<67:29:12, 5.71s/it]
3%|▎ | 1416/43939 [2:18:09<67:24:53, 5.71s/it]
3%|▎ | 1417/43939 [2:18:15<67:18:41, 5.70s/it]
3%|▎ | 1418/43939 [2:18:20<67:27:43, 5.71s/it]
3%|▎ | 1419/43939 [2:18:26<67:17:20, 5.70s/it]
3%|▎ | 1420/43939 [2:18:32<67:09:59, 5.69s/it]
3%|▎ | 1421/43939 [2:18:37<67:11:41, 5.69s/it]
3%|▎ | 1422/43939 [2:18:43<67:46:30, 5.74s/it]
3%|▎ | 1423/43939 [2:18:49<67:31:29, 5.72s/it]
3%|▎ | 1424/43939 [2:18:55<67:21:13, 5.70s/it]
3%|▎ | 1425/43939 [2:19:00<67:30:32, 5.72s/it]
3%|▎ | 1426/43939 [2:19:06<67:26:45, 5.71s/it]
3%|▎ | 1427/43939 [2:19:12<67:41:47, 5.73s/it]
3%|▎ | 1428/43939 [2:19:17<67:32:38, 5.72s/it]
3%|▎ | 1429/43939 [2:19:23<67:28:31, 5.71s/it]
3%|▎ | 1430/43939 [2:19:29<67:26:56, 5.71s/it]
3%|▎ | 1431/43939 [2:19:35<67:29:01, 5.72s/it]
3%|▎ | 1432/43939 [2:19:40<67:16:45, 5.70s/it]
3%|▎ | 1433/43939 [2:19:46<67:12:30, 5.69s/it]
3%|▎ | 1434/43939 [2:19:52<67:33:34, 5.72s/it]
3%|▎ | 1435/43939 [2:19:57<67:31:29, 5.72s/it]
3%|▎ | 1436/43939 [2:20:03<67:29:55, 5.72s/it]
3%|▎ | 1437/43939 [2:20:09<67:34:06, 5.72s/it]
3%|▎ | 1438/43939 [2:20:15<67:28:00, 5.71s/it]
3%|▎ | 1439/43939 [2:20:20<67:35:04, 5.72s/it]
3%|▎ | 1440/43939 [2:20:26<67:30:50, 5.72s/it]
3%|▎ | 1441/43939 [2:20:32<67:18:44, 5.70s/it]
3%|▎ | 1442/43939 [2:20:37<67:25:11, 5.71s/it]
3%|▎ | 1443/43939 [2:20:43<68:04:33, 5.77s/it]
3%|▎ | 1444/43939 [2:20:49<67:45:21, 5.74s/it]
3%|▎ | 1445/43939 [2:20:55<67:42:07, 5.74s/it]
3%|▎ | 1446/43939 [2:21:00<67:36:24, 5.73s/it]
3%|▎ | 1447/43939 [2:21:06<67:31:02, 5.72s/it]
3%|▎ | 1448/43939 [2:21:12<67:31:23, 5.72s/it]
3%|▎ | 1449/43939 [2:21:18<67:23:29, 5.71s/it]
3%|▎ | 1450/43939 [2:21:23<67:31:15, 5.72s/it]
{'loss': 1.1767, 'grad_norm': 1.9961028099060059, 'learning_rate': 1.649214659685864e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 494.13, 'epoch': 0.13}
3%|▎ | 1450/43939 [2:21:23<67:31:15, 5.72s/it]
3%|▎ | 1451/43939 [2:21:29<67:33:44, 5.72s/it]
3%|▎ | 1452/43939 [2:21:35<67:39:48, 5.73s/it]
3%|▎ | 1453/43939 [2:21:41<67:47:49, 5.74s/it]
3%|▎ | 1454/43939 [2:21:46<67:35:57, 5.73s/it]
3%|▎ | 1455/43939 [2:21:52<67:30:11, 5.72s/it]
3%|▎ | 1456/43939 [2:21:58<67:22:00, 5.71s/it]
3%|▎ | 1457/43939 [2:22:03<67:26:04, 5.71s/it]
3%|▎ | 1458/43939 [2:22:09<67:19:48, 5.71s/it]
3%|▎ | 1459/43939 [2:22:15<67:10:01, 5.69s/it]
3%|▎ | 1460/43939 [2:22:20<67:08:56, 5.69s/it]
3%|▎ | 1461/43939 [2:22:26<67:04:17, 5.68s/it]
3%|▎ | 1462/43939 [2:22:32<67:15:07, 5.70s/it]
3%|▎ | 1463/43939 [2:22:38<68:02:17, 5.77s/it]
3%|▎ | 1464/43939 [2:22:43<67:49:46, 5.75s/it]
3%|▎ | 1465/43939 [2:22:49<67:39:13, 5.73s/it]
3%|▎ | 1466/43939 [2:22:55<67:30:09, 5.72s/it]
3%|▎ | 1467/43939 [2:23:00<67:17:59, 5.70s/it]
3%|▎ | 1468/43939 [2:23:06<67:16:19, 5.70s/it]
3%|▎ | 1469/43939 [2:23:12<67:09:11, 5.69s/it]
3%|▎ | 1470/43939 [2:23:17<67:07:06, 5.69s/it]
3%|▎ | 1471/43939 [2:23:23<67:00:42, 5.68s/it]
3%|▎ | 1472/43939 [2:23:29<67:04:38, 5.69s/it]
3%|▎ | 1473/43939 [2:23:35<67:01:28, 5.68s/it]
3%|▎ | 1474/43939 [2:23:40<67:04:03, 5.69s/it]
3%|▎ | 1475/43939 [2:23:46<67:08:48, 5.69s/it]
3%|▎ | 1476/43939 [2:23:52<67:05:37, 5.69s/it]
3%|▎ | 1477/43939 [2:23:57<67:07:08, 5.69s/it]
3%|▎ | 1478/43939 [2:24:03<67:13:24, 5.70s/it]
3%|▎ | 1479/43939 [2:24:09<67:20:57, 5.71s/it]
3%|▎ | 1480/43939 [2:24:14<67:23:43, 5.71s/it]
3%|▎ | 1481/43939 [2:24:20<67:22:04, 5.71s/it]
3%|▎ | 1482/43939 [2:24:26<67:35:55, 5.73s/it]
3%|▎ | 1483/43939 [2:24:32<67:29:22, 5.72s/it]
3%|▎ | 1484/43939 [2:24:38<68:11:39, 5.78s/it]
3%|▎ | 1485/43939 [2:24:43<67:51:55, 5.75s/it]
3%|▎ | 1486/43939 [2:24:49<67:41:19, 5.74s/it]
3%|▎ | 1487/43939 [2:24:55<67:41:24, 5.74s/it]
3%|▎ | 1488/43939 [2:25:00<67:34:16, 5.73s/it]
3%|▎ | 1489/43939 [2:25:06<67:20:19, 5.71s/it]
3%|▎ | 1490/43939 [2:25:12<67:21:58, 5.71s/it]
3%|▎ | 1491/43939 [2:25:18<67:21:37, 5.71s/it]
3%|▎ | 1492/43939 [2:25:23<67:17:24, 5.71s/it]
3%|▎ | 1493/43939 [2:25:29<67:09:25, 5.70s/it]
3%|▎ | 1494/43939 [2:25:35<67:09:22, 5.70s/it]
3%|▎ | 1495/43939 [2:25:40<67:12:50, 5.70s/it]
3%|▎ | 1496/43939 [2:25:46<67:20:32, 5.71s/it]
3%|▎ | 1497/43939 [2:25:52<67:17:31, 5.71s/it]
3%|▎ | 1498/43939 [2:25:57<67:14:20, 5.70s/it]
3%|▎ | 1499/43939 [2:26:03<67:25:59, 5.72s/it]
3%|▎ | 1500/43939 [2:26:09<67:25:13, 5.72s/it]
{'loss': 1.1622, 'grad_norm': 1.7714760303497314, 'learning_rate': 1.7061233781015252e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 482.39, 'epoch': 0.14}
3%|▎ | 1500/43939 [2:26:09<67:25:13, 5.72s/it]
3%|▎ | 1501/43939 [2:26:15<67:15:59, 5.71s/it]
3%|▎ | 1502/43939 [2:26:20<67:18:45, 5.71s/it]
3%|▎ | 1503/43939 [2:26:26<67:10:26, 5.70s/it]
3%|▎ | 1504/43939 [2:26:32<67:15:08, 5.71s/it]
3%|▎ | 1505/43939 [2:26:37<67:05:24, 5.69s/it]
3%|▎ | 1506/43939 [2:26:43<67:56:24, 5.76s/it]
3%|▎ | 1507/43939 [2:26:49<67:41:40, 5.74s/it]
3%|▎ | 1508/43939 [2:26:55<67:40:57, 5.74s/it]
3%|▎ | 1509/43939 [2:27:00<67:33:23, 5.73s/it]
3%|▎ | 1510/43939 [2:27:06<67:23:29, 5.72s/it]
3%|▎ | 1511/43939 [2:27:12<67:18:05, 5.71s/it]
3%|▎ | 1512/43939 [2:27:18<67:17:10, 5.71s/it]
3%|▎ | 1513/43939 [2:27:23<67:23:15, 5.72s/it]
3%|▎ | 1514/43939 [2:27:29<67:14:26, 5.71s/it]
3%|▎ | 1515/43939 [2:27:35<67:15:21, 5.71s/it]
3%|▎ | 1516/43939 [2:27:40<67:07:44, 5.70s/it]
3%|▎ | 1517/43939 [2:27:46<67:07:21, 5.70s/it]
3%|▎ | 1518/43939 [2:27:52<67:08:17, 5.70s/it]
3%|▎ | 1519/43939 [2:27:57<67:15:06, 5.71s/it]
3%|▎ | 1520/43939 [2:28:03<67:10:36, 5.70s/it]
3%|▎ | 1521/43939 [2:28:09<67:09:40, 5.70s/it]
3%|▎ | 1522/43939 [2:28:15<67:17:17, 5.71s/it]
3%|▎ | 1523/43939 [2:28:20<67:15:04, 5.71s/it]
3%|▎ | 1524/43939 [2:28:26<67:18:29, 5.71s/it]
3%|▎ | 1525/43939 [2:28:32<67:14:44, 5.71s/it]
3%|▎ | 1526/43939 [2:28:38<68:07:55, 5.78s/it]
3%|▎ | 1527/43939 [2:28:43<67:57:00, 5.77s/it]
3%|▎ | 1528/43939 [2:28:49<67:37:53, 5.74s/it]
3%|▎ | 1529/43939 [2:28:55<67:36:07, 5.74s/it]
3%|▎ | 1530/43939 [2:29:00<67:25:40, 5.72s/it]
3%|▎ | 1531/43939 [2:29:06<67:24:13, 5.72s/it]
3%|▎ | 1532/43939 [2:29:12<67:14:21, 5.71s/it]
3%|▎ | 1533/43939 [2:29:18<67:10:51, 5.70s/it]
3%|▎ | 1534/43939 [2:29:23<67:03:43, 5.69s/it]
3%|▎ | 1535/43939 [2:29:29<67:04:39, 5.69s/it]
3%|▎ | 1536/43939 [2:29:35<67:16:03, 5.71s/it]
3%|▎ | 1537/43939 [2:29:40<67:16:19, 5.71s/it]
4%|▎ | 1538/43939 [2:29:46<67:21:13, 5.72s/it]
4%|▎ | 1539/43939 [2:29:52<67:17:07, 5.71s/it]
4%|▎ | 1540/43939 [2:29:58<67:13:14, 5.71s/it]
4%|▎ | 1541/43939 [2:30:03<67:14:45, 5.71s/it]
4%|▎ | 1542/43939 [2:30:09<67:16:54, 5.71s/it]
4%|▎ | 1543/43939 [2:30:15<67:14:09, 5.71s/it]
4%|▎ | 1544/43939 [2:30:20<67:14:43, 5.71s/it]
4%|▎ | 1545/43939 [2:30:26<67:10:12, 5.70s/it]
4%|▎ | 1546/43939 [2:30:32<67:05:31, 5.70s/it]
4%|▎ | 1547/43939 [2:30:38<67:56:40, 5.77s/it]
4%|▎ | 1548/43939 [2:30:43<67:50:06, 5.76s/it]
4%|▎ | 1549/43939 [2:30:49<67:42:24, 5.75s/it]
4%|▎ | 1550/43939 [2:30:55<67:42:30, 5.75s/it]
{'loss': 1.1754, 'grad_norm': 1.944865345954895, 'learning_rate': 1.7630320965171866e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 482.41, 'epoch': 0.14}
4%|▎ | 1550/43939 [2:30:55<67:42:30, 5.75s/it]
4%|▎ | 1551/43939 [2:31:01<67:39:38, 5.75s/it]
4%|▎ | 1552/43939 [2:31:06<67:33:27, 5.74s/it]
4%|▎ | 1553/43939 [2:31:12<67:30:26, 5.73s/it]
4%|▎ | 1554/43939 [2:31:18<67:23:33, 5.72s/it]
4%|▎ | 1555/43939 [2:31:24<67:28:24, 5.73s/it]
4%|▎ | 1556/43939 [2:31:29<67:30:54, 5.73s/it]
4%|▎ | 1557/43939 [2:31:35<67:20:06, 5.72s/it]
4%|▎ | 1558/43939 [2:31:41<67:19:33, 5.72s/it]
4%|▎ | 1559/43939 [2:31:46<67:15:22, 5.71s/it]
4%|▎ | 1560/43939 [2:31:52<67:13:53, 5.71s/it]
4%|▎ | 1561/43939 [2:31:58<67:11:11, 5.71s/it]
4%|▎ | 1562/43939 [2:32:04<67:20:20, 5.72s/it]
4%|▎ | 1563/43939 [2:32:09<67:16:36, 5.72s/it]
4%|▎ | 1564/43939 [2:32:15<67:11:41, 5.71s/it]
4%|▎ | 1565/43939 [2:32:21<67:10:38, 5.71s/it]
4%|▎ | 1566/43939 [2:32:26<67:04:44, 5.70s/it]
4%|▎ | 1567/43939 [2:32:32<67:07:47, 5.70s/it]
4%|▎ | 1568/43939 [2:32:38<67:10:37, 5.71s/it]
4%|▎ | 1569/43939 [2:32:44<67:54:56, 5.77s/it]
4%|▎ | 1570/43939 [2:32:49<67:40:59, 5.75s/it]
4%|▎ | 1571/43939 [2:32:55<67:28:03, 5.73s/it]
4%|▎ | 1572/43939 [2:33:01<67:25:19, 5.73s/it]
4%|▎ | 1573/43939 [2:33:06<67:11:07, 5.71s/it]
4%|▎ | 1574/43939 [2:33:12<67:06:53, 5.70s/it]
4%|▎ | 1575/43939 [2:33:18<67:04:30, 5.70s/it]
4%|▎ | 1576/43939 [2:33:24<67:10:29, 5.71s/it]
4%|▎ | 1577/43939 [2:33:29<67:11:07, 5.71s/it]
4%|▎ | 1578/43939 [2:33:35<67:08:51, 5.71s/it]
4%|▎ | 1579/43939 [2:33:41<67:02:43, 5.70s/it]
4%|▎ | 1580/43939 [2:33:46<66:56:43, 5.69s/it]
4%|▎ | 1581/43939 [2:33:52<66:55:56, 5.69s/it]
4%|▎ | 1582/43939 [2:33:58<66:55:40, 5.69s/it]
4%|▎ | 1583/43939 [2:34:03<66:56:12, 5.69s/it]
4%|▎ | 1584/43939 [2:34:09<66:57:45, 5.69s/it]
4%|▎ | 1585/43939 [2:34:15<66:54:11, 5.69s/it]
4%|▎ | 1586/43939 [2:34:20<67:09:15, 5.71s/it]
4%|▎ | 1587/43939 [2:34:26<66:56:53, 5.69s/it]
4%|▎ | 1588/43939 [2:34:32<66:46:39, 5.68s/it]
4%|▎ | 1589/43939 [2:34:38<67:45:20, 5.76s/it]
4%|▎ | 1590/43939 [2:34:43<67:44:05, 5.76s/it]
4%|▎ | 1591/43939 [2:34:49<67:26:36, 5.73s/it]
4%|▎ | 1592/43939 [2:34:55<67:14:50, 5.72s/it]
4%|▎ | 1593/43939 [2:35:01<67:10:52, 5.71s/it]
4%|▎ | 1594/43939 [2:35:06<67:17:30, 5.72s/it]
4%|▎ | 1595/43939 [2:35:12<67:14:55, 5.72s/it]
4%|▎ | 1596/43939 [2:35:18<67:01:38, 5.70s/it]
4%|▎ | 1597/43939 [2:35:23<66:58:22, 5.69s/it]
4%|▎ | 1598/43939 [2:35:29<66:48:43, 5.68s/it]
4%|▎ | 1599/43939 [2:35:35<66:49:47, 5.68s/it]
4%|▎ | 1600/43939 [2:35:40<66:55:49, 5.69s/it]
{'loss': 1.1597, 'grad_norm': 1.8010004758834839, 'learning_rate': 1.819940814932848e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 477.14, 'epoch': 0.15}
4%|▎ | 1600/43939 [2:35:40<66:55:49, 5.69s/it]
4%|▎ | 1601/43939 [2:35:46<66:51:06, 5.68s/it]
4%|▎ | 1602/43939 [2:35:52<66:55:31, 5.69s/it]
4%|▎ | 1603/43939 [2:35:57<66:56:07, 5.69s/it]
4%|▎ | 1604/43939 [2:36:03<66:54:32, 5.69s/it]
4%|▎ | 1605/43939 [2:36:09<66:57:04, 5.69s/it]
4%|▎ | 1606/43939 [2:36:15<66:54:28, 5.69s/it]
4%|▎ | 1607/43939 [2:36:20<66:57:20, 5.69s/it]
4%|▎ | 1608/43939 [2:36:26<66:58:14, 5.70s/it]
4%|▎ | 1609/43939 [2:36:32<67:00:52, 5.70s/it]
4%|▎ | 1610/43939 [2:36:38<67:56:27, 5.78s/it]
4%|▎ | 1611/43939 [2:36:43<67:44:23, 5.76s/it]
4%|▎ | 1612/43939 [2:36:49<67:40:20, 5.76s/it]
4%|▎ | 1613/43939 [2:36:55<67:31:13, 5.74s/it]
4%|▎ | 1614/43939 [2:37:00<67:21:29, 5.73s/it]
4%|▎ | 1615/43939 [2:37:06<67:19:06, 5.73s/it]
4%|▎ | 1616/43939 [2:37:12<67:08:33, 5.71s/it]
4%|▎ | 1617/43939 [2:37:18<67:09:21, 5.71s/it]
4%|▎ | 1618/43939 [2:37:23<67:06:17, 5.71s/it]
4%|▎ | 1619/43939 [2:37:29<67:03:10, 5.70s/it]
4%|▎ | 1620/43939 [2:37:35<66:50:59, 5.69s/it]
4%|▎ | 1621/43939 [2:37:40<66:55:57, 5.69s/it]
4%|▎ | 1622/43939 [2:37:46<66:50:59, 5.69s/it]
4%|▎ | 1623/43939 [2:37:52<66:55:15, 5.69s/it]
4%|▎ | 1624/43939 [2:37:57<66:57:18, 5.70s/it]
4%|▎ | 1625/43939 [2:38:03<67:05:30, 5.71s/it]
4%|▎ | 1626/43939 [2:38:09<67:03:39, 5.71s/it]
4%|▎ | 1627/43939 [2:38:15<66:56:39, 5.70s/it]
4%|▎ | 1628/43939 [2:38:20<67:07:29, 5.71s/it]
4%|▎ | 1629/43939 [2:38:26<67:06:46, 5.71s/it]
4%|▎ | 1630/43939 [2:38:32<67:07:00, 5.71s/it]
4%|▎ | 1631/43939 [2:38:37<67:10:13, 5.72s/it]
4%|▎ | 1632/43939 [2:38:43<67:37:28, 5.75s/it]
4%|▎ | 1633/43939 [2:38:49<67:26:11, 5.74s/it]
4%|▎ | 1634/43939 [2:38:55<67:08:56, 5.71s/it]
4%|▎ | 1635/43939 [2:39:00<66:59:37, 5.70s/it]
4%|▎ | 1636/43939 [2:39:06<67:00:03, 5.70s/it]
4%|▎ | 1637/43939 [2:39:12<66:58:25, 5.70s/it]
4%|▎ | 1638/43939 [2:39:17<66:56:54, 5.70s/it]
4%|▎ | 1639/43939 [2:39:23<66:58:54, 5.70s/it]
4%|▎ | 1640/43939 [2:39:29<67:13:44, 5.72s/it]
4%|▎ | 1641/43939 [2:39:35<67:11:15, 5.72s/it]
4%|▎ | 1642/43939 [2:39:40<67:04:34, 5.71s/it]
4%|▎ | 1643/43939 [2:39:46<67:02:24, 5.71s/it]
4%|▎ | 1644/43939 [2:39:52<67:03:26, 5.71s/it]
4%|▎ | 1645/43939 [2:39:57<66:58:02, 5.70s/it]
4%|▎ | 1646/43939 [2:40:03<66:51:39, 5.69s/it]
4%|▎ | 1647/43939 [2:40:09<66:52:32, 5.69s/it]
4%|▍ | 1648/43939 [2:40:14<66:54:27, 5.70s/it]
4%|▍ | 1649/43939 [2:40:20<66:59:45, 5.70s/it]
4%|▍ | 1650/43939 [2:40:26<66:57:46, 5.70s/it]
{'loss': 1.1533, 'grad_norm': 2.0305533409118652, 'learning_rate': 1.8768495333485092e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 484.22, 'epoch': 0.15}
4%|▍ | 1650/43939 [2:40:26<66:57:46, 5.70s/it]
4%|▍ | 1651/43939 [2:40:32<66:57:59, 5.70s/it]
4%|▍ | 1652/43939 [2:40:37<67:20:31, 5.73s/it]
4%|▍ | 1653/43939 [2:40:43<68:16:19, 5.81s/it]
4%|▍ | 1654/43939 [2:40:49<67:50:47, 5.78s/it]
4%|▍ | 1655/43939 [2:40:55<67:35:28, 5.75s/it]
4%|▍ | 1656/43939 [2:41:00<67:13:27, 5.72s/it]
4%|▍ | 1657/43939 [2:41:06<67:04:38, 5.71s/it]
4%|▍ | 1658/43939 [2:41:12<67:10:09, 5.72s/it]
4%|▍ | 1659/43939 [2:41:18<67:10:52, 5.72s/it]
4%|▍ | 1660/43939 [2:41:23<67:03:47, 5.71s/it]
4%|▍ | 1661/43939 [2:41:29<67:01:42, 5.71s/it]
4%|▍ | 1662/43939 [2:41:35<67:05:21, 5.71s/it]
4%|▍ | 1663/43939 [2:41:40<66:54:28, 5.70s/it]
4%|▍ | 1664/43939 [2:41:46<66:57:17, 5.70s/it]
4%|▍ | 1665/43939 [2:41:52<66:57:37, 5.70s/it]
4%|▍ | 1666/43939 [2:41:57<67:00:18, 5.71s/it]
4%|▍ | 1667/43939 [2:42:03<67:00:22, 5.71s/it]
4%|▍ | 1668/43939 [2:42:09<67:03:53, 5.71s/it]
4%|▍ | 1669/43939 [2:42:15<67:01:46, 5.71s/it]
4%|▍ | 1670/43939 [2:42:20<66:59:49, 5.71s/it]
4%|▍ | 1671/43939 [2:42:26<66:53:45, 5.70s/it]
4%|▍ | 1672/43939 [2:42:32<66:43:07, 5.68s/it]
4%|▍ | 1673/43939 [2:42:37<67:27:48, 5.75s/it]
4%|▍ | 1674/43939 [2:42:43<67:24:31, 5.74s/it]
4%|▍ | 1675/43939 [2:42:49<67:15:41, 5.73s/it]
4%|▍ | 1676/43939 [2:42:55<67:07:23, 5.72s/it]
4%|▍ | 1677/43939 [2:43:00<66:56:23, 5.70s/it]
4%|▍ | 1678/43939 [2:43:06<67:08:29, 5.72s/it]
4%|▍ | 1679/43939 [2:43:12<67:10:54, 5.72s/it]
4%|▍ | 1680/43939 [2:43:17<67:00:21, 5.71s/it]
4%|▍ | 1681/43939 [2:43:23<67:03:06, 5.71s/it]
4%|▍ | 1682/43939 [2:43:29<67:06:54, 5.72s/it]
4%|▍ | 1683/43939 [2:43:35<67:06:33, 5.72s/it]
4%|▍ | 1684/43939 [2:43:40<67:00:19, 5.71s/it]
4%|▍ | 1685/43939 [2:43:46<67:14:43, 5.73s/it]
4%|▍ | 1686/43939 [2:43:52<67:06:18, 5.72s/it]
4%|▍ | 1687/43939 [2:43:58<67:17:18, 5.73s/it]
4%|▍ | 1688/43939 [2:44:03<67:18:50, 5.74s/it]
4%|▍ | 1689/43939 [2:44:09<67:18:21, 5.73s/it]
4%|▍ | 1690/43939 [2:44:15<67:16:29, 5.73s/it]
4%|▍ | 1691/43939 [2:44:20<67:03:51, 5.71s/it]
4%|▍ | 1692/43939 [2:44:26<67:01:37, 5.71s/it]
4%|▍ | 1693/43939 [2:44:32<66:55:56, 5.70s/it]
4%|▍ | 1694/43939 [2:44:38<66:57:27, 5.71s/it]
4%|▍ | 1695/43939 [2:44:43<67:44:00, 5.77s/it]
4%|▍ | 1696/43939 [2:44:49<67:28:13, 5.75s/it]
4%|▍ | 1697/43939 [2:44:55<67:17:24, 5.73s/it]
4%|▍ | 1698/43939 [2:45:00<66:56:12, 5.70s/it]
4%|▍ | 1699/43939 [2:45:06<67:04:20, 5.72s/it]
4%|▍ | 1700/43939 [2:45:12<67:11:25, 5.73s/it]
{'loss': 1.1548, 'grad_norm': 1.8290001153945923, 'learning_rate': 1.9337582517641702e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 492.04, 'epoch': 0.15}
4%|▍ | 1700/43939 [2:45:12<67:11:25, 5.73s/it]
4%|▍ | 1701/43939 [2:45:18<67:05:16, 5.72s/it]
4%|▍ | 1702/43939 [2:45:23<67:01:48, 5.71s/it]
4%|▍ | 1703/43939 [2:45:29<67:00:18, 5.71s/it]
4%|▍ | 1704/43939 [2:45:35<66:56:04, 5.71s/it]
4%|▍ | 1705/43939 [2:45:40<66:58:42, 5.71s/it]
4%|▍ | 1706/43939 [2:45:46<66:55:34, 5.70s/it]
4%|▍ | 1707/43939 [2:45:52<67:02:06, 5.71s/it]
4%|▍ | 1708/43939 [2:45:58<67:08:36, 5.72s/it]
4%|▍ | 1709/43939 [2:46:03<67:12:14, 5.73s/it]
4%|▍ | 1710/43939 [2:46:09<67:08:28, 5.72s/it]
4%|▍ | 1711/43939 [2:46:15<67:08:43, 5.72s/it]
4%|▍ | 1712/43939 [2:46:21<67:07:40, 5.72s/it]
4%|▍ | 1713/43939 [2:46:26<66:59:43, 5.71s/it]
4%|▍ | 1714/43939 [2:46:32<67:17:00, 5.74s/it]
4%|▍ | 1715/43939 [2:46:38<67:10:12, 5.73s/it]
4%|▍ | 1716/43939 [2:46:44<67:38:19, 5.77s/it]
4%|▍ | 1717/43939 [2:46:49<67:24:12, 5.75s/it]
4%|▍ | 1718/43939 [2:46:55<67:19:44, 5.74s/it]
4%|▍ | 1719/43939 [2:47:01<67:09:42, 5.73s/it]
4%|▍ | 1720/43939 [2:47:06<67:04:26, 5.72s/it]
4%|▍ | 1721/43939 [2:47:12<67:01:19, 5.72s/it]
4%|▍ | 1722/43939 [2:47:18<67:00:34, 5.71s/it]
4%|▍ | 1723/43939 [2:47:24<66:57:18, 5.71s/it]
4%|▍ | 1724/43939 [2:47:29<66:59:09, 5.71s/it]
4%|▍ | 1725/43939 [2:47:35<66:53:39, 5.70s/it]
4%|▍ | 1726/43939 [2:47:41<66:50:40, 5.70s/it]
4%|▍ | 1727/43939 [2:47:46<66:54:53, 5.71s/it]
4%|▍ | 1728/43939 [2:47:52<66:52:41, 5.70s/it]
4%|▍ | 1729/43939 [2:47:58<66:59:26, 5.71s/it]
4%|▍ | 1730/43939 [2:48:03<67:00:03, 5.71s/it]
4%|▍ | 1731/43939 [2:48:09<66:57:37, 5.71s/it]
4%|▍ | 1732/43939 [2:48:15<66:49:40, 5.70s/it]
4%|▍ | 1733/43939 [2:48:21<66:47:58, 5.70s/it]
4%|▍ | 1734/43939 [2:48:26<66:48:41, 5.70s/it]
4%|▍ | 1735/43939 [2:48:32<66:49:56, 5.70s/it]
4%|▍ | 1736/43939 [2:48:38<67:48:28, 5.78s/it]
4%|▍ | 1737/43939 [2:48:44<67:33:52, 5.76s/it]
4%|▍ | 1738/43939 [2:48:49<67:23:40, 5.75s/it]
4%|▍ | 1739/43939 [2:48:55<67:10:12, 5.73s/it]
4%|▍ | 1740/43939 [2:49:01<67:14:40, 5.74s/it]
4%|▍ | 1741/43939 [2:49:07<67:05:14, 5.72s/it]
4%|▍ | 1742/43939 [2:49:12<66:56:41, 5.71s/it]
4%|▍ | 1743/43939 [2:49:18<66:53:15, 5.71s/it]
4%|▍ | 1744/43939 [2:49:24<66:48:50, 5.70s/it]
4%|▍ | 1745/43939 [2:49:29<66:56:49, 5.71s/it]
4%|▍ | 1746/43939 [2:49:35<66:57:17, 5.71s/it]
4%|▍ | 1747/43939 [2:49:41<66:50:39, 5.70s/it]
4%|▍ | 1748/43939 [2:49:46<66:53:05, 5.71s/it]
4%|▍ | 1749/43939 [2:49:52<66:46:01, 5.70s/it]
4%|▍ | 1750/43939 [2:49:58<66:51:05, 5.70s/it]
{'loss': 1.1421, 'grad_norm': 1.9180041551589966, 'learning_rate': 1.9906669701798315e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 446.31, 'epoch': 0.16}
4%|▍ | 1750/43939 [2:49:58<66:51:05, 5.70s/it]
4%|▍ | 1751/43939 [2:50:03<66:37:27, 5.69s/it]
4%|▍ | 1752/43939 [2:50:09<66:39:45, 5.69s/it]
4%|▍ | 1753/43939 [2:50:15<66:40:44, 5.69s/it]
4%|▍ | 1754/43939 [2:50:21<66:41:00, 5.69s/it]
4%|▍ | 1755/43939 [2:50:26<66:54:47, 5.71s/it]
4%|▍ | 1756/43939 [2:50:32<66:59:55, 5.72s/it]
4%|▍ | 1757/43939 [2:50:38<67:40:04, 5.78s/it]
4%|▍ | 1758/43939 [2:50:44<67:34:01, 5.77s/it]
4%|▍ | 1759/43939 [2:50:49<67:31:44, 5.76s/it]
4%|▍ | 1760/43939 [2:50:55<67:12:25, 5.74s/it]
4%|▍ | 1761/43939 [2:51:01<66:59:56, 5.72s/it]
4%|▍ | 1762/43939 [2:51:06<66:49:49, 5.70s/it]
4%|▍ | 1763/43939 [2:51:12<66:39:56, 5.69s/it]
4%|▍ | 1764/43939 [2:51:18<66:41:56, 5.69s/it]
4%|▍ | 1765/43939 [2:51:24<66:43:09, 5.70s/it]
4%|▍ | 1766/43939 [2:51:29<66:41:14, 5.69s/it]
4%|▍ | 1767/43939 [2:51:35<66:42:29, 5.69s/it]
4%|▍ | 1768/43939 [2:51:41<66:41:18, 5.69s/it]
4%|▍ | 1769/43939 [2:51:46<66:43:47, 5.70s/it]
4%|▍ | 1770/43939 [2:51:52<66:41:31, 5.69s/it]
4%|▍ | 1771/43939 [2:51:58<66:36:49, 5.69s/it]
4%|▍ | 1772/43939 [2:52:03<66:36:26, 5.69s/it]
4%|▍ | 1773/43939 [2:52:09<66:32:15, 5.68s/it]
4%|▍ | 1774/43939 [2:52:15<66:48:39, 5.70s/it]
4%|▍ | 1775/43939 [2:52:20<66:50:52, 5.71s/it]
4%|▍ | 1776/43939 [2:52:26<66:53:39, 5.71s/it]
4%|▍ | 1777/43939 [2:52:32<66:53:08, 5.71s/it]
4%|▍ | 1778/43939 [2:52:38<66:55:57, 5.72s/it]
4%|▍ | 1779/43939 [2:52:44<67:38:50, 5.78s/it]
4%|▍ | 1780/43939 [2:52:49<67:13:42, 5.74s/it]
4%|▍ | 1781/43939 [2:52:55<67:05:42, 5.73s/it]
4%|▍ | 1782/43939 [2:53:01<67:10:26, 5.74s/it]
4%|▍ | 1783/43939 [2:53:06<66:58:38, 5.72s/it]
4%|▍ | 1784/43939 [2:53:12<66:58:31, 5.72s/it]
4%|▍ | 1785/43939 [2:53:18<66:55:55, 5.72s/it]
4%|▍ | 1786/43939 [2:53:23<66:50:29, 5.71s/it]
4%|▍ | 1787/43939 [2:53:29<66:40:11, 5.69s/it]
4%|▍ | 1788/43939 [2:53:35<66:50:53, 5.71s/it]
4%|▍ | 1789/43939 [2:53:41<66:52:17, 5.71s/it]
4%|▍ | 1790/43939 [2:53:46<66:53:16, 5.71s/it]
4%|▍ | 1791/43939 [2:53:52<66:54:07, 5.71s/it]
4%|▍ | 1792/43939 [2:53:58<66:49:21, 5.71s/it]
4%|▍ | 1793/43939 [2:54:03<66:44:28, 5.70s/it]
4%|▍ | 1794/43939 [2:54:09<66:43:18, 5.70s/it]
4%|▍ | 1795/43939 [2:54:15<66:50:28, 5.71s/it]
4%|▍ | 1796/43939 [2:54:21<66:48:00, 5.71s/it]
4%|▍ | 1797/43939 [2:54:26<66:47:13, 5.71s/it]
4%|▍ | 1798/43939 [2:54:32<66:42:22, 5.70s/it]
4%|▍ | 1799/43939 [2:54:38<67:32:32, 5.77s/it]
4%|▍ | 1800/43939 [2:54:44<67:15:50, 5.75s/it]
{'loss': 1.1166, 'grad_norm': 1.6121618747711182, 'learning_rate': 2.047575688595493e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 454.94, 'epoch': 0.16}
4%|▍ | 1800/43939 [2:54:44<67:15:50, 5.75s/it]
4%|▍ | 1801/43939 [2:54:49<67:12:28, 5.74s/it]
4%|▍ | 1802/43939 [2:54:55<67:17:03, 5.75s/it]
4%|▍ | 1803/43939 [2:55:01<67:12:50, 5.74s/it]
4%|▍ | 1804/43939 [2:55:06<67:05:52, 5.73s/it]
4%|▍ | 1805/43939 [2:55:12<67:04:51, 5.73s/it]
4%|▍ | 1806/43939 [2:55:18<66:52:46, 5.71s/it]
4%|▍ | 1807/43939 [2:55:24<66:51:01, 5.71s/it]
4%|▍ | 1808/43939 [2:55:29<66:40:57, 5.70s/it]
4%|▍ | 1809/43939 [2:55:35<66:42:01, 5.70s/it]
4%|▍ | 1810/43939 [2:55:41<66:52:06, 5.71s/it]
4%|▍ | 1811/43939 [2:55:46<66:46:23, 5.71s/it]
4%|▍ | 1812/43939 [2:55:52<66:48:24, 5.71s/it]
4%|▍ | 1813/43939 [2:55:58<66:42:53, 5.70s/it]
4%|▍ | 1814/43939 [2:56:04<66:45:45, 5.71s/it]
4%|▍ | 1815/43939 [2:56:09<66:45:10, 5.70s/it]
4%|▍ | 1816/43939 [2:56:15<66:50:56, 5.71s/it]
4%|▍ | 1817/43939 [2:56:21<66:44:32, 5.70s/it]
4%|▍ | 1818/43939 [2:56:26<66:39:23, 5.70s/it]
4%|▍ | 1819/43939 [2:56:32<66:39:58, 5.70s/it]
4%|▍ | 1820/43939 [2:56:38<67:37:50, 5.78s/it]
4%|▍ | 1821/43939 [2:56:44<67:18:14, 5.75s/it]
4%|▍ | 1822/43939 [2:56:49<67:14:07, 5.75s/it]
4%|▍ | 1823/43939 [2:56:55<67:05:00, 5.73s/it]
4%|▍ | 1824/43939 [2:57:01<67:02:18, 5.73s/it]
4%|▍ | 1825/43939 [2:57:07<66:58:30, 5.73s/it]
4%|▍ | 1826/43939 [2:57:12<67:07:01, 5.74s/it]
4%|▍ | 1827/43939 [2:57:18<67:03:45, 5.73s/it]
4%|▍ | 1828/43939 [2:57:24<66:49:37, 5.71s/it]
4%|▍ | 1829/43939 [2:57:29<66:45:17, 5.71s/it]
4%|▍ | 1830/43939 [2:57:35<66:47:09, 5.71s/it]
4%|▍ | 1831/43939 [2:57:41<66:53:51, 5.72s/it]
4%|▍ | 1832/43939 [2:57:47<66:59:16, 5.73s/it]
4%|▍ | 1833/43939 [2:57:52<66:56:56, 5.72s/it]
4%|▍ | 1834/43939 [2:57:58<66:53:04, 5.72s/it]
4%|▍ | 1835/43939 [2:58:04<66:52:25, 5.72s/it]
4%|▍ | 1836/43939 [2:58:09<66:39:00, 5.70s/it]
4%|▍ | 1837/43939 [2:58:15<66:33:46, 5.69s/it]
4%|▍ | 1838/43939 [2:58:21<66:34:29, 5.69s/it]
4%|▍ | 1839/43939 [2:58:26<66:35:56, 5.69s/it]
4%|▍ | 1840/43939 [2:58:32<66:29:10, 5.69s/it]
4%|▍ | 1841/43939 [2:58:38<66:29:17, 5.69s/it]
4%|▍ | 1842/43939 [2:58:44<67:25:18, 5.77s/it]
4%|▍ | 1843/43939 [2:58:49<67:03:00, 5.73s/it]
4%|▍ | 1844/43939 [2:58:55<66:52:10, 5.72s/it]
4%|▍ | 1845/43939 [2:59:01<66:39:57, 5.70s/it]
4%|▍ | 1846/43939 [2:59:06<66:32:20, 5.69s/it]
4%|▍ | 1847/43939 [2:59:12<66:43:36, 5.71s/it]
4%|▍ | 1848/43939 [2:59:18<66:38:18, 5.70s/it]
4%|▍ | 1849/43939 [2:59:24<66:35:29, 5.70s/it]
4%|▍ | 1850/43939 [2:59:29<66:36:01, 5.70s/it]
{'loss': 1.1582, 'grad_norm': 1.5404553413391113, 'learning_rate': 2.1044844070111542e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 507.31, 'epoch': 0.17}
4%|▍ | 1850/43939 [2:59:29<66:36:01, 5.70s/it]
4%|▍ | 1851/43939 [2:59:35<66:31:27, 5.69s/it]
4%|▍ | 1852/43939 [2:59:41<66:33:47, 5.69s/it]
4%|▍ | 1853/43939 [2:59:46<66:34:50, 5.70s/it]
4%|▍ | 1854/43939 [2:59:52<66:36:41, 5.70s/it]
4%|▍ | 1855/43939 [2:59:58<66:43:40, 5.71s/it]
4%|▍ | 1856/43939 [3:00:03<66:38:43, 5.70s/it]
4%|▍ | 1857/43939 [3:00:09<66:56:38, 5.73s/it]
4%|▍ | 1858/43939 [3:00:15<66:42:30, 5.71s/it]
4%|▍ | 1859/43939 [3:00:21<66:35:46, 5.70s/it]
4%|▍ | 1860/43939 [3:00:26<66:35:03, 5.70s/it]
4%|▍ | 1861/43939 [3:00:32<66:33:16, 5.69s/it]
4%|▍ | 1862/43939 [3:00:38<67:27:46, 5.77s/it]
4%|▍ | 1863/43939 [3:00:44<67:13:02, 5.75s/it]
4%|▍ | 1864/43939 [3:00:49<67:09:35, 5.75s/it]
4%|▍ | 1865/43939 [3:00:55<67:11:18, 5.75s/it]
4%|▍ | 1866/43939 [3:01:01<66:58:16, 5.73s/it]
4%|▍ | 1867/43939 [3:01:07<66:54:43, 5.73s/it]
4%|▍ | 1868/43939 [3:01:12<66:48:58, 5.72s/it]
4%|▍ | 1869/43939 [3:01:18<66:48:02, 5.72s/it]
4%|▍ | 1870/43939 [3:01:24<66:43:35, 5.71s/it]
4%|▍ | 1871/43939 [3:01:29<66:48:42, 5.72s/it]
4%|▍ | 1872/43939 [3:01:35<66:53:25, 5.72s/it]
4%|▍ | 1873/43939 [3:01:41<66:49:50, 5.72s/it]
4%|▍ | 1874/43939 [3:01:46<66:46:35, 5.71s/it]
4%|▍ | 1875/43939 [3:01:52<66:35:25, 5.70s/it]
4%|▍ | 1876/43939 [3:01:58<66:25:43, 5.69s/it]
4%|▍ | 1877/43939 [3:02:04<66:30:53, 5.69s/it]
4%|▍ | 1878/43939 [3:02:09<66:28:15, 5.69s/it]
4%|▍ | 1879/43939 [3:02:15<66:27:54, 5.69s/it]
4%|▍ | 1880/43939 [3:02:21<66:28:52, 5.69s/it]
4%|▍ | 1881/43939 [3:02:26<66:25:17, 5.69s/it]
4%|▍ | 1882/43939 [3:02:32<66:34:42, 5.70s/it]
4%|▍ | 1883/43939 [3:02:38<67:27:33, 5.77s/it]
4%|▍ | 1884/43939 [3:02:44<67:00:36, 5.74s/it]
4%|▍ | 1885/43939 [3:02:49<66:48:45, 5.72s/it]
4%|▍ | 1886/43939 [3:02:55<66:38:07, 5.70s/it]
4%|▍ | 1887/43939 [3:03:01<66:34:16, 5.70s/it]
4%|▍ | 1888/43939 [3:03:06<66:42:54, 5.71s/it]
4%|▍ | 1889/43939 [3:03:12<66:39:43, 5.71s/it]
4%|▍ | 1890/43939 [3:03:18<66:44:20, 5.71s/it]
4%|▍ | 1891/43939 [3:03:23<66:38:10, 5.71s/it]
4%|▍ | 1892/43939 [3:03:29<66:36:59, 5.70s/it]
4%|▍ | 1893/43939 [3:03:35<66:41:24, 5.71s/it]
4%|▍ | 1894/43939 [3:03:41<66:42:26, 5.71s/it]
4%|▍ | 1895/43939 [3:03:46<66:39:20, 5.71s/it]
4%|▍ | 1896/43939 [3:03:52<66:36:26, 5.70s/it]
4%|▍ | 1897/43939 [3:03:58<66:42:12, 5.71s/it]
4%|▍ | 1898/43939 [3:04:03<66:31:24, 5.70s/it]
4%|▍ | 1899/43939 [3:04:09<66:44:27, 5.72s/it]
4%|▍ | 1900/43939 [3:04:15<66:36:17, 5.70s/it]
{'loss': 1.1591, 'grad_norm': 1.6776846647262573, 'learning_rate': 2.1613931254268155e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 503.53, 'epoch': 0.17}
4%|▍ | 1900/43939 [3:04:15<66:36:17, 5.70s/it]
4%|▍ | 1901/43939 [3:04:21<66:38:24, 5.71s/it]
4%|▍ | 1902/43939 [3:04:26<66:39:37, 5.71s/it]
4%|▍ | 1903/43939 [3:04:32<66:28:34, 5.69s/it]
4%|▍ | 1904/43939 [3:04:38<66:24:54, 5.69s/it]
4%|▍ | 1905/43939 [3:04:44<67:10:27, 5.75s/it]
4%|▍ | 1906/43939 [3:04:49<67:05:15, 5.75s/it]
4%|▍ | 1907/43939 [3:04:55<66:54:24, 5.73s/it]
4%|▍ | 1908/43939 [3:05:01<66:50:07, 5.72s/it]
4%|▍ | 1909/43939 [3:05:06<66:43:52, 5.72s/it]
4%|▍ | 1910/43939 [3:05:12<66:35:48, 5.70s/it]
4%|▍ | 1911/43939 [3:05:18<66:36:18, 5.71s/it]
4%|▍ | 1912/43939 [3:05:23<66:38:04, 5.71s/it]
4%|▍ | 1913/43939 [3:05:29<66:31:15, 5.70s/it]
4%|▍ | 1914/43939 [3:05:35<66:33:41, 5.70s/it]
4%|▍ | 1915/43939 [3:05:41<66:36:26, 5.71s/it]
4%|▍ | 1916/43939 [3:05:46<66:33:18, 5.70s/it]
4%|▍ | 1917/43939 [3:05:52<66:37:49, 5.71s/it]
4%|▍ | 1918/43939 [3:05:58<66:35:38, 5.71s/it]
4%|▍ | 1919/43939 [3:06:03<66:33:03, 5.70s/it]
4%|▍ | 1920/43939 [3:06:09<66:36:38, 5.71s/it]
4%|▍ | 1921/43939 [3:06:15<66:26:13, 5.69s/it]
4%|▍ | 1922/43939 [3:06:20<66:34:54, 5.70s/it]
4%|▍ | 1923/43939 [3:06:26<66:31:27, 5.70s/it]
4%|▍ | 1924/43939 [3:06:32<66:24:25, 5.69s/it]
4%|▍ | 1925/43939 [3:06:38<66:33:31, 5.70s/it]
4%|▍ | 1926/43939 [3:06:43<67:18:47, 5.77s/it]
4%|▍ | 1927/43939 [3:06:49<67:12:56, 5.76s/it]
4%|▍ | 1928/43939 [3:06:55<67:07:39, 5.75s/it]
4%|▍ | 1929/43939 [3:07:01<66:51:26, 5.73s/it]
4%|▍ | 1930/43939 [3:07:06<66:42:47, 5.72s/it]
4%|▍ | 1931/43939 [3:07:12<66:38:46, 5.71s/it]
4%|▍ | 1932/43939 [3:07:18<66:43:05, 5.72s/it]
4%|▍ | 1933/43939 [3:07:23<66:40:48, 5.71s/it]
4%|▍ | 1934/43939 [3:07:29<66:38:33, 5.71s/it]
4%|▍ | 1935/43939 [3:07:35<66:37:34, 5.71s/it]
4%|▍ | 1936/43939 [3:07:41<66:39:08, 5.71s/it]
4%|▍ | 1937/43939 [3:07:46<66:35:23, 5.71s/it]
4%|▍ | 1938/43939 [3:07:52<66:30:25, 5.70s/it]
4%|▍ | 1939/43939 [3:07:58<66:30:17, 5.70s/it]
4%|▍ | 1940/43939 [3:08:03<66:34:02, 5.71s/it]
4%|▍ | 1941/43939 [3:08:09<66:31:23, 5.70s/it]
4%|▍ | 1942/43939 [3:08:15<66:29:20, 5.70s/it]
4%|▍ | 1943/43939 [3:08:20<66:24:21, 5.69s/it]
4%|▍ | 1944/43939 [3:08:26<66:23:48, 5.69s/it]
4%|▍ | 1945/43939 [3:08:32<66:24:43, 5.69s/it]
4%|▍ | 1946/43939 [3:08:38<67:10:08, 5.76s/it]
4%|▍ | 1947/43939 [3:08:43<66:59:21, 5.74s/it]
4%|▍ | 1948/43939 [3:08:49<66:45:19, 5.72s/it]
4%|▍ | 1949/43939 [3:08:55<66:43:53, 5.72s/it]
4%|▍ | 1950/43939 [3:09:00<66:29:26, 5.70s/it]
{'loss': 1.1293, 'grad_norm': 1.8899917602539062, 'learning_rate': 2.218301843842477e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 467.41, 'epoch': 0.18}
4%|▍ | 1950/43939 [3:09:00<66:29:26, 5.70s/it]
4%|▍ | 1951/43939 [3:09:06<66:40:51, 5.72s/it]
4%|▍ | 1952/43939 [3:09:12<66:40:07, 5.72s/it]
4%|▍ | 1953/43939 [3:09:18<66:31:01, 5.70s/it]
4%|▍ | 1954/43939 [3:09:23<66:28:34, 5.70s/it]
4%|▍ | 1955/43939 [3:09:29<66:29:13, 5.70s/it]
4%|▍ | 1956/43939 [3:09:35<66:27:39, 5.70s/it]
4%|▍ | 1957/43939 [3:09:40<66:23:15, 5.69s/it]
4%|▍ | 1958/43939 [3:09:46<66:28:51, 5.70s/it]
4%|▍ | 1959/43939 [3:09:52<66:27:21, 5.70s/it]
4%|▍ | 1960/43939 [3:09:58<66:31:26, 5.70s/it]
4%|▍ | 1961/43939 [3:10:03<66:36:12, 5.71s/it]
4%|▍ | 1962/43939 [3:10:09<66:34:23, 5.71s/it]
4%|▍ | 1963/43939 [3:10:15<66:33:56, 5.71s/it]
4%|▍ | 1964/43939 [3:10:20<66:35:18, 5.71s/it]
4%|▍ | 1965/43939 [3:10:26<66:32:43, 5.71s/it]
4%|▍ | 1966/43939 [3:10:32<66:31:20, 5.71s/it]
4%|▍ | 1967/43939 [3:10:38<67:19:35, 5.77s/it]
4%|▍ | 1968/43939 [3:10:43<67:03:36, 5.75s/it]
4%|▍ | 1969/43939 [3:10:49<66:58:32, 5.74s/it]
4%|▍ | 1970/43939 [3:10:55<66:46:39, 5.73s/it]
4%|▍ | 1971/43939 [3:11:01<66:41:26, 5.72s/it]
4%|▍ | 1972/43939 [3:11:06<66:30:28, 5.71s/it]
4%|▍ | 1973/43939 [3:11:12<66:27:15, 5.70s/it]
4%|▍ | 1974/43939 [3:11:18<66:32:35, 5.71s/it]
4%|▍ | 1975/43939 [3:11:23<66:27:22, 5.70s/it]
4%|▍ | 1976/43939 [3:11:29<66:32:47, 5.71s/it]
4%|▍ | 1977/43939 [3:11:35<66:29:27, 5.70s/it]
5%|▍ | 1978/43939 [3:11:40<66:23:53, 5.70s/it]
5%|▍ | 1979/43939 [3:11:46<66:44:57, 5.73s/it]
5%|▍ | 1980/43939 [3:11:52<66:39:06, 5.72s/it]
5%|▍ | 1981/43939 [3:11:58<66:34:41, 5.71s/it]
5%|▍ | 1982/43939 [3:12:03<66:28:26, 5.70s/it]
5%|▍ | 1983/43939 [3:12:09<66:23:44, 5.70s/it]
5%|▍ | 1984/43939 [3:12:15<66:23:37, 5.70s/it]
5%|▍ | 1985/43939 [3:12:20<66:21:09, 5.69s/it]
5%|▍ | 1986/43939 [3:12:26<66:32:33, 5.71s/it]
5%|▍ | 1987/43939 [3:12:32<66:37:58, 5.72s/it]
5%|▍ | 1988/43939 [3:12:38<66:31:50, 5.71s/it]
5%|▍ | 1989/43939 [3:12:43<67:17:52, 5.78s/it]
5%|▍ | 1990/43939 [3:12:49<67:04:28, 5.76s/it]
5%|▍ | 1991/43939 [3:12:55<66:58:54, 5.75s/it]
5%|▍ | 1992/43939 [3:13:01<66:47:47, 5.73s/it]
5%|▍ | 1993/43939 [3:13:06<66:59:17, 5.75s/it]
5%|▍ | 1994/43939 [3:13:12<66:41:47, 5.72s/it]
5%|▍ | 1995/43939 [3:13:18<66:52:40, 5.74s/it]
5%|▍ | 1996/43939 [3:13:24<66:46:18, 5.73s/it]
5%|▍ | 1997/43939 [3:13:29<66:39:44, 5.72s/it]
5%|▍ | 1998/43939 [3:13:35<66:36:14, 5.72s/it]
5%|▍ | 1999/43939 [3:13:41<66:33:08, 5.71s/it]
5%|▍ | 2000/43939 [3:13:46<66:31:23, 5.71s/it]
{'loss': 1.1541, 'grad_norm': 1.4181222915649414, 'learning_rate': 2.275210562258138e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 494.36, 'epoch': 0.18}
5%|▍ | 2000/43939 [3:13:46<66:31:23, 5.71s/it]
5%|▍ | 2001/43939 [3:13:52<66:28:13, 5.71s/it]
5%|▍ | 2002/43939 [3:13:58<66:26:23, 5.70s/it]
5%|▍ | 2003/43939 [3:14:03<66:30:28, 5.71s/it]
5%|▍ | 2004/43939 [3:14:09<66:26:20, 5.70s/it]
5%|▍ | 2005/43939 [3:14:15<66:34:54, 5.72s/it]
5%|▍ | 2006/43939 [3:14:21<66:33:25, 5.71s/it]
5%|▍ | 2007/43939 [3:14:26<66:30:32, 5.71s/it]
5%|▍ | 2008/43939 [3:14:32<67:20:04, 5.78s/it]
5%|▍ | 2009/43939 [3:14:38<67:01:18, 5.75s/it]
5%|▍ | 2010/43939 [3:14:44<66:50:09, 5.74s/it]
5%|▍ | 2011/43939 [3:14:49<66:45:52, 5.73s/it]
5%|▍ | 2012/43939 [3:14:55<66:37:18, 5.72s/it]
5%|▍ | 2013/43939 [3:15:01<66:34:24, 5.72s/it]
5%|▍ | 2014/43939 [3:15:06<66:26:36, 5.71s/it]
5%|▍ | 2015/43939 [3:15:12<66:18:33, 5.69s/it]
5%|▍ | 2016/43939 [3:15:18<66:14:05, 5.69s/it]
5%|▍ | 2017/43939 [3:15:23<66:08:41, 5.68s/it]
5%|▍ | 2018/43939 [3:15:29<66:10:51, 5.68s/it]
5%|▍ | 2019/43939 [3:15:35<66:09:37, 5.68s/it]
5%|▍ | 2020/43939 [3:15:40<66:11:02, 5.68s/it]
5%|▍ | 2021/43939 [3:15:46<66:17:48, 5.69s/it]
5%|▍ | 2022/43939 [3:15:52<66:24:10, 5.70s/it]
5%|▍ | 2023/43939 [3:15:58<66:25:19, 5.70s/it]
5%|▍ | 2024/43939 [3:16:03<66:22:58, 5.70s/it]
5%|▍ | 2025/43939 [3:16:09<66:17:27, 5.69s/it]
5%|▍ | 2026/43939 [3:16:15<66:28:52, 5.71s/it]
5%|▍ | 2027/43939 [3:16:20<66:23:27, 5.70s/it]
5%|▍ | 2028/43939 [3:16:26<66:24:09, 5.70s/it]
5%|▍ | 2029/43939 [3:16:32<67:07:31, 5.77s/it]
5%|▍ | 2030/43939 [3:16:38<66:50:53, 5.74s/it]
5%|▍ | 2031/43939 [3:16:44<66:51:48, 5.74s/it]
5%|▍ | 2032/43939 [3:16:49<66:46:21, 5.74s/it]
5%|▍ | 2033/43939 [3:16:55<66:36:23, 5.72s/it]
5%|▍ | 2034/43939 [3:17:01<66:36:38, 5.72s/it]
5%|▍ | 2035/43939 [3:17:06<66:28:52, 5.71s/it]
5%|▍ | 2036/43939 [3:17:12<66:33:28, 5.72s/it]
5%|▍ | 2037/43939 [3:17:18<66:18:01, 5.70s/it]
5%|▍ | 2038/43939 [3:17:23<66:18:23, 5.70s/it]
5%|▍ | 2039/43939 [3:17:29<66:38:05, 5.73s/it]
5%|▍ | 2040/43939 [3:17:35<66:33:35, 5.72s/it]
5%|▍ | 2041/43939 [3:17:41<66:29:25, 5.71s/it]
5%|▍ | 2042/43939 [3:17:46<66:24:02, 5.71s/it]
5%|▍ | 2043/43939 [3:17:52<66:25:27, 5.71s/it]
5%|▍ | 2044/43939 [3:17:58<66:15:04, 5.69s/it]
5%|▍ | 2045/43939 [3:18:03<66:13:48, 5.69s/it]
5%|▍ | 2046/43939 [3:18:09<66:14:28, 5.69s/it]
5%|▍ | 2047/43939 [3:18:15<66:15:56, 5.69s/it]
5%|▍ | 2048/43939 [3:18:20<66:30:12, 5.72s/it]
5%|▍ | 2049/43939 [3:18:26<66:27:36, 5.71s/it]
5%|▍ | 2050/43939 [3:18:32<66:27:57, 5.71s/it]
{'loss': 1.1207, 'grad_norm': 1.6970280408859253, 'learning_rate': 2.3321192806737992e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 457.57, 'epoch': 0.19}
5%|▍ | 2050/43939 [3:18:32<66:27:57, 5.71s/it]
5%|▍ | 2051/43939 [3:18:38<67:06:30, 5.77s/it]
5%|▍ | 2052/43939 [3:18:44<66:55:44, 5.75s/it]
5%|▍ | 2053/43939 [3:18:49<66:59:14, 5.76s/it]
5%|▍ | 2054/43939 [3:18:55<66:49:15, 5.74s/it]
5%|▍ | 2055/43939 [3:19:01<66:35:59, 5.72s/it]
5%|▍ | 2056/43939 [3:19:06<66:32:15, 5.72s/it]
5%|▍ | 2057/43939 [3:19:12<66:28:44, 5.71s/it]
5%|▍ | 2058/43939 [3:19:18<66:30:59, 5.72s/it]
5%|▍ | 2059/43939 [3:19:24<66:32:22, 5.72s/it]
5%|▍ | 2060/43939 [3:19:29<66:35:40, 5.72s/it]
5%|▍ | 2061/43939 [3:19:35<66:21:48, 5.70s/it]
5%|▍ | 2062/43939 [3:19:41<66:19:12, 5.70s/it]
5%|▍ | 2063/43939 [3:19:46<66:21:56, 5.71s/it]
5%|▍ | 2064/43939 [3:19:52<66:23:19, 5.71s/it]
5%|▍ | 2065/43939 [3:19:58<66:15:40, 5.70s/it]
5%|▍ | 2066/43939 [3:20:03<66:28:48, 5.72s/it]
5%|▍ | 2067/43939 [3:20:09<66:26:41, 5.71s/it]
5%|▍ | 2068/43939 [3:20:15<66:23:03, 5.71s/it]
5%|▍ | 2069/43939 [3:20:21<66:16:50, 5.70s/it]
5%|▍ | 2070/43939 [3:20:26<66:23:16, 5.71s/it]
5%|▍ | 2071/43939 [3:20:32<67:06:50, 5.77s/it]
5%|▍ | 2072/43939 [3:20:38<66:51:17, 5.75s/it]
5%|▍ | 2073/43939 [3:20:44<66:38:54, 5.73s/it]
5%|▍ | 2074/43939 [3:20:49<66:38:46, 5.73s/it]
5%|▍ | 2075/43939 [3:20:55<66:31:29, 5.72s/it]
5%|▍ | 2076/43939 [3:21:01<66:22:32, 5.71s/it]
5%|▍ | 2077/43939 [3:21:06<66:30:08, 5.72s/it]
5%|▍ | 2078/43939 [3:21:12<66:31:09, 5.72s/it]
5%|▍ | 2079/43939 [3:21:18<66:35:23, 5.73s/it]
5%|▍ | 2080/43939 [3:21:24<66:18:43, 5.70s/it]
5%|▍ | 2081/43939 [3:21:29<66:23:00, 5.71s/it]
5%|▍ | 2082/43939 [3:21:35<66:14:59, 5.70s/it]
5%|▍ | 2083/43939 [3:21:41<66:22:00, 5.71s/it]
5%|▍ | 2084/43939 [3:21:46<66:15:14, 5.70s/it]
5%|▍ | 2085/43939 [3:21:52<66:11:41, 5.69s/it]
5%|▍ | 2086/43939 [3:21:58<66:07:54, 5.69s/it]
5%|▍ | 2087/43939 [3:22:03<66:10:08, 5.69s/it]
5%|▍ | 2088/43939 [3:22:09<66:05:53, 5.69s/it]
5%|▍ | 2089/43939 [3:22:15<66:09:46, 5.69s/it]
5%|▍ | 2090/43939 [3:22:21<66:13:03, 5.70s/it]
5%|▍ | 2091/43939 [3:22:26<66:26:43, 5.72s/it]
5%|▍ | 2092/43939 [3:22:32<67:29:03, 5.81s/it]
5%|▍ | 2093/43939 [3:22:38<67:02:38, 5.77s/it]
5%|▍ | 2094/43939 [3:22:44<66:54:10, 5.76s/it]
5%|▍ | 2095/43939 [3:22:49<66:41:38, 5.74s/it]
5%|▍ | 2096/43939 [3:22:55<66:30:30, 5.72s/it]
5%|▍ | 2097/43939 [3:23:01<66:25:41, 5.72s/it]
5%|▍ | 2098/43939 [3:23:07<66:34:11, 5.73s/it]
5%|▍ | 2099/43939 [3:23:12<66:32:04, 5.72s/it]
5%|▍ | 2100/43939 [3:23:18<66:28:29, 5.72s/it]
{'loss': 1.1484, 'grad_norm': 1.7740880250930786, 'learning_rate': 2.3890279990894605e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 484.6, 'epoch': 0.19}
5%|▍ | 2100/43939 [3:23:18<66:28:29, 5.72s/it]
5%|▍ | 2101/43939 [3:23:24<66:31:15, 5.72s/it]
5%|▍ | 2102/43939 [3:23:29<66:27:15, 5.72s/it]
5%|▍ | 2103/43939 [3:23:35<66:30:18, 5.72s/it]
5%|▍ | 2104/43939 [3:23:41<66:26:54, 5.72s/it]
5%|▍ | 2105/43939 [3:23:47<66:24:44, 5.72s/it]
5%|▍ | 2106/43939 [3:23:52<66:22:54, 5.71s/it]
5%|▍ | 2107/43939 [3:23:58<66:30:03, 5.72s/it]
5%|▍ | 2108/43939 [3:24:04<66:27:24, 5.72s/it]
5%|▍ | 2109/43939 [3:24:09<66:23:58, 5.71s/it]
5%|▍ | 2110/43939 [3:24:15<66:19:27, 5.71s/it]
5%|▍ | 2111/43939 [3:24:21<66:19:47, 5.71s/it]
5%|▍ | 2112/43939 [3:24:27<66:27:59, 5.72s/it]
5%|▍ | 2113/43939 [3:24:33<67:23:57, 5.80s/it]
5%|▍ | 2114/43939 [3:24:38<67:13:30, 5.79s/it]
5%|▍ | 2115/43939 [3:24:44<67:03:28, 5.77s/it]
5%|▍ | 2116/43939 [3:24:50<66:44:33, 5.75s/it]
5%|▍ | 2117/43939 [3:24:55<66:32:34, 5.73s/it]
5%|▍ | 2118/43939 [3:25:01<66:21:26, 5.71s/it]
5%|▍ | 2119/43939 [3:25:07<66:10:22, 5.70s/it]
5%|▍ | 2120/43939 [3:25:12<66:06:45, 5.69s/it]
5%|▍ | 2121/43939 [3:25:18<66:05:23, 5.69s/it]
5%|▍ | 2122/43939 [3:25:24<66:09:32, 5.70s/it]
5%|▍ | 2123/43939 [3:25:30<66:09:32, 5.70s/it]
5%|▍ | 2124/43939 [3:25:35<66:09:57, 5.70s/it]
5%|▍ | 2125/43939 [3:25:41<66:15:30, 5.70s/it]
5%|▍ | 2126/43939 [3:25:47<66:13:46, 5.70s/it]
5%|▍ | 2127/43939 [3:25:52<66:22:49, 5.72s/it]
5%|▍ | 2128/43939 [3:25:58<66:28:03, 5.72s/it]
5%|▍ | 2129/43939 [3:26:04<66:27:17, 5.72s/it]
5%|▍ | 2130/43939 [3:26:10<66:24:36, 5.72s/it]
5%|▍ | 2131/43939 [3:26:15<66:17:59, 5.71s/it]
5%|▍ | 2132/43939 [3:26:21<66:10:04, 5.70s/it]
5%|▍ | 2133/43939 [3:26:27<66:08:15, 5.70s/it]
5%|▍ | 2134/43939 [3:26:33<67:00:29, 5.77s/it]
5%|▍ | 2135/43939 [3:26:38<66:59:26, 5.77s/it]
5%|▍ | 2136/43939 [3:26:44<66:52:06, 5.76s/it]
5%|▍ | 2137/43939 [3:26:50<66:44:16, 5.75s/it]
5%|▍ | 2138/43939 [3:26:55<66:34:10, 5.73s/it]
5%|▍ | 2139/43939 [3:27:01<66:36:14, 5.74s/it]
5%|▍ | 2140/43939 [3:27:07<66:28:53, 5.73s/it]
5%|▍ | 2141/43939 [3:27:13<66:25:20, 5.72s/it]
5%|▍ | 2142/43939 [3:27:18<66:20:32, 5.71s/it]
5%|▍ | 2143/43939 [3:27:24<66:15:25, 5.71s/it]
5%|▍ | 2144/43939 [3:27:30<66:20:39, 5.71s/it]
5%|▍ | 2145/43939 [3:27:35<66:14:42, 5.71s/it]
5%|▍ | 2146/43939 [3:27:41<66:24:22, 5.72s/it]
5%|▍ | 2147/43939 [3:27:47<66:22:55, 5.72s/it]
5%|▍ | 2148/43939 [3:27:53<66:21:44, 5.72s/it]
5%|▍ | 2149/43939 [3:27:58<66:19:34, 5.71s/it]
5%|▍ | 2150/43939 [3:28:04<66:10:47, 5.70s/it]
{'loss': 1.1451, 'grad_norm': 1.6202048063278198, 'learning_rate': 2.445936717505122e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.76, 'tokens_per_second_per_gpu': 532.16, 'epoch': 0.2}
5%|▍ | 2150/43939 [3:28:04<66:10:47, 5.70s/it]
5%|▍ | 2151/43939 [3:28:10<66:11:00, 5.70s/it]
5%|▍ | 2152/43939 [3:28:15<66:14:57, 5.71s/it]
5%|▍ | 2153/43939 [3:28:21<66:05:39, 5.69s/it]
5%|▍ | 2154/43939 [3:28:27<67:01:18, 5.77s/it]
5%|▍ | 2155/43939 [3:28:33<66:53:00, 5.76s/it]
5%|▍ | 2156/43939 [3:28:38<66:41:07, 5.75s/it]
5%|▍ | 2157/43939 [3:28:44<66:29:36, 5.73s/it]
5%|▍ | 2158/43939 [3:28:50<66:27:59, 5.73s/it]
5%|▍ | 2159/43939 [3:28:56<66:25:37, 5.72s/it]
5%|▍ | 2160/43939 [3:29:01<66:24:38, 5.72s/it]
5%|▍ | 2161/43939 [3:29:07<66:20:39, 5.72s/it]
5%|▍ | 2162/43939 [3:29:13<66:21:01, 5.72s/it]
5%|▍ | 2163/43939 [3:29:18<66:07:46, 5.70s/it]
5%|▍ | 2164/43939 [3:29:24<66:04:45, 5.69s/it]
5%|▍ | 2165/43939 [3:29:30<66:05:52, 5.70s/it]
5%|▍ | 2166/43939 [3:29:35<66:06:11, 5.70s/it]
5%|▍ | 2167/43939 [3:29:41<66:17:14, 5.71s/it]
5%|▍ | 2168/43939 [3:29:47<66:14:23, 5.71s/it]
5%|▍ | 2169/43939 [3:29:53<66:11:26, 5.70s/it]
5%|▍ | 2170/43939 [3:29:58<66:13:33, 5.71s/it]
5%|▍ | 2171/43939 [3:30:04<66:13:51, 5.71s/it]
5%|▍ | 2172/43939 [3:30:10<66:18:51, 5.72s/it]
5%|▍ | 2173/43939 [3:30:15<66:05:19, 5.70s/it]
5%|▍ | 2174/43939 [3:30:21<66:03:31, 5.69s/it]
5%|▍ | 2175/43939 [3:30:27<66:08:39, 5.70s/it]
5%|▍ | 2176/43939 [3:30:33<66:51:07, 5.76s/it]
5%|▍ | 2177/43939 [3:30:38<66:38:43, 5.75s/it]
5%|▍ | 2178/43939 [3:30:44<66:29:31, 5.73s/it]
5%|▍ | 2179/43939 [3:30:50<66:16:26, 5.71s/it]
5%|▍ | 2180/43939 [3:30:56<66:14:06, 5.71s/it]
5%|▍ | 2181/43939 [3:31:01<66:13:53, 5.71s/it]
5%|▍ | 2182/43939 [3:31:07<66:08:11, 5.70s/it]
5%|▍ | 2183/43939 [3:31:13<66:08:13, 5.70s/it]
5%|▍ | 2184/43939 [3:31:18<66:08:44, 5.70s/it]
5%|▍ | 2185/43939 [3:31:24<66:01:28, 5.69s/it]
5%|▍ | 2186/43939 [3:31:30<66:18:55, 5.72s/it]
5%|▍ | 2187/43939 [3:31:35<66:13:37, 5.71s/it]
5%|▍ | 2188/43939 [3:31:41<66:10:01, 5.71s/it]
5%|▍ | 2189/43939 [3:31:47<66:11:04, 5.71s/it]
5%|▍ | 2190/43939 [3:31:53<66:08:48, 5.70s/it]
5%|▍ | 2191/43939 [3:31:58<66:09:40, 5.71s/it]
5%|▍ | 2192/43939 [3:32:04<66:08:07, 5.70s/it]
5%|▍ | 2193/43939 [3:32:10<66:23:17, 5.73s/it]
5%|▍ | 2194/43939 [3:32:15<66:13:43, 5.71s/it]
5%|▍ | 2195/43939 [3:32:21<66:09:20, 5.71s/it]
5%|▍ | 2196/43939 [3:32:27<66:55:49, 5.77s/it]
5%|▌ | 2197/43939 [3:32:33<66:38:15, 5.75s/it][2025-11-28 21:36:46,564] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.38s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:41<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<01:00, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.74s/it][A
79%|███████▉ | 57/72 [02:32<00:41, 2.74s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.74s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.74s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.74s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.74s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 1.119379997253418, 'eval_runtime': 195.6152, 'eval_samples_per_second': 5.107, 'eval_steps_per_second': 0.368, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 0.2}
5%|▌ | 2197/43939 [3:35:48<66:38:15, 5.75s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
5%|▌ | 2198/43939 [3:35:54<747:23:22, 64.46s/it]
5%|▌ | 2199/43939 [3:36:00<542:56:28, 46.83s/it]
5%|▌ | 2200/43939 [3:36:06<399:52:17, 34.49s/it]
{'loss': 1.1345, 'grad_norm': 1.9262148141860962, 'learning_rate': 2.502845435920783e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.06, 'tokens_per_second_per_gpu': 24.4, 'epoch': 0.2}
5%|▌ | 2200/43939 [3:36:06<399:52:17, 34.49s/it]
5%|▌ | 2201/43939 [3:36:11<299:47:34, 25.86s/it]
5%|▌ | 2202/43939 [3:36:17<229:40:52, 19.81s/it]
5%|▌ | 2203/43939 [3:36:23<180:27:49, 15.57s/it]
5%|▌ | 2204/43939 [3:36:28<146:07:52, 12.61s/it]
5%|▌ | 2205/43939 [3:36:34<122:05:05, 10.53s/it]
5%|▌ | 2206/43939 [3:36:40<105:34:26, 9.11s/it]
5%|▌ | 2207/43939 [3:36:46<93:41:28, 8.08s/it]
5%|▌ | 2208/43939 [3:36:51<85:28:06, 7.37s/it]
5%|▌ | 2209/43939 [3:36:57<79:37:15, 6.87s/it]
5%|▌ | 2210/43939 [3:37:03<75:29:07, 6.51s/it]
5%|▌ | 2211/43939 [3:37:08<72:33:45, 6.26s/it]
5%|▌ | 2212/43939 [3:37:14<70:42:12, 6.10s/it]
5%|▌ | 2213/43939 [3:37:20<69:14:46, 5.97s/it]
5%|▌ | 2214/43939 [3:37:25<68:19:08, 5.89s/it]
5%|▌ | 2215/43939 [3:37:31<67:31:54, 5.83s/it]
5%|▌ | 2216/43939 [3:37:37<67:58:41, 5.87s/it]
5%|▌ | 2217/43939 [3:37:43<67:24:12, 5.82s/it]
5%|▌ | 2218/43939 [3:37:48<66:54:12, 5.77s/it]
5%|▌ | 2219/43939 [3:37:54<66:40:02, 5.75s/it]
5%|▌ | 2220/43939 [3:38:00<66:36:46, 5.75s/it]
5%|▌ | 2221/43939 [3:38:06<66:25:46, 5.73s/it]
5%|▌ | 2222/43939 [3:38:11<66:26:09, 5.73s/it]
5%|▌ | 2223/43939 [3:38:17<66:17:02, 5.72s/it]
5%|▌ | 2224/43939 [3:38:23<66:05:42, 5.70s/it]
5%|▌ | 2225/43939 [3:38:28<65:57:21, 5.69s/it]
5%|▌ | 2226/43939 [3:38:34<66:05:55, 5.70s/it]
5%|▌ | 2227/43939 [3:38:40<65:57:34, 5.69s/it]
5%|▌ | 2228/43939 [3:38:45<65:59:08, 5.70s/it]
5%|▌ | 2229/43939 [3:38:51<65:59:44, 5.70s/it]
5%|▌ | 2230/43939 [3:38:57<66:04:50, 5.70s/it]
5%|▌ | 2231/43939 [3:39:03<66:06:26, 5.71s/it]
5%|▌ | 2232/43939 [3:39:08<66:05:54, 5.71s/it]
5%|▌ | 2233/43939 [3:39:14<66:00:49, 5.70s/it]
5%|▌ | 2234/43939 [3:39:20<66:08:20, 5.71s/it]
5%|▌ | 2235/43939 [3:39:25<66:03:52, 5.70s/it]
5%|▌ | 2236/43939 [3:39:31<65:57:27, 5.69s/it]
5%|▌ | 2237/43939 [3:39:37<65:58:43, 5.70s/it]
5%|▌ | 2238/43939 [3:39:43<66:39:55, 5.76s/it]
5%|▌ | 2239/43939 [3:39:48<66:29:21, 5.74s/it]
5%|▌ | 2240/43939 [3:39:54<66:25:19, 5.73s/it]
5%|▌ | 2241/43939 [3:40:00<66:15:43, 5.72s/it]
5%|▌ | 2242/43939 [3:40:05<66:15:15, 5.72s/it]
5%|▌ | 2243/43939 [3:40:11<66:06:08, 5.71s/it]
5%|▌ | 2244/43939 [3:40:17<66:16:23, 5.72s/it]
5%|▌ | 2245/43939 [3:40:23<66:09:53, 5.71s/it]
5%|▌ | 2246/43939 [3:40:28<66:03:16, 5.70s/it]
5%|▌ | 2247/43939 [3:40:34<66:13:02, 5.72s/it]
5%|▌ | 2248/43939 [3:40:40<66:03:53, 5.70s/it]
5%|▌ | 2249/43939 [3:40:45<66:04:00, 5.70s/it]
5%|▌ | 2250/43939 [3:40:51<66:04:26, 5.71s/it]
{'loss': 1.1275, 'grad_norm': 1.6392292976379395, 'learning_rate': 2.5597541543364445e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.24, 'epoch': 0.2}
5%|▌ | 2250/43939 [3:40:51<66:04:26, 5.71s/it]
5%|▌ | 2251/43939 [3:40:57<66:06:09, 5.71s/it]
5%|▌ | 2252/43939 [3:41:03<66:08:55, 5.71s/it]
5%|▌ | 2253/43939 [3:41:08<66:06:35, 5.71s/it]
5%|▌ | 2254/43939 [3:41:14<66:08:21, 5.71s/it]
5%|▌ | 2255/43939 [3:41:20<66:10:35, 5.72s/it]
5%|▌ | 2256/43939 [3:41:25<66:13:12, 5.72s/it]
5%|▌ | 2257/43939 [3:41:31<66:06:48, 5.71s/it]
5%|▌ | 2258/43939 [3:41:37<65:57:48, 5.70s/it]
5%|▌ | 2259/43939 [3:41:43<66:45:02, 5.77s/it]
5%|▌ | 2260/43939 [3:41:48<66:31:25, 5.75s/it]
5%|▌ | 2261/43939 [3:41:54<66:28:14, 5.74s/it]
5%|▌ | 2262/43939 [3:42:00<66:18:09, 5.73s/it]
5%|▌ | 2263/43939 [3:42:05<66:09:48, 5.72s/it]
5%|▌ | 2264/43939 [3:42:11<66:07:27, 5.71s/it]
5%|▌ | 2265/43939 [3:42:17<66:00:30, 5.70s/it]
5%|▌ | 2266/43939 [3:42:23<65:52:00, 5.69s/it]
5%|▌ | 2267/43939 [3:42:28<65:49:47, 5.69s/it]
5%|▌ | 2268/43939 [3:42:34<65:47:57, 5.68s/it]
5%|▌ | 2269/43939 [3:42:40<66:00:02, 5.70s/it]
5%|▌ | 2270/43939 [3:42:45<65:51:40, 5.69s/it]
5%|▌ | 2271/43939 [3:42:51<65:59:59, 5.70s/it]
5%|▌ | 2272/43939 [3:42:57<66:08:38, 5.71s/it]
5%|▌ | 2273/43939 [3:43:02<66:06:59, 5.71s/it]
5%|▌ | 2274/43939 [3:43:08<66:05:11, 5.71s/it]
5%|▌ | 2275/43939 [3:43:14<65:58:20, 5.70s/it]
5%|▌ | 2276/43939 [3:43:20<65:55:06, 5.70s/it]
5%|▌ | 2277/43939 [3:43:25<65:49:24, 5.69s/it]
5%|▌ | 2278/43939 [3:43:31<65:50:17, 5.69s/it]
5%|▌ | 2279/43939 [3:43:37<66:39:51, 5.76s/it]
5%|▌ | 2280/43939 [3:43:43<66:32:16, 5.75s/it]
5%|▌ | 2281/43939 [3:43:48<66:22:06, 5.74s/it]
5%|▌ | 2282/43939 [3:43:54<66:18:51, 5.73s/it]
5%|▌ | 2283/43939 [3:44:00<66:13:24, 5.72s/it]
5%|▌ | 2284/43939 [3:44:05<66:12:25, 5.72s/it]
5%|▌ | 2285/43939 [3:44:11<66:15:05, 5.73s/it]
5%|▌ | 2286/43939 [3:44:17<66:11:17, 5.72s/it]
5%|▌ | 2287/43939 [3:44:23<66:05:12, 5.71s/it]
5%|▌ | 2288/43939 [3:44:28<66:09:20, 5.72s/it]
5%|▌ | 2289/43939 [3:44:34<66:12:13, 5.72s/it]
5%|▌ | 2290/43939 [3:44:40<66:08:09, 5.72s/it]
5%|▌ | 2291/43939 [3:44:45<66:07:57, 5.72s/it]
5%|▌ | 2292/43939 [3:44:51<66:15:58, 5.73s/it]
5%|▌ | 2293/43939 [3:44:57<66:11:43, 5.72s/it]
5%|▌ | 2294/43939 [3:45:03<66:18:16, 5.73s/it]
5%|▌ | 2295/43939 [3:45:08<66:09:11, 5.72s/it]
5%|▌ | 2296/43939 [3:45:14<66:13:51, 5.73s/it]
5%|▌ | 2297/43939 [3:45:20<66:04:08, 5.71s/it]
5%|▌ | 2298/43939 [3:45:25<66:10:57, 5.72s/it]
5%|▌ | 2299/43939 [3:45:31<66:06:06, 5.71s/it]
5%|▌ | 2300/43939 [3:45:37<65:53:10, 5.70s/it]
{'loss': 1.1246, 'grad_norm': 1.8397603034973145, 'learning_rate': 2.6166628727521058e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.72, 'epoch': 0.21}
5%|▌ | 2300/43939 [3:45:37<65:53:10, 5.70s/it]
5%|▌ | 2301/43939 [3:45:43<66:46:46, 5.77s/it]
5%|▌ | 2302/43939 [3:45:48<66:27:57, 5.75s/it]
5%|▌ | 2303/43939 [3:45:54<66:34:30, 5.76s/it]
5%|▌ | 2304/43939 [3:46:00<66:13:26, 5.73s/it]
5%|▌ | 2305/43939 [3:46:06<66:07:40, 5.72s/it]
5%|▌ | 2306/43939 [3:46:11<66:06:22, 5.72s/it]
5%|▌ | 2307/43939 [3:46:17<66:00:59, 5.71s/it]
5%|▌ | 2308/43939 [3:46:23<66:03:18, 5.71s/it]
5%|▌ | 2309/43939 [3:46:28<66:01:50, 5.71s/it]
5%|▌ | 2310/43939 [3:46:34<66:09:38, 5.72s/it]
5%|▌ | 2311/43939 [3:46:40<66:12:01, 5.73s/it]
5%|▌ | 2312/43939 [3:46:46<66:06:40, 5.72s/it]
5%|▌ | 2313/43939 [3:46:51<65:56:33, 5.70s/it]
5%|▌ | 2314/43939 [3:46:57<65:53:00, 5.70s/it]
5%|▌ | 2315/43939 [3:47:03<66:04:55, 5.72s/it]
5%|▌ | 2316/43939 [3:47:08<66:03:54, 5.71s/it]
5%|▌ | 2317/43939 [3:47:14<66:08:14, 5.72s/it]
5%|▌ | 2318/43939 [3:47:20<66:00:23, 5.71s/it]
5%|▌ | 2319/43939 [3:47:26<66:03:41, 5.71s/it]
5%|▌ | 2320/43939 [3:47:31<66:05:50, 5.72s/it]
5%|▌ | 2321/43939 [3:47:37<65:57:00, 5.70s/it]
5%|▌ | 2322/43939 [3:47:43<66:33:08, 5.76s/it]
5%|▌ | 2323/43939 [3:47:49<66:19:52, 5.74s/it]
5%|▌ | 2324/43939 [3:47:54<66:10:52, 5.73s/it]
5%|▌ | 2325/43939 [3:48:00<66:12:39, 5.73s/it]
5%|▌ | 2326/43939 [3:48:06<66:11:53, 5.73s/it]
5%|▌ | 2327/43939 [3:48:11<66:10:36, 5.73s/it]
5%|▌ | 2328/43939 [3:48:17<66:04:38, 5.72s/it]
5%|▌ | 2329/43939 [3:48:23<65:55:39, 5.70s/it]
5%|▌ | 2330/43939 [3:48:29<65:52:07, 5.70s/it]
5%|▌ | 2331/43939 [3:48:34<65:55:45, 5.70s/it]
5%|▌ | 2332/43939 [3:48:40<66:03:33, 5.72s/it]
5%|▌ | 2333/43939 [3:48:46<65:58:27, 5.71s/it]
5%|▌ | 2334/43939 [3:48:51<65:58:02, 5.71s/it]
5%|▌ | 2335/43939 [3:48:57<65:55:14, 5.70s/it]
5%|▌ | 2336/43939 [3:49:03<65:52:02, 5.70s/it]
5%|▌ | 2337/43939 [3:49:08<65:51:15, 5.70s/it]
5%|▌ | 2338/43939 [3:49:14<66:03:39, 5.72s/it]
5%|▌ | 2339/43939 [3:49:20<66:05:01, 5.72s/it]
5%|▌ | 2340/43939 [3:49:26<66:08:30, 5.72s/it]
5%|▌ | 2341/43939 [3:49:31<66:04:03, 5.72s/it]
5%|▌ | 2342/43939 [3:49:37<66:01:44, 5.71s/it]
5%|▌ | 2343/43939 [3:49:43<66:46:53, 5.78s/it]
5%|▌ | 2344/43939 [3:49:49<66:31:28, 5.76s/it]
5%|▌ | 2345/43939 [3:49:54<66:22:52, 5.75s/it]
5%|▌ | 2346/43939 [3:50:00<66:16:24, 5.74s/it]
5%|▌ | 2347/43939 [3:50:06<66:06:18, 5.72s/it]
5%|▌ | 2348/43939 [3:50:12<66:00:17, 5.71s/it]
5%|▌ | 2349/43939 [3:50:17<66:02:24, 5.72s/it]
5%|▌ | 2350/43939 [3:50:23<65:55:46, 5.71s/it]
{'loss': 1.1322, 'grad_norm': 1.2947629690170288, 'learning_rate': 2.673571591167767e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.82, 'epoch': 0.21}
5%|▌ | 2350/43939 [3:50:23<65:55:46, 5.71s/it]
5%|▌ | 2351/43939 [3:50:29<66:13:34, 5.73s/it]
5%|▌ | 2352/43939 [3:50:34<66:14:34, 5.73s/it]
5%|▌ | 2353/43939 [3:50:40<66:18:49, 5.74s/it]
5%|▌ | 2354/43939 [3:50:46<66:07:33, 5.72s/it]
5%|▌ | 2355/43939 [3:50:52<66:14:58, 5.74s/it]
5%|▌ | 2356/43939 [3:50:57<66:09:53, 5.73s/it]
5%|▌ | 2357/43939 [3:51:03<66:08:48, 5.73s/it]
5%|▌ | 2358/43939 [3:51:09<66:17:58, 5.74s/it]
5%|▌ | 2359/43939 [3:51:15<66:16:57, 5.74s/it]
5%|▌ | 2360/43939 [3:51:20<66:18:00, 5.74s/it]
5%|▌ | 2361/43939 [3:51:26<66:07:50, 5.73s/it]
5%|▌ | 2362/43939 [3:51:32<66:03:16, 5.72s/it]
5%|▌ | 2363/43939 [3:51:38<66:16:43, 5.74s/it]
5%|▌ | 2364/43939 [3:51:44<67:07:08, 5.81s/it]
5%|▌ | 2365/43939 [3:51:49<66:43:20, 5.78s/it]
5%|▌ | 2366/43939 [3:51:55<66:25:14, 5.75s/it]
5%|▌ | 2367/43939 [3:52:01<66:14:34, 5.74s/it]
5%|▌ | 2368/43939 [3:52:06<66:14:18, 5.74s/it]
5%|▌ | 2369/43939 [3:52:12<66:09:50, 5.73s/it]
5%|▌ | 2370/43939 [3:52:18<66:04:45, 5.72s/it]
5%|▌ | 2371/43939 [3:52:23<65:59:43, 5.72s/it]
5%|▌ | 2372/43939 [3:52:29<66:09:26, 5.73s/it]
5%|▌ | 2373/43939 [3:52:35<66:05:35, 5.72s/it]
5%|▌ | 2374/43939 [3:52:41<65:55:43, 5.71s/it]
5%|▌ | 2375/43939 [3:52:46<66:01:05, 5.72s/it]
5%|▌ | 2376/43939 [3:52:52<65:58:18, 5.71s/it]
5%|▌ | 2377/43939 [3:52:58<65:55:26, 5.71s/it]
5%|▌ | 2378/43939 [3:53:03<65:51:44, 5.70s/it]
5%|▌ | 2379/43939 [3:53:09<65:52:19, 5.71s/it]
5%|▌ | 2380/43939 [3:53:15<65:54:36, 5.71s/it]
5%|▌ | 2381/43939 [3:53:21<65:56:00, 5.71s/it]
5%|▌ | 2382/43939 [3:53:26<65:54:07, 5.71s/it]
5%|▌ | 2383/43939 [3:53:32<65:51:12, 5.70s/it]
5%|▌ | 2384/43939 [3:53:38<65:43:50, 5.69s/it]
5%|▌ | 2385/43939 [3:53:44<66:40:19, 5.78s/it]
5%|▌ | 2386/43939 [3:53:49<66:28:00, 5.76s/it]
5%|▌ | 2387/43939 [3:53:55<66:17:47, 5.74s/it]
5%|▌ | 2388/43939 [3:54:01<66:01:42, 5.72s/it]
5%|▌ | 2389/43939 [3:54:06<66:07:28, 5.73s/it]
5%|▌ | 2390/43939 [3:54:12<66:07:02, 5.73s/it]
5%|▌ | 2391/43939 [3:54:18<66:02:28, 5.72s/it]
5%|▌ | 2392/43939 [3:54:24<65:56:09, 5.71s/it]
5%|▌ | 2393/43939 [3:54:29<65:55:58, 5.71s/it]
5%|▌ | 2394/43939 [3:54:35<66:00:07, 5.72s/it]
5%|▌ | 2395/43939 [3:54:41<65:56:08, 5.71s/it]
5%|▌ | 2396/43939 [3:54:46<65:48:47, 5.70s/it]
5%|▌ | 2397/43939 [3:54:52<65:50:01, 5.71s/it]
5%|▌ | 2398/43939 [3:54:58<65:49:26, 5.70s/it]
5%|▌ | 2399/43939 [3:55:04<65:48:57, 5.70s/it]
5%|▌ | 2400/43939 [3:55:09<65:43:59, 5.70s/it]
{'loss': 1.1028, 'grad_norm': 1.5403435230255127, 'learning_rate': 2.730480309583428e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.72, 'epoch': 0.22}
5%|▌ | 2400/43939 [3:55:09<65:43:59, 5.70s/it]
5%|▌ | 2401/43939 [3:55:15<65:53:54, 5.71s/it]
5%|▌ | 2402/43939 [3:55:21<65:53:12, 5.71s/it]
5%|▌ | 2403/43939 [3:55:26<65:52:10, 5.71s/it]
5%|▌ | 2404/43939 [3:55:32<65:51:45, 5.71s/it]
5%|▌ | 2405/43939 [3:55:38<65:44:11, 5.70s/it]
5%|▌ | 2406/43939 [3:55:44<66:31:51, 5.77s/it]
5%|▌ | 2407/43939 [3:55:49<66:07:45, 5.73s/it]
5%|▌ | 2408/43939 [3:55:55<65:59:33, 5.72s/it]
5%|▌ | 2409/43939 [3:56:01<66:00:09, 5.72s/it]
5%|▌ | 2410/43939 [3:56:06<65:59:30, 5.72s/it]
5%|▌ | 2411/43939 [3:56:12<65:55:13, 5.71s/it]
5%|▌ | 2412/43939 [3:56:18<65:53:10, 5.71s/it]
5%|▌ | 2413/43939 [3:56:24<65:55:28, 5.72s/it]
5%|▌ | 2414/43939 [3:56:29<65:48:01, 5.70s/it]
5%|▌ | 2415/43939 [3:56:35<66:03:24, 5.73s/it]
5%|▌ | 2416/43939 [3:56:41<65:48:03, 5.70s/it]
6%|▌ | 2417/43939 [3:56:46<65:51:25, 5.71s/it]
6%|▌ | 2418/43939 [3:56:52<65:43:34, 5.70s/it]
6%|▌ | 2419/43939 [3:56:58<65:46:37, 5.70s/it]
6%|▌ | 2420/43939 [3:57:04<66:00:32, 5.72s/it]
6%|▌ | 2421/43939 [3:57:09<65:55:01, 5.72s/it]
6%|▌ | 2422/43939 [3:57:15<65:56:39, 5.72s/it]
6%|▌ | 2423/43939 [3:57:21<65:48:57, 5.71s/it]
6%|▌ | 2424/43939 [3:57:26<65:56:29, 5.72s/it]
6%|▌ | 2425/43939 [3:57:32<65:49:41, 5.71s/it]
6%|▌ | 2426/43939 [3:57:38<65:48:11, 5.71s/it]
6%|▌ | 2427/43939 [3:57:44<66:35:09, 5.77s/it]
6%|▌ | 2428/43939 [3:57:49<66:19:40, 5.75s/it]
6%|▌ | 2429/43939 [3:57:55<66:19:43, 5.75s/it]
6%|▌ | 2430/43939 [3:58:01<66:02:53, 5.73s/it]
6%|▌ | 2431/43939 [3:58:07<65:53:45, 5.72s/it]
6%|▌ | 2432/43939 [3:58:12<65:46:55, 5.71s/it]
6%|▌ | 2433/43939 [3:58:18<65:51:11, 5.71s/it]
6%|▌ | 2434/43939 [3:58:24<65:58:00, 5.72s/it]
6%|▌ | 2435/43939 [3:58:29<65:59:45, 5.72s/it]
6%|▌ | 2436/43939 [3:58:35<65:53:24, 5.72s/it]
6%|▌ | 2437/43939 [3:58:41<65:51:08, 5.71s/it]
6%|▌ | 2438/43939 [3:58:47<65:47:21, 5.71s/it]
6%|▌ | 2439/43939 [3:58:52<65:42:10, 5.70s/it]
6%|▌ | 2440/43939 [3:58:58<65:43:12, 5.70s/it]
6%|▌ | 2441/43939 [3:59:04<65:44:40, 5.70s/it]
6%|▌ | 2442/43939 [3:59:09<65:45:40, 5.70s/it]
6%|▌ | 2443/43939 [3:59:15<65:40:42, 5.70s/it]
6%|▌ | 2444/43939 [3:59:21<65:40:43, 5.70s/it]
6%|▌ | 2445/43939 [3:59:26<65:48:45, 5.71s/it]
6%|▌ | 2446/43939 [3:59:32<65:48:52, 5.71s/it]
6%|▌ | 2447/43939 [3:59:38<65:46:36, 5.71s/it]
6%|▌ | 2448/43939 [3:59:44<66:22:29, 5.76s/it]
6%|▌ | 2449/43939 [3:59:49<66:09:40, 5.74s/it]
6%|▌ | 2450/43939 [3:59:55<66:01:12, 5.73s/it]
{'loss': 1.1469, 'grad_norm': 1.2514817714691162, 'learning_rate': 2.7873890279990895e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 525.63, 'epoch': 0.22}
6%|▌ | 2450/43939 [3:59:55<66:01:12, 5.73s/it]
6%|▌ | 2451/43939 [4:00:01<65:54:44, 5.72s/it]
6%|▌ | 2452/43939 [4:00:07<65:46:37, 5.71s/it]
6%|▌ | 2453/43939 [4:00:12<65:50:57, 5.71s/it]
6%|▌ | 2454/43939 [4:00:18<65:52:48, 5.72s/it]
6%|▌ | 2455/43939 [4:00:24<65:59:24, 5.73s/it]
6%|▌ | 2456/43939 [4:00:29<65:58:03, 5.72s/it]
6%|▌ | 2457/43939 [4:00:35<65:54:50, 5.72s/it]
6%|▌ | 2458/43939 [4:00:41<65:55:50, 5.72s/it]
6%|▌ | 2459/43939 [4:00:47<65:54:45, 5.72s/it]
6%|▌ | 2460/43939 [4:00:52<65:48:01, 5.71s/it]
6%|▌ | 2461/43939 [4:00:58<65:54:52, 5.72s/it]
6%|▌ | 2462/43939 [4:01:04<65:45:24, 5.71s/it]
6%|▌ | 2463/43939 [4:01:09<65:37:18, 5.70s/it]
6%|▌ | 2464/43939 [4:01:15<65:34:35, 5.69s/it]
6%|▌ | 2465/43939 [4:01:21<65:39:19, 5.70s/it]
6%|▌ | 2466/43939 [4:01:27<65:50:34, 5.72s/it]
6%|▌ | 2467/43939 [4:01:32<65:53:02, 5.72s/it]
6%|▌ | 2468/43939 [4:01:38<65:47:53, 5.71s/it]
6%|▌ | 2469/43939 [4:01:44<66:29:57, 5.77s/it]
6%|▌ | 2470/43939 [4:01:50<66:07:14, 5.74s/it]
6%|▌ | 2471/43939 [4:01:55<66:03:26, 5.73s/it]
6%|▌ | 2472/43939 [4:02:01<65:56:36, 5.72s/it]
6%|▌ | 2473/43939 [4:02:07<66:07:43, 5.74s/it]
6%|▌ | 2474/43939 [4:02:12<66:04:05, 5.74s/it]
6%|▌ | 2475/43939 [4:02:18<65:55:36, 5.72s/it]
6%|▌ | 2476/43939 [4:02:24<65:48:58, 5.71s/it]
6%|▌ | 2477/43939 [4:02:30<65:44:56, 5.71s/it]
6%|▌ | 2478/43939 [4:02:35<65:42:24, 5.71s/it]
6%|▌ | 2479/43939 [4:02:41<65:41:01, 5.70s/it]
6%|▌ | 2480/43939 [4:02:47<65:37:10, 5.70s/it]
6%|▌ | 2481/43939 [4:02:52<65:40:51, 5.70s/it]
6%|▌ | 2482/43939 [4:02:58<65:42:53, 5.71s/it]
6%|▌ | 2483/43939 [4:03:04<65:35:22, 5.70s/it]
6%|▌ | 2484/43939 [4:03:09<65:35:46, 5.70s/it]
6%|▌ | 2485/43939 [4:03:15<65:37:14, 5.70s/it]
6%|▌ | 2486/43939 [4:03:21<65:29:26, 5.69s/it]
6%|▌ | 2487/43939 [4:03:26<65:28:56, 5.69s/it]
6%|▌ | 2488/43939 [4:03:32<65:20:08, 5.67s/it]
6%|▌ | 2489/43939 [4:03:38<65:22:00, 5.68s/it]
6%|▌ | 2490/43939 [4:03:44<66:13:51, 5.75s/it]
6%|▌ | 2491/43939 [4:03:49<66:04:23, 5.74s/it]
6%|▌ | 2492/43939 [4:03:55<65:57:30, 5.73s/it]
6%|▌ | 2493/43939 [4:04:01<65:58:42, 5.73s/it]
6%|▌ | 2494/43939 [4:04:07<66:03:57, 5.74s/it]
6%|▌ | 2495/43939 [4:04:12<65:52:19, 5.72s/it]
6%|▌ | 2496/43939 [4:04:18<65:51:06, 5.72s/it]
6%|▌ | 2497/43939 [4:04:24<65:39:53, 5.70s/it]
6%|▌ | 2498/43939 [4:04:29<65:39:01, 5.70s/it]
6%|▌ | 2499/43939 [4:04:35<65:31:31, 5.69s/it]
6%|▌ | 2500/43939 [4:04:41<65:32:01, 5.69s/it]
{'loss': 1.1068, 'grad_norm': 1.5391417741775513, 'learning_rate': 2.8442977464147508e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.06, 'epoch': 0.23}
6%|▌ | 2500/43939 [4:04:41<65:32:01, 5.69s/it]
6%|▌ | 2501/43939 [4:04:46<65:33:08, 5.69s/it]
6%|▌ | 2502/43939 [4:04:52<65:34:51, 5.70s/it]
6%|▌ | 2503/43939 [4:04:58<65:32:24, 5.69s/it]
6%|▌ | 2504/43939 [4:05:04<65:32:22, 5.69s/it]
6%|▌ | 2505/43939 [4:05:09<65:26:50, 5.69s/it]
6%|▌ | 2506/43939 [4:05:15<65:26:50, 5.69s/it]
6%|▌ | 2507/43939 [4:05:21<65:28:33, 5.69s/it]
6%|▌ | 2508/43939 [4:05:26<65:32:56, 5.70s/it]
6%|▌ | 2509/43939 [4:05:32<65:29:24, 5.69s/it]
6%|▌ | 2510/43939 [4:05:38<65:37:20, 5.70s/it]
6%|▌ | 2511/43939 [4:05:44<66:28:48, 5.78s/it]
6%|▌ | 2512/43939 [4:05:49<66:10:19, 5.75s/it]
6%|▌ | 2513/43939 [4:05:55<66:03:01, 5.74s/it]
6%|▌ | 2514/43939 [4:06:01<66:07:49, 5.75s/it]
6%|▌ | 2515/43939 [4:06:07<65:59:10, 5.73s/it]
6%|▌ | 2516/43939 [4:06:12<65:52:30, 5.73s/it]
6%|▌ | 2517/43939 [4:06:18<65:53:57, 5.73s/it]
6%|▌ | 2518/43939 [4:06:24<65:45:25, 5.72s/it]
6%|▌ | 2519/43939 [4:06:29<65:46:25, 5.72s/it]
6%|▌ | 2520/43939 [4:06:35<65:47:17, 5.72s/it]
6%|▌ | 2521/43939 [4:06:41<65:41:55, 5.71s/it]
6%|▌ | 2522/43939 [4:06:47<65:54:50, 5.73s/it]
6%|▌ | 2523/43939 [4:06:52<65:53:07, 5.73s/it]
6%|▌ | 2524/43939 [4:06:58<65:55:13, 5.73s/it]
6%|▌ | 2525/43939 [4:07:04<65:50:11, 5.72s/it]
6%|▌ | 2526/43939 [4:07:09<65:42:49, 5.71s/it]
6%|▌ | 2527/43939 [4:07:15<65:40:45, 5.71s/it]
6%|▌ | 2528/43939 [4:07:21<65:46:16, 5.72s/it]
6%|▌ | 2529/43939 [4:07:27<65:44:53, 5.72s/it]
6%|▌ | 2530/43939 [4:07:32<65:36:53, 5.70s/it]
6%|▌ | 2531/43939 [4:07:38<65:35:18, 5.70s/it]
6%|▌ | 2532/43939 [4:07:44<66:15:45, 5.76s/it]
6%|▌ | 2533/43939 [4:07:50<66:05:20, 5.75s/it]
6%|▌ | 2534/43939 [4:07:55<66:01:56, 5.74s/it]
6%|▌ | 2535/43939 [4:08:01<65:45:44, 5.72s/it]
6%|▌ | 2536/43939 [4:08:07<65:41:10, 5.71s/it]
6%|▌ | 2537/43939 [4:08:12<65:28:52, 5.69s/it]
6%|▌ | 2538/43939 [4:08:18<65:41:42, 5.71s/it]
6%|▌ | 2539/43939 [4:08:24<65:43:04, 5.71s/it]
6%|▌ | 2540/43939 [4:08:29<65:42:43, 5.71s/it]
6%|▌ | 2541/43939 [4:08:35<65:40:31, 5.71s/it]
6%|▌ | 2542/43939 [4:08:41<65:47:47, 5.72s/it]
6%|▌ | 2543/43939 [4:08:47<65:41:54, 5.71s/it]
6%|▌ | 2544/43939 [4:08:52<65:37:45, 5.71s/it]
6%|▌ | 2545/43939 [4:08:58<65:35:53, 5.71s/it]
6%|▌ | 2546/43939 [4:09:04<65:37:00, 5.71s/it]
6%|▌ | 2547/43939 [4:09:09<65:36:04, 5.71s/it]
6%|▌ | 2548/43939 [4:09:15<65:33:02, 5.70s/it]
6%|▌ | 2549/43939 [4:09:21<65:36:10, 5.71s/it]
6%|▌ | 2550/43939 [4:09:27<65:56:03, 5.73s/it]
{'loss': 1.1453, 'grad_norm': 1.393747091293335, 'learning_rate': 2.901206464830412e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.26, 'epoch': 0.23}
6%|▌ | 2550/43939 [4:09:27<65:56:03, 5.73s/it]
6%|▌ | 2551/43939 [4:09:32<65:58:36, 5.74s/it]
6%|▌ | 2552/43939 [4:09:38<65:51:21, 5.73s/it]
6%|▌ | 2553/43939 [4:09:44<66:44:20, 5.81s/it]
6%|▌ | 2554/43939 [4:09:50<66:11:15, 5.76s/it]
6%|▌ | 2555/43939 [4:09:55<65:53:00, 5.73s/it]
6%|▌ | 2556/43939 [4:10:01<65:41:59, 5.72s/it]
6%|▌ | 2557/43939 [4:10:07<65:45:24, 5.72s/it]
6%|▌ | 2558/43939 [4:10:12<65:31:53, 5.70s/it]
6%|▌ | 2559/43939 [4:10:18<65:34:16, 5.70s/it]
6%|▌ | 2560/43939 [4:10:24<65:35:35, 5.71s/it]
6%|▌ | 2561/43939 [4:10:30<65:39:03, 5.71s/it]
6%|▌ | 2562/43939 [4:10:35<65:28:42, 5.70s/it]
6%|▌ | 2563/43939 [4:10:41<65:36:06, 5.71s/it]
6%|▌ | 2564/43939 [4:10:47<65:34:43, 5.71s/it]
6%|▌ | 2565/43939 [4:10:52<65:32:40, 5.70s/it]
6%|▌ | 2566/43939 [4:10:58<65:26:54, 5.69s/it]
6%|▌ | 2567/43939 [4:11:04<65:23:33, 5.69s/it]
6%|▌ | 2568/43939 [4:11:09<65:22:34, 5.69s/it]
6%|▌ | 2569/43939 [4:11:15<65:33:46, 5.71s/it]
6%|▌ | 2570/43939 [4:11:21<65:34:02, 5.71s/it]
6%|▌ | 2571/43939 [4:11:27<65:32:38, 5.70s/it]
6%|▌ | 2572/43939 [4:11:32<65:25:41, 5.69s/it]
6%|▌ | 2573/43939 [4:11:38<65:29:18, 5.70s/it]
6%|▌ | 2574/43939 [4:11:44<66:19:57, 5.77s/it]
6%|▌ | 2575/43939 [4:11:50<66:10:35, 5.76s/it]
6%|▌ | 2576/43939 [4:11:55<66:02:21, 5.75s/it]
6%|▌ | 2577/43939 [4:12:01<65:43:25, 5.72s/it]
6%|▌ | 2578/43939 [4:12:07<65:41:06, 5.72s/it]
6%|▌ | 2579/43939 [4:12:12<65:35:31, 5.71s/it]
6%|▌ | 2580/43939 [4:12:18<65:34:23, 5.71s/it]
6%|▌ | 2581/43939 [4:12:24<65:34:43, 5.71s/it]
6%|▌ | 2582/43939 [4:12:30<65:36:20, 5.71s/it]
6%|▌ | 2583/43939 [4:12:35<65:42:39, 5.72s/it]
6%|▌ | 2584/43939 [4:12:41<65:36:14, 5.71s/it]
6%|▌ | 2585/43939 [4:12:47<65:33:56, 5.71s/it]
6%|▌ | 2586/43939 [4:12:52<65:34:13, 5.71s/it]
6%|▌ | 2587/43939 [4:12:58<65:33:43, 5.71s/it]
6%|▌ | 2588/43939 [4:13:04<65:30:46, 5.70s/it]
6%|▌ | 2589/43939 [4:13:10<65:31:06, 5.70s/it]
6%|▌ | 2590/43939 [4:13:15<65:32:33, 5.71s/it]
6%|▌ | 2591/43939 [4:13:21<65:25:38, 5.70s/it]
6%|▌ | 2592/43939 [4:13:27<65:27:19, 5.70s/it]
6%|▌ | 2593/43939 [4:13:32<65:30:45, 5.70s/it]
6%|▌ | 2594/43939 [4:13:38<65:27:08, 5.70s/it]
6%|▌ | 2595/43939 [4:13:44<66:04:36, 5.75s/it]
6%|▌ | 2596/43939 [4:13:50<65:48:18, 5.73s/it]
6%|▌ | 2597/43939 [4:13:55<65:44:22, 5.72s/it]
6%|▌ | 2598/43939 [4:14:01<65:46:09, 5.73s/it]
6%|▌ | 2599/43939 [4:14:07<65:51:22, 5.73s/it]
6%|▌ | 2600/43939 [4:14:12<65:41:30, 5.72s/it]
{'loss': 1.116, 'grad_norm': 1.5987119674682617, 'learning_rate': 2.9581151832460735e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.22, 'epoch': 0.24}
6%|▌ | 2600/43939 [4:14:12<65:41:30, 5.72s/it]
6%|▌ | 2601/43939 [4:14:18<65:44:41, 5.73s/it]
6%|▌ | 2602/43939 [4:14:24<65:39:35, 5.72s/it]
6%|▌ | 2603/43939 [4:14:30<65:39:17, 5.72s/it]
6%|▌ | 2604/43939 [4:14:35<65:46:17, 5.73s/it]
6%|▌ | 2605/43939 [4:14:41<65:38:42, 5.72s/it]
6%|▌ | 2606/43939 [4:14:47<65:37:17, 5.72s/it]
6%|▌ | 2607/43939 [4:14:52<65:34:54, 5.71s/it]
6%|▌ | 2608/43939 [4:14:58<65:26:36, 5.70s/it]
6%|▌ | 2609/43939 [4:15:04<65:25:31, 5.70s/it]
6%|▌ | 2610/43939 [4:15:10<65:29:33, 5.70s/it]
6%|▌ | 2611/43939 [4:15:15<65:43:27, 5.73s/it]
6%|▌ | 2612/43939 [4:15:21<65:51:51, 5.74s/it]
6%|▌ | 2613/43939 [4:15:27<65:50:07, 5.74s/it]
6%|▌ | 2614/43939 [4:15:33<65:49:51, 5.73s/it]
6%|▌ | 2615/43939 [4:15:38<65:37:17, 5.72s/it]
6%|▌ | 2616/43939 [4:15:44<66:20:23, 5.78s/it]
6%|▌ | 2617/43939 [4:15:50<66:00:36, 5.75s/it]
6%|▌ | 2618/43939 [4:15:56<65:44:58, 5.73s/it]
6%|▌ | 2619/43939 [4:16:01<65:49:00, 5.73s/it]
6%|▌ | 2620/43939 [4:16:07<65:32:35, 5.71s/it]
6%|▌ | 2621/43939 [4:16:13<65:33:29, 5.71s/it]
6%|▌ | 2622/43939 [4:16:18<65:29:39, 5.71s/it]
6%|▌ | 2623/43939 [4:16:24<65:29:55, 5.71s/it]
6%|▌ | 2624/43939 [4:16:30<65:27:55, 5.70s/it]
6%|▌ | 2625/43939 [4:16:35<65:25:02, 5.70s/it]
6%|▌ | 2626/43939 [4:16:41<65:27:06, 5.70s/it]
6%|▌ | 2627/43939 [4:16:47<65:24:41, 5.70s/it]
6%|▌ | 2628/43939 [4:16:53<65:30:07, 5.71s/it]
6%|▌ | 2629/43939 [4:16:58<65:21:00, 5.69s/it]
6%|▌ | 2630/43939 [4:17:04<65:19:28, 5.69s/it]
6%|▌ | 2631/43939 [4:17:10<65:32:13, 5.71s/it]
6%|▌ | 2632/43939 [4:17:15<65:28:09, 5.71s/it]
6%|▌ | 2633/43939 [4:17:21<65:23:50, 5.70s/it]
6%|▌ | 2634/43939 [4:17:27<65:26:26, 5.70s/it]
6%|▌ | 2635/43939 [4:17:32<65:24:40, 5.70s/it]
6%|▌ | 2636/43939 [4:17:38<66:12:02, 5.77s/it]
6%|▌ | 2637/43939 [4:17:44<65:56:10, 5.75s/it]
6%|▌ | 2638/43939 [4:17:50<65:58:13, 5.75s/it]
6%|▌ | 2639/43939 [4:17:56<65:53:37, 5.74s/it]
6%|▌ | 2640/43939 [4:18:01<65:47:49, 5.74s/it]
6%|▌ | 2641/43939 [4:18:07<65:42:56, 5.73s/it]
6%|▌ | 2642/43939 [4:18:13<65:42:37, 5.73s/it]
6%|▌ | 2643/43939 [4:18:18<65:35:30, 5.72s/it]
6%|▌ | 2644/43939 [4:18:24<65:36:13, 5.72s/it]
6%|▌ | 2645/43939 [4:18:30<65:26:02, 5.70s/it]
6%|▌ | 2646/43939 [4:18:36<65:25:04, 5.70s/it]
6%|▌ | 2647/43939 [4:18:41<65:26:49, 5.71s/it]
6%|▌ | 2648/43939 [4:18:47<65:27:32, 5.71s/it]
6%|▌ | 2649/43939 [4:18:53<65:26:41, 5.71s/it]
6%|▌ | 2650/43939 [4:18:58<65:25:11, 5.70s/it]
{'loss': 1.1107, 'grad_norm': 1.2702186107635498, 'learning_rate': 3.0150239016617348e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.9, 'epoch': 0.24}
6%|▌ | 2650/43939 [4:18:58<65:25:11, 5.70s/it]
6%|▌ | 2651/43939 [4:19:04<65:32:00, 5.71s/it]
6%|▌ | 2652/43939 [4:19:10<65:26:10, 5.71s/it]
6%|▌ | 2653/43939 [4:19:15<65:22:19, 5.70s/it]
6%|▌ | 2654/43939 [4:19:21<65:13:18, 5.69s/it]
6%|▌ | 2655/43939 [4:19:27<65:14:02, 5.69s/it]
6%|▌ | 2656/43939 [4:19:32<65:14:43, 5.69s/it]
6%|▌ | 2657/43939 [4:19:38<65:13:24, 5.69s/it]
6%|▌ | 2658/43939 [4:19:44<65:58:26, 5.75s/it]
6%|▌ | 2659/43939 [4:19:50<66:00:48, 5.76s/it]
6%|▌ | 2660/43939 [4:19:56<65:43:48, 5.73s/it]
6%|▌ | 2661/43939 [4:20:01<65:33:09, 5.72s/it]
6%|▌ | 2662/43939 [4:20:07<65:25:10, 5.71s/it]
6%|▌ | 2663/43939 [4:20:13<65:21:01, 5.70s/it]
6%|▌ | 2664/43939 [4:20:18<65:30:57, 5.71s/it]
6%|▌ | 2665/43939 [4:20:24<65:36:51, 5.72s/it]
6%|▌ | 2666/43939 [4:20:30<65:29:30, 5.71s/it]
6%|▌ | 2667/43939 [4:20:35<65:29:05, 5.71s/it]
6%|▌ | 2668/43939 [4:20:41<65:28:34, 5.71s/it]
6%|▌ | 2669/43939 [4:20:47<65:26:47, 5.71s/it]
6%|▌ | 2670/43939 [4:20:53<65:34:49, 5.72s/it]
6%|▌ | 2671/43939 [4:20:58<65:45:51, 5.74s/it]
6%|▌ | 2672/43939 [4:21:04<65:39:40, 5.73s/it]
6%|▌ | 2673/43939 [4:21:10<65:37:54, 5.73s/it]
6%|▌ | 2674/43939 [4:21:16<65:35:37, 5.72s/it]
6%|▌ | 2675/43939 [4:21:21<65:28:28, 5.71s/it]
6%|▌ | 2676/43939 [4:21:27<65:30:36, 5.72s/it]
6%|▌ | 2677/43939 [4:21:33<65:30:05, 5.71s/it]
6%|▌ | 2678/43939 [4:21:39<66:15:35, 5.78s/it]
6%|▌ | 2679/43939 [4:21:44<65:51:17, 5.75s/it]
6%|▌ | 2680/43939 [4:21:50<65:56:15, 5.75s/it]
6%|▌ | 2681/43939 [4:21:56<65:36:45, 5.73s/it]
6%|▌ | 2682/43939 [4:22:01<65:37:21, 5.73s/it]
6%|▌ | 2683/43939 [4:22:07<65:35:16, 5.72s/it]
6%|▌ | 2684/43939 [4:22:13<65:34:13, 5.72s/it]
6%|▌ | 2685/43939 [4:22:19<65:26:04, 5.71s/it]
6%|▌ | 2686/43939 [4:22:24<65:24:31, 5.71s/it]
6%|▌ | 2687/43939 [4:22:30<65:21:06, 5.70s/it]
6%|▌ | 2688/43939 [4:22:36<65:26:25, 5.71s/it]
6%|▌ | 2689/43939 [4:22:41<65:15:41, 5.70s/it]
6%|▌ | 2690/43939 [4:22:47<65:32:57, 5.72s/it]
6%|▌ | 2691/43939 [4:22:53<65:18:57, 5.70s/it]
6%|▌ | 2692/43939 [4:22:58<65:23:40, 5.71s/it]
6%|▌ | 2693/43939 [4:23:04<65:23:09, 5.71s/it]
6%|▌ | 2694/43939 [4:23:10<65:21:26, 5.70s/it]
6%|▌ | 2695/43939 [4:23:16<65:30:11, 5.72s/it]
6%|▌ | 2696/43939 [4:23:21<65:27:42, 5.71s/it]
6%|▌ | 2697/43939 [4:23:27<65:36:22, 5.73s/it]
6%|▌ | 2698/43939 [4:23:33<65:23:00, 5.71s/it]
6%|▌ | 2699/43939 [4:23:39<66:08:51, 5.77s/it]
6%|▌ | 2700/43939 [4:23:44<65:58:02, 5.76s/it]
{'loss': 1.1302, 'grad_norm': 1.4149582386016846, 'learning_rate': 3.071932620077396e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.9, 'epoch': 0.25}
6%|▌ | 2700/43939 [4:23:44<65:58:02, 5.76s/it]
6%|▌ | 2701/43939 [4:23:50<65:44:11, 5.74s/it]
6%|▌ | 2702/43939 [4:23:56<65:28:04, 5.72s/it]
6%|▌ | 2703/43939 [4:24:01<65:33:59, 5.72s/it]
6%|▌ | 2704/43939 [4:24:07<65:24:14, 5.71s/it]
6%|▌ | 2705/43939 [4:24:13<65:23:36, 5.71s/it]
6%|▌ | 2706/43939 [4:24:19<65:23:04, 5.71s/it]
6%|▌ | 2707/43939 [4:24:24<65:15:59, 5.70s/it]
6%|▌ | 2708/43939 [4:24:30<65:19:56, 5.70s/it]
6%|▌ | 2709/43939 [4:24:36<65:17:44, 5.70s/it]
6%|▌ | 2710/43939 [4:24:41<65:12:24, 5.69s/it]
6%|▌ | 2711/43939 [4:24:47<65:16:46, 5.70s/it]
6%|▌ | 2712/43939 [4:24:53<65:09:47, 5.69s/it]
6%|▌ | 2713/43939 [4:24:58<65:12:05, 5.69s/it]
6%|▌ | 2714/43939 [4:25:04<65:19:02, 5.70s/it]
6%|▌ | 2715/43939 [4:25:10<65:17:44, 5.70s/it]
6%|▌ | 2716/43939 [4:25:16<65:23:23, 5.71s/it]
6%|▌ | 2717/43939 [4:25:21<65:27:32, 5.72s/it]
6%|▌ | 2718/43939 [4:25:27<65:26:51, 5.72s/it]
6%|▌ | 2719/43939 [4:25:33<65:32:38, 5.72s/it]
6%|▌ | 2720/43939 [4:25:38<65:28:25, 5.72s/it]
6%|▌ | 2721/43939 [4:25:44<66:16:56, 5.79s/it]
6%|▌ | 2722/43939 [4:25:50<66:05:19, 5.77s/it]
6%|▌ | 2723/43939 [4:25:56<65:48:44, 5.75s/it]
6%|▌ | 2724/43939 [4:26:02<65:32:40, 5.73s/it]
6%|▌ | 2725/43939 [4:26:07<65:25:55, 5.72s/it]
6%|▌ | 2726/43939 [4:26:13<65:19:21, 5.71s/it]
6%|▌ | 2727/43939 [4:26:19<65:09:06, 5.69s/it]
6%|▌ | 2728/43939 [4:26:24<65:05:38, 5.69s/it]
6%|▌ | 2729/43939 [4:26:30<65:00:10, 5.68s/it]
6%|▌ | 2730/43939 [4:26:36<65:02:31, 5.68s/it]
6%|▌ | 2731/43939 [4:26:41<65:06:24, 5.69s/it]
6%|▌ | 2732/43939 [4:26:47<65:08:46, 5.69s/it]
6%|▌ | 2733/43939 [4:26:53<65:18:25, 5.71s/it]
6%|▌ | 2734/43939 [4:26:58<65:12:41, 5.70s/it]
6%|▌ | 2735/43939 [4:27:04<65:15:14, 5.70s/it]
6%|▌ | 2736/43939 [4:27:10<65:17:52, 5.71s/it]
6%|▌ | 2737/43939 [4:27:16<65:22:13, 5.71s/it]
6%|▌ | 2738/43939 [4:27:21<65:19:10, 5.71s/it]
6%|▌ | 2739/43939 [4:27:27<65:20:38, 5.71s/it]
6%|▌ | 2740/43939 [4:27:33<65:18:13, 5.71s/it]
6%|▌ | 2741/43939 [4:27:38<65:18:08, 5.71s/it]
6%|▌ | 2742/43939 [4:27:44<66:16:03, 5.79s/it]
6%|▌ | 2743/43939 [4:27:50<66:03:22, 5.77s/it]
6%|▌ | 2744/43939 [4:27:56<65:59:02, 5.77s/it]
6%|▌ | 2745/43939 [4:28:02<65:50:15, 5.75s/it]
6%|▌ | 2746/43939 [4:28:07<65:42:28, 5.74s/it]
6%|▋ | 2747/43939 [4:28:13<65:41:38, 5.74s/it]
6%|▋ | 2748/43939 [4:28:19<65:32:35, 5.73s/it]
6%|▋ | 2749/43939 [4:28:24<65:39:44, 5.74s/it]
6%|▋ | 2750/43939 [4:28:30<65:25:49, 5.72s/it]
{'loss': 1.1311, 'grad_norm': 3.761045455932617, 'learning_rate': 3.128841338493057e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.72, 'epoch': 0.25}
6%|▋ | 2750/43939 [4:28:30<65:25:49, 5.72s/it]
6%|▋ | 2751/43939 [4:28:36<65:17:50, 5.71s/it]
6%|▋ | 2752/43939 [4:28:42<65:15:13, 5.70s/it]
6%|▋ | 2753/43939 [4:28:47<65:25:20, 5.72s/it]
6%|▋ | 2754/43939 [4:28:53<65:45:55, 5.75s/it]
6%|▋ | 2755/43939 [4:28:59<65:33:17, 5.73s/it]
6%|▋ | 2756/43939 [4:29:04<65:25:35, 5.72s/it]
6%|▋ | 2757/43939 [4:29:10<65:19:24, 5.71s/it]
6%|▋ | 2758/43939 [4:29:16<65:30:25, 5.73s/it]
6%|▋ | 2759/43939 [4:29:22<65:22:45, 5.72s/it]
6%|▋ | 2760/43939 [4:29:27<65:16:54, 5.71s/it]
6%|▋ | 2761/43939 [4:29:33<65:15:25, 5.71s/it]
6%|▋ | 2762/43939 [4:29:39<66:07:19, 5.78s/it]
6%|▋ | 2763/43939 [4:29:45<65:45:48, 5.75s/it]
6%|▋ | 2764/43939 [4:29:50<65:33:43, 5.73s/it]
6%|▋ | 2765/43939 [4:29:56<65:26:47, 5.72s/it]
6%|▋ | 2766/43939 [4:30:02<65:21:43, 5.71s/it]
6%|▋ | 2767/43939 [4:30:07<65:20:11, 5.71s/it]
6%|▋ | 2768/43939 [4:30:13<65:11:36, 5.70s/it]
6%|▋ | 2769/43939 [4:30:19<65:13:19, 5.70s/it]
6%|▋ | 2770/43939 [4:30:25<65:12:41, 5.70s/it]
6%|▋ | 2771/43939 [4:30:30<65:11:35, 5.70s/it]
6%|▋ | 2772/43939 [4:30:36<65:14:34, 5.71s/it]
6%|▋ | 2773/43939 [4:30:42<65:15:04, 5.71s/it]
6%|▋ | 2774/43939 [4:30:47<65:10:38, 5.70s/it]
6%|▋ | 2775/43939 [4:30:53<65:04:05, 5.69s/it]
6%|▋ | 2776/43939 [4:30:59<65:05:38, 5.69s/it]
6%|▋ | 2777/43939 [4:31:04<64:59:34, 5.68s/it]
6%|▋ | 2778/43939 [4:31:10<65:03:01, 5.69s/it]
6%|▋ | 2779/43939 [4:31:16<65:04:31, 5.69s/it]
6%|▋ | 2780/43939 [4:31:21<65:03:50, 5.69s/it]
6%|▋ | 2781/43939 [4:31:27<64:58:30, 5.68s/it]
6%|▋ | 2782/43939 [4:31:33<64:56:45, 5.68s/it]
6%|▋ | 2783/43939 [4:31:38<64:56:37, 5.68s/it]
6%|▋ | 2784/43939 [4:31:44<65:54:22, 5.77s/it]
6%|▋ | 2785/43939 [4:31:50<65:44:55, 5.75s/it]
6%|▋ | 2786/43939 [4:31:56<65:39:19, 5.74s/it]
6%|▋ | 2787/43939 [4:32:02<65:27:14, 5.73s/it]
6%|▋ | 2788/43939 [4:32:07<65:15:18, 5.71s/it]
6%|▋ | 2789/43939 [4:32:13<65:06:14, 5.70s/it]
6%|▋ | 2790/43939 [4:32:19<65:14:54, 5.71s/it]
6%|▋ | 2791/43939 [4:32:24<65:13:46, 5.71s/it]
6%|▋ | 2792/43939 [4:32:30<65:07:14, 5.70s/it]
6%|▋ | 2793/43939 [4:32:36<64:58:39, 5.69s/it]
6%|▋ | 2794/43939 [4:32:41<65:02:31, 5.69s/it]
6%|▋ | 2795/43939 [4:32:47<65:05:16, 5.70s/it]
6%|▋ | 2796/43939 [4:32:53<65:13:15, 5.71s/it]
6%|▋ | 2797/43939 [4:32:58<65:03:53, 5.69s/it]
6%|▋ | 2798/43939 [4:33:04<65:03:03, 5.69s/it]
6%|▋ | 2799/43939 [4:33:10<65:08:42, 5.70s/it]
6%|▋ | 2800/43939 [4:33:16<65:10:29, 5.70s/it]
{'loss': 1.1243, 'grad_norm': 1.2627049684524536, 'learning_rate': 3.185750056908718e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.22, 'epoch': 0.25}
6%|▋ | 2800/43939 [4:33:16<65:10:29, 5.70s/it]
6%|▋ | 2801/43939 [4:33:21<65:11:52, 5.71s/it]
6%|▋ | 2802/43939 [4:33:27<65:12:23, 5.71s/it]
6%|▋ | 2803/43939 [4:33:33<65:07:01, 5.70s/it]
6%|▋ | 2804/43939 [4:33:38<65:00:46, 5.69s/it]
6%|▋ | 2805/43939 [4:33:44<65:57:48, 5.77s/it]
6%|▋ | 2806/43939 [4:33:50<65:48:52, 5.76s/it]
6%|▋ | 2807/43939 [4:33:56<65:34:31, 5.74s/it]
6%|▋ | 2808/43939 [4:34:01<65:20:10, 5.72s/it]
6%|▋ | 2809/43939 [4:34:07<65:12:18, 5.71s/it]
6%|▋ | 2810/43939 [4:34:13<65:15:23, 5.71s/it]
6%|▋ | 2811/43939 [4:34:19<65:14:42, 5.71s/it]
6%|▋ | 2812/43939 [4:34:24<65:13:27, 5.71s/it]
6%|▋ | 2813/43939 [4:34:30<65:07:53, 5.70s/it]
6%|▋ | 2814/43939 [4:34:36<65:09:51, 5.70s/it]
6%|▋ | 2815/43939 [4:34:41<65:11:50, 5.71s/it]
6%|▋ | 2816/43939 [4:34:47<65:09:52, 5.70s/it]
6%|▋ | 2817/43939 [4:34:53<65:07:18, 5.70s/it]
6%|▋ | 2818/43939 [4:34:58<65:09:57, 5.71s/it]
6%|▋ | 2819/43939 [4:35:04<65:09:01, 5.70s/it]
6%|▋ | 2820/43939 [4:35:10<65:11:04, 5.71s/it]
6%|▋ | 2821/43939 [4:35:16<65:10:25, 5.71s/it]
6%|▋ | 2822/43939 [4:35:21<65:04:32, 5.70s/it]
6%|▋ | 2823/43939 [4:35:27<64:59:50, 5.69s/it]
6%|▋ | 2824/43939 [4:35:33<65:01:50, 5.69s/it]
6%|▋ | 2825/43939 [4:35:39<65:48:06, 5.76s/it]
6%|▋ | 2826/43939 [4:35:44<65:35:37, 5.74s/it]
6%|▋ | 2827/43939 [4:35:50<65:17:53, 5.72s/it]
6%|▋ | 2828/43939 [4:35:56<65:10:31, 5.71s/it]
6%|▋ | 2829/43939 [4:36:01<65:11:39, 5.71s/it]
6%|▋ | 2830/43939 [4:36:07<65:09:32, 5.71s/it]
6%|▋ | 2831/43939 [4:36:13<65:20:55, 5.72s/it]
6%|▋ | 2832/43939 [4:36:18<65:12:42, 5.71s/it]
6%|▋ | 2833/43939 [4:36:24<65:09:05, 5.71s/it]
6%|▋ | 2834/43939 [4:36:30<65:01:36, 5.70s/it]
6%|▋ | 2835/43939 [4:36:35<64:53:58, 5.68s/it]
6%|▋ | 2836/43939 [4:36:41<64:55:17, 5.69s/it]
6%|▋ | 2837/43939 [4:36:47<64:51:05, 5.68s/it]
6%|▋ | 2838/43939 [4:36:53<64:54:11, 5.68s/it]
6%|▋ | 2839/43939 [4:36:58<65:06:28, 5.70s/it]
6%|▋ | 2840/43939 [4:37:04<65:07:27, 5.70s/it]
6%|▋ | 2841/43939 [4:37:10<65:10:15, 5.71s/it]
6%|▋ | 2842/43939 [4:37:15<65:08:14, 5.71s/it]
6%|▋ | 2843/43939 [4:37:21<65:07:37, 5.71s/it]
6%|▋ | 2844/43939 [4:37:27<65:04:25, 5.70s/it]
6%|▋ | 2845/43939 [4:37:32<65:01:25, 5.70s/it]
6%|▋ | 2846/43939 [4:37:38<64:58:26, 5.69s/it]
6%|▋ | 2847/43939 [4:37:44<65:35:27, 5.75s/it]
6%|▋ | 2848/43939 [4:37:50<65:27:18, 5.73s/it]
6%|▋ | 2849/43939 [4:37:55<65:18:26, 5.72s/it]
6%|▋ | 2850/43939 [4:38:01<65:18:33, 5.72s/it]
{'loss': 1.1066, 'grad_norm': 1.3173185586929321, 'learning_rate': 3.24265877532438e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.07, 'epoch': 0.26}
6%|▋ | 2850/43939 [4:38:01<65:18:33, 5.72s/it]
6%|▋ | 2851/43939 [4:38:07<65:21:27, 5.73s/it]
6%|▋ | 2852/43939 [4:38:13<65:21:53, 5.73s/it]
6%|▋ | 2853/43939 [4:38:18<65:13:32, 5.72s/it]
6%|▋ | 2854/43939 [4:38:24<65:03:50, 5.70s/it]
6%|▋ | 2855/43939 [4:38:30<65:14:52, 5.72s/it]
6%|▋ | 2856/43939 [4:38:35<65:18:30, 5.72s/it]
7%|▋ | 2857/43939 [4:38:41<65:14:50, 5.72s/it]
7%|▋ | 2858/43939 [4:38:47<65:11:33, 5.71s/it]
7%|▋ | 2859/43939 [4:38:53<65:12:56, 5.72s/it]
7%|▋ | 2860/43939 [4:38:58<65:26:52, 5.74s/it]
7%|▋ | 2861/43939 [4:39:04<65:16:53, 5.72s/it]
7%|▋ | 2862/43939 [4:39:10<65:13:54, 5.72s/it]
7%|▋ | 2863/43939 [4:39:16<65:20:28, 5.73s/it]
7%|▋ | 2864/43939 [4:39:21<65:11:43, 5.71s/it]
7%|▋ | 2865/43939 [4:39:27<65:12:29, 5.72s/it]
7%|▋ | 2866/43939 [4:39:33<65:10:27, 5.71s/it]
7%|▋ | 2867/43939 [4:39:38<65:06:53, 5.71s/it]
7%|▋ | 2868/43939 [4:39:44<65:52:41, 5.77s/it]
7%|▋ | 2869/43939 [4:39:50<65:32:26, 5.74s/it]
7%|▋ | 2870/43939 [4:39:56<65:25:42, 5.74s/it]
7%|▋ | 2871/43939 [4:40:01<65:27:52, 5.74s/it]
7%|▋ | 2872/43939 [4:40:07<65:20:17, 5.73s/it]
7%|▋ | 2873/43939 [4:40:13<65:14:22, 5.72s/it]
7%|▋ | 2874/43939 [4:40:19<65:12:58, 5.72s/it]
7%|▋ | 2875/43939 [4:40:24<65:18:48, 5.73s/it]
7%|▋ | 2876/43939 [4:40:30<65:19:32, 5.73s/it]
7%|▋ | 2877/43939 [4:40:36<65:21:46, 5.73s/it]
7%|▋ | 2878/43939 [4:40:41<65:22:54, 5.73s/it]
7%|▋ | 2879/43939 [4:40:47<65:07:26, 5.71s/it]
7%|▋ | 2880/43939 [4:40:53<65:06:44, 5.71s/it]
7%|▋ | 2881/43939 [4:40:59<65:09:33, 5.71s/it]
7%|▋ | 2882/43939 [4:41:04<65:12:39, 5.72s/it]
7%|▋ | 2883/43939 [4:41:10<65:02:00, 5.70s/it]
7%|▋ | 2884/43939 [4:41:16<65:00:57, 5.70s/it]
7%|▋ | 2885/43939 [4:41:21<65:04:11, 5.71s/it]
7%|▋ | 2886/43939 [4:41:27<65:00:51, 5.70s/it]
7%|▋ | 2887/43939 [4:41:33<65:00:12, 5.70s/it]
7%|▋ | 2888/43939 [4:41:38<64:57:25, 5.70s/it]
7%|▋ | 2889/43939 [4:41:44<65:41:00, 5.76s/it]
7%|▋ | 2890/43939 [4:41:50<65:27:31, 5.74s/it]
7%|▋ | 2891/43939 [4:41:56<65:27:52, 5.74s/it]
7%|▋ | 2892/43939 [4:42:01<65:19:41, 5.73s/it]
7%|▋ | 2893/43939 [4:42:07<65:11:43, 5.72s/it]
7%|▋ | 2894/43939 [4:42:13<65:14:11, 5.72s/it]
7%|▋ | 2895/43939 [4:42:19<65:04:08, 5.71s/it]
7%|▋ | 2896/43939 [4:42:24<65:02:32, 5.71s/it]
7%|▋ | 2897/43939 [4:42:30<65:07:56, 5.71s/it]
7%|▋ | 2898/43939 [4:42:36<65:07:57, 5.71s/it]
7%|▋ | 2899/43939 [4:42:41<65:07:09, 5.71s/it]
7%|▋ | 2900/43939 [4:42:47<65:00:08, 5.70s/it]
{'loss': 1.1183, 'grad_norm': 1.4638679027557373, 'learning_rate': 3.2995674937400414e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.32, 'epoch': 0.26}
7%|▋ | 2900/43939 [4:42:47<65:00:08, 5.70s/it]
7%|▋ | 2901/43939 [4:42:53<65:16:41, 5.73s/it]
7%|▋ | 2902/43939 [4:42:59<65:09:18, 5.72s/it]
7%|▋ | 2903/43939 [4:43:04<64:57:39, 5.70s/it]
7%|▋ | 2904/43939 [4:43:10<64:54:56, 5.70s/it]
7%|▋ | 2905/43939 [4:43:16<64:58:32, 5.70s/it]
7%|▋ | 2906/43939 [4:43:21<65:02:40, 5.71s/it]
7%|▋ | 2907/43939 [4:43:27<65:08:27, 5.72s/it]
7%|▋ | 2908/43939 [4:43:33<65:02:52, 5.71s/it]
7%|▋ | 2909/43939 [4:43:39<65:59:15, 5.79s/it]
7%|▋ | 2910/43939 [4:43:44<65:46:04, 5.77s/it]
7%|▋ | 2911/43939 [4:43:50<65:36:43, 5.76s/it]
7%|▋ | 2912/43939 [4:43:56<65:20:13, 5.73s/it]
7%|▋ | 2913/43939 [4:44:02<65:13:40, 5.72s/it]
7%|▋ | 2914/43939 [4:44:07<65:03:44, 5.71s/it]
7%|▋ | 2915/43939 [4:44:13<65:00:04, 5.70s/it]
7%|▋ | 2916/43939 [4:44:19<64:59:27, 5.70s/it]
7%|▋ | 2917/43939 [4:44:24<64:58:49, 5.70s/it]
7%|▋ | 2918/43939 [4:44:30<65:00:41, 5.71s/it]
7%|▋ | 2919/43939 [4:44:36<65:04:06, 5.71s/it]
7%|▋ | 2920/43939 [4:44:42<65:02:19, 5.71s/it]
7%|▋ | 2921/43939 [4:44:47<65:03:40, 5.71s/it]
7%|▋ | 2922/43939 [4:44:53<65:15:41, 5.73s/it]
7%|▋ | 2923/43939 [4:44:59<65:06:32, 5.71s/it]
7%|▋ | 2924/43939 [4:45:04<65:00:07, 5.71s/it]
7%|▋ | 2925/43939 [4:45:10<64:59:25, 5.70s/it]
7%|▋ | 2926/43939 [4:45:16<64:57:00, 5.70s/it]
7%|▋ | 2927/43939 [4:45:21<64:59:13, 5.70s/it]
7%|▋ | 2928/43939 [4:45:27<64:48:43, 5.69s/it]
7%|▋ | 2929/43939 [4:45:33<64:44:06, 5.68s/it]
7%|▋ | 2930/43939 [4:45:38<64:47:50, 5.69s/it]
7%|▋ | 2931/43939 [4:45:44<65:41:12, 5.77s/it]
7%|▋ | 2932/43939 [4:45:50<65:20:29, 5.74s/it]
7%|▋ | 2933/43939 [4:45:56<65:12:33, 5.72s/it]
7%|▋ | 2934/43939 [4:46:01<65:05:28, 5.71s/it]
7%|▋ | 2935/43939 [4:46:07<65:01:14, 5.71s/it]
7%|▋ | 2936/43939 [4:46:13<65:05:10, 5.71s/it]
7%|▋ | 2937/43939 [4:46:19<65:10:19, 5.72s/it]
7%|▋ | 2938/43939 [4:46:24<65:04:23, 5.71s/it]
7%|▋ | 2939/43939 [4:46:30<65:06:02, 5.72s/it]
7%|▋ | 2940/43939 [4:46:36<65:03:11, 5.71s/it]
7%|▋ | 2941/43939 [4:46:41<65:01:00, 5.71s/it]
7%|▋ | 2942/43939 [4:46:47<64:59:14, 5.71s/it]
7%|▋ | 2943/43939 [4:46:53<65:02:07, 5.71s/it]
7%|▋ | 2944/43939 [4:46:59<65:21:49, 5.74s/it]
7%|▋ | 2945/43939 [4:47:04<65:17:22, 5.73s/it]
7%|▋ | 2946/43939 [4:47:10<65:06:17, 5.72s/it]
7%|▋ | 2947/43939 [4:47:16<65:06:44, 5.72s/it]
7%|▋ | 2948/43939 [4:47:22<64:58:58, 5.71s/it]
7%|▋ | 2949/43939 [4:47:27<65:11:28, 5.73s/it]
7%|▋ | 2950/43939 [4:47:33<65:09:01, 5.72s/it]
{'loss': 1.0697, 'grad_norm': 1.3660715818405151, 'learning_rate': 3.3564762121557024e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.36, 'epoch': 0.27}
7%|▋ | 2950/43939 [4:47:33<65:09:01, 5.72s/it]
7%|▋ | 2951/43939 [4:47:39<65:55:25, 5.79s/it]
7%|▋ | 2952/43939 [4:47:45<65:40:55, 5.77s/it]
7%|▋ | 2953/43939 [4:47:50<65:27:33, 5.75s/it]
7%|▋ | 2954/43939 [4:47:56<65:15:35, 5.73s/it]
7%|▋ | 2955/43939 [4:48:02<65:03:23, 5.71s/it]
7%|▋ | 2956/43939 [4:48:07<64:54:03, 5.70s/it]
7%|▋ | 2957/43939 [4:48:13<64:48:12, 5.69s/it]
7%|▋ | 2958/43939 [4:48:19<64:50:38, 5.70s/it]
7%|▋ | 2959/43939 [4:48:24<64:53:17, 5.70s/it]
7%|▋ | 2960/43939 [4:48:30<64:58:00, 5.71s/it]
7%|▋ | 2961/43939 [4:48:36<64:51:01, 5.70s/it]
7%|▋ | 2962/43939 [4:48:42<64:46:47, 5.69s/it]
7%|▋ | 2963/43939 [4:48:47<64:55:29, 5.70s/it]
7%|▋ | 2964/43939 [4:48:53<64:56:34, 5.71s/it]
7%|▋ | 2965/43939 [4:48:59<64:59:00, 5.71s/it]
7%|▋ | 2966/43939 [4:49:04<64:50:05, 5.70s/it]
7%|▋ | 2967/43939 [4:49:10<64:47:21, 5.69s/it]
7%|▋ | 2968/43939 [4:49:16<65:05:14, 5.72s/it]
7%|▋ | 2969/43939 [4:49:22<65:09:29, 5.73s/it]
7%|▋ | 2970/43939 [4:49:27<65:08:28, 5.72s/it]
7%|▋ | 2971/43939 [4:49:33<65:00:38, 5.71s/it]
7%|▋ | 2972/43939 [4:49:39<65:44:45, 5.78s/it]
7%|▋ | 2973/43939 [4:49:45<65:51:59, 5.79s/it]
7%|▋ | 2974/43939 [4:49:50<65:35:56, 5.76s/it]
7%|▋ | 2975/43939 [4:49:56<65:22:16, 5.74s/it]
7%|▋ | 2976/43939 [4:50:02<65:10:47, 5.73s/it]
7%|▋ | 2977/43939 [4:50:08<65:18:30, 5.74s/it]
7%|▋ | 2978/43939 [4:50:13<65:20:25, 5.74s/it]
7%|▋ | 2979/43939 [4:50:19<65:10:37, 5.73s/it]
7%|▋ | 2980/43939 [4:50:25<65:05:54, 5.72s/it]
7%|▋ | 2981/43939 [4:50:30<65:03:16, 5.72s/it]
7%|▋ | 2982/43939 [4:50:36<64:57:15, 5.71s/it]
7%|▋ | 2983/43939 [4:50:42<64:49:15, 5.70s/it]
7%|▋ | 2984/43939 [4:50:48<64:46:37, 5.69s/it]
7%|▋ | 2985/43939 [4:50:53<64:45:28, 5.69s/it]
7%|▋ | 2986/43939 [4:50:59<64:41:22, 5.69s/it]
7%|▋ | 2987/43939 [4:51:05<64:36:38, 5.68s/it]
7%|▋ | 2988/43939 [4:51:10<64:30:33, 5.67s/it]
7%|▋ | 2989/43939 [4:51:16<64:27:06, 5.67s/it]
7%|▋ | 2990/43939 [4:51:22<64:42:04, 5.69s/it]
7%|▋ | 2991/43939 [4:51:27<64:37:56, 5.68s/it]
7%|▋ | 2992/43939 [4:51:33<64:45:49, 5.69s/it]
7%|▋ | 2993/43939 [4:51:39<64:47:13, 5.70s/it]
7%|▋ | 2994/43939 [4:51:45<65:35:55, 5.77s/it]
7%|▋ | 2995/43939 [4:51:50<65:19:24, 5.74s/it]
7%|▋ | 2996/43939 [4:51:56<65:18:55, 5.74s/it]
7%|▋ | 2997/43939 [4:52:02<65:03:37, 5.72s/it]
7%|▋ | 2998/43939 [4:52:07<64:59:01, 5.71s/it]
7%|▋ | 2999/43939 [4:52:13<65:03:05, 5.72s/it]
7%|▋ | 3000/43939 [4:52:19<64:59:05, 5.71s/it]
{'loss': 1.1047, 'grad_norm': 1.175022840499878, 'learning_rate': 3.413384930571364e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.11, 'epoch': 0.27}
7%|▋ | 3000/43939 [4:52:19<64:59:05, 5.71s/it]
7%|▋ | 3001/43939 [4:52:25<64:51:32, 5.70s/it]
7%|▋ | 3002/43939 [4:52:30<64:45:37, 5.70s/it]
7%|▋ | 3003/43939 [4:52:36<64:56:09, 5.71s/it]
7%|▋ | 3004/43939 [4:52:42<65:07:19, 5.73s/it]
7%|▋ | 3005/43939 [4:52:47<65:01:53, 5.72s/it]
7%|▋ | 3006/43939 [4:52:53<64:59:19, 5.72s/it]
7%|▋ | 3007/43939 [4:52:59<64:59:52, 5.72s/it]
7%|▋ | 3008/43939 [4:53:05<64:59:38, 5.72s/it]
7%|▋ | 3009/43939 [4:53:10<64:53:19, 5.71s/it]
7%|▋ | 3010/43939 [4:53:16<64:51:33, 5.70s/it]
7%|▋ | 3011/43939 [4:53:22<64:45:14, 5.70s/it]
7%|▋ | 3012/43939 [4:53:27<64:49:04, 5.70s/it]
7%|▋ | 3013/43939 [4:53:33<64:43:46, 5.69s/it]
7%|▋ | 3014/43939 [4:53:39<65:38:53, 5.77s/it]
7%|▋ | 3015/43939 [4:53:45<65:25:45, 5.76s/it]
7%|▋ | 3016/43939 [4:53:50<65:14:27, 5.74s/it]
7%|▋ | 3017/43939 [4:53:56<65:06:14, 5.73s/it]
7%|▋ | 3018/43939 [4:54:02<65:02:20, 5.72s/it]
7%|▋ | 3019/43939 [4:54:08<65:03:04, 5.72s/it]
7%|▋ | 3020/43939 [4:54:13<64:54:29, 5.71s/it]
7%|▋ | 3021/43939 [4:54:19<65:10:31, 5.73s/it]
7%|▋ | 3022/43939 [4:54:25<64:56:11, 5.71s/it]
7%|▋ | 3023/43939 [4:54:30<64:49:41, 5.70s/it]
7%|▋ | 3024/43939 [4:54:36<64:53:43, 5.71s/it]
7%|▋ | 3025/43939 [4:54:42<64:54:34, 5.71s/it]
7%|▋ | 3026/43939 [4:54:47<64:57:17, 5.72s/it]
7%|▋ | 3027/43939 [4:54:53<64:51:54, 5.71s/it]
7%|▋ | 3028/43939 [4:54:59<64:54:32, 5.71s/it]
7%|▋ | 3029/43939 [4:55:05<65:01:13, 5.72s/it]
7%|▋ | 3030/43939 [4:55:10<64:59:27, 5.72s/it]
7%|▋ | 3031/43939 [4:55:16<65:02:20, 5.72s/it]
7%|▋ | 3032/43939 [4:55:22<65:03:15, 5.73s/it]
7%|▋ | 3033/43939 [4:55:28<64:54:25, 5.71s/it]
7%|▋ | 3034/43939 [4:55:33<64:49:32, 5.71s/it]
7%|▋ | 3035/43939 [4:55:39<65:28:23, 5.76s/it]
7%|▋ | 3036/43939 [4:55:45<65:10:50, 5.74s/it]
7%|▋ | 3037/43939 [4:55:50<64:55:21, 5.71s/it]
7%|▋ | 3038/43939 [4:55:56<65:04:56, 5.73s/it]
7%|▋ | 3039/43939 [4:56:02<64:58:48, 5.72s/it]
7%|▋ | 3040/43939 [4:56:08<64:51:29, 5.71s/it]
7%|▋ | 3041/43939 [4:56:13<64:50:27, 5.71s/it]
7%|▋ | 3042/43939 [4:56:19<64:52:12, 5.71s/it]
7%|▋ | 3043/43939 [4:56:25<64:49:13, 5.71s/it]
7%|▋ | 3044/43939 [4:56:30<64:41:14, 5.69s/it]
7%|▋ | 3045/43939 [4:56:36<64:45:07, 5.70s/it]
7%|▋ | 3046/43939 [4:56:42<64:47:16, 5.70s/it]
7%|▋ | 3047/43939 [4:56:48<64:51:26, 5.71s/it]
7%|▋ | 3048/43939 [4:56:53<64:43:52, 5.70s/it]
7%|▋ | 3049/43939 [4:56:59<64:50:57, 5.71s/it]
7%|▋ | 3050/43939 [4:57:05<64:51:14, 5.71s/it]
{'loss': 1.1115, 'grad_norm': 1.247564673423767, 'learning_rate': 3.470293648987025e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.44, 'epoch': 0.28}
7%|▋ | 3050/43939 [4:57:05<64:51:14, 5.71s/it]
7%|▋ | 3051/43939 [4:57:10<64:54:19, 5.71s/it]
7%|▋ | 3052/43939 [4:57:16<64:56:51, 5.72s/it]
7%|▋ | 3053/43939 [4:57:22<64:51:18, 5.71s/it]
7%|▋ | 3054/43939 [4:57:27<64:45:00, 5.70s/it]
7%|▋ | 3055/43939 [4:57:33<64:44:25, 5.70s/it]
7%|▋ | 3056/43939 [4:57:39<64:48:17, 5.71s/it]
7%|▋ | 3057/43939 [4:57:45<65:35:58, 5.78s/it]
7%|▋ | 3058/43939 [4:57:51<65:27:50, 5.76s/it]
7%|▋ | 3059/43939 [4:57:56<65:19:26, 5.75s/it]
7%|▋ | 3060/43939 [4:58:02<65:02:45, 5.73s/it]
7%|▋ | 3061/43939 [4:58:08<64:57:19, 5.72s/it]
7%|▋ | 3062/43939 [4:58:13<64:44:36, 5.70s/it]
7%|▋ | 3063/43939 [4:58:19<64:40:01, 5.70s/it]
7%|▋ | 3064/43939 [4:58:25<64:45:49, 5.70s/it]
7%|▋ | 3065/43939 [4:58:30<64:39:39, 5.70s/it]
7%|▋ | 3066/43939 [4:58:36<64:34:46, 5.69s/it]
7%|▋ | 3067/43939 [4:58:42<64:33:19, 5.69s/it]
7%|▋ | 3068/43939 [4:58:47<64:37:43, 5.69s/it]
7%|▋ | 3069/43939 [4:58:53<64:37:19, 5.69s/it]
7%|▋ | 3070/43939 [4:58:59<64:34:22, 5.69s/it]
7%|▋ | 3071/43939 [4:59:05<64:39:33, 5.70s/it]
7%|▋ | 3072/43939 [4:59:10<64:55:54, 5.72s/it]
7%|▋ | 3073/43939 [4:59:16<64:48:14, 5.71s/it]
7%|▋ | 3074/43939 [4:59:22<64:51:07, 5.71s/it]
7%|▋ | 3075/43939 [4:59:27<64:49:05, 5.71s/it]
7%|▋ | 3076/43939 [4:59:33<64:49:59, 5.71s/it]
7%|▋ | 3077/43939 [4:59:39<64:48:39, 5.71s/it]
7%|▋ | 3078/43939 [4:59:45<65:32:33, 5.77s/it]
7%|▋ | 3079/43939 [4:59:50<65:16:22, 5.75s/it]
7%|▋ | 3080/43939 [4:59:56<65:08:02, 5.74s/it]
7%|▋ | 3081/43939 [5:00:02<65:00:59, 5.73s/it]
7%|▋ | 3082/43939 [5:00:08<64:56:47, 5.72s/it]
7%|▋ | 3083/43939 [5:00:13<64:53:23, 5.72s/it]
7%|▋ | 3084/43939 [5:00:19<64:49:34, 5.71s/it]
7%|▋ | 3085/43939 [5:00:25<64:51:00, 5.71s/it]
7%|▋ | 3086/43939 [5:00:30<64:38:46, 5.70s/it]
7%|▋ | 3087/43939 [5:00:36<64:36:21, 5.69s/it]
7%|▋ | 3088/43939 [5:00:42<64:34:24, 5.69s/it]
7%|▋ | 3089/43939 [5:00:47<64:35:31, 5.69s/it]
7%|▋ | 3090/43939 [5:00:53<64:41:21, 5.70s/it]
7%|▋ | 3091/43939 [5:00:59<64:33:52, 5.69s/it]
7%|▋ | 3092/43939 [5:01:05<64:34:30, 5.69s/it]
7%|▋ | 3093/43939 [5:01:10<64:30:02, 5.68s/it]
7%|▋ | 3094/43939 [5:01:16<64:27:58, 5.68s/it]
7%|▋ | 3095/43939 [5:01:22<64:28:46, 5.68s/it]
7%|▋ | 3096/43939 [5:01:27<64:41:03, 5.70s/it]
7%|▋ | 3097/43939 [5:01:33<64:43:26, 5.71s/it]
7%|▋ | 3098/43939 [5:01:39<65:18:48, 5.76s/it]
7%|▋ | 3099/43939 [5:01:45<65:10:17, 5.74s/it]
7%|▋ | 3100/43939 [5:01:50<65:00:21, 5.73s/it]
{'loss': 1.1257, 'grad_norm': 1.1579777002334595, 'learning_rate': 3.527202367402686e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.81, 'epoch': 0.28}
7%|▋ | 3100/43939 [5:01:50<65:00:21, 5.73s/it]
7%|▋ | 3101/43939 [5:01:56<64:55:58, 5.72s/it]
7%|▋ | 3102/43939 [5:02:02<64:52:01, 5.72s/it]
7%|▋ | 3103/43939 [5:02:07<64:45:36, 5.71s/it]
7%|▋ | 3104/43939 [5:02:13<65:03:28, 5.74s/it]
7%|▋ | 3105/43939 [5:02:19<65:00:32, 5.73s/it]
7%|▋ | 3106/43939 [5:02:25<64:53:55, 5.72s/it]
7%|▋ | 3107/43939 [5:02:30<64:50:08, 5.72s/it]
7%|▋ | 3108/43939 [5:02:36<64:48:55, 5.71s/it]
7%|▋ | 3109/43939 [5:02:42<64:50:46, 5.72s/it]
7%|▋ | 3110/43939 [5:02:47<64:47:01, 5.71s/it]
7%|▋ | 3111/43939 [5:02:53<64:53:58, 5.72s/it]
7%|▋ | 3112/43939 [5:02:59<64:39:56, 5.70s/it]
7%|▋ | 3113/43939 [5:03:05<64:37:35, 5.70s/it]
7%|▋ | 3114/43939 [5:03:10<64:29:15, 5.69s/it]
7%|▋ | 3115/43939 [5:03:16<64:33:05, 5.69s/it]
7%|▋ | 3116/43939 [5:03:22<64:36:14, 5.70s/it]
7%|▋ | 3117/43939 [5:03:27<64:30:42, 5.69s/it]
7%|▋ | 3118/43939 [5:03:33<64:22:51, 5.68s/it]
7%|▋ | 3119/43939 [5:03:39<65:19:04, 5.76s/it]
7%|▋ | 3120/43939 [5:03:45<65:12:07, 5.75s/it]
7%|▋ | 3121/43939 [5:03:50<64:59:50, 5.73s/it]
7%|▋ | 3122/43939 [5:03:56<64:52:26, 5.72s/it]
7%|▋ | 3123/43939 [5:04:02<64:46:55, 5.71s/it]
7%|▋ | 3124/43939 [5:04:07<64:54:05, 5.72s/it]
7%|▋ | 3125/43939 [5:04:13<64:45:54, 5.71s/it]
7%|▋ | 3126/43939 [5:04:19<64:46:49, 5.71s/it]
7%|▋ | 3127/43939 [5:04:25<64:46:18, 5.71s/it]
7%|▋ | 3128/43939 [5:04:30<64:38:32, 5.70s/it]
7%|▋ | 3129/43939 [5:04:36<64:37:00, 5.70s/it]
7%|▋ | 3130/43939 [5:04:42<64:39:34, 5.70s/it]
7%|▋ | 3131/43939 [5:04:47<64:36:07, 5.70s/it]
7%|▋ | 3132/43939 [5:04:53<64:37:04, 5.70s/it]
7%|▋ | 3133/43939 [5:04:59<64:37:20, 5.70s/it]
7%|▋ | 3134/43939 [5:05:04<64:37:32, 5.70s/it]
7%|▋ | 3135/43939 [5:05:10<64:41:51, 5.71s/it]
7%|▋ | 3136/43939 [5:05:16<64:37:46, 5.70s/it]
7%|▋ | 3137/43939 [5:05:22<64:38:24, 5.70s/it]
7%|▋ | 3138/43939 [5:05:27<64:37:24, 5.70s/it]
7%|▋ | 3139/43939 [5:05:33<64:39:55, 5.71s/it]
7%|▋ | 3140/43939 [5:05:39<64:35:02, 5.70s/it]
7%|▋ | 3141/43939 [5:05:45<65:22:32, 5.77s/it]
7%|▋ | 3142/43939 [5:05:50<65:07:50, 5.75s/it]
7%|▋ | 3143/43939 [5:05:56<64:52:39, 5.73s/it]
7%|▋ | 3144/43939 [5:06:02<64:49:00, 5.72s/it]
7%|▋ | 3145/43939 [5:06:07<64:46:22, 5.72s/it]
7%|▋ | 3146/43939 [5:06:13<64:46:18, 5.72s/it]
7%|▋ | 3147/43939 [5:06:19<64:50:08, 5.72s/it]
7%|▋ | 3148/43939 [5:06:25<64:49:54, 5.72s/it]
7%|▋ | 3149/43939 [5:06:30<64:51:24, 5.72s/it]
7%|▋ | 3150/43939 [5:06:36<64:45:34, 5.72s/it]
{'loss': 1.1078, 'grad_norm': 1.1780682802200317, 'learning_rate': 3.584111085818348e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.72, 'epoch': 0.29}
7%|▋ | 3150/43939 [5:06:36<64:45:34, 5.72s/it]
7%|▋ | 3151/43939 [5:06:42<64:54:41, 5.73s/it]
7%|▋ | 3152/43939 [5:06:47<64:46:29, 5.72s/it]
7%|▋ | 3153/43939 [5:06:53<64:46:49, 5.72s/it]
7%|▋ | 3154/43939 [5:06:59<64:57:02, 5.73s/it]
7%|▋ | 3155/43939 [5:07:05<64:48:24, 5.72s/it]
7%|▋ | 3156/43939 [5:07:10<64:53:02, 5.73s/it]
7%|▋ | 3157/43939 [5:07:16<64:56:55, 5.73s/it]
7%|▋ | 3158/43939 [5:07:22<64:52:58, 5.73s/it]
7%|▋ | 3159/43939 [5:07:28<64:53:12, 5.73s/it]
7%|▋ | 3160/43939 [5:07:33<64:47:37, 5.72s/it]
7%|▋ | 3161/43939 [5:07:39<65:23:16, 5.77s/it]
7%|▋ | 3162/43939 [5:07:45<65:09:40, 5.75s/it]
7%|▋ | 3163/43939 [5:07:51<64:57:37, 5.74s/it]
7%|▋ | 3164/43939 [5:07:56<64:49:59, 5.72s/it]
7%|▋ | 3165/43939 [5:08:02<64:39:06, 5.71s/it]
7%|▋ | 3166/43939 [5:08:08<64:30:44, 5.70s/it]
7%|▋ | 3167/43939 [5:08:13<64:33:06, 5.70s/it]
7%|▋ | 3168/43939 [5:08:19<64:34:17, 5.70s/it]
7%|▋ | 3169/43939 [5:08:25<64:24:34, 5.69s/it]
7%|▋ | 3170/43939 [5:08:30<64:33:45, 5.70s/it]
7%|▋ | 3171/43939 [5:08:36<64:33:45, 5.70s/it]
7%|▋ | 3172/43939 [5:08:42<64:32:10, 5.70s/it]
7%|▋ | 3173/43939 [5:08:47<64:30:24, 5.70s/it]
7%|▋ | 3174/43939 [5:08:53<64:28:03, 5.69s/it]
7%|▋ | 3175/43939 [5:08:59<64:28:17, 5.69s/it]
7%|▋ | 3176/43939 [5:09:05<64:29:27, 5.70s/it]
7%|▋ | 3177/43939 [5:09:10<64:36:17, 5.71s/it]
7%|▋ | 3178/43939 [5:09:16<64:33:45, 5.70s/it]
7%|▋ | 3179/43939 [5:09:22<64:25:49, 5.69s/it]
7%|▋ | 3180/43939 [5:09:27<64:36:08, 5.71s/it]
7%|▋ | 3181/43939 [5:09:33<64:38:03, 5.71s/it]
7%|▋ | 3182/43939 [5:09:39<65:32:29, 5.79s/it]
7%|▋ | 3183/43939 [5:09:45<65:23:53, 5.78s/it]
7%|▋ | 3184/43939 [5:09:50<65:04:48, 5.75s/it]
7%|▋ | 3185/43939 [5:09:56<64:53:18, 5.73s/it]
7%|▋ | 3186/43939 [5:10:02<64:40:08, 5.71s/it]
7%|▋ | 3187/43939 [5:10:08<64:35:45, 5.71s/it]
7%|▋ | 3188/43939 [5:10:13<64:38:22, 5.71s/it]
7%|▋ | 3189/43939 [5:10:19<64:38:31, 5.71s/it]
7%|▋ | 3190/43939 [5:10:25<64:30:46, 5.70s/it]
7%|▋ | 3191/43939 [5:10:30<64:24:01, 5.69s/it]
7%|▋ | 3192/43939 [5:10:36<64:24:51, 5.69s/it]
7%|▋ | 3193/43939 [5:10:42<64:31:05, 5.70s/it]
7%|▋ | 3194/43939 [5:10:47<64:29:20, 5.70s/it]
7%|▋ | 3195/43939 [5:10:53<64:30:41, 5.70s/it]
7%|▋ | 3196/43939 [5:10:59<64:25:16, 5.69s/it]
7%|▋ | 3197/43939 [5:11:04<64:26:48, 5.69s/it]
7%|▋ | 3198/43939 [5:11:10<64:22:38, 5.69s/it]
7%|▋ | 3199/43939 [5:11:16<64:28:02, 5.70s/it]
7%|▋ | 3200/43939 [5:11:22<64:24:03, 5.69s/it]
{'loss': 1.1464, 'grad_norm': 1.224279761314392, 'learning_rate': 3.641019804234009e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 524.15, 'epoch': 0.29}
7%|▋ | 3200/43939 [5:11:22<64:24:03, 5.69s/it]
7%|▋ | 3201/43939 [5:11:27<64:38:51, 5.71s/it]
7%|▋ | 3202/43939 [5:11:33<64:34:39, 5.71s/it]
7%|▋ | 3203/43939 [5:11:39<65:34:21, 5.79s/it]
7%|▋ | 3204/43939 [5:11:45<65:24:57, 5.78s/it]
7%|▋ | 3205/43939 [5:11:51<65:19:18, 5.77s/it]
7%|▋ | 3206/43939 [5:11:56<65:06:55, 5.75s/it]
7%|▋ | 3207/43939 [5:12:02<64:47:37, 5.73s/it]
7%|▋ | 3208/43939 [5:12:08<64:48:29, 5.73s/it]
7%|▋ | 3209/43939 [5:12:13<64:35:53, 5.71s/it]
7%|▋ | 3210/43939 [5:12:19<64:44:32, 5.72s/it]
7%|▋ | 3211/43939 [5:12:25<64:40:32, 5.72s/it]
7%|▋ | 3212/43939 [5:12:31<64:54:18, 5.74s/it]
7%|▋ | 3213/43939 [5:12:36<64:47:59, 5.73s/it]
7%|▋ | 3214/43939 [5:12:42<64:46:56, 5.73s/it]
7%|▋ | 3215/43939 [5:12:48<64:41:18, 5.72s/it]
7%|▋ | 3216/43939 [5:12:53<64:32:02, 5.70s/it]
7%|▋ | 3217/43939 [5:12:59<64:33:16, 5.71s/it]
7%|▋ | 3218/43939 [5:13:05<64:27:42, 5.70s/it]
7%|▋ | 3219/43939 [5:13:10<64:21:16, 5.69s/it]
7%|▋ | 3220/43939 [5:13:16<64:14:34, 5.68s/it]
7%|▋ | 3221/43939 [5:13:22<64:27:09, 5.70s/it]
7%|▋ | 3222/43939 [5:13:27<64:27:16, 5.70s/it]
7%|▋ | 3223/43939 [5:13:33<64:23:15, 5.69s/it]
7%|▋ | 3224/43939 [5:13:39<65:12:37, 5.77s/it]
7%|▋ | 3225/43939 [5:13:45<65:05:10, 5.76s/it]
7%|▋ | 3226/43939 [5:13:51<64:47:44, 5.73s/it]
7%|▋ | 3227/43939 [5:13:56<64:36:07, 5.71s/it]
7%|▋ | 3228/43939 [5:14:02<64:27:15, 5.70s/it]
7%|▋ | 3229/43939 [5:14:08<64:24:37, 5.70s/it]
7%|▋ | 3230/43939 [5:14:13<64:39:22, 5.72s/it]
7%|▋ | 3231/43939 [5:14:19<64:40:14, 5.72s/it]
7%|▋ | 3232/43939 [5:14:25<64:36:29, 5.71s/it]
7%|▋ | 3233/43939 [5:14:30<64:25:29, 5.70s/it]
7%|▋ | 3234/43939 [5:14:36<64:31:28, 5.71s/it]
7%|▋ | 3235/43939 [5:14:42<64:28:08, 5.70s/it]
7%|▋ | 3236/43939 [5:14:48<64:27:15, 5.70s/it]
7%|▋ | 3237/43939 [5:14:53<64:31:12, 5.71s/it]
7%|▋ | 3238/43939 [5:14:59<64:25:37, 5.70s/it]
7%|▋ | 3239/43939 [5:15:05<64:27:54, 5.70s/it]
7%|▋ | 3240/43939 [5:15:10<64:25:00, 5.70s/it]
7%|▋ | 3241/43939 [5:15:16<64:30:23, 5.71s/it]
7%|▋ | 3242/43939 [5:15:22<64:40:06, 5.72s/it]
7%|▋ | 3243/43939 [5:15:27<64:36:35, 5.72s/it]
7%|▋ | 3244/43939 [5:15:33<65:21:03, 5.78s/it]
7%|▋ | 3245/43939 [5:15:39<65:08:13, 5.76s/it]
7%|▋ | 3246/43939 [5:15:45<64:53:48, 5.74s/it]
7%|▋ | 3247/43939 [5:15:51<64:46:53, 5.73s/it]
7%|▋ | 3248/43939 [5:15:56<64:38:34, 5.72s/it]
7%|▋ | 3249/43939 [5:16:02<64:35:14, 5.71s/it]
7%|▋ | 3250/43939 [5:16:08<64:31:05, 5.71s/it]
{'loss': 1.0947, 'grad_norm': 1.2794740200042725, 'learning_rate': 3.6979285226496704e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 517.16, 'epoch': 0.3}
7%|▋ | 3250/43939 [5:16:08<64:31:05, 5.71s/it]
7%|▋ | 3251/43939 [5:16:13<64:25:40, 5.70s/it]
7%|▋ | 3252/43939 [5:16:19<64:17:49, 5.69s/it]
7%|▋ | 3253/43939 [5:16:25<64:14:43, 5.68s/it]
7%|▋ | 3254/43939 [5:16:30<64:09:03, 5.68s/it]
7%|▋ | 3255/43939 [5:16:36<64:10:16, 5.68s/it]
7%|▋ | 3256/43939 [5:16:42<64:13:13, 5.68s/it]
7%|▋ | 3257/43939 [5:16:47<64:12:34, 5.68s/it]
7%|▋ | 3258/43939 [5:16:53<64:11:24, 5.68s/it]
7%|▋ | 3259/43939 [5:16:59<64:11:44, 5.68s/it]
7%|▋ | 3260/43939 [5:17:04<64:19:40, 5.69s/it]
7%|▋ | 3261/43939 [5:17:10<64:15:09, 5.69s/it]
7%|▋ | 3262/43939 [5:17:16<64:21:46, 5.70s/it]
7%|▋ | 3263/43939 [5:17:22<64:23:18, 5.70s/it]
7%|▋ | 3264/43939 [5:17:27<64:23:49, 5.70s/it]
7%|▋ | 3265/43939 [5:17:33<65:05:57, 5.76s/it]
7%|▋ | 3266/43939 [5:17:39<64:59:21, 5.75s/it]
7%|▋ | 3267/43939 [5:17:45<64:48:01, 5.74s/it]
7%|▋ | 3268/43939 [5:17:50<64:42:45, 5.73s/it]
7%|▋ | 3269/43939 [5:17:56<64:46:48, 5.73s/it]
7%|▋ | 3270/43939 [5:18:02<64:36:18, 5.72s/it]
7%|▋ | 3271/43939 [5:18:07<64:32:31, 5.71s/it]
7%|▋ | 3272/43939 [5:18:13<64:27:12, 5.71s/it]
7%|▋ | 3273/43939 [5:18:19<64:22:08, 5.70s/it]
7%|▋ | 3274/43939 [5:18:24<64:13:01, 5.69s/it]
7%|▋ | 3275/43939 [5:18:30<64:18:26, 5.69s/it]
7%|▋ | 3276/43939 [5:18:36<64:09:49, 5.68s/it]
7%|▋ | 3277/43939 [5:18:41<64:12:08, 5.68s/it]
7%|▋ | 3278/43939 [5:18:47<64:06:10, 5.68s/it]
7%|▋ | 3279/43939 [5:18:53<64:06:28, 5.68s/it]
7%|▋ | 3280/43939 [5:18:59<64:18:12, 5.69s/it]
7%|▋ | 3281/43939 [5:19:04<64:23:55, 5.70s/it]
7%|▋ | 3282/43939 [5:19:10<64:37:34, 5.72s/it]
7%|▋ | 3283/43939 [5:19:16<64:38:02, 5.72s/it]
7%|▋ | 3284/43939 [5:19:21<64:37:57, 5.72s/it]
7%|▋ | 3285/43939 [5:19:27<64:29:00, 5.71s/it]
7%|▋ | 3286/43939 [5:19:33<65:17:38, 5.78s/it]
7%|▋ | 3287/43939 [5:19:39<65:00:36, 5.76s/it]
7%|▋ | 3288/43939 [5:19:44<64:41:17, 5.73s/it]
7%|▋ | 3289/43939 [5:19:50<64:31:52, 5.71s/it]
7%|▋ | 3290/43939 [5:19:56<64:21:52, 5.70s/it]
7%|▋ | 3291/43939 [5:20:02<64:19:56, 5.70s/it]
7%|▋ | 3292/43939 [5:20:07<64:18:36, 5.70s/it]
7%|▋ | 3293/43939 [5:20:13<64:17:24, 5.69s/it]
7%|▋ | 3294/43939 [5:20:19<64:22:47, 5.70s/it]
7%|▋ | 3295/43939 [5:20:24<64:19:30, 5.70s/it]
8%|▊ | 3296/43939 [5:20:30<64:15:38, 5.69s/it]
8%|▊ | 3297/43939 [5:20:36<64:10:06, 5.68s/it]
8%|▊ | 3298/43939 [5:20:41<64:07:26, 5.68s/it]
8%|▊ | 3299/43939 [5:20:47<64:09:31, 5.68s/it]
8%|▊ | 3300/43939 [5:20:53<64:05:45, 5.68s/it]
{'loss': 1.1261, 'grad_norm': 1.2177249193191528, 'learning_rate': 3.7548372410653314e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.35, 'epoch': 0.3}
8%|▊ | 3300/43939 [5:20:53<64:05:45, 5.68s/it]
8%|▊ | 3301/43939 [5:20:58<64:10:36, 5.69s/it]
8%|▊ | 3302/43939 [5:21:04<64:20:52, 5.70s/it]
8%|▊ | 3303/43939 [5:21:10<64:35:00, 5.72s/it]
8%|▊ | 3304/43939 [5:21:16<64:20:17, 5.70s/it]
8%|▊ | 3305/43939 [5:21:21<64:32:26, 5.72s/it]
8%|▊ | 3306/43939 [5:21:27<64:27:25, 5.71s/it]
8%|▊ | 3307/43939 [5:21:33<65:01:53, 5.76s/it]
8%|▊ | 3308/43939 [5:21:39<64:57:34, 5.76s/it]
8%|▊ | 3309/43939 [5:21:44<64:50:39, 5.75s/it]
8%|▊ | 3310/43939 [5:21:50<64:34:25, 5.72s/it]
8%|▊ | 3311/43939 [5:21:56<64:31:14, 5.72s/it]
8%|▊ | 3312/43939 [5:22:01<64:26:26, 5.71s/it]
8%|▊ | 3313/43939 [5:22:07<64:26:35, 5.71s/it]
8%|▊ | 3314/43939 [5:22:13<64:32:37, 5.72s/it]
8%|▊ | 3315/43939 [5:22:19<64:33:01, 5.72s/it]
8%|▊ | 3316/43939 [5:22:24<64:27:18, 5.71s/it]
8%|▊ | 3317/43939 [5:22:30<64:26:35, 5.71s/it]
8%|▊ | 3318/43939 [5:22:36<64:24:12, 5.71s/it]
8%|▊ | 3319/43939 [5:22:41<64:25:23, 5.71s/it]
8%|▊ | 3320/43939 [5:22:47<64:10:47, 5.69s/it]
8%|▊ | 3321/43939 [5:22:53<64:14:42, 5.69s/it]
8%|▊ | 3322/43939 [5:22:58<64:15:25, 5.70s/it]
8%|▊ | 3323/43939 [5:23:04<64:17:47, 5.70s/it]
8%|▊ | 3324/43939 [5:23:10<64:27:49, 5.71s/it]
8%|▊ | 3325/43939 [5:23:16<64:22:09, 5.71s/it]
8%|▊ | 3326/43939 [5:23:21<64:21:44, 5.71s/it]
8%|▊ | 3327/43939 [5:23:27<64:25:26, 5.71s/it]
8%|▊ | 3328/43939 [5:23:33<64:16:04, 5.70s/it]
8%|▊ | 3329/43939 [5:23:39<65:02:35, 5.77s/it]
8%|▊ | 3330/43939 [5:23:44<64:50:12, 5.75s/it]
8%|▊ | 3331/43939 [5:23:50<64:40:35, 5.73s/it]
8%|▊ | 3332/43939 [5:23:56<64:35:18, 5.73s/it]
8%|▊ | 3333/43939 [5:24:01<64:23:17, 5.71s/it]
8%|▊ | 3334/43939 [5:24:07<64:24:42, 5.71s/it]
8%|▊ | 3335/43939 [5:24:13<64:22:18, 5.71s/it]
8%|▊ | 3336/43939 [5:24:19<64:30:10, 5.72s/it]
8%|▊ | 3337/43939 [5:24:24<64:25:05, 5.71s/it]
8%|▊ | 3338/43939 [5:24:30<64:26:31, 5.71s/it]
8%|▊ | 3339/43939 [5:24:36<64:27:35, 5.72s/it]
8%|▊ | 3340/43939 [5:24:41<64:20:53, 5.71s/it]
8%|▊ | 3341/43939 [5:24:47<64:24:13, 5.71s/it]
8%|▊ | 3342/43939 [5:24:53<64:20:39, 5.71s/it]
8%|▊ | 3343/43939 [5:24:59<64:29:10, 5.72s/it]
8%|▊ | 3344/43939 [5:25:04<64:24:40, 5.71s/it]
8%|▊ | 3345/43939 [5:25:10<64:21:20, 5.71s/it]
8%|▊ | 3346/43939 [5:25:16<64:16:34, 5.70s/it]
8%|▊ | 3347/43939 [5:25:21<64:18:17, 5.70s/it]
8%|▊ | 3348/43939 [5:25:27<64:28:46, 5.72s/it]
8%|▊ | 3349/43939 [5:25:33<64:23:09, 5.71s/it]
8%|▊ | 3350/43939 [5:25:39<65:09:47, 5.78s/it]
{'loss': 1.1125, 'grad_norm': 1.1074647903442383, 'learning_rate': 3.811745959480993e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 505.32, 'epoch': 0.3}
8%|▊ | 3350/43939 [5:25:39<65:09:47, 5.78s/it]
8%|▊ | 3351/43939 [5:25:44<64:46:55, 5.75s/it]
8%|▊ | 3352/43939 [5:25:50<64:33:45, 5.73s/it]
8%|▊ | 3353/43939 [5:25:56<64:30:38, 5.72s/it]
8%|▊ | 3354/43939 [5:26:01<64:22:35, 5.71s/it]
8%|▊ | 3355/43939 [5:26:07<64:18:15, 5.70s/it]
8%|▊ | 3356/43939 [5:26:13<64:11:00, 5.69s/it]
8%|▊ | 3357/43939 [5:26:19<64:13:24, 5.70s/it]
8%|▊ | 3358/43939 [5:26:24<64:16:22, 5.70s/it]
8%|▊ | 3359/43939 [5:26:30<64:08:22, 5.69s/it]
8%|▊ | 3360/43939 [5:26:36<64:03:27, 5.68s/it]
8%|▊ | 3361/43939 [5:26:41<64:02:47, 5.68s/it]
8%|▊ | 3362/43939 [5:26:47<64:06:32, 5.69s/it]
8%|▊ | 3363/43939 [5:26:53<64:01:07, 5.68s/it]
8%|▊ | 3364/43939 [5:26:58<64:03:42, 5.68s/it]
8%|▊ | 3365/43939 [5:27:04<64:01:50, 5.68s/it]
8%|▊ | 3366/43939 [5:27:10<64:06:48, 5.69s/it]
8%|▊ | 3367/43939 [5:27:15<64:01:33, 5.68s/it]
8%|▊ | 3368/43939 [5:27:21<64:07:41, 5.69s/it]
8%|▊ | 3369/43939 [5:27:27<64:02:34, 5.68s/it]
8%|▊ | 3370/43939 [5:27:33<64:57:21, 5.76s/it]
8%|▊ | 3371/43939 [5:27:38<64:40:51, 5.74s/it]
8%|▊ | 3372/43939 [5:27:44<64:33:20, 5.73s/it]
8%|▊ | 3373/43939 [5:27:50<64:27:42, 5.72s/it]
8%|▊ | 3374/43939 [5:27:55<64:25:26, 5.72s/it]
8%|▊ | 3375/43939 [5:28:01<64:19:48, 5.71s/it]
8%|▊ | 3376/43939 [5:28:07<64:20:39, 5.71s/it]
8%|▊ | 3377/43939 [5:28:13<64:12:30, 5.70s/it]
8%|▊ | 3378/43939 [5:28:18<64:12:18, 5.70s/it]
8%|▊ | 3379/43939 [5:28:24<64:14:46, 5.70s/it]
8%|▊ | 3380/43939 [5:28:30<64:21:06, 5.71s/it]
8%|▊ | 3381/43939 [5:28:35<64:19:31, 5.71s/it]
8%|▊ | 3382/43939 [5:28:41<64:16:27, 5.71s/it]
8%|▊ | 3383/43939 [5:28:47<64:04:24, 5.69s/it]
8%|▊ | 3384/43939 [5:28:52<64:13:57, 5.70s/it]
8%|▊ | 3385/43939 [5:28:58<64:08:30, 5.69s/it]
8%|▊ | 3386/43939 [5:29:04<64:10:41, 5.70s/it]
8%|▊ | 3387/43939 [5:29:10<64:12:07, 5.70s/it]
8%|▊ | 3388/43939 [5:29:15<64:13:31, 5.70s/it]
8%|▊ | 3389/43939 [5:29:21<64:13:05, 5.70s/it]
8%|▊ | 3390/43939 [5:29:27<64:10:53, 5.70s/it]
8%|▊ | 3391/43939 [5:29:33<64:48:25, 5.75s/it]
8%|▊ | 3392/43939 [5:29:38<64:39:28, 5.74s/it]
8%|▊ | 3393/43939 [5:29:44<64:32:52, 5.73s/it]
8%|▊ | 3394/43939 [5:29:50<64:23:36, 5.72s/it]
8%|▊ | 3395/43939 [5:29:55<64:31:05, 5.73s/it]
8%|▊ | 3396/43939 [5:30:01<64:24:45, 5.72s/it]
8%|▊ | 3397/43939 [5:30:07<64:13:18, 5.70s/it]
8%|▊ | 3398/43939 [5:30:12<64:10:49, 5.70s/it]
8%|▊ | 3399/43939 [5:30:18<64:05:09, 5.69s/it]
8%|▊ | 3400/43939 [5:30:24<64:11:33, 5.70s/it]
{'loss': 1.1102, 'grad_norm': 1.2007449865341187, 'learning_rate': 3.8686546778966534e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.46, 'epoch': 0.31}
8%|▊ | 3400/43939 [5:30:24<64:11:33, 5.70s/it]
8%|▊ | 3401/43939 [5:30:30<64:14:52, 5.71s/it]
8%|▊ | 3402/43939 [5:30:35<64:14:53, 5.71s/it]
8%|▊ | 3403/43939 [5:30:41<64:19:17, 5.71s/it]
8%|▊ | 3404/43939 [5:30:47<64:16:43, 5.71s/it]
8%|▊ | 3405/43939 [5:30:52<64:15:59, 5.71s/it]
8%|▊ | 3406/43939 [5:30:58<64:05:01, 5.69s/it]
8%|▊ | 3407/43939 [5:31:04<64:08:17, 5.70s/it]
8%|▊ | 3408/43939 [5:31:09<64:13:14, 5.70s/it]
8%|▊ | 3409/43939 [5:31:15<64:14:41, 5.71s/it]
8%|▊ | 3410/43939 [5:31:21<64:12:06, 5.70s/it]
8%|▊ | 3411/43939 [5:31:27<64:05:59, 5.69s/it]
8%|▊ | 3412/43939 [5:31:32<64:39:09, 5.74s/it]
8%|▊ | 3413/43939 [5:31:38<64:28:58, 5.73s/it]
8%|▊ | 3414/43939 [5:31:44<64:38:59, 5.74s/it]
8%|▊ | 3415/43939 [5:31:50<64:26:17, 5.72s/it]
8%|▊ | 3416/43939 [5:31:55<64:19:58, 5.72s/it]
8%|▊ | 3417/43939 [5:32:01<64:08:20, 5.70s/it]
8%|▊ | 3418/43939 [5:32:07<64:06:06, 5.69s/it]
8%|▊ | 3419/43939 [5:32:12<64:07:29, 5.70s/it]
8%|▊ | 3420/43939 [5:32:18<64:07:30, 5.70s/it]
8%|▊ | 3421/43939 [5:32:24<64:12:27, 5.70s/it]
8%|▊ | 3422/43939 [5:32:29<64:09:06, 5.70s/it]
8%|▊ | 3423/43939 [5:32:35<63:58:28, 5.68s/it]
8%|▊ | 3424/43939 [5:32:41<64:02:27, 5.69s/it]
8%|▊ | 3425/43939 [5:32:46<64:03:45, 5.69s/it]
8%|▊ | 3426/43939 [5:32:52<64:00:13, 5.69s/it]
8%|▊ | 3427/43939 [5:32:58<64:03:52, 5.69s/it]
8%|▊ | 3428/43939 [5:33:04<64:00:13, 5.69s/it]
8%|▊ | 3429/43939 [5:33:09<64:04:39, 5.69s/it]
8%|▊ | 3430/43939 [5:33:15<64:07:13, 5.70s/it]
8%|▊ | 3431/43939 [5:33:21<64:05:37, 5.70s/it]
8%|▊ | 3432/43939 [5:33:27<64:55:25, 5.77s/it]
8%|▊ | 3433/43939 [5:33:32<64:53:22, 5.77s/it]
8%|▊ | 3434/43939 [5:33:38<64:51:25, 5.76s/it]
8%|▊ | 3435/43939 [5:33:44<64:44:45, 5.75s/it]
8%|▊ | 3436/43939 [5:33:50<64:35:39, 5.74s/it]
8%|▊ | 3437/43939 [5:33:55<64:26:07, 5.73s/it]
8%|▊ | 3438/43939 [5:34:01<64:21:34, 5.72s/it]
8%|▊ | 3439/43939 [5:34:07<64:26:50, 5.73s/it]
8%|▊ | 3440/43939 [5:34:12<64:30:25, 5.73s/it]
8%|▊ | 3441/43939 [5:34:18<64:19:42, 5.72s/it]
8%|▊ | 3442/43939 [5:34:24<64:08:21, 5.70s/it]
8%|▊ | 3443/43939 [5:34:29<64:10:01, 5.70s/it]
8%|▊ | 3444/43939 [5:34:35<64:07:02, 5.70s/it]
8%|▊ | 3445/43939 [5:34:41<64:02:50, 5.69s/it]
8%|▊ | 3446/43939 [5:34:47<64:02:44, 5.69s/it]
8%|▊ | 3447/43939 [5:34:52<64:02:47, 5.69s/it]
8%|▊ | 3448/43939 [5:34:58<63:53:31, 5.68s/it]
8%|▊ | 3449/43939 [5:35:04<63:53:18, 5.68s/it]
8%|▊ | 3450/43939 [5:35:09<63:58:09, 5.69s/it]
{'loss': 1.0961, 'grad_norm': 1.5369510650634766, 'learning_rate': 3.925563396312315e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.95, 'epoch': 0.31}
8%|▊ | 3450/43939 [5:35:09<63:58:09, 5.69s/it]
8%|▊ | 3451/43939 [5:35:15<64:08:40, 5.70s/it]
8%|▊ | 3452/43939 [5:35:21<64:07:51, 5.70s/it]
8%|▊ | 3453/43939 [5:35:27<65:06:40, 5.79s/it]
8%|▊ | 3454/43939 [5:35:32<64:46:33, 5.76s/it]
8%|▊ | 3455/43939 [5:35:38<64:36:03, 5.74s/it]
8%|▊ | 3456/43939 [5:35:44<64:27:31, 5.73s/it]
8%|▊ | 3457/43939 [5:35:50<64:30:31, 5.74s/it]
8%|▊ | 3458/43939 [5:35:55<64:30:08, 5.74s/it]
8%|▊ | 3459/43939 [5:36:01<64:33:27, 5.74s/it]
8%|▊ | 3460/43939 [5:36:07<64:28:24, 5.73s/it]
8%|▊ | 3461/43939 [5:36:12<64:22:11, 5.72s/it]
8%|▊ | 3462/43939 [5:36:18<64:07:54, 5.70s/it]
8%|▊ | 3463/43939 [5:36:24<64:07:02, 5.70s/it]
8%|▊ | 3464/43939 [5:36:30<64:03:50, 5.70s/it]
8%|▊ | 3465/43939 [5:36:35<63:58:17, 5.69s/it]
8%|▊ | 3466/43939 [5:36:41<63:49:50, 5.68s/it]
8%|▊ | 3467/43939 [5:36:47<63:55:31, 5.69s/it]
8%|▊ | 3468/43939 [5:36:52<63:53:56, 5.68s/it]
8%|▊ | 3469/43939 [5:36:58<64:00:30, 5.69s/it]
8%|▊ | 3470/43939 [5:37:04<64:07:11, 5.70s/it]
8%|▊ | 3471/43939 [5:37:09<64:01:22, 5.70s/it]
8%|▊ | 3472/43939 [5:37:15<64:01:24, 5.70s/it]
8%|▊ | 3473/43939 [5:37:21<64:08:03, 5.71s/it]
8%|▊ | 3474/43939 [5:37:27<64:52:08, 5.77s/it]
8%|▊ | 3475/43939 [5:37:32<64:34:59, 5.75s/it]
8%|▊ | 3476/43939 [5:37:38<64:31:54, 5.74s/it]
8%|▊ | 3477/43939 [5:37:44<64:22:34, 5.73s/it]
8%|▊ | 3478/43939 [5:37:49<64:05:05, 5.70s/it]
8%|▊ | 3479/43939 [5:37:55<64:03:00, 5.70s/it]
8%|▊ | 3480/43939 [5:38:01<64:02:43, 5.70s/it]
8%|▊ | 3481/43939 [5:38:07<64:13:42, 5.72s/it]
8%|▊ | 3482/43939 [5:38:12<64:09:52, 5.71s/it]
8%|▊ | 3483/43939 [5:38:18<64:13:37, 5.72s/it]
8%|▊ | 3484/43939 [5:38:24<64:09:39, 5.71s/it]
8%|▊ | 3485/43939 [5:38:29<64:12:57, 5.71s/it]
8%|▊ | 3486/43939 [5:38:35<64:15:05, 5.72s/it]
8%|▊ | 3487/43939 [5:38:41<64:08:06, 5.71s/it]
8%|▊ | 3488/43939 [5:38:47<64:04:30, 5.70s/it]
8%|▊ | 3489/43939 [5:38:52<64:03:26, 5.70s/it]
8%|▊ | 3490/43939 [5:38:58<64:02:23, 5.70s/it]
8%|▊ | 3491/43939 [5:39:04<64:00:39, 5.70s/it]
8%|▊ | 3492/43939 [5:39:09<64:07:35, 5.71s/it]
8%|▊ | 3493/43939 [5:39:15<64:07:52, 5.71s/it]
8%|▊ | 3494/43939 [5:39:21<64:02:54, 5.70s/it]
8%|▊ | 3495/43939 [5:39:27<64:49:08, 5.77s/it]
8%|▊ | 3496/43939 [5:39:32<64:30:04, 5.74s/it]
8%|▊ | 3497/43939 [5:39:38<64:32:05, 5.74s/it]
8%|▊ | 3498/43939 [5:39:44<64:24:14, 5.73s/it]
8%|▊ | 3499/43939 [5:39:49<64:07:05, 5.71s/it]
8%|▊ | 3500/43939 [5:39:55<64:04:34, 5.70s/it]
{'loss': 1.1108, 'grad_norm': 0.9427760243415833, 'learning_rate': 3.982472114727977e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.4, 'epoch': 0.32}
8%|▊ | 3500/43939 [5:39:55<64:04:34, 5.70s/it]
8%|▊ | 3501/43939 [5:40:01<64:02:05, 5.70s/it]
8%|▊ | 3502/43939 [5:40:07<63:55:09, 5.69s/it]
8%|▊ | 3503/43939 [5:40:12<64:07:07, 5.71s/it]
8%|▊ | 3504/43939 [5:40:18<64:03:14, 5.70s/it]
8%|▊ | 3505/43939 [5:40:24<64:00:41, 5.70s/it]
8%|▊ | 3506/43939 [5:40:29<64:11:52, 5.72s/it]
8%|▊ | 3507/43939 [5:40:35<64:13:33, 5.72s/it]
8%|▊ | 3508/43939 [5:40:41<64:10:34, 5.71s/it]
8%|▊ | 3509/43939 [5:40:47<64:16:56, 5.72s/it]
8%|▊ | 3510/43939 [5:40:52<64:11:48, 5.72s/it]
8%|▊ | 3511/43939 [5:40:58<64:16:55, 5.72s/it]
8%|▊ | 3512/43939 [5:41:04<64:27:22, 5.74s/it]
8%|▊ | 3513/43939 [5:41:10<64:24:11, 5.74s/it]
8%|▊ | 3514/43939 [5:41:15<64:19:06, 5.73s/it]
8%|▊ | 3515/43939 [5:41:21<64:13:58, 5.72s/it]
8%|▊ | 3516/43939 [5:41:27<64:53:14, 5.78s/it]
8%|▊ | 3517/43939 [5:41:33<64:31:45, 5.75s/it]
8%|▊ | 3518/43939 [5:41:38<64:14:06, 5.72s/it]
8%|▊ | 3519/43939 [5:41:44<64:09:28, 5.71s/it]
8%|▊ | 3520/43939 [5:41:50<64:17:03, 5.73s/it]
8%|▊ | 3521/43939 [5:41:55<64:14:33, 5.72s/it]
8%|▊ | 3522/43939 [5:42:01<64:10:12, 5.72s/it]
8%|▊ | 3523/43939 [5:42:07<64:25:39, 5.74s/it]
8%|▊ | 3524/43939 [5:42:13<64:26:09, 5.74s/it]
8%|▊ | 3525/43939 [5:42:18<64:15:06, 5.72s/it]
8%|▊ | 3526/43939 [5:42:24<63:57:27, 5.70s/it]
8%|▊ | 3527/43939 [5:42:30<63:57:02, 5.70s/it]
8%|▊ | 3528/43939 [5:42:35<63:53:50, 5.69s/it]
8%|▊ | 3529/43939 [5:42:41<63:49:24, 5.69s/it]
8%|▊ | 3530/43939 [5:42:47<63:48:11, 5.68s/it]
8%|▊ | 3531/43939 [5:42:52<64:00:13, 5.70s/it]
8%|▊ | 3532/43939 [5:42:58<64:06:36, 5.71s/it]
8%|▊ | 3533/43939 [5:43:04<64:03:44, 5.71s/it]
8%|▊ | 3534/43939 [5:43:09<64:00:42, 5.70s/it]
8%|▊ | 3535/43939 [5:43:15<63:59:25, 5.70s/it]
8%|▊ | 3536/43939 [5:43:21<64:08:03, 5.71s/it]
8%|▊ | 3537/43939 [5:43:27<64:50:45, 5.78s/it]
8%|▊ | 3538/43939 [5:43:33<64:33:44, 5.75s/it]
8%|▊ | 3539/43939 [5:43:38<64:23:33, 5.74s/it]
8%|▊ | 3540/43939 [5:43:44<64:19:47, 5.73s/it]
8%|▊ | 3541/43939 [5:43:50<64:10:25, 5.72s/it]
8%|▊ | 3542/43939 [5:43:55<64:06:06, 5.71s/it]
8%|▊ | 3543/43939 [5:44:01<63:59:48, 5.70s/it]
8%|▊ | 3544/43939 [5:44:07<63:54:46, 5.70s/it]
8%|▊ | 3545/43939 [5:44:12<63:56:08, 5.70s/it]
8%|▊ | 3546/43939 [5:44:18<64:00:10, 5.70s/it]
8%|▊ | 3547/43939 [5:44:24<63:55:09, 5.70s/it]
8%|▊ | 3548/43939 [5:44:30<63:54:31, 5.70s/it]
8%|▊ | 3549/43939 [5:44:35<63:52:14, 5.69s/it]
8%|▊ | 3550/43939 [5:44:41<63:50:52, 5.69s/it]
{'loss': 1.1014, 'grad_norm': 1.334630012512207, 'learning_rate': 4.039380833143638e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.55, 'epoch': 0.32}
8%|▊ | 3550/43939 [5:44:41<63:50:52, 5.69s/it]
8%|▊ | 3551/43939 [5:44:47<63:44:03, 5.68s/it]
8%|▊ | 3552/43939 [5:44:52<63:49:42, 5.69s/it]
8%|▊ | 3553/43939 [5:44:58<63:55:55, 5.70s/it]
8%|▊ | 3554/43939 [5:45:04<63:48:05, 5.69s/it]
8%|▊ | 3555/43939 [5:45:09<63:45:06, 5.68s/it]
8%|▊ | 3556/43939 [5:45:15<63:48:11, 5.69s/it]
8%|▊ | 3557/43939 [5:45:21<63:53:35, 5.70s/it]
8%|▊ | 3558/43939 [5:45:27<64:38:00, 5.76s/it]
8%|▊ | 3559/43939 [5:45:32<64:24:05, 5.74s/it]
8%|▊ | 3560/43939 [5:45:38<64:15:53, 5.73s/it]
8%|▊ | 3561/43939 [5:45:44<64:08:08, 5.72s/it]
8%|▊ | 3562/43939 [5:45:49<63:54:12, 5.70s/it]
8%|▊ | 3563/43939 [5:45:55<63:56:54, 5.70s/it]
8%|▊ | 3564/43939 [5:46:01<63:59:16, 5.71s/it]
8%|▊ | 3565/43939 [5:46:07<63:59:32, 5.71s/it]
8%|▊ | 3566/43939 [5:46:12<63:55:44, 5.70s/it]
8%|▊ | 3567/43939 [5:46:18<63:59:55, 5.71s/it]
8%|▊ | 3568/43939 [5:46:24<63:59:07, 5.71s/it]
8%|▊ | 3569/43939 [5:46:29<63:57:49, 5.70s/it]
8%|▊ | 3570/43939 [5:46:35<63:51:02, 5.69s/it]
8%|▊ | 3571/43939 [5:46:41<63:55:09, 5.70s/it]
8%|▊ | 3572/43939 [5:46:46<63:55:42, 5.70s/it]
8%|▊ | 3573/43939 [5:46:52<63:51:28, 5.70s/it]
8%|▊ | 3574/43939 [5:46:58<63:51:45, 5.70s/it]
8%|▊ | 3575/43939 [5:47:04<63:55:08, 5.70s/it]
8%|▊ | 3576/43939 [5:47:09<64:05:00, 5.72s/it]
8%|▊ | 3577/43939 [5:47:15<64:01:36, 5.71s/it]
8%|▊ | 3578/43939 [5:47:21<64:52:57, 5.79s/it]
8%|▊ | 3579/43939 [5:47:27<64:42:10, 5.77s/it]
8%|▊ | 3580/43939 [5:47:32<64:25:53, 5.75s/it]
8%|▊ | 3581/43939 [5:47:38<64:22:01, 5.74s/it]
8%|▊ | 3582/43939 [5:47:44<64:14:26, 5.73s/it]
8%|▊ | 3583/43939 [5:47:49<64:07:31, 5.72s/it]
8%|▊ | 3584/43939 [5:47:55<64:04:32, 5.72s/it]
8%|▊ | 3585/43939 [5:48:01<64:03:42, 5.71s/it]
8%|▊ | 3586/43939 [5:48:07<64:00:43, 5.71s/it]
8%|▊ | 3587/43939 [5:48:12<63:59:25, 5.71s/it]
8%|▊ | 3588/43939 [5:48:18<63:52:13, 5.70s/it]
8%|▊ | 3589/43939 [5:48:24<63:51:47, 5.70s/it]
8%|▊ | 3590/43939 [5:48:29<63:57:06, 5.71s/it]
8%|▊ | 3591/43939 [5:48:35<63:50:20, 5.70s/it]
8%|▊ | 3592/43939 [5:48:41<63:45:41, 5.69s/it]
8%|▊ | 3593/43939 [5:48:46<63:44:45, 5.69s/it]
8%|▊ | 3594/43939 [5:48:52<63:40:13, 5.68s/it]
8%|▊ | 3595/43939 [5:48:58<63:40:36, 5.68s/it]
8%|▊ | 3596/43939 [5:49:03<63:42:06, 5.68s/it]
8%|▊ | 3597/43939 [5:49:09<63:41:50, 5.68s/it]
8%|▊ | 3598/43939 [5:49:15<63:50:43, 5.70s/it]
8%|▊ | 3599/43939 [5:49:21<63:57:30, 5.71s/it]
8%|▊ | 3600/43939 [5:49:27<64:51:01, 5.79s/it]
{'loss': 1.11, 'grad_norm': 1.1430654525756836, 'learning_rate': 4.0962895515592994e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 437.81, 'epoch': 0.33}
8%|▊ | 3600/43939 [5:49:27<64:51:01, 5.79s/it]
8%|▊ | 3601/43939 [5:49:32<64:45:58, 5.78s/it]
8%|▊ | 3602/43939 [5:49:38<64:34:41, 5.76s/it]
8%|▊ | 3603/43939 [5:49:44<64:19:34, 5.74s/it]
8%|▊ | 3604/43939 [5:49:49<64:17:21, 5.74s/it]
8%|▊ | 3605/43939 [5:49:55<64:17:29, 5.74s/it]
8%|▊ | 3606/43939 [5:50:01<64:02:16, 5.72s/it]
8%|▊ | 3607/43939 [5:50:07<63:53:39, 5.70s/it]
8%|▊ | 3608/43939 [5:50:12<63:58:49, 5.71s/it]
8%|▊ | 3609/43939 [5:50:18<63:46:00, 5.69s/it]
8%|▊ | 3610/43939 [5:50:24<63:47:13, 5.69s/it]
8%|▊ | 3611/43939 [5:50:29<63:43:45, 5.69s/it]
8%|▊ | 3612/43939 [5:50:35<63:48:16, 5.70s/it]
8%|▊ | 3613/43939 [5:50:41<63:47:58, 5.70s/it]
8%|▊ | 3614/43939 [5:50:46<63:48:54, 5.70s/it]
8%|▊ | 3615/43939 [5:50:52<63:50:21, 5.70s/it]
8%|▊ | 3616/43939 [5:50:58<63:51:40, 5.70s/it]
8%|▊ | 3617/43939 [5:51:04<63:50:17, 5.70s/it]
8%|▊ | 3618/43939 [5:51:09<63:49:13, 5.70s/it]
8%|▊ | 3619/43939 [5:51:15<63:52:06, 5.70s/it]
8%|▊ | 3620/43939 [5:51:21<64:41:41, 5.78s/it]
8%|▊ | 3621/43939 [5:51:27<64:22:07, 5.75s/it]
8%|▊ | 3622/43939 [5:51:32<64:16:12, 5.74s/it]
8%|▊ | 3623/43939 [5:51:38<64:04:09, 5.72s/it]
8%|▊ | 3624/43939 [5:51:44<63:55:55, 5.71s/it]
8%|▊ | 3625/43939 [5:51:49<63:52:23, 5.70s/it]
8%|▊ | 3626/43939 [5:51:55<63:50:12, 5.70s/it]
8%|▊ | 3627/43939 [5:52:01<63:55:12, 5.71s/it]
8%|▊ | 3628/43939 [5:52:06<63:53:48, 5.71s/it]
8%|▊ | 3629/43939 [5:52:12<63:48:23, 5.70s/it]
8%|▊ | 3630/43939 [5:52:18<63:57:12, 5.71s/it]
8%|▊ | 3631/43939 [5:52:24<63:56:57, 5.71s/it]
8%|▊ | 3632/43939 [5:52:29<63:50:21, 5.70s/it]
8%|▊ | 3633/43939 [5:52:35<63:46:01, 5.70s/it]
8%|▊ | 3634/43939 [5:52:41<63:47:56, 5.70s/it]
8%|▊ | 3635/43939 [5:52:46<63:48:39, 5.70s/it]
8%|▊ | 3636/43939 [5:52:52<63:41:18, 5.69s/it]
8%|▊ | 3637/43939 [5:52:58<63:47:05, 5.70s/it]
8%|▊ | 3638/43939 [5:53:03<63:52:56, 5.71s/it]
8%|▊ | 3639/43939 [5:53:09<63:55:19, 5.71s/it]
8%|▊ | 3640/43939 [5:53:15<63:48:20, 5.70s/it]
8%|▊ | 3641/43939 [5:53:21<64:36:01, 5.77s/it]
8%|▊ | 3642/43939 [5:53:27<64:31:18, 5.76s/it]
8%|▊ | 3643/43939 [5:53:32<64:13:40, 5.74s/it]
8%|▊ | 3644/43939 [5:53:38<63:58:10, 5.72s/it]
8%|▊ | 3645/43939 [5:53:44<63:53:04, 5.71s/it]
8%|▊ | 3646/43939 [5:53:49<63:51:48, 5.71s/it]
8%|▊ | 3647/43939 [5:53:55<63:48:39, 5.70s/it]
8%|▊ | 3648/43939 [5:54:01<63:44:25, 5.70s/it]
8%|▊ | 3649/43939 [5:54:06<63:47:26, 5.70s/it]
8%|▊ | 3650/43939 [5:54:12<63:45:17, 5.70s/it]
{'loss': 1.0991, 'grad_norm': 1.4943946599960327, 'learning_rate': 4.1531982699749604e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.41, 'epoch': 0.33}
8%|▊ | 3650/43939 [5:54:12<63:45:17, 5.70s/it]
8%|▊ | 3651/43939 [5:54:18<63:48:06, 5.70s/it]
8%|▊ | 3652/43939 [5:54:23<63:41:57, 5.69s/it]
8%|▊ | 3653/43939 [5:54:29<63:46:46, 5.70s/it]
8%|▊ | 3654/43939 [5:54:35<63:48:34, 5.70s/it]
8%|▊ | 3655/43939 [5:54:41<64:05:37, 5.73s/it]
8%|▊ | 3656/43939 [5:54:46<63:55:23, 5.71s/it]
8%|▊ | 3657/43939 [5:54:52<63:59:26, 5.72s/it]
8%|▊ | 3658/43939 [5:54:58<63:52:13, 5.71s/it]
8%|▊ | 3659/43939 [5:55:03<63:41:58, 5.69s/it]
8%|▊ | 3660/43939 [5:55:09<63:49:53, 5.71s/it]
8%|▊ | 3661/43939 [5:55:15<63:44:01, 5.70s/it]
8%|▊ | 3662/43939 [5:55:21<63:46:00, 5.70s/it]
8%|▊ | 3663/43939 [5:55:26<64:20:05, 5.75s/it]
8%|▊ | 3664/43939 [5:55:32<64:00:02, 5.72s/it]
8%|▊ | 3665/43939 [5:55:38<63:52:32, 5.71s/it]
8%|▊ | 3666/43939 [5:55:43<63:49:40, 5.71s/it]
8%|▊ | 3667/43939 [5:55:49<63:56:48, 5.72s/it]
8%|▊ | 3668/43939 [5:55:55<63:50:19, 5.71s/it]
8%|▊ | 3669/43939 [5:56:01<63:45:55, 5.70s/it]
8%|▊ | 3670/43939 [5:56:06<63:39:57, 5.69s/it]
8%|▊ | 3671/43939 [5:56:12<63:42:18, 5.70s/it]
8%|▊ | 3672/43939 [5:56:18<63:52:19, 5.71s/it]
8%|▊ | 3673/43939 [5:56:23<63:44:15, 5.70s/it]
8%|▊ | 3674/43939 [5:56:29<63:39:39, 5.69s/it]
8%|▊ | 3675/43939 [5:56:35<63:44:42, 5.70s/it]
8%|▊ | 3676/43939 [5:56:40<63:39:11, 5.69s/it]
8%|▊ | 3677/43939 [5:56:46<63:43:13, 5.70s/it]
8%|▊ | 3678/43939 [5:56:52<63:46:43, 5.70s/it]
8%|▊ | 3679/43939 [5:56:58<63:45:05, 5.70s/it]
8%|▊ | 3680/43939 [5:57:03<63:38:24, 5.69s/it]
8%|▊ | 3681/43939 [5:57:09<63:35:51, 5.69s/it]
8%|▊ | 3682/43939 [5:57:15<63:36:25, 5.69s/it]
8%|▊ | 3683/43939 [5:57:20<63:36:05, 5.69s/it]
8%|▊ | 3684/43939 [5:57:26<64:19:24, 5.75s/it]
8%|▊ | 3685/43939 [5:57:32<64:13:27, 5.74s/it]
8%|▊ | 3686/43939 [5:57:38<63:59:00, 5.72s/it]
8%|▊ | 3687/43939 [5:57:43<63:51:02, 5.71s/it]
8%|▊ | 3688/43939 [5:57:49<63:50:49, 5.71s/it]
8%|▊ | 3689/43939 [5:57:55<63:44:00, 5.70s/it]
8%|▊ | 3690/43939 [5:58:00<63:46:13, 5.70s/it]
8%|▊ | 3691/43939 [5:58:06<63:53:26, 5.71s/it]
8%|▊ | 3692/43939 [5:58:12<63:50:58, 5.71s/it]
8%|▊ | 3693/43939 [5:58:18<64:00:40, 5.73s/it]
8%|▊ | 3694/43939 [5:58:23<63:49:36, 5.71s/it]
8%|▊ | 3695/43939 [5:58:29<63:46:46, 5.71s/it]
8%|▊ | 3696/43939 [5:58:35<63:42:13, 5.70s/it]
8%|▊ | 3697/43939 [5:58:40<63:37:14, 5.69s/it]
8%|▊ | 3698/43939 [5:58:46<63:43:25, 5.70s/it]
8%|▊ | 3699/43939 [5:58:52<63:39:40, 5.70s/it]
8%|▊ | 3700/43939 [5:58:57<63:49:15, 5.71s/it]
{'loss': 1.0669, 'grad_norm': 1.2534953355789185, 'learning_rate': 4.210106988390622e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.29, 'epoch': 0.34}
8%|▊ | 3700/43939 [5:58:57<63:49:15, 5.71s/it]
8%|▊ | 3701/43939 [5:59:03<63:39:30, 5.70s/it]
8%|▊ | 3702/43939 [5:59:09<63:40:41, 5.70s/it]
8%|▊ | 3703/43939 [5:59:14<63:41:39, 5.70s/it]
8%|▊ | 3704/43939 [5:59:20<64:32:01, 5.77s/it]
8%|▊ | 3705/43939 [5:59:26<64:09:04, 5.74s/it]
8%|▊ | 3706/43939 [5:59:32<64:02:59, 5.73s/it]
8%|▊ | 3707/43939 [5:59:37<63:51:00, 5.71s/it]
8%|▊ | 3708/43939 [5:59:43<63:42:56, 5.70s/it]
8%|▊ | 3709/43939 [5:59:49<63:41:22, 5.70s/it]
8%|▊ | 3710/43939 [5:59:54<63:37:00, 5.69s/it]
8%|▊ | 3711/43939 [6:00:00<63:29:27, 5.68s/it]
8%|▊ | 3712/43939 [6:00:06<63:30:04, 5.68s/it]
8%|▊ | 3713/43939 [6:00:12<63:43:59, 5.70s/it]
8%|▊ | 3714/43939 [6:00:17<63:45:09, 5.71s/it]
8%|▊ | 3715/43939 [6:00:23<63:50:43, 5.71s/it]
8%|▊ | 3716/43939 [6:00:29<63:48:56, 5.71s/it]
8%|▊ | 3717/43939 [6:00:34<63:49:38, 5.71s/it]
8%|▊ | 3718/43939 [6:00:40<63:46:34, 5.71s/it]
8%|▊ | 3719/43939 [6:00:46<63:47:04, 5.71s/it]
8%|▊ | 3720/43939 [6:00:52<63:40:55, 5.70s/it]
8%|▊ | 3721/43939 [6:00:57<63:56:01, 5.72s/it]
8%|▊ | 3722/43939 [6:01:03<63:44:54, 5.71s/it]
8%|▊ | 3723/43939 [6:01:09<63:40:19, 5.70s/it]
8%|▊ | 3724/43939 [6:01:14<63:48:20, 5.71s/it]
8%|▊ | 3725/43939 [6:01:20<64:22:36, 5.76s/it]
8%|▊ | 3726/43939 [6:01:26<64:10:15, 5.74s/it]
8%|▊ | 3727/43939 [6:01:32<63:57:11, 5.73s/it]
8%|▊ | 3728/43939 [6:01:37<63:53:51, 5.72s/it]
8%|▊ | 3729/43939 [6:01:43<63:42:51, 5.70s/it]
8%|▊ | 3730/43939 [6:01:49<63:38:59, 5.70s/it]
8%|▊ | 3731/43939 [6:01:54<63:43:50, 5.71s/it]
8%|▊ | 3732/43939 [6:02:00<63:38:03, 5.70s/it]
8%|▊ | 3733/43939 [6:02:06<63:44:45, 5.71s/it]
8%|▊ | 3734/43939 [6:02:12<63:42:06, 5.70s/it]
9%|▊ | 3735/43939 [6:02:17<63:47:41, 5.71s/it]
9%|▊ | 3736/43939 [6:02:23<63:42:55, 5.71s/it]
9%|▊ | 3737/43939 [6:02:29<63:42:52, 5.71s/it]
9%|▊ | 3738/43939 [6:02:34<63:31:34, 5.69s/it]
9%|▊ | 3739/43939 [6:02:40<63:39:33, 5.70s/it]
9%|▊ | 3740/43939 [6:02:46<63:37:10, 5.70s/it]
9%|▊ | 3741/43939 [6:02:51<63:39:02, 5.70s/it]
9%|▊ | 3742/43939 [6:02:57<63:34:15, 5.69s/it]
9%|▊ | 3743/43939 [6:03:03<63:41:41, 5.70s/it]
9%|▊ | 3744/43939 [6:03:09<63:41:13, 5.70s/it]
9%|▊ | 3745/43939 [6:03:14<63:39:05, 5.70s/it]
9%|▊ | 3746/43939 [6:03:20<63:37:34, 5.70s/it]
9%|▊ | 3747/43939 [6:03:26<64:10:22, 5.75s/it]
9%|▊ | 3748/43939 [6:03:32<63:56:09, 5.73s/it]
9%|▊ | 3749/43939 [6:03:37<63:42:33, 5.71s/it]
9%|▊ | 3750/43939 [6:03:43<63:44:01, 5.71s/it]
{'loss': 1.0984, 'grad_norm': 1.0308473110198975, 'learning_rate': 4.267015706806283e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 455.95, 'epoch': 0.34}
9%|▊ | 3750/43939 [6:03:43<63:44:01, 5.71s/it]
9%|▊ | 3751/43939 [6:03:49<63:41:52, 5.71s/it]
9%|▊ | 3752/43939 [6:03:54<63:41:05, 5.70s/it]
9%|▊ | 3753/43939 [6:04:00<63:38:02, 5.70s/it]
9%|▊ | 3754/43939 [6:04:06<63:28:20, 5.69s/it]
9%|▊ | 3755/43939 [6:04:11<63:35:39, 5.70s/it]
9%|▊ | 3756/43939 [6:04:17<63:38:52, 5.70s/it]
9%|▊ | 3757/43939 [6:04:23<63:37:55, 5.70s/it]
9%|▊ | 3758/43939 [6:04:28<63:37:40, 5.70s/it]
9%|▊ | 3759/43939 [6:04:34<63:41:56, 5.71s/it]
9%|▊ | 3760/43939 [6:04:40<63:40:10, 5.70s/it]
9%|▊ | 3761/43939 [6:04:46<63:42:38, 5.71s/it]
9%|▊ | 3762/43939 [6:04:51<63:40:24, 5.71s/it]
9%|▊ | 3763/43939 [6:04:57<63:40:30, 5.71s/it]
9%|▊ | 3764/43939 [6:05:03<63:37:05, 5.70s/it]
9%|▊ | 3765/43939 [6:05:08<63:34:15, 5.70s/it]
9%|▊ | 3766/43939 [6:05:14<63:39:25, 5.70s/it]
9%|▊ | 3767/43939 [6:05:20<63:32:57, 5.69s/it]
9%|▊ | 3768/43939 [6:05:26<64:09:29, 5.75s/it]
9%|▊ | 3769/43939 [6:05:31<63:54:18, 5.73s/it]
9%|▊ | 3770/43939 [6:05:37<63:44:13, 5.71s/it]
9%|▊ | 3771/43939 [6:05:43<63:31:26, 5.69s/it]
9%|▊ | 3772/43939 [6:05:48<63:28:30, 5.69s/it]
9%|▊ | 3773/43939 [6:05:54<63:32:12, 5.69s/it]
9%|▊ | 3774/43939 [6:06:00<63:28:18, 5.69s/it]
9%|▊ | 3775/43939 [6:06:05<63:28:33, 5.69s/it]
9%|▊ | 3776/43939 [6:06:11<63:27:52, 5.69s/it]
9%|▊ | 3777/43939 [6:06:17<63:28:54, 5.69s/it]
9%|▊ | 3778/43939 [6:06:22<63:27:44, 5.69s/it]
9%|▊ | 3779/43939 [6:06:28<63:28:33, 5.69s/it]
9%|▊ | 3780/43939 [6:06:34<63:35:34, 5.70s/it]
9%|▊ | 3781/43939 [6:06:40<63:37:59, 5.70s/it]
9%|▊ | 3782/43939 [6:06:45<63:42:54, 5.71s/it]
9%|▊ | 3783/43939 [6:06:51<63:37:07, 5.70s/it]
9%|▊ | 3784/43939 [6:06:57<63:31:31, 5.70s/it]
9%|▊ | 3785/43939 [6:07:02<63:25:44, 5.69s/it]
9%|▊ | 3786/43939 [6:07:08<63:35:28, 5.70s/it]
9%|▊ | 3787/43939 [6:07:14<63:27:08, 5.69s/it]
9%|▊ | 3788/43939 [6:07:19<63:25:06, 5.69s/it]
9%|▊ | 3789/43939 [6:07:25<64:14:26, 5.76s/it]
9%|▊ | 3790/43939 [6:07:31<64:11:01, 5.76s/it]
9%|▊ | 3791/43939 [6:07:37<64:16:32, 5.76s/it]
9%|▊ | 3792/43939 [6:07:43<63:51:12, 5.73s/it]
9%|▊ | 3793/43939 [6:07:48<63:44:17, 5.72s/it]
9%|▊ | 3794/43939 [6:07:54<63:37:29, 5.71s/it]
9%|▊ | 3795/43939 [6:08:00<63:28:25, 5.69s/it]
9%|▊ | 3796/43939 [6:08:05<63:35:18, 5.70s/it]
9%|▊ | 3797/43939 [6:08:11<63:30:17, 5.70s/it]
9%|▊ | 3798/43939 [6:08:17<63:25:13, 5.69s/it]
9%|▊ | 3799/43939 [6:08:22<63:30:20, 5.70s/it]
9%|▊ | 3800/43939 [6:08:28<63:39:12, 5.71s/it]
{'loss': 1.0733, 'grad_norm': 1.2468562126159668, 'learning_rate': 4.323924425221944e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.65, 'epoch': 0.35}
9%|▊ | 3800/43939 [6:08:28<63:39:12, 5.71s/it]
9%|▊ | 3801/43939 [6:08:34<63:36:37, 5.71s/it]
9%|▊ | 3802/43939 [6:08:40<63:38:47, 5.71s/it]
9%|▊ | 3803/43939 [6:08:45<63:34:04, 5.70s/it]
9%|▊ | 3804/43939 [6:08:51<63:32:06, 5.70s/it]
9%|▊ | 3805/43939 [6:08:57<63:22:02, 5.68s/it]
9%|▊ | 3806/43939 [6:09:02<63:29:49, 5.70s/it]
9%|▊ | 3807/43939 [6:09:08<63:31:39, 5.70s/it]
9%|▊ | 3808/43939 [6:09:14<63:19:32, 5.68s/it]
9%|▊ | 3809/43939 [6:09:19<63:26:02, 5.69s/it]
9%|▊ | 3810/43939 [6:09:25<63:34:47, 5.70s/it]
9%|▊ | 3811/43939 [6:09:31<64:21:23, 5.77s/it]
9%|▊ | 3812/43939 [6:09:37<64:11:45, 5.76s/it]
9%|▊ | 3813/43939 [6:09:42<63:56:43, 5.74s/it]
9%|▊ | 3814/43939 [6:09:48<63:46:34, 5.72s/it]
9%|▊ | 3815/43939 [6:09:54<63:39:07, 5.71s/it]
9%|▊ | 3816/43939 [6:10:00<63:48:27, 5.73s/it]
9%|▊ | 3817/43939 [6:10:05<63:49:35, 5.73s/it]
9%|▊ | 3818/43939 [6:10:11<63:44:16, 5.72s/it]
9%|▊ | 3819/43939 [6:10:17<63:45:55, 5.72s/it]
9%|▊ | 3820/43939 [6:10:22<63:42:12, 5.72s/it]
9%|▊ | 3821/43939 [6:10:28<63:34:05, 5.70s/it]
9%|▊ | 3822/43939 [6:10:34<63:37:28, 5.71s/it]
9%|▊ | 3823/43939 [6:10:39<63:31:04, 5.70s/it]
9%|▊ | 3824/43939 [6:10:45<63:27:30, 5.69s/it]
9%|▊ | 3825/43939 [6:10:51<63:22:42, 5.69s/it]
9%|▊ | 3826/43939 [6:10:57<63:30:59, 5.70s/it]
9%|▊ | 3827/43939 [6:11:02<63:26:16, 5.69s/it]
9%|▊ | 3828/43939 [6:11:08<63:27:28, 5.70s/it]
9%|▊ | 3829/43939 [6:11:14<63:26:00, 5.69s/it]
9%|▊ | 3830/43939 [6:11:19<63:35:19, 5.71s/it]
9%|▊ | 3831/43939 [6:11:25<64:17:25, 5.77s/it]
9%|▊ | 3832/43939 [6:11:31<64:05:41, 5.75s/it]
9%|▊ | 3833/43939 [6:11:37<63:53:13, 5.73s/it]
9%|▊ | 3834/43939 [6:11:42<63:45:00, 5.72s/it]
9%|▊ | 3835/43939 [6:11:48<63:47:08, 5.73s/it]
9%|▊ | 3836/43939 [6:11:54<63:37:44, 5.71s/it]
9%|▊ | 3837/43939 [6:11:59<63:35:28, 5.71s/it]
9%|▊ | 3838/43939 [6:12:05<63:29:43, 5.70s/it]
9%|▊ | 3839/43939 [6:12:11<63:31:46, 5.70s/it]
9%|▊ | 3840/43939 [6:12:17<63:28:59, 5.70s/it]
9%|▊ | 3841/43939 [6:12:22<63:25:17, 5.69s/it]
9%|▊ | 3842/43939 [6:12:28<63:23:26, 5.69s/it]
9%|▊ | 3843/43939 [6:12:34<63:21:52, 5.69s/it]
9%|▊ | 3844/43939 [6:12:39<63:28:34, 5.70s/it]
9%|▉ | 3845/43939 [6:12:45<63:28:42, 5.70s/it]
9%|▉ | 3846/43939 [6:12:51<63:21:58, 5.69s/it]
9%|▉ | 3847/43939 [6:12:56<63:18:31, 5.68s/it]
9%|▉ | 3848/43939 [6:13:02<63:13:57, 5.68s/it]
9%|▉ | 3849/43939 [6:13:08<63:17:51, 5.68s/it]
9%|▉ | 3850/43939 [6:13:13<63:24:39, 5.69s/it]
{'loss': 1.098, 'grad_norm': 1.1006118059158325, 'learning_rate': 4.380833143637606e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.27, 'epoch': 0.35}
9%|▉ | 3850/43939 [6:13:13<63:24:39, 5.69s/it]
9%|▉ | 3851/43939 [6:13:19<63:19:45, 5.69s/it]
9%|▉ | 3852/43939 [6:13:25<64:03:39, 5.75s/it]
9%|▉ | 3853/43939 [6:13:31<63:59:20, 5.75s/it]
9%|▉ | 3854/43939 [6:13:36<63:44:38, 5.72s/it]
9%|▉ | 3855/43939 [6:13:42<63:39:13, 5.72s/it]
9%|▉ | 3856/43939 [6:13:48<63:43:46, 5.72s/it]
9%|▉ | 3857/43939 [6:13:54<63:50:29, 5.73s/it]
9%|▉ | 3858/43939 [6:13:59<63:39:58, 5.72s/it]
9%|▉ | 3859/43939 [6:14:05<63:51:12, 5.74s/it]
9%|▉ | 3860/43939 [6:14:11<63:47:38, 5.73s/it]
9%|▉ | 3861/43939 [6:14:17<63:42:27, 5.72s/it]
9%|▉ | 3862/43939 [6:14:22<63:42:09, 5.72s/it]
9%|▉ | 3863/43939 [6:14:28<63:43:13, 5.72s/it]
9%|▉ | 3864/43939 [6:14:34<63:34:22, 5.71s/it]
9%|▉ | 3865/43939 [6:14:39<63:34:30, 5.71s/it]
9%|▉ | 3866/43939 [6:14:45<63:43:35, 5.72s/it]
9%|▉ | 3867/43939 [6:14:51<63:34:33, 5.71s/it]
9%|▉ | 3868/43939 [6:14:56<63:29:31, 5.70s/it]
9%|▉ | 3869/43939 [6:15:02<63:26:04, 5.70s/it]
9%|▉ | 3870/43939 [6:15:08<63:19:08, 5.69s/it]
9%|▉ | 3871/43939 [6:15:14<63:28:07, 5.70s/it]
9%|▉ | 3872/43939 [6:15:19<63:26:25, 5.70s/it]
9%|▉ | 3873/43939 [6:15:25<63:30:52, 5.71s/it]
9%|▉ | 3874/43939 [6:15:31<64:17:11, 5.78s/it]
9%|▉ | 3875/43939 [6:15:37<64:01:44, 5.75s/it]
9%|▉ | 3876/43939 [6:15:42<63:47:55, 5.73s/it]
9%|▉ | 3877/43939 [6:15:48<63:36:19, 5.72s/it]
9%|▉ | 3878/43939 [6:15:54<63:35:52, 5.72s/it]
9%|▉ | 3879/43939 [6:15:59<63:36:56, 5.72s/it]
9%|▉ | 3880/43939 [6:16:05<63:29:40, 5.71s/it]
9%|▉ | 3881/43939 [6:16:11<63:32:03, 5.71s/it]
9%|▉ | 3882/43939 [6:16:17<63:27:19, 5.70s/it]
9%|▉ | 3883/43939 [6:16:22<63:18:52, 5.69s/it]
9%|▉ | 3884/43939 [6:16:28<63:16:42, 5.69s/it]
9%|▉ | 3885/43939 [6:16:34<63:29:43, 5.71s/it]
9%|▉ | 3886/43939 [6:16:39<63:26:58, 5.70s/it]
9%|▉ | 3887/43939 [6:16:45<63:15:09, 5.69s/it]
9%|▉ | 3888/43939 [6:16:51<63:16:35, 5.69s/it]
9%|▉ | 3889/43939 [6:16:56<63:21:47, 5.70s/it]
9%|▉ | 3890/43939 [6:17:02<63:37:03, 5.72s/it]
9%|▉ | 3891/43939 [6:17:08<63:39:59, 5.72s/it]
9%|▉ | 3892/43939 [6:17:14<63:33:52, 5.71s/it]
9%|▉ | 3893/43939 [6:17:19<63:28:20, 5.71s/it]
9%|▉ | 3894/43939 [6:17:25<64:10:38, 5.77s/it]
9%|▉ | 3895/43939 [6:17:31<63:55:57, 5.75s/it]
9%|▉ | 3896/43939 [6:17:37<63:40:05, 5.72s/it]
9%|▉ | 3897/43939 [6:17:42<63:46:45, 5.73s/it]
9%|▉ | 3898/43939 [6:17:48<63:38:15, 5.72s/it]
9%|▉ | 3899/43939 [6:17:54<63:35:54, 5.72s/it]
9%|▉ | 3900/43939 [6:17:59<63:31:34, 5.71s/it]
{'loss': 1.1089, 'grad_norm': 1.1994727849960327, 'learning_rate': 4.437741862053267e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.68, 'epoch': 0.36}
9%|▉ | 3900/43939 [6:17:59<63:31:34, 5.71s/it]
9%|▉ | 3901/43939 [6:18:05<63:33:25, 5.71s/it]
9%|▉ | 3902/43939 [6:18:11<63:28:40, 5.71s/it]
9%|▉ | 3903/43939 [6:18:16<63:23:41, 5.70s/it]
9%|▉ | 3904/43939 [6:18:22<63:27:45, 5.71s/it]
9%|▉ | 3905/43939 [6:18:28<63:20:57, 5.70s/it]
9%|▉ | 3906/43939 [6:18:34<63:33:29, 5.72s/it]
9%|▉ | 3907/43939 [6:18:39<63:26:58, 5.71s/it]
9%|▉ | 3908/43939 [6:18:45<63:20:05, 5.70s/it]
9%|▉ | 3909/43939 [6:18:51<63:19:47, 5.70s/it]
9%|▉ | 3910/43939 [6:18:56<63:16:30, 5.69s/it]
9%|▉ | 3911/43939 [6:19:02<63:14:16, 5.69s/it]
9%|▉ | 3912/43939 [6:19:08<63:12:46, 5.69s/it]
9%|▉ | 3913/43939 [6:19:13<63:10:21, 5.68s/it]
9%|▉ | 3914/43939 [6:19:19<63:08:29, 5.68s/it]
9%|▉ | 3915/43939 [6:19:25<63:56:42, 5.75s/it]
9%|▉ | 3916/43939 [6:19:31<63:42:18, 5.73s/it]
9%|▉ | 3917/43939 [6:19:36<63:33:32, 5.72s/it]
9%|▉ | 3918/43939 [6:19:42<63:28:20, 5.71s/it]
9%|▉ | 3919/43939 [6:19:48<63:38:50, 5.73s/it]
9%|▉ | 3920/43939 [6:19:54<63:38:03, 5.72s/it]
9%|▉ | 3921/43939 [6:19:59<63:29:25, 5.71s/it]
9%|▉ | 3922/43939 [6:20:05<63:28:41, 5.71s/it]
9%|▉ | 3923/43939 [6:20:11<63:19:35, 5.70s/it]
9%|▉ | 3924/43939 [6:20:16<63:22:27, 5.70s/it]
9%|▉ | 3925/43939 [6:20:22<63:17:18, 5.69s/it]
9%|▉ | 3926/43939 [6:20:28<63:13:38, 5.69s/it]
9%|▉ | 3927/43939 [6:20:33<63:15:03, 5.69s/it]
9%|▉ | 3928/43939 [6:20:39<63:20:59, 5.70s/it]
9%|▉ | 3929/43939 [6:20:45<63:12:03, 5.69s/it]
9%|▉ | 3930/43939 [6:20:50<63:21:11, 5.70s/it]
9%|▉ | 3931/43939 [6:20:56<63:11:59, 5.69s/it]
9%|▉ | 3932/43939 [6:21:02<63:13:14, 5.69s/it]
9%|▉ | 3933/43939 [6:21:08<63:17:55, 5.70s/it]
9%|▉ | 3934/43939 [6:21:13<63:17:52, 5.70s/it]
9%|▉ | 3935/43939 [6:21:19<63:11:25, 5.69s/it]
9%|▉ | 3936/43939 [6:21:25<63:13:53, 5.69s/it]
9%|▉ | 3937/43939 [6:21:30<63:57:55, 5.76s/it]
9%|▉ | 3938/43939 [6:21:36<63:48:53, 5.74s/it]
9%|▉ | 3939/43939 [6:21:42<63:40:32, 5.73s/it]
9%|▉ | 3940/43939 [6:21:48<63:38:26, 5.73s/it]
9%|▉ | 3941/43939 [6:21:53<63:30:18, 5.72s/it]
9%|▉ | 3942/43939 [6:21:59<63:25:39, 5.71s/it]
9%|▉ | 3943/43939 [6:22:05<63:22:06, 5.70s/it]
9%|▉ | 3944/43939 [6:22:10<63:19:57, 5.70s/it]
9%|▉ | 3945/43939 [6:22:16<63:33:09, 5.72s/it]
9%|▉ | 3946/43939 [6:22:22<63:28:45, 5.71s/it]
9%|▉ | 3947/43939 [6:22:28<63:26:54, 5.71s/it]
9%|▉ | 3948/43939 [6:22:33<63:18:42, 5.70s/it]
9%|▉ | 3949/43939 [6:22:39<63:10:09, 5.69s/it]
9%|▉ | 3950/43939 [6:22:45<63:11:49, 5.69s/it]
{'loss': 1.0693, 'grad_norm': 1.166426181793213, 'learning_rate': 4.4946505804689283e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 447.81, 'epoch': 0.36}
9%|▉ | 3950/43939 [6:22:45<63:11:49, 5.69s/it]
9%|▉ | 3951/43939 [6:22:50<63:17:13, 5.70s/it]
9%|▉ | 3952/43939 [6:22:56<63:15:20, 5.69s/it]
9%|▉ | 3953/43939 [6:23:02<63:17:42, 5.70s/it]
9%|▉ | 3954/43939 [6:23:07<63:21:32, 5.70s/it]
9%|▉ | 3955/43939 [6:23:13<63:15:20, 5.70s/it]
9%|▉ | 3956/43939 [6:23:19<63:11:12, 5.69s/it]
9%|▉ | 3957/43939 [6:23:25<63:22:41, 5.71s/it]
9%|▉ | 3958/43939 [6:23:30<63:56:52, 5.76s/it]
9%|▉ | 3959/43939 [6:23:36<63:40:03, 5.73s/it]
9%|▉ | 3960/43939 [6:23:42<63:35:37, 5.73s/it]
9%|▉ | 3961/43939 [6:23:47<63:33:01, 5.72s/it]
9%|▉ | 3962/43939 [6:23:53<63:23:42, 5.71s/it]
9%|▉ | 3963/43939 [6:23:59<63:22:57, 5.71s/it]
9%|▉ | 3964/43939 [6:24:05<63:22:46, 5.71s/it]
9%|▉ | 3965/43939 [6:24:10<63:23:38, 5.71s/it]
9%|▉ | 3966/43939 [6:24:16<63:25:46, 5.71s/it]
9%|▉ | 3967/43939 [6:24:22<63:11:46, 5.69s/it]
9%|▉ | 3968/43939 [6:24:27<63:10:40, 5.69s/it]
9%|▉ | 3969/43939 [6:24:33<63:20:16, 5.70s/it]
9%|▉ | 3970/43939 [6:24:39<63:17:40, 5.70s/it]
9%|▉ | 3971/43939 [6:24:44<63:16:28, 5.70s/it]
9%|▉ | 3972/43939 [6:24:50<63:27:02, 5.72s/it]
9%|▉ | 3973/43939 [6:24:56<63:23:28, 5.71s/it]
9%|▉ | 3974/43939 [6:25:02<63:17:02, 5.70s/it]
9%|▉ | 3975/43939 [6:25:07<63:21:18, 5.71s/it]
9%|▉ | 3976/43939 [6:25:13<63:15:35, 5.70s/it]
9%|▉ | 3977/43939 [6:25:19<63:10:59, 5.69s/it]
9%|▉ | 3978/43939 [6:25:24<63:14:26, 5.70s/it]
9%|▉ | 3979/43939 [6:25:30<63:54:50, 5.76s/it]
9%|▉ | 3980/43939 [6:25:36<63:45:03, 5.74s/it]
9%|▉ | 3981/43939 [6:25:42<63:27:45, 5.72s/it]
9%|▉ | 3982/43939 [6:25:47<63:27:41, 5.72s/it]
9%|▉ | 3983/43939 [6:25:53<63:16:38, 5.70s/it]
9%|▉ | 3984/43939 [6:25:59<63:17:55, 5.70s/it]
9%|▉ | 3985/43939 [6:26:04<63:14:08, 5.70s/it]
9%|▉ | 3986/43939 [6:26:10<63:08:55, 5.69s/it]
9%|▉ | 3987/43939 [6:26:16<63:16:32, 5.70s/it]
9%|▉ | 3988/43939 [6:26:22<63:23:49, 5.71s/it]
9%|▉ | 3989/43939 [6:26:27<63:22:38, 5.71s/it]
9%|▉ | 3990/43939 [6:26:33<63:36:09, 5.73s/it]
9%|▉ | 3991/43939 [6:26:39<63:29:50, 5.72s/it]
9%|▉ | 3992/43939 [6:26:44<63:20:26, 5.71s/it]
9%|▉ | 3993/43939 [6:26:50<63:17:34, 5.70s/it]
9%|▉ | 3994/43939 [6:26:56<63:07:09, 5.69s/it]
9%|▉ | 3995/43939 [6:27:01<63:11:49, 5.70s/it]
9%|▉ | 3996/43939 [6:27:07<63:11:06, 5.69s/it]
9%|▉ | 3997/43939 [6:27:13<63:17:49, 5.71s/it]
9%|▉ | 3998/43939 [6:27:19<63:18:21, 5.71s/it]
9%|▉ | 3999/43939 [6:27:24<63:20:25, 5.71s/it]
9%|▉ | 4000/43939 [6:27:30<64:14:02, 5.79s/it]
{'loss': 1.1004, 'grad_norm': 1.003192663192749, 'learning_rate': 4.551559298884589e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.48, 'epoch': 0.36}
9%|▉ | 4000/43939 [6:27:30<64:14:02, 5.79s/it]
9%|▉ | 4001/43939 [6:27:36<64:00:50, 5.77s/it]
9%|▉ | 4002/43939 [6:27:42<63:41:52, 5.74s/it]
9%|▉ | 4003/43939 [6:27:47<63:35:17, 5.73s/it]
9%|▉ | 4004/43939 [6:27:53<63:33:07, 5.73s/it]
9%|▉ | 4005/43939 [6:27:59<63:23:45, 5.72s/it]
9%|▉ | 4006/43939 [6:28:05<63:26:03, 5.72s/it]
9%|▉ | 4007/43939 [6:28:10<63:23:51, 5.72s/it]
9%|▉ | 4008/43939 [6:28:16<63:18:51, 5.71s/it]
9%|▉ | 4009/43939 [6:28:22<63:22:37, 5.71s/it]
9%|▉ | 4010/43939 [6:28:27<63:22:49, 5.71s/it]
9%|▉ | 4011/43939 [6:28:33<63:22:47, 5.71s/it]
9%|▉ | 4012/43939 [6:28:39<63:27:15, 5.72s/it]
9%|▉ | 4013/43939 [6:28:45<63:18:53, 5.71s/it]
9%|▉ | 4014/43939 [6:28:50<63:22:34, 5.71s/it]
9%|▉ | 4015/43939 [6:28:56<63:19:18, 5.71s/it]
9%|▉ | 4016/43939 [6:29:02<63:18:26, 5.71s/it]
9%|▉ | 4017/43939 [6:29:07<63:19:34, 5.71s/it]
9%|▉ | 4018/43939 [6:29:13<63:21:17, 5.71s/it]
9%|▉ | 4019/43939 [6:29:19<63:16:17, 5.71s/it]
9%|▉ | 4020/43939 [6:29:24<63:10:22, 5.70s/it]
9%|▉ | 4021/43939 [6:29:30<63:53:53, 5.76s/it]
9%|▉ | 4022/43939 [6:29:36<63:41:24, 5.74s/it]
9%|▉ | 4023/43939 [6:29:42<63:33:37, 5.73s/it]
9%|▉ | 4024/43939 [6:29:48<63:37:19, 5.74s/it]
9%|▉ | 4025/43939 [6:29:53<63:23:10, 5.72s/it]
9%|▉ | 4026/43939 [6:29:59<63:21:49, 5.72s/it]
9%|▉ | 4027/43939 [6:30:05<63:19:16, 5.71s/it]
9%|▉ | 4028/43939 [6:30:10<63:17:29, 5.71s/it]
9%|▉ | 4029/43939 [6:30:16<63:17:33, 5.71s/it]
9%|▉ | 4030/43939 [6:30:22<63:07:37, 5.69s/it]
9%|▉ | 4031/43939 [6:30:27<63:11:37, 5.70s/it]
9%|▉ | 4032/43939 [6:30:33<63:19:01, 5.71s/it]
9%|▉ | 4033/43939 [6:30:39<63:29:06, 5.73s/it]
9%|▉ | 4034/43939 [6:30:45<63:21:55, 5.72s/it]
9%|▉ | 4035/43939 [6:30:50<63:12:52, 5.70s/it]
9%|▉ | 4036/43939 [6:30:56<63:10:50, 5.70s/it]
9%|▉ | 4037/43939 [6:31:02<63:14:27, 5.71s/it]
9%|▉ | 4038/43939 [6:31:07<63:04:47, 5.69s/it]
9%|▉ | 4039/43939 [6:31:13<63:00:00, 5.68s/it]
9%|▉ | 4040/43939 [6:31:19<63:03:00, 5.69s/it]
9%|▉ | 4041/43939 [6:31:24<63:09:10, 5.70s/it]
9%|▉ | 4042/43939 [6:31:30<63:51:25, 5.76s/it]
9%|▉ | 4043/43939 [6:31:36<63:38:20, 5.74s/it]
9%|▉ | 4044/43939 [6:31:42<63:21:15, 5.72s/it]
9%|▉ | 4045/43939 [6:31:47<63:15:28, 5.71s/it]
9%|▉ | 4046/43939 [6:31:53<63:11:26, 5.70s/it]
9%|▉ | 4047/43939 [6:31:59<63:07:18, 5.70s/it]
9%|▉ | 4048/43939 [6:32:04<63:01:15, 5.69s/it]
9%|▉ | 4049/43939 [6:32:10<62:57:11, 5.68s/it]
9%|▉ | 4050/43939 [6:32:16<63:05:21, 5.69s/it]
{'loss': 1.0936, 'grad_norm': 0.9891825318336487, 'learning_rate': 4.60846801730025e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.97, 'epoch': 0.37}
9%|▉ | 4050/43939 [6:32:16<63:05:21, 5.69s/it]
9%|▉ | 4051/43939 [6:32:22<63:12:10, 5.70s/it]
9%|▉ | 4052/43939 [6:32:27<63:10:50, 5.70s/it]
9%|▉ | 4053/43939 [6:32:33<63:02:13, 5.69s/it]
9%|▉ | 4054/43939 [6:32:39<62:59:09, 5.69s/it]
9%|▉ | 4055/43939 [6:32:44<62:59:47, 5.69s/it]
9%|▉ | 4056/43939 [6:32:50<63:02:10, 5.69s/it]
9%|▉ | 4057/43939 [6:32:56<63:03:41, 5.69s/it]
9%|▉ | 4058/43939 [6:33:01<63:13:26, 5.71s/it]
9%|▉ | 4059/43939 [6:33:07<63:17:14, 5.71s/it]
9%|▉ | 4060/43939 [6:33:13<63:11:11, 5.70s/it]
9%|▉ | 4061/43939 [6:33:18<63:07:16, 5.70s/it]
9%|▉ | 4062/43939 [6:33:24<63:45:55, 5.76s/it]
9%|▉ | 4063/43939 [6:33:30<63:39:48, 5.75s/it]
9%|▉ | 4064/43939 [6:33:36<63:33:54, 5.74s/it]
9%|▉ | 4065/43939 [6:33:42<63:26:16, 5.73s/it]
9%|▉ | 4066/43939 [6:33:47<63:12:43, 5.71s/it]
9%|▉ | 4067/43939 [6:33:53<63:08:19, 5.70s/it]
9%|▉ | 4068/43939 [6:33:59<63:08:05, 5.70s/it]
9%|▉ | 4069/43939 [6:34:04<63:18:19, 5.72s/it]
9%|▉ | 4070/43939 [6:34:10<63:23:45, 5.72s/it]
9%|▉ | 4071/43939 [6:34:16<63:20:32, 5.72s/it]
9%|▉ | 4072/43939 [6:34:21<63:10:33, 5.70s/it]
9%|▉ | 4073/43939 [6:34:27<63:09:04, 5.70s/it]
9%|▉ | 4074/43939 [6:34:33<63:12:40, 5.71s/it]
9%|▉ | 4075/43939 [6:34:39<63:09:11, 5.70s/it]
9%|▉ | 4076/43939 [6:34:44<63:22:05, 5.72s/it]
9%|▉ | 4077/43939 [6:34:50<63:17:07, 5.72s/it]
9%|▉ | 4078/43939 [6:34:56<63:09:15, 5.70s/it]
9%|▉ | 4079/43939 [6:35:01<63:20:33, 5.72s/it]
9%|▉ | 4080/43939 [6:35:07<63:19:45, 5.72s/it]
9%|▉ | 4081/43939 [6:35:13<63:09:44, 5.70s/it]
9%|▉ | 4082/43939 [6:35:19<63:10:11, 5.71s/it]
9%|▉ | 4083/43939 [6:35:24<63:13:35, 5.71s/it]
9%|▉ | 4084/43939 [6:35:30<63:55:07, 5.77s/it]
9%|▉ | 4085/43939 [6:35:36<63:37:04, 5.75s/it]
9%|▉ | 4086/43939 [6:35:42<63:34:27, 5.74s/it]
9%|▉ | 4087/43939 [6:35:47<63:20:15, 5.72s/it]
9%|▉ | 4088/43939 [6:35:53<63:18:53, 5.72s/it]
9%|▉ | 4089/43939 [6:35:59<63:12:29, 5.71s/it]
9%|▉ | 4090/43939 [6:36:04<63:05:35, 5.70s/it]
9%|▉ | 4091/43939 [6:36:10<63:02:16, 5.70s/it]
9%|▉ | 4092/43939 [6:36:16<63:14:50, 5.71s/it]
9%|▉ | 4093/43939 [6:36:22<63:18:11, 5.72s/it]
9%|▉ | 4094/43939 [6:36:27<63:09:51, 5.71s/it]
9%|▉ | 4095/43939 [6:36:33<63:06:47, 5.70s/it]
9%|▉ | 4096/43939 [6:36:39<63:09:29, 5.71s/it]
9%|▉ | 4097/43939 [6:36:44<63:19:57, 5.72s/it]
9%|▉ | 4098/43939 [6:36:50<63:19:24, 5.72s/it]
9%|▉ | 4099/43939 [6:36:56<63:25:15, 5.73s/it]
9%|▉ | 4100/43939 [6:37:02<63:16:18, 5.72s/it]
{'loss': 1.0909, 'grad_norm': 1.1373133659362793, 'learning_rate': 4.665376735715912e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.57, 'epoch': 0.37}
9%|▉ | 4100/43939 [6:37:02<63:16:18, 5.72s/it]
9%|▉ | 4101/43939 [6:37:07<63:15:16, 5.72s/it]
9%|▉ | 4102/43939 [6:37:13<63:15:45, 5.72s/it]
9%|▉ | 4103/43939 [6:37:19<63:11:07, 5.71s/it]
9%|▉ | 4104/43939 [6:37:24<63:10:33, 5.71s/it]
9%|▉ | 4105/43939 [6:37:30<63:51:03, 5.77s/it]
9%|▉ | 4106/43939 [6:37:36<63:32:38, 5.74s/it]
9%|▉ | 4107/43939 [6:37:42<63:25:05, 5.73s/it]
9%|▉ | 4108/43939 [6:37:47<63:16:52, 5.72s/it]
9%|▉ | 4109/43939 [6:37:53<63:09:23, 5.71s/it]
9%|▉ | 4110/43939 [6:37:59<63:15:38, 5.72s/it]
9%|▉ | 4111/43939 [6:38:04<63:10:14, 5.71s/it]
9%|▉ | 4112/43939 [6:38:10<63:10:20, 5.71s/it]
9%|▉ | 4113/43939 [6:38:16<63:13:58, 5.72s/it]
9%|▉ | 4114/43939 [6:38:22<63:10:02, 5.71s/it]
9%|▉ | 4115/43939 [6:38:27<62:57:54, 5.69s/it]
9%|▉ | 4116/43939 [6:38:33<63:06:37, 5.71s/it]
9%|▉ | 4117/43939 [6:38:39<63:04:55, 5.70s/it]
9%|▉ | 4118/43939 [6:38:44<63:11:02, 5.71s/it]
9%|▉ | 4119/43939 [6:38:50<63:03:38, 5.70s/it]
9%|▉ | 4120/43939 [6:38:56<62:57:10, 5.69s/it]
9%|▉ | 4121/43939 [6:39:01<62:59:50, 5.70s/it]
9%|▉ | 4122/43939 [6:39:07<63:00:43, 5.70s/it]
9%|▉ | 4123/43939 [6:39:13<62:57:04, 5.69s/it]
9%|▉ | 4124/43939 [6:39:19<62:57:06, 5.69s/it]
9%|▉ | 4125/43939 [6:39:24<63:36:38, 5.75s/it]
9%|▉ | 4126/43939 [6:39:30<63:26:40, 5.74s/it]
9%|▉ | 4127/43939 [6:39:36<63:29:51, 5.74s/it]
9%|▉ | 4128/43939 [6:39:42<63:21:59, 5.73s/it]
9%|▉ | 4129/43939 [6:39:47<63:15:42, 5.72s/it]
9%|▉ | 4130/43939 [6:39:53<63:11:08, 5.71s/it]
9%|▉ | 4131/43939 [6:39:59<63:07:24, 5.71s/it]
9%|▉ | 4132/43939 [6:40:04<63:04:45, 5.70s/it]
9%|▉ | 4133/43939 [6:40:10<63:07:30, 5.71s/it]
9%|▉ | 4134/43939 [6:40:16<62:56:30, 5.69s/it]
9%|▉ | 4135/43939 [6:40:21<62:57:24, 5.69s/it]
9%|▉ | 4136/43939 [6:40:27<63:03:29, 5.70s/it]
9%|▉ | 4137/43939 [6:40:33<63:00:24, 5.70s/it]
9%|▉ | 4138/43939 [6:40:39<63:01:42, 5.70s/it]
9%|▉ | 4139/43939 [6:40:44<63:04:38, 5.71s/it]
9%|▉ | 4140/43939 [6:40:50<62:56:14, 5.69s/it]
9%|▉ | 4141/43939 [6:40:56<63:04:05, 5.70s/it]
9%|▉ | 4142/43939 [6:41:01<63:08:07, 5.71s/it]
9%|▉ | 4143/43939 [6:41:07<62:58:50, 5.70s/it]
9%|▉ | 4144/43939 [6:41:13<62:52:59, 5.69s/it]
9%|▉ | 4145/43939 [6:41:18<62:52:57, 5.69s/it]
9%|▉ | 4146/43939 [6:41:24<62:44:00, 5.68s/it]
9%|▉ | 4147/43939 [6:41:30<63:25:49, 5.74s/it]
9%|▉ | 4148/43939 [6:41:36<63:20:15, 5.73s/it]
9%|▉ | 4149/43939 [6:41:41<63:10:44, 5.72s/it]
9%|▉ | 4150/43939 [6:41:47<63:00:45, 5.70s/it]
{'loss': 1.0913, 'grad_norm': 1.149749517440796, 'learning_rate': 4.722285454131573e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.49, 'epoch': 0.38}
9%|▉ | 4150/43939 [6:41:47<63:00:45, 5.70s/it]
9%|▉ | 4151/43939 [6:41:53<63:03:20, 5.71s/it]
9%|▉ | 4152/43939 [6:41:58<63:05:31, 5.71s/it]
9%|▉ | 4153/43939 [6:42:04<63:01:51, 5.70s/it]
9%|▉ | 4154/43939 [6:42:10<63:05:18, 5.71s/it]
9%|▉ | 4155/43939 [6:42:16<62:56:43, 5.70s/it]
9%|▉ | 4156/43939 [6:42:21<62:52:18, 5.69s/it]
9%|▉ | 4157/43939 [6:42:27<62:56:43, 5.70s/it]
9%|▉ | 4158/43939 [6:42:33<62:56:17, 5.70s/it]
9%|▉ | 4159/43939 [6:42:38<62:57:13, 5.70s/it]
9%|▉ | 4160/43939 [6:42:44<63:02:31, 5.71s/it]
9%|▉ | 4161/43939 [6:42:50<63:15:43, 5.73s/it]
9%|▉ | 4162/43939 [6:42:56<63:08:00, 5.71s/it]
9%|▉ | 4163/43939 [6:43:01<63:01:25, 5.70s/it]
9%|▉ | 4164/43939 [6:43:07<62:59:11, 5.70s/it]
9%|▉ | 4165/43939 [6:43:13<63:03:40, 5.71s/it]
9%|▉ | 4166/43939 [6:43:18<63:08:05, 5.71s/it]
9%|▉ | 4167/43939 [6:43:24<63:09:16, 5.72s/it]
9%|▉ | 4168/43939 [6:43:30<63:48:20, 5.78s/it]
9%|▉ | 4169/43939 [6:43:36<63:33:29, 5.75s/it]
9%|▉ | 4170/43939 [6:43:41<63:19:14, 5.73s/it]
9%|▉ | 4171/43939 [6:43:47<63:20:04, 5.73s/it]
9%|▉ | 4172/43939 [6:43:53<63:25:44, 5.74s/it]
9%|▉ | 4173/43939 [6:43:59<63:19:38, 5.73s/it]
9%|▉ | 4174/43939 [6:44:04<63:25:02, 5.74s/it]
10%|▉ | 4175/43939 [6:44:10<63:16:42, 5.73s/it]
10%|▉ | 4176/43939 [6:44:16<63:23:46, 5.74s/it]
10%|▉ | 4177/43939 [6:44:21<63:11:19, 5.72s/it]
10%|▉ | 4178/43939 [6:44:27<63:11:56, 5.72s/it]
10%|▉ | 4179/43939 [6:44:33<63:09:35, 5.72s/it]
10%|▉ | 4180/43939 [6:44:39<63:12:02, 5.72s/it]
10%|▉ | 4181/43939 [6:44:44<63:15:02, 5.73s/it]
10%|▉ | 4182/43939 [6:44:50<63:03:54, 5.71s/it]
10%|▉ | 4183/43939 [6:44:56<63:10:16, 5.72s/it]
10%|▉ | 4184/43939 [6:45:01<63:02:59, 5.71s/it]
10%|▉ | 4185/43939 [6:45:07<62:53:55, 5.70s/it]
10%|▉ | 4186/43939 [6:45:13<62:49:40, 5.69s/it]
10%|▉ | 4187/43939 [6:45:19<62:51:35, 5.69s/it]
10%|▉ | 4188/43939 [6:45:24<62:53:48, 5.70s/it]
10%|▉ | 4189/43939 [6:45:30<63:46:01, 5.78s/it]
10%|▉ | 4190/43939 [6:45:36<63:26:25, 5.75s/it]
10%|▉ | 4191/43939 [6:45:42<63:18:31, 5.73s/it]
10%|▉ | 4192/43939 [6:45:47<63:15:17, 5.73s/it]
10%|▉ | 4193/43939 [6:45:53<63:10:55, 5.72s/it]
10%|▉ | 4194/43939 [6:45:59<63:02:28, 5.71s/it]
10%|▉ | 4195/43939 [6:46:04<63:05:06, 5.71s/it]
10%|▉ | 4196/43939 [6:46:10<63:01:41, 5.71s/it]
10%|▉ | 4197/43939 [6:46:16<62:57:21, 5.70s/it]
10%|▉ | 4198/43939 [6:46:21<62:59:04, 5.71s/it]
10%|▉ | 4199/43939 [6:46:27<62:57:30, 5.70s/it]
10%|▉ | 4200/43939 [6:46:33<62:55:02, 5.70s/it]
{'loss': 1.0951, 'grad_norm': 1.1793601512908936, 'learning_rate': 4.7791941725472346e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 532.82, 'epoch': 0.38}
10%|▉ | 4200/43939 [6:46:33<62:55:02, 5.70s/it]
10%|▉ | 4201/43939 [6:46:39<62:54:06, 5.70s/it]
10%|▉ | 4202/43939 [6:46:44<63:10:08, 5.72s/it]
10%|▉ | 4203/43939 [6:46:50<63:00:39, 5.71s/it]
10%|▉ | 4204/43939 [6:46:56<62:59:03, 5.71s/it]
10%|▉ | 4205/43939 [6:47:01<62:54:27, 5.70s/it]
10%|▉ | 4206/43939 [6:47:07<62:48:59, 5.69s/it]
10%|▉ | 4207/43939 [6:47:13<62:55:47, 5.70s/it]
10%|▉ | 4208/43939 [6:47:19<62:55:17, 5.70s/it]
10%|▉ | 4209/43939 [6:47:24<63:46:04, 5.78s/it]
10%|▉ | 4210/43939 [6:47:30<63:24:14, 5.75s/it]
10%|▉ | 4211/43939 [6:47:36<63:15:39, 5.73s/it]
10%|▉ | 4212/43939 [6:47:42<63:06:02, 5.72s/it]
10%|▉ | 4213/43939 [6:47:47<62:56:53, 5.70s/it]
10%|▉ | 4214/43939 [6:47:53<63:03:38, 5.71s/it]
10%|▉ | 4215/43939 [6:47:59<62:58:46, 5.71s/it]
10%|▉ | 4216/43939 [6:48:04<62:55:25, 5.70s/it]
10%|▉ | 4217/43939 [6:48:10<62:44:17, 5.69s/it]
10%|▉ | 4218/43939 [6:48:16<62:46:08, 5.69s/it]
10%|▉ | 4219/43939 [6:48:21<62:49:33, 5.69s/it]
10%|▉ | 4220/43939 [6:48:27<62:50:20, 5.70s/it]
10%|▉ | 4221/43939 [6:48:33<62:54:27, 5.70s/it]
10%|▉ | 4222/43939 [6:48:38<62:52:58, 5.70s/it]
10%|▉ | 4223/43939 [6:48:44<62:51:02, 5.70s/it]
10%|▉ | 4224/43939 [6:48:50<62:55:14, 5.70s/it]
10%|▉ | 4225/43939 [6:48:56<62:56:40, 5.71s/it]
10%|▉ | 4226/43939 [6:49:01<62:49:04, 5.69s/it]
10%|▉ | 4227/43939 [6:49:07<62:45:38, 5.69s/it]
10%|▉ | 4228/43939 [6:49:13<62:45:37, 5.69s/it]
10%|▉ | 4229/43939 [6:49:18<62:56:12, 5.71s/it]
10%|▉ | 4230/43939 [6:49:24<62:57:04, 5.71s/it]
10%|▉ | 4231/43939 [6:49:30<63:42:09, 5.78s/it]
10%|▉ | 4232/43939 [6:49:36<63:29:49, 5.76s/it]
10%|▉ | 4233/43939 [6:49:41<63:18:31, 5.74s/it]
10%|▉ | 4234/43939 [6:49:47<63:15:32, 5.74s/it]
10%|▉ | 4235/43939 [6:49:53<63:08:37, 5.73s/it]
10%|▉ | 4236/43939 [6:49:59<63:04:43, 5.72s/it]
10%|▉ | 4237/43939 [6:50:04<63:01:58, 5.72s/it]
10%|▉ | 4238/43939 [6:50:10<62:51:59, 5.70s/it]
10%|▉ | 4239/43939 [6:50:16<62:53:34, 5.70s/it]
10%|▉ | 4240/43939 [6:50:21<63:03:14, 5.72s/it]
10%|▉ | 4241/43939 [6:50:27<63:01:08, 5.71s/it]
10%|▉ | 4242/43939 [6:50:33<62:56:27, 5.71s/it]
10%|▉ | 4243/43939 [6:50:38<62:48:58, 5.70s/it]
10%|▉ | 4244/43939 [6:50:44<62:43:02, 5.69s/it]
10%|▉ | 4245/43939 [6:50:50<62:38:25, 5.68s/it]
10%|▉ | 4246/43939 [6:50:55<62:41:24, 5.69s/it]
10%|▉ | 4247/43939 [6:51:01<62:39:57, 5.68s/it]
10%|▉ | 4248/43939 [6:51:07<62:43:11, 5.69s/it]
10%|▉ | 4249/43939 [6:51:13<62:36:51, 5.68s/it]
10%|▉ | 4250/43939 [6:51:18<62:52:42, 5.70s/it]
{'loss': 1.1117, 'grad_norm': 0.982142448425293, 'learning_rate': 4.8361028909628956e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 518.2, 'epoch': 0.39}
10%|▉ | 4250/43939 [6:51:18<62:52:42, 5.70s/it]
10%|▉ | 4251/43939 [6:51:24<63:36:31, 5.77s/it]
10%|▉ | 4252/43939 [6:51:30<63:26:09, 5.75s/it]
10%|▉ | 4253/43939 [6:51:36<63:16:19, 5.74s/it]
10%|▉ | 4254/43939 [6:51:41<63:06:05, 5.72s/it]
10%|▉ | 4255/43939 [6:51:47<63:01:27, 5.72s/it]
10%|▉ | 4256/43939 [6:51:53<63:02:07, 5.72s/it]
10%|▉ | 4257/43939 [6:51:59<63:08:06, 5.73s/it]
10%|▉ | 4258/43939 [6:52:04<62:57:19, 5.71s/it]
10%|▉ | 4259/43939 [6:52:10<62:59:11, 5.71s/it]
10%|▉ | 4260/43939 [6:52:16<63:01:17, 5.72s/it]
10%|▉ | 4261/43939 [6:52:21<63:01:27, 5.72s/it]
10%|▉ | 4262/43939 [6:52:27<63:02:59, 5.72s/it]
10%|▉ | 4263/43939 [6:52:33<63:00:33, 5.72s/it]
10%|▉ | 4264/43939 [6:52:38<62:53:44, 5.71s/it]
10%|▉ | 4265/43939 [6:52:44<62:53:24, 5.71s/it]
10%|▉ | 4266/43939 [6:52:50<62:56:30, 5.71s/it]
10%|▉ | 4267/43939 [6:52:56<62:56:27, 5.71s/it]
10%|▉ | 4268/43939 [6:53:01<62:53:32, 5.71s/it]
10%|▉ | 4269/43939 [6:53:07<62:50:42, 5.70s/it]
10%|▉ | 4270/43939 [6:53:13<62:49:20, 5.70s/it]
10%|▉ | 4271/43939 [6:53:18<62:49:22, 5.70s/it]
10%|▉ | 4272/43939 [6:53:24<63:34:25, 5.77s/it]
10%|▉ | 4273/43939 [6:53:30<63:19:19, 5.75s/it]
10%|▉ | 4274/43939 [6:53:36<63:07:20, 5.73s/it]
10%|▉ | 4275/43939 [6:53:41<63:02:51, 5.72s/it]
10%|▉ | 4276/43939 [6:53:47<62:59:25, 5.72s/it]
10%|▉ | 4277/43939 [6:53:53<63:03:28, 5.72s/it]
10%|▉ | 4278/43939 [6:53:59<62:58:56, 5.72s/it]
10%|▉ | 4279/43939 [6:54:04<62:57:50, 5.72s/it]
10%|▉ | 4280/43939 [6:54:10<62:54:38, 5.71s/it]
10%|▉ | 4281/43939 [6:54:16<62:58:31, 5.72s/it]
10%|▉ | 4282/43939 [6:54:21<63:05:17, 5.73s/it]
10%|▉ | 4283/43939 [6:54:27<62:54:52, 5.71s/it]
10%|▉ | 4284/43939 [6:54:33<63:02:05, 5.72s/it]
10%|▉ | 4285/43939 [6:54:39<62:56:40, 5.71s/it]
10%|▉ | 4286/43939 [6:54:44<62:54:00, 5.71s/it]
10%|▉ | 4287/43939 [6:54:50<62:54:54, 5.71s/it]
10%|▉ | 4288/43939 [6:54:56<62:52:53, 5.71s/it]
10%|▉ | 4289/43939 [6:55:01<62:52:48, 5.71s/it]
10%|▉ | 4290/43939 [6:55:07<62:53:29, 5.71s/it]
10%|▉ | 4291/43939 [6:55:13<62:43:29, 5.70s/it]
10%|▉ | 4292/43939 [6:55:18<62:42:32, 5.69s/it]
10%|▉ | 4293/43939 [6:55:24<62:45:19, 5.70s/it]
10%|▉ | 4294/43939 [6:55:30<63:31:45, 5.77s/it]
10%|▉ | 4295/43939 [6:55:36<63:16:47, 5.75s/it]
10%|▉ | 4296/43939 [6:55:41<62:59:53, 5.72s/it]
10%|▉ | 4297/43939 [6:55:47<62:57:14, 5.72s/it]
10%|▉ | 4298/43939 [6:55:53<62:48:50, 5.70s/it]
10%|▉ | 4299/43939 [6:55:59<62:51:39, 5.71s/it]
10%|▉ | 4300/43939 [6:56:04<62:48:19, 5.70s/it]
{'loss': 1.0869, 'grad_norm': 1.0042027235031128, 'learning_rate': 4.893011609378557e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.27, 'epoch': 0.39}
10%|▉ | 4300/43939 [6:56:04<62:48:19, 5.70s/it]
10%|▉ | 4301/43939 [6:56:10<62:46:17, 5.70s/it]
10%|▉ | 4302/43939 [6:56:16<62:45:54, 5.70s/it]
10%|▉ | 4303/43939 [6:56:21<62:57:12, 5.72s/it]
10%|▉ | 4304/43939 [6:56:27<62:43:30, 5.70s/it]
10%|▉ | 4305/43939 [6:56:33<62:43:23, 5.70s/it]
10%|▉ | 4306/43939 [6:56:38<62:42:01, 5.70s/it]
10%|▉ | 4307/43939 [6:56:44<62:40:02, 5.69s/it]
10%|▉ | 4308/43939 [6:56:50<62:36:03, 5.69s/it]
10%|▉ | 4309/43939 [6:56:56<62:40:38, 5.69s/it]
10%|▉ | 4310/43939 [6:57:01<62:36:35, 5.69s/it]
10%|▉ | 4311/43939 [6:57:07<62:39:52, 5.69s/it]
10%|▉ | 4312/43939 [6:57:13<62:38:14, 5.69s/it]
10%|▉ | 4313/43939 [6:57:18<62:38:59, 5.69s/it]
10%|▉ | 4314/43939 [6:57:24<63:24:00, 5.76s/it]
10%|▉ | 4315/43939 [6:57:30<63:09:01, 5.74s/it]
10%|▉ | 4316/43939 [6:57:36<62:57:41, 5.72s/it]
10%|▉ | 4317/43939 [6:57:41<62:52:37, 5.71s/it]
10%|▉ | 4318/43939 [6:57:47<62:51:48, 5.71s/it]
10%|▉ | 4319/43939 [6:57:53<62:50:13, 5.71s/it]
10%|▉ | 4320/43939 [6:57:58<62:49:50, 5.71s/it]
10%|▉ | 4321/43939 [6:58:04<62:49:42, 5.71s/it]
10%|▉ | 4322/43939 [6:58:10<62:48:05, 5.71s/it]
10%|▉ | 4323/43939 [6:58:15<62:40:47, 5.70s/it]
10%|▉ | 4324/43939 [6:58:21<62:55:03, 5.72s/it]
10%|▉ | 4325/43939 [6:58:27<62:46:36, 5.70s/it]
10%|▉ | 4326/43939 [6:58:33<62:40:11, 5.70s/it]
10%|▉ | 4327/43939 [6:58:38<62:45:48, 5.70s/it]
10%|▉ | 4328/43939 [6:58:44<62:42:26, 5.70s/it]
10%|▉ | 4329/43939 [6:58:50<62:37:02, 5.69s/it]
10%|▉ | 4330/43939 [6:58:55<62:34:33, 5.69s/it]
10%|▉ | 4331/43939 [6:59:01<62:40:05, 5.70s/it]
10%|▉ | 4332/43939 [6:59:07<62:32:04, 5.68s/it]
10%|▉ | 4333/43939 [6:59:12<62:38:47, 5.69s/it]
10%|▉ | 4334/43939 [6:59:18<62:36:58, 5.69s/it]
10%|▉ | 4335/43939 [6:59:24<63:21:59, 5.76s/it]
10%|▉ | 4336/43939 [6:59:30<63:14:04, 5.75s/it]
10%|▉ | 4337/43939 [6:59:35<63:08:16, 5.74s/it]
10%|▉ | 4338/43939 [6:59:41<63:01:42, 5.73s/it]
10%|▉ | 4339/43939 [6:59:47<62:55:53, 5.72s/it]
10%|▉ | 4340/43939 [6:59:53<62:51:27, 5.71s/it]
10%|▉ | 4341/43939 [6:59:58<62:46:06, 5.71s/it]
10%|▉ | 4342/43939 [7:00:04<62:47:05, 5.71s/it]
10%|▉ | 4343/43939 [7:00:10<62:36:44, 5.69s/it]
10%|▉ | 4344/43939 [7:00:15<62:30:37, 5.68s/it]
10%|▉ | 4345/43939 [7:00:21<62:25:01, 5.68s/it]
10%|▉ | 4346/43939 [7:00:27<62:30:27, 5.68s/it]
10%|▉ | 4347/43939 [7:00:32<62:29:33, 5.68s/it]
10%|▉ | 4348/43939 [7:00:38<62:23:28, 5.67s/it]
10%|▉ | 4349/43939 [7:00:44<62:27:39, 5.68s/it]
10%|▉ | 4350/43939 [7:00:49<62:22:15, 5.67s/it]
{'loss': 1.0682, 'grad_norm': 1.2012032270431519, 'learning_rate': 4.949920327794218e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.47, 'epoch': 0.4}
10%|▉ | 4350/43939 [7:00:49<62:22:15, 5.67s/it]
10%|▉ | 4351/43939 [7:00:55<62:20:28, 5.67s/it]
10%|▉ | 4352/43939 [7:01:01<62:28:41, 5.68s/it]
10%|▉ | 4353/43939 [7:01:06<62:29:37, 5.68s/it]
10%|▉ | 4354/43939 [7:01:12<62:28:10, 5.68s/it]
10%|▉ | 4355/43939 [7:01:18<62:36:46, 5.69s/it]
10%|▉ | 4356/43939 [7:01:24<62:47:01, 5.71s/it]
10%|▉ | 4357/43939 [7:01:29<63:28:40, 5.77s/it]
10%|▉ | 4358/43939 [7:01:35<63:11:04, 5.75s/it]
10%|▉ | 4359/43939 [7:01:41<63:04:04, 5.74s/it]
10%|▉ | 4360/43939 [7:01:47<62:54:32, 5.72s/it]
10%|▉ | 4361/43939 [7:01:52<62:51:46, 5.72s/it]
10%|▉ | 4362/43939 [7:01:58<62:53:30, 5.72s/it]
10%|▉ | 4363/43939 [7:02:04<62:52:27, 5.72s/it]
10%|▉ | 4364/43939 [7:02:09<62:49:22, 5.71s/it]
10%|▉ | 4365/43939 [7:02:15<62:45:45, 5.71s/it]
10%|▉ | 4366/43939 [7:02:21<62:35:55, 5.69s/it]
10%|▉ | 4367/43939 [7:02:26<62:29:22, 5.68s/it]
10%|▉ | 4368/43939 [7:02:32<62:30:36, 5.69s/it]
10%|▉ | 4369/43939 [7:02:38<62:24:18, 5.68s/it]
10%|▉ | 4370/43939 [7:02:43<62:30:29, 5.69s/it]
10%|▉ | 4371/43939 [7:02:49<62:41:03, 5.70s/it]
10%|▉ | 4372/43939 [7:02:55<62:35:58, 5.70s/it]
10%|▉ | 4373/43939 [7:03:01<62:37:39, 5.70s/it]
10%|▉ | 4374/43939 [7:03:06<62:29:29, 5.69s/it]
10%|▉ | 4375/43939 [7:03:12<62:37:25, 5.70s/it]
10%|▉ | 4376/43939 [7:03:18<62:37:45, 5.70s/it]
10%|▉ | 4377/43939 [7:03:23<62:44:42, 5.71s/it]
10%|▉ | 4378/43939 [7:03:29<63:31:14, 5.78s/it]
10%|▉ | 4379/43939 [7:03:35<63:35:50, 5.79s/it]
10%|▉ | 4380/43939 [7:03:41<63:09:46, 5.75s/it]
10%|▉ | 4381/43939 [7:03:47<63:01:15, 5.74s/it]
10%|▉ | 4382/43939 [7:03:52<62:43:10, 5.71s/it]
10%|▉ | 4383/43939 [7:03:58<62:38:22, 5.70s/it]
10%|▉ | 4384/43939 [7:04:04<62:39:39, 5.70s/it]
10%|▉ | 4385/43939 [7:04:09<62:51:07, 5.72s/it]
10%|▉ | 4386/43939 [7:04:15<62:44:34, 5.71s/it]
10%|▉ | 4387/43939 [7:04:21<62:31:47, 5.69s/it]
10%|▉ | 4388/43939 [7:04:26<62:42:05, 5.71s/it]
10%|▉ | 4389/43939 [7:04:32<62:31:50, 5.69s/it]
10%|▉ | 4390/43939 [7:04:38<62:26:43, 5.68s/it]
10%|▉ | 4391/43939 [7:04:43<62:31:28, 5.69s/it]
10%|▉ | 4392/43939 [7:04:49<62:37:16, 5.70s/it]
10%|▉ | 4393/43939 [7:04:55<62:31:01, 5.69s/it]
10%|█ | 4394/43939 [7:05:01<62:40:15, 5.71s/it][2025-11-29 01:09:14,381] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:40<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.72s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.72s/it][A
75%|███████▌ | 54/72 [02:24<00:48, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A{'eval_loss': 1.0708580017089844, 'eval_runtime': 195.2836, 'eval_samples_per_second': 5.116, 'eval_steps_per_second': 0.369, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 0.4}
10%|█ | 4394/43939 [7:08:16<62:40:15, 5.71s/it]
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A
10%|█ | 4395/43939 [7:08:22<706:39:50, 64.33s/it]
10%|█ | 4396/43939 [7:08:27<513:41:08, 46.77s/it]
10%|█ | 4397/43939 [7:08:33<378:18:21, 34.44s/it]
10%|█ | 4398/43939 [7:08:39<284:25:29, 25.90s/it]
10%|█ | 4399/43939 [7:08:45<217:53:07, 19.84s/it]
10%|█ | 4400/43939 [7:08:51<171:16:01, 15.59s/it]
{'loss': 1.1033, 'grad_norm': 1.1110295057296753, 'learning_rate': 4.999999716007333e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.06, 'tokens_per_second_per_gpu': 62.64, 'epoch': 0.4}
10%|█ | 4400/43939 [7:08:51<171:16:01, 15.59s/it]
10%|█ | 4401/43939 [7:08:56<138:34:10, 12.62s/it]
10%|█ | 4402/43939 [7:09:02<115:48:44, 10.55s/it]
10%|█ | 4403/43939 [7:09:08<99:49:09, 9.09s/it]
10%|█ | 4404/43939 [7:09:13<88:54:09, 8.10s/it]
10%|█ | 4405/43939 [7:09:19<80:53:21, 7.37s/it]
10%|█ | 4406/43939 [7:09:25<75:17:04, 6.86s/it]
10%|█ | 4407/43939 [7:09:30<71:27:47, 6.51s/it]
10%|█ | 4408/43939 [7:09:36<68:44:25, 6.26s/it]
10%|█ | 4409/43939 [7:09:42<66:53:37, 6.09s/it]
10%|█ | 4410/43939 [7:09:47<65:25:27, 5.96s/it]
10%|█ | 4411/43939 [7:09:53<64:36:35, 5.88s/it]
10%|█ | 4412/43939 [7:09:59<63:55:03, 5.82s/it]
10%|█ | 4413/43939 [7:10:04<63:30:20, 5.78s/it]
10%|█ | 4414/43939 [7:10:10<63:15:44, 5.76s/it]
10%|█ | 4415/43939 [7:10:16<63:02:49, 5.74s/it]
10%|█ | 4416/43939 [7:10:22<62:52:43, 5.73s/it]
10%|█ | 4417/43939 [7:10:27<62:53:08, 5.73s/it]
10%|█ | 4418/43939 [7:10:33<62:44:38, 5.72s/it]
10%|█ | 4419/43939 [7:10:39<63:37:41, 5.80s/it]
10%|█ | 4420/43939 [7:10:45<63:24:44, 5.78s/it]
10%|█ | 4421/43939 [7:10:50<63:08:12, 5.75s/it]
10%|█ | 4422/43939 [7:10:56<63:06:53, 5.75s/it]
10%|█ | 4423/43939 [7:11:02<62:57:58, 5.74s/it]
10%|█ | 4424/43939 [7:11:08<62:47:56, 5.72s/it]
10%|█ | 4425/43939 [7:11:13<62:45:34, 5.72s/it]
10%|█ | 4426/43939 [7:11:19<62:36:22, 5.70s/it]
10%|█ | 4427/43939 [7:11:25<62:24:04, 5.69s/it]
10%|█ | 4428/43939 [7:11:30<62:31:17, 5.70s/it]
10%|█ | 4429/43939 [7:11:36<62:35:32, 5.70s/it]
10%|█ | 4430/43939 [7:11:42<62:40:30, 5.71s/it]
10%|█ | 4431/43939 [7:11:47<62:33:44, 5.70s/it]
10%|█ | 4432/43939 [7:11:53<62:47:01, 5.72s/it]
10%|█ | 4433/43939 [7:11:59<62:41:08, 5.71s/it]
10%|█ | 4434/43939 [7:12:05<62:39:13, 5.71s/it]
10%|█ | 4435/43939 [7:12:10<62:36:32, 5.71s/it]
10%|█ | 4436/43939 [7:12:16<62:40:59, 5.71s/it]
10%|█ | 4437/43939 [7:12:22<62:38:38, 5.71s/it]
10%|█ | 4438/43939 [7:12:27<62:44:08, 5.72s/it]
10%|█ | 4439/43939 [7:12:33<63:27:42, 5.78s/it]
10%|█ | 4440/43939 [7:12:39<63:09:55, 5.76s/it]
10%|█ | 4441/43939 [7:12:45<63:02:26, 5.75s/it]
10%|█ | 4442/43939 [7:12:51<62:58:04, 5.74s/it]
10%|█ | 4443/43939 [7:12:56<62:50:16, 5.73s/it]
10%|█ | 4444/43939 [7:13:02<62:42:44, 5.72s/it]
10%|█ | 4445/43939 [7:13:08<62:37:08, 5.71s/it]
10%|█ | 4446/43939 [7:13:13<62:30:46, 5.70s/it]
10%|█ | 4447/43939 [7:13:19<62:33:45, 5.70s/it]
10%|█ | 4448/43939 [7:13:25<62:26:09, 5.69s/it]
10%|█ | 4449/43939 [7:13:30<62:29:44, 5.70s/it]
10%|█ | 4450/43939 [7:13:36<62:25:14, 5.69s/it]
{'loss': 1.088, 'grad_norm': 1.0224543809890747, 'learning_rate': 4.999975261123525e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.07, 'epoch': 0.41}
10%|█ | 4450/43939 [7:13:36<62:25:14, 5.69s/it]
10%|█ | 4451/43939 [7:13:42<62:23:56, 5.69s/it]
10%|█ | 4452/43939 [7:13:47<62:37:57, 5.71s/it]
10%|█ | 4453/43939 [7:13:53<62:35:37, 5.71s/it]
10%|█ | 4454/43939 [7:13:59<62:32:13, 5.70s/it]
10%|█ | 4455/43939 [7:14:05<62:31:27, 5.70s/it]
10%|█ | 4456/43939 [7:14:10<62:35:49, 5.71s/it]
10%|█ | 4457/43939 [7:14:16<62:42:47, 5.72s/it]
10%|█ | 4458/43939 [7:14:22<62:41:19, 5.72s/it]
10%|█ | 4459/43939 [7:14:27<62:37:00, 5.71s/it]
10%|█ | 4460/43939 [7:14:33<62:32:42, 5.70s/it]
10%|█ | 4461/43939 [7:14:39<63:22:29, 5.78s/it]
10%|█ | 4462/43939 [7:14:45<63:04:25, 5.75s/it]
10%|█ | 4463/43939 [7:14:50<62:48:09, 5.73s/it]
10%|█ | 4464/43939 [7:14:56<62:31:37, 5.70s/it]
10%|█ | 4465/43939 [7:15:02<62:33:43, 5.71s/it]
10%|█ | 4466/43939 [7:15:08<62:39:34, 5.71s/it]
10%|█ | 4467/43939 [7:15:13<62:37:54, 5.71s/it]
10%|█ | 4468/43939 [7:15:19<62:46:04, 5.72s/it]
10%|█ | 4469/43939 [7:15:25<62:41:28, 5.72s/it]
10%|█ | 4470/43939 [7:15:30<62:38:43, 5.71s/it]
10%|█ | 4471/43939 [7:15:36<62:41:30, 5.72s/it]
10%|█ | 4472/43939 [7:15:42<62:41:47, 5.72s/it]
10%|█ | 4473/43939 [7:15:48<62:38:46, 5.71s/it]
10%|█ | 4474/43939 [7:15:53<62:28:27, 5.70s/it]
10%|█ | 4475/43939 [7:15:59<62:37:18, 5.71s/it]
10%|█ | 4476/43939 [7:16:05<62:40:03, 5.72s/it]
10%|█ | 4477/43939 [7:16:10<62:30:54, 5.70s/it]
10%|█ | 4478/43939 [7:16:16<62:31:46, 5.70s/it]
10%|█ | 4479/43939 [7:16:22<62:27:44, 5.70s/it]
10%|█ | 4480/43939 [7:16:27<62:27:04, 5.70s/it]
10%|█ | 4481/43939 [7:16:33<62:30:14, 5.70s/it]
10%|█ | 4482/43939 [7:16:39<63:13:04, 5.77s/it]
10%|█ | 4483/43939 [7:16:45<63:02:38, 5.75s/it]
10%|█ | 4484/43939 [7:16:50<62:46:08, 5.73s/it]
10%|█ | 4485/43939 [7:16:56<62:42:20, 5.72s/it]
10%|█ | 4486/43939 [7:17:02<62:36:10, 5.71s/it]
10%|█ | 4487/43939 [7:17:08<62:31:56, 5.71s/it]
10%|█ | 4488/43939 [7:17:13<62:29:27, 5.70s/it]
10%|█ | 4489/43939 [7:17:19<62:22:10, 5.69s/it]
10%|█ | 4490/43939 [7:17:25<62:24:47, 5.70s/it]
10%|█ | 4491/43939 [7:17:30<62:25:31, 5.70s/it]
10%|█ | 4492/43939 [7:17:36<62:27:06, 5.70s/it]
10%|█ | 4493/43939 [7:17:42<62:25:37, 5.70s/it]
10%|█ | 4494/43939 [7:17:47<62:23:16, 5.69s/it]
10%|█ | 4495/43939 [7:17:53<62:21:44, 5.69s/it]
10%|█ | 4496/43939 [7:17:59<62:23:30, 5.69s/it]
10%|█ | 4497/43939 [7:18:04<62:23:16, 5.69s/it]
10%|█ | 4498/43939 [7:18:10<62:25:44, 5.70s/it]
10%|█ | 4499/43939 [7:18:16<62:27:23, 5.70s/it]
10%|█ | 4500/43939 [7:18:22<62:26:17, 5.70s/it]
{'loss': 1.0761, 'grad_norm': 0.8717514872550964, 'learning_rate': 4.999911363255113e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.38, 'epoch': 0.41}
10%|█ | 4500/43939 [7:18:22<62:26:17, 5.70s/it]
10%|█ | 4501/43939 [7:18:27<62:18:57, 5.69s/it]
10%|█ | 4502/43939 [7:18:33<63:09:08, 5.76s/it]
10%|█ | 4503/43939 [7:18:39<62:51:08, 5.74s/it]
10%|█ | 4504/43939 [7:18:45<62:40:15, 5.72s/it]
10%|█ | 4505/43939 [7:18:50<62:35:22, 5.71s/it]
10%|█ | 4506/43939 [7:18:56<62:31:05, 5.71s/it]
10%|█ | 4507/43939 [7:19:02<62:24:36, 5.70s/it]
10%|█ | 4508/43939 [7:19:07<62:24:22, 5.70s/it]
10%|█ | 4509/43939 [7:19:13<62:22:27, 5.69s/it]
10%|█ | 4510/43939 [7:19:19<62:18:39, 5.69s/it]
10%|█ | 4511/43939 [7:19:24<62:20:49, 5.69s/it]
10%|█ | 4512/43939 [7:19:30<62:18:28, 5.69s/it]
10%|█ | 4513/43939 [7:19:36<62:23:35, 5.70s/it]
10%|█ | 4514/43939 [7:19:41<62:22:51, 5.70s/it]
10%|█ | 4515/43939 [7:19:47<62:24:39, 5.70s/it]
10%|█ | 4516/43939 [7:19:53<62:34:11, 5.71s/it]
10%|█ | 4517/43939 [7:19:59<62:26:51, 5.70s/it]
10%|█ | 4518/43939 [7:20:04<62:31:06, 5.71s/it]
10%|█ | 4519/43939 [7:20:10<62:34:35, 5.71s/it]
10%|█ | 4520/43939 [7:20:16<62:27:13, 5.70s/it]
10%|█ | 4521/43939 [7:20:21<62:29:37, 5.71s/it]
10%|█ | 4522/43939 [7:20:27<62:26:51, 5.70s/it]
10%|█ | 4523/43939 [7:20:33<62:24:22, 5.70s/it]
10%|█ | 4524/43939 [7:20:39<63:10:35, 5.77s/it]
10%|█ | 4525/43939 [7:20:44<62:59:14, 5.75s/it]
10%|█ | 4526/43939 [7:20:50<62:46:09, 5.73s/it]
10%|█ | 4527/43939 [7:20:56<62:36:30, 5.72s/it]
10%|█ | 4528/43939 [7:21:02<62:24:28, 5.70s/it]
10%|█ | 4529/43939 [7:21:07<62:17:13, 5.69s/it]
10%|█ | 4530/43939 [7:21:13<62:13:51, 5.68s/it]
10%|█ | 4531/43939 [7:21:19<62:19:34, 5.69s/it]
10%|█ | 4532/43939 [7:21:24<62:21:41, 5.70s/it]
10%|█ | 4533/43939 [7:21:30<62:25:06, 5.70s/it]
10%|█ | 4534/43939 [7:21:36<62:23:44, 5.70s/it]
10%|█ | 4535/43939 [7:21:41<62:20:10, 5.70s/it]
10%|█ | 4536/43939 [7:21:47<62:19:44, 5.69s/it]
10%|█ | 4537/43939 [7:21:53<62:23:13, 5.70s/it]
10%|█ | 4538/43939 [7:21:58<62:16:39, 5.69s/it]
10%|█ | 4539/43939 [7:22:04<62:12:11, 5.68s/it]
10%|█ | 4540/43939 [7:22:10<62:07:38, 5.68s/it]
10%|█ | 4541/43939 [7:22:16<62:17:05, 5.69s/it]
10%|█ | 4542/43939 [7:22:21<62:11:02, 5.68s/it]
10%|█ | 4543/43939 [7:22:27<62:10:15, 5.68s/it]
10%|█ | 4544/43939 [7:22:33<62:07:39, 5.68s/it]
10%|█ | 4545/43939 [7:22:38<62:54:59, 5.75s/it]
10%|█ | 4546/43939 [7:22:44<62:54:34, 5.75s/it]
10%|█ | 4547/43939 [7:22:50<62:49:02, 5.74s/it]
10%|█ | 4548/43939 [7:22:56<62:48:31, 5.74s/it]
10%|█ | 4549/43939 [7:23:01<62:31:21, 5.71s/it]
10%|█ | 4550/43939 [7:23:07<62:26:22, 5.71s/it]
{'loss': 1.0455, 'grad_norm': 1.2397717237472534, 'learning_rate': 4.999808023410233e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.23, 'epoch': 0.41}
10%|█ | 4550/43939 [7:23:07<62:26:22, 5.71s/it]
10%|█ | 4551/43939 [7:23:13<62:29:28, 5.71s/it]
10%|█ | 4552/43939 [7:23:18<62:28:22, 5.71s/it]
10%|█ | 4553/43939 [7:23:24<62:22:44, 5.70s/it]
10%|█ | 4554/43939 [7:23:30<62:20:26, 5.70s/it]
10%|█ | 4555/43939 [7:23:35<62:20:15, 5.70s/it]
10%|█ | 4556/43939 [7:23:41<62:26:26, 5.71s/it]
10%|█ | 4557/43939 [7:23:47<62:20:45, 5.70s/it]
10%|█ | 4558/43939 [7:23:53<62:25:54, 5.71s/it]
10%|█ | 4559/43939 [7:23:58<62:18:39, 5.70s/it]
10%|█ | 4560/43939 [7:24:04<62:18:28, 5.70s/it]
10%|█ | 4561/43939 [7:24:10<62:24:01, 5.70s/it]
10%|█ | 4562/43939 [7:24:15<62:22:07, 5.70s/it]
10%|█ | 4563/43939 [7:24:21<62:18:36, 5.70s/it]
10%|█ | 4564/43939 [7:24:27<62:21:27, 5.70s/it]
10%|█ | 4565/43939 [7:24:33<63:11:11, 5.78s/it]
10%|█ | 4566/43939 [7:24:38<63:01:10, 5.76s/it]
10%|█ | 4567/43939 [7:24:44<62:42:58, 5.73s/it]
10%|█ | 4568/43939 [7:24:50<62:38:45, 5.73s/it]
10%|█ | 4569/43939 [7:24:56<62:32:00, 5.72s/it]
10%|█ | 4570/43939 [7:25:01<62:26:19, 5.71s/it]
10%|█ | 4571/43939 [7:25:07<62:23:24, 5.71s/it]
10%|█ | 4572/43939 [7:25:13<62:28:55, 5.71s/it]
10%|█ | 4573/43939 [7:25:18<62:24:42, 5.71s/it]
10%|█ | 4574/43939 [7:25:24<62:26:08, 5.71s/it]
10%|█ | 4575/43939 [7:25:30<62:18:33, 5.70s/it]
10%|█ | 4576/43939 [7:25:35<62:15:57, 5.69s/it]
10%|█ | 4577/43939 [7:25:41<62:19:29, 5.70s/it]
10%|█ | 4578/43939 [7:25:47<62:13:07, 5.69s/it]
10%|█ | 4579/43939 [7:25:53<62:12:33, 5.69s/it]
10%|█ | 4580/43939 [7:25:58<62:22:18, 5.70s/it]
10%|█ | 4581/43939 [7:26:04<62:29:33, 5.72s/it]
10%|█ | 4582/43939 [7:26:10<62:28:46, 5.72s/it]
10%|█ | 4583/43939 [7:26:15<62:28:42, 5.72s/it]
10%|█ | 4584/43939 [7:26:21<62:30:22, 5.72s/it]
10%|█ | 4585/43939 [7:26:27<62:22:37, 5.71s/it]
10%|█ | 4586/43939 [7:26:33<63:00:45, 5.76s/it]
10%|█ | 4587/43939 [7:26:38<62:53:07, 5.75s/it]
10%|█ | 4588/43939 [7:26:44<62:42:33, 5.74s/it]
10%|█ | 4589/43939 [7:26:50<62:25:26, 5.71s/it]
10%|█ | 4590/43939 [7:26:56<62:29:48, 5.72s/it]
10%|█ | 4591/43939 [7:27:01<62:21:44, 5.71s/it]
10%|█ | 4592/43939 [7:27:07<62:18:51, 5.70s/it]
10%|█ | 4593/43939 [7:27:13<62:13:47, 5.69s/it]
10%|█ | 4594/43939 [7:27:18<62:09:00, 5.69s/it]
10%|█ | 4595/43939 [7:27:24<62:16:02, 5.70s/it]
10%|█ | 4596/43939 [7:27:30<62:08:36, 5.69s/it]
10%|█ | 4597/43939 [7:27:35<62:01:08, 5.68s/it]
10%|█ | 4598/43939 [7:27:41<62:03:29, 5.68s/it]
10%|█ | 4599/43939 [7:27:47<61:55:53, 5.67s/it]
10%|█ | 4600/43939 [7:27:52<61:54:24, 5.67s/it]
{'loss': 1.0789, 'grad_norm': 1.0875898599624634, 'learning_rate': 4.999665243219317e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.7, 'epoch': 0.42}
10%|█ | 4600/43939 [7:27:52<61:54:24, 5.67s/it]
10%|█ | 4601/43939 [7:27:58<61:54:26, 5.67s/it]
10%|█ | 4602/43939 [7:28:04<61:53:23, 5.66s/it]
10%|█ | 4603/43939 [7:28:09<61:55:52, 5.67s/it]
10%|█ | 4604/43939 [7:28:15<61:58:05, 5.67s/it]
10%|█ | 4605/43939 [7:28:21<62:12:02, 5.69s/it]
10%|█ | 4606/43939 [7:28:26<62:15:19, 5.70s/it]
10%|█ | 4607/43939 [7:28:32<62:16:23, 5.70s/it]
10%|█ | 4608/43939 [7:28:38<63:08:58, 5.78s/it]
10%|█ | 4609/43939 [7:28:44<62:46:43, 5.75s/it]
10%|█ | 4610/43939 [7:28:49<62:41:09, 5.74s/it]
10%|█ | 4611/43939 [7:28:55<62:37:17, 5.73s/it]
10%|█ | 4612/43939 [7:29:01<62:31:03, 5.72s/it]
10%|█ | 4613/43939 [7:29:07<62:23:35, 5.71s/it]
11%|█ | 4614/43939 [7:29:12<62:21:21, 5.71s/it]
11%|█ | 4615/43939 [7:29:18<62:17:15, 5.70s/it]
11%|█ | 4616/43939 [7:29:24<62:14:06, 5.70s/it]
11%|█ | 4617/43939 [7:29:29<62:08:01, 5.69s/it]
11%|█ | 4618/43939 [7:29:35<62:10:15, 5.69s/it]
11%|█ | 4619/43939 [7:29:41<62:09:15, 5.69s/it]
11%|█ | 4620/43939 [7:29:46<62:07:41, 5.69s/it]
11%|█ | 4621/43939 [7:29:52<62:02:59, 5.68s/it]
11%|█ | 4622/43939 [7:29:58<62:07:36, 5.69s/it]
11%|█ | 4623/43939 [7:30:03<62:12:40, 5.70s/it]
11%|█ | 4624/43939 [7:30:09<62:07:54, 5.69s/it]
11%|█ | 4625/43939 [7:30:15<62:02:21, 5.68s/it]
11%|█ | 4626/43939 [7:30:21<62:05:41, 5.69s/it]
11%|█ | 4627/43939 [7:30:26<62:07:34, 5.69s/it]
11%|█ | 4628/43939 [7:30:32<62:58:18, 5.77s/it]
11%|█ | 4629/43939 [7:30:38<62:45:03, 5.75s/it]
11%|█ | 4630/43939 [7:30:44<62:36:29, 5.73s/it]
11%|█ | 4631/43939 [7:30:49<62:30:50, 5.73s/it]
11%|█ | 4632/43939 [7:30:55<62:25:54, 5.72s/it]
11%|█ | 4633/43939 [7:31:01<62:16:20, 5.70s/it]
11%|█ | 4634/43939 [7:31:06<62:17:40, 5.71s/it]
11%|█ | 4635/43939 [7:31:12<62:12:11, 5.70s/it]
11%|█ | 4636/43939 [7:31:18<62:17:58, 5.71s/it]
11%|█ | 4637/43939 [7:31:23<62:18:37, 5.71s/it]
11%|█ | 4638/43939 [7:31:29<62:21:59, 5.71s/it]
11%|█ | 4639/43939 [7:31:35<62:13:29, 5.70s/it]
11%|█ | 4640/43939 [7:31:41<62:13:17, 5.70s/it]
11%|█ | 4641/43939 [7:31:46<62:09:24, 5.69s/it]
11%|█ | 4642/43939 [7:31:52<62:12:19, 5.70s/it]
11%|█ | 4643/43939 [7:31:58<62:15:50, 5.70s/it]
11%|█ | 4644/43939 [7:32:03<62:05:12, 5.69s/it]
11%|█ | 4645/43939 [7:32:09<62:06:50, 5.69s/it]
11%|█ | 4646/43939 [7:32:15<62:16:43, 5.71s/it]
11%|█ | 4647/43939 [7:32:20<62:20:03, 5.71s/it]
11%|█ | 4648/43939 [7:32:26<62:08:30, 5.69s/it]
11%|█ | 4649/43939 [7:32:32<62:55:53, 5.77s/it]
11%|█ | 4650/43939 [7:32:38<62:49:25, 5.76s/it]
{'loss': 1.0736, 'grad_norm': 0.9245534539222717, 'learning_rate': 4.9994830249350555e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.91, 'epoch': 0.42}
11%|█ | 4650/43939 [7:32:38<62:49:25, 5.76s/it]
11%|█ | 4651/43939 [7:32:44<62:41:38, 5.74s/it]
11%|█ | 4652/43939 [7:32:49<62:39:34, 5.74s/it]
11%|█ | 4653/43939 [7:32:55<62:29:22, 5.73s/it]
11%|█ | 4654/43939 [7:33:01<62:23:25, 5.72s/it]
11%|█ | 4655/43939 [7:33:06<62:14:21, 5.70s/it]
11%|█ | 4656/43939 [7:33:12<62:20:24, 5.71s/it]
11%|█ | 4657/43939 [7:33:18<62:12:55, 5.70s/it]
11%|█ | 4658/43939 [7:33:23<62:06:20, 5.69s/it]
11%|█ | 4659/43939 [7:33:29<62:09:59, 5.70s/it]
11%|█ | 4660/43939 [7:33:35<62:11:44, 5.70s/it]
11%|█ | 4661/43939 [7:33:40<62:06:11, 5.69s/it]
11%|█ | 4662/43939 [7:33:46<61:59:42, 5.68s/it]
11%|█ | 4663/43939 [7:33:52<62:02:03, 5.69s/it]
11%|█ | 4664/43939 [7:33:57<61:57:30, 5.68s/it]
11%|█ | 4665/43939 [7:34:03<62:02:23, 5.69s/it]
11%|█ | 4666/43939 [7:34:09<62:04:51, 5.69s/it]
11%|█ | 4667/43939 [7:34:15<61:59:56, 5.68s/it]
11%|█ | 4668/43939 [7:34:20<62:14:24, 5.71s/it]
11%|█ | 4669/43939 [7:34:26<62:15:24, 5.71s/it]
11%|█ | 4670/43939 [7:34:32<62:11:53, 5.70s/it]
11%|█ | 4671/43939 [7:34:38<63:03:34, 5.78s/it]
11%|█ | 4672/43939 [7:34:43<62:42:35, 5.75s/it]
11%|█ | 4673/43939 [7:34:49<62:32:15, 5.73s/it]
11%|█ | 4674/43939 [7:34:55<62:27:30, 5.73s/it]
11%|█ | 4675/43939 [7:35:00<62:23:21, 5.72s/it]
11%|█ | 4676/43939 [7:35:06<62:20:51, 5.72s/it]
11%|█ | 4677/43939 [7:35:12<62:13:29, 5.71s/it]
11%|█ | 4678/43939 [7:35:18<62:10:48, 5.70s/it]
11%|█ | 4679/43939 [7:35:23<62:09:27, 5.70s/it]
11%|█ | 4680/43939 [7:35:29<62:02:21, 5.69s/it]
11%|█ | 4681/43939 [7:35:35<62:05:19, 5.69s/it]
11%|█ | 4682/43939 [7:35:40<62:04:04, 5.69s/it]
11%|█ | 4683/43939 [7:35:46<62:09:48, 5.70s/it]
11%|█ | 4684/43939 [7:35:52<62:08:44, 5.70s/it]
11%|█ | 4685/43939 [7:35:57<62:13:49, 5.71s/it]
11%|█ | 4686/43939 [7:36:03<62:11:03, 5.70s/it]
11%|█ | 4687/43939 [7:36:09<62:06:22, 5.70s/it]
11%|█ | 4688/43939 [7:36:15<62:14:47, 5.71s/it]
11%|█ | 4689/43939 [7:36:20<62:09:31, 5.70s/it]
11%|█ | 4690/43939 [7:36:26<62:03:04, 5.69s/it]
11%|█ | 4691/43939 [7:36:32<62:48:16, 5.76s/it]
11%|█ | 4692/43939 [7:36:38<62:33:39, 5.74s/it]
11%|█ | 4693/43939 [7:36:43<62:20:02, 5.72s/it]
11%|█ | 4694/43939 [7:36:49<62:15:59, 5.71s/it]
11%|█ | 4695/43939 [7:36:55<62:13:06, 5.71s/it]
11%|█ | 4696/43939 [7:37:00<62:03:07, 5.69s/it]
11%|█ | 4697/43939 [7:37:06<61:59:35, 5.69s/it]
11%|█ | 4698/43939 [7:37:12<61:59:53, 5.69s/it]
11%|█ | 4699/43939 [7:37:17<62:02:12, 5.69s/it]
11%|█ | 4700/43939 [7:37:23<62:04:57, 5.70s/it]
{'loss': 1.0639, 'grad_norm': 1.1447221040725708, 'learning_rate': 4.9992613714323723e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.85, 'epoch': 0.43}
11%|█ | 4700/43939 [7:37:23<62:04:57, 5.70s/it]
11%|█ | 4701/43939 [7:37:29<62:00:58, 5.69s/it]
11%|█ | 4702/43939 [7:37:34<62:07:08, 5.70s/it]
11%|█ | 4703/43939 [7:37:40<62:12:14, 5.71s/it]
11%|█ | 4704/43939 [7:37:46<62:06:36, 5.70s/it]
11%|█ | 4705/43939 [7:37:51<62:02:44, 5.69s/it]
11%|█ | 4706/43939 [7:37:57<62:00:43, 5.69s/it]
11%|█ | 4707/43939 [7:38:03<61:53:25, 5.68s/it]
11%|█ | 4708/43939 [7:38:09<61:56:40, 5.68s/it]
11%|█ | 4709/43939 [7:38:14<61:54:56, 5.68s/it]
11%|█ | 4710/43939 [7:38:20<62:05:18, 5.70s/it]
11%|█ | 4711/43939 [7:38:26<61:58:47, 5.69s/it]
11%|█ | 4712/43939 [7:38:32<62:53:22, 5.77s/it]
11%|█ | 4713/43939 [7:38:37<62:36:46, 5.75s/it]
11%|█ | 4714/43939 [7:38:43<62:20:24, 5.72s/it]
11%|█ | 4715/43939 [7:38:49<62:29:26, 5.74s/it]
11%|█ | 4716/43939 [7:38:54<62:22:13, 5.72s/it]
11%|█ | 4717/43939 [7:39:00<62:13:02, 5.71s/it]
11%|█ | 4718/43939 [7:39:06<62:09:42, 5.71s/it]
11%|█ | 4719/43939 [7:39:11<62:07:16, 5.70s/it]
11%|█ | 4720/43939 [7:39:17<62:01:07, 5.69s/it]
11%|█ | 4721/43939 [7:39:23<61:57:17, 5.69s/it]
11%|█ | 4722/43939 [7:39:28<61:54:35, 5.68s/it]
11%|█ | 4723/43939 [7:39:34<61:59:28, 5.69s/it]
11%|█ | 4724/43939 [7:39:40<62:00:44, 5.69s/it]
11%|█ | 4725/43939 [7:39:46<62:17:42, 5.72s/it]
11%|█ | 4726/43939 [7:39:51<62:14:15, 5.71s/it]
11%|█ | 4727/43939 [7:39:57<62:24:23, 5.73s/it]
11%|█ | 4728/43939 [7:40:03<62:12:48, 5.71s/it]
11%|█ | 4729/43939 [7:40:09<62:22:42, 5.73s/it]
11%|█ | 4730/43939 [7:40:14<62:16:19, 5.72s/it]
11%|█ | 4731/43939 [7:40:20<62:14:15, 5.71s/it]
11%|█ | 4732/43939 [7:40:26<62:05:26, 5.70s/it]
11%|█ | 4733/43939 [7:40:31<61:59:47, 5.69s/it]
11%|█ | 4734/43939 [7:40:37<62:41:50, 5.76s/it]
11%|█ | 4735/43939 [7:40:43<62:29:54, 5.74s/it]
11%|█ | 4736/43939 [7:40:49<62:19:06, 5.72s/it]
11%|█ | 4737/43939 [7:40:54<62:14:13, 5.72s/it]
11%|█ | 4738/43939 [7:41:00<62:15:39, 5.72s/it]
11%|█ | 4739/43939 [7:41:06<62:11:44, 5.71s/it]
11%|█ | 4740/43939 [7:41:11<62:09:37, 5.71s/it]
11%|█ | 4741/43939 [7:41:17<62:11:24, 5.71s/it]
11%|█ | 4742/43939 [7:41:23<62:09:11, 5.71s/it]
11%|█ | 4743/43939 [7:41:29<62:09:29, 5.71s/it]
11%|█ | 4744/43939 [7:41:34<62:02:02, 5.70s/it]
11%|█ | 4745/43939 [7:41:40<62:03:08, 5.70s/it]
11%|█ | 4746/43939 [7:41:46<62:09:36, 5.71s/it]
11%|█ | 4747/43939 [7:41:51<62:09:09, 5.71s/it]
11%|█ | 4748/43939 [7:41:57<62:08:42, 5.71s/it]
11%|█ | 4749/43939 [7:42:03<62:05:08, 5.70s/it]
11%|█ | 4750/43939 [7:42:09<62:15:43, 5.72s/it]
{'loss': 1.077, 'grad_norm': 1.0307674407958984, 'learning_rate': 4.99900028620837e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.88, 'epoch': 0.43}
11%|█ | 4750/43939 [7:42:09<62:15:43, 5.72s/it]
11%|█ | 4751/43939 [7:42:14<62:12:29, 5.71s/it]
11%|█ | 4752/43939 [7:42:20<62:08:15, 5.71s/it]
11%|█ | 4753/43939 [7:42:26<62:18:26, 5.72s/it]
11%|█ | 4754/43939 [7:42:32<62:52:28, 5.78s/it]
11%|█ | 4755/43939 [7:42:37<62:41:52, 5.76s/it]
11%|█ | 4756/43939 [7:42:43<62:26:03, 5.74s/it]
11%|█ | 4757/43939 [7:42:49<62:18:49, 5.73s/it]
11%|█ | 4758/43939 [7:42:54<62:16:38, 5.72s/it]
11%|█ | 4759/43939 [7:43:00<62:12:07, 5.72s/it]
11%|█ | 4760/43939 [7:43:06<62:13:45, 5.72s/it]
11%|█ | 4761/43939 [7:43:11<62:04:31, 5.70s/it]
11%|█ | 4762/43939 [7:43:17<62:00:11, 5.70s/it]
11%|█ | 4763/43939 [7:43:23<61:57:14, 5.69s/it]
11%|█ | 4764/43939 [7:43:29<61:57:36, 5.69s/it]
11%|█ | 4765/43939 [7:43:34<61:58:40, 5.70s/it]
11%|█ | 4766/43939 [7:43:40<61:53:39, 5.69s/it]
11%|█ | 4767/43939 [7:43:46<61:52:15, 5.69s/it]
11%|█ | 4768/43939 [7:43:51<62:04:15, 5.70s/it]
11%|█ | 4769/43939 [7:43:57<61:50:51, 5.68s/it]
11%|█ | 4770/43939 [7:44:03<61:52:00, 5.69s/it]
11%|█ | 4771/43939 [7:44:08<61:49:35, 5.68s/it]
11%|█ | 4772/43939 [7:44:14<61:56:11, 5.69s/it]
11%|█ | 4773/43939 [7:44:20<61:55:57, 5.69s/it]
11%|█ | 4774/43939 [7:44:25<61:56:56, 5.69s/it]
11%|█ | 4775/43939 [7:44:31<62:43:47, 5.77s/it]
11%|█ | 4776/43939 [7:44:37<62:29:52, 5.75s/it]
11%|█ | 4777/43939 [7:44:43<62:30:26, 5.75s/it]
11%|█ | 4778/43939 [7:44:49<62:18:05, 5.73s/it]
11%|█ | 4779/43939 [7:44:54<62:19:27, 5.73s/it]
11%|█ | 4780/43939 [7:45:00<62:20:08, 5.73s/it]
11%|█ | 4781/43939 [7:45:06<62:13:35, 5.72s/it]
11%|█ | 4782/43939 [7:45:11<62:08:40, 5.71s/it]
11%|█ | 4783/43939 [7:45:17<62:03:35, 5.71s/it]
11%|█ | 4784/43939 [7:45:23<61:58:57, 5.70s/it]
11%|█ | 4785/43939 [7:45:28<61:59:07, 5.70s/it]
11%|█ | 4786/43939 [7:45:34<61:56:57, 5.70s/it]
11%|█ | 4787/43939 [7:45:40<61:50:36, 5.69s/it]
11%|█ | 4788/43939 [7:45:46<61:59:36, 5.70s/it]
11%|█ | 4789/43939 [7:45:51<62:03:22, 5.71s/it]
11%|█ | 4790/43939 [7:45:57<62:13:19, 5.72s/it]
11%|█ | 4791/43939 [7:46:03<62:10:41, 5.72s/it]
11%|█ | 4792/43939 [7:46:08<62:04:26, 5.71s/it]
11%|█ | 4793/43939 [7:46:14<62:03:05, 5.71s/it]
11%|█ | 4794/43939 [7:46:20<61:55:58, 5.70s/it]
11%|█ | 4795/43939 [7:46:26<62:02:13, 5.71s/it]
11%|█ | 4796/43939 [7:46:31<62:04:31, 5.71s/it]
11%|█ | 4797/43939 [7:46:37<62:48:00, 5.78s/it]
11%|█ | 4798/43939 [7:46:43<62:38:18, 5.76s/it]
11%|█ | 4799/43939 [7:46:49<62:22:47, 5.74s/it]
11%|█ | 4800/43939 [7:46:54<62:18:45, 5.73s/it]
{'loss': 1.088, 'grad_norm': 0.9278285503387451, 'learning_rate': 4.9986997733822826e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.35, 'epoch': 0.44}
11%|█ | 4800/43939 [7:46:54<62:18:45, 5.73s/it]
11%|█ | 4801/43939 [7:47:00<62:09:12, 5.72s/it]
11%|█ | 4802/43939 [7:47:06<62:07:40, 5.71s/it]
11%|█ | 4803/43939 [7:47:11<62:05:43, 5.71s/it]
11%|█ | 4804/43939 [7:47:17<62:05:18, 5.71s/it]
11%|█ | 4805/43939 [7:47:23<62:01:59, 5.71s/it]
11%|█ | 4806/43939 [7:47:28<62:00:06, 5.70s/it]
11%|█ | 4807/43939 [7:47:34<62:02:11, 5.71s/it]
11%|█ | 4808/43939 [7:47:40<62:05:45, 5.71s/it]
11%|█ | 4809/43939 [7:47:46<62:01:15, 5.71s/it]
11%|█ | 4810/43939 [7:47:51<62:05:42, 5.71s/it]
11%|█ | 4811/43939 [7:47:57<62:05:46, 5.71s/it]
11%|█ | 4812/43939 [7:48:03<62:01:06, 5.71s/it]
11%|█ | 4813/43939 [7:48:08<61:50:40, 5.69s/it]
11%|█ | 4814/43939 [7:48:14<61:53:41, 5.70s/it]
11%|█ | 4815/43939 [7:48:20<61:49:31, 5.69s/it]
11%|█ | 4816/43939 [7:48:26<61:58:56, 5.70s/it]
11%|█ | 4817/43939 [7:48:31<62:48:25, 5.78s/it]
11%|█ | 4818/43939 [7:48:37<62:29:46, 5.75s/it]
11%|█ | 4819/43939 [7:48:43<62:16:19, 5.73s/it]
11%|█ | 4820/43939 [7:48:49<62:11:20, 5.72s/it]
11%|█ | 4821/43939 [7:48:54<62:01:54, 5.71s/it]
11%|█ | 4822/43939 [7:49:00<61:56:23, 5.70s/it]
11%|█ | 4823/43939 [7:49:06<61:52:06, 5.69s/it]
11%|█ | 4824/43939 [7:49:11<61:54:47, 5.70s/it]
11%|█ | 4825/43939 [7:49:17<61:59:35, 5.71s/it]
11%|█ | 4826/43939 [7:49:23<61:53:31, 5.70s/it]
11%|█ | 4827/43939 [7:49:28<62:05:33, 5.72s/it]
11%|█ | 4828/43939 [7:49:34<62:03:48, 5.71s/it]
11%|█ | 4829/43939 [7:49:40<61:52:02, 5.69s/it]
11%|█ | 4830/43939 [7:49:46<61:51:26, 5.69s/it]
11%|█ | 4831/43939 [7:49:51<61:55:58, 5.70s/it]
11%|█ | 4832/43939 [7:49:57<62:05:29, 5.72s/it]
11%|█ | 4833/43939 [7:50:03<62:04:26, 5.71s/it]
11%|█ | 4834/43939 [7:50:08<61:57:58, 5.70s/it]
11%|█ | 4835/43939 [7:50:14<61:52:52, 5.70s/it]
11%|█ | 4836/43939 [7:50:20<61:50:33, 5.69s/it]
11%|█ | 4837/43939 [7:50:25<61:58:30, 5.71s/it]
11%|█ | 4838/43939 [7:50:31<62:41:00, 5.77s/it]
11%|█ | 4839/43939 [7:50:37<62:24:23, 5.75s/it]
11%|█ | 4840/43939 [7:50:43<62:09:05, 5.72s/it]
11%|█ | 4841/43939 [7:50:48<62:01:52, 5.71s/it]
11%|█ | 4842/43939 [7:50:54<62:04:19, 5.72s/it]
11%|█ | 4843/43939 [7:51:00<62:02:10, 5.71s/it]
11%|█ | 4844/43939 [7:51:06<62:03:03, 5.71s/it]
11%|█ | 4845/43939 [7:51:11<61:57:12, 5.71s/it]
11%|█ | 4846/43939 [7:51:17<61:48:40, 5.69s/it]
11%|█ | 4847/43939 [7:51:23<61:57:12, 5.71s/it]
11%|█ | 4848/43939 [7:51:28<61:57:20, 5.71s/it]
11%|█ | 4849/43939 [7:51:34<61:51:23, 5.70s/it]
11%|█ | 4850/43939 [7:51:40<61:53:55, 5.70s/it]
{'loss': 1.0781, 'grad_norm': 1.2356767654418945, 'learning_rate': 4.998359837695407e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.56, 'epoch': 0.44}
11%|█ | 4850/43939 [7:51:40<61:53:55, 5.70s/it]
11%|█ | 4851/43939 [7:51:45<61:52:22, 5.70s/it]
11%|█ | 4852/43939 [7:51:51<61:55:13, 5.70s/it]
11%|█ | 4853/43939 [7:51:57<62:02:42, 5.71s/it]
11%|█ | 4854/43939 [7:52:03<62:02:34, 5.71s/it]
11%|█ | 4855/43939 [7:52:08<61:58:11, 5.71s/it]
11%|█ | 4856/43939 [7:52:14<61:54:30, 5.70s/it]
11%|█ | 4857/43939 [7:52:20<61:50:33, 5.70s/it]
11%|█ | 4858/43939 [7:52:25<61:54:16, 5.70s/it]
11%|█ | 4859/43939 [7:52:31<62:42:31, 5.78s/it]
11%|█ | 4860/43939 [7:52:37<62:28:58, 5.76s/it]
11%|█ | 4861/43939 [7:52:43<62:12:18, 5.73s/it]
11%|█ | 4862/43939 [7:52:48<62:06:55, 5.72s/it]
11%|█ | 4863/43939 [7:52:54<62:10:20, 5.73s/it]
11%|█ | 4864/43939 [7:53:00<62:08:10, 5.72s/it]
11%|█ | 4865/43939 [7:53:06<62:06:57, 5.72s/it]
11%|█ | 4866/43939 [7:53:11<62:03:30, 5.72s/it]
11%|█ | 4867/43939 [7:53:17<61:58:57, 5.71s/it]
11%|█ | 4868/43939 [7:53:23<61:51:40, 5.70s/it]
11%|█ | 4869/43939 [7:53:28<61:52:57, 5.70s/it]
11%|█ | 4870/43939 [7:53:34<61:57:01, 5.71s/it]
11%|█ | 4871/43939 [7:53:40<61:49:38, 5.70s/it]
11%|█ | 4872/43939 [7:53:45<61:50:04, 5.70s/it]
11%|█ | 4873/43939 [7:53:51<61:48:30, 5.70s/it]
11%|█ | 4874/43939 [7:53:57<61:52:41, 5.70s/it]
11%|█ | 4875/43939 [7:54:03<61:56:17, 5.71s/it]
11%|█ | 4876/43939 [7:54:08<61:54:13, 5.70s/it]
11%|█ | 4877/43939 [7:54:14<61:43:34, 5.69s/it]
11%|█ | 4878/43939 [7:54:20<61:48:27, 5.70s/it]
11%|█ | 4879/43939 [7:54:25<61:51:51, 5.70s/it]
11%|█ | 4880/43939 [7:54:31<61:51:19, 5.70s/it]
11%|█ | 4881/43939 [7:54:37<62:24:32, 5.75s/it]
11%|█ | 4882/43939 [7:54:43<62:10:50, 5.73s/it]
11%|█ | 4883/43939 [7:54:48<62:08:49, 5.73s/it]
11%|█ | 4884/43939 [7:54:54<62:02:13, 5.72s/it]
11%|█ | 4885/43939 [7:55:00<61:59:44, 5.71s/it]
11%|█ | 4886/43939 [7:55:05<61:57:04, 5.71s/it]
11%|█ | 4887/43939 [7:55:11<61:53:56, 5.71s/it]
11%|█ | 4888/43939 [7:55:17<61:48:22, 5.70s/it]
11%|█ | 4889/43939 [7:55:23<61:46:50, 5.70s/it]
11%|█ | 4890/43939 [7:55:28<61:53:55, 5.71s/it]
11%|█ | 4891/43939 [7:55:34<61:44:51, 5.69s/it]
11%|█ | 4892/43939 [7:55:40<61:45:16, 5.69s/it]
11%|█ | 4893/43939 [7:55:45<61:42:53, 5.69s/it]
11%|█ | 4894/43939 [7:55:51<61:38:18, 5.68s/it]
11%|█ | 4895/43939 [7:55:57<61:45:16, 5.69s/it]
11%|█ | 4896/43939 [7:56:02<61:47:03, 5.70s/it]
11%|█ | 4897/43939 [7:56:08<61:51:02, 5.70s/it]
11%|█ | 4898/43939 [7:56:14<61:47:13, 5.70s/it]
11%|█ | 4899/43939 [7:56:19<61:47:58, 5.70s/it]
11%|█ | 4900/43939 [7:56:25<61:54:23, 5.71s/it]
{'loss': 1.1009, 'grad_norm': 0.9927319884300232, 'learning_rate': 4.997980484511027e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.86, 'epoch': 0.45}
11%|█ | 4900/43939 [7:56:25<61:54:23, 5.71s/it]
11%|█ | 4901/43939 [7:56:31<62:04:48, 5.72s/it]
11%|█ | 4902/43939 [7:56:37<62:34:07, 5.77s/it]
11%|█ | 4903/43939 [7:56:43<62:20:30, 5.75s/it]
11%|█ | 4904/43939 [7:56:48<62:05:37, 5.73s/it]
11%|█ | 4905/43939 [7:56:54<61:56:50, 5.71s/it]
11%|█ | 4906/43939 [7:57:00<61:48:02, 5.70s/it]
11%|█ | 4907/43939 [7:57:05<61:49:02, 5.70s/it]
11%|█ | 4908/43939 [7:57:11<61:48:43, 5.70s/it]
11%|█ | 4909/43939 [7:57:17<61:57:03, 5.71s/it]
11%|█ | 4910/43939 [7:57:22<61:53:26, 5.71s/it]
11%|█ | 4911/43939 [7:57:28<61:52:00, 5.71s/it]
11%|█ | 4912/43939 [7:57:34<61:48:47, 5.70s/it]
11%|█ | 4913/43939 [7:57:40<61:45:42, 5.70s/it]
11%|█ | 4914/43939 [7:57:45<61:42:27, 5.69s/it]
11%|█ | 4915/43939 [7:57:51<61:40:43, 5.69s/it]
11%|█ | 4916/43939 [7:57:57<61:48:59, 5.70s/it]
11%|█ | 4917/43939 [7:58:02<61:57:21, 5.72s/it]
11%|█ | 4918/43939 [7:58:08<61:54:38, 5.71s/it]
11%|█ | 4919/43939 [7:58:14<61:48:08, 5.70s/it]
11%|█ | 4920/43939 [7:58:19<61:53:37, 5.71s/it]
11%|█ | 4921/43939 [7:58:25<61:59:13, 5.72s/it]
11%|█ | 4922/43939 [7:58:31<62:44:34, 5.79s/it]
11%|█ | 4923/43939 [7:58:37<62:16:10, 5.75s/it]
11%|█ | 4924/43939 [7:58:43<62:09:53, 5.74s/it]
11%|█ | 4925/43939 [7:58:48<61:57:15, 5.72s/it]
11%|█ | 4926/43939 [7:58:54<61:53:10, 5.71s/it]
11%|█ | 4927/43939 [7:59:00<61:43:40, 5.70s/it]
11%|█ | 4928/43939 [7:59:05<61:38:31, 5.69s/it]
11%|█ | 4929/43939 [7:59:11<61:42:08, 5.69s/it]
11%|█ | 4930/43939 [7:59:17<61:48:11, 5.70s/it]
11%|█ | 4931/43939 [7:59:22<61:46:06, 5.70s/it]
11%|█ | 4932/43939 [7:59:28<61:55:36, 5.72s/it]
11%|█ | 4933/43939 [7:59:34<61:56:15, 5.72s/it]
11%|█ | 4934/43939 [7:59:40<62:07:05, 5.73s/it]
11%|█ | 4935/43939 [7:59:45<61:56:45, 5.72s/it]
11%|█ | 4936/43939 [7:59:51<62:04:12, 5.73s/it]
11%|█ | 4937/43939 [7:59:57<62:00:38, 5.72s/it]
11%|█ | 4938/43939 [8:00:02<61:55:33, 5.72s/it]
11%|█ | 4939/43939 [8:00:08<61:49:47, 5.71s/it]
11%|█ | 4940/43939 [8:00:14<61:48:36, 5.71s/it]
11%|█ | 4941/43939 [8:00:20<61:49:56, 5.71s/it]
11%|█ | 4942/43939 [8:00:25<61:59:01, 5.72s/it]
11%|█ | 4943/43939 [8:00:31<61:50:58, 5.71s/it]
11%|█▏ | 4944/43939 [8:00:37<62:39:14, 5.78s/it]
11%|█▏ | 4945/43939 [8:00:43<62:17:11, 5.75s/it]
11%|█▏ | 4946/43939 [8:00:48<62:05:13, 5.73s/it]
11%|█▏ | 4947/43939 [8:00:54<62:11:27, 5.74s/it]
11%|█▏ | 4948/43939 [8:01:00<61:58:33, 5.72s/it]
11%|█▏ | 4949/43939 [8:01:05<61:50:48, 5.71s/it]
11%|█▏ | 4950/43939 [8:01:11<61:53:28, 5.71s/it]
{'loss': 1.0898, 'grad_norm': 0.9227324724197388, 'learning_rate': 4.997561719814329e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 520.64, 'epoch': 0.45}
11%|█▏ | 4950/43939 [8:01:11<61:53:28, 5.71s/it]
11%|█▏ | 4951/43939 [8:01:17<61:40:39, 5.70s/it]
11%|█▏ | 4952/43939 [8:01:22<61:42:42, 5.70s/it]
11%|█▏ | 4953/43939 [8:01:28<61:34:36, 5.69s/it]
11%|█▏ | 4954/43939 [8:01:34<61:43:14, 5.70s/it]
11%|█▏ | 4955/43939 [8:01:40<61:42:49, 5.70s/it]
11%|█▏ | 4956/43939 [8:01:45<61:43:16, 5.70s/it]
11%|█▏ | 4957/43939 [8:01:51<61:36:16, 5.69s/it]
11%|█▏ | 4958/43939 [8:01:57<61:38:45, 5.69s/it]
11%|█▏ | 4959/43939 [8:02:02<61:41:45, 5.70s/it]
11%|█▏ | 4960/43939 [8:02:08<61:36:56, 5.69s/it]
11%|█▏ | 4961/43939 [8:02:14<61:44:20, 5.70s/it]
11%|█▏ | 4962/43939 [8:02:19<61:50:35, 5.71s/it]
11%|█▏ | 4963/43939 [8:02:25<61:46:51, 5.71s/it]
11%|█▏ | 4964/43939 [8:02:31<62:26:24, 5.77s/it]
11%|█▏ | 4965/43939 [8:02:37<62:12:13, 5.75s/it]
11%|█▏ | 4966/43939 [8:02:42<62:03:53, 5.73s/it]
11%|█▏ | 4967/43939 [8:02:48<61:59:45, 5.73s/it]
11%|█▏ | 4968/43939 [8:02:54<61:56:43, 5.72s/it]
11%|█▏ | 4969/43939 [8:03:00<61:50:08, 5.71s/it]
11%|█▏ | 4970/43939 [8:03:05<61:44:07, 5.70s/it]
11%|█▏ | 4971/43939 [8:03:11<61:45:28, 5.71s/it]
11%|█▏ | 4972/43939 [8:03:17<61:42:23, 5.70s/it]
11%|█▏ | 4973/43939 [8:03:22<61:36:29, 5.69s/it]
11%|█▏ | 4974/43939 [8:03:28<61:41:09, 5.70s/it]
11%|█▏ | 4975/43939 [8:03:34<61:43:13, 5.70s/it]
11%|█▏ | 4976/43939 [8:03:39<61:42:00, 5.70s/it]
11%|█▏ | 4977/43939 [8:03:45<61:35:39, 5.69s/it]
11%|█▏ | 4978/43939 [8:03:51<61:54:56, 5.72s/it]
11%|█▏ | 4979/43939 [8:03:57<61:49:50, 5.71s/it]
11%|█▏ | 4980/43939 [8:04:02<61:45:53, 5.71s/it]
11%|█▏ | 4981/43939 [8:04:08<61:40:33, 5.70s/it]
11%|█▏ | 4982/43939 [8:04:14<61:41:51, 5.70s/it]
11%|█▏ | 4983/43939 [8:04:19<61:36:23, 5.69s/it]
11%|█▏ | 4984/43939 [8:04:25<61:34:38, 5.69s/it]
11%|█▏ | 4985/43939 [8:04:31<62:22:16, 5.76s/it]
11%|█▏ | 4986/43939 [8:04:37<62:12:17, 5.75s/it]
11%|█▏ | 4987/43939 [8:04:42<62:06:48, 5.74s/it]
11%|█▏ | 4988/43939 [8:04:48<62:06:44, 5.74s/it]
11%|█▏ | 4989/43939 [8:04:54<61:52:27, 5.72s/it]
11%|█▏ | 4990/43939 [8:05:00<61:48:13, 5.71s/it]
11%|█▏ | 4991/43939 [8:05:05<61:45:16, 5.71s/it]
11%|█▏ | 4992/43939 [8:05:11<61:38:10, 5.70s/it]
11%|█▏ | 4993/43939 [8:05:17<61:38:47, 5.70s/it]
11%|█▏ | 4994/43939 [8:05:22<61:27:33, 5.68s/it]
11%|█▏ | 4995/43939 [8:05:28<61:27:28, 5.68s/it]
11%|█▏ | 4996/43939 [8:05:34<61:33:10, 5.69s/it]
11%|█▏ | 4997/43939 [8:05:39<61:25:23, 5.68s/it]
11%|█▏ | 4998/43939 [8:05:45<61:25:37, 5.68s/it]
11%|█▏ | 4999/43939 [8:05:51<61:30:43, 5.69s/it]
11%|█▏ | 5000/43939 [8:05:56<61:27:17, 5.68s/it]
{'loss': 1.0729, 'grad_norm': 1.0663466453552246, 'learning_rate': 4.9971035502123126e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.18, 'epoch': 0.46}
11%|█▏ | 5000/43939 [8:05:56<61:27:17, 5.68s/it]
11%|█▏ | 5001/43939 [8:06:02<61:34:21, 5.69s/it]
11%|█▏ | 5002/43939 [8:06:08<61:40:56, 5.70s/it]
11%|█▏ | 5003/43939 [8:06:14<61:39:52, 5.70s/it]
11%|█▏ | 5004/43939 [8:06:19<61:45:43, 5.71s/it]
11%|█▏ | 5005/43939 [8:06:25<61:46:57, 5.71s/it]
11%|█▏ | 5006/43939 [8:06:31<61:43:20, 5.71s/it]
11%|█▏ | 5007/43939 [8:06:37<62:30:49, 5.78s/it]
11%|█▏ | 5008/43939 [8:06:42<62:11:54, 5.75s/it]
11%|█▏ | 5009/43939 [8:06:48<62:04:12, 5.74s/it]
11%|█▏ | 5010/43939 [8:06:54<62:04:06, 5.74s/it]
11%|█▏ | 5011/43939 [8:06:59<61:55:05, 5.73s/it]
11%|█▏ | 5012/43939 [8:07:05<61:52:50, 5.72s/it]
11%|█▏ | 5013/43939 [8:07:11<61:42:26, 5.71s/it]
11%|█▏ | 5014/43939 [8:07:17<61:46:24, 5.71s/it]
11%|█▏ | 5015/43939 [8:07:22<61:38:30, 5.70s/it]
11%|█▏ | 5016/43939 [8:07:28<61:34:46, 5.70s/it]
11%|█▏ | 5017/43939 [8:07:34<61:36:17, 5.70s/it]
11%|█▏ | 5018/43939 [8:07:39<61:42:55, 5.71s/it]
11%|█▏ | 5019/43939 [8:07:45<61:37:20, 5.70s/it]
11%|█▏ | 5020/43939 [8:07:51<61:31:38, 5.69s/it]
11%|█▏ | 5021/43939 [8:07:56<61:34:23, 5.70s/it]
11%|█▏ | 5022/43939 [8:08:02<61:31:27, 5.69s/it]
11%|█▏ | 5023/43939 [8:08:08<61:28:04, 5.69s/it]
11%|█▏ | 5024/43939 [8:08:13<61:23:34, 5.68s/it]
11%|█▏ | 5025/43939 [8:08:19<61:28:54, 5.69s/it]
11%|█▏ | 5026/43939 [8:08:25<61:30:59, 5.69s/it]
11%|█▏ | 5027/43939 [8:08:31<62:08:24, 5.75s/it]
11%|█▏ | 5028/43939 [8:08:36<62:04:45, 5.74s/it]
11%|█▏ | 5029/43939 [8:08:42<62:02:55, 5.74s/it]
11%|█▏ | 5030/43939 [8:08:48<62:06:11, 5.75s/it]
11%|█▏ | 5031/43939 [8:08:54<61:57:33, 5.73s/it]
11%|█▏ | 5032/43939 [8:08:59<61:54:44, 5.73s/it]
11%|█▏ | 5033/43939 [8:09:05<61:43:19, 5.71s/it]
11%|█▏ | 5034/43939 [8:09:11<61:46:06, 5.72s/it]
11%|█▏ | 5035/43939 [8:09:16<61:52:02, 5.72s/it]
11%|█▏ | 5036/43939 [8:09:22<61:44:18, 5.71s/it]
11%|█▏ | 5037/43939 [8:09:28<61:32:16, 5.69s/it]
11%|█▏ | 5038/43939 [8:09:34<61:35:49, 5.70s/it]
11%|█▏ | 5039/43939 [8:09:39<61:35:17, 5.70s/it]
11%|█▏ | 5040/43939 [8:09:45<61:38:03, 5.70s/it]
11%|█▏ | 5041/43939 [8:09:51<61:36:51, 5.70s/it]
11%|█▏ | 5042/43939 [8:09:56<61:40:25, 5.71s/it]
11%|█▏ | 5043/43939 [8:10:02<61:36:45, 5.70s/it]
11%|█▏ | 5044/43939 [8:10:08<61:31:58, 5.70s/it]
11%|█▏ | 5045/43939 [8:10:13<61:33:53, 5.70s/it]
11%|█▏ | 5046/43939 [8:10:19<61:31:12, 5.69s/it]
11%|█▏ | 5047/43939 [8:10:25<61:46:59, 5.72s/it]
11%|█▏ | 5048/43939 [8:10:31<62:28:16, 5.78s/it]
11%|█▏ | 5049/43939 [8:10:37<62:14:35, 5.76s/it]
11%|█▏ | 5050/43939 [8:10:42<62:05:21, 5.75s/it]
{'loss': 1.069, 'grad_norm': 0.9325670599937439, 'learning_rate': 4.9966059829336786e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.12, 'epoch': 0.46}
11%|█▏ | 5050/43939 [8:10:42<62:05:21, 5.75s/it]
11%|█▏ | 5051/43939 [8:10:48<61:51:32, 5.73s/it]
11%|█▏ | 5052/43939 [8:10:54<61:41:09, 5.71s/it]
12%|█▏ | 5053/43939 [8:10:59<61:36:38, 5.70s/it]
12%|█▏ | 5054/43939 [8:11:05<61:35:47, 5.70s/it]
12%|█▏ | 5055/43939 [8:11:11<61:26:39, 5.69s/it]
12%|█▏ | 5056/43939 [8:11:16<61:24:14, 5.69s/it]
12%|█▏ | 5057/43939 [8:11:22<61:17:35, 5.68s/it]
12%|█▏ | 5058/43939 [8:11:28<61:18:48, 5.68s/it]
12%|█▏ | 5059/43939 [8:11:33<61:33:57, 5.70s/it]
12%|█▏ | 5060/43939 [8:11:39<61:36:26, 5.70s/it]
12%|█▏ | 5061/43939 [8:11:45<61:32:21, 5.70s/it]
12%|█▏ | 5062/43939 [8:11:51<61:27:27, 5.69s/it]
12%|█▏ | 5063/43939 [8:11:56<61:33:44, 5.70s/it]
12%|█▏ | 5064/43939 [8:12:02<61:36:09, 5.70s/it]
12%|█▏ | 5065/43939 [8:12:08<61:33:48, 5.70s/it]
12%|█▏ | 5066/43939 [8:12:13<61:33:10, 5.70s/it]
12%|█▏ | 5067/43939 [8:12:19<61:31:30, 5.70s/it]
12%|█▏ | 5068/43939 [8:12:25<61:29:00, 5.69s/it]
12%|█▏ | 5069/43939 [8:12:30<61:32:58, 5.70s/it]
12%|█▏ | 5070/43939 [8:12:36<62:10:37, 5.76s/it]
12%|█▏ | 5071/43939 [8:12:42<62:00:12, 5.74s/it]
12%|█▏ | 5072/43939 [8:12:48<61:49:57, 5.73s/it]
12%|█▏ | 5073/43939 [8:12:53<61:32:53, 5.70s/it]
12%|█▏ | 5074/43939 [8:12:59<61:38:17, 5.71s/it]
12%|█▏ | 5075/43939 [8:13:05<61:36:00, 5.71s/it]
12%|█▏ | 5076/43939 [8:13:11<61:38:47, 5.71s/it]
12%|█▏ | 5077/43939 [8:13:16<61:39:37, 5.71s/it]
12%|█▏ | 5078/43939 [8:13:22<61:34:09, 5.70s/it]
12%|█▏ | 5079/43939 [8:13:28<61:25:39, 5.69s/it]
12%|█▏ | 5080/43939 [8:13:33<61:20:29, 5.68s/it]
12%|█▏ | 5081/43939 [8:13:39<61:29:44, 5.70s/it]
12%|█▏ | 5082/43939 [8:13:45<61:27:08, 5.69s/it]
12%|█▏ | 5083/43939 [8:13:50<61:25:33, 5.69s/it]
12%|█▏ | 5084/43939 [8:13:56<61:24:15, 5.69s/it]
12%|█▏ | 5085/43939 [8:14:02<61:28:11, 5.70s/it]
12%|█▏ | 5086/43939 [8:14:07<61:36:31, 5.71s/it]
12%|█▏ | 5087/43939 [8:14:13<61:29:41, 5.70s/it]
12%|█▏ | 5088/43939 [8:14:19<61:31:42, 5.70s/it]
12%|█▏ | 5089/43939 [8:14:25<61:29:23, 5.70s/it]
12%|█▏ | 5090/43939 [8:14:30<62:02:54, 5.75s/it]
12%|█▏ | 5091/43939 [8:14:36<61:48:42, 5.73s/it]
12%|█▏ | 5092/43939 [8:14:42<61:33:18, 5.70s/it]
12%|█▏ | 5093/43939 [8:14:47<61:31:56, 5.70s/it]
12%|█▏ | 5094/43939 [8:14:53<61:29:10, 5.70s/it]
12%|█▏ | 5095/43939 [8:14:59<61:30:47, 5.70s/it]
12%|█▏ | 5096/43939 [8:15:05<61:30:00, 5.70s/it]
12%|█▏ | 5097/43939 [8:15:10<61:28:54, 5.70s/it]
12%|█▏ | 5098/43939 [8:15:16<61:23:26, 5.69s/it]
12%|█▏ | 5099/43939 [8:15:22<61:31:32, 5.70s/it]
12%|█▏ | 5100/43939 [8:15:27<61:38:08, 5.71s/it]
{'loss': 1.0801, 'grad_norm': 1.1680859327316284, 'learning_rate': 4.9960690258287204e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.73, 'epoch': 0.46}
12%|█▏ | 5100/43939 [8:15:27<61:38:08, 5.71s/it]
12%|█▏ | 5101/43939 [8:15:33<61:35:20, 5.71s/it]
12%|█▏ | 5102/43939 [8:15:39<61:31:31, 5.70s/it]
12%|█▏ | 5103/43939 [8:15:44<61:30:16, 5.70s/it]
12%|█▏ | 5104/43939 [8:15:50<61:27:44, 5.70s/it]
12%|█▏ | 5105/43939 [8:15:56<61:28:00, 5.70s/it]
12%|█▏ | 5106/43939 [8:16:02<61:35:06, 5.71s/it]
12%|█▏ | 5107/43939 [8:16:07<61:32:34, 5.71s/it]
12%|█▏ | 5108/43939 [8:16:13<61:42:26, 5.72s/it]
12%|█▏ | 5109/43939 [8:16:19<61:29:39, 5.70s/it]
12%|█▏ | 5110/43939 [8:16:24<61:34:02, 5.71s/it]
12%|█▏ | 5111/43939 [8:16:30<62:09:58, 5.76s/it]
12%|█▏ | 5112/43939 [8:16:36<62:00:54, 5.75s/it]
12%|█▏ | 5113/43939 [8:16:42<61:46:56, 5.73s/it]
12%|█▏ | 5114/43939 [8:16:47<61:46:44, 5.73s/it]
12%|█▏ | 5115/43939 [8:16:53<61:32:30, 5.71s/it]
12%|█▏ | 5116/43939 [8:16:59<61:33:20, 5.71s/it]
12%|█▏ | 5117/43939 [8:17:04<61:27:37, 5.70s/it]
12%|█▏ | 5118/43939 [8:17:10<61:36:06, 5.71s/it]
12%|█▏ | 5119/43939 [8:17:16<61:30:18, 5.70s/it]
12%|█▏ | 5120/43939 [8:17:22<61:30:27, 5.70s/it]
12%|█▏ | 5121/43939 [8:17:27<61:35:28, 5.71s/it]
12%|█▏ | 5122/43939 [8:17:33<61:30:35, 5.70s/it]
12%|█▏ | 5123/43939 [8:17:39<61:25:09, 5.70s/it]
12%|█▏ | 5124/43939 [8:17:44<61:21:06, 5.69s/it]
12%|█▏ | 5125/43939 [8:17:50<61:22:49, 5.69s/it]
12%|█▏ | 5126/43939 [8:17:56<61:23:48, 5.69s/it]
12%|█▏ | 5127/43939 [8:18:01<61:27:01, 5.70s/it]
12%|█▏ | 5128/43939 [8:18:07<61:32:47, 5.71s/it]
12%|█▏ | 5129/43939 [8:18:13<61:36:37, 5.71s/it]
12%|█▏ | 5130/43939 [8:18:19<61:33:04, 5.71s/it]
12%|█▏ | 5131/43939 [8:18:24<61:37:42, 5.72s/it]
12%|█▏ | 5132/43939 [8:18:30<61:34:02, 5.71s/it]
12%|█▏ | 5133/43939 [8:18:36<62:07:50, 5.76s/it]
12%|█▏ | 5134/43939 [8:18:42<61:51:17, 5.74s/it]
12%|█▏ | 5135/43939 [8:18:47<61:35:41, 5.71s/it]
12%|█▏ | 5136/43939 [8:18:53<61:33:27, 5.71s/it]
12%|█▏ | 5137/43939 [8:18:59<61:25:14, 5.70s/it]
12%|█▏ | 5138/43939 [8:19:04<61:23:25, 5.70s/it]
12%|█▏ | 5139/43939 [8:19:10<61:22:40, 5.69s/it]
12%|█▏ | 5140/43939 [8:19:16<61:21:05, 5.69s/it]
12%|█▏ | 5141/43939 [8:19:21<61:13:19, 5.68s/it]
12%|█▏ | 5142/43939 [8:19:27<61:11:27, 5.68s/it]
12%|█▏ | 5143/43939 [8:19:33<61:10:34, 5.68s/it]
12%|█▏ | 5144/43939 [8:19:38<61:09:51, 5.68s/it]
12%|█▏ | 5145/43939 [8:19:44<61:10:22, 5.68s/it]
12%|█▏ | 5146/43939 [8:19:50<61:18:16, 5.69s/it]
12%|█▏ | 5147/43939 [8:19:55<61:19:30, 5.69s/it]
12%|█▏ | 5148/43939 [8:20:01<61:20:02, 5.69s/it]
12%|█▏ | 5149/43939 [8:20:07<61:23:05, 5.70s/it]
12%|█▏ | 5150/43939 [8:20:13<61:28:38, 5.71s/it]
{'loss': 1.0803, 'grad_norm': 1.132150411605835, 'learning_rate': 4.995492687369198e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.93, 'epoch': 0.47}
12%|█▏ | 5150/43939 [8:20:13<61:28:38, 5.71s/it]
12%|█▏ | 5151/43939 [8:20:18<61:31:42, 5.71s/it]
12%|█▏ | 5152/43939 [8:20:24<61:19:07, 5.69s/it]
12%|█▏ | 5153/43939 [8:20:30<61:17:35, 5.69s/it]
12%|█▏ | 5154/43939 [8:20:36<61:57:28, 5.75s/it]
12%|█▏ | 5155/43939 [8:20:41<61:48:00, 5.74s/it]
12%|█▏ | 5156/43939 [8:20:47<61:38:21, 5.72s/it]
12%|█▏ | 5157/43939 [8:20:53<61:30:48, 5.71s/it]
12%|█▏ | 5158/43939 [8:20:58<61:36:07, 5.72s/it]
12%|█▏ | 5159/43939 [8:21:04<61:27:34, 5.71s/it]
12%|█▏ | 5160/43939 [8:21:10<61:34:26, 5.72s/it]
12%|█▏ | 5161/43939 [8:21:15<61:26:10, 5.70s/it]
12%|█▏ | 5162/43939 [8:21:21<61:32:55, 5.71s/it]
12%|█▏ | 5163/43939 [8:21:27<61:23:27, 5.70s/it]
12%|█▏ | 5164/43939 [8:21:33<61:26:49, 5.70s/it]
12%|█▏ | 5165/43939 [8:21:38<61:23:47, 5.70s/it]
12%|█▏ | 5166/43939 [8:21:44<61:33:02, 5.71s/it]
12%|█▏ | 5167/43939 [8:21:50<61:38:19, 5.72s/it]
12%|█▏ | 5168/43939 [8:21:55<61:35:36, 5.72s/it]
12%|█▏ | 5169/43939 [8:22:01<61:33:51, 5.72s/it]
12%|█▏ | 5170/43939 [8:22:07<61:29:14, 5.71s/it]
12%|█▏ | 5171/43939 [8:22:13<61:23:25, 5.70s/it]
12%|█▏ | 5172/43939 [8:22:18<61:18:06, 5.69s/it]
12%|█▏ | 5173/43939 [8:22:24<61:12:40, 5.68s/it]
12%|█▏ | 5174/43939 [8:22:30<61:11:06, 5.68s/it]
12%|█▏ | 5175/43939 [8:22:36<62:01:12, 5.76s/it]
12%|█▏ | 5176/43939 [8:22:41<61:48:41, 5.74s/it]
12%|█▏ | 5177/43939 [8:22:47<61:34:14, 5.72s/it]
12%|█▏ | 5178/43939 [8:22:53<61:31:20, 5.71s/it]
12%|█▏ | 5179/43939 [8:22:58<61:25:58, 5.71s/it]
12%|█▏ | 5180/43939 [8:23:04<61:30:49, 5.71s/it]
12%|█▏ | 5181/43939 [8:23:10<61:27:27, 5.71s/it]
12%|█▏ | 5182/43939 [8:23:15<61:26:48, 5.71s/it]
12%|█▏ | 5183/43939 [8:23:21<61:24:45, 5.70s/it]
12%|█▏ | 5184/43939 [8:23:27<61:13:36, 5.69s/it]
12%|█▏ | 5185/43939 [8:23:32<61:19:10, 5.70s/it]
12%|█▏ | 5186/43939 [8:23:38<61:24:59, 5.71s/it]
12%|█▏ | 5187/43939 [8:23:44<61:20:46, 5.70s/it]
12%|█▏ | 5188/43939 [8:23:50<61:24:41, 5.71s/it]
12%|█▏ | 5189/43939 [8:23:55<61:30:00, 5.71s/it]
12%|█▏ | 5190/43939 [8:24:01<61:27:47, 5.71s/it]
12%|█▏ | 5191/43939 [8:24:07<61:26:38, 5.71s/it]
12%|█▏ | 5192/43939 [8:24:12<61:25:55, 5.71s/it]
12%|█▏ | 5193/43939 [8:24:18<61:26:08, 5.71s/it]
12%|█▏ | 5194/43939 [8:24:24<61:18:35, 5.70s/it]
12%|█▏ | 5195/43939 [8:24:30<61:19:44, 5.70s/it]
12%|█▏ | 5196/43939 [8:24:36<62:15:43, 5.79s/it]
12%|█▏ | 5197/43939 [8:24:41<62:05:26, 5.77s/it]
12%|█▏ | 5198/43939 [8:24:47<61:51:50, 5.75s/it]
12%|█▏ | 5199/43939 [8:24:53<61:41:19, 5.73s/it]
12%|█▏ | 5200/43939 [8:24:58<61:30:29, 5.72s/it]
{'loss': 1.0829, 'grad_norm': 1.0051616430282593, 'learning_rate': 4.994876976648204e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.51, 'epoch': 0.47}
12%|█▏ | 5200/43939 [8:24:58<61:30:29, 5.72s/it]
12%|█▏ | 5201/43939 [8:25:04<61:21:57, 5.70s/it]
12%|█▏ | 5202/43939 [8:25:10<61:25:09, 5.71s/it]
12%|█▏ | 5203/43939 [8:25:15<61:19:43, 5.70s/it]
12%|█▏ | 5204/43939 [8:25:21<61:18:54, 5.70s/it]
12%|█▏ | 5205/43939 [8:25:27<61:27:00, 5.71s/it]
12%|█▏ | 5206/43939 [8:25:33<61:21:59, 5.70s/it]
12%|█▏ | 5207/43939 [8:25:38<61:17:50, 5.70s/it]
12%|█▏ | 5208/43939 [8:25:44<61:26:59, 5.71s/it]
12%|█▏ | 5209/43939 [8:25:50<61:29:26, 5.72s/it]
12%|█▏ | 5210/43939 [8:25:55<61:24:37, 5.71s/it]
12%|█▏ | 5211/43939 [8:26:01<61:19:43, 5.70s/it]
12%|█▏ | 5212/43939 [8:26:07<61:28:28, 5.71s/it]
12%|█▏ | 5213/43939 [8:26:12<61:21:58, 5.70s/it]
12%|█▏ | 5214/43939 [8:26:18<61:19:40, 5.70s/it]
12%|█▏ | 5215/43939 [8:26:24<61:35:45, 5.73s/it]
12%|█▏ | 5216/43939 [8:26:30<61:36:49, 5.73s/it]
12%|█▏ | 5217/43939 [8:26:36<62:08:20, 5.78s/it]
12%|█▏ | 5218/43939 [8:26:41<61:50:21, 5.75s/it]
12%|█▏ | 5219/43939 [8:26:47<61:41:14, 5.74s/it]
12%|█▏ | 5220/43939 [8:26:53<61:33:19, 5.72s/it]
12%|█▏ | 5221/43939 [8:26:58<61:35:02, 5.73s/it]
12%|█▏ | 5222/43939 [8:27:04<61:20:45, 5.70s/it]
12%|█▏ | 5223/43939 [8:27:10<61:19:07, 5.70s/it]
12%|█▏ | 5224/43939 [8:27:15<61:28:32, 5.72s/it]
12%|█▏ | 5225/43939 [8:27:21<61:26:30, 5.71s/it]
12%|█▏ | 5226/43939 [8:27:27<61:20:55, 5.70s/it]
12%|█▏ | 5227/43939 [8:27:33<61:27:48, 5.72s/it]
12%|█▏ | 5228/43939 [8:27:38<61:31:34, 5.72s/it]
12%|█▏ | 5229/43939 [8:27:44<61:23:04, 5.71s/it]
12%|█▏ | 5230/43939 [8:27:50<61:22:52, 5.71s/it]
12%|█▏ | 5231/43939 [8:27:55<61:19:34, 5.70s/it]
12%|█▏ | 5232/43939 [8:28:01<61:11:45, 5.69s/it]
12%|█▏ | 5233/43939 [8:28:07<61:15:45, 5.70s/it]
12%|█▏ | 5234/43939 [8:28:13<61:19:25, 5.70s/it]
12%|█▏ | 5235/43939 [8:28:18<61:08:26, 5.69s/it]
12%|█▏ | 5236/43939 [8:28:24<61:08:25, 5.69s/it]
12%|█▏ | 5237/43939 [8:28:30<61:10:22, 5.69s/it]
12%|█▏ | 5238/43939 [8:28:36<62:04:04, 5.77s/it]
12%|█▏ | 5239/43939 [8:28:41<61:48:27, 5.75s/it]
12%|█▏ | 5240/43939 [8:28:47<61:36:19, 5.73s/it]
12%|█▏ | 5241/43939 [8:28:53<61:29:54, 5.72s/it]
12%|█▏ | 5242/43939 [8:28:58<61:30:20, 5.72s/it]
12%|█▏ | 5243/43939 [8:29:04<61:25:23, 5.71s/it]
12%|█▏ | 5244/43939 [8:29:10<61:27:23, 5.72s/it]
12%|█▏ | 5245/43939 [8:29:15<61:28:18, 5.72s/it]
12%|█▏ | 5246/43939 [8:29:21<61:29:09, 5.72s/it]
12%|█▏ | 5247/43939 [8:29:27<61:19:56, 5.71s/it]
12%|█▏ | 5248/43939 [8:29:33<61:30:35, 5.72s/it]
12%|█▏ | 5249/43939 [8:29:38<61:20:31, 5.71s/it]
12%|█▏ | 5250/43939 [8:29:44<61:17:37, 5.70s/it]
{'loss': 1.0709, 'grad_norm': 1.000447392463684, 'learning_rate': 4.994221903380024e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.0, 'epoch': 0.48}
12%|█▏ | 5250/43939 [8:29:44<61:17:37, 5.70s/it]
12%|█▏ | 5251/43939 [8:29:50<61:17:16, 5.70s/it]
12%|█▏ | 5252/43939 [8:29:55<61:17:51, 5.70s/it]
12%|█▏ | 5253/43939 [8:30:01<61:17:08, 5.70s/it]
12%|█▏ | 5254/43939 [8:30:07<61:18:34, 5.71s/it]
12%|█▏ | 5255/43939 [8:30:13<61:19:56, 5.71s/it]
12%|█▏ | 5256/43939 [8:30:18<61:18:57, 5.71s/it]
12%|█▏ | 5257/43939 [8:30:24<61:16:25, 5.70s/it]
12%|█▏ | 5258/43939 [8:30:30<61:07:00, 5.69s/it]
12%|█▏ | 5259/43939 [8:30:36<61:59:15, 5.77s/it]
12%|█▏ | 5260/43939 [8:30:41<61:41:18, 5.74s/it]
12%|█▏ | 5261/43939 [8:30:47<61:38:45, 5.74s/it]
12%|█▏ | 5262/43939 [8:30:53<61:33:58, 5.73s/it]
12%|█▏ | 5263/43939 [8:30:58<61:27:54, 5.72s/it]
12%|█▏ | 5264/43939 [8:31:04<61:21:22, 5.71s/it]
12%|█▏ | 5265/43939 [8:31:10<61:20:44, 5.71s/it]
12%|█▏ | 5266/43939 [8:31:15<61:23:32, 5.71s/it]
12%|█▏ | 5267/43939 [8:31:21<61:16:33, 5.70s/it]
12%|█▏ | 5268/43939 [8:31:27<61:10:38, 5.70s/it]
12%|█▏ | 5269/43939 [8:31:33<61:11:35, 5.70s/it]
12%|█▏ | 5270/43939 [8:31:38<61:07:25, 5.69s/it]
12%|█▏ | 5271/43939 [8:31:44<61:08:03, 5.69s/it]
12%|█▏ | 5272/43939 [8:31:50<61:13:47, 5.70s/it]
12%|█▏ | 5273/43939 [8:31:55<61:12:50, 5.70s/it]
12%|█▏ | 5274/43939 [8:32:01<61:20:31, 5.71s/it]
12%|█▏ | 5275/43939 [8:32:07<61:19:55, 5.71s/it]
12%|█▏ | 5276/43939 [8:32:12<61:13:56, 5.70s/it]
12%|█▏ | 5277/43939 [8:32:18<61:20:44, 5.71s/it]
12%|█▏ | 5278/43939 [8:32:24<61:23:56, 5.72s/it]
12%|█▏ | 5279/43939 [8:32:30<61:25:56, 5.72s/it]
12%|█▏ | 5280/43939 [8:32:36<62:04:47, 5.78s/it]
12%|█▏ | 5281/43939 [8:32:41<61:48:57, 5.76s/it]
12%|█▏ | 5282/43939 [8:32:47<61:37:28, 5.74s/it]
12%|█▏ | 5283/43939 [8:32:53<61:20:39, 5.71s/it]
12%|█▏ | 5284/43939 [8:32:58<61:17:06, 5.71s/it]
12%|█▏ | 5285/43939 [8:33:04<61:32:31, 5.73s/it]
12%|█▏ | 5286/43939 [8:33:10<61:23:08, 5.72s/it]
12%|█▏ | 5287/43939 [8:33:15<61:17:25, 5.71s/it]
12%|█▏ | 5288/43939 [8:33:21<61:18:22, 5.71s/it]
12%|█▏ | 5289/43939 [8:33:27<61:10:52, 5.70s/it]
12%|█▏ | 5290/43939 [8:33:33<61:03:00, 5.69s/it]
12%|█▏ | 5291/43939 [8:33:38<61:04:55, 5.69s/it]
12%|█▏ | 5292/43939 [8:33:44<61:12:00, 5.70s/it]
12%|█▏ | 5293/43939 [8:33:50<61:10:05, 5.70s/it]
12%|█▏ | 5294/43939 [8:33:55<61:09:16, 5.70s/it]
12%|█▏ | 5295/43939 [8:34:01<61:11:08, 5.70s/it]
12%|█▏ | 5296/43939 [8:34:07<61:21:35, 5.72s/it]
12%|█▏ | 5297/43939 [8:34:13<61:22:46, 5.72s/it]
12%|█▏ | 5298/43939 [8:34:18<61:14:53, 5.71s/it]
12%|█▏ | 5299/43939 [8:34:24<61:32:41, 5.73s/it]
12%|█▏ | 5300/43939 [8:34:30<61:23:05, 5.72s/it]
{'loss': 1.0676, 'grad_norm': 1.1662852764129639, 'learning_rate': 4.993527477899977e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.04, 'epoch': 0.48}
12%|█▏ | 5300/43939 [8:34:30<61:23:05, 5.72s/it]
12%|█▏ | 5301/43939 [8:34:36<62:06:06, 5.79s/it]
12%|█▏ | 5302/43939 [8:34:41<61:54:45, 5.77s/it]
12%|█▏ | 5303/43939 [8:34:47<61:36:17, 5.74s/it]
12%|█▏ | 5304/43939 [8:34:53<61:34:52, 5.74s/it]
12%|█▏ | 5305/43939 [8:34:58<61:21:08, 5.72s/it]
12%|█▏ | 5306/43939 [8:35:04<61:17:10, 5.71s/it]
12%|█▏ | 5307/43939 [8:35:10<61:22:46, 5.72s/it]
12%|█▏ | 5308/43939 [8:35:16<61:23:53, 5.72s/it]
12%|█▏ | 5309/43939 [8:35:21<61:26:29, 5.73s/it]
12%|█▏ | 5310/43939 [8:35:27<61:19:43, 5.72s/it]
12%|█▏ | 5311/43939 [8:35:33<61:27:24, 5.73s/it]
12%|█▏ | 5312/43939 [8:35:39<61:31:53, 5.73s/it]
12%|█▏ | 5313/43939 [8:35:44<61:33:13, 5.74s/it]
12%|█▏ | 5314/43939 [8:35:50<61:23:51, 5.72s/it]
12%|█▏ | 5315/43939 [8:35:56<61:13:45, 5.71s/it]
12%|█▏ | 5316/43939 [8:36:01<61:11:49, 5.70s/it]
12%|█▏ | 5317/43939 [8:36:07<61:12:11, 5.70s/it]
12%|█▏ | 5318/43939 [8:36:13<61:12:38, 5.71s/it]
12%|█▏ | 5319/43939 [8:36:18<61:09:34, 5.70s/it]
12%|█▏ | 5320/43939 [8:36:24<61:02:45, 5.69s/it]
12%|█▏ | 5321/43939 [8:36:30<61:46:11, 5.76s/it]
12%|█▏ | 5322/43939 [8:36:36<61:32:08, 5.74s/it]
12%|█▏ | 5323/43939 [8:36:41<61:24:30, 5.72s/it]
12%|█▏ | 5324/43939 [8:36:47<61:21:52, 5.72s/it]
12%|█▏ | 5325/43939 [8:36:53<61:17:25, 5.71s/it]
12%|█▏ | 5326/43939 [8:36:58<61:07:26, 5.70s/it]
12%|█▏ | 5327/43939 [8:37:04<61:00:45, 5.69s/it]
12%|█▏ | 5328/43939 [8:37:10<60:57:31, 5.68s/it]
12%|█▏ | 5329/43939 [8:37:16<61:03:43, 5.69s/it]
12%|█▏ | 5330/43939 [8:37:21<61:03:26, 5.69s/it]
12%|█▏ | 5331/43939 [8:37:27<61:02:56, 5.69s/it]
12%|█▏ | 5332/43939 [8:37:33<61:02:11, 5.69s/it]
12%|█▏ | 5333/43939 [8:37:38<61:06:17, 5.70s/it]
12%|█▏ | 5334/43939 [8:37:44<61:08:08, 5.70s/it]
12%|█▏ | 5335/43939 [8:37:50<61:13:36, 5.71s/it]
12%|█▏ | 5336/43939 [8:37:55<61:07:37, 5.70s/it]
12%|█▏ | 5337/43939 [8:38:01<61:07:52, 5.70s/it]
12%|█▏ | 5338/43939 [8:38:07<61:06:42, 5.70s/it]
12%|█▏ | 5339/43939 [8:38:13<61:02:59, 5.69s/it]
12%|█▏ | 5340/43939 [8:38:18<61:03:10, 5.69s/it]
12%|█▏ | 5341/43939 [8:38:24<61:04:30, 5.70s/it]
12%|█▏ | 5342/43939 [8:38:30<61:01:21, 5.69s/it]
12%|█▏ | 5343/43939 [8:38:36<61:53:35, 5.77s/it]
12%|█▏ | 5344/43939 [8:38:41<61:39:36, 5.75s/it]
12%|█▏ | 5345/43939 [8:38:47<61:26:01, 5.73s/it]
12%|█▏ | 5346/43939 [8:38:53<61:17:23, 5.72s/it]
12%|█▏ | 5347/43939 [8:38:58<61:15:10, 5.71s/it]
12%|█▏ | 5348/43939 [8:39:04<61:11:15, 5.71s/it]
12%|█▏ | 5349/43939 [8:39:10<61:03:04, 5.70s/it]
12%|█▏ | 5350/43939 [8:39:15<60:59:21, 5.69s/it]
{'loss': 1.0825, 'grad_norm': 1.0618816614151, 'learning_rate': 4.992793711164256e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.62, 'epoch': 0.49}
12%|█▏ | 5350/43939 [8:39:15<60:59:21, 5.69s/it]
12%|█▏ | 5351/43939 [8:39:21<60:55:49, 5.68s/it]
12%|█▏ | 5352/43939 [8:39:27<61:09:12, 5.71s/it]
12%|█▏ | 5353/43939 [8:39:33<61:18:01, 5.72s/it]
12%|█▏ | 5354/43939 [8:39:38<61:15:58, 5.72s/it]
12%|█▏ | 5355/43939 [8:39:44<61:10:40, 5.71s/it]
12%|█▏ | 5356/43939 [8:39:50<61:11:21, 5.71s/it]
12%|█▏ | 5357/43939 [8:39:55<61:03:17, 5.70s/it]
12%|█▏ | 5358/43939 [8:40:01<61:04:08, 5.70s/it]
12%|█▏ | 5359/43939 [8:40:07<61:06:15, 5.70s/it]
12%|█▏ | 5360/43939 [8:40:12<61:09:26, 5.71s/it]
12%|█▏ | 5361/43939 [8:40:18<61:08:44, 5.71s/it]
12%|█▏ | 5362/43939 [8:40:24<61:16:44, 5.72s/it]
12%|█▏ | 5363/43939 [8:40:30<61:52:47, 5.77s/it]
12%|█▏ | 5364/43939 [8:40:35<61:34:34, 5.75s/it]
12%|█▏ | 5365/43939 [8:40:41<61:21:15, 5.73s/it]
12%|█▏ | 5366/43939 [8:40:47<61:08:52, 5.71s/it]
12%|█▏ | 5367/43939 [8:40:52<61:00:46, 5.69s/it]
12%|█▏ | 5368/43939 [8:40:58<60:59:47, 5.69s/it]
12%|█▏ | 5369/43939 [8:41:04<61:03:23, 5.70s/it]
12%|█▏ | 5370/43939 [8:41:10<60:57:59, 5.69s/it]
12%|█▏ | 5371/43939 [8:41:15<60:56:55, 5.69s/it]
12%|█▏ | 5372/43939 [8:41:21<61:02:16, 5.70s/it]
12%|█▏ | 5373/43939 [8:41:27<61:06:05, 5.70s/it]
12%|█▏ | 5374/43939 [8:41:32<61:02:32, 5.70s/it]
12%|█▏ | 5375/43939 [8:41:38<61:07:11, 5.71s/it]
12%|█▏ | 5376/43939 [8:41:44<61:00:43, 5.70s/it]
12%|█▏ | 5377/43939 [8:41:49<61:05:50, 5.70s/it]
12%|█▏ | 5378/43939 [8:41:55<61:06:45, 5.71s/it]
12%|█▏ | 5379/43939 [8:42:01<61:06:40, 5.71s/it]
12%|█▏ | 5380/43939 [8:42:07<61:10:53, 5.71s/it]
12%|█▏ | 5381/43939 [8:42:12<61:08:58, 5.71s/it]
12%|█▏ | 5382/43939 [8:42:18<61:00:36, 5.70s/it]
12%|█▏ | 5383/43939 [8:42:24<60:54:48, 5.69s/it]
12%|█▏ | 5384/43939 [8:42:30<61:44:54, 5.77s/it]
12%|█▏ | 5385/43939 [8:42:35<61:35:40, 5.75s/it]
12%|█▏ | 5386/43939 [8:42:41<61:16:20, 5.72s/it]
12%|█▏ | 5387/43939 [8:42:47<61:07:12, 5.71s/it]
12%|█▏ | 5388/43939 [8:42:52<61:07:52, 5.71s/it]
12%|█▏ | 5389/43939 [8:42:58<61:00:58, 5.70s/it]
12%|█▏ | 5390/43939 [8:43:04<60:59:58, 5.70s/it]
12%|█▏ | 5391/43939 [8:43:09<60:53:42, 5.69s/it]
12%|█▏ | 5392/43939 [8:43:15<60:58:10, 5.69s/it]
12%|█▏ | 5393/43939 [8:43:21<60:57:32, 5.69s/it]
12%|█▏ | 5394/43939 [8:43:27<61:02:16, 5.70s/it]
12%|█▏ | 5395/43939 [8:43:32<60:54:44, 5.69s/it]
12%|█▏ | 5396/43939 [8:43:38<60:51:07, 5.68s/it]
12%|█▏ | 5397/43939 [8:43:44<60:54:17, 5.69s/it]
12%|█▏ | 5398/43939 [8:43:49<60:54:47, 5.69s/it]
12%|█▏ | 5399/43939 [8:43:55<60:49:59, 5.68s/it]
12%|█▏ | 5400/43939 [8:44:01<60:52:22, 5.69s/it]
{'loss': 1.0664, 'grad_norm': 1.1399046182632446, 'learning_rate': 4.992020614749758e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.69, 'epoch': 0.49}
12%|█▏ | 5400/43939 [8:44:01<60:52:22, 5.69s/it]
12%|█▏ | 5401/43939 [8:44:06<61:01:03, 5.70s/it]
12%|█▏ | 5402/43939 [8:44:12<61:02:41, 5.70s/it]
12%|█▏ | 5403/43939 [8:44:18<61:04:25, 5.71s/it]
12%|█▏ | 5404/43939 [8:44:23<61:00:15, 5.70s/it]
12%|█▏ | 5405/43939 [8:44:29<61:45:11, 5.77s/it]
12%|█▏ | 5406/43939 [8:44:35<61:24:50, 5.74s/it]
12%|█▏ | 5407/43939 [8:44:41<61:17:05, 5.73s/it]
12%|█▏ | 5408/43939 [8:44:46<61:14:41, 5.72s/it]
12%|█▏ | 5409/43939 [8:44:52<61:12:56, 5.72s/it]
12%|█▏ | 5410/43939 [8:44:58<61:07:12, 5.71s/it]
12%|█▏ | 5411/43939 [8:45:04<61:05:41, 5.71s/it]
12%|█▏ | 5412/43939 [8:45:09<61:03:42, 5.71s/it]
12%|█▏ | 5413/43939 [8:45:15<60:58:35, 5.70s/it]
12%|█▏ | 5414/43939 [8:45:21<61:00:20, 5.70s/it]
12%|█▏ | 5415/43939 [8:45:26<61:04:04, 5.71s/it]
12%|█▏ | 5416/43939 [8:45:32<61:05:03, 5.71s/it]
12%|█▏ | 5417/43939 [8:45:38<61:02:22, 5.70s/it]
12%|█▏ | 5418/43939 [8:45:43<61:02:38, 5.70s/it]
12%|█▏ | 5419/43939 [8:45:49<61:04:00, 5.71s/it]
12%|█▏ | 5420/43939 [8:45:55<61:00:58, 5.70s/it]
12%|█▏ | 5421/43939 [8:46:01<60:58:06, 5.70s/it]
12%|█▏ | 5422/43939 [8:46:06<60:47:44, 5.68s/it]
12%|█▏ | 5423/43939 [8:46:12<60:48:27, 5.68s/it]
12%|█▏ | 5424/43939 [8:46:18<60:47:49, 5.68s/it]
12%|█▏ | 5425/43939 [8:46:23<60:56:14, 5.70s/it]
12%|█▏ | 5426/43939 [8:46:29<61:39:23, 5.76s/it]
12%|█▏ | 5427/43939 [8:46:35<61:19:26, 5.73s/it]
12%|█▏ | 5428/43939 [8:46:41<61:18:17, 5.73s/it]
12%|█▏ | 5429/43939 [8:46:46<61:13:32, 5.72s/it]
12%|█▏ | 5430/43939 [8:46:52<61:16:11, 5.73s/it]
12%|█▏ | 5431/43939 [8:46:58<61:04:11, 5.71s/it]
12%|█▏ | 5432/43939 [8:47:03<61:08:31, 5.72s/it]
12%|█▏ | 5433/43939 [8:47:09<61:15:29, 5.73s/it]
12%|█▏ | 5434/43939 [8:47:15<61:00:52, 5.70s/it]
12%|█▏ | 5435/43939 [8:47:21<60:52:55, 5.69s/it]
12%|█▏ | 5436/43939 [8:47:26<60:55:38, 5.70s/it]
12%|█▏ | 5437/43939 [8:47:32<60:54:53, 5.70s/it]
12%|█▏ | 5438/43939 [8:47:38<60:47:03, 5.68s/it]
12%|█▏ | 5439/43939 [8:47:43<60:51:35, 5.69s/it]
12%|█▏ | 5440/43939 [8:47:49<60:56:54, 5.70s/it]
12%|█▏ | 5441/43939 [8:47:55<60:59:48, 5.70s/it]
12%|█▏ | 5442/43939 [8:48:00<60:57:09, 5.70s/it]
12%|█▏ | 5443/43939 [8:48:06<60:56:06, 5.70s/it]
12%|█▏ | 5444/43939 [8:48:12<60:59:07, 5.70s/it]
12%|█▏ | 5445/43939 [8:48:18<60:57:05, 5.70s/it]
12%|█▏ | 5446/43939 [8:48:23<61:05:28, 5.71s/it]
12%|█▏ | 5447/43939 [8:48:29<61:49:09, 5.78s/it]
12%|█▏ | 5448/43939 [8:48:35<61:29:53, 5.75s/it]
12%|█▏ | 5449/43939 [8:48:41<61:28:46, 5.75s/it]
12%|█▏ | 5450/43939 [8:48:46<61:24:00, 5.74s/it]
{'loss': 1.0657, 'grad_norm': 1.2065247297286987, 'learning_rate': 4.991208200853894e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.34, 'epoch': 0.5}
12%|█▏ | 5450/43939 [8:48:46<61:24:00, 5.74s/it]
12%|█▏ | 5451/43939 [8:48:52<61:25:17, 5.75s/it]
12%|█▏ | 5452/43939 [8:48:58<61:22:20, 5.74s/it]
12%|█▏ | 5453/43939 [8:49:04<61:18:29, 5.73s/it]
12%|█▏ | 5454/43939 [8:49:09<61:08:39, 5.72s/it]
12%|█▏ | 5455/43939 [8:49:15<61:03:19, 5.71s/it]
12%|█▏ | 5456/43939 [8:49:21<60:57:40, 5.70s/it]
12%|█▏ | 5457/43939 [8:49:26<60:54:02, 5.70s/it]
12%|█▏ | 5458/43939 [8:49:32<60:54:12, 5.70s/it]
12%|█▏ | 5459/43939 [8:49:38<60:53:31, 5.70s/it]
12%|█▏ | 5460/43939 [8:49:43<60:55:11, 5.70s/it]
12%|█▏ | 5461/43939 [8:49:49<60:49:20, 5.69s/it]
12%|█▏ | 5462/43939 [8:49:55<60:44:48, 5.68s/it]
12%|█▏ | 5463/43939 [8:50:00<60:48:50, 5.69s/it]
12%|█▏ | 5464/43939 [8:50:06<60:41:18, 5.68s/it]
12%|█▏ | 5465/43939 [8:50:12<60:44:20, 5.68s/it]
12%|█▏ | 5466/43939 [8:50:17<60:48:42, 5.69s/it]
12%|█▏ | 5467/43939 [8:50:23<60:47:56, 5.69s/it]
12%|█▏ | 5468/43939 [8:50:29<60:55:21, 5.70s/it]
12%|█▏ | 5469/43939 [8:50:35<61:39:08, 5.77s/it]
12%|█▏ | 5470/43939 [8:50:41<61:23:56, 5.75s/it]
12%|█▏ | 5471/43939 [8:50:46<61:14:45, 5.73s/it]
12%|█▏ | 5472/43939 [8:50:52<61:16:52, 5.74s/it]
12%|█▏ | 5473/43939 [8:50:58<61:15:50, 5.73s/it]
12%|█▏ | 5474/43939 [8:51:03<61:10:29, 5.73s/it]
12%|█▏ | 5475/43939 [8:51:09<61:19:40, 5.74s/it]
12%|█▏ | 5476/43939 [8:51:15<61:10:44, 5.73s/it]
12%|█▏ | 5477/43939 [8:51:21<61:02:48, 5.71s/it]
12%|█▏ | 5478/43939 [8:51:26<61:04:23, 5.72s/it]
12%|█▏ | 5479/43939 [8:51:32<60:55:39, 5.70s/it]
12%|█▏ | 5480/43939 [8:51:38<60:47:23, 5.69s/it]
12%|█▏ | 5481/43939 [8:51:43<60:56:00, 5.70s/it]
12%|█▏ | 5482/43939 [8:51:49<60:59:10, 5.71s/it]
12%|█▏ | 5483/43939 [8:51:55<60:56:12, 5.70s/it]
12%|█▏ | 5484/43939 [8:52:00<60:53:12, 5.70s/it]
12%|█▏ | 5485/43939 [8:52:06<60:49:45, 5.69s/it]
12%|█▏ | 5486/43939 [8:52:12<60:50:03, 5.70s/it]
12%|█▏ | 5487/43939 [8:52:17<60:44:01, 5.69s/it]
12%|█▏ | 5488/43939 [8:52:23<60:51:59, 5.70s/it]
12%|█▏ | 5489/43939 [8:52:29<60:52:07, 5.70s/it]
12%|█▏ | 5490/43939 [8:52:35<61:41:18, 5.78s/it]
12%|█▏ | 5491/43939 [8:52:41<61:29:39, 5.76s/it]
12%|█▏ | 5492/43939 [8:52:46<61:22:25, 5.75s/it]
13%|█▎ | 5493/43939 [8:52:52<61:11:12, 5.73s/it][2025-11-29 02:57:05,827] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-5493
13%|█▎ | 5494/43939 [8:53:00<67:16:35, 6.30s/it]
13%|█▎ | 5495/43939 [8:53:05<65:23:35, 6.12s/it]
13%|█▎ | 5496/43939 [8:53:11<63:51:58, 5.98s/it]
13%|█▎ | 5497/43939 [8:53:17<62:51:35, 5.89s/it]
13%|█▎ | 5498/43939 [8:53:22<62:16:35, 5.83s/it]
13%|█▎ | 5499/43939 [8:53:28<61:50:55, 5.79s/it]
13%|█▎ | 5500/43939 [8:53:34<61:39:56, 5.78s/it]
{'loss': 1.0677, 'grad_norm': 1.0179318189620972, 'learning_rate': 4.9903564822944024e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.23, 'epoch': 0.5}
13%|█▎ | 5500/43939 [8:53:34<61:39:56, 5.78s/it]
13%|█▎ | 5501/43939 [8:53:39<61:22:55, 5.75s/it]
13%|█▎ | 5502/43939 [8:53:45<61:11:27, 5.73s/it]
13%|█▎ | 5503/43939 [8:53:51<61:04:26, 5.72s/it]
13%|█▎ | 5504/43939 [8:53:57<61:00:20, 5.71s/it]
13%|█▎ | 5505/43939 [8:54:02<60:52:55, 5.70s/it]
13%|█▎ | 5506/43939 [8:54:08<60:49:48, 5.70s/it]
13%|█▎ | 5507/43939 [8:54:14<60:51:51, 5.70s/it]
13%|█▎ | 5508/43939 [8:54:19<60:47:12, 5.69s/it]
13%|█▎ | 5509/43939 [8:54:25<60:55:09, 5.71s/it]
13%|█▎ | 5510/43939 [8:54:31<61:48:55, 5.79s/it]
13%|█▎ | 5511/43939 [8:54:37<61:23:52, 5.75s/it]
13%|█▎ | 5512/43939 [8:54:42<61:11:00, 5.73s/it]
13%|█▎ | 5513/43939 [8:54:48<61:08:27, 5.73s/it]
13%|█▎ | 5514/43939 [8:54:54<61:05:39, 5.72s/it]
13%|█▎ | 5515/43939 [8:55:00<61:02:23, 5.72s/it]
13%|█▎ | 5516/43939 [8:55:05<60:53:28, 5.71s/it]
13%|█▎ | 5517/43939 [8:55:11<60:54:06, 5.71s/it]
13%|█▎ | 5518/43939 [8:55:17<60:53:54, 5.71s/it]
13%|█▎ | 5519/43939 [8:55:22<60:54:49, 5.71s/it]
13%|█▎ | 5520/43939 [8:55:28<60:50:29, 5.70s/it]
13%|█▎ | 5521/43939 [8:55:34<60:41:46, 5.69s/it]
13%|█▎ | 5522/43939 [8:55:39<60:45:27, 5.69s/it]
13%|█▎ | 5523/43939 [8:55:45<61:01:15, 5.72s/it]
13%|█▎ | 5524/43939 [8:55:51<61:05:22, 5.72s/it]
13%|█▎ | 5525/43939 [8:55:57<60:58:51, 5.71s/it]
13%|█▎ | 5526/43939 [8:56:02<60:58:02, 5.71s/it]
13%|█▎ | 5527/43939 [8:56:08<60:56:29, 5.71s/it]
13%|█▎ | 5528/43939 [8:56:14<61:01:28, 5.72s/it]
13%|█▎ | 5529/43939 [8:56:19<60:59:51, 5.72s/it]
13%|█▎ | 5530/43939 [8:56:25<60:54:04, 5.71s/it]
13%|█▎ | 5531/43939 [8:56:31<61:47:53, 5.79s/it]
13%|█▎ | 5532/43939 [8:56:37<61:27:14, 5.76s/it]
13%|█▎ | 5533/43939 [8:56:43<61:15:12, 5.74s/it]
13%|█▎ | 5534/43939 [8:56:48<61:06:24, 5.73s/it]
13%|█▎ | 5535/43939 [8:56:54<60:59:38, 5.72s/it]
13%|█▎ | 5536/43939 [8:57:00<60:58:37, 5.72s/it]
13%|█▎ | 5537/43939 [8:57:05<60:53:07, 5.71s/it]
13%|█▎ | 5538/43939 [8:57:11<60:41:02, 5.69s/it]
13%|█▎ | 5539/43939 [8:57:17<60:47:35, 5.70s/it]
13%|█▎ | 5540/43939 [8:57:22<60:45:31, 5.70s/it]
13%|█▎ | 5541/43939 [8:57:28<60:47:07, 5.70s/it]
13%|█▎ | 5542/43939 [8:57:34<60:51:49, 5.71s/it]
13%|█▎ | 5543/43939 [8:57:40<60:54:32, 5.71s/it]
13%|█▎ | 5544/43939 [8:57:45<60:52:32, 5.71s/it]
13%|█▎ | 5545/43939 [8:57:51<60:56:56, 5.71s/it]
13%|█▎ | 5546/43939 [8:57:57<60:47:55, 5.70s/it]
13%|█▎ | 5547/43939 [8:58:02<60:46:23, 5.70s/it]
13%|█▎ | 5548/43939 [8:58:08<60:48:01, 5.70s/it]
13%|█▎ | 5549/43939 [8:58:14<60:43:34, 5.69s/it]
13%|█▎ | 5550/43939 [8:58:19<60:31:44, 5.68s/it]
{'loss': 1.0534, 'grad_norm': 1.192871332168579, 'learning_rate': 4.989465472509145e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.98, 'epoch': 0.51}
13%|█▎ | 5550/43939 [8:58:19<60:31:44, 5.68s/it]
13%|█▎ | 5551/43939 [8:58:25<61:14:34, 5.74s/it]
13%|█▎ | 5552/43939 [8:58:31<61:07:41, 5.73s/it]
13%|█▎ | 5553/43939 [8:58:37<60:58:49, 5.72s/it]
13%|█▎ | 5554/43939 [8:58:42<60:52:34, 5.71s/it]
13%|█▎ | 5555/43939 [8:58:48<60:49:12, 5.70s/it]
13%|█▎ | 5556/43939 [8:58:54<60:43:46, 5.70s/it]
13%|█▎ | 5557/43939 [8:58:59<60:50:03, 5.71s/it]
13%|█▎ | 5558/43939 [8:59:05<60:41:24, 5.69s/it]
13%|█▎ | 5559/43939 [8:59:11<60:36:50, 5.69s/it]
13%|█▎ | 5560/43939 [8:59:16<60:38:00, 5.69s/it]
13%|█▎ | 5561/43939 [8:59:22<60:36:20, 5.69s/it]
13%|█▎ | 5562/43939 [8:59:28<60:36:12, 5.68s/it]
13%|█▎ | 5563/43939 [8:59:33<60:35:53, 5.68s/it]
13%|█▎ | 5564/43939 [8:59:39<60:28:25, 5.67s/it]
13%|█▎ | 5565/43939 [8:59:45<60:35:44, 5.68s/it]
13%|█▎ | 5566/43939 [8:59:50<60:29:35, 5.68s/it]
13%|█▎ | 5567/43939 [8:59:56<60:42:28, 5.70s/it]
13%|█▎ | 5568/43939 [9:00:02<60:39:26, 5.69s/it]
13%|█▎ | 5569/43939 [9:00:08<60:38:43, 5.69s/it]
13%|█▎ | 5570/43939 [9:00:13<60:40:13, 5.69s/it]
13%|█▎ | 5571/43939 [9:00:19<60:39:27, 5.69s/it]
13%|█▎ | 5572/43939 [9:00:25<60:38:20, 5.69s/it]
13%|█▎ | 5573/43939 [9:00:31<61:27:10, 5.77s/it]
13%|█▎ | 5574/43939 [9:00:36<61:10:30, 5.74s/it]
13%|█▎ | 5575/43939 [9:00:42<61:00:28, 5.72s/it]
13%|█▎ | 5576/43939 [9:00:48<61:02:41, 5.73s/it]
13%|█▎ | 5577/43939 [9:00:53<60:59:27, 5.72s/it]
13%|█▎ | 5578/43939 [9:00:59<60:59:40, 5.72s/it]
13%|█▎ | 5579/43939 [9:01:05<60:57:56, 5.72s/it]
13%|█▎ | 5580/43939 [9:01:11<60:45:21, 5.70s/it]
13%|█▎ | 5581/43939 [9:01:16<60:54:44, 5.72s/it]
13%|█▎ | 5582/43939 [9:01:22<60:47:56, 5.71s/it]
13%|█▎ | 5583/43939 [9:01:28<60:49:39, 5.71s/it]
13%|█▎ | 5584/43939 [9:01:33<60:47:21, 5.71s/it]
13%|█▎ | 5585/43939 [9:01:39<60:46:51, 5.71s/it]
13%|█▎ | 5586/43939 [9:01:45<60:45:17, 5.70s/it]
13%|█▎ | 5587/43939 [9:01:50<60:38:13, 5.69s/it]
13%|█▎ | 5588/43939 [9:01:56<60:29:34, 5.68s/it]
13%|█▎ | 5589/43939 [9:02:02<60:41:14, 5.70s/it]
13%|█▎ | 5590/43939 [9:02:08<60:40:23, 5.70s/it]
13%|█▎ | 5591/43939 [9:02:13<60:40:43, 5.70s/it]
13%|█▎ | 5592/43939 [9:02:19<60:49:19, 5.71s/it]
13%|█▎ | 5593/43939 [9:02:25<61:25:12, 5.77s/it]
13%|█▎ | 5594/43939 [9:02:31<61:05:23, 5.74s/it]
13%|█▎ | 5595/43939 [9:02:36<61:04:24, 5.73s/it]
13%|█▎ | 5596/43939 [9:02:42<60:57:22, 5.72s/it]
13%|█▎ | 5597/43939 [9:02:48<60:49:33, 5.71s/it]
13%|█▎ | 5598/43939 [9:02:53<60:46:26, 5.71s/it]
13%|█▎ | 5599/43939 [9:02:59<60:38:19, 5.69s/it]
13%|█▎ | 5600/43939 [9:03:05<60:40:16, 5.70s/it]
{'loss': 1.0625, 'grad_norm': 1.072653889656067, 'learning_rate': 4.988535185555896e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.49, 'epoch': 0.51}
13%|█▎ | 5600/43939 [9:03:05<60:40:16, 5.70s/it]
13%|█▎ | 5601/43939 [9:03:10<60:41:12, 5.70s/it]
13%|█▎ | 5602/43939 [9:03:16<60:47:14, 5.71s/it]
13%|█▎ | 5603/43939 [9:03:22<60:47:26, 5.71s/it]
13%|█▎ | 5604/43939 [9:03:28<60:36:44, 5.69s/it]
13%|█▎ | 5605/43939 [9:03:33<60:34:30, 5.69s/it]
13%|█▎ | 5606/43939 [9:03:39<60:32:32, 5.69s/it]
13%|█▎ | 5607/43939 [9:03:45<60:36:38, 5.69s/it]
13%|█▎ | 5608/43939 [9:03:50<60:32:10, 5.69s/it]
13%|█▎ | 5609/43939 [9:03:56<60:28:12, 5.68s/it]
13%|█▎ | 5610/43939 [9:04:02<60:32:33, 5.69s/it]
13%|█▎ | 5611/43939 [9:04:07<60:35:34, 5.69s/it]
13%|█▎ | 5612/43939 [9:04:13<60:37:20, 5.69s/it]
13%|█▎ | 5613/43939 [9:04:19<60:44:49, 5.71s/it]
13%|█▎ | 5614/43939 [9:04:25<61:19:30, 5.76s/it]
13%|█▎ | 5615/43939 [9:04:30<61:16:15, 5.76s/it]
13%|█▎ | 5616/43939 [9:04:36<61:04:42, 5.74s/it]
13%|█▎ | 5617/43939 [9:04:42<60:51:46, 5.72s/it]
13%|█▎ | 5618/43939 [9:04:47<60:49:54, 5.71s/it]
13%|█▎ | 5619/43939 [9:04:53<60:50:15, 5.72s/it]
13%|█▎ | 5620/43939 [9:04:59<60:41:22, 5.70s/it]
13%|█▎ | 5621/43939 [9:05:05<60:38:43, 5.70s/it]
13%|█▎ | 5622/43939 [9:05:10<60:40:15, 5.70s/it]
13%|█▎ | 5623/43939 [9:05:16<60:42:55, 5.70s/it]
13%|█▎ | 5624/43939 [9:05:22<60:39:23, 5.70s/it]
13%|█▎ | 5625/43939 [9:05:27<60:30:39, 5.69s/it]
13%|█▎ | 5626/43939 [9:05:33<60:27:15, 5.68s/it]
13%|█▎ | 5627/43939 [9:05:39<60:36:50, 5.70s/it]
13%|█▎ | 5628/43939 [9:05:44<60:37:32, 5.70s/it]
13%|█▎ | 5629/43939 [9:05:50<60:40:28, 5.70s/it]
13%|█▎ | 5630/43939 [9:05:56<60:42:15, 5.70s/it]
13%|█▎ | 5631/43939 [9:06:02<60:43:17, 5.71s/it]
13%|█▎ | 5632/43939 [9:06:07<60:56:18, 5.73s/it]
13%|█▎ | 5633/43939 [9:06:13<60:51:45, 5.72s/it]
13%|█▎ | 5634/43939 [9:06:19<60:53:12, 5.72s/it]
13%|█▎ | 5635/43939 [9:06:24<60:49:37, 5.72s/it]
13%|█▎ | 5636/43939 [9:06:30<61:23:39, 5.77s/it]
13%|█▎ | 5637/43939 [9:06:36<61:08:07, 5.75s/it]
13%|█▎ | 5638/43939 [9:06:42<61:00:35, 5.73s/it]
13%|█▎ | 5639/43939 [9:06:47<60:55:47, 5.73s/it]
13%|█▎ | 5640/43939 [9:06:53<60:46:51, 5.71s/it]
13%|█▎ | 5641/43939 [9:06:59<60:51:04, 5.72s/it]
13%|█▎ | 5642/43939 [9:07:05<60:46:54, 5.71s/it]
13%|█▎ | 5643/43939 [9:07:10<60:38:34, 5.70s/it]
13%|█▎ | 5644/43939 [9:07:16<60:41:40, 5.71s/it]
13%|█▎ | 5645/43939 [9:07:22<60:34:28, 5.69s/it]
13%|█▎ | 5646/43939 [9:07:27<60:54:49, 5.73s/it]
13%|█▎ | 5647/43939 [9:07:33<60:52:03, 5.72s/it]
13%|█▎ | 5648/43939 [9:07:39<60:48:41, 5.72s/it]
13%|█▎ | 5649/43939 [9:07:45<60:52:16, 5.72s/it]
13%|█▎ | 5650/43939 [9:07:50<61:01:11, 5.74s/it]
{'loss': 1.0596, 'grad_norm': 1.1457641124725342, 'learning_rate': 4.987565636112117e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.03, 'epoch': 0.51}
13%|█▎ | 5650/43939 [9:07:50<61:01:11, 5.74s/it]
13%|█▎ | 5651/43939 [9:07:56<60:59:01, 5.73s/it]
13%|█▎ | 5652/43939 [9:08:02<61:01:02, 5.74s/it]
13%|█▎ | 5653/43939 [9:08:07<60:51:00, 5.72s/it]
13%|█▎ | 5654/43939 [9:08:13<60:52:37, 5.72s/it]
13%|█▎ | 5655/43939 [9:08:19<60:46:55, 5.72s/it]
13%|█▎ | 5656/43939 [9:08:25<60:53:23, 5.73s/it]
13%|█▎ | 5657/43939 [9:08:31<61:17:35, 5.76s/it]
13%|█▎ | 5658/43939 [9:08:36<60:58:31, 5.73s/it]
13%|█▎ | 5659/43939 [9:08:42<60:51:16, 5.72s/it]
13%|█▎ | 5660/43939 [9:08:48<60:45:38, 5.71s/it]
13%|█▎ | 5661/43939 [9:08:53<60:42:03, 5.71s/it]
13%|█▎ | 5662/43939 [9:08:59<60:41:27, 5.71s/it]
13%|█▎ | 5663/43939 [9:09:05<60:43:36, 5.71s/it]
13%|█▎ | 5664/43939 [9:09:10<60:41:07, 5.71s/it]
13%|█▎ | 5665/43939 [9:09:16<60:42:04, 5.71s/it]
13%|█▎ | 5666/43939 [9:09:22<60:47:33, 5.72s/it]
13%|█▎ | 5667/43939 [9:09:28<60:43:23, 5.71s/it]
13%|█▎ | 5668/43939 [9:09:33<60:42:12, 5.71s/it]
13%|█▎ | 5669/43939 [9:09:39<60:41:07, 5.71s/it]
13%|█▎ | 5670/43939 [9:09:45<60:30:01, 5.69s/it]
13%|█▎ | 5671/43939 [9:09:50<60:26:28, 5.69s/it]
13%|█▎ | 5672/43939 [9:09:56<60:22:45, 5.68s/it]
13%|█▎ | 5673/43939 [9:10:02<60:23:57, 5.68s/it]
13%|█▎ | 5674/43939 [9:10:07<60:29:16, 5.69s/it]
13%|█▎ | 5675/43939 [9:10:13<60:32:11, 5.70s/it]
13%|█▎ | 5676/43939 [9:10:19<60:32:53, 5.70s/it]
13%|█▎ | 5677/43939 [9:10:24<60:33:06, 5.70s/it]
13%|█▎ | 5678/43939 [9:10:30<61:12:47, 5.76s/it]
13%|█▎ | 5679/43939 [9:10:36<60:55:12, 5.73s/it]
13%|█▎ | 5680/43939 [9:10:42<60:43:55, 5.71s/it]
13%|█▎ | 5681/43939 [9:10:47<60:39:37, 5.71s/it]
13%|█▎ | 5682/43939 [9:10:53<60:31:40, 5.70s/it]
13%|█▎ | 5683/43939 [9:10:59<60:22:37, 5.68s/it]
13%|█▎ | 5684/43939 [9:11:04<60:22:21, 5.68s/it]
13%|█▎ | 5685/43939 [9:11:10<60:24:15, 5.68s/it]
13%|█▎ | 5686/43939 [9:11:16<60:33:37, 5.70s/it]
13%|█▎ | 5687/43939 [9:11:22<60:35:47, 5.70s/it]
13%|█▎ | 5688/43939 [9:11:27<60:31:49, 5.70s/it]
13%|█▎ | 5689/43939 [9:11:33<60:29:25, 5.69s/it]
13%|█▎ | 5690/43939 [9:11:39<60:27:41, 5.69s/it]
13%|█▎ | 5691/43939 [9:11:44<60:33:24, 5.70s/it]
13%|█▎ | 5692/43939 [9:11:50<60:44:00, 5.72s/it]
13%|█▎ | 5693/43939 [9:11:56<60:36:55, 5.71s/it]
13%|█▎ | 5694/43939 [9:12:01<60:37:08, 5.71s/it]
13%|█▎ | 5695/43939 [9:12:07<60:32:36, 5.70s/it]
13%|█▎ | 5696/43939 [9:12:13<60:23:39, 5.69s/it]
13%|█▎ | 5697/43939 [9:12:18<60:29:06, 5.69s/it]
13%|█▎ | 5698/43939 [9:12:24<60:32:10, 5.70s/it]
13%|█▎ | 5699/43939 [9:12:30<61:18:01, 5.77s/it]
13%|█▎ | 5700/43939 [9:12:36<61:05:41, 5.75s/it]
{'loss': 1.0502, 'grad_norm': 0.9638017416000366, 'learning_rate': 4.986556839474729e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.55, 'epoch': 0.52}
13%|█▎ | 5700/43939 [9:12:36<61:05:41, 5.75s/it]
13%|█▎ | 5701/43939 [9:12:42<61:06:51, 5.75s/it]
13%|█▎ | 5702/43939 [9:12:47<60:47:01, 5.72s/it]
13%|█▎ | 5703/43939 [9:12:53<60:49:03, 5.73s/it]
13%|█▎ | 5704/43939 [9:12:59<60:46:36, 5.72s/it]
13%|█▎ | 5705/43939 [9:13:04<60:45:23, 5.72s/it]
13%|█▎ | 5706/43939 [9:13:10<60:33:01, 5.70s/it]
13%|█▎ | 5707/43939 [9:13:16<60:39:08, 5.71s/it]
13%|█▎ | 5708/43939 [9:13:21<60:35:47, 5.71s/it]
13%|█▎ | 5709/43939 [9:13:27<60:29:08, 5.70s/it]
13%|█▎ | 5710/43939 [9:13:33<60:31:45, 5.70s/it]
13%|█▎ | 5711/43939 [9:13:39<60:34:32, 5.70s/it]
13%|█▎ | 5712/43939 [9:13:44<60:30:44, 5.70s/it]
13%|█▎ | 5713/43939 [9:13:50<60:21:41, 5.68s/it]
13%|█▎ | 5714/43939 [9:13:56<60:26:32, 5.69s/it]
13%|█▎ | 5715/43939 [9:14:01<60:28:12, 5.70s/it]
13%|█▎ | 5716/43939 [9:14:07<60:39:13, 5.71s/it]
13%|█▎ | 5717/43939 [9:14:13<60:33:52, 5.70s/it]
13%|█▎ | 5718/43939 [9:14:18<60:22:23, 5.69s/it]
13%|█▎ | 5719/43939 [9:14:24<60:23:30, 5.69s/it]
13%|█▎ | 5720/43939 [9:14:30<61:10:55, 5.76s/it]
13%|█▎ | 5721/43939 [9:14:36<60:48:43, 5.73s/it]
13%|█▎ | 5722/43939 [9:14:41<60:40:30, 5.72s/it]
13%|█▎ | 5723/43939 [9:14:47<60:33:52, 5.71s/it]
13%|█▎ | 5724/43939 [9:14:53<60:25:24, 5.69s/it]
13%|█▎ | 5725/43939 [9:14:58<60:28:09, 5.70s/it]
13%|█▎ | 5726/43939 [9:15:04<60:25:32, 5.69s/it]
13%|█▎ | 5727/43939 [9:15:10<60:23:45, 5.69s/it]
13%|█▎ | 5728/43939 [9:15:15<60:19:42, 5.68s/it]
13%|█▎ | 5729/43939 [9:15:21<60:20:15, 5.68s/it]
13%|█▎ | 5730/43939 [9:15:27<60:25:25, 5.69s/it]
13%|█▎ | 5731/43939 [9:15:33<60:26:00, 5.69s/it]
13%|█▎ | 5732/43939 [9:15:38<60:25:29, 5.69s/it]
13%|█▎ | 5733/43939 [9:15:44<60:15:19, 5.68s/it]
13%|█▎ | 5734/43939 [9:15:50<60:26:56, 5.70s/it]
13%|█▎ | 5735/43939 [9:15:55<60:25:43, 5.69s/it]
13%|█▎ | 5736/43939 [9:16:01<60:25:30, 5.69s/it]
13%|█▎ | 5737/43939 [9:16:07<60:25:03, 5.69s/it]
13%|█▎ | 5738/43939 [9:16:12<60:29:07, 5.70s/it]
13%|█▎ | 5739/43939 [9:16:18<60:22:54, 5.69s/it]
13%|█▎ | 5740/43939 [9:16:24<60:12:29, 5.67s/it]
13%|█▎ | 5741/43939 [9:16:30<61:06:04, 5.76s/it]
13%|█▎ | 5742/43939 [9:16:35<61:01:35, 5.75s/it]
13%|█▎ | 5743/43939 [9:16:41<60:45:48, 5.73s/it]
13%|█▎ | 5744/43939 [9:16:47<60:42:59, 5.72s/it]
13%|█▎ | 5745/43939 [9:16:52<60:33:15, 5.71s/it]
13%|█▎ | 5746/43939 [9:16:58<60:26:39, 5.70s/it]
13%|█▎ | 5747/43939 [9:17:04<60:22:55, 5.69s/it]
13%|█▎ | 5748/43939 [9:17:10<60:26:48, 5.70s/it]
13%|█▎ | 5749/43939 [9:17:15<60:24:46, 5.69s/it]
13%|█▎ | 5750/43939 [9:17:21<60:23:55, 5.69s/it]
{'loss': 1.0695, 'grad_norm': 1.1078736782073975, 'learning_rate': 4.98550881155987e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.32, 'epoch': 0.52}
13%|█▎ | 5750/43939 [9:17:21<60:23:55, 5.69s/it]
13%|█▎ | 5751/43939 [9:17:27<60:18:13, 5.68s/it]
13%|█▎ | 5752/43939 [9:17:32<60:26:01, 5.70s/it]
13%|█▎ | 5753/43939 [9:17:38<60:27:06, 5.70s/it]
13%|█▎ | 5754/43939 [9:17:44<60:20:13, 5.69s/it]
13%|█▎ | 5755/43939 [9:17:49<60:19:22, 5.69s/it]
13%|█▎ | 5756/43939 [9:17:55<60:24:40, 5.70s/it]
13%|█▎ | 5757/43939 [9:18:01<60:24:51, 5.70s/it]
13%|█▎ | 5758/43939 [9:18:06<60:25:05, 5.70s/it]
13%|█▎ | 5759/43939 [9:18:12<60:22:26, 5.69s/it]
13%|█▎ | 5760/43939 [9:18:18<60:17:40, 5.69s/it]
13%|█▎ | 5761/43939 [9:18:24<60:58:35, 5.75s/it]
13%|█▎ | 5762/43939 [9:18:29<60:52:09, 5.74s/it]
13%|█▎ | 5763/43939 [9:18:35<60:48:09, 5.73s/it]
13%|█▎ | 5764/43939 [9:18:41<60:41:36, 5.72s/it]
13%|█▎ | 5765/43939 [9:18:47<60:40:13, 5.72s/it]
13%|█▎ | 5766/43939 [9:18:52<60:35:14, 5.71s/it]
13%|█▎ | 5767/43939 [9:18:58<60:22:16, 5.69s/it]
13%|█▎ | 5768/43939 [9:19:04<60:17:29, 5.69s/it]
13%|█▎ | 5769/43939 [9:19:09<60:16:03, 5.68s/it]
13%|█▎ | 5770/43939 [9:19:15<60:22:13, 5.69s/it]
13%|█▎ | 5771/43939 [9:19:21<60:21:56, 5.69s/it]
13%|█▎ | 5772/43939 [9:19:26<60:18:59, 5.69s/it]
13%|█▎ | 5773/43939 [9:19:32<60:13:22, 5.68s/it]
13%|█▎ | 5774/43939 [9:19:38<60:11:25, 5.68s/it]
13%|█▎ | 5775/43939 [9:19:43<60:16:51, 5.69s/it]
13%|█▎ | 5776/43939 [9:19:49<60:25:09, 5.70s/it]
13%|█▎ | 5777/43939 [9:19:55<60:28:36, 5.71s/it]
13%|█▎ | 5778/43939 [9:20:01<60:28:47, 5.71s/it]
13%|█▎ | 5779/43939 [9:20:06<60:34:25, 5.71s/it]
13%|█▎ | 5780/43939 [9:20:12<60:26:30, 5.70s/it]
13%|█▎ | 5781/43939 [9:20:18<60:19:48, 5.69s/it]
13%|█▎ | 5782/43939 [9:20:23<60:28:20, 5.71s/it]
13%|█▎ | 5783/43939 [9:20:29<61:13:49, 5.78s/it]
13%|█▎ | 5784/43939 [9:20:35<61:06:07, 5.77s/it]
13%|█▎ | 5785/43939 [9:20:41<60:50:08, 5.74s/it]
13%|█▎ | 5786/43939 [9:20:46<60:39:18, 5.72s/it]
13%|█▎ | 5787/43939 [9:20:52<60:29:52, 5.71s/it]
13%|█▎ | 5788/43939 [9:20:58<60:27:20, 5.70s/it]
13%|█▎ | 5789/43939 [9:21:04<60:38:16, 5.72s/it]
13%|█▎ | 5790/43939 [9:21:09<60:30:07, 5.71s/it]
13%|█▎ | 5791/43939 [9:21:15<60:32:28, 5.71s/it]
13%|█▎ | 5792/43939 [9:21:21<60:33:00, 5.71s/it]
13%|█▎ | 5793/43939 [9:21:26<60:24:42, 5.70s/it]
13%|█▎ | 5794/43939 [9:21:32<60:20:45, 5.70s/it]
13%|█▎ | 5795/43939 [9:21:38<60:26:00, 5.70s/it]
13%|█▎ | 5796/43939 [9:21:43<60:19:27, 5.69s/it]
13%|█▎ | 5797/43939 [9:21:49<60:29:10, 5.71s/it]
13%|█▎ | 5798/43939 [9:21:55<60:18:24, 5.69s/it]
13%|█▎ | 5799/43939 [9:22:01<60:25:39, 5.70s/it]
13%|█▎ | 5800/43939 [9:22:06<60:17:02, 5.69s/it]
{'loss': 1.0253, 'grad_norm': 1.3936585187911987, 'learning_rate': 4.984421568902643e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.27, 'epoch': 0.53}
13%|█▎ | 5800/43939 [9:22:06<60:17:02, 5.69s/it]
13%|█▎ | 5801/43939 [9:22:12<60:29:13, 5.71s/it]
13%|█▎ | 5802/43939 [9:22:18<60:31:11, 5.71s/it]
13%|█▎ | 5803/43939 [9:22:24<61:19:07, 5.79s/it]
13%|█▎ | 5804/43939 [9:22:29<60:56:21, 5.75s/it]
13%|█▎ | 5805/43939 [9:22:35<60:41:40, 5.73s/it]
13%|█▎ | 5806/43939 [9:22:41<60:30:01, 5.71s/it]
13%|█▎ | 5807/43939 [9:22:46<60:22:50, 5.70s/it]
13%|█▎ | 5808/43939 [9:22:52<60:27:58, 5.71s/it]
13%|█▎ | 5809/43939 [9:22:58<60:32:22, 5.72s/it]
13%|█▎ | 5810/43939 [9:23:04<60:31:29, 5.71s/it]
13%|█▎ | 5811/43939 [9:23:09<60:28:46, 5.71s/it]
13%|█▎ | 5812/43939 [9:23:15<60:26:48, 5.71s/it]
13%|█▎ | 5813/43939 [9:23:21<60:26:29, 5.71s/it]
13%|█▎ | 5814/43939 [9:23:26<60:23:36, 5.70s/it]
13%|█▎ | 5815/43939 [9:23:32<60:20:00, 5.70s/it]
13%|█▎ | 5816/43939 [9:23:38<60:26:36, 5.71s/it]
13%|█▎ | 5817/43939 [9:23:43<60:20:29, 5.70s/it]
13%|█▎ | 5818/43939 [9:23:49<60:21:58, 5.70s/it]
13%|█▎ | 5819/43939 [9:23:55<60:26:20, 5.71s/it]
13%|█▎ | 5820/43939 [9:24:00<60:12:48, 5.69s/it]
13%|█▎ | 5821/43939 [9:24:06<60:17:30, 5.69s/it]
13%|█▎ | 5822/43939 [9:24:12<60:30:04, 5.71s/it]
13%|█▎ | 5823/43939 [9:24:18<60:20:01, 5.70s/it]
13%|█▎ | 5824/43939 [9:24:24<61:12:21, 5.78s/it]
13%|█▎ | 5825/43939 [9:24:29<60:54:03, 5.75s/it]
13%|█▎ | 5826/43939 [9:24:35<60:39:04, 5.73s/it]
13%|█▎ | 5827/43939 [9:24:41<60:28:39, 5.71s/it]
13%|█▎ | 5828/43939 [9:24:46<60:25:35, 5.71s/it]
13%|█▎ | 5829/43939 [9:24:52<60:19:03, 5.70s/it]
13%|█▎ | 5830/43939 [9:24:58<60:17:02, 5.69s/it]
13%|█▎ | 5831/43939 [9:25:03<60:09:42, 5.68s/it]
13%|█▎ | 5832/43939 [9:25:09<60:14:34, 5.69s/it]
13%|█▎ | 5833/43939 [9:25:15<60:23:28, 5.71s/it]
13%|█▎ | 5834/43939 [9:25:20<60:22:30, 5.70s/it]
13%|█▎ | 5835/43939 [9:25:26<60:18:54, 5.70s/it]
13%|█▎ | 5836/43939 [9:25:32<60:24:56, 5.71s/it]
13%|█▎ | 5837/43939 [9:25:38<60:19:39, 5.70s/it]
13%|█▎ | 5838/43939 [9:25:43<60:20:17, 5.70s/it]
13%|█▎ | 5839/43939 [9:25:49<60:14:53, 5.69s/it]
13%|█▎ | 5840/43939 [9:25:55<60:13:58, 5.69s/it]
13%|█▎ | 5841/43939 [9:26:00<60:20:52, 5.70s/it]
13%|█▎ | 5842/43939 [9:26:06<60:22:07, 5.70s/it]
13%|█▎ | 5843/43939 [9:26:12<60:15:28, 5.69s/it]
13%|█▎ | 5844/43939 [9:26:17<60:10:01, 5.69s/it]
13%|█▎ | 5845/43939 [9:26:23<60:13:21, 5.69s/it]
13%|█▎ | 5846/43939 [9:26:29<60:56:28, 5.76s/it]
13%|█▎ | 5847/43939 [9:26:35<61:01:23, 5.77s/it]
13%|█▎ | 5848/43939 [9:26:40<60:42:22, 5.74s/it]
13%|█▎ | 5849/43939 [9:26:46<60:44:53, 5.74s/it]
13%|█▎ | 5850/43939 [9:26:52<60:27:07, 5.71s/it]
{'loss': 1.0549, 'grad_norm': 1.12526535987854, 'learning_rate': 4.983295128656856e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.14, 'epoch': 0.53}
13%|█▎ | 5850/43939 [9:26:52<60:27:07, 5.71s/it]
13%|█▎ | 5851/43939 [9:26:58<60:24:46, 5.71s/it]
13%|█▎ | 5852/43939 [9:27:03<60:22:09, 5.71s/it]
13%|█▎ | 5853/43939 [9:27:09<60:24:25, 5.71s/it]
13%|█▎ | 5854/43939 [9:27:15<60:14:34, 5.69s/it]
13%|█▎ | 5855/43939 [9:27:20<60:17:25, 5.70s/it]
13%|█▎ | 5856/43939 [9:27:26<60:16:27, 5.70s/it]
13%|█▎ | 5857/43939 [9:27:32<60:13:14, 5.69s/it]
13%|█▎ | 5858/43939 [9:27:37<60:14:13, 5.69s/it]
13%|█▎ | 5859/43939 [9:27:43<60:14:54, 5.70s/it]
13%|█▎ | 5860/43939 [9:27:49<60:12:44, 5.69s/it]
13%|█▎ | 5861/43939 [9:27:55<60:17:44, 5.70s/it]
13%|█▎ | 5862/43939 [9:28:00<60:15:58, 5.70s/it]
13%|█▎ | 5863/43939 [9:28:06<60:16:12, 5.70s/it]
13%|█▎ | 5864/43939 [9:28:12<60:16:25, 5.70s/it]
13%|█▎ | 5865/43939 [9:28:17<60:19:05, 5.70s/it]
13%|█▎ | 5866/43939 [9:28:23<61:01:39, 5.77s/it]
13%|█▎ | 5867/43939 [9:28:29<60:53:40, 5.76s/it]
13%|█▎ | 5868/43939 [9:28:35<60:45:34, 5.75s/it]
13%|█▎ | 5869/43939 [9:28:40<60:35:04, 5.73s/it]
13%|█▎ | 5870/43939 [9:28:46<60:22:45, 5.71s/it]
13%|█▎ | 5871/43939 [9:28:52<60:15:21, 5.70s/it]
13%|█▎ | 5872/43939 [9:28:58<60:30:12, 5.72s/it]
13%|█▎ | 5873/43939 [9:29:03<60:25:16, 5.71s/it]
13%|█▎ | 5874/43939 [9:29:09<60:21:28, 5.71s/it]
13%|█▎ | 5875/43939 [9:29:15<60:23:55, 5.71s/it]
13%|█▎ | 5876/43939 [9:29:20<60:27:04, 5.72s/it]
13%|█▎ | 5877/43939 [9:29:26<60:16:12, 5.70s/it]
13%|█▎ | 5878/43939 [9:29:32<60:15:26, 5.70s/it]
13%|█▎ | 5879/43939 [9:29:37<60:06:48, 5.69s/it]
13%|█▎ | 5880/43939 [9:29:43<60:04:30, 5.68s/it]
13%|█▎ | 5881/43939 [9:29:49<60:04:40, 5.68s/it]
13%|█▎ | 5882/43939 [9:29:54<60:15:54, 5.70s/it]
13%|█▎ | 5883/43939 [9:30:00<60:08:05, 5.69s/it]
13%|█▎ | 5884/43939 [9:30:06<60:04:17, 5.68s/it]
13%|█▎ | 5885/43939 [9:30:11<60:05:12, 5.68s/it]
13%|█▎ | 5886/43939 [9:30:17<60:19:11, 5.71s/it]
13%|█▎ | 5887/43939 [9:30:23<61:01:53, 5.77s/it]
13%|█▎ | 5888/43939 [9:30:29<60:48:28, 5.75s/it]
13%|█▎ | 5889/43939 [9:30:35<60:37:15, 5.74s/it]
13%|█▎ | 5890/43939 [9:30:40<60:33:16, 5.73s/it]
13%|█▎ | 5891/43939 [9:30:46<60:32:19, 5.73s/it]
13%|█▎ | 5892/43939 [9:30:52<60:25:49, 5.72s/it]
13%|█▎ | 5893/43939 [9:30:57<60:19:39, 5.71s/it]
13%|█▎ | 5894/43939 [9:31:03<60:21:42, 5.71s/it]
13%|█▎ | 5895/43939 [9:31:09<60:18:44, 5.71s/it]
13%|█▎ | 5896/43939 [9:31:15<60:19:33, 5.71s/it]
13%|█▎ | 5897/43939 [9:31:20<60:16:28, 5.70s/it]
13%|█▎ | 5898/43939 [9:31:26<60:17:42, 5.71s/it]
13%|█▎ | 5899/43939 [9:31:32<60:10:23, 5.69s/it]
13%|█▎ | 5900/43939 [9:31:37<60:07:36, 5.69s/it]
{'loss': 1.0723, 'grad_norm': 1.0340914726257324, 'learning_rate': 4.9821295085947515e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.14, 'epoch': 0.54}
13%|█▎ | 5900/43939 [9:31:37<60:07:36, 5.69s/it]
13%|█▎ | 5901/43939 [9:31:43<60:10:00, 5.69s/it]
13%|█▎ | 5902/43939 [9:31:49<60:13:32, 5.70s/it]
13%|█▎ | 5903/43939 [9:31:54<60:11:52, 5.70s/it]
13%|█▎ | 5904/43939 [9:32:00<60:16:55, 5.71s/it]
13%|█▎ | 5905/43939 [9:32:06<60:11:16, 5.70s/it]
13%|█▎ | 5906/43939 [9:32:11<60:07:00, 5.69s/it]
13%|█▎ | 5907/43939 [9:32:17<60:07:32, 5.69s/it]
13%|█▎ | 5908/43939 [9:32:23<60:48:40, 5.76s/it]
13%|█▎ | 5909/43939 [9:32:29<60:33:58, 5.73s/it]
13%|█▎ | 5910/43939 [9:32:34<60:29:48, 5.73s/it]
13%|█▎ | 5911/43939 [9:32:40<60:18:21, 5.71s/it]
13%|█▎ | 5912/43939 [9:32:46<60:17:34, 5.71s/it]
13%|█▎ | 5913/43939 [9:32:52<60:14:37, 5.70s/it]
13%|█▎ | 5914/43939 [9:32:57<60:16:36, 5.71s/it]
13%|█▎ | 5915/43939 [9:33:03<60:13:27, 5.70s/it]
13%|█▎ | 5916/43939 [9:33:09<60:08:07, 5.69s/it]
13%|█▎ | 5917/43939 [9:33:14<60:09:23, 5.70s/it]
13%|█▎ | 5918/43939 [9:33:20<60:02:28, 5.68s/it]
13%|█▎ | 5919/43939 [9:33:26<60:09:41, 5.70s/it]
13%|█▎ | 5920/43939 [9:33:31<60:12:25, 5.70s/it]
13%|█▎ | 5921/43939 [9:33:37<60:06:11, 5.69s/it]
13%|█▎ | 5922/43939 [9:33:43<60:07:40, 5.69s/it]
13%|█▎ | 5923/43939 [9:33:48<60:04:12, 5.69s/it]
13%|█▎ | 5924/43939 [9:33:54<60:07:57, 5.69s/it]
13%|█▎ | 5925/43939 [9:34:00<60:00:47, 5.68s/it]
13%|█▎ | 5926/43939 [9:34:06<60:11:30, 5.70s/it]
13%|█▎ | 5927/43939 [9:34:11<60:13:14, 5.70s/it]
13%|█▎ | 5928/43939 [9:34:17<60:14:58, 5.71s/it]
13%|█▎ | 5929/43939 [9:34:23<60:11:01, 5.70s/it]
13%|█▎ | 5930/43939 [9:34:29<60:47:49, 5.76s/it]
13%|█▎ | 5931/43939 [9:34:34<60:28:10, 5.73s/it]
14%|█▎ | 5932/43939 [9:34:40<60:22:41, 5.72s/it]
14%|█▎ | 5933/43939 [9:34:46<60:21:04, 5.72s/it]
14%|█▎ | 5934/43939 [9:34:51<60:22:09, 5.72s/it]
14%|█▎ | 5935/43939 [9:34:57<60:24:38, 5.72s/it]
14%|█▎ | 5936/43939 [9:35:03<60:26:56, 5.73s/it]
14%|█▎ | 5937/43939 [9:35:09<60:22:14, 5.72s/it]
14%|█▎ | 5938/43939 [9:35:14<60:16:39, 5.71s/it]
14%|█▎ | 5939/43939 [9:35:20<60:15:08, 5.71s/it]
14%|█▎ | 5940/43939 [9:35:26<60:17:08, 5.71s/it]
14%|█▎ | 5941/43939 [9:35:31<60:11:17, 5.70s/it]
14%|█▎ | 5942/43939 [9:35:37<60:11:46, 5.70s/it]
14%|█▎ | 5943/43939 [9:35:43<60:22:06, 5.72s/it]
14%|█▎ | 5944/43939 [9:35:48<60:18:21, 5.71s/it]
14%|█▎ | 5945/43939 [9:35:54<60:14:56, 5.71s/it]
14%|█▎ | 5946/43939 [9:36:00<60:17:10, 5.71s/it]
14%|█▎ | 5947/43939 [9:36:06<60:05:50, 5.69s/it]
14%|█▎ | 5948/43939 [9:36:11<60:08:15, 5.70s/it]
14%|█▎ | 5949/43939 [9:36:17<60:03:37, 5.69s/it]
14%|█▎ | 5950/43939 [9:36:23<60:13:40, 5.71s/it]
{'loss': 1.0622, 'grad_norm': 1.0414705276489258, 'learning_rate': 4.980924727106725e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.2, 'epoch': 0.54}
14%|█▎ | 5950/43939 [9:36:23<60:13:40, 5.71s/it]
14%|█▎ | 5951/43939 [9:36:29<60:53:31, 5.77s/it]
14%|█▎ | 5952/43939 [9:36:34<60:36:40, 5.74s/it]
14%|█▎ | 5953/43939 [9:36:40<60:20:28, 5.72s/it]
14%|█▎ | 5954/43939 [9:36:46<60:10:58, 5.70s/it]
14%|█▎ | 5955/43939 [9:36:51<60:03:31, 5.69s/it]
14%|█▎ | 5956/43939 [9:36:57<60:03:07, 5.69s/it]
14%|█▎ | 5957/43939 [9:37:03<60:01:24, 5.69s/it]
14%|█▎ | 5958/43939 [9:37:08<60:05:46, 5.70s/it]
14%|█▎ | 5959/43939 [9:37:14<60:06:21, 5.70s/it]
14%|█▎ | 5960/43939 [9:37:20<60:05:36, 5.70s/it]
14%|█▎ | 5961/43939 [9:37:25<60:05:35, 5.70s/it]
14%|█▎ | 5962/43939 [9:37:31<60:11:46, 5.71s/it]
14%|█▎ | 5963/43939 [9:37:37<60:18:33, 5.72s/it]
14%|█▎ | 5964/43939 [9:37:43<60:30:46, 5.74s/it]
14%|█▎ | 5965/43939 [9:37:48<60:24:52, 5.73s/it]
14%|█▎ | 5966/43939 [9:37:54<60:15:42, 5.71s/it]
14%|█▎ | 5967/43939 [9:38:00<60:14:13, 5.71s/it]
14%|█▎ | 5968/43939 [9:38:05<60:13:29, 5.71s/it]
14%|█▎ | 5969/43939 [9:38:11<60:10:57, 5.71s/it]
14%|█▎ | 5970/43939 [9:38:17<60:11:29, 5.71s/it]
14%|█▎ | 5971/43939 [9:38:23<60:58:24, 5.78s/it]
14%|█▎ | 5972/43939 [9:38:29<60:41:10, 5.75s/it]
14%|█▎ | 5973/43939 [9:38:34<60:26:20, 5.73s/it]
14%|█▎ | 5974/43939 [9:38:40<60:29:20, 5.74s/it]
14%|█▎ | 5975/43939 [9:38:46<60:28:26, 5.73s/it]
14%|█▎ | 5976/43939 [9:38:51<60:21:30, 5.72s/it]
14%|█▎ | 5977/43939 [9:38:57<60:14:05, 5.71s/it]
14%|█▎ | 5978/43939 [9:39:03<60:12:50, 5.71s/it]
14%|█▎ | 5979/43939 [9:39:08<60:08:15, 5.70s/it]
14%|█▎ | 5980/43939 [9:39:14<60:18:38, 5.72s/it]
14%|█▎ | 5981/43939 [9:39:20<60:14:22, 5.71s/it]
14%|█▎ | 5982/43939 [9:39:26<60:08:31, 5.70s/it]
14%|█▎ | 5983/43939 [9:39:31<60:15:21, 5.72s/it]
14%|█▎ | 5984/43939 [9:39:37<60:16:15, 5.72s/it]
14%|█▎ | 5985/43939 [9:39:43<60:22:34, 5.73s/it]
14%|█▎ | 5986/43939 [9:39:49<60:21:05, 5.72s/it]
14%|█▎ | 5987/43939 [9:39:54<60:08:29, 5.70s/it]
14%|█▎ | 5988/43939 [9:40:00<60:18:57, 5.72s/it]
14%|█▎ | 5989/43939 [9:40:06<60:16:58, 5.72s/it]
14%|█▎ | 5990/43939 [9:40:11<60:24:16, 5.73s/it]
14%|█▎ | 5991/43939 [9:40:17<60:21:41, 5.73s/it]
14%|█▎ | 5992/43939 [9:40:23<60:19:08, 5.72s/it]
14%|█▎ | 5993/43939 [9:40:29<61:07:56, 5.80s/it]
14%|█▎ | 5994/43939 [9:40:35<60:50:36, 5.77s/it]
14%|█▎ | 5995/43939 [9:40:40<60:31:46, 5.74s/it]
14%|█▎ | 5996/43939 [9:40:46<60:17:22, 5.72s/it]
14%|█▎ | 5997/43939 [9:40:52<60:11:02, 5.71s/it]
14%|█▎ | 5998/43939 [9:40:57<60:14:02, 5.72s/it]
14%|█▎ | 5999/43939 [9:41:03<60:17:07, 5.72s/it]
14%|█▎ | 6000/43939 [9:41:09<60:20:28, 5.73s/it]
{'loss': 1.1061, 'grad_norm': 1.0546574592590332, 'learning_rate': 4.979680803201034e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 528.09, 'epoch': 0.55}
14%|█▎ | 6000/43939 [9:41:09<60:20:28, 5.73s/it]
14%|█▎ | 6001/43939 [9:41:15<60:26:26, 5.74s/it]
14%|█▎ | 6002/43939 [9:41:20<60:15:37, 5.72s/it]
14%|█▎ | 6003/43939 [9:41:26<60:18:26, 5.72s/it]
14%|█▎ | 6004/43939 [9:41:32<60:15:03, 5.72s/it]
14%|█▎ | 6005/43939 [9:41:37<60:07:29, 5.71s/it]
14%|█▎ | 6006/43939 [9:41:43<60:10:17, 5.71s/it]
14%|█▎ | 6007/43939 [9:41:49<60:02:33, 5.70s/it]
14%|█▎ | 6008/43939 [9:41:54<60:06:52, 5.71s/it]
14%|█▎ | 6009/43939 [9:42:00<60:12:52, 5.72s/it]
14%|█▎ | 6010/43939 [9:42:06<60:08:08, 5.71s/it]
14%|█▎ | 6011/43939 [9:42:12<60:07:07, 5.71s/it]
14%|█▎ | 6012/43939 [9:42:17<60:09:07, 5.71s/it]
14%|█▎ | 6013/43939 [9:42:23<60:41:26, 5.76s/it]
14%|█▎ | 6014/43939 [9:42:29<60:29:09, 5.74s/it]
14%|█▎ | 6015/43939 [9:42:35<60:23:35, 5.73s/it]
14%|█▎ | 6016/43939 [9:42:40<60:17:03, 5.72s/it]
14%|█▎ | 6017/43939 [9:42:46<60:11:19, 5.71s/it]
14%|█▎ | 6018/43939 [9:42:52<60:05:10, 5.70s/it]
14%|█▎ | 6019/43939 [9:42:57<60:06:25, 5.71s/it]
14%|█▎ | 6020/43939 [9:43:03<60:02:50, 5.70s/it]
14%|█▎ | 6021/43939 [9:43:09<60:09:17, 5.71s/it]
14%|█▎ | 6022/43939 [9:43:14<60:05:47, 5.71s/it]
14%|█▎ | 6023/43939 [9:43:20<60:01:29, 5.70s/it]
14%|█▎ | 6024/43939 [9:43:26<60:08:22, 5.71s/it]
14%|█▎ | 6025/43939 [9:43:32<60:06:18, 5.71s/it]
14%|█▎ | 6026/43939 [9:43:37<60:08:22, 5.71s/it]
14%|█▎ | 6027/43939 [9:43:43<60:15:59, 5.72s/it]
14%|█▎ | 6028/43939 [9:43:49<60:04:40, 5.70s/it]
14%|█▎ | 6029/43939 [9:43:54<60:03:40, 5.70s/it]
14%|█▎ | 6030/43939 [9:44:00<60:09:27, 5.71s/it]
14%|█▎ | 6031/43939 [9:44:06<60:04:33, 5.71s/it]
14%|█▎ | 6032/43939 [9:44:12<60:08:08, 5.71s/it]
14%|█▎ | 6033/43939 [9:44:17<60:15:02, 5.72s/it]
14%|█▎ | 6034/43939 [9:44:23<60:53:18, 5.78s/it]
14%|█▎ | 6035/43939 [9:44:29<60:43:06, 5.77s/it]
14%|█▎ | 6036/43939 [9:44:35<60:48:36, 5.78s/it]
14%|█▎ | 6037/43939 [9:44:40<60:30:31, 5.75s/it]
14%|█▎ | 6038/43939 [9:44:46<60:20:04, 5.73s/it]
14%|█▎ | 6039/43939 [9:44:52<60:20:01, 5.73s/it]
14%|█▎ | 6040/43939 [9:44:58<60:11:53, 5.72s/it]
14%|█▎ | 6041/43939 [9:45:03<60:06:41, 5.71s/it]
14%|█▍ | 6042/43939 [9:45:09<60:06:49, 5.71s/it]
14%|█▍ | 6043/43939 [9:45:15<60:01:57, 5.70s/it]
14%|█▍ | 6044/43939 [9:45:20<59:54:58, 5.69s/it]
14%|█▍ | 6045/43939 [9:45:26<59:58:19, 5.70s/it]
14%|█▍ | 6046/43939 [9:45:32<60:05:51, 5.71s/it]
14%|█▍ | 6047/43939 [9:45:37<59:58:59, 5.70s/it]
14%|█▍ | 6048/43939 [9:45:43<59:58:48, 5.70s/it]
14%|█▍ | 6049/43939 [9:45:49<59:52:28, 5.69s/it]
14%|█▍ | 6050/43939 [9:45:55<59:57:43, 5.70s/it]
{'loss': 1.0554, 'grad_norm': 1.0350837707519531, 'learning_rate': 4.978397756503504e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.79, 'epoch': 0.55}
14%|█▍ | 6050/43939 [9:45:55<59:57:43, 5.70s/it]
14%|█▍ | 6051/43939 [9:46:00<59:50:56, 5.69s/it]
14%|█▍ | 6052/43939 [9:46:06<59:44:22, 5.68s/it]
14%|█▍ | 6053/43939 [9:46:12<59:52:28, 5.69s/it]
14%|█▍ | 6054/43939 [9:46:17<59:50:06, 5.69s/it]
14%|█▍ | 6055/43939 [9:46:23<59:53:01, 5.69s/it]
14%|█▍ | 6056/43939 [9:46:29<60:35:53, 5.76s/it]
14%|█▍ | 6057/43939 [9:46:35<60:23:40, 5.74s/it]
14%|█▍ | 6058/43939 [9:46:40<60:14:58, 5.73s/it]
14%|█▍ | 6059/43939 [9:46:46<60:04:02, 5.71s/it]
14%|█▍ | 6060/43939 [9:46:52<60:07:43, 5.71s/it]
14%|█▍ | 6061/43939 [9:46:57<60:02:31, 5.71s/it]
14%|█▍ | 6062/43939 [9:47:03<60:03:05, 5.71s/it]
14%|█▍ | 6063/43939 [9:47:09<59:59:50, 5.70s/it]
14%|█▍ | 6064/43939 [9:47:14<60:06:31, 5.71s/it]
14%|█▍ | 6065/43939 [9:47:20<59:57:25, 5.70s/it]
14%|█▍ | 6066/43939 [9:47:26<60:04:38, 5.71s/it]
14%|█▍ | 6067/43939 [9:47:32<59:59:55, 5.70s/it]
14%|█▍ | 6068/43939 [9:47:37<59:54:55, 5.70s/it]
14%|█▍ | 6069/43939 [9:47:43<59:59:05, 5.70s/it]
14%|█▍ | 6070/43939 [9:47:49<60:01:05, 5.71s/it]
14%|█▍ | 6071/43939 [9:47:54<59:51:19, 5.69s/it]
14%|█▍ | 6072/43939 [9:48:00<59:50:35, 5.69s/it]
14%|█▍ | 6073/43939 [9:48:06<59:59:17, 5.70s/it]
14%|█▍ | 6074/43939 [9:48:11<59:55:28, 5.70s/it]
14%|█▍ | 6075/43939 [9:48:17<59:56:14, 5.70s/it]
14%|█▍ | 6076/43939 [9:48:23<60:44:39, 5.78s/it]
14%|█▍ | 6077/43939 [9:48:29<60:31:10, 5.75s/it]
14%|█▍ | 6078/43939 [9:48:34<60:12:43, 5.73s/it]
14%|█▍ | 6079/43939 [9:48:40<60:09:11, 5.72s/it]
14%|█▍ | 6080/43939 [9:48:46<60:02:21, 5.71s/it]
14%|█▍ | 6081/43939 [9:48:52<60:01:41, 5.71s/it]
14%|█▍ | 6082/43939 [9:48:57<60:15:28, 5.73s/it]
14%|█▍ | 6083/43939 [9:49:03<60:12:53, 5.73s/it]
14%|█▍ | 6084/43939 [9:49:09<60:08:17, 5.72s/it]
14%|█▍ | 6085/43939 [9:49:15<60:14:06, 5.73s/it]
14%|█▍ | 6086/43939 [9:49:20<59:58:46, 5.70s/it]
14%|█▍ | 6087/43939 [9:49:26<60:00:10, 5.71s/it]
14%|█▍ | 6088/43939 [9:49:32<60:06:47, 5.72s/it]
14%|█▍ | 6089/43939 [9:49:37<60:05:46, 5.72s/it]
14%|█▍ | 6090/43939 [9:49:43<60:02:04, 5.71s/it]
14%|█▍ | 6091/43939 [9:49:49<59:58:36, 5.70s/it]
14%|█▍ | 6092/43939 [9:49:54<59:59:56, 5.71s/it]
14%|█▍ | 6093/43939 [9:50:00<60:02:27, 5.71s/it]
14%|█▍ | 6094/43939 [9:50:06<59:57:48, 5.70s/it]
14%|█▍ | 6095/43939 [9:50:12<59:52:01, 5.69s/it]
14%|█▍ | 6096/43939 [9:50:17<59:43:57, 5.68s/it]
14%|█▍ | 6097/43939 [9:50:23<60:35:59, 5.77s/it]
14%|█▍ | 6098/43939 [9:50:29<60:27:50, 5.75s/it]
14%|█▍ | 6099/43939 [9:50:35<60:12:09, 5.73s/it]
14%|█▍ | 6100/43939 [9:50:40<60:12:03, 5.73s/it]
{'loss': 1.056, 'grad_norm': 0.9223297238349915, 'learning_rate': 4.97707560725721e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.58, 'epoch': 0.56}
14%|█▍ | 6100/43939 [9:50:40<60:12:03, 5.73s/it]
14%|█▍ | 6101/43939 [9:50:46<60:11:50, 5.73s/it]
14%|█▍ | 6102/43939 [9:50:52<60:14:49, 5.73s/it]
14%|█▍ | 6103/43939 [9:50:57<60:14:55, 5.73s/it]
14%|█▍ | 6104/43939 [9:51:03<60:14:35, 5.73s/it]
14%|█▍ | 6105/43939 [9:51:09<60:09:32, 5.72s/it]
14%|█▍ | 6106/43939 [9:51:15<60:02:11, 5.71s/it]
14%|█▍ | 6107/43939 [9:51:20<59:58:41, 5.71s/it]
14%|█▍ | 6108/43939 [9:51:26<59:54:53, 5.70s/it]
14%|█▍ | 6109/43939 [9:51:32<60:00:53, 5.71s/it]
14%|█▍ | 6110/43939 [9:51:37<60:05:01, 5.72s/it]
14%|█▍ | 6111/43939 [9:51:43<60:05:04, 5.72s/it]
14%|█▍ | 6112/43939 [9:51:49<59:55:21, 5.70s/it]
14%|█▍ | 6113/43939 [9:51:54<59:54:21, 5.70s/it]
14%|█▍ | 6114/43939 [9:52:00<59:52:10, 5.70s/it]
14%|█▍ | 6115/43939 [9:52:06<59:52:26, 5.70s/it]
14%|█▍ | 6116/43939 [9:52:12<59:45:42, 5.69s/it]
14%|█▍ | 6117/43939 [9:52:17<59:44:56, 5.69s/it]
14%|█▍ | 6118/43939 [9:52:23<59:44:36, 5.69s/it]
14%|█▍ | 6119/43939 [9:52:29<60:37:19, 5.77s/it]
14%|█▍ | 6120/43939 [9:52:35<60:19:54, 5.74s/it]
14%|█▍ | 6121/43939 [9:52:40<60:02:42, 5.72s/it]
14%|█▍ | 6122/43939 [9:52:46<59:58:22, 5.71s/it]
14%|█▍ | 6123/43939 [9:52:52<59:51:02, 5.70s/it]
14%|█▍ | 6124/43939 [9:52:57<59:51:53, 5.70s/it]
14%|█▍ | 6125/43939 [9:53:03<59:45:11, 5.69s/it]
14%|█▍ | 6126/43939 [9:53:09<59:47:36, 5.69s/it]
14%|█▍ | 6127/43939 [9:53:14<59:42:37, 5.68s/it]
14%|█▍ | 6128/43939 [9:53:20<59:38:11, 5.68s/it]
14%|█▍ | 6129/43939 [9:53:26<59:37:16, 5.68s/it]
14%|█▍ | 6130/43939 [9:53:31<59:47:25, 5.69s/it]
14%|█▍ | 6131/43939 [9:53:37<59:43:53, 5.69s/it]
14%|█▍ | 6132/43939 [9:53:43<59:37:21, 5.68s/it]
14%|█▍ | 6133/43939 [9:53:48<59:38:29, 5.68s/it]
14%|█▍ | 6134/43939 [9:53:54<59:50:58, 5.70s/it]
14%|█▍ | 6135/43939 [9:54:00<59:51:43, 5.70s/it]
14%|█▍ | 6136/43939 [9:54:06<59:46:34, 5.69s/it]
14%|█▍ | 6137/43939 [9:54:11<59:44:53, 5.69s/it]
14%|█▍ | 6138/43939 [9:54:17<59:55:27, 5.71s/it]
14%|█▍ | 6139/43939 [9:54:23<60:28:59, 5.76s/it]
14%|█▍ | 6140/43939 [9:54:29<60:25:19, 5.75s/it]
14%|█▍ | 6141/43939 [9:54:34<60:15:04, 5.74s/it]
14%|█▍ | 6142/43939 [9:54:40<60:06:29, 5.73s/it]
14%|█▍ | 6143/43939 [9:54:46<60:05:56, 5.72s/it]
14%|█▍ | 6144/43939 [9:54:51<59:57:05, 5.71s/it]
14%|█▍ | 6145/43939 [9:54:57<60:06:49, 5.73s/it]
14%|█▍ | 6146/43939 [9:55:03<59:53:30, 5.71s/it]
14%|█▍ | 6147/43939 [9:55:09<59:57:47, 5.71s/it]
14%|█▍ | 6148/43939 [9:55:14<59:57:27, 5.71s/it]
14%|█▍ | 6149/43939 [9:55:20<59:53:36, 5.71s/it]
14%|█▍ | 6150/43939 [9:55:26<59:44:34, 5.69s/it]
{'loss': 1.0438, 'grad_norm': 1.9479364156723022, 'learning_rate': 4.9757143763221635e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.52, 'epoch': 0.56}
14%|█▍ | 6150/43939 [9:55:26<59:44:34, 5.69s/it]
14%|█▍ | 6151/43939 [9:55:31<59:45:40, 5.69s/it]
14%|█▍ | 6152/43939 [9:55:37<59:52:06, 5.70s/it]
14%|█▍ | 6153/43939 [9:55:43<59:49:42, 5.70s/it]
14%|█▍ | 6154/43939 [9:55:48<59:48:48, 5.70s/it]
14%|█▍ | 6155/43939 [9:55:54<59:42:11, 5.69s/it]
14%|█▍ | 6156/43939 [9:56:00<59:39:03, 5.68s/it]
14%|█▍ | 6157/43939 [9:56:05<59:37:54, 5.68s/it]
14%|█▍ | 6158/43939 [9:56:11<59:40:29, 5.69s/it]
14%|█▍ | 6159/43939 [9:56:17<59:55:53, 5.71s/it]
14%|█▍ | 6160/43939 [9:56:23<60:26:39, 5.76s/it]
14%|█▍ | 6161/43939 [9:56:28<60:17:15, 5.75s/it]
14%|█▍ | 6162/43939 [9:56:34<60:07:36, 5.73s/it]
14%|█▍ | 6163/43939 [9:56:40<59:59:49, 5.72s/it]
14%|█▍ | 6164/43939 [9:56:46<59:49:26, 5.70s/it]
14%|█▍ | 6165/43939 [9:56:51<59:47:37, 5.70s/it]
14%|█▍ | 6166/43939 [9:56:57<59:45:54, 5.70s/it]
14%|█▍ | 6167/43939 [9:57:03<59:48:04, 5.70s/it]
14%|█▍ | 6168/43939 [9:57:08<59:53:16, 5.71s/it]
14%|█▍ | 6169/43939 [9:57:14<59:50:13, 5.70s/it]
14%|█▍ | 6170/43939 [9:57:20<59:47:44, 5.70s/it]
14%|█▍ | 6171/43939 [9:57:25<59:46:36, 5.70s/it]
14%|█▍ | 6172/43939 [9:57:31<59:51:29, 5.71s/it]
14%|█▍ | 6173/43939 [9:57:37<60:09:41, 5.73s/it]
14%|█▍ | 6174/43939 [9:57:43<60:02:46, 5.72s/it]
14%|█▍ | 6175/43939 [9:57:48<59:56:11, 5.71s/it]
14%|█▍ | 6176/43939 [9:57:54<59:50:07, 5.70s/it]
14%|█▍ | 6177/43939 [9:58:00<59:46:19, 5.70s/it]
14%|█▍ | 6178/43939 [9:58:05<59:45:05, 5.70s/it]
14%|█▍ | 6179/43939 [9:58:11<59:40:40, 5.69s/it]
14%|█▍ | 6180/43939 [9:58:17<59:39:57, 5.69s/it]
14%|█▍ | 6181/43939 [9:58:22<59:35:56, 5.68s/it]
14%|█▍ | 6182/43939 [9:58:28<60:17:51, 5.75s/it]
14%|█▍ | 6183/43939 [9:58:34<60:11:55, 5.74s/it]
14%|█▍ | 6184/43939 [9:58:40<59:56:52, 5.72s/it]
14%|█▍ | 6185/43939 [9:58:45<59:53:04, 5.71s/it]
14%|█▍ | 6186/43939 [9:58:51<59:54:46, 5.71s/it]
14%|█▍ | 6187/43939 [9:58:57<59:56:33, 5.72s/it]
14%|█▍ | 6188/43939 [9:59:02<59:48:25, 5.70s/it]
14%|█▍ | 6189/43939 [9:59:08<59:47:41, 5.70s/it]
14%|█▍ | 6190/43939 [9:59:14<59:45:06, 5.70s/it]
14%|█▍ | 6191/43939 [9:59:20<59:41:39, 5.69s/it]
14%|█▍ | 6192/43939 [9:59:25<59:44:10, 5.70s/it]
14%|█▍ | 6193/43939 [9:59:31<59:34:07, 5.68s/it]
14%|█▍ | 6194/43939 [9:59:37<59:40:53, 5.69s/it]
14%|█▍ | 6195/43939 [9:59:42<59:53:02, 5.71s/it]
14%|█▍ | 6196/43939 [9:59:48<59:52:07, 5.71s/it]
14%|█▍ | 6197/43939 [9:59:54<59:57:56, 5.72s/it]
14%|█▍ | 6198/43939 [10:00:00<60:03:56, 5.73s/it]
14%|█▍ | 6199/43939 [10:00:05<59:56:31, 5.72s/it]
14%|█▍ | 6200/43939 [10:00:11<59:52:01, 5.71s/it]
{'loss': 1.0378, 'grad_norm': 1.0383278131484985, 'learning_rate': 4.9743140851749814e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.36, 'epoch': 0.56}
14%|█▍ | 6200/43939 [10:00:11<59:52:01, 5.71s/it]
14%|█▍ | 6201/43939 [10:00:17<59:50:19, 5.71s/it]
14%|█▍ | 6202/43939 [10:00:22<59:53:56, 5.71s/it]
14%|█▍ | 6203/43939 [10:00:28<60:30:43, 5.77s/it]
14%|█▍ | 6204/43939 [10:00:34<60:12:58, 5.74s/it]
14%|█▍ | 6205/43939 [10:00:40<60:06:03, 5.73s/it]
14%|█▍ | 6206/43939 [10:00:45<60:04:38, 5.73s/it]
14%|█▍ | 6207/43939 [10:00:51<59:54:31, 5.72s/it]
14%|█▍ | 6208/43939 [10:00:57<59:52:58, 5.71s/it]
14%|█▍ | 6209/43939 [10:01:03<59:51:14, 5.71s/it]
14%|█▍ | 6210/43939 [10:01:08<59:42:57, 5.70s/it]
14%|█▍ | 6211/43939 [10:01:14<59:41:25, 5.70s/it]
14%|█▍ | 6212/43939 [10:01:20<59:39:11, 5.69s/it]
14%|█▍ | 6213/43939 [10:01:25<59:36:20, 5.69s/it]
14%|█▍ | 6214/43939 [10:01:31<59:31:10, 5.68s/it]
14%|█▍ | 6215/43939 [10:01:37<59:27:19, 5.67s/it]
14%|█▍ | 6216/43939 [10:01:42<59:37:19, 5.69s/it]
14%|█▍ | 6217/43939 [10:01:48<59:41:45, 5.70s/it]
14%|█▍ | 6218/43939 [10:01:54<59:49:02, 5.71s/it]
14%|█▍ | 6219/43939 [10:01:59<59:45:13, 5.70s/it]
14%|█▍ | 6220/43939 [10:02:05<60:00:19, 5.73s/it]
14%|█▍ | 6221/43939 [10:02:11<59:57:46, 5.72s/it]
14%|█▍ | 6222/43939 [10:02:17<59:48:51, 5.71s/it]
14%|█▍ | 6223/43939 [10:02:22<59:51:43, 5.71s/it]
14%|█▍ | 6224/43939 [10:02:28<60:41:44, 5.79s/it]
14%|█▍ | 6225/43939 [10:02:34<60:17:12, 5.75s/it]
14%|█▍ | 6226/43939 [10:02:40<60:06:56, 5.74s/it]
14%|█▍ | 6227/43939 [10:02:45<60:00:40, 5.73s/it]
14%|█▍ | 6228/43939 [10:02:51<59:53:32, 5.72s/it]
14%|█▍ | 6229/43939 [10:02:57<59:45:40, 5.71s/it]
14%|█▍ | 6230/43939 [10:03:02<59:39:07, 5.69s/it]
14%|█▍ | 6231/43939 [10:03:08<59:45:46, 5.71s/it]
14%|█▍ | 6232/43939 [10:03:14<59:41:13, 5.70s/it]
14%|█▍ | 6233/43939 [10:03:20<59:45:33, 5.71s/it]
14%|█▍ | 6234/43939 [10:03:25<59:43:35, 5.70s/it]
14%|█▍ | 6235/43939 [10:03:31<59:39:01, 5.70s/it]
14%|█▍ | 6236/43939 [10:03:37<59:41:58, 5.70s/it]
14%|█▍ | 6237/43939 [10:03:42<59:45:11, 5.71s/it]
14%|█▍ | 6238/43939 [10:03:48<59:44:32, 5.70s/it]
14%|█▍ | 6239/43939 [10:03:54<59:44:51, 5.71s/it]
14%|█▍ | 6240/43939 [10:03:59<59:48:14, 5.71s/it]
14%|█▍ | 6241/43939 [10:04:05<59:49:03, 5.71s/it]
14%|█▍ | 6242/43939 [10:04:11<59:45:42, 5.71s/it]
14%|█▍ | 6243/43939 [10:04:17<59:43:14, 5.70s/it]
14%|█▍ | 6244/43939 [10:04:22<59:45:51, 5.71s/it]
14%|█▍ | 6245/43939 [10:04:28<60:27:23, 5.77s/it]
14%|█▍ | 6246/43939 [10:04:34<60:08:42, 5.74s/it]
14%|█▍ | 6247/43939 [10:04:40<59:55:33, 5.72s/it]
14%|█▍ | 6248/43939 [10:04:45<59:47:42, 5.71s/it]
14%|█▍ | 6249/43939 [10:04:51<59:44:49, 5.71s/it]
14%|█▍ | 6250/43939 [10:04:57<59:47:31, 5.71s/it]
{'loss': 1.0385, 'grad_norm': 1.022878646850586, 'learning_rate': 4.972874755908548e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.72, 'epoch': 0.57}
14%|█▍ | 6250/43939 [10:04:57<59:47:31, 5.71s/it]
14%|█▍ | 6251/43939 [10:05:02<59:55:10, 5.72s/it]
14%|█▍ | 6252/43939 [10:05:08<59:51:53, 5.72s/it]
14%|█▍ | 6253/43939 [10:05:14<59:50:53, 5.72s/it]
14%|█▍ | 6254/43939 [10:05:20<59:46:34, 5.71s/it]
14%|█▍ | 6255/43939 [10:05:25<59:44:16, 5.71s/it]
14%|█▍ | 6256/43939 [10:05:31<59:49:34, 5.72s/it]
14%|█▍ | 6257/43939 [10:05:37<59:49:29, 5.72s/it]
14%|█▍ | 6258/43939 [10:05:42<59:48:22, 5.71s/it]
14%|█▍ | 6259/43939 [10:05:48<60:03:41, 5.74s/it]
14%|█▍ | 6260/43939 [10:05:54<59:57:56, 5.73s/it]
14%|█▍ | 6261/43939 [10:06:00<59:46:42, 5.71s/it]
14%|█▍ | 6262/43939 [10:06:05<59:45:29, 5.71s/it]
14%|█▍ | 6263/43939 [10:06:11<59:49:45, 5.72s/it]
14%|█▍ | 6264/43939 [10:06:17<59:47:51, 5.71s/it]
14%|█▍ | 6265/43939 [10:06:22<59:47:46, 5.71s/it]
14%|█▍ | 6266/43939 [10:06:28<60:23:42, 5.77s/it]
14%|█▍ | 6267/43939 [10:06:34<60:07:36, 5.75s/it]
14%|█▍ | 6268/43939 [10:06:40<60:11:34, 5.75s/it]
14%|█▍ | 6269/43939 [10:06:46<60:02:58, 5.74s/it]
14%|█▍ | 6270/43939 [10:06:51<59:53:37, 5.72s/it]
14%|█▍ | 6271/43939 [10:06:57<59:50:51, 5.72s/it]
14%|█▍ | 6272/43939 [10:07:03<59:45:51, 5.71s/it]
14%|█▍ | 6273/43939 [10:07:08<59:43:36, 5.71s/it]
14%|█▍ | 6274/43939 [10:07:14<59:38:24, 5.70s/it]
14%|█▍ | 6275/43939 [10:07:20<59:37:53, 5.70s/it]
14%|█▍ | 6276/43939 [10:07:25<59:36:59, 5.70s/it]
14%|█▍ | 6277/43939 [10:07:31<59:36:54, 5.70s/it]
14%|█▍ | 6278/43939 [10:07:37<59:52:24, 5.72s/it]
14%|█▍ | 6279/43939 [10:07:43<59:41:04, 5.71s/it]
14%|█▍ | 6280/43939 [10:07:48<59:33:06, 5.69s/it]
14%|█▍ | 6281/43939 [10:07:54<59:35:26, 5.70s/it]
14%|█▍ | 6282/43939 [10:08:00<59:35:45, 5.70s/it]
14%|█▍ | 6283/43939 [10:08:05<59:36:37, 5.70s/it]
14%|█▍ | 6284/43939 [10:08:11<59:38:50, 5.70s/it]
14%|█▍ | 6285/43939 [10:08:17<59:41:59, 5.71s/it]
14%|█▍ | 6286/43939 [10:08:22<59:38:16, 5.70s/it]
14%|█▍ | 6287/43939 [10:08:28<60:19:45, 5.77s/it]
14%|█▍ | 6288/43939 [10:08:34<60:01:03, 5.74s/it]
14%|█▍ | 6289/43939 [10:08:40<59:56:21, 5.73s/it]
14%|█▍ | 6290/43939 [10:08:45<59:47:26, 5.72s/it]
14%|█▍ | 6291/43939 [10:08:51<59:49:29, 5.72s/it]
14%|█▍ | 6292/43939 [10:08:57<59:45:07, 5.71s/it]
14%|█▍ | 6293/43939 [10:09:03<59:39:59, 5.71s/it]
14%|█▍ | 6294/43939 [10:09:08<59:43:07, 5.71s/it]
14%|█▍ | 6295/43939 [10:09:14<59:42:10, 5.71s/it]
14%|█▍ | 6296/43939 [10:09:20<59:32:06, 5.69s/it]
14%|█▍ | 6297/43939 [10:09:25<59:32:28, 5.69s/it]
14%|█▍ | 6298/43939 [10:09:31<59:32:24, 5.69s/it]
14%|█▍ | 6299/43939 [10:09:37<59:33:00, 5.70s/it]
14%|█▍ | 6300/43939 [10:09:42<59:32:29, 5.69s/it]
{'loss': 1.0658, 'grad_norm': 1.0296788215637207, 'learning_rate': 4.971396411231664e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.78, 'epoch': 0.57}
14%|█▍ | 6300/43939 [10:09:42<59:32:29, 5.69s/it]
14%|█▍ | 6301/43939 [10:09:48<59:34:42, 5.70s/it]
14%|█▍ | 6302/43939 [10:09:54<59:34:03, 5.70s/it]
14%|█▍ | 6303/43939 [10:09:59<59:27:26, 5.69s/it]
14%|█▍ | 6304/43939 [10:10:05<59:24:56, 5.68s/it]
14%|█▍ | 6305/43939 [10:10:11<59:21:57, 5.68s/it]
14%|█▍ | 6306/43939 [10:10:16<59:23:01, 5.68s/it]
14%|█▍ | 6307/43939 [10:10:22<59:21:02, 5.68s/it]
14%|█▍ | 6308/43939 [10:10:28<60:10:51, 5.76s/it]
14%|█▍ | 6309/43939 [10:10:34<59:55:58, 5.73s/it]
14%|█▍ | 6310/43939 [10:10:39<59:48:58, 5.72s/it]
14%|█▍ | 6311/43939 [10:10:45<59:50:17, 5.72s/it]
14%|█▍ | 6312/43939 [10:10:51<59:39:27, 5.71s/it]
14%|█▍ | 6313/43939 [10:10:57<59:43:14, 5.71s/it]
14%|█▍ | 6314/43939 [10:11:02<59:38:55, 5.71s/it]
14%|█▍ | 6315/43939 [10:11:08<59:37:13, 5.70s/it]
14%|█▍ | 6316/43939 [10:11:14<59:38:18, 5.71s/it]
14%|█▍ | 6317/43939 [10:11:19<59:31:50, 5.70s/it]
14%|█▍ | 6318/43939 [10:11:25<59:30:00, 5.69s/it]
14%|█▍ | 6319/43939 [10:11:31<59:23:23, 5.68s/it]
14%|█▍ | 6320/43939 [10:11:36<59:26:55, 5.69s/it]
14%|█▍ | 6321/43939 [10:11:42<59:32:39, 5.70s/it]
14%|█▍ | 6322/43939 [10:11:48<59:43:12, 5.72s/it]
14%|█▍ | 6323/43939 [10:11:54<59:40:39, 5.71s/it]
14%|█▍ | 6324/43939 [10:11:59<59:32:15, 5.70s/it]
14%|█▍ | 6325/43939 [10:12:05<59:30:42, 5.70s/it]
14%|█▍ | 6326/43939 [10:12:11<59:35:29, 5.70s/it]
14%|█▍ | 6327/43939 [10:12:16<59:32:57, 5.70s/it]
14%|█▍ | 6328/43939 [10:12:22<59:35:38, 5.70s/it]
14%|█▍ | 6329/43939 [10:12:28<60:11:46, 5.76s/it]
14%|█▍ | 6330/43939 [10:12:34<60:09:47, 5.76s/it]
14%|█▍ | 6331/43939 [10:12:39<60:02:39, 5.75s/it]
14%|█▍ | 6332/43939 [10:12:45<59:53:00, 5.73s/it]
14%|█▍ | 6333/43939 [10:12:51<59:42:42, 5.72s/it]
14%|█▍ | 6334/43939 [10:12:57<59:41:09, 5.71s/it]
14%|█▍ | 6335/43939 [10:13:02<59:48:40, 5.73s/it]
14%|█▍ | 6336/43939 [10:13:08<59:38:42, 5.71s/it]
14%|█▍ | 6337/43939 [10:13:14<59:36:42, 5.71s/it]
14%|█▍ | 6338/43939 [10:13:19<59:36:02, 5.71s/it]
14%|█▍ | 6339/43939 [10:13:25<59:29:22, 5.70s/it]
14%|█▍ | 6340/43939 [10:13:31<59:25:08, 5.69s/it]
14%|█▍ | 6341/43939 [10:13:36<59:34:15, 5.70s/it]
14%|█▍ | 6342/43939 [10:13:42<59:40:47, 5.71s/it]
14%|█▍ | 6343/43939 [10:13:48<59:37:42, 5.71s/it]
14%|█▍ | 6344/43939 [10:13:54<59:30:27, 5.70s/it]
14%|█▍ | 6345/43939 [10:13:59<59:35:13, 5.71s/it]
14%|█▍ | 6346/43939 [10:14:05<59:28:01, 5.69s/it]
14%|█▍ | 6347/43939 [10:14:11<59:23:58, 5.69s/it]
14%|█▍ | 6348/43939 [10:14:16<59:33:43, 5.70s/it]
14%|█▍ | 6349/43939 [10:14:22<59:32:07, 5.70s/it]
14%|█▍ | 6350/43939 [10:14:28<60:14:03, 5.77s/it]
{'loss': 1.0536, 'grad_norm': 1.2907683849334717, 'learning_rate': 4.9698790744686903e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.21, 'epoch': 0.58}
14%|█▍ | 6350/43939 [10:14:28<60:14:03, 5.77s/it]
14%|█▍ | 6351/43939 [10:14:34<60:12:37, 5.77s/it]
14%|█▍ | 6352/43939 [10:14:39<60:02:24, 5.75s/it]
14%|█▍ | 6353/43939 [10:14:45<59:51:25, 5.73s/it]
14%|█▍ | 6354/43939 [10:14:51<59:47:49, 5.73s/it]
14%|█▍ | 6355/43939 [10:14:57<59:37:47, 5.71s/it]
14%|█▍ | 6356/43939 [10:15:02<59:41:47, 5.72s/it]
14%|█▍ | 6357/43939 [10:15:08<59:38:40, 5.71s/it]
14%|█▍ | 6358/43939 [10:15:14<59:35:12, 5.71s/it]
14%|█▍ | 6359/43939 [10:15:19<59:27:54, 5.70s/it]
14%|█▍ | 6360/43939 [10:15:25<59:31:01, 5.70s/it]
14%|█▍ | 6361/43939 [10:15:31<59:27:40, 5.70s/it]
14%|█▍ | 6362/43939 [10:15:36<59:29:20, 5.70s/it]
14%|█▍ | 6363/43939 [10:15:42<59:31:23, 5.70s/it]
14%|█▍ | 6364/43939 [10:15:48<59:40:59, 5.72s/it]
14%|█▍ | 6365/43939 [10:15:54<59:37:55, 5.71s/it]
14%|█▍ | 6366/43939 [10:15:59<59:45:58, 5.73s/it]
14%|█▍ | 6367/43939 [10:16:05<59:37:22, 5.71s/it]
14%|█▍ | 6368/43939 [10:16:11<59:38:08, 5.71s/it]
14%|█▍ | 6369/43939 [10:16:16<59:32:07, 5.70s/it]
14%|█▍ | 6370/43939 [10:16:22<60:12:58, 5.77s/it]
14%|█▍ | 6371/43939 [10:16:28<59:58:09, 5.75s/it]
15%|█▍ | 6372/43939 [10:16:34<59:52:13, 5.74s/it]
15%|█▍ | 6373/43939 [10:16:40<59:47:11, 5.73s/it]
15%|█▍ | 6374/43939 [10:16:45<59:44:28, 5.73s/it]
15%|█▍ | 6375/43939 [10:16:51<59:36:57, 5.71s/it]
15%|█▍ | 6376/43939 [10:16:57<59:35:46, 5.71s/it]
15%|█▍ | 6377/43939 [10:17:02<59:40:06, 5.72s/it]
15%|█▍ | 6378/43939 [10:17:08<59:41:55, 5.72s/it]
15%|█▍ | 6379/43939 [10:17:14<59:34:34, 5.71s/it]
15%|█▍ | 6380/43939 [10:17:19<59:30:47, 5.70s/it]
15%|█▍ | 6381/43939 [10:17:25<59:28:26, 5.70s/it]
15%|█▍ | 6382/43939 [10:17:31<59:24:46, 5.69s/it]
15%|█▍ | 6383/43939 [10:17:36<59:19:30, 5.69s/it]
15%|█▍ | 6384/43939 [10:17:42<59:21:51, 5.69s/it]
15%|█▍ | 6385/43939 [10:17:48<59:23:42, 5.69s/it]
15%|█▍ | 6386/43939 [10:17:54<59:32:30, 5.71s/it]
15%|█▍ | 6387/43939 [10:17:59<59:40:02, 5.72s/it]
15%|█▍ | 6388/43939 [10:18:05<59:31:32, 5.71s/it]
15%|█▍ | 6389/43939 [10:18:11<59:35:51, 5.71s/it]
15%|█▍ | 6390/43939 [10:18:16<59:31:33, 5.71s/it]
15%|█▍ | 6391/43939 [10:18:22<59:32:57, 5.71s/it]
15%|█▍ | 6392/43939 [10:18:28<60:14:21, 5.78s/it]
15%|█▍ | 6393/43939 [10:18:34<60:09:07, 5.77s/it]
15%|█▍ | 6394/43939 [10:18:40<59:56:10, 5.75s/it]
15%|█▍ | 6395/43939 [10:18:45<59:50:09, 5.74s/it]
15%|█▍ | 6396/43939 [10:18:51<59:43:26, 5.73s/it]
15%|█▍ | 6397/43939 [10:18:57<59:40:15, 5.72s/it]
15%|█▍ | 6398/43939 [10:19:02<59:41:51, 5.72s/it]
15%|█▍ | 6399/43939 [10:19:08<59:36:39, 5.72s/it]
15%|█▍ | 6400/43939 [10:19:14<59:32:38, 5.71s/it]
{'loss': 1.0722, 'grad_norm': 1.1980242729187012, 'learning_rate': 4.968322769559181e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.21, 'epoch': 0.58}
15%|█▍ | 6400/43939 [10:19:14<59:32:38, 5.71s/it]
15%|█▍ | 6401/43939 [10:19:20<59:45:06, 5.73s/it]
15%|█▍ | 6402/43939 [10:19:25<59:41:06, 5.72s/it]
15%|█▍ | 6403/43939 [10:19:31<59:35:08, 5.71s/it]
15%|█▍ | 6404/43939 [10:19:37<59:27:28, 5.70s/it]
15%|█▍ | 6405/43939 [10:19:42<59:28:40, 5.70s/it]
15%|█▍ | 6406/43939 [10:19:48<59:23:53, 5.70s/it]
15%|█▍ | 6407/43939 [10:19:54<59:25:06, 5.70s/it]
15%|█▍ | 6408/43939 [10:19:59<59:22:46, 5.70s/it]
15%|█▍ | 6409/43939 [10:20:05<59:22:33, 5.70s/it]
15%|█▍ | 6410/43939 [10:20:11<59:16:41, 5.69s/it]
15%|█▍ | 6411/43939 [10:20:17<59:26:59, 5.70s/it]
15%|█▍ | 6412/43939 [10:20:22<60:04:51, 5.76s/it]
15%|█▍ | 6413/43939 [10:20:28<59:50:12, 5.74s/it]
15%|█▍ | 6414/43939 [10:20:34<59:37:28, 5.72s/it]
15%|█▍ | 6415/43939 [10:20:40<59:33:40, 5.71s/it]
15%|█▍ | 6416/43939 [10:20:45<59:32:35, 5.71s/it]
15%|█▍ | 6417/43939 [10:20:51<59:22:34, 5.70s/it]
15%|█▍ | 6418/43939 [10:20:57<59:17:53, 5.69s/it]
15%|█▍ | 6419/43939 [10:21:02<59:15:47, 5.69s/it]
15%|█▍ | 6420/43939 [10:21:08<59:15:30, 5.69s/it]
15%|█▍ | 6421/43939 [10:21:14<59:14:40, 5.68s/it]
15%|█▍ | 6422/43939 [10:21:19<59:26:03, 5.70s/it]
15%|█▍ | 6423/43939 [10:21:25<59:19:52, 5.69s/it]
15%|█▍ | 6424/43939 [10:21:31<59:25:40, 5.70s/it]
15%|█▍ | 6425/43939 [10:21:37<59:37:22, 5.72s/it]
15%|█▍ | 6426/43939 [10:21:42<59:27:13, 5.71s/it]
15%|█▍ | 6427/43939 [10:21:48<59:30:02, 5.71s/it]
15%|█▍ | 6428/43939 [10:21:54<59:24:57, 5.70s/it]
15%|█▍ | 6429/43939 [10:21:59<59:20:20, 5.70s/it]
15%|█▍ | 6430/43939 [10:22:05<59:24:33, 5.70s/it]
15%|█▍ | 6431/43939 [10:22:11<59:30:21, 5.71s/it]
15%|█▍ | 6432/43939 [10:22:16<59:27:41, 5.71s/it]
15%|█▍ | 6433/43939 [10:22:22<60:14:41, 5.78s/it]
15%|█▍ | 6434/43939 [10:22:28<60:06:09, 5.77s/it]
15%|█▍ | 6435/43939 [10:22:34<59:52:29, 5.75s/it]
15%|█▍ | 6436/43939 [10:22:40<59:52:55, 5.75s/it]
15%|█▍ | 6437/43939 [10:22:45<59:56:16, 5.75s/it]
15%|█▍ | 6438/43939 [10:22:51<59:43:43, 5.73s/it]
15%|█▍ | 6439/43939 [10:22:57<59:33:22, 5.72s/it]
15%|█▍ | 6440/43939 [10:23:02<59:26:54, 5.71s/it]
15%|█▍ | 6441/43939 [10:23:08<59:27:07, 5.71s/it]
15%|█▍ | 6442/43939 [10:23:14<59:21:33, 5.70s/it]
15%|█▍ | 6443/43939 [10:23:19<59:17:52, 5.69s/it]
15%|█▍ | 6444/43939 [10:23:25<59:20:15, 5.70s/it]
15%|█▍ | 6445/43939 [10:23:31<59:19:40, 5.70s/it]
15%|█▍ | 6446/43939 [10:23:37<59:18:53, 5.70s/it]
15%|█▍ | 6447/43939 [10:23:42<59:23:46, 5.70s/it]
15%|█▍ | 6448/43939 [10:23:48<59:21:35, 5.70s/it]
15%|█▍ | 6449/43939 [10:23:54<59:19:54, 5.70s/it]
15%|█▍ | 6450/43939 [10:23:59<59:30:00, 5.71s/it]
{'loss': 1.0533, 'grad_norm': 1.0062897205352783, 'learning_rate': 4.9667275210575036e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.36, 'epoch': 0.59}
15%|█▍ | 6450/43939 [10:23:59<59:30:00, 5.71s/it]
15%|█▍ | 6451/43939 [10:24:05<59:32:23, 5.72s/it]
15%|█▍ | 6452/43939 [10:24:11<59:28:14, 5.71s/it]
15%|█▍ | 6453/43939 [10:24:17<59:22:33, 5.70s/it]
15%|█▍ | 6454/43939 [10:24:22<60:14:37, 5.79s/it]
15%|█▍ | 6455/43939 [10:24:28<59:59:38, 5.76s/it]
15%|█▍ | 6456/43939 [10:24:34<59:40:03, 5.73s/it]
15%|█▍ | 6457/43939 [10:24:40<59:35:34, 5.72s/it]
15%|█▍ | 6458/43939 [10:24:45<59:30:22, 5.72s/it]
15%|█▍ | 6459/43939 [10:24:51<59:28:01, 5.71s/it]
15%|█▍ | 6460/43939 [10:24:57<59:19:42, 5.70s/it]
15%|█▍ | 6461/43939 [10:25:02<59:10:58, 5.68s/it]
15%|█▍ | 6462/43939 [10:25:08<59:09:34, 5.68s/it]
15%|█▍ | 6463/43939 [10:25:14<59:06:49, 5.68s/it]
15%|█▍ | 6464/43939 [10:25:19<59:09:34, 5.68s/it]
15%|█▍ | 6465/43939 [10:25:25<59:08:42, 5.68s/it]
15%|█▍ | 6466/43939 [10:25:31<59:12:17, 5.69s/it]
15%|█▍ | 6467/43939 [10:25:36<59:13:13, 5.69s/it]
15%|█▍ | 6468/43939 [10:25:42<59:17:31, 5.70s/it]
15%|█▍ | 6469/43939 [10:25:48<59:19:53, 5.70s/it]
15%|█▍ | 6470/43939 [10:25:54<59:19:38, 5.70s/it]
15%|█▍ | 6471/43939 [10:25:59<59:18:26, 5.70s/it]
15%|█▍ | 6472/43939 [10:26:05<59:17:17, 5.70s/it]
15%|█▍ | 6473/43939 [10:26:11<59:11:39, 5.69s/it]
15%|█▍ | 6474/43939 [10:26:16<59:31:05, 5.72s/it]
15%|█▍ | 6475/43939 [10:26:22<60:10:50, 5.78s/it]
15%|█▍ | 6476/43939 [10:26:28<59:51:00, 5.75s/it]
15%|█▍ | 6477/43939 [10:26:34<59:43:59, 5.74s/it]
15%|█▍ | 6478/43939 [10:26:39<59:41:45, 5.74s/it]
15%|█▍ | 6479/43939 [10:26:45<59:28:04, 5.72s/it]
15%|█▍ | 6480/43939 [10:26:51<59:28:22, 5.72s/it]
15%|█▍ | 6481/43939 [10:26:57<59:34:43, 5.73s/it]
15%|█▍ | 6482/43939 [10:27:02<59:33:03, 5.72s/it]
15%|█▍ | 6483/43939 [10:27:08<59:26:19, 5.71s/it]
15%|█▍ | 6484/43939 [10:27:14<59:21:44, 5.71s/it]
15%|█▍ | 6485/43939 [10:27:19<59:18:09, 5.70s/it]
15%|█▍ | 6486/43939 [10:27:25<59:09:29, 5.69s/it]
15%|█▍ | 6487/43939 [10:27:31<59:11:33, 5.69s/it]
15%|█▍ | 6488/43939 [10:27:36<59:12:29, 5.69s/it]
15%|█▍ | 6489/43939 [10:27:42<59:15:22, 5.70s/it]
15%|█▍ | 6490/43939 [10:27:48<59:14:35, 5.70s/it]
15%|█▍ | 6491/43939 [10:27:53<59:16:13, 5.70s/it]
15%|█▍ | 6492/43939 [10:27:59<59:18:45, 5.70s/it]
15%|█▍ | 6493/43939 [10:28:05<59:09:22, 5.69s/it]
15%|█▍ | 6494/43939 [10:28:11<59:04:31, 5.68s/it]
15%|█▍ | 6495/43939 [10:28:16<59:06:16, 5.68s/it]
15%|█▍ | 6496/43939 [10:28:22<59:51:33, 5.76s/it]
15%|█▍ | 6497/43939 [10:28:28<59:49:12, 5.75s/it]
15%|█▍ | 6498/43939 [10:28:34<59:43:30, 5.74s/it]
15%|█▍ | 6499/43939 [10:28:39<59:32:52, 5.73s/it]
15%|█▍ | 6500/43939 [10:28:45<59:28:20, 5.72s/it]
{'loss': 1.0687, 'grad_norm': 1.1628015041351318, 'learning_rate': 4.9650933541324506e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.69, 'epoch': 0.59}
15%|█▍ | 6500/43939 [10:28:45<59:28:20, 5.72s/it]
15%|█▍ | 6501/43939 [10:28:51<59:22:48, 5.71s/it]
15%|█▍ | 6502/43939 [10:28:56<59:20:36, 5.71s/it]
15%|█▍ | 6503/43939 [10:29:02<59:33:39, 5.73s/it]
15%|█▍ | 6504/43939 [10:29:08<59:26:04, 5.72s/it]
15%|█▍ | 6505/43939 [10:29:14<59:24:33, 5.71s/it]
15%|█▍ | 6506/43939 [10:29:19<59:18:19, 5.70s/it]
15%|█▍ | 6507/43939 [10:29:25<59:16:45, 5.70s/it]
15%|█▍ | 6508/43939 [10:29:31<59:14:33, 5.70s/it]
15%|█▍ | 6509/43939 [10:29:36<59:17:41, 5.70s/it]
15%|█▍ | 6510/43939 [10:29:42<59:06:36, 5.69s/it]
15%|█▍ | 6511/43939 [10:29:48<59:13:45, 5.70s/it]
15%|█▍ | 6512/43939 [10:29:53<59:15:48, 5.70s/it]
15%|█▍ | 6513/43939 [10:29:59<59:15:05, 5.70s/it]
15%|█▍ | 6514/43939 [10:30:05<59:10:14, 5.69s/it]
15%|█▍ | 6515/43939 [10:30:10<59:17:55, 5.70s/it]
15%|█▍ | 6516/43939 [10:30:16<59:09:05, 5.69s/it]
15%|█▍ | 6517/43939 [10:30:22<59:05:50, 5.69s/it]
15%|█▍ | 6518/43939 [10:30:28<59:54:19, 5.76s/it]
15%|█▍ | 6519/43939 [10:30:33<59:45:43, 5.75s/it]
15%|█▍ | 6520/43939 [10:30:39<59:47:52, 5.75s/it]
15%|█▍ | 6521/43939 [10:30:45<59:40:35, 5.74s/it]
15%|█▍ | 6522/43939 [10:30:51<59:38:22, 5.74s/it]
15%|█▍ | 6523/43939 [10:30:56<59:41:04, 5.74s/it]
15%|█▍ | 6524/43939 [10:31:02<59:38:26, 5.74s/it]
15%|█▍ | 6525/43939 [10:31:08<59:30:54, 5.73s/it]
15%|█▍ | 6526/43939 [10:31:14<59:24:51, 5.72s/it]
15%|█▍ | 6527/43939 [10:31:19<59:24:53, 5.72s/it]
15%|█▍ | 6528/43939 [10:31:25<59:15:37, 5.70s/it]
15%|█▍ | 6529/43939 [10:31:31<59:14:29, 5.70s/it]
15%|█▍ | 6530/43939 [10:31:36<59:10:01, 5.69s/it]
15%|█▍ | 6531/43939 [10:31:42<59:16:53, 5.71s/it]
15%|█▍ | 6532/43939 [10:31:48<59:13:11, 5.70s/it]
15%|█▍ | 6533/43939 [10:31:53<59:09:42, 5.69s/it]
15%|█▍ | 6534/43939 [10:31:59<59:06:11, 5.69s/it]
15%|█▍ | 6535/43939 [10:32:05<59:06:35, 5.69s/it]
15%|█▍ | 6536/43939 [10:32:10<59:06:05, 5.69s/it]
15%|█▍ | 6537/43939 [10:32:16<59:06:13, 5.69s/it]
15%|█▍ | 6538/43939 [10:32:22<59:17:56, 5.71s/it]
15%|█▍ | 6539/43939 [10:32:28<60:02:09, 5.78s/it]
15%|█▍ | 6540/43939 [10:32:34<59:43:07, 5.75s/it]
15%|█▍ | 6541/43939 [10:32:39<59:32:44, 5.73s/it]
15%|█▍ | 6542/43939 [10:32:45<59:30:38, 5.73s/it]
15%|█▍ | 6543/43939 [10:32:51<59:28:29, 5.73s/it]
15%|█▍ | 6544/43939 [10:32:56<59:26:37, 5.72s/it]
15%|█▍ | 6545/43939 [10:33:02<59:13:36, 5.70s/it]
15%|█▍ | 6546/43939 [10:33:08<59:06:56, 5.69s/it]
15%|█▍ | 6547/43939 [10:33:13<59:10:09, 5.70s/it]
15%|█▍ | 6548/43939 [10:33:19<59:07:31, 5.69s/it]
15%|█▍ | 6549/43939 [10:33:25<59:04:01, 5.69s/it]
15%|█▍ | 6550/43939 [10:33:31<59:11:56, 5.70s/it]
{'loss': 1.0512, 'grad_norm': 1.0880138874053955, 'learning_rate': 4.9634202945668476e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.51, 'epoch': 0.6}
15%|█▍ | 6550/43939 [10:33:31<59:11:56, 5.70s/it]
15%|█▍ | 6551/43939 [10:33:36<59:11:08, 5.70s/it]
15%|█▍ | 6552/43939 [10:33:42<59:17:19, 5.71s/it]
15%|█▍ | 6553/43939 [10:33:48<59:25:46, 5.72s/it]
15%|█▍ | 6554/43939 [10:33:53<59:25:04, 5.72s/it]
15%|█▍ | 6555/43939 [10:33:59<59:16:37, 5.71s/it]
15%|█▍ | 6556/43939 [10:34:05<59:15:27, 5.71s/it]
15%|█▍ | 6557/43939 [10:34:11<59:18:18, 5.71s/it]
15%|█▍ | 6558/43939 [10:34:16<59:16:06, 5.71s/it]
15%|█▍ | 6559/43939 [10:34:22<59:53:44, 5.77s/it]
15%|█▍ | 6560/43939 [10:34:28<59:40:09, 5.75s/it]
15%|█▍ | 6561/43939 [10:34:34<59:33:54, 5.74s/it]
15%|█▍ | 6562/43939 [10:34:39<59:29:25, 5.73s/it]
15%|█▍ | 6563/43939 [10:34:45<59:26:07, 5.72s/it]
15%|█▍ | 6564/43939 [10:34:51<59:14:46, 5.71s/it]
15%|█▍ | 6565/43939 [10:34:56<59:04:31, 5.69s/it]
15%|█▍ | 6566/43939 [10:35:02<59:07:41, 5.70s/it]
15%|█▍ | 6567/43939 [10:35:08<59:03:19, 5.69s/it]
15%|█▍ | 6568/43939 [10:35:13<59:07:07, 5.70s/it]
15%|█▍ | 6569/43939 [10:35:19<59:10:39, 5.70s/it]
15%|█▍ | 6570/43939 [10:35:25<59:10:17, 5.70s/it]
15%|█▍ | 6571/43939 [10:35:30<59:13:02, 5.70s/it]
15%|█▍ | 6572/43939 [10:35:36<59:11:29, 5.70s/it]
15%|█▍ | 6573/43939 [10:35:42<59:14:23, 5.71s/it]
15%|█▍ | 6574/43939 [10:35:48<59:13:01, 5.71s/it]
15%|█▍ | 6575/43939 [10:35:53<59:07:29, 5.70s/it]
15%|█▍ | 6576/43939 [10:35:59<59:02:11, 5.69s/it]
15%|█▍ | 6577/43939 [10:36:05<59:01:58, 5.69s/it]
15%|█▍ | 6578/43939 [10:36:10<59:04:26, 5.69s/it]
15%|█▍ | 6579/43939 [10:36:16<59:06:53, 5.70s/it]
15%|█▍ | 6580/43939 [10:36:22<59:09:03, 5.70s/it]
15%|█▍ | 6581/43939 [10:36:28<59:43:34, 5.76s/it]
15%|█▍ | 6582/43939 [10:36:33<59:34:04, 5.74s/it]
15%|█▍ | 6583/43939 [10:36:39<59:26:39, 5.73s/it]
15%|█▍ | 6584/43939 [10:36:45<59:24:19, 5.73s/it]
15%|█▍ | 6585/43939 [10:36:50<59:20:42, 5.72s/it]
15%|█▍ | 6586/43939 [10:36:56<59:22:14, 5.72s/it]
15%|█▍ | 6587/43939 [10:37:02<59:10:09, 5.70s/it]
15%|█▍ | 6588/43939 [10:37:08<59:06:39, 5.70s/it]
15%|█▍ | 6589/43939 [10:37:13<59:18:24, 5.72s/it]
15%|█▍ | 6590/43939 [10:37:19<59:19:03, 5.72s/it]
15%|█▌ | 6591/43939 [10:37:25<59:29:09, 5.73s/it][2025-11-29 04:41:38,617] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.36s/it][A
4%|▍ | 3/72 [00:05<02:12, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:50, 2.62s/it][A
11%|█ | 8/72 [00:19<02:50, 2.66s/it][A
12%|█▎ | 9/72 [00:21<02:49, 2.68s/it][A
14%|█▍ | 10/72 [00:24<02:47, 2.70s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.71s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.71s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:18, 2.72s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [01:00<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.72s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.72s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.72s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.72s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.72s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.72s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.72s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.72s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.72s/it][A
54%|█████▍ | 39/72 [01:43<01:29, 2.72s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.72s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.72s/it][A
60%|█████▉ | 43/72 [01:54<01:18, 2.72s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.72s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.72s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.72s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.72s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.72s/it][A
75%|███████▌ | 54/72 [02:24<00:48, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.72s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.72s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.72s/it][A
85%|████████▍ | 61/72 [02:43<00:29, 2.72s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.72s/it][A
88%|████████▊ | 63/72 [02:48<00:24, 2.72s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.71s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.71s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:07<00:05, 2.72s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.71s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A{'eval_loss': 1.0370653867721558, 'eval_runtime': 195.0721, 'eval_samples_per_second': 5.121, 'eval_steps_per_second': 0.369, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 0.6}
15%|█▌ | 6591/43939 [10:40:40<59:29:09, 5.73s/it]
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A
15%|█▌ | 6592/43939 [10:40:46<666:59:21, 64.29s/it]
15%|█▌ | 6593/43939 [10:40:51<484:36:56, 46.71s/it]
15%|█▌ | 6594/43939 [10:40:57<356:54:12, 34.40s/it]
15%|█▌ | 6595/43939 [10:41:03<267:29:42, 25.79s/it]
15%|█▌ | 6596/43939 [10:41:08<204:57:55, 19.76s/it]
15%|█▌ | 6597/43939 [10:41:14<161:12:18, 15.54s/it]
15%|█▌ | 6598/43939 [10:41:20<130:29:16, 12.58s/it]
15%|█▌ | 6599/43939 [10:41:26<109:04:39, 10.52s/it]
15%|█▌ | 6600/43939 [10:41:31<94:01:18, 9.07s/it]
{'loss': 1.0435, 'grad_norm': 1.089520812034607, 'learning_rate': 4.961708368757138e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 77.93, 'epoch': 0.6}
15%|█▌ | 6600/43939 [10:41:31<94:01:18, 9.07s/it]
15%|█▌ | 6601/43939 [10:41:37<83:27:50, 8.05s/it]
15%|█▌ | 6602/43939 [10:41:43<76:58:27, 7.42s/it]
15%|█▌ | 6603/43939 [10:41:49<71:38:11, 6.91s/it]
15%|█▌ | 6604/43939 [10:41:54<67:51:19, 6.54s/it]
15%|█▌ | 6605/43939 [10:42:00<65:11:19, 6.29s/it]
15%|█▌ | 6606/43939 [10:42:06<63:28:07, 6.12s/it]
15%|█▌ | 6607/43939 [10:42:11<62:07:38, 5.99s/it]
15%|█▌ | 6608/43939 [10:42:17<61:07:06, 5.89s/it]
15%|█▌ | 6609/43939 [10:42:23<60:27:39, 5.83s/it]
15%|█▌ | 6610/43939 [10:42:28<60:08:05, 5.80s/it]
15%|█▌ | 6611/43939 [10:42:34<59:50:56, 5.77s/it]
15%|█▌ | 6612/43939 [10:42:40<59:34:26, 5.75s/it]
15%|█▌ | 6613/43939 [10:42:46<59:36:26, 5.75s/it]
15%|█▌ | 6614/43939 [10:42:51<59:28:17, 5.74s/it]
15%|█▌ | 6615/43939 [10:42:57<59:17:58, 5.72s/it]
15%|█▌ | 6616/43939 [10:43:03<59:14:21, 5.71s/it]
15%|█▌ | 6617/43939 [10:43:08<59:21:05, 5.72s/it]
15%|█▌ | 6618/43939 [10:43:14<59:15:22, 5.72s/it]
15%|█▌ | 6619/43939 [10:43:20<59:11:50, 5.71s/it]
15%|█▌ | 6620/43939 [10:43:26<59:11:30, 5.71s/it]
15%|█▌ | 6621/43939 [10:43:31<59:01:22, 5.69s/it]
15%|█▌ | 6622/43939 [10:43:37<59:46:25, 5.77s/it]
15%|█▌ | 6623/43939 [10:43:43<59:32:59, 5.74s/it]
15%|█▌ | 6624/43939 [10:43:49<59:21:16, 5.73s/it]
15%|█▌ | 6625/43939 [10:43:54<59:20:35, 5.73s/it]
15%|█▌ | 6626/43939 [10:44:00<59:15:50, 5.72s/it]
15%|█▌ | 6627/43939 [10:44:06<59:10:59, 5.71s/it]
15%|█▌ | 6628/43939 [10:44:11<59:05:34, 5.70s/it]
15%|█▌ | 6629/43939 [10:44:17<59:12:32, 5.71s/it]
15%|█▌ | 6630/43939 [10:44:23<59:08:28, 5.71s/it]
15%|█▌ | 6631/43939 [10:44:28<58:59:37, 5.69s/it]
15%|█▌ | 6632/43939 [10:44:34<59:09:38, 5.71s/it]
15%|█▌ | 6633/43939 [10:44:40<59:13:09, 5.71s/it]
15%|█▌ | 6634/43939 [10:44:46<59:09:14, 5.71s/it]
15%|█▌ | 6635/43939 [10:44:51<59:08:56, 5.71s/it]
15%|█▌ | 6636/43939 [10:44:57<59:05:49, 5.70s/it]
15%|█▌ | 6637/43939 [10:45:03<58:58:13, 5.69s/it]
15%|█▌ | 6638/43939 [10:45:08<59:10:49, 5.71s/it]
15%|█▌ | 6639/43939 [10:45:14<59:05:22, 5.70s/it]
15%|█▌ | 6640/43939 [10:45:20<59:07:28, 5.71s/it]
15%|█▌ | 6641/43939 [10:45:25<59:07:23, 5.71s/it]
15%|█▌ | 6642/43939 [10:45:31<59:07:10, 5.71s/it]
15%|█▌ | 6643/43939 [10:45:37<59:06:25, 5.71s/it]
15%|█▌ | 6644/43939 [10:45:43<59:42:19, 5.76s/it]
15%|█▌ | 6645/43939 [10:45:48<59:28:16, 5.74s/it]
15%|█▌ | 6646/43939 [10:45:54<59:14:05, 5.72s/it]
15%|█▌ | 6647/43939 [10:46:00<59:09:45, 5.71s/it]
15%|█▌ | 6648/43939 [10:46:06<59:09:51, 5.71s/it]
15%|█▌ | 6649/43939 [10:46:11<59:11:49, 5.71s/it]
15%|█▌ | 6650/43939 [10:46:17<59:10:45, 5.71s/it]
{'loss': 1.0353, 'grad_norm': 0.904808759689331, 'learning_rate': 4.959957603712977e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.54, 'epoch': 0.61}
15%|█▌ | 6650/43939 [10:46:17<59:10:45, 5.71s/it]
15%|█▌ | 6651/43939 [10:46:23<59:10:32, 5.71s/it]
15%|█▌ | 6652/43939 [10:46:28<59:02:13, 5.70s/it]
15%|█▌ | 6653/43939 [10:46:34<59:02:01, 5.70s/it]
15%|█▌ | 6654/43939 [10:46:40<59:02:51, 5.70s/it]
15%|█▌ | 6655/43939 [10:46:46<59:08:39, 5.71s/it]
15%|█▌ | 6656/43939 [10:46:51<59:07:39, 5.71s/it]
15%|█▌ | 6657/43939 [10:46:57<59:11:07, 5.72s/it]
15%|█▌ | 6658/43939 [10:47:03<59:22:10, 5.73s/it]
15%|█▌ | 6659/43939 [10:47:08<59:12:44, 5.72s/it]
15%|█▌ | 6660/43939 [10:47:14<59:14:18, 5.72s/it]
15%|█▌ | 6661/43939 [10:47:20<59:09:18, 5.71s/it]
15%|█▌ | 6662/43939 [10:47:26<59:10:55, 5.72s/it]
15%|█▌ | 6663/43939 [10:47:31<59:12:01, 5.72s/it]
15%|█▌ | 6664/43939 [10:47:37<59:44:21, 5.77s/it]
15%|█▌ | 6665/43939 [10:47:43<59:27:21, 5.74s/it]
15%|█▌ | 6666/43939 [10:47:49<59:16:59, 5.73s/it]
15%|█▌ | 6667/43939 [10:47:54<59:06:54, 5.71s/it]
15%|█▌ | 6668/43939 [10:48:00<58:57:18, 5.69s/it]
15%|█▌ | 6669/43939 [10:48:06<59:00:01, 5.70s/it]
15%|█▌ | 6670/43939 [10:48:11<59:11:28, 5.72s/it]
15%|█▌ | 6671/43939 [10:48:17<59:09:09, 5.71s/it]
15%|█▌ | 6672/43939 [10:48:23<59:12:55, 5.72s/it]
15%|█▌ | 6673/43939 [10:48:29<59:16:46, 5.73s/it]
15%|█▌ | 6674/43939 [10:48:34<59:08:45, 5.71s/it]
15%|█▌ | 6675/43939 [10:48:40<58:56:41, 5.69s/it]
15%|█▌ | 6676/43939 [10:48:46<58:53:16, 5.69s/it]
15%|█▌ | 6677/43939 [10:48:51<59:05:22, 5.71s/it]
15%|█▌ | 6678/43939 [10:48:57<59:08:14, 5.71s/it]
15%|█▌ | 6679/43939 [10:49:03<59:05:41, 5.71s/it]
15%|█▌ | 6680/43939 [10:49:08<59:00:46, 5.70s/it]
15%|█▌ | 6681/43939 [10:49:14<59:01:13, 5.70s/it]
15%|█▌ | 6682/43939 [10:49:20<58:49:55, 5.68s/it]
15%|█▌ | 6683/43939 [10:49:25<59:00:09, 5.70s/it]
15%|█▌ | 6684/43939 [10:49:31<59:03:53, 5.71s/it]
15%|█▌ | 6685/43939 [10:49:37<59:56:41, 5.79s/it]
15%|█▌ | 6686/43939 [10:49:43<59:35:16, 5.76s/it]
15%|█▌ | 6687/43939 [10:49:49<59:27:51, 5.75s/it]
15%|█▌ | 6688/43939 [10:49:54<59:20:13, 5.73s/it]
15%|█▌ | 6689/43939 [10:50:00<59:11:11, 5.72s/it]
15%|█▌ | 6690/43939 [10:50:06<59:05:35, 5.71s/it]
15%|█▌ | 6691/43939 [10:50:11<59:08:17, 5.72s/it]
15%|█▌ | 6692/43939 [10:50:17<59:06:26, 5.71s/it]
15%|█▌ | 6693/43939 [10:50:23<59:07:17, 5.71s/it]
15%|█▌ | 6694/43939 [10:50:29<59:03:18, 5.71s/it]
15%|█▌ | 6695/43939 [10:50:34<59:08:27, 5.72s/it]
15%|█▌ | 6696/43939 [10:50:40<59:03:59, 5.71s/it]
15%|█▌ | 6697/43939 [10:50:46<59:06:32, 5.71s/it]
15%|█▌ | 6698/43939 [10:50:51<59:00:49, 5.70s/it]
15%|█▌ | 6699/43939 [10:50:57<58:52:18, 5.69s/it]
15%|█▌ | 6700/43939 [10:51:03<58:45:39, 5.68s/it]
{'loss': 1.0425, 'grad_norm': 1.1445884704589844, 'learning_rate': 4.9581680270567957e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.08, 'epoch': 0.61}
15%|█▌ | 6700/43939 [10:51:03<58:45:39, 5.68s/it]
15%|█▌ | 6701/43939 [10:51:08<58:55:49, 5.70s/it]
15%|█▌ | 6702/43939 [10:51:14<58:44:48, 5.68s/it]
15%|█▌ | 6703/43939 [10:51:20<58:54:07, 5.69s/it]
15%|█▌ | 6704/43939 [10:51:25<58:52:45, 5.69s/it]
15%|█▌ | 6705/43939 [10:51:31<58:50:23, 5.69s/it]
15%|█▌ | 6706/43939 [10:51:37<58:49:09, 5.69s/it]
15%|█▌ | 6707/43939 [10:51:43<59:39:36, 5.77s/it]
15%|█▌ | 6708/43939 [10:51:48<59:25:41, 5.75s/it]
15%|█▌ | 6709/43939 [10:51:54<59:23:17, 5.74s/it]
15%|█▌ | 6710/43939 [10:52:00<59:09:08, 5.72s/it]
15%|█▌ | 6711/43939 [10:52:06<59:06:12, 5.72s/it]
15%|█▌ | 6712/43939 [10:52:11<58:57:22, 5.70s/it]
15%|█▌ | 6713/43939 [10:52:17<58:58:11, 5.70s/it]
15%|█▌ | 6714/43939 [10:52:23<58:51:59, 5.69s/it]
15%|█▌ | 6715/43939 [10:52:28<59:06:48, 5.72s/it]
15%|█▌ | 6716/43939 [10:52:34<59:01:41, 5.71s/it]
15%|█▌ | 6717/43939 [10:52:40<59:03:50, 5.71s/it]
15%|█▌ | 6718/43939 [10:52:45<58:58:08, 5.70s/it]
15%|█▌ | 6719/43939 [10:52:51<58:58:40, 5.70s/it]
15%|█▌ | 6720/43939 [10:52:57<58:55:24, 5.70s/it]
15%|█▌ | 6721/43939 [10:53:03<58:50:01, 5.69s/it]
15%|█▌ | 6722/43939 [10:53:08<58:50:57, 5.69s/it]
15%|█▌ | 6723/43939 [10:53:14<59:02:57, 5.71s/it]
15%|█▌ | 6724/43939 [10:53:20<59:10:17, 5.72s/it]
15%|█▌ | 6725/43939 [10:53:25<58:54:11, 5.70s/it]
15%|█▌ | 6726/43939 [10:53:31<58:48:39, 5.69s/it]
15%|█▌ | 6727/43939 [10:53:37<58:55:20, 5.70s/it]
15%|█▌ | 6728/43939 [10:53:43<59:49:49, 5.79s/it]
15%|█▌ | 6729/43939 [10:53:48<59:28:34, 5.75s/it]
15%|█▌ | 6730/43939 [10:53:54<59:20:40, 5.74s/it]
15%|█▌ | 6731/43939 [10:54:00<59:12:18, 5.73s/it]
15%|█▌ | 6732/43939 [10:54:06<59:12:49, 5.73s/it]
15%|█▌ | 6733/43939 [10:54:11<59:18:28, 5.74s/it]
15%|█▌ | 6734/43939 [10:54:17<59:10:37, 5.73s/it]
15%|█▌ | 6735/43939 [10:54:23<58:58:21, 5.71s/it]
15%|█▌ | 6736/43939 [10:54:28<59:02:10, 5.71s/it]
15%|█▌ | 6737/43939 [10:54:34<58:53:11, 5.70s/it]
15%|█▌ | 6738/43939 [10:54:40<58:47:59, 5.69s/it]
15%|█▌ | 6739/43939 [10:54:45<58:48:34, 5.69s/it]
15%|█▌ | 6740/43939 [10:54:51<58:49:27, 5.69s/it]
15%|█▌ | 6741/43939 [10:54:57<58:47:37, 5.69s/it]
15%|█▌ | 6742/43939 [10:55:03<58:48:57, 5.69s/it]
15%|█▌ | 6743/43939 [10:55:08<58:54:55, 5.70s/it]
15%|█▌ | 6744/43939 [10:55:14<58:50:08, 5.69s/it]
15%|█▌ | 6745/43939 [10:55:20<58:46:03, 5.69s/it]
15%|█▌ | 6746/43939 [10:55:25<58:54:21, 5.70s/it]
15%|█▌ | 6747/43939 [10:55:31<58:52:02, 5.70s/it]
15%|█▌ | 6748/43939 [10:55:37<59:32:16, 5.76s/it]
15%|█▌ | 6749/43939 [10:55:43<59:22:54, 5.75s/it]
15%|█▌ | 6750/43939 [10:55:48<59:21:35, 5.75s/it]
{'loss': 1.0563, 'grad_norm': 1.157138466835022, 'learning_rate': 4.956339667023373e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 511.61, 'epoch': 0.61}
15%|█▌ | 6750/43939 [10:55:48<59:21:35, 5.75s/it]
15%|█▌ | 6751/43939 [10:55:54<59:28:57, 5.76s/it]
15%|█▌ | 6752/43939 [10:56:00<59:18:50, 5.74s/it]
15%|█▌ | 6753/43939 [10:56:06<59:02:58, 5.72s/it]
15%|█▌ | 6754/43939 [10:56:11<58:59:55, 5.71s/it]
15%|█▌ | 6755/43939 [10:56:17<58:56:39, 5.71s/it]
15%|█▌ | 6756/43939 [10:56:23<58:54:10, 5.70s/it]
15%|█▌ | 6757/43939 [10:56:28<58:51:51, 5.70s/it]
15%|█▌ | 6758/43939 [10:56:34<59:01:44, 5.72s/it]
15%|█▌ | 6759/43939 [10:56:40<58:56:36, 5.71s/it]
15%|█▌ | 6760/43939 [10:56:45<58:55:16, 5.71s/it]
15%|█▌ | 6761/43939 [10:56:51<58:55:49, 5.71s/it]
15%|█▌ | 6762/43939 [10:56:57<58:50:13, 5.70s/it]
15%|█▌ | 6763/43939 [10:57:03<58:56:22, 5.71s/it]
15%|█▌ | 6764/43939 [10:57:08<59:03:33, 5.72s/it]
15%|█▌ | 6765/43939 [10:57:14<58:58:25, 5.71s/it]
15%|█▌ | 6766/43939 [10:57:20<59:02:51, 5.72s/it]
15%|█▌ | 6767/43939 [10:57:25<59:00:58, 5.72s/it]
15%|█▌ | 6768/43939 [10:57:31<58:57:59, 5.71s/it]
15%|█▌ | 6769/43939 [10:57:37<59:32:09, 5.77s/it]
15%|█▌ | 6770/43939 [10:57:43<59:18:46, 5.74s/it]
15%|█▌ | 6771/43939 [10:57:48<59:08:33, 5.73s/it]
15%|█▌ | 6772/43939 [10:57:54<59:02:14, 5.72s/it]
15%|█▌ | 6773/43939 [10:58:00<59:10:35, 5.73s/it]
15%|█▌ | 6774/43939 [10:58:06<59:04:33, 5.72s/it]
15%|█▌ | 6775/43939 [10:58:11<59:03:03, 5.72s/it]
15%|█▌ | 6776/43939 [10:58:17<58:57:39, 5.71s/it]
15%|█▌ | 6777/43939 [10:58:23<58:50:35, 5.70s/it]
15%|█▌ | 6778/43939 [10:58:28<58:50:55, 5.70s/it]
15%|█▌ | 6779/43939 [10:58:34<58:47:35, 5.70s/it]
15%|█▌ | 6780/43939 [10:58:40<58:52:42, 5.70s/it]
15%|█▌ | 6781/43939 [10:58:46<58:48:43, 5.70s/it]
15%|█▌ | 6782/43939 [10:58:51<58:52:53, 5.70s/it]
15%|█▌ | 6783/43939 [10:58:57<58:50:32, 5.70s/it]
15%|█▌ | 6784/43939 [10:59:03<58:51:27, 5.70s/it]
15%|█▌ | 6785/43939 [10:59:08<58:54:01, 5.71s/it]
15%|█▌ | 6786/43939 [10:59:14<58:56:06, 5.71s/it]
15%|█▌ | 6787/43939 [10:59:20<58:56:17, 5.71s/it]
15%|█▌ | 6788/43939 [10:59:25<58:57:32, 5.71s/it]
15%|█▌ | 6789/43939 [10:59:31<58:59:13, 5.72s/it]
15%|█▌ | 6790/43939 [10:59:37<59:02:22, 5.72s/it]
15%|█▌ | 6791/43939 [10:59:43<59:36:44, 5.78s/it]
15%|█▌ | 6792/43939 [10:59:49<59:22:01, 5.75s/it]
15%|█▌ | 6793/43939 [10:59:54<59:13:23, 5.74s/it]
15%|█▌ | 6794/43939 [11:00:00<59:07:49, 5.73s/it]
15%|█▌ | 6795/43939 [11:00:06<58:59:45, 5.72s/it]
15%|█▌ | 6796/43939 [11:00:11<59:04:25, 5.73s/it]
15%|█▌ | 6797/43939 [11:00:17<59:06:42, 5.73s/it]
15%|█▌ | 6798/43939 [11:00:23<58:59:02, 5.72s/it]
15%|█▌ | 6799/43939 [11:00:29<58:57:55, 5.72s/it]
15%|█▌ | 6800/43939 [11:00:34<58:56:29, 5.71s/it]
{'loss': 1.0364, 'grad_norm': 0.8729771375656128, 'learning_rate': 4.954472552459383e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.17, 'epoch': 0.62}
15%|█▌ | 6800/43939 [11:00:34<58:56:29, 5.71s/it]
15%|█▌ | 6801/43939 [11:00:40<59:01:08, 5.72s/it]
15%|█▌ | 6802/43939 [11:00:46<58:57:41, 5.72s/it]
15%|█▌ | 6803/43939 [11:00:51<58:54:06, 5.71s/it]
15%|█▌ | 6804/43939 [11:00:57<58:52:05, 5.71s/it]
15%|█▌ | 6805/43939 [11:01:03<59:00:40, 5.72s/it]
15%|█▌ | 6806/43939 [11:01:09<58:52:05, 5.71s/it]
15%|█▌ | 6807/43939 [11:01:14<59:03:17, 5.73s/it]
15%|█▌ | 6808/43939 [11:01:20<58:59:56, 5.72s/it]
15%|█▌ | 6809/43939 [11:01:26<58:49:01, 5.70s/it]
15%|█▌ | 6810/43939 [11:01:31<58:48:17, 5.70s/it]
16%|█▌ | 6811/43939 [11:01:37<58:46:23, 5.70s/it]
16%|█▌ | 6812/43939 [11:01:43<59:19:33, 5.75s/it]
16%|█▌ | 6813/43939 [11:01:49<59:05:25, 5.73s/it]
16%|█▌ | 6814/43939 [11:01:54<58:54:23, 5.71s/it]
16%|█▌ | 6815/43939 [11:02:00<58:59:44, 5.72s/it]
16%|█▌ | 6816/43939 [11:02:06<58:52:13, 5.71s/it]
16%|█▌ | 6817/43939 [11:02:11<58:58:43, 5.72s/it]
16%|█▌ | 6818/43939 [11:02:17<58:55:25, 5.71s/it]
16%|█▌ | 6819/43939 [11:02:23<58:49:01, 5.70s/it]
16%|█▌ | 6820/43939 [11:02:29<58:48:16, 5.70s/it]
16%|█▌ | 6821/43939 [11:02:34<58:43:48, 5.70s/it]
16%|█▌ | 6822/43939 [11:02:40<58:41:34, 5.69s/it]
16%|█▌ | 6823/43939 [11:02:46<58:50:45, 5.71s/it]
16%|█▌ | 6824/43939 [11:02:51<58:46:51, 5.70s/it]
16%|█▌ | 6825/43939 [11:02:57<58:42:40, 5.69s/it]
16%|█▌ | 6826/43939 [11:03:03<58:50:48, 5.71s/it]
16%|█▌ | 6827/43939 [11:03:08<58:51:03, 5.71s/it]
16%|█▌ | 6828/43939 [11:03:14<58:59:10, 5.72s/it]
16%|█▌ | 6829/43939 [11:03:20<58:52:48, 5.71s/it]
16%|█▌ | 6830/43939 [11:03:26<58:48:09, 5.70s/it]
16%|█▌ | 6831/43939 [11:03:31<58:43:30, 5.70s/it]
16%|█▌ | 6832/43939 [11:03:37<59:26:53, 5.77s/it]
16%|█▌ | 6833/43939 [11:03:43<59:12:50, 5.74s/it]
16%|█▌ | 6834/43939 [11:03:49<59:05:50, 5.73s/it]
16%|█▌ | 6835/43939 [11:03:54<58:59:11, 5.72s/it]
16%|█▌ | 6836/43939 [11:04:00<58:52:09, 5.71s/it]
16%|█▌ | 6837/43939 [11:04:06<58:53:08, 5.71s/it]
16%|█▌ | 6838/43939 [11:04:11<58:52:03, 5.71s/it]
16%|█▌ | 6839/43939 [11:04:17<58:39:25, 5.69s/it]
16%|█▌ | 6840/43939 [11:04:23<58:38:30, 5.69s/it]
16%|█▌ | 6841/43939 [11:04:28<58:40:52, 5.69s/it]
16%|█▌ | 6842/43939 [11:04:34<58:40:14, 5.69s/it]
16%|█▌ | 6843/43939 [11:04:40<58:38:47, 5.69s/it]
16%|█▌ | 6844/43939 [11:04:45<58:34:01, 5.68s/it]
16%|█▌ | 6845/43939 [11:04:51<58:33:36, 5.68s/it]
16%|█▌ | 6846/43939 [11:04:57<58:37:33, 5.69s/it]
16%|█▌ | 6847/43939 [11:05:03<58:35:55, 5.69s/it]
16%|█▌ | 6848/43939 [11:05:08<58:35:54, 5.69s/it]
16%|█▌ | 6849/43939 [11:05:14<58:31:07, 5.68s/it]
16%|█▌ | 6850/43939 [11:05:20<58:30:00, 5.68s/it]
{'loss': 1.0605, 'grad_norm': 1.065958857536316, 'learning_rate': 4.9525667128229474e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.49, 'epoch': 0.62}
16%|█▌ | 6850/43939 [11:05:20<58:30:00, 5.68s/it]
16%|█▌ | 6851/43939 [11:05:25<58:32:09, 5.68s/it]
16%|█▌ | 6852/43939 [11:05:31<58:38:57, 5.69s/it]
16%|█▌ | 6853/43939 [11:05:37<58:45:19, 5.70s/it]
16%|█▌ | 6854/43939 [11:05:43<59:25:59, 5.77s/it]
16%|█▌ | 6855/43939 [11:05:48<59:19:56, 5.76s/it]
16%|█▌ | 6856/43939 [11:05:54<59:06:57, 5.74s/it]
16%|█▌ | 6857/43939 [11:06:00<58:56:56, 5.72s/it]
16%|█▌ | 6858/43939 [11:06:05<58:56:56, 5.72s/it]
16%|█▌ | 6859/43939 [11:06:11<58:48:35, 5.71s/it]
16%|█▌ | 6860/43939 [11:06:17<58:49:21, 5.71s/it]
16%|█▌ | 6861/43939 [11:06:23<58:48:33, 5.71s/it]
16%|█▌ | 6862/43939 [11:06:28<58:41:49, 5.70s/it]
16%|█▌ | 6863/43939 [11:06:34<58:49:24, 5.71s/it]
16%|█▌ | 6864/43939 [11:06:40<58:47:42, 5.71s/it]
16%|█▌ | 6865/43939 [11:06:45<58:51:28, 5.72s/it]
16%|█▌ | 6866/43939 [11:06:51<58:47:40, 5.71s/it]
16%|█▌ | 6867/43939 [11:06:57<58:55:13, 5.72s/it]
16%|█▌ | 6868/43939 [11:07:03<58:46:11, 5.71s/it]
16%|█▌ | 6869/43939 [11:07:08<58:36:33, 5.69s/it]
16%|█▌ | 6870/43939 [11:07:14<58:33:57, 5.69s/it]
16%|█▌ | 6871/43939 [11:07:20<58:34:06, 5.69s/it]
16%|█▌ | 6872/43939 [11:07:25<58:43:19, 5.70s/it]
16%|█▌ | 6873/43939 [11:07:31<58:42:14, 5.70s/it]
16%|█▌ | 6874/43939 [11:07:37<59:33:12, 5.78s/it]
16%|█▌ | 6875/43939 [11:07:43<59:26:04, 5.77s/it]
16%|█▌ | 6876/43939 [11:07:48<59:13:14, 5.75s/it]
16%|█▌ | 6877/43939 [11:07:54<59:04:41, 5.74s/it]
16%|█▌ | 6878/43939 [11:08:00<59:03:23, 5.74s/it]
16%|█▌ | 6879/43939 [11:08:06<59:06:51, 5.74s/it]
16%|█▌ | 6880/43939 [11:08:11<58:53:36, 5.72s/it]
16%|█▌ | 6881/43939 [11:08:17<58:48:22, 5.71s/it]
16%|█▌ | 6882/43939 [11:08:23<58:45:00, 5.71s/it]
16%|█▌ | 6883/43939 [11:08:28<58:56:45, 5.73s/it]
16%|█▌ | 6884/43939 [11:08:34<58:43:46, 5.71s/it]
16%|█▌ | 6885/43939 [11:08:40<58:47:16, 5.71s/it]
16%|█▌ | 6886/43939 [11:08:46<58:46:49, 5.71s/it]
16%|█▌ | 6887/43939 [11:08:51<58:49:55, 5.72s/it]
16%|█▌ | 6888/43939 [11:08:57<58:49:15, 5.72s/it]
16%|█▌ | 6889/43939 [11:09:03<58:49:14, 5.72s/it]
16%|█▌ | 6890/43939 [11:09:08<58:46:57, 5.71s/it]
16%|█▌ | 6891/43939 [11:09:14<58:51:37, 5.72s/it]
16%|█▌ | 6892/43939 [11:09:20<58:48:29, 5.71s/it]
16%|█▌ | 6893/43939 [11:09:26<58:45:47, 5.71s/it]
16%|█▌ | 6894/43939 [11:09:31<58:42:19, 5.70s/it]
16%|█▌ | 6895/43939 [11:09:37<59:24:42, 5.77s/it]
16%|█▌ | 6896/43939 [11:09:43<59:11:02, 5.75s/it]
16%|█▌ | 6897/43939 [11:09:49<58:56:51, 5.73s/it]
16%|█▌ | 6898/43939 [11:09:54<58:56:54, 5.73s/it]
16%|█▌ | 6899/43939 [11:10:00<58:45:16, 5.71s/it]
16%|█▌ | 6900/43939 [11:10:06<58:46:53, 5.71s/it]
{'loss': 1.0743, 'grad_norm': 0.9055165648460388, 'learning_rate': 4.950622178183164e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.48, 'epoch': 0.63}
16%|█▌ | 6900/43939 [11:10:06<58:46:53, 5.71s/it]
16%|█▌ | 6901/43939 [11:10:11<58:51:57, 5.72s/it]
16%|█▌ | 6902/43939 [11:10:17<58:49:10, 5.72s/it]
16%|█▌ | 6903/43939 [11:10:23<58:42:29, 5.71s/it]
16%|█▌ | 6904/43939 [11:10:29<58:53:35, 5.72s/it]
16%|█▌ | 6905/43939 [11:10:34<58:53:38, 5.72s/it]
16%|█▌ | 6906/43939 [11:10:40<58:46:29, 5.71s/it]
16%|█▌ | 6907/43939 [11:10:46<58:47:01, 5.71s/it]
16%|█▌ | 6908/43939 [11:10:51<58:42:10, 5.71s/it]
16%|█▌ | 6909/43939 [11:10:57<58:44:52, 5.71s/it]
16%|█▌ | 6910/43939 [11:11:03<58:47:38, 5.72s/it]
16%|█▌ | 6911/43939 [11:11:09<58:46:42, 5.71s/it]
16%|█▌ | 6912/43939 [11:11:14<58:42:33, 5.71s/it]
16%|█▌ | 6913/43939 [11:11:20<58:52:22, 5.72s/it]
16%|█▌ | 6914/43939 [11:11:26<58:57:19, 5.73s/it]
16%|█▌ | 6915/43939 [11:11:31<58:49:26, 5.72s/it]
16%|█▌ | 6916/43939 [11:11:37<58:42:23, 5.71s/it]
16%|█▌ | 6917/43939 [11:11:43<59:19:09, 5.77s/it]
16%|█▌ | 6918/43939 [11:11:49<59:02:23, 5.74s/it]
16%|█▌ | 6919/43939 [11:11:54<58:54:02, 5.73s/it]
16%|█▌ | 6920/43939 [11:12:00<58:44:23, 5.71s/it]
16%|█▌ | 6921/43939 [11:12:06<58:43:55, 5.71s/it]
16%|█▌ | 6922/43939 [11:12:11<58:40:20, 5.71s/it]
16%|█▌ | 6923/43939 [11:12:17<58:45:51, 5.72s/it]
16%|█▌ | 6924/43939 [11:12:23<58:47:40, 5.72s/it]
16%|█▌ | 6925/43939 [11:12:29<58:49:25, 5.72s/it]
16%|█▌ | 6926/43939 [11:12:34<58:48:41, 5.72s/it]
16%|█▌ | 6927/43939 [11:12:40<58:45:27, 5.72s/it]
16%|█▌ | 6928/43939 [11:12:46<58:48:27, 5.72s/it]
16%|█▌ | 6929/43939 [11:12:52<58:47:41, 5.72s/it]
16%|█▌ | 6930/43939 [11:12:57<58:43:39, 5.71s/it]
16%|█▌ | 6931/43939 [11:13:03<58:46:58, 5.72s/it]
16%|█▌ | 6932/43939 [11:13:09<58:49:57, 5.72s/it]
16%|█▌ | 6933/43939 [11:13:14<58:51:01, 5.73s/it]
16%|█▌ | 6934/43939 [11:13:20<58:54:00, 5.73s/it]
16%|█▌ | 6935/43939 [11:13:26<58:45:47, 5.72s/it]
16%|█▌ | 6936/43939 [11:13:32<58:36:56, 5.70s/it]
16%|█▌ | 6937/43939 [11:13:37<59:15:58, 5.77s/it]
16%|█▌ | 6938/43939 [11:13:43<59:01:56, 5.74s/it]
16%|█▌ | 6939/43939 [11:13:49<58:48:54, 5.72s/it]
16%|█▌ | 6940/43939 [11:13:55<58:46:36, 5.72s/it]
16%|█▌ | 6941/43939 [11:14:00<58:44:48, 5.72s/it]
16%|█▌ | 6942/43939 [11:14:06<58:51:23, 5.73s/it]
16%|█▌ | 6943/43939 [11:14:12<58:46:34, 5.72s/it]
16%|█▌ | 6944/43939 [11:14:17<58:37:57, 5.71s/it]
16%|█▌ | 6945/43939 [11:14:23<58:38:40, 5.71s/it]
16%|█▌ | 6946/43939 [11:14:29<58:35:14, 5.70s/it]
16%|█▌ | 6947/43939 [11:14:34<58:29:37, 5.69s/it]
16%|█▌ | 6948/43939 [11:14:40<58:31:28, 5.70s/it]
16%|█▌ | 6949/43939 [11:14:46<58:33:48, 5.70s/it]
16%|█▌ | 6950/43939 [11:14:52<58:38:13, 5.71s/it]
{'loss': 1.035, 'grad_norm': 1.157254934310913, 'learning_rate': 4.948638979219637e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.23, 'epoch': 0.63}
16%|█▌ | 6950/43939 [11:14:52<58:38:13, 5.71s/it]
16%|█▌ | 6951/43939 [11:14:57<58:32:37, 5.70s/it]
16%|█▌ | 6952/43939 [11:15:03<58:31:36, 5.70s/it]
16%|█▌ | 6953/43939 [11:15:09<58:34:39, 5.70s/it]
16%|█▌ | 6954/43939 [11:15:14<58:50:02, 5.73s/it]
16%|█▌ | 6955/43939 [11:15:20<58:47:47, 5.72s/it]
16%|█▌ | 6956/43939 [11:15:26<58:36:09, 5.70s/it]
16%|█▌ | 6957/43939 [11:15:31<58:24:28, 5.69s/it]
16%|█▌ | 6958/43939 [11:15:37<58:57:18, 5.74s/it]
16%|█▌ | 6959/43939 [11:15:43<58:48:40, 5.73s/it]
16%|█▌ | 6960/43939 [11:15:49<58:45:07, 5.72s/it]
16%|█▌ | 6961/43939 [11:15:54<58:40:01, 5.71s/it]
16%|█▌ | 6962/43939 [11:16:00<58:39:02, 5.71s/it]
16%|█▌ | 6963/43939 [11:16:06<58:33:20, 5.70s/it]
16%|█▌ | 6964/43939 [11:16:11<58:26:35, 5.69s/it]
16%|█▌ | 6965/43939 [11:16:17<58:26:21, 5.69s/it]
16%|█▌ | 6966/43939 [11:16:23<58:29:49, 5.70s/it]
16%|█▌ | 6967/43939 [11:16:28<58:21:21, 5.68s/it]
16%|█▌ | 6968/43939 [11:16:34<58:33:04, 5.70s/it]
16%|█▌ | 6969/43939 [11:16:40<58:29:17, 5.70s/it]
16%|█▌ | 6970/43939 [11:16:46<58:31:52, 5.70s/it]
16%|█▌ | 6971/43939 [11:16:51<58:27:29, 5.69s/it]
16%|█▌ | 6972/43939 [11:16:57<58:39:04, 5.71s/it]
16%|█▌ | 6973/43939 [11:17:03<58:36:20, 5.71s/it]
16%|█▌ | 6974/43939 [11:17:08<58:32:42, 5.70s/it]
16%|█▌ | 6975/43939 [11:17:14<58:34:42, 5.71s/it]
16%|█▌ | 6976/43939 [11:17:20<58:25:19, 5.69s/it]
16%|█▌ | 6977/43939 [11:17:26<58:32:08, 5.70s/it]
16%|█▌ | 6978/43939 [11:17:31<58:31:00, 5.70s/it]
16%|█▌ | 6979/43939 [11:17:37<58:27:39, 5.69s/it]
16%|█▌ | 6980/43939 [11:17:43<59:10:23, 5.76s/it]
16%|█▌ | 6981/43939 [11:17:49<59:08:07, 5.76s/it]
16%|█▌ | 6982/43939 [11:17:54<58:47:05, 5.73s/it]
16%|█▌ | 6983/43939 [11:18:00<58:40:41, 5.72s/it]
16%|█▌ | 6984/43939 [11:18:06<58:46:12, 5.73s/it]
16%|█▌ | 6985/43939 [11:18:11<58:37:51, 5.71s/it]
16%|█▌ | 6986/43939 [11:18:17<58:41:08, 5.72s/it]
16%|█▌ | 6987/43939 [11:18:23<58:43:22, 5.72s/it]
16%|█▌ | 6988/43939 [11:18:29<58:38:55, 5.71s/it]
16%|█▌ | 6989/43939 [11:18:34<58:32:15, 5.70s/it]
16%|█▌ | 6990/43939 [11:18:40<58:30:46, 5.70s/it]
16%|█▌ | 6991/43939 [11:18:46<58:31:43, 5.70s/it]
16%|█▌ | 6992/43939 [11:18:51<58:27:22, 5.70s/it]
16%|█▌ | 6993/43939 [11:18:57<58:23:37, 5.69s/it]
16%|█▌ | 6994/43939 [11:19:03<58:20:47, 5.69s/it]
16%|█▌ | 6995/43939 [11:19:08<58:29:29, 5.70s/it]
16%|█▌ | 6996/43939 [11:19:14<58:31:10, 5.70s/it]
16%|█▌ | 6997/43939 [11:19:20<58:30:37, 5.70s/it]
16%|█▌ | 6998/43939 [11:19:25<58:28:26, 5.70s/it]
16%|█▌ | 6999/43939 [11:19:31<58:29:11, 5.70s/it]
16%|█▌ | 7000/43939 [11:19:37<58:22:50, 5.69s/it]
{'loss': 1.0365, 'grad_norm': 1.0744335651397705, 'learning_rate': 4.9466171472219904e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.74, 'epoch': 0.64}
16%|█▌ | 7000/43939 [11:19:37<58:22:50, 5.69s/it]
16%|█▌ | 7001/43939 [11:19:43<59:05:14, 5.76s/it]
16%|█▌ | 7002/43939 [11:19:48<58:55:19, 5.74s/it]
16%|█▌ | 7003/43939 [11:19:54<58:55:56, 5.74s/it]
16%|█▌ | 7004/43939 [11:20:00<58:40:35, 5.72s/it]
16%|█▌ | 7005/43939 [11:20:06<58:46:10, 5.73s/it]
16%|█▌ | 7006/43939 [11:20:11<58:38:46, 5.72s/it]
16%|█▌ | 7007/43939 [11:20:17<58:34:58, 5.71s/it]
16%|█▌ | 7008/43939 [11:20:23<58:26:58, 5.70s/it]
16%|█▌ | 7009/43939 [11:20:28<58:21:04, 5.69s/it]
16%|█▌ | 7010/43939 [11:20:34<58:34:15, 5.71s/it]
16%|█▌ | 7011/43939 [11:20:40<58:35:19, 5.71s/it]
16%|█▌ | 7012/43939 [11:20:45<58:27:36, 5.70s/it]
16%|█▌ | 7013/43939 [11:20:51<58:32:26, 5.71s/it]
16%|█▌ | 7014/43939 [11:20:57<58:33:55, 5.71s/it]
16%|█▌ | 7015/43939 [11:21:03<58:36:18, 5.71s/it]
16%|█▌ | 7016/43939 [11:21:08<58:39:41, 5.72s/it]
16%|█▌ | 7017/43939 [11:21:14<58:34:40, 5.71s/it]
16%|█▌ | 7018/43939 [11:21:20<58:34:18, 5.71s/it]
16%|█▌ | 7019/43939 [11:21:25<58:24:05, 5.69s/it]
16%|█▌ | 7020/43939 [11:21:31<58:26:42, 5.70s/it]
16%|█▌ | 7021/43939 [11:21:37<58:33:00, 5.71s/it]
16%|█▌ | 7022/43939 [11:21:43<59:13:54, 5.78s/it]
16%|█▌ | 7023/43939 [11:21:49<59:04:44, 5.76s/it]
16%|█▌ | 7024/43939 [11:21:54<58:53:09, 5.74s/it]
16%|█▌ | 7025/43939 [11:22:00<58:35:41, 5.71s/it]
16%|█▌ | 7026/43939 [11:22:06<58:31:25, 5.71s/it]
16%|█▌ | 7027/43939 [11:22:11<58:27:46, 5.70s/it]
16%|█▌ | 7028/43939 [11:22:17<58:35:08, 5.71s/it]
16%|█▌ | 7029/43939 [11:22:23<58:34:33, 5.71s/it]
16%|█▌ | 7030/43939 [11:22:28<58:31:29, 5.71s/it]
16%|█▌ | 7031/43939 [11:22:34<58:29:51, 5.71s/it]
16%|█▌ | 7032/43939 [11:22:40<58:43:09, 5.73s/it]
16%|█▌ | 7033/43939 [11:22:46<58:39:06, 5.72s/it]
16%|█▌ | 7034/43939 [11:22:51<58:32:27, 5.71s/it]
16%|█▌ | 7035/43939 [11:22:57<58:27:45, 5.70s/it]
16%|█▌ | 7036/43939 [11:23:03<58:20:46, 5.69s/it]
16%|█▌ | 7037/43939 [11:23:08<58:20:36, 5.69s/it]
16%|█▌ | 7038/43939 [11:23:14<58:22:18, 5.69s/it]
16%|█▌ | 7039/43939 [11:23:20<58:14:08, 5.68s/it]
16%|█▌ | 7040/43939 [11:23:25<58:23:29, 5.70s/it]
16%|█▌ | 7041/43939 [11:23:31<58:32:45, 5.71s/it]
16%|█▌ | 7042/43939 [11:23:37<58:28:13, 5.70s/it]
16%|█▌ | 7043/43939 [11:23:43<59:27:42, 5.80s/it]
16%|█▌ | 7044/43939 [11:23:49<59:09:10, 5.77s/it]
16%|█▌ | 7045/43939 [11:23:54<58:54:26, 5.75s/it]
16%|█▌ | 7046/43939 [11:24:00<58:46:58, 5.74s/it]
16%|█▌ | 7047/43939 [11:24:06<58:42:45, 5.73s/it]
16%|█▌ | 7048/43939 [11:24:11<58:37:10, 5.72s/it]
16%|█▌ | 7049/43939 [11:24:17<58:38:09, 5.72s/it]
16%|█▌ | 7050/43939 [11:24:23<58:43:23, 5.73s/it]
{'loss': 1.0769, 'grad_norm': 0.9064004421234131, 'learning_rate': 4.944556714089374e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.16, 'epoch': 0.64}
16%|█▌ | 7050/43939 [11:24:23<58:43:23, 5.73s/it]
16%|█▌ | 7051/43939 [11:24:29<58:46:54, 5.74s/it]
16%|█▌ | 7052/43939 [11:24:34<58:42:49, 5.73s/it]
16%|█▌ | 7053/43939 [11:24:40<58:37:16, 5.72s/it]
16%|█▌ | 7054/43939 [11:24:46<58:24:43, 5.70s/it]
16%|█▌ | 7055/43939 [11:24:51<58:33:39, 5.72s/it]
16%|█▌ | 7056/43939 [11:24:57<58:25:02, 5.70s/it]
16%|█▌ | 7057/43939 [11:25:03<58:21:34, 5.70s/it]
16%|█▌ | 7058/43939 [11:25:09<58:26:07, 5.70s/it]
16%|█▌ | 7059/43939 [11:25:14<58:32:57, 5.72s/it]
16%|█▌ | 7060/43939 [11:25:20<58:38:35, 5.72s/it]
16%|█▌ | 7061/43939 [11:25:26<58:29:26, 5.71s/it]
16%|█▌ | 7062/43939 [11:25:31<58:20:56, 5.70s/it]
16%|█▌ | 7063/43939 [11:25:37<58:20:43, 5.70s/it]
16%|█▌ | 7064/43939 [11:25:43<59:11:41, 5.78s/it]
16%|█▌ | 7065/43939 [11:25:49<58:56:10, 5.75s/it]
16%|█▌ | 7066/43939 [11:25:54<58:42:20, 5.73s/it]
16%|█▌ | 7067/43939 [11:26:00<58:28:00, 5.71s/it]
16%|█▌ | 7068/43939 [11:26:06<58:32:51, 5.72s/it]
16%|█▌ | 7069/43939 [11:26:11<58:21:31, 5.70s/it]
16%|█▌ | 7070/43939 [11:26:17<58:20:00, 5.70s/it]
16%|█▌ | 7071/43939 [11:26:23<58:20:37, 5.70s/it]
16%|█▌ | 7072/43939 [11:26:29<58:23:24, 5.70s/it]
16%|█▌ | 7073/43939 [11:26:34<58:20:17, 5.70s/it]
16%|█▌ | 7074/43939 [11:26:40<58:19:50, 5.70s/it]
16%|█▌ | 7075/43939 [11:26:46<58:19:20, 5.70s/it]
16%|█▌ | 7076/43939 [11:26:51<58:21:31, 5.70s/it]
16%|█▌ | 7077/43939 [11:26:57<58:25:37, 5.71s/it]
16%|█▌ | 7078/43939 [11:27:03<58:21:23, 5.70s/it]
16%|█▌ | 7079/43939 [11:27:08<58:15:36, 5.69s/it]
16%|█▌ | 7080/43939 [11:27:14<58:10:02, 5.68s/it]
16%|█▌ | 7081/43939 [11:27:20<58:07:37, 5.68s/it]
16%|█▌ | 7082/43939 [11:27:25<58:00:27, 5.67s/it]
16%|█▌ | 7083/43939 [11:27:31<58:04:20, 5.67s/it]
16%|█▌ | 7084/43939 [11:27:37<58:48:45, 5.74s/it]
16%|█▌ | 7085/43939 [11:27:43<58:45:02, 5.74s/it]
16%|█▌ | 7086/43939 [11:27:48<58:39:27, 5.73s/it]
16%|█▌ | 7087/43939 [11:27:54<58:25:41, 5.71s/it]
16%|█▌ | 7088/43939 [11:28:00<58:20:41, 5.70s/it]
16%|█▌ | 7089/43939 [11:28:05<58:28:06, 5.71s/it]
16%|█▌ | 7090/43939 [11:28:11<58:20:14, 5.70s/it]
16%|█▌ | 7091/43939 [11:28:17<58:18:58, 5.70s/it]
16%|█▌ | 7092/43939 [11:28:23<58:30:31, 5.72s/it]
16%|█▌ | 7093/43939 [11:28:28<58:27:02, 5.71s/it]
16%|█▌ | 7094/43939 [11:28:34<58:26:26, 5.71s/it]
16%|█▌ | 7095/43939 [11:28:40<58:24:47, 5.71s/it]
16%|█▌ | 7096/43939 [11:28:45<58:20:26, 5.70s/it]
16%|█▌ | 7097/43939 [11:28:51<58:19:41, 5.70s/it]
16%|█▌ | 7098/43939 [11:28:57<58:15:22, 5.69s/it]
16%|█▌ | 7099/43939 [11:29:02<58:19:01, 5.70s/it]
16%|█▌ | 7100/43939 [11:29:08<58:11:49, 5.69s/it]
{'loss': 1.0223, 'grad_norm': 0.9930420517921448, 'learning_rate': 4.942457712329964e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.81, 'epoch': 0.65}
16%|█▌ | 7100/43939 [11:29:08<58:11:49, 5.69s/it]
16%|█▌ | 7101/43939 [11:29:14<58:17:24, 5.70s/it]
16%|█▌ | 7102/43939 [11:29:20<58:13:01, 5.69s/it]
16%|█▌ | 7103/43939 [11:29:25<58:14:53, 5.69s/it]
16%|█▌ | 7104/43939 [11:29:31<58:14:14, 5.69s/it]
16%|█▌ | 7105/43939 [11:29:37<58:59:25, 5.77s/it]
16%|█▌ | 7106/43939 [11:29:43<58:41:55, 5.74s/it]
16%|█▌ | 7107/43939 [11:29:48<58:33:11, 5.72s/it]
16%|█▌ | 7108/43939 [11:29:54<58:31:04, 5.72s/it]
16%|█▌ | 7109/43939 [11:30:00<58:27:56, 5.71s/it]
16%|█▌ | 7110/43939 [11:30:05<58:23:21, 5.71s/it]
16%|█▌ | 7111/43939 [11:30:11<58:17:14, 5.70s/it]
16%|█▌ | 7112/43939 [11:30:17<58:17:07, 5.70s/it]
16%|█▌ | 7113/43939 [11:30:22<58:23:08, 5.71s/it]
16%|█▌ | 7114/43939 [11:30:28<58:31:35, 5.72s/it]
16%|█▌ | 7115/43939 [11:30:34<58:33:33, 5.72s/it]
16%|█▌ | 7116/43939 [11:30:40<58:26:48, 5.71s/it]
16%|█▌ | 7117/43939 [11:30:45<58:26:31, 5.71s/it]
16%|█▌ | 7118/43939 [11:30:51<58:23:13, 5.71s/it]
16%|█▌ | 7119/43939 [11:30:57<58:24:10, 5.71s/it]
16%|█▌ | 7120/43939 [11:31:02<58:27:51, 5.72s/it]
16%|█▌ | 7121/43939 [11:31:08<58:17:27, 5.70s/it]
16%|█▌ | 7122/43939 [11:31:14<58:14:06, 5.69s/it]
16%|█▌ | 7123/43939 [11:31:19<58:09:49, 5.69s/it]
16%|█▌ | 7124/43939 [11:31:25<58:12:35, 5.69s/it]
16%|█▌ | 7125/43939 [11:31:31<58:13:09, 5.69s/it]
16%|█▌ | 7126/43939 [11:31:37<58:15:07, 5.70s/it]
16%|█▌ | 7127/43939 [11:31:42<58:56:21, 5.76s/it]
16%|█▌ | 7128/43939 [11:31:48<58:54:29, 5.76s/it]
16%|█▌ | 7129/43939 [11:31:54<58:48:06, 5.75s/it]
16%|█▌ | 7130/43939 [11:32:00<58:33:39, 5.73s/it]
16%|█▌ | 7131/43939 [11:32:05<58:37:33, 5.73s/it]
16%|█▌ | 7132/43939 [11:32:11<58:31:49, 5.72s/it]
16%|█▌ | 7133/43939 [11:32:17<58:28:47, 5.72s/it]
16%|█▌ | 7134/43939 [11:32:22<58:18:30, 5.70s/it]
16%|█▌ | 7135/43939 [11:32:28<58:17:26, 5.70s/it]
16%|█▌ | 7136/43939 [11:32:34<58:22:13, 5.71s/it]
16%|█▌ | 7137/43939 [11:32:40<58:19:58, 5.71s/it]
16%|█▌ | 7138/43939 [11:32:45<58:26:25, 5.72s/it]
16%|█▌ | 7139/43939 [11:32:51<58:18:01, 5.70s/it]
16%|█▌ | 7140/43939 [11:32:57<58:08:29, 5.69s/it]
16%|█▋ | 7141/43939 [11:33:02<58:09:48, 5.69s/it]
16%|█▋ | 7142/43939 [11:33:08<58:14:01, 5.70s/it]
16%|█▋ | 7143/43939 [11:33:14<58:13:06, 5.70s/it]
16%|█▋ | 7144/43939 [11:33:19<58:10:49, 5.69s/it]
16%|█▋ | 7145/43939 [11:33:25<58:14:30, 5.70s/it]
16%|█▋ | 7146/43939 [11:33:31<58:20:31, 5.71s/it]
16%|█▋ | 7147/43939 [11:33:37<59:01:29, 5.78s/it]
16%|█▋ | 7148/43939 [11:33:43<58:49:11, 5.76s/it]
16%|█▋ | 7149/43939 [11:33:48<58:42:36, 5.74s/it]
16%|█▋ | 7150/43939 [11:33:54<58:38:49, 5.74s/it]
{'loss': 1.0566, 'grad_norm': 0.9996587634086609, 'learning_rate': 4.940320175060446e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.44, 'epoch': 0.65}
16%|█▋ | 7150/43939 [11:33:54<58:38:49, 5.74s/it]
16%|█▋ | 7151/43939 [11:34:00<58:31:23, 5.73s/it]
16%|█▋ | 7152/43939 [11:34:05<58:23:35, 5.71s/it]
16%|█▋ | 7153/43939 [11:34:11<58:17:18, 5.70s/it]
16%|█▋ | 7154/43939 [11:34:17<58:21:05, 5.71s/it]
16%|█▋ | 7155/43939 [11:34:22<58:20:39, 5.71s/it]
16%|█▋ | 7156/43939 [11:34:28<58:10:53, 5.69s/it]
16%|█▋ | 7157/43939 [11:34:34<58:14:07, 5.70s/it]
16%|█▋ | 7158/43939 [11:34:40<58:08:34, 5.69s/it]
16%|█▋ | 7159/43939 [11:34:45<58:10:00, 5.69s/it]
16%|█▋ | 7160/43939 [11:34:51<58:15:35, 5.70s/it]
16%|█▋ | 7161/43939 [11:34:57<58:08:00, 5.69s/it]
16%|█▋ | 7162/43939 [11:35:02<58:11:07, 5.70s/it]
16%|█▋ | 7163/43939 [11:35:08<58:16:13, 5.70s/it]
16%|█▋ | 7164/43939 [11:35:14<58:14:38, 5.70s/it]
16%|█▋ | 7165/43939 [11:35:19<58:15:58, 5.70s/it]
16%|█▋ | 7166/43939 [11:35:25<58:15:57, 5.70s/it]
16%|█▋ | 7167/43939 [11:35:31<58:16:18, 5.70s/it]
16%|█▋ | 7168/43939 [11:35:37<58:58:34, 5.77s/it]
16%|█▋ | 7169/43939 [11:35:42<58:39:44, 5.74s/it]
16%|█▋ | 7170/43939 [11:35:48<58:32:48, 5.73s/it]
16%|█▋ | 7171/43939 [11:35:54<58:16:15, 5.71s/it]
16%|█▋ | 7172/43939 [11:35:59<58:08:18, 5.69s/it]
16%|█▋ | 7173/43939 [11:36:05<58:06:11, 5.69s/it]
16%|█▋ | 7174/43939 [11:36:11<58:15:34, 5.70s/it]
16%|█▋ | 7175/43939 [11:36:17<58:12:31, 5.70s/it]
16%|█▋ | 7176/43939 [11:36:22<58:19:13, 5.71s/it]
16%|█▋ | 7177/43939 [11:36:28<58:18:23, 5.71s/it]
16%|█▋ | 7178/43939 [11:36:34<58:16:26, 5.71s/it]
16%|█▋ | 7179/43939 [11:36:39<58:22:21, 5.72s/it]
16%|█▋ | 7180/43939 [11:36:45<58:18:07, 5.71s/it]
16%|█▋ | 7181/43939 [11:36:51<58:15:57, 5.71s/it]
16%|█▋ | 7182/43939 [11:36:57<58:19:29, 5.71s/it]
16%|█▋ | 7183/43939 [11:37:02<58:09:36, 5.70s/it]
16%|█▋ | 7184/43939 [11:37:08<58:15:39, 5.71s/it]
16%|█▋ | 7185/43939 [11:37:14<58:12:55, 5.70s/it]
16%|█▋ | 7186/43939 [11:37:19<58:16:04, 5.71s/it]
16%|█▋ | 7187/43939 [11:37:25<58:13:41, 5.70s/it]
16%|█▋ | 7188/43939 [11:37:31<58:17:37, 5.71s/it]
16%|█▋ | 7189/43939 [11:37:37<58:14:37, 5.71s/it]
16%|█▋ | 7190/43939 [11:37:42<58:57:04, 5.77s/it]
16%|█▋ | 7191/43939 [11:37:48<58:45:34, 5.76s/it]
16%|█▋ | 7192/43939 [11:37:54<58:35:51, 5.74s/it]
16%|█▋ | 7193/43939 [11:38:00<58:29:16, 5.73s/it]
16%|█▋ | 7194/43939 [11:38:05<58:25:13, 5.72s/it]
16%|█▋ | 7195/43939 [11:38:11<58:28:50, 5.73s/it]
16%|█▋ | 7196/43939 [11:38:17<58:22:12, 5.72s/it]
16%|█▋ | 7197/43939 [11:38:22<58:17:43, 5.71s/it]
16%|█▋ | 7198/43939 [11:38:28<58:14:43, 5.71s/it]
16%|█▋ | 7199/43939 [11:38:34<58:14:23, 5.71s/it]
16%|█▋ | 7200/43939 [11:38:39<58:07:58, 5.70s/it]
{'loss': 1.0522, 'grad_norm': 1.7110569477081299, 'learning_rate': 4.9381441360054915e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.36, 'epoch': 0.66}
16%|█▋ | 7200/43939 [11:38:39<58:07:58, 5.70s/it]
16%|█▋ | 7201/43939 [11:38:45<58:08:29, 5.70s/it]
16%|█▋ | 7202/43939 [11:38:51<58:10:01, 5.70s/it]
16%|█▋ | 7203/43939 [11:38:57<58:09:42, 5.70s/it]
16%|█▋ | 7204/43939 [11:39:02<58:08:18, 5.70s/it]
16%|█▋ | 7205/43939 [11:39:08<58:00:40, 5.69s/it]
16%|█▋ | 7206/43939 [11:39:14<58:11:53, 5.70s/it]
16%|█▋ | 7207/43939 [11:39:19<58:17:26, 5.71s/it]
16%|█▋ | 7208/43939 [11:39:25<58:19:28, 5.72s/it]
16%|█▋ | 7209/43939 [11:39:31<58:14:44, 5.71s/it]
16%|█▋ | 7210/43939 [11:39:37<58:58:50, 5.78s/it]
16%|█▋ | 7211/43939 [11:39:42<58:37:23, 5.75s/it]
16%|█▋ | 7212/43939 [11:39:48<58:30:17, 5.73s/it]
16%|█▋ | 7213/43939 [11:39:54<58:22:53, 5.72s/it]
16%|█▋ | 7214/43939 [11:40:00<58:19:23, 5.72s/it]
16%|█▋ | 7215/43939 [11:40:05<58:16:40, 5.71s/it]
16%|█▋ | 7216/43939 [11:40:11<58:13:47, 5.71s/it]
16%|█▋ | 7217/43939 [11:40:17<58:12:51, 5.71s/it]
16%|█▋ | 7218/43939 [11:40:22<58:14:18, 5.71s/it]
16%|█▋ | 7219/43939 [11:40:28<58:07:51, 5.70s/it]
16%|█▋ | 7220/43939 [11:40:34<58:05:52, 5.70s/it]
16%|█▋ | 7221/43939 [11:40:39<57:58:23, 5.68s/it]
16%|█▋ | 7222/43939 [11:40:45<58:00:27, 5.69s/it]
16%|█▋ | 7223/43939 [11:40:51<58:05:25, 5.70s/it]
16%|█▋ | 7224/43939 [11:40:56<58:01:22, 5.69s/it]
16%|█▋ | 7225/43939 [11:41:02<57:59:14, 5.69s/it]
16%|█▋ | 7226/43939 [11:41:08<57:56:07, 5.68s/it]
16%|█▋ | 7227/43939 [11:41:14<58:01:08, 5.69s/it]
16%|█▋ | 7228/43939 [11:41:19<58:03:52, 5.69s/it]
16%|█▋ | 7229/43939 [11:41:25<58:02:59, 5.69s/it]
16%|█▋ | 7230/43939 [11:41:31<58:04:35, 5.70s/it]
16%|█▋ | 7231/43939 [11:41:37<58:44:22, 5.76s/it]
16%|█▋ | 7232/43939 [11:41:42<58:24:10, 5.73s/it]
16%|█▋ | 7233/43939 [11:41:48<58:10:43, 5.71s/it]
16%|█▋ | 7234/43939 [11:41:54<58:14:02, 5.71s/it]
16%|█▋ | 7235/43939 [11:41:59<58:12:56, 5.71s/it]
16%|█▋ | 7236/43939 [11:42:05<58:03:42, 5.69s/it]
16%|█▋ | 7237/43939 [11:42:11<58:03:05, 5.69s/it]
16%|█▋ | 7238/43939 [11:42:16<58:06:20, 5.70s/it]
16%|█▋ | 7239/43939 [11:42:22<58:01:48, 5.69s/it]
16%|█▋ | 7240/43939 [11:42:28<58:02:23, 5.69s/it]
16%|█▋ | 7241/43939 [11:42:34<58:19:41, 5.72s/it]
16%|█▋ | 7242/43939 [11:42:39<58:14:14, 5.71s/it]
16%|█▋ | 7243/43939 [11:42:45<58:12:33, 5.71s/it]
16%|█▋ | 7244/43939 [11:42:51<58:15:32, 5.72s/it]
16%|█▋ | 7245/43939 [11:42:56<58:15:11, 5.72s/it]
16%|█▋ | 7246/43939 [11:43:02<58:12:13, 5.71s/it]
16%|█▋ | 7247/43939 [11:43:08<58:13:48, 5.71s/it]
16%|█▋ | 7248/43939 [11:43:14<58:21:39, 5.73s/it]
16%|█▋ | 7249/43939 [11:43:19<58:14:54, 5.72s/it]
17%|█▋ | 7250/43939 [11:43:25<58:11:39, 5.71s/it]
{'loss': 1.0541, 'grad_norm': 1.2259682416915894, 'learning_rate': 4.9359296294972315e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.99, 'epoch': 0.66}
17%|█▋ | 7250/43939 [11:43:25<58:11:39, 5.71s/it]
17%|█▋ | 7251/43939 [11:43:31<58:14:29, 5.71s/it]
17%|█▋ | 7252/43939 [11:43:36<58:09:13, 5.71s/it]
17%|█▋ | 7253/43939 [11:43:42<58:46:31, 5.77s/it]
17%|█▋ | 7254/43939 [11:43:48<58:35:01, 5.75s/it]
17%|█▋ | 7255/43939 [11:43:54<58:30:00, 5.74s/it]
17%|█▋ | 7256/43939 [11:43:59<58:25:26, 5.73s/it]
17%|█▋ | 7257/43939 [11:44:05<58:19:49, 5.72s/it]
17%|█▋ | 7258/43939 [11:44:11<58:08:24, 5.71s/it]
17%|█▋ | 7259/43939 [11:44:16<58:03:29, 5.70s/it]
17%|█▋ | 7260/43939 [11:44:22<58:06:37, 5.70s/it]
17%|█▋ | 7261/43939 [11:44:28<58:05:44, 5.70s/it]
17%|█▋ | 7262/43939 [11:44:34<58:09:09, 5.71s/it]
17%|█▋ | 7263/43939 [11:44:39<58:01:12, 5.70s/it]
17%|█▋ | 7264/43939 [11:44:45<58:00:57, 5.69s/it]
17%|█▋ | 7265/43939 [11:44:51<57:59:04, 5.69s/it]
17%|█▋ | 7266/43939 [11:44:56<57:57:39, 5.69s/it]
17%|█▋ | 7267/43939 [11:45:02<58:01:13, 5.70s/it]
17%|█▋ | 7268/43939 [11:45:08<58:04:51, 5.70s/it]
17%|█▋ | 7269/43939 [11:45:13<57:58:20, 5.69s/it]
17%|█▋ | 7270/43939 [11:45:19<58:08:14, 5.71s/it]
17%|█▋ | 7271/43939 [11:45:25<58:02:23, 5.70s/it]
17%|█▋ | 7272/43939 [11:45:30<57:57:14, 5.69s/it]
17%|█▋ | 7273/43939 [11:45:36<57:57:00, 5.69s/it]
17%|█▋ | 7274/43939 [11:45:42<58:33:03, 5.75s/it]
17%|█▋ | 7275/43939 [11:45:48<58:23:21, 5.73s/it]
17%|█▋ | 7276/43939 [11:45:53<58:11:12, 5.71s/it]
17%|█▋ | 7277/43939 [11:45:59<58:08:39, 5.71s/it]
17%|█▋ | 7278/43939 [11:46:05<58:07:55, 5.71s/it]
17%|█▋ | 7279/43939 [11:46:11<58:07:43, 5.71s/it]
17%|█▋ | 7280/43939 [11:46:16<58:01:31, 5.70s/it]
17%|█▋ | 7281/43939 [11:46:22<58:01:34, 5.70s/it]
17%|█▋ | 7282/43939 [11:46:28<58:01:14, 5.70s/it]
17%|█▋ | 7283/43939 [11:46:33<58:00:32, 5.70s/it]
17%|█▋ | 7284/43939 [11:46:39<58:09:34, 5.71s/it]
17%|█▋ | 7285/43939 [11:46:45<58:03:42, 5.70s/it]
17%|█▋ | 7286/43939 [11:46:50<58:03:18, 5.70s/it]
17%|█▋ | 7287/43939 [11:46:56<57:55:24, 5.69s/it]
17%|█▋ | 7288/43939 [11:47:02<57:55:24, 5.69s/it]
17%|█▋ | 7289/43939 [11:47:07<57:59:02, 5.70s/it]
17%|█▋ | 7290/43939 [11:47:13<58:01:51, 5.70s/it]
17%|█▋ | 7291/43939 [11:47:19<58:01:49, 5.70s/it]
17%|█▋ | 7292/43939 [11:47:25<58:00:47, 5.70s/it]
17%|█▋ | 7293/43939 [11:47:30<58:05:12, 5.71s/it]
17%|█▋ | 7294/43939 [11:47:36<58:00:17, 5.70s/it]
17%|█▋ | 7295/43939 [11:47:42<58:42:12, 5.77s/it]
17%|█▋ | 7296/43939 [11:47:48<58:43:09, 5.77s/it]
17%|█▋ | 7297/43939 [11:47:53<58:30:43, 5.75s/it]
17%|█▋ | 7298/43939 [11:47:59<58:17:59, 5.73s/it]
17%|█▋ | 7299/43939 [11:48:05<58:03:54, 5.71s/it]
17%|█▋ | 7300/43939 [11:48:10<58:06:16, 5.71s/it]
{'loss': 1.0551, 'grad_norm': 1.290621042251587, 'learning_rate': 4.933676690474711e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.9, 'epoch': 0.66}
17%|█▋ | 7300/43939 [11:48:10<58:06:16, 5.71s/it]
17%|█▋ | 7301/43939 [11:48:16<58:10:08, 5.72s/it]
17%|█▋ | 7302/43939 [11:48:22<58:02:42, 5.70s/it]
17%|█▋ | 7303/43939 [11:48:28<57:57:54, 5.70s/it]
17%|█▋ | 7304/43939 [11:48:33<57:52:22, 5.69s/it]
17%|█▋ | 7305/43939 [11:48:39<57:57:02, 5.69s/it]
17%|█▋ | 7306/43939 [11:48:45<57:58:05, 5.70s/it]
17%|█▋ | 7307/43939 [11:48:50<57:58:25, 5.70s/it]
17%|█▋ | 7308/43939 [11:48:56<57:58:29, 5.70s/it]
17%|█▋ | 7309/43939 [11:49:02<57:51:41, 5.69s/it]
17%|█▋ | 7310/43939 [11:49:07<57:55:19, 5.69s/it]
17%|█▋ | 7311/43939 [11:49:13<58:04:07, 5.71s/it]
17%|█▋ | 7312/43939 [11:49:19<58:11:21, 5.72s/it]
17%|█▋ | 7313/43939 [11:49:25<58:02:03, 5.70s/it]
17%|█▋ | 7314/43939 [11:49:30<58:01:35, 5.70s/it]
17%|█▋ | 7315/43939 [11:49:36<57:59:48, 5.70s/it]
17%|█▋ | 7316/43939 [11:49:42<58:44:45, 5.77s/it]
17%|█▋ | 7317/43939 [11:49:48<58:38:45, 5.77s/it]
17%|█▋ | 7318/43939 [11:49:53<58:29:33, 5.75s/it]
17%|█▋ | 7319/43939 [11:49:59<58:24:54, 5.74s/it]
17%|█▋ | 7320/43939 [11:50:05<58:12:40, 5.72s/it]
17%|█▋ | 7321/43939 [11:50:10<58:08:46, 5.72s/it]
17%|█▋ | 7322/43939 [11:50:16<58:10:02, 5.72s/it]
17%|█▋ | 7323/43939 [11:50:22<58:06:16, 5.71s/it]
17%|█▋ | 7324/43939 [11:50:28<58:04:28, 5.71s/it]
17%|█▋ | 7325/43939 [11:50:33<57:58:03, 5.70s/it]
17%|█▋ | 7326/43939 [11:50:39<57:57:17, 5.70s/it]
17%|█▋ | 7327/43939 [11:50:45<57:59:00, 5.70s/it]
17%|█▋ | 7328/43939 [11:50:50<57:53:03, 5.69s/it]
17%|█▋ | 7329/43939 [11:50:56<57:51:27, 5.69s/it]
17%|█▋ | 7330/43939 [11:51:02<57:58:41, 5.70s/it]
17%|█▋ | 7331/43939 [11:51:07<58:02:27, 5.71s/it]
17%|█▋ | 7332/43939 [11:51:13<58:08:22, 5.72s/it]
17%|█▋ | 7333/43939 [11:51:19<58:07:25, 5.72s/it]
17%|█▋ | 7334/43939 [11:51:25<58:05:02, 5.71s/it]
17%|█▋ | 7335/43939 [11:51:30<57:58:11, 5.70s/it]
17%|█▋ | 7336/43939 [11:51:36<58:00:08, 5.70s/it]
17%|█▋ | 7337/43939 [11:51:42<58:39:04, 5.77s/it]
17%|█▋ | 7338/43939 [11:51:48<58:25:08, 5.75s/it]
17%|█▋ | 7339/43939 [11:51:53<58:13:37, 5.73s/it]
17%|█▋ | 7340/43939 [11:51:59<58:06:23, 5.72s/it]
17%|█▋ | 7341/43939 [11:52:05<58:10:13, 5.72s/it]
17%|█▋ | 7342/43939 [11:52:10<58:09:53, 5.72s/it]
17%|█▋ | 7343/43939 [11:52:16<58:07:39, 5.72s/it]
17%|█▋ | 7344/43939 [11:52:22<58:07:57, 5.72s/it]
17%|█▋ | 7345/43939 [11:52:28<58:04:11, 5.71s/it]
17%|█▋ | 7346/43939 [11:52:33<58:11:50, 5.73s/it]
17%|█▋ | 7347/43939 [11:52:39<58:09:15, 5.72s/it]
17%|█▋ | 7348/43939 [11:52:45<58:06:51, 5.72s/it]
17%|█▋ | 7349/43939 [11:52:50<58:05:35, 5.72s/it]
17%|█▋ | 7350/43939 [11:52:56<58:11:52, 5.73s/it]
{'loss': 1.0512, 'grad_norm': 0.8518486022949219, 'learning_rate': 4.931385354483336e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.75, 'epoch': 0.67}
17%|█▋ | 7350/43939 [11:52:56<58:11:52, 5.73s/it]
17%|█▋ | 7351/43939 [11:53:02<58:10:05, 5.72s/it]
17%|█▋ | 7352/43939 [11:53:08<57:58:01, 5.70s/it]
17%|█▋ | 7353/43939 [11:53:13<58:08:28, 5.72s/it]
17%|█▋ | 7354/43939 [11:53:19<57:59:09, 5.71s/it]
17%|█▋ | 7355/43939 [11:53:25<57:59:21, 5.71s/it]
17%|█▋ | 7356/43939 [11:53:30<57:54:56, 5.70s/it]
17%|█▋ | 7357/43939 [11:53:36<57:51:01, 5.69s/it]
17%|█▋ | 7358/43939 [11:53:42<58:39:35, 5.77s/it]
17%|█▋ | 7359/43939 [11:53:48<58:21:49, 5.74s/it]
17%|█▋ | 7360/43939 [11:53:53<58:06:39, 5.72s/it]
17%|█▋ | 7361/43939 [11:53:59<58:05:02, 5.72s/it]
17%|█▋ | 7362/43939 [11:54:05<58:07:36, 5.72s/it]
17%|█▋ | 7363/43939 [11:54:10<57:57:12, 5.70s/it]
17%|█▋ | 7364/43939 [11:54:16<57:59:52, 5.71s/it]
17%|█▋ | 7365/43939 [11:54:22<57:56:04, 5.70s/it]
17%|█▋ | 7366/43939 [11:54:28<58:03:39, 5.72s/it]
17%|█▋ | 7367/43939 [11:54:33<58:04:17, 5.72s/it]
17%|█▋ | 7368/43939 [11:54:39<57:59:35, 5.71s/it]
17%|█▋ | 7369/43939 [11:54:45<57:59:36, 5.71s/it]
17%|█▋ | 7370/43939 [11:54:50<57:55:59, 5.70s/it]
17%|█▋ | 7371/43939 [11:54:56<57:47:20, 5.69s/it]
17%|█▋ | 7372/43939 [11:55:02<57:44:58, 5.69s/it]
17%|█▋ | 7373/43939 [11:55:07<57:50:28, 5.69s/it]
17%|█▋ | 7374/43939 [11:55:13<57:50:21, 5.69s/it]
17%|█▋ | 7375/43939 [11:55:19<57:57:59, 5.71s/it]
17%|█▋ | 7376/43939 [11:55:25<58:05:18, 5.72s/it]
17%|█▋ | 7377/43939 [11:55:30<58:06:41, 5.72s/it]
17%|█▋ | 7378/43939 [11:55:36<57:59:42, 5.71s/it]
17%|█▋ | 7379/43939 [11:55:42<58:39:44, 5.78s/it]
17%|█▋ | 7380/43939 [11:55:48<58:19:16, 5.74s/it]
17%|█▋ | 7381/43939 [11:55:53<58:10:19, 5.73s/it]
17%|█▋ | 7382/43939 [11:55:59<58:10:04, 5.73s/it]
17%|█▋ | 7383/43939 [11:56:05<58:06:34, 5.72s/it]
17%|█▋ | 7384/43939 [11:56:10<57:57:45, 5.71s/it]
17%|█▋ | 7385/43939 [11:56:16<57:51:24, 5.70s/it]
17%|█▋ | 7386/43939 [11:56:22<57:47:35, 5.69s/it]
17%|█▋ | 7387/43939 [11:56:28<57:54:17, 5.70s/it]
17%|█▋ | 7388/43939 [11:56:33<57:54:02, 5.70s/it]
17%|█▋ | 7389/43939 [11:56:39<57:53:27, 5.70s/it]
17%|█▋ | 7390/43939 [11:56:45<57:51:22, 5.70s/it]
17%|█▋ | 7391/43939 [11:56:50<57:48:27, 5.69s/it]
17%|█▋ | 7392/43939 [11:56:56<57:58:30, 5.71s/it]
17%|█▋ | 7393/43939 [11:57:02<57:47:46, 5.69s/it]
17%|█▋ | 7394/43939 [11:57:07<57:48:01, 5.69s/it]
17%|█▋ | 7395/43939 [11:57:13<57:43:37, 5.69s/it]
17%|█▋ | 7396/43939 [11:57:19<57:41:38, 5.68s/it]
17%|█▋ | 7397/43939 [11:57:25<57:49:02, 5.70s/it]
17%|█▋ | 7398/43939 [11:57:30<57:45:39, 5.69s/it]
17%|█▋ | 7399/43939 [11:57:36<57:44:03, 5.69s/it]
17%|█▋ | 7400/43939 [11:57:42<58:34:14, 5.77s/it]
{'loss': 1.0463, 'grad_norm': 1.0784375667572021, 'learning_rate': 4.929055657674316e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 451.58, 'epoch': 0.67}
17%|█▋ | 7400/43939 [11:57:42<58:34:14, 5.77s/it]
17%|█▋ | 7401/43939 [11:57:48<58:17:04, 5.74s/it]
17%|█▋ | 7402/43939 [11:57:53<58:10:00, 5.73s/it]
17%|█▋ | 7403/43939 [11:57:59<58:00:34, 5.72s/it]
17%|█▋ | 7404/43939 [11:58:05<58:00:08, 5.72s/it]
17%|█▋ | 7405/43939 [11:58:10<57:57:22, 5.71s/it]
17%|█▋ | 7406/43939 [11:58:16<57:57:45, 5.71s/it]
17%|█▋ | 7407/43939 [11:58:22<57:50:48, 5.70s/it]
17%|█▋ | 7408/43939 [11:58:27<57:49:34, 5.70s/it]
17%|█▋ | 7409/43939 [11:58:33<57:50:05, 5.70s/it]
17%|█▋ | 7410/43939 [11:58:39<57:45:06, 5.69s/it]
17%|█▋ | 7411/43939 [11:58:44<57:40:23, 5.68s/it]
17%|█▋ | 7412/43939 [11:58:50<57:50:50, 5.70s/it]
17%|█▋ | 7413/43939 [11:58:56<57:51:31, 5.70s/it]
17%|█▋ | 7414/43939 [11:59:02<57:54:48, 5.71s/it]
17%|█▋ | 7415/43939 [11:59:07<57:51:57, 5.70s/it]
17%|█▋ | 7416/43939 [11:59:13<57:49:08, 5.70s/it]
17%|█▋ | 7417/43939 [11:59:19<57:49:09, 5.70s/it]
17%|█▋ | 7418/43939 [11:59:24<57:50:53, 5.70s/it]
17%|█▋ | 7419/43939 [11:59:30<57:50:50, 5.70s/it]
17%|█▋ | 7420/43939 [11:59:36<57:50:13, 5.70s/it]
17%|█▋ | 7421/43939 [11:59:42<58:29:57, 5.77s/it]
17%|█▋ | 7422/43939 [11:59:47<58:11:54, 5.74s/it]
17%|█▋ | 7423/43939 [11:59:53<58:05:06, 5.73s/it]
17%|█▋ | 7424/43939 [11:59:59<57:59:39, 5.72s/it]
17%|█▋ | 7425/43939 [12:00:04<58:00:12, 5.72s/it]
17%|█▋ | 7426/43939 [12:00:10<58:00:15, 5.72s/it]
17%|█▋ | 7427/43939 [12:00:16<57:56:39, 5.71s/it]
17%|█▋ | 7428/43939 [12:00:22<57:50:33, 5.70s/it]
17%|█▋ | 7429/43939 [12:00:27<57:58:58, 5.72s/it]
17%|█▋ | 7430/43939 [12:00:33<58:00:09, 5.72s/it]
17%|█▋ | 7431/43939 [12:00:39<57:55:01, 5.71s/it]
17%|█▋ | 7432/43939 [12:00:44<57:50:33, 5.70s/it]
17%|█▋ | 7433/43939 [12:00:50<57:49:31, 5.70s/it]
17%|█▋ | 7434/43939 [12:00:56<57:48:47, 5.70s/it]
17%|█▋ | 7435/43939 [12:01:02<57:49:11, 5.70s/it]
17%|█▋ | 7436/43939 [12:01:07<57:46:04, 5.70s/it]
17%|█▋ | 7437/43939 [12:01:13<57:41:22, 5.69s/it]
17%|█▋ | 7438/43939 [12:01:19<57:41:09, 5.69s/it]
17%|█▋ | 7439/43939 [12:01:24<57:43:00, 5.69s/it]
17%|█▋ | 7440/43939 [12:01:30<57:40:03, 5.69s/it]
17%|█▋ | 7441/43939 [12:01:36<57:40:30, 5.69s/it]
17%|█▋ | 7442/43939 [12:01:42<58:25:38, 5.76s/it]
17%|█▋ | 7443/43939 [12:01:47<58:15:36, 5.75s/it]
17%|█▋ | 7444/43939 [12:01:53<58:05:22, 5.73s/it]
17%|█▋ | 7445/43939 [12:01:59<57:57:39, 5.72s/it]
17%|█▋ | 7446/43939 [12:02:04<57:53:12, 5.71s/it]
17%|█▋ | 7447/43939 [12:02:10<57:49:58, 5.71s/it]
17%|█▋ | 7448/43939 [12:02:16<58:05:25, 5.73s/it]
17%|█▋ | 7449/43939 [12:02:22<57:58:06, 5.72s/it]
17%|█▋ | 7450/43939 [12:02:27<57:48:40, 5.70s/it]
{'loss': 1.0281, 'grad_norm': 1.1245126724243164, 'learning_rate': 4.9266876368040954e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.61, 'epoch': 0.68}
17%|█▋ | 7450/43939 [12:02:27<57:48:40, 5.70s/it]
17%|█▋ | 7451/43939 [12:02:33<57:40:43, 5.69s/it]
17%|█▋ | 7452/43939 [12:02:39<57:44:44, 5.70s/it]
17%|█▋ | 7453/43939 [12:02:44<57:50:03, 5.71s/it]
17%|█▋ | 7454/43939 [12:02:50<57:54:53, 5.71s/it]
17%|█▋ | 7455/43939 [12:02:56<57:50:18, 5.71s/it]
17%|█▋ | 7456/43939 [12:03:01<57:55:16, 5.72s/it]
17%|█▋ | 7457/43939 [12:03:07<57:48:07, 5.70s/it]
17%|█▋ | 7458/43939 [12:03:13<57:44:21, 5.70s/it]
17%|█▋ | 7459/43939 [12:03:19<57:48:37, 5.70s/it]
17%|█▋ | 7460/43939 [12:03:24<57:43:15, 5.70s/it]
17%|█▋ | 7461/43939 [12:03:30<57:44:43, 5.70s/it]
17%|█▋ | 7462/43939 [12:03:36<57:42:58, 5.70s/it]
17%|█▋ | 7463/43939 [12:03:42<58:36:07, 5.78s/it]
17%|█▋ | 7464/43939 [12:03:47<58:12:41, 5.75s/it]
17%|█▋ | 7465/43939 [12:03:53<58:10:07, 5.74s/it]
17%|█▋ | 7466/43939 [12:03:59<58:05:32, 5.73s/it]
17%|█▋ | 7467/43939 [12:04:04<57:56:30, 5.72s/it]
17%|█▋ | 7468/43939 [12:04:10<57:42:30, 5.70s/it]
17%|█▋ | 7469/43939 [12:04:16<57:33:54, 5.68s/it]
17%|█▋ | 7470/43939 [12:04:21<57:28:59, 5.67s/it]
17%|█▋ | 7471/43939 [12:04:27<57:42:56, 5.70s/it]
17%|█▋ | 7472/43939 [12:04:33<57:46:39, 5.70s/it]
17%|█▋ | 7473/43939 [12:04:39<57:50:48, 5.71s/it]
17%|█▋ | 7474/43939 [12:04:44<57:49:09, 5.71s/it]
17%|█▋ | 7475/43939 [12:04:50<57:47:53, 5.71s/it]
17%|█▋ | 7476/43939 [12:04:56<57:49:26, 5.71s/it]
17%|█▋ | 7477/43939 [12:05:01<57:55:14, 5.72s/it]
17%|█▋ | 7478/43939 [12:05:07<57:54:30, 5.72s/it]
17%|█▋ | 7479/43939 [12:05:13<57:45:34, 5.70s/it]
17%|█▋ | 7480/43939 [12:05:18<57:43:15, 5.70s/it]
17%|█▋ | 7481/43939 [12:05:24<57:43:37, 5.70s/it]
17%|█▋ | 7482/43939 [12:05:30<57:43:13, 5.70s/it]
17%|█▋ | 7483/43939 [12:05:36<57:39:38, 5.69s/it]
17%|█▋ | 7484/43939 [12:05:42<58:22:13, 5.76s/it]
17%|█▋ | 7485/43939 [12:05:47<58:05:19, 5.74s/it]
17%|█▋ | 7486/43939 [12:05:53<57:58:20, 5.73s/it]
17%|█▋ | 7487/43939 [12:05:59<57:54:12, 5.72s/it]
17%|█▋ | 7488/43939 [12:06:04<57:58:02, 5.73s/it]
17%|█▋ | 7489/43939 [12:06:10<57:55:12, 5.72s/it]
17%|█▋ | 7490/43939 [12:06:16<57:45:06, 5.70s/it]
17%|█▋ | 7491/43939 [12:06:21<57:46:47, 5.71s/it]
17%|█▋ | 7492/43939 [12:06:27<57:49:07, 5.71s/it]
17%|█▋ | 7493/43939 [12:06:33<57:49:49, 5.71s/it]
17%|█▋ | 7494/43939 [12:06:39<57:48:15, 5.71s/it]
17%|█▋ | 7495/43939 [12:06:44<57:43:55, 5.70s/it]
17%|█▋ | 7496/43939 [12:06:50<57:46:16, 5.71s/it]
17%|█▋ | 7497/43939 [12:06:56<57:45:31, 5.71s/it]
17%|█▋ | 7498/43939 [12:07:01<57:38:29, 5.69s/it]
17%|█▋ | 7499/43939 [12:07:07<57:39:16, 5.70s/it]
17%|█▋ | 7500/43939 [12:07:13<57:42:37, 5.70s/it]
{'loss': 1.036, 'grad_norm': 1.037510633468628, 'learning_rate': 4.924281329233766e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.11, 'epoch': 0.68}
17%|█▋ | 7500/43939 [12:07:13<57:42:37, 5.70s/it]
17%|█▋ | 7501/43939 [12:07:18<57:46:39, 5.71s/it]
17%|█▋ | 7502/43939 [12:07:24<57:33:37, 5.69s/it]
17%|█▋ | 7503/43939 [12:07:30<57:29:00, 5.68s/it]
17%|█▋ | 7504/43939 [12:07:35<57:32:17, 5.69s/it]
17%|█▋ | 7505/43939 [12:07:41<58:18:25, 5.76s/it]
17%|█▋ | 7506/43939 [12:07:47<58:00:31, 5.73s/it]
17%|█▋ | 7507/43939 [12:07:53<57:54:05, 5.72s/it]
17%|█▋ | 7508/43939 [12:07:58<57:53:44, 5.72s/it]
17%|█▋ | 7509/43939 [12:08:04<57:51:05, 5.72s/it]
17%|█▋ | 7510/43939 [12:08:10<57:44:50, 5.71s/it]
17%|█▋ | 7511/43939 [12:08:16<57:42:13, 5.70s/it]
17%|█▋ | 7512/43939 [12:08:21<57:46:18, 5.71s/it]
17%|█▋ | 7513/43939 [12:08:27<57:50:25, 5.72s/it]
17%|█▋ | 7514/43939 [12:08:33<57:49:37, 5.72s/it]
17%|█▋ | 7515/43939 [12:08:38<57:47:29, 5.71s/it]
17%|█▋ | 7516/43939 [12:08:44<57:44:39, 5.71s/it]
17%|█▋ | 7517/43939 [12:08:50<57:46:17, 5.71s/it]
17%|█▋ | 7518/43939 [12:08:56<57:44:49, 5.71s/it]
17%|█▋ | 7519/43939 [12:09:01<57:43:51, 5.71s/it]
17%|█▋ | 7520/43939 [12:09:07<57:35:14, 5.69s/it]
17%|█▋ | 7521/43939 [12:09:13<57:39:49, 5.70s/it]
17%|█▋ | 7522/43939 [12:09:18<57:51:26, 5.72s/it]
17%|█▋ | 7523/43939 [12:09:24<57:46:05, 5.71s/it]
17%|█▋ | 7524/43939 [12:09:30<57:37:33, 5.70s/it]
17%|█▋ | 7525/43939 [12:09:35<57:40:32, 5.70s/it]
17%|█▋ | 7526/43939 [12:09:41<58:17:57, 5.76s/it]
17%|█▋ | 7527/43939 [12:09:47<58:03:14, 5.74s/it]
17%|█▋ | 7528/43939 [12:09:53<57:52:31, 5.72s/it]
17%|█▋ | 7529/43939 [12:09:58<57:52:03, 5.72s/it]
17%|█▋ | 7530/43939 [12:10:04<57:50:34, 5.72s/it]
17%|█▋ | 7531/43939 [12:10:10<57:41:22, 5.70s/it]
17%|█▋ | 7532/43939 [12:10:16<57:44:29, 5.71s/it]
17%|█▋ | 7533/43939 [12:10:21<57:41:59, 5.71s/it]
17%|█▋ | 7534/43939 [12:10:27<57:44:50, 5.71s/it]
17%|█▋ | 7535/43939 [12:10:33<57:38:09, 5.70s/it]
17%|█▋ | 7536/43939 [12:10:38<57:33:15, 5.69s/it]
17%|█▋ | 7537/43939 [12:10:44<57:39:00, 5.70s/it]
17%|█▋ | 7538/43939 [12:10:50<57:40:32, 5.70s/it]
17%|█▋ | 7539/43939 [12:10:55<57:39:36, 5.70s/it]
17%|█▋ | 7540/43939 [12:11:01<57:43:05, 5.71s/it]
17%|█▋ | 7541/43939 [12:11:07<57:43:30, 5.71s/it]
17%|█▋ | 7542/43939 [12:11:13<57:41:51, 5.71s/it]
17%|█▋ | 7543/43939 [12:11:18<57:41:25, 5.71s/it]
17%|█▋ | 7544/43939 [12:11:24<57:39:07, 5.70s/it]
17%|█▋ | 7545/43939 [12:11:30<57:38:29, 5.70s/it]
17%|█▋ | 7546/43939 [12:11:35<57:38:45, 5.70s/it]
17%|█▋ | 7547/43939 [12:11:41<57:37:50, 5.70s/it]
17%|█▋ | 7548/43939 [12:11:47<58:23:38, 5.78s/it]
17%|█▋ | 7549/43939 [12:11:53<58:13:15, 5.76s/it]
17%|█▋ | 7550/43939 [12:11:58<57:54:46, 5.73s/it]
{'loss': 1.0504, 'grad_norm': 0.9862679839134216, 'learning_rate': 4.921836772928484e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.9, 'epoch': 0.69}
17%|█▋ | 7550/43939 [12:11:58<57:54:46, 5.73s/it]
17%|█▋ | 7551/43939 [12:12:04<57:45:38, 5.71s/it]
17%|█▋ | 7552/43939 [12:12:10<57:51:08, 5.72s/it]
17%|█▋ | 7553/43939 [12:12:16<57:55:42, 5.73s/it]
17%|█▋ | 7554/43939 [12:12:21<57:48:26, 5.72s/it]
17%|█▋ | 7555/43939 [12:12:27<57:40:07, 5.71s/it]
17%|█▋ | 7556/43939 [12:12:33<57:47:25, 5.72s/it]
17%|█▋ | 7557/43939 [12:12:38<57:51:03, 5.72s/it]
17%|█▋ | 7558/43939 [12:12:44<57:49:13, 5.72s/it]
17%|█▋ | 7559/43939 [12:12:50<57:48:10, 5.72s/it]
17%|█▋ | 7560/43939 [12:12:56<57:43:55, 5.71s/it]
17%|█▋ | 7561/43939 [12:13:01<57:37:40, 5.70s/it]
17%|█▋ | 7562/43939 [12:13:07<57:33:58, 5.70s/it]
17%|█▋ | 7563/43939 [12:13:13<57:30:01, 5.69s/it]
17%|█▋ | 7564/43939 [12:13:18<57:32:19, 5.69s/it]
17%|█▋ | 7565/43939 [12:13:24<57:30:14, 5.69s/it]
17%|█▋ | 7566/43939 [12:13:30<57:22:35, 5.68s/it]
17%|█▋ | 7567/43939 [12:13:35<57:24:29, 5.68s/it]
17%|█▋ | 7568/43939 [12:13:41<57:25:38, 5.68s/it]
17%|█▋ | 7569/43939 [12:13:47<58:10:51, 5.76s/it]
17%|█▋ | 7570/43939 [12:13:53<57:49:45, 5.72s/it]
17%|█▋ | 7571/43939 [12:13:58<57:44:38, 5.72s/it]
17%|█▋ | 7572/43939 [12:14:04<57:38:05, 5.71s/it]
17%|█▋ | 7573/43939 [12:14:10<57:34:25, 5.70s/it]
17%|█▋ | 7574/43939 [12:14:15<57:28:39, 5.69s/it]
17%|█▋ | 7575/43939 [12:14:21<57:31:06, 5.69s/it]
17%|█▋ | 7576/43939 [12:14:27<57:33:49, 5.70s/it]
17%|█▋ | 7577/43939 [12:14:32<57:35:04, 5.70s/it]
17%|█▋ | 7578/43939 [12:14:38<57:42:27, 5.71s/it]
17%|█▋ | 7579/43939 [12:14:44<57:41:53, 5.71s/it]
17%|█▋ | 7580/43939 [12:14:50<57:34:08, 5.70s/it]
17%|█▋ | 7581/43939 [12:14:55<57:33:21, 5.70s/it]
17%|█▋ | 7582/43939 [12:15:01<57:34:12, 5.70s/it]
17%|█▋ | 7583/43939 [12:15:07<57:30:07, 5.69s/it]
17%|█▋ | 7584/43939 [12:15:12<57:23:37, 5.68s/it]
17%|█▋ | 7585/43939 [12:15:18<57:20:18, 5.68s/it]
17%|█▋ | 7586/43939 [12:15:24<57:21:33, 5.68s/it]
17%|█▋ | 7587/43939 [12:15:29<57:21:56, 5.68s/it]
17%|█▋ | 7588/43939 [12:15:35<57:18:05, 5.67s/it]
17%|█▋ | 7589/43939 [12:15:41<58:03:22, 5.75s/it]
17%|█▋ | 7590/43939 [12:15:47<57:50:30, 5.73s/it]
17%|█▋ | 7591/43939 [12:15:52<57:46:58, 5.72s/it]
17%|█▋ | 7592/43939 [12:15:58<57:40:39, 5.71s/it]
17%|█▋ | 7593/43939 [12:16:04<57:38:46, 5.71s/it]
17%|█▋ | 7594/43939 [12:16:09<57:34:00, 5.70s/it]
17%|█▋ | 7595/43939 [12:16:15<57:31:05, 5.70s/it]
17%|█▋ | 7596/43939 [12:16:21<57:33:13, 5.70s/it]
17%|█▋ | 7597/43939 [12:16:27<57:37:59, 5.71s/it]
17%|█▋ | 7598/43939 [12:16:32<57:33:19, 5.70s/it]
17%|█▋ | 7599/43939 [12:16:38<57:22:49, 5.68s/it]
17%|█▋ | 7600/43939 [12:16:44<57:26:00, 5.69s/it]
{'loss': 1.0229, 'grad_norm': 1.153113603591919, 'learning_rate': 4.9193540064568724e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.22, 'epoch': 0.69}
17%|█▋ | 7600/43939 [12:16:44<57:26:00, 5.69s/it]
17%|█▋ | 7601/43939 [12:16:49<57:33:53, 5.70s/it]
17%|█▋ | 7602/43939 [12:16:55<57:27:58, 5.69s/it]
17%|█▋ | 7603/43939 [12:17:01<57:30:31, 5.70s/it]
17%|█▋ | 7604/43939 [12:17:06<57:30:03, 5.70s/it]
17%|█▋ | 7605/43939 [12:17:12<57:40:32, 5.71s/it]
17%|█▋ | 7606/43939 [12:17:18<57:27:02, 5.69s/it]
17%|█▋ | 7607/43939 [12:17:24<57:37:52, 5.71s/it]
17%|█▋ | 7608/43939 [12:17:29<57:36:22, 5.71s/it]
17%|█▋ | 7609/43939 [12:17:35<57:28:01, 5.69s/it]
17%|█▋ | 7610/43939 [12:17:41<57:34:27, 5.71s/it]
17%|█▋ | 7611/43939 [12:17:47<58:26:02, 5.79s/it]
17%|█▋ | 7612/43939 [12:17:52<58:07:55, 5.76s/it]
17%|█▋ | 7613/43939 [12:17:58<57:59:16, 5.75s/it]
17%|█▋ | 7614/43939 [12:18:04<57:59:04, 5.75s/it]
17%|█▋ | 7615/43939 [12:18:09<57:50:36, 5.73s/it]
17%|█▋ | 7616/43939 [12:18:15<57:48:11, 5.73s/it]
17%|█▋ | 7617/43939 [12:18:21<57:40:51, 5.72s/it]
17%|█▋ | 7618/43939 [12:18:27<57:40:45, 5.72s/it]
17%|█▋ | 7619/43939 [12:18:32<57:35:05, 5.71s/it]
17%|█▋ | 7620/43939 [12:18:38<57:29:18, 5.70s/it]
17%|█▋ | 7621/43939 [12:18:44<57:28:44, 5.70s/it]
17%|█▋ | 7622/43939 [12:18:49<57:28:27, 5.70s/it]
17%|█▋ | 7623/43939 [12:18:55<57:26:18, 5.69s/it]
17%|█▋ | 7624/43939 [12:19:01<57:25:22, 5.69s/it]
17%|█▋ | 7625/43939 [12:19:06<57:23:25, 5.69s/it]
17%|█▋ | 7626/43939 [12:19:12<57:24:22, 5.69s/it]
17%|█▋ | 7627/43939 [12:19:18<57:38:01, 5.71s/it]
17%|█▋ | 7628/43939 [12:19:24<57:26:39, 5.70s/it]
17%|█▋ | 7629/43939 [12:19:29<57:26:16, 5.69s/it]
17%|█▋ | 7630/43939 [12:19:35<57:33:14, 5.71s/it]
17%|█▋ | 7631/43939 [12:19:41<57:34:00, 5.71s/it]
17%|█▋ | 7632/43939 [12:19:47<58:12:36, 5.77s/it]
17%|█▋ | 7633/43939 [12:19:52<57:53:23, 5.74s/it]
17%|█▋ | 7634/43939 [12:19:58<57:47:39, 5.73s/it]
17%|█▋ | 7635/43939 [12:20:04<57:42:01, 5.72s/it]
17%|█▋ | 7636/43939 [12:20:09<57:33:21, 5.71s/it]
17%|█▋ | 7637/43939 [12:20:15<57:27:42, 5.70s/it]
17%|█▋ | 7638/43939 [12:20:21<57:21:53, 5.69s/it]
17%|█▋ | 7639/43939 [12:20:26<57:18:08, 5.68s/it]
17%|█▋ | 7640/43939 [12:20:32<57:37:04, 5.71s/it]
17%|█▋ | 7641/43939 [12:20:38<57:38:31, 5.72s/it]
17%|█▋ | 7642/43939 [12:20:43<57:26:47, 5.70s/it]
17%|█▋ | 7643/43939 [12:20:49<57:34:39, 5.71s/it]
17%|█▋ | 7644/43939 [12:20:55<57:28:08, 5.70s/it]
17%|█▋ | 7645/43939 [12:21:01<57:25:13, 5.70s/it]
17%|█▋ | 7646/43939 [12:21:06<57:25:23, 5.70s/it]
17%|█▋ | 7647/43939 [12:21:12<57:35:52, 5.71s/it]
17%|█▋ | 7648/43939 [12:21:18<57:42:18, 5.72s/it]
17%|█▋ | 7649/43939 [12:21:23<57:35:30, 5.71s/it]
17%|█▋ | 7650/43939 [12:21:29<57:27:19, 5.70s/it]
{'loss': 1.0183, 'grad_norm': 1.1115753650665283, 'learning_rate': 4.916833068990404e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.87, 'epoch': 0.7}
17%|█▋ | 7650/43939 [12:21:29<57:27:19, 5.70s/it]
17%|█▋ | 7651/43939 [12:21:35<57:33:04, 5.71s/it]
17%|█▋ | 7652/43939 [12:21:41<57:27:15, 5.70s/it]
17%|█▋ | 7653/43939 [12:21:46<58:04:31, 5.76s/it]
17%|█▋ | 7654/43939 [12:21:52<57:58:05, 5.75s/it]
17%|█▋ | 7655/43939 [12:21:58<57:46:27, 5.73s/it]
17%|█▋ | 7656/43939 [12:22:04<57:33:13, 5.71s/it]
17%|█▋ | 7657/43939 [12:22:09<57:35:34, 5.71s/it]
17%|█▋ | 7658/43939 [12:22:15<57:28:58, 5.70s/it]
17%|█▋ | 7659/43939 [12:22:21<57:28:57, 5.70s/it]
17%|█▋ | 7660/43939 [12:22:26<57:31:27, 5.71s/it]
17%|█▋ | 7661/43939 [12:22:32<57:36:18, 5.72s/it]
17%|█▋ | 7662/43939 [12:22:38<57:33:24, 5.71s/it]
17%|█▋ | 7663/43939 [12:22:43<57:28:43, 5.70s/it]
17%|█▋ | 7664/43939 [12:22:49<57:25:20, 5.70s/it]
17%|█▋ | 7665/43939 [12:22:55<57:31:47, 5.71s/it]
17%|█▋ | 7666/43939 [12:23:01<57:26:35, 5.70s/it]
17%|█▋ | 7667/43939 [12:23:06<57:24:30, 5.70s/it]
17%|█▋ | 7668/43939 [12:23:12<57:29:20, 5.71s/it]
17%|█▋ | 7669/43939 [12:23:18<57:29:52, 5.71s/it]
17%|█▋ | 7670/43939 [12:23:23<57:28:29, 5.70s/it]
17%|█▋ | 7671/43939 [12:23:29<57:31:35, 5.71s/it]
17%|█▋ | 7672/43939 [12:23:35<57:27:39, 5.70s/it]
17%|█▋ | 7673/43939 [12:23:41<57:31:59, 5.71s/it]
17%|█▋ | 7674/43939 [12:23:47<58:13:44, 5.78s/it]
17%|█▋ | 7675/43939 [12:23:52<57:53:38, 5.75s/it]
17%|█▋ | 7676/43939 [12:23:58<57:43:54, 5.73s/it]
17%|█▋ | 7677/43939 [12:24:04<57:41:02, 5.73s/it]
17%|█▋ | 7678/43939 [12:24:09<57:32:57, 5.71s/it]
17%|█▋ | 7679/43939 [12:24:15<57:36:11, 5.72s/it]
17%|█▋ | 7680/43939 [12:24:21<57:36:03, 5.72s/it]
17%|█▋ | 7681/43939 [12:24:26<57:24:16, 5.70s/it]
17%|█▋ | 7682/43939 [12:24:32<57:23:56, 5.70s/it]
17%|█▋ | 7683/43939 [12:24:38<57:21:07, 5.69s/it]
17%|█▋ | 7684/43939 [12:24:44<57:33:02, 5.71s/it]
17%|█▋ | 7685/43939 [12:24:49<57:29:12, 5.71s/it]
17%|█▋ | 7686/43939 [12:24:55<57:30:17, 5.71s/it]
17%|█▋ | 7687/43939 [12:25:01<57:28:05, 5.71s/it]
17%|█▋ | 7688/43939 [12:25:06<57:29:25, 5.71s/it]
17%|█▋ | 7689/43939 [12:25:12<57:26:55, 5.71s/it]
18%|█▊ | 7690/43939 [12:25:18<57:25:34, 5.70s/it]
18%|█▊ | 7691/43939 [12:25:23<57:29:16, 5.71s/it]
18%|█▊ | 7692/43939 [12:25:29<57:27:32, 5.71s/it]
18%|█▊ | 7693/43939 [12:25:35<57:29:39, 5.71s/it]
18%|█▊ | 7694/43939 [12:25:41<57:28:58, 5.71s/it]
18%|█▊ | 7695/43939 [12:25:47<58:08:22, 5.77s/it]
18%|█▊ | 7696/43939 [12:25:52<57:52:02, 5.75s/it]
18%|█▊ | 7697/43939 [12:25:58<57:41:33, 5.73s/it]
18%|█▊ | 7698/43939 [12:26:04<57:42:57, 5.73s/it]
18%|█▊ | 7699/43939 [12:26:09<57:46:50, 5.74s/it]
18%|█▊ | 7700/43939 [12:26:15<57:38:28, 5.73s/it]
{'loss': 1.0284, 'grad_norm': 1.1606495380401611, 'learning_rate': 4.914274000302793e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.9, 'epoch': 0.7}
18%|█▊ | 7700/43939 [12:26:15<57:38:28, 5.73s/it]
18%|█▊ | 7701/43939 [12:26:21<57:30:32, 5.71s/it]
18%|█▊ | 7702/43939 [12:26:26<57:26:37, 5.71s/it]
18%|█▊ | 7703/43939 [12:26:32<57:25:46, 5.71s/it]
18%|█▊ | 7704/43939 [12:26:38<57:19:55, 5.70s/it]
18%|█▊ | 7705/43939 [12:26:44<57:20:27, 5.70s/it]
18%|█▊ | 7706/43939 [12:26:49<57:11:32, 5.68s/it]
18%|█▊ | 7707/43939 [12:26:55<57:07:57, 5.68s/it]
18%|█▊ | 7708/43939 [12:27:01<57:18:26, 5.69s/it]
18%|█▊ | 7709/43939 [12:27:06<57:18:04, 5.69s/it]
18%|█▊ | 7710/43939 [12:27:12<57:28:53, 5.71s/it]
18%|█▊ | 7711/43939 [12:27:18<57:22:26, 5.70s/it]
18%|█▊ | 7712/43939 [12:27:23<57:18:14, 5.69s/it]
18%|█▊ | 7713/43939 [12:27:29<57:13:23, 5.69s/it]
18%|█▊ | 7714/43939 [12:27:35<57:15:13, 5.69s/it]
18%|█▊ | 7715/43939 [12:27:40<57:19:09, 5.70s/it]
18%|█▊ | 7716/43939 [12:27:46<57:59:49, 5.76s/it]
18%|█▊ | 7717/43939 [12:27:52<57:43:40, 5.74s/it]
18%|█▊ | 7718/43939 [12:27:58<57:38:31, 5.73s/it]
18%|█▊ | 7719/43939 [12:28:03<57:30:26, 5.72s/it]
18%|█▊ | 7720/43939 [12:28:09<57:24:02, 5.71s/it]
18%|█▊ | 7721/43939 [12:28:15<57:20:33, 5.70s/it]
18%|█▊ | 7722/43939 [12:28:20<57:12:38, 5.69s/it]
18%|█▊ | 7723/43939 [12:28:26<57:12:34, 5.69s/it]
18%|█▊ | 7724/43939 [12:28:32<57:13:35, 5.69s/it]
18%|█▊ | 7725/43939 [12:28:38<57:18:09, 5.70s/it]
18%|█▊ | 7726/43939 [12:28:43<57:10:53, 5.68s/it]
18%|█▊ | 7727/43939 [12:28:49<57:05:04, 5.68s/it]
18%|█▊ | 7728/43939 [12:28:55<57:07:09, 5.68s/it]
18%|█▊ | 7729/43939 [12:29:00<57:15:44, 5.69s/it]
18%|█▊ | 7730/43939 [12:29:06<57:09:36, 5.68s/it]
18%|█▊ | 7731/43939 [12:29:12<57:10:24, 5.68s/it]
18%|█▊ | 7732/43939 [12:29:17<57:09:12, 5.68s/it]
18%|█▊ | 7733/43939 [12:29:23<57:11:16, 5.69s/it]
18%|█▊ | 7734/43939 [12:29:29<57:19:09, 5.70s/it]
18%|█▊ | 7735/43939 [12:29:34<57:29:27, 5.72s/it]
18%|█▊ | 7736/43939 [12:29:40<58:08:32, 5.78s/it]
18%|█▊ | 7737/43939 [12:29:46<57:51:52, 5.75s/it]
18%|█▊ | 7738/43939 [12:29:52<57:38:27, 5.73s/it]
18%|█▊ | 7739/43939 [12:29:57<57:31:15, 5.72s/it]
18%|█▊ | 7740/43939 [12:30:03<57:34:25, 5.73s/it]
18%|█▊ | 7741/43939 [12:30:09<57:24:06, 5.71s/it]
18%|█▊ | 7742/43939 [12:30:15<57:13:13, 5.69s/it]
18%|█▊ | 7743/43939 [12:30:20<57:25:14, 5.71s/it]
18%|█▊ | 7744/43939 [12:30:26<57:18:09, 5.70s/it]
18%|█▊ | 7745/43939 [12:30:32<57:16:38, 5.70s/it]
18%|█▊ | 7746/43939 [12:30:37<57:10:09, 5.69s/it]
18%|█▊ | 7747/43939 [12:30:43<57:20:51, 5.70s/it]
18%|█▊ | 7748/43939 [12:30:49<57:14:52, 5.69s/it]
18%|█▊ | 7749/43939 [12:30:54<57:07:27, 5.68s/it]
18%|█▊ | 7750/43939 [12:31:00<57:10:13, 5.69s/it]
{'loss': 1.0377, 'grad_norm': 0.9907020330429077, 'learning_rate': 4.911676840769361e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.59, 'epoch': 0.71}
18%|█▊ | 7750/43939 [12:31:00<57:10:13, 5.69s/it]
18%|█▊ | 7751/43939 [12:31:06<57:14:37, 5.69s/it]
18%|█▊ | 7752/43939 [12:31:12<57:18:25, 5.70s/it]
18%|█▊ | 7753/43939 [12:31:17<57:15:13, 5.70s/it]
18%|█▊ | 7754/43939 [12:31:23<57:18:34, 5.70s/it]
18%|█▊ | 7755/43939 [12:31:29<57:12:19, 5.69s/it]
18%|█▊ | 7756/43939 [12:31:34<57:18:20, 5.70s/it]
18%|█▊ | 7757/43939 [12:31:40<57:18:27, 5.70s/it]
18%|█▊ | 7758/43939 [12:31:46<58:01:34, 5.77s/it]
18%|█▊ | 7759/43939 [12:31:52<57:51:29, 5.76s/it]
18%|█▊ | 7760/43939 [12:31:57<57:48:11, 5.75s/it]
18%|█▊ | 7761/43939 [12:32:03<57:50:58, 5.76s/it]
18%|█▊ | 7762/43939 [12:32:09<57:49:16, 5.75s/it]
18%|█▊ | 7763/43939 [12:32:15<57:38:09, 5.74s/it]
18%|█▊ | 7764/43939 [12:32:20<57:36:08, 5.73s/it]
18%|█▊ | 7765/43939 [12:32:26<57:30:55, 5.72s/it]
18%|█▊ | 7766/43939 [12:32:32<57:23:06, 5.71s/it]
18%|█▊ | 7767/43939 [12:32:37<57:16:59, 5.70s/it]
18%|█▊ | 7768/43939 [12:32:43<57:09:08, 5.69s/it]
18%|█▊ | 7769/43939 [12:32:49<57:08:12, 5.69s/it]
18%|█▊ | 7770/43939 [12:32:54<57:17:07, 5.70s/it]
18%|█▊ | 7771/43939 [12:33:00<57:18:13, 5.70s/it]
18%|█▊ | 7772/43939 [12:33:06<57:13:10, 5.70s/it]
18%|█▊ | 7773/43939 [12:33:12<57:14:46, 5.70s/it]
18%|█▊ | 7774/43939 [12:33:17<57:07:45, 5.69s/it]
18%|█▊ | 7775/43939 [12:33:23<57:14:06, 5.70s/it]
18%|█▊ | 7776/43939 [12:33:29<57:15:11, 5.70s/it]
18%|█▊ | 7777/43939 [12:33:34<57:12:59, 5.70s/it]
18%|█▊ | 7778/43939 [12:33:40<57:56:46, 5.77s/it]
18%|█▊ | 7779/43939 [12:33:46<57:36:47, 5.74s/it]
18%|█▊ | 7780/43939 [12:33:52<57:30:09, 5.72s/it]
18%|█▊ | 7781/43939 [12:33:57<57:23:50, 5.71s/it]
18%|█▊ | 7782/43939 [12:34:03<57:21:42, 5.71s/it]
18%|█▊ | 7783/43939 [12:34:09<57:20:18, 5.71s/it]
18%|█▊ | 7784/43939 [12:34:14<57:21:29, 5.71s/it]
18%|█▊ | 7785/43939 [12:34:20<57:18:58, 5.71s/it]
18%|█▊ | 7786/43939 [12:34:26<57:19:40, 5.71s/it]
18%|█▊ | 7787/43939 [12:34:32<57:15:45, 5.70s/it]
18%|█▊ | 7788/43939 [12:34:37<57:24:17, 5.72s/it]
18%|█▊ | 7789/43939 [12:34:43<57:15:29, 5.70s/it]
18%|█▊ | 7790/43939 [12:34:49<57:26:33, 5.72s/it]
18%|█▊ | 7791/43939 [12:34:54<57:25:54, 5.72s/it]
18%|█▊ | 7792/43939 [12:35:00<57:23:22, 5.72s/it]
18%|█▊ | 7793/43939 [12:35:06<57:15:22, 5.70s/it]
18%|█▊ | 7794/43939 [12:35:12<57:16:08, 5.70s/it]
18%|█▊ | 7795/43939 [12:35:17<57:17:57, 5.71s/it]
18%|█▊ | 7796/43939 [12:35:23<57:23:58, 5.72s/it]
18%|█▊ | 7797/43939 [12:35:29<57:22:49, 5.72s/it]
18%|█▊ | 7798/43939 [12:35:34<57:14:26, 5.70s/it]
18%|█▊ | 7799/43939 [12:35:40<57:59:02, 5.78s/it]
18%|█▊ | 7800/43939 [12:35:46<57:45:21, 5.75s/it]
{'loss': 1.0421, 'grad_norm': 1.0251307487487793, 'learning_rate': 4.9090416313664036e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.37, 'epoch': 0.71}
18%|█▊ | 7800/43939 [12:35:46<57:45:21, 5.75s/it]
18%|█▊ | 7801/43939 [12:35:52<57:38:58, 5.74s/it]
18%|█▊ | 7802/43939 [12:35:57<57:31:16, 5.73s/it]
18%|█▊ | 7803/43939 [12:36:03<57:20:13, 5.71s/it]
18%|█▊ | 7804/43939 [12:36:09<57:17:29, 5.71s/it]
18%|█▊ | 7805/43939 [12:36:15<57:12:10, 5.70s/it]
18%|█▊ | 7806/43939 [12:36:20<57:12:28, 5.70s/it]
18%|█▊ | 7807/43939 [12:36:26<57:11:43, 5.70s/it]
18%|█▊ | 7808/43939 [12:36:32<57:11:45, 5.70s/it]
18%|█▊ | 7809/43939 [12:36:37<57:15:09, 5.70s/it]
18%|█▊ | 7810/43939 [12:36:43<57:20:56, 5.71s/it]
18%|█▊ | 7811/43939 [12:36:49<57:27:51, 5.73s/it]
18%|█▊ | 7812/43939 [12:36:54<57:16:58, 5.71s/it]
18%|█▊ | 7813/43939 [12:37:00<57:15:51, 5.71s/it]
18%|█▊ | 7814/43939 [12:37:06<57:19:29, 5.71s/it]
18%|█▊ | 7815/43939 [12:37:12<57:11:10, 5.70s/it]
18%|█▊ | 7816/43939 [12:37:17<57:10:43, 5.70s/it]
18%|█▊ | 7817/43939 [12:37:23<57:12:41, 5.70s/it]
18%|█▊ | 7818/43939 [12:37:29<57:09:19, 5.70s/it]
18%|█▊ | 7819/43939 [12:37:34<57:12:47, 5.70s/it]
18%|█▊ | 7820/43939 [12:37:40<57:10:09, 5.70s/it]
18%|█▊ | 7821/43939 [12:37:46<58:00:25, 5.78s/it]
18%|█▊ | 7822/43939 [12:37:52<57:51:18, 5.77s/it]
18%|█▊ | 7823/43939 [12:37:57<57:34:54, 5.74s/it]
18%|█▊ | 7824/43939 [12:38:03<57:31:57, 5.73s/it]
18%|█▊ | 7825/43939 [12:38:09<57:20:02, 5.72s/it]
18%|█▊ | 7826/43939 [12:38:15<57:13:42, 5.70s/it]
18%|█▊ | 7827/43939 [12:38:20<57:11:47, 5.70s/it]
18%|█▊ | 7828/43939 [12:38:26<57:23:47, 5.72s/it]
18%|█▊ | 7829/43939 [12:38:32<57:26:35, 5.73s/it]
18%|█▊ | 7830/43939 [12:38:37<57:19:12, 5.71s/it]
18%|█▊ | 7831/43939 [12:38:43<57:10:39, 5.70s/it]
18%|█▊ | 7832/43939 [12:38:49<57:17:26, 5.71s/it]
18%|█▊ | 7833/43939 [12:38:54<57:09:21, 5.70s/it]
18%|█▊ | 7834/43939 [12:39:00<57:12:29, 5.70s/it]
18%|█▊ | 7835/43939 [12:39:06<57:11:43, 5.70s/it]
18%|█▊ | 7836/43939 [12:39:12<57:05:15, 5.69s/it]
18%|█▊ | 7837/43939 [12:39:17<57:14:53, 5.71s/it]
18%|█▊ | 7838/43939 [12:39:23<57:06:58, 5.70s/it]
18%|█▊ | 7839/43939 [12:39:29<57:09:24, 5.70s/it]
18%|█▊ | 7840/43939 [12:39:34<57:01:31, 5.69s/it]
18%|█▊ | 7841/43939 [12:39:40<57:18:53, 5.72s/it]
18%|█▊ | 7842/43939 [12:39:46<57:55:03, 5.78s/it]
18%|█▊ | 7843/43939 [12:39:52<57:44:35, 5.76s/it]
18%|█▊ | 7844/43939 [12:39:57<57:36:28, 5.75s/it]
18%|█▊ | 7845/43939 [12:40:03<57:32:14, 5.74s/it]
18%|█▊ | 7846/43939 [12:40:09<57:28:00, 5.73s/it]
18%|█▊ | 7847/43939 [12:40:15<57:22:30, 5.72s/it]
18%|█▊ | 7848/43939 [12:40:20<57:13:34, 5.71s/it]
18%|█▊ | 7849/43939 [12:40:26<57:09:17, 5.70s/it]
18%|█▊ | 7850/43939 [12:40:32<57:12:23, 5.71s/it]
{'loss': 1.0792, 'grad_norm': 1.1614022254943848, 'learning_rate': 4.9063684136705404e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 518.05, 'epoch': 0.71}
18%|█▊ | 7850/43939 [12:40:32<57:12:23, 5.71s/it]
18%|█▊ | 7851/43939 [12:40:37<57:10:56, 5.70s/it]
18%|█▊ | 7852/43939 [12:40:43<57:07:50, 5.70s/it]
18%|█▊ | 7853/43939 [12:40:49<57:10:12, 5.70s/it]
18%|█▊ | 7854/43939 [12:40:54<57:08:47, 5.70s/it]
18%|█▊ | 7855/43939 [12:41:00<57:12:45, 5.71s/it]
18%|█▊ | 7856/43939 [12:41:06<57:07:42, 5.70s/it]
18%|█▊ | 7857/43939 [12:41:12<57:12:48, 5.71s/it]
18%|█▊ | 7858/43939 [12:41:17<57:12:04, 5.71s/it]
18%|█▊ | 7859/43939 [12:41:23<57:06:21, 5.70s/it]
18%|█▊ | 7860/43939 [12:41:29<57:19:19, 5.72s/it]
18%|█▊ | 7861/43939 [12:41:34<57:18:26, 5.72s/it]
18%|█▊ | 7862/43939 [12:41:40<58:08:24, 5.80s/it]
18%|█▊ | 7863/43939 [12:41:46<57:51:10, 5.77s/it]
18%|█▊ | 7864/43939 [12:41:52<57:38:57, 5.75s/it]
18%|█▊ | 7865/43939 [12:41:58<57:24:26, 5.73s/it]
18%|█▊ | 7866/43939 [12:42:03<57:24:17, 5.73s/it]
18%|█▊ | 7867/43939 [12:42:09<57:12:50, 5.71s/it]
18%|█▊ | 7868/43939 [12:42:15<57:13:01, 5.71s/it]
18%|█▊ | 7869/43939 [12:42:20<57:10:13, 5.71s/it]
18%|█▊ | 7870/43939 [12:42:26<57:05:41, 5.70s/it]
18%|█▊ | 7871/43939 [12:42:32<57:11:21, 5.71s/it]
18%|█▊ | 7872/43939 [12:42:37<57:11:00, 5.71s/it]
18%|█▊ | 7873/43939 [12:42:43<57:09:12, 5.70s/it]
18%|█▊ | 7874/43939 [12:42:49<57:13:40, 5.71s/it]
18%|█▊ | 7875/43939 [12:42:55<57:14:25, 5.71s/it]
18%|█▊ | 7876/43939 [12:43:00<57:10:43, 5.71s/it]
18%|█▊ | 7877/43939 [12:43:06<57:14:31, 5.71s/it]
18%|█▊ | 7878/43939 [12:43:12<57:16:55, 5.72s/it]
18%|█▊ | 7879/43939 [12:43:18<57:16:23, 5.72s/it]
18%|█▊ | 7880/43939 [12:43:23<57:16:42, 5.72s/it]
18%|█▊ | 7881/43939 [12:43:29<57:09:06, 5.71s/it]
18%|█▊ | 7882/43939 [12:43:35<57:04:14, 5.70s/it]
18%|█▊ | 7883/43939 [12:43:40<57:02:21, 5.70s/it]
18%|█▊ | 7884/43939 [12:43:46<57:54:30, 5.78s/it]
18%|█▊ | 7885/43939 [12:43:52<57:44:25, 5.77s/it]
18%|█▊ | 7886/43939 [12:43:58<57:37:41, 5.75s/it]
18%|█▊ | 7887/43939 [12:44:03<57:18:52, 5.72s/it]
18%|█▊ | 7888/43939 [12:44:09<57:12:19, 5.71s/it]
18%|█▊ | 7889/43939 [12:44:15<57:16:02, 5.72s/it]
18%|█▊ | 7890/43939 [12:44:20<57:12:08, 5.71s/it]
18%|█▊ | 7891/43939 [12:44:26<57:06:29, 5.70s/it]
18%|█▊ | 7892/43939 [12:44:32<57:13:47, 5.72s/it]
18%|█▊ | 7893/43939 [12:44:38<57:20:27, 5.73s/it]
18%|█▊ | 7894/43939 [12:44:43<57:18:08, 5.72s/it]
18%|█▊ | 7895/43939 [12:44:49<57:09:09, 5.71s/it]
18%|█▊ | 7896/43939 [12:44:55<57:07:37, 5.71s/it]
18%|█▊ | 7897/43939 [12:45:00<57:05:19, 5.70s/it]
18%|█▊ | 7898/43939 [12:45:06<57:07:05, 5.71s/it]
18%|█▊ | 7899/43939 [12:45:12<57:08:35, 5.71s/it]
18%|█▊ | 7900/43939 [12:45:18<57:07:41, 5.71s/it]
{'loss': 1.0356, 'grad_norm': 0.9551727175712585, 'learning_rate': 4.903657229858063e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.4, 'epoch': 0.72}
18%|█▊ | 7900/43939 [12:45:18<57:07:41, 5.71s/it]
18%|█▊ | 7901/43939 [12:45:23<56:58:18, 5.69s/it]
18%|█▊ | 7902/43939 [12:45:29<56:59:44, 5.69s/it]
18%|█▊ | 7903/43939 [12:45:35<56:59:04, 5.69s/it]
18%|█▊ | 7904/43939 [12:45:40<57:02:33, 5.70s/it]
18%|█▊ | 7905/43939 [12:45:46<57:56:16, 5.79s/it]
18%|█▊ | 7906/43939 [12:45:52<57:38:46, 5.76s/it]
18%|█▊ | 7907/43939 [12:45:58<57:24:53, 5.74s/it]
18%|█▊ | 7908/43939 [12:46:03<57:19:28, 5.73s/it]
18%|█▊ | 7909/43939 [12:46:09<57:03:24, 5.70s/it]
18%|█▊ | 7910/43939 [12:46:15<57:02:59, 5.70s/it]
18%|█▊ | 7911/43939 [12:46:20<57:01:44, 5.70s/it]
18%|█▊ | 7912/43939 [12:46:26<57:11:19, 5.71s/it]
18%|█▊ | 7913/43939 [12:46:32<57:10:48, 5.71s/it]
18%|█▊ | 7914/43939 [12:46:38<57:04:58, 5.70s/it]
18%|█▊ | 7915/43939 [12:46:43<57:00:37, 5.70s/it]
18%|█▊ | 7916/43939 [12:46:49<57:09:09, 5.71s/it]
18%|█▊ | 7917/43939 [12:46:55<57:01:23, 5.70s/it]
18%|█▊ | 7918/43939 [12:47:00<56:58:22, 5.69s/it]
18%|█▊ | 7919/43939 [12:47:06<57:00:45, 5.70s/it]
18%|█▊ | 7920/43939 [12:47:12<56:59:18, 5.70s/it]
18%|█▊ | 7921/43939 [12:47:18<57:06:47, 5.71s/it]
18%|█▊ | 7922/43939 [12:47:23<57:08:59, 5.71s/it]
18%|█▊ | 7923/43939 [12:47:29<57:05:32, 5.71s/it]
18%|█▊ | 7924/43939 [12:47:35<57:05:11, 5.71s/it]
18%|█▊ | 7925/43939 [12:47:40<57:03:09, 5.70s/it]
18%|█▊ | 7926/43939 [12:47:46<57:49:25, 5.78s/it]
18%|█▊ | 7927/43939 [12:47:52<57:38:21, 5.76s/it]
18%|█▊ | 7928/43939 [12:47:58<57:28:16, 5.75s/it]
18%|█▊ | 7929/43939 [12:48:03<57:21:57, 5.74s/it]
18%|█▊ | 7930/43939 [12:48:09<57:17:16, 5.73s/it]
18%|█▊ | 7931/43939 [12:48:15<57:10:33, 5.72s/it]
18%|█▊ | 7932/43939 [12:48:20<57:00:28, 5.70s/it]
18%|█▊ | 7933/43939 [12:48:26<57:03:01, 5.70s/it]
18%|█▊ | 7934/43939 [12:48:32<57:02:25, 5.70s/it]
18%|█▊ | 7935/43939 [12:48:38<56:59:59, 5.70s/it]
18%|█▊ | 7936/43939 [12:48:43<57:06:58, 5.71s/it]
18%|█▊ | 7937/43939 [12:48:49<57:05:18, 5.71s/it]
18%|█▊ | 7938/43939 [12:48:55<56:58:00, 5.70s/it]
18%|█▊ | 7939/43939 [12:49:00<56:52:05, 5.69s/it]
18%|█▊ | 7940/43939 [12:49:06<56:54:31, 5.69s/it]
18%|█▊ | 7941/43939 [12:49:12<56:53:32, 5.69s/it]
18%|█▊ | 7942/43939 [12:49:17<57:01:07, 5.70s/it]
18%|█▊ | 7943/43939 [12:49:23<56:59:41, 5.70s/it]
18%|█▊ | 7944/43939 [12:49:29<57:09:42, 5.72s/it]
18%|█▊ | 7945/43939 [12:49:35<57:06:13, 5.71s/it]
18%|█▊ | 7946/43939 [12:49:40<57:05:49, 5.71s/it]
18%|█▊ | 7947/43939 [12:49:46<57:45:31, 5.78s/it]
18%|█▊ | 7948/43939 [12:49:52<57:31:45, 5.75s/it]
18%|█▊ | 7949/43939 [12:49:58<57:19:49, 5.73s/it]
18%|█▊ | 7950/43939 [12:50:03<57:14:48, 5.73s/it]
{'loss': 1.0267, 'grad_norm': 0.9774997234344482, 'learning_rate': 4.900908122704269e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.2, 'epoch': 0.72}
18%|█▊ | 7950/43939 [12:50:03<57:14:48, 5.73s/it]
18%|█▊ | 7951/43939 [12:50:09<57:11:36, 5.72s/it]
18%|█▊ | 7952/43939 [12:50:15<57:08:28, 5.72s/it]
18%|█▊ | 7953/43939 [12:50:21<57:15:31, 5.73s/it]
18%|█▊ | 7954/43939 [12:50:26<57:09:46, 5.72s/it]
18%|█▊ | 7955/43939 [12:50:32<57:09:48, 5.72s/it]
18%|█▊ | 7956/43939 [12:50:38<57:06:32, 5.71s/it]
18%|█▊ | 7957/43939 [12:50:43<57:01:10, 5.70s/it]
18%|█▊ | 7958/43939 [12:50:49<56:59:23, 5.70s/it]
18%|█▊ | 7959/43939 [12:50:55<56:58:16, 5.70s/it]
18%|█▊ | 7960/43939 [12:51:00<56:54:16, 5.69s/it]
18%|█▊ | 7961/43939 [12:51:06<56:54:06, 5.69s/it]
18%|█▊ | 7962/43939 [12:51:12<56:56:18, 5.70s/it]
18%|█▊ | 7963/43939 [12:51:18<56:57:56, 5.70s/it]
18%|█▊ | 7964/43939 [12:51:23<56:55:13, 5.70s/it]
18%|█▊ | 7965/43939 [12:51:29<57:01:19, 5.71s/it]
18%|█▊ | 7966/43939 [12:51:35<56:57:22, 5.70s/it]
18%|█▊ | 7967/43939 [12:51:40<57:01:28, 5.71s/it]
18%|█▊ | 7968/43939 [12:51:46<57:37:18, 5.77s/it]
18%|█▊ | 7969/43939 [12:51:52<57:18:30, 5.74s/it]
18%|█▊ | 7970/43939 [12:51:58<57:06:09, 5.72s/it]
18%|█▊ | 7971/43939 [12:52:03<57:04:16, 5.71s/it]
18%|█▊ | 7972/43939 [12:52:09<56:58:58, 5.70s/it]
18%|█▊ | 7973/43939 [12:52:15<57:03:02, 5.71s/it]
18%|█▊ | 7974/43939 [12:52:20<56:53:27, 5.69s/it]
18%|█▊ | 7975/43939 [12:52:26<56:53:23, 5.69s/it]
18%|█▊ | 7976/43939 [12:52:32<56:55:56, 5.70s/it]
18%|█▊ | 7977/43939 [12:52:37<56:53:45, 5.70s/it]
18%|█▊ | 7978/43939 [12:52:43<56:53:37, 5.70s/it]
18%|█▊ | 7979/43939 [12:52:49<56:48:29, 5.69s/it]
18%|█▊ | 7980/43939 [12:52:55<56:55:01, 5.70s/it]
18%|█▊ | 7981/43939 [12:53:00<56:57:31, 5.70s/it]
18%|█▊ | 7982/43939 [12:53:06<57:00:12, 5.71s/it]
18%|█▊ | 7983/43939 [12:53:12<57:00:23, 5.71s/it]
18%|█▊ | 7984/43939 [12:53:17<56:54:30, 5.70s/it]
18%|█▊ | 7985/43939 [12:53:23<56:57:11, 5.70s/it]
18%|█▊ | 7986/43939 [12:53:29<56:57:59, 5.70s/it]
18%|█▊ | 7987/43939 [12:53:34<56:58:04, 5.70s/it]
18%|█▊ | 7988/43939 [12:53:40<57:07:13, 5.72s/it]
18%|█▊ | 7989/43939 [12:53:46<57:45:32, 5.78s/it]
18%|█▊ | 7990/43939 [12:53:52<57:40:59, 5.78s/it]
18%|█▊ | 7991/43939 [12:53:58<57:24:17, 5.75s/it]
18%|█▊ | 7992/43939 [12:54:03<57:15:50, 5.73s/it]
18%|█▊ | 7993/43939 [12:54:09<57:14:38, 5.73s/it]
18%|█▊ | 7994/43939 [12:54:15<57:09:16, 5.72s/it]
18%|█▊ | 7995/43939 [12:54:20<57:05:11, 5.72s/it]
18%|█▊ | 7996/43939 [12:54:26<56:53:45, 5.70s/it]
18%|█▊ | 7997/43939 [12:54:32<56:53:41, 5.70s/it]
18%|█▊ | 7998/43939 [12:54:37<56:50:47, 5.69s/it]
18%|█▊ | 7999/43939 [12:54:43<56:53:28, 5.70s/it]
18%|█▊ | 8000/43939 [12:54:49<56:53:00, 5.70s/it]
{'loss': 1.0426, 'grad_norm': 1.1599117517471313, 'learning_rate': 4.898121135582782e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.25, 'epoch': 0.73}
18%|█▊ | 8000/43939 [12:54:49<56:53:00, 5.70s/it]
18%|█▊ | 8001/43939 [12:54:55<56:50:39, 5.69s/it]
18%|█▊ | 8002/43939 [12:55:00<56:53:34, 5.70s/it]
18%|█▊ | 8003/43939 [12:55:06<56:54:45, 5.70s/it]
18%|█▊ | 8004/43939 [12:55:12<57:07:25, 5.72s/it]
18%|█▊ | 8005/43939 [12:55:17<57:06:10, 5.72s/it]
18%|█▊ | 8006/43939 [12:55:23<56:59:58, 5.71s/it]
18%|█▊ | 8007/43939 [12:55:29<57:02:30, 5.71s/it]
18%|█▊ | 8008/43939 [12:55:35<57:03:42, 5.72s/it]
18%|█▊ | 8009/43939 [12:55:40<57:31:27, 5.76s/it]
18%|█▊ | 8010/43939 [12:55:46<57:21:20, 5.75s/it]
18%|█▊ | 8011/43939 [12:55:52<57:11:05, 5.73s/it]
18%|█▊ | 8012/43939 [12:55:58<57:01:02, 5.71s/it]
18%|█▊ | 8013/43939 [12:56:03<57:06:19, 5.72s/it]
18%|█▊ | 8014/43939 [12:56:09<57:05:44, 5.72s/it]
18%|█▊ | 8015/43939 [12:56:15<57:04:28, 5.72s/it]
18%|█▊ | 8016/43939 [12:56:20<56:59:46, 5.71s/it]
18%|█▊ | 8017/43939 [12:56:26<56:59:01, 5.71s/it]
18%|█▊ | 8018/43939 [12:56:32<56:53:32, 5.70s/it]
18%|█▊ | 8019/43939 [12:56:38<56:55:18, 5.70s/it]
18%|█▊ | 8020/43939 [12:56:43<56:49:32, 5.70s/it]
18%|█▊ | 8021/43939 [12:56:49<56:53:42, 5.70s/it]
18%|█▊ | 8022/43939 [12:56:55<56:52:17, 5.70s/it]
18%|█▊ | 8023/43939 [12:57:00<56:46:41, 5.69s/it]
18%|█▊ | 8024/43939 [12:57:06<56:57:20, 5.71s/it]
18%|█▊ | 8025/43939 [12:57:12<56:55:26, 5.71s/it]
18%|█▊ | 8026/43939 [12:57:17<56:55:05, 5.71s/it]
18%|█▊ | 8027/43939 [12:57:23<56:51:14, 5.70s/it]
18%|█▊ | 8028/43939 [12:57:29<56:45:28, 5.69s/it]
18%|█▊ | 8029/43939 [12:57:35<56:51:41, 5.70s/it]
18%|█▊ | 8030/43939 [12:57:40<56:52:09, 5.70s/it]
18%|█▊ | 8031/43939 [12:57:46<57:24:09, 5.75s/it]
18%|█▊ | 8032/43939 [12:57:52<57:14:41, 5.74s/it]
18%|█▊ | 8033/43939 [12:57:58<57:08:23, 5.73s/it]
18%|█▊ | 8034/43939 [12:58:03<57:09:12, 5.73s/it]
18%|█▊ | 8035/43939 [12:58:09<57:02:46, 5.72s/it]
18%|█▊ | 8036/43939 [12:58:15<57:01:00, 5.72s/it]
18%|█▊ | 8037/43939 [12:58:20<57:07:12, 5.73s/it]
18%|█▊ | 8038/43939 [12:58:26<57:01:24, 5.72s/it]
18%|█▊ | 8039/43939 [12:58:32<56:55:24, 5.71s/it]
18%|█▊ | 8040/43939 [12:58:37<56:53:54, 5.71s/it]
18%|█▊ | 8041/43939 [12:58:43<56:59:47, 5.72s/it]
18%|█▊ | 8042/43939 [12:58:49<56:52:59, 5.70s/it]
18%|█▊ | 8043/43939 [12:58:55<56:55:06, 5.71s/it]
18%|█▊ | 8044/43939 [12:59:00<56:59:55, 5.72s/it]
18%|█▊ | 8045/43939 [12:59:06<56:57:46, 5.71s/it]
18%|█▊ | 8046/43939 [12:59:12<56:52:30, 5.70s/it]
18%|█▊ | 8047/43939 [12:59:17<56:51:45, 5.70s/it]
18%|█▊ | 8048/43939 [12:59:23<56:50:01, 5.70s/it]
18%|█▊ | 8049/43939 [12:59:29<56:48:43, 5.70s/it]
18%|█▊ | 8050/43939 [12:59:35<56:44:47, 5.69s/it]
{'loss': 1.0496, 'grad_norm': 1.0983076095581055, 'learning_rate': 4.895296312464874e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.5, 'epoch': 0.73}
18%|█▊ | 8050/43939 [12:59:35<56:44:47, 5.69s/it]
18%|█▊ | 8051/43939 [12:59:40<56:51:13, 5.70s/it]
18%|█▊ | 8052/43939 [12:59:46<57:24:13, 5.76s/it]
18%|█▊ | 8053/43939 [12:59:52<57:09:42, 5.73s/it]
18%|█▊ | 8054/43939 [12:59:57<56:57:51, 5.71s/it]
18%|█▊ | 8055/43939 [13:00:03<56:55:58, 5.71s/it]
18%|█▊ | 8056/43939 [13:00:09<56:48:43, 5.70s/it]
18%|█▊ | 8057/43939 [13:00:15<56:47:38, 5.70s/it]
18%|█▊ | 8058/43939 [13:00:20<56:40:45, 5.69s/it]
18%|█▊ | 8059/43939 [13:00:26<56:41:34, 5.69s/it]
18%|█▊ | 8060/43939 [13:00:32<56:42:08, 5.69s/it]
18%|█▊ | 8061/43939 [13:00:37<56:44:53, 5.69s/it]
18%|█▊ | 8062/43939 [13:00:43<56:45:29, 5.70s/it]
18%|█▊ | 8063/43939 [13:00:49<56:46:26, 5.70s/it]
18%|█▊ | 8064/43939 [13:00:54<56:48:14, 5.70s/it]
18%|█▊ | 8065/43939 [13:01:00<56:37:53, 5.68s/it]
18%|█▊ | 8066/43939 [13:01:06<56:49:57, 5.70s/it]
18%|█▊ | 8067/43939 [13:01:11<56:39:13, 5.69s/it]
18%|█▊ | 8068/43939 [13:01:17<56:40:06, 5.69s/it]
18%|█▊ | 8069/43939 [13:01:23<56:41:39, 5.69s/it]
18%|█▊ | 8070/43939 [13:01:29<56:44:46, 5.70s/it]
18%|█▊ | 8071/43939 [13:01:34<56:47:42, 5.70s/it]
18%|█▊ | 8072/43939 [13:01:40<56:47:25, 5.70s/it]
18%|█▊ | 8073/43939 [13:01:46<57:32:33, 5.78s/it]
18%|█▊ | 8074/43939 [13:01:52<57:17:40, 5.75s/it]
18%|█▊ | 8075/43939 [13:01:57<57:02:07, 5.73s/it]
18%|█▊ | 8076/43939 [13:02:03<57:08:40, 5.74s/it]
18%|█▊ | 8077/43939 [13:02:09<57:03:15, 5.73s/it]
18%|█▊ | 8078/43939 [13:02:14<56:54:52, 5.71s/it]
18%|█▊ | 8079/43939 [13:02:20<56:50:43, 5.71s/it]
18%|█▊ | 8080/43939 [13:02:26<56:49:11, 5.70s/it]
18%|█▊ | 8081/43939 [13:02:32<56:52:49, 5.71s/it]
18%|█▊ | 8082/43939 [13:02:37<56:51:55, 5.71s/it]
18%|█▊ | 8083/43939 [13:02:43<56:46:38, 5.70s/it]
18%|█▊ | 8084/43939 [13:02:49<56:46:58, 5.70s/it]
18%|█▊ | 8085/43939 [13:02:54<56:46:02, 5.70s/it]
18%|█▊ | 8086/43939 [13:03:00<56:47:26, 5.70s/it]
18%|█▊ | 8087/43939 [13:03:06<56:50:20, 5.71s/it]
18%|█▊ | 8088/43939 [13:03:11<56:54:00, 5.71s/it]
18%|█▊ | 8089/43939 [13:03:17<57:02:36, 5.73s/it]
18%|█▊ | 8090/43939 [13:03:23<57:01:55, 5.73s/it]
18%|█▊ | 8091/43939 [13:03:29<57:11:54, 5.74s/it]
18%|█▊ | 8092/43939 [13:03:34<57:00:11, 5.72s/it]
18%|█▊ | 8093/43939 [13:03:40<56:51:51, 5.71s/it]
18%|█▊ | 8094/43939 [13:03:46<57:25:16, 5.77s/it]
18%|█▊ | 8095/43939 [13:03:52<57:10:35, 5.74s/it]
18%|█▊ | 8096/43939 [13:03:57<56:55:00, 5.72s/it]
18%|█▊ | 8097/43939 [13:04:03<56:50:58, 5.71s/it]
18%|█▊ | 8098/43939 [13:04:09<56:47:51, 5.70s/it]
18%|█▊ | 8099/43939 [13:04:14<56:50:23, 5.71s/it]
18%|█▊ | 8100/43939 [13:04:20<56:45:03, 5.70s/it]
{'loss': 0.9954, 'grad_norm': 1.1329846382141113, 'learning_rate': 4.892433697918768e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.45, 'epoch': 0.74}
18%|█▊ | 8100/43939 [13:04:20<56:45:03, 5.70s/it]
18%|█▊ | 8101/43939 [13:04:26<56:46:12, 5.70s/it]
18%|█▊ | 8102/43939 [13:04:32<56:43:18, 5.70s/it]
18%|█▊ | 8103/43939 [13:04:37<56:39:58, 5.69s/it]
18%|█▊ | 8104/43939 [13:04:43<56:41:27, 5.70s/it]
18%|█▊ | 8105/43939 [13:04:49<56:41:59, 5.70s/it]
18%|█▊ | 8106/43939 [13:04:54<56:51:02, 5.71s/it]
18%|█▊ | 8107/43939 [13:05:00<56:46:05, 5.70s/it]
18%|█▊ | 8108/43939 [13:05:06<56:48:26, 5.71s/it]
18%|█▊ | 8109/43939 [13:05:11<56:46:54, 5.71s/it]
18%|█▊ | 8110/43939 [13:05:17<57:00:44, 5.73s/it]
18%|█▊ | 8111/43939 [13:05:23<57:03:17, 5.73s/it]
18%|█▊ | 8112/43939 [13:05:29<56:58:29, 5.72s/it]
18%|█▊ | 8113/43939 [13:05:34<56:51:38, 5.71s/it]
18%|█▊ | 8114/43939 [13:05:40<56:56:28, 5.72s/it]
18%|█▊ | 8115/43939 [13:05:46<57:32:12, 5.78s/it]
18%|█▊ | 8116/43939 [13:05:52<57:16:25, 5.76s/it]
18%|█▊ | 8117/43939 [13:05:57<57:05:19, 5.74s/it]
18%|█▊ | 8118/43939 [13:06:03<56:52:11, 5.72s/it]
18%|█▊ | 8119/43939 [13:06:09<56:54:24, 5.72s/it]
18%|█▊ | 8120/43939 [13:06:14<56:50:50, 5.71s/it]
18%|█▊ | 8121/43939 [13:06:20<56:45:30, 5.70s/it]
18%|█▊ | 8122/43939 [13:06:26<56:43:55, 5.70s/it]
18%|█▊ | 8123/43939 [13:06:32<56:42:37, 5.70s/it]
18%|█▊ | 8124/43939 [13:06:37<56:38:43, 5.69s/it]
18%|█▊ | 8125/43939 [13:06:43<56:37:16, 5.69s/it]
18%|█▊ | 8126/43939 [13:06:49<56:38:52, 5.69s/it]
18%|█▊ | 8127/43939 [13:06:54<56:34:12, 5.69s/it]
18%|█▊ | 8128/43939 [13:07:00<56:36:04, 5.69s/it]
19%|█▊ | 8129/43939 [13:07:06<56:35:38, 5.69s/it]
19%|█▊ | 8130/43939 [13:07:11<56:42:27, 5.70s/it]
19%|█▊ | 8131/43939 [13:07:17<56:41:40, 5.70s/it]
19%|█▊ | 8132/43939 [13:07:23<56:39:05, 5.70s/it]
19%|█▊ | 8133/43939 [13:07:29<56:43:38, 5.70s/it]
19%|█▊ | 8134/43939 [13:07:34<56:43:27, 5.70s/it]
19%|█▊ | 8135/43939 [13:07:40<57:15:21, 5.76s/it]
19%|█▊ | 8136/43939 [13:07:46<57:00:38, 5.73s/it]
19%|█▊ | 8137/43939 [13:07:51<56:52:34, 5.72s/it]
19%|█▊ | 8138/43939 [13:07:57<56:48:19, 5.71s/it]
19%|█▊ | 8139/43939 [13:08:03<56:40:22, 5.70s/it]
19%|█▊ | 8140/43939 [13:08:09<56:39:15, 5.70s/it]
19%|█▊ | 8141/43939 [13:08:14<56:38:29, 5.70s/it]
19%|█▊ | 8142/43939 [13:08:20<56:42:47, 5.70s/it]
19%|█▊ | 8143/43939 [13:08:26<56:35:19, 5.69s/it]
19%|█▊ | 8144/43939 [13:08:31<56:36:21, 5.69s/it]
19%|█▊ | 8145/43939 [13:08:37<56:36:39, 5.69s/it]
19%|█▊ | 8146/43939 [13:08:43<56:42:01, 5.70s/it]
19%|█▊ | 8147/43939 [13:08:48<56:46:42, 5.71s/it]
19%|█▊ | 8148/43939 [13:08:54<56:47:03, 5.71s/it]
19%|█▊ | 8149/43939 [13:09:00<56:54:00, 5.72s/it]
19%|█▊ | 8150/43939 [13:09:06<56:52:04, 5.72s/it]
{'loss': 1.039, 'grad_norm': 1.291815161705017, 'learning_rate': 4.889533337108935e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.98, 'epoch': 0.74}
19%|█▊ | 8150/43939 [13:09:06<56:52:04, 5.72s/it]
19%|█▊ | 8151/43939 [13:09:11<56:57:56, 5.73s/it]
19%|█▊ | 8152/43939 [13:09:17<56:55:05, 5.73s/it]
19%|█▊ | 8153/43939 [13:09:23<56:53:29, 5.72s/it]
19%|█▊ | 8154/43939 [13:09:29<56:53:44, 5.72s/it]
19%|█▊ | 8155/43939 [13:09:34<56:48:44, 5.72s/it]
19%|█▊ | 8156/43939 [13:09:40<57:28:51, 5.78s/it]
19%|█▊ | 8157/43939 [13:09:46<57:16:23, 5.76s/it]
19%|█▊ | 8158/43939 [13:09:52<57:14:07, 5.76s/it]
19%|█▊ | 8159/43939 [13:09:57<56:58:08, 5.73s/it]
19%|█▊ | 8160/43939 [13:10:03<56:57:24, 5.73s/it]
19%|█▊ | 8161/43939 [13:10:09<56:50:10, 5.72s/it]
19%|█▊ | 8162/43939 [13:10:14<56:56:59, 5.73s/it]
19%|█▊ | 8163/43939 [13:10:20<56:45:36, 5.71s/it]
19%|█▊ | 8164/43939 [13:10:26<56:48:49, 5.72s/it]
19%|█▊ | 8165/43939 [13:10:32<56:52:18, 5.72s/it]
19%|█▊ | 8166/43939 [13:10:37<56:50:54, 5.72s/it]
19%|█▊ | 8167/43939 [13:10:43<56:50:33, 5.72s/it]
19%|█▊ | 8168/43939 [13:10:49<56:51:50, 5.72s/it]
19%|█▊ | 8169/43939 [13:10:54<56:41:47, 5.71s/it]
19%|█▊ | 8170/43939 [13:11:00<56:42:42, 5.71s/it]
19%|█▊ | 8171/43939 [13:11:06<56:32:48, 5.69s/it]
19%|█▊ | 8172/43939 [13:11:11<56:26:14, 5.68s/it]
19%|█▊ | 8173/43939 [13:11:17<56:31:37, 5.69s/it]
19%|█▊ | 8174/43939 [13:11:23<56:27:10, 5.68s/it]
19%|█▊ | 8175/43939 [13:11:29<56:27:54, 5.68s/it]
19%|█▊ | 8176/43939 [13:11:34<56:30:18, 5.69s/it]
19%|█▊ | 8177/43939 [13:11:40<56:23:53, 5.68s/it]
19%|█▊ | 8178/43939 [13:11:46<57:17:02, 5.77s/it]
19%|█▊ | 8179/43939 [13:11:52<56:59:45, 5.74s/it]
19%|█▊ | 8180/43939 [13:11:57<56:54:30, 5.73s/it]
19%|█▊ | 8181/43939 [13:12:03<56:43:13, 5.71s/it]
19%|█▊ | 8182/43939 [13:12:09<56:40:49, 5.71s/it]
19%|█▊ | 8183/43939 [13:12:14<56:39:38, 5.70s/it]
19%|█▊ | 8184/43939 [13:12:20<56:35:47, 5.70s/it]
19%|█▊ | 8185/43939 [13:12:26<56:32:32, 5.69s/it]
19%|█▊ | 8186/43939 [13:12:31<56:38:50, 5.70s/it]
19%|█▊ | 8187/43939 [13:12:37<56:36:02, 5.70s/it]
19%|█▊ | 8188/43939 [13:12:43<56:36:25, 5.70s/it]
19%|█▊ | 8189/43939 [13:12:49<56:38:47, 5.70s/it]
19%|█▊ | 8190/43939 [13:12:54<56:39:17, 5.71s/it]
19%|█▊ | 8191/43939 [13:13:00<56:39:39, 5.71s/it]
19%|█▊ | 8192/43939 [13:13:06<56:34:34, 5.70s/it]
19%|█▊ | 8193/43939 [13:13:11<56:30:15, 5.69s/it]
19%|█▊ | 8194/43939 [13:13:17<56:32:55, 5.70s/it]
19%|█▊ | 8195/43939 [13:13:23<56:35:04, 5.70s/it]
19%|█▊ | 8196/43939 [13:13:28<56:29:27, 5.69s/it]
19%|█▊ | 8197/43939 [13:13:34<56:30:46, 5.69s/it]
19%|█▊ | 8198/43939 [13:13:40<57:22:23, 5.78s/it]
19%|█▊ | 8199/43939 [13:13:46<57:03:06, 5.75s/it]
19%|█▊ | 8200/43939 [13:13:51<57:07:06, 5.75s/it]
{'loss': 1.0431, 'grad_norm': 0.8809608817100525, 'learning_rate': 4.886595275795383e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.41, 'epoch': 0.75}
19%|█▊ | 8200/43939 [13:13:51<57:07:06, 5.75s/it]
19%|█▊ | 8201/43939 [13:13:57<56:53:51, 5.73s/it]
19%|█▊ | 8202/43939 [13:14:03<56:45:00, 5.72s/it]
19%|█▊ | 8203/43939 [13:14:09<56:41:00, 5.71s/it]
19%|█▊ | 8204/43939 [13:14:14<56:40:42, 5.71s/it]
19%|█▊ | 8205/43939 [13:14:20<56:42:30, 5.71s/it]
19%|█▊ | 8206/43939 [13:14:26<56:41:07, 5.71s/it]
19%|█▊ | 8207/43939 [13:14:31<56:33:58, 5.70s/it]
19%|█▊ | 8208/43939 [13:14:37<56:41:55, 5.71s/it]
19%|█▊ | 8209/43939 [13:14:43<56:36:07, 5.70s/it]
19%|█▊ | 8210/43939 [13:14:48<56:40:37, 5.71s/it]
19%|█▊ | 8211/43939 [13:14:54<56:37:17, 5.71s/it]
19%|█▊ | 8212/43939 [13:15:00<56:43:39, 5.72s/it]
19%|█▊ | 8213/43939 [13:15:06<56:41:06, 5.71s/it]
19%|█▊ | 8214/43939 [13:15:11<56:37:01, 5.71s/it]
19%|█▊ | 8215/43939 [13:15:17<56:52:36, 5.73s/it]
19%|█▊ | 8216/43939 [13:15:23<56:41:51, 5.71s/it]
19%|█▊ | 8217/43939 [13:15:28<56:32:19, 5.70s/it]
19%|█▊ | 8218/43939 [13:15:34<56:37:47, 5.71s/it]
19%|█▊ | 8219/43939 [13:15:40<57:13:44, 5.77s/it]
19%|█▊ | 8220/43939 [13:15:46<56:59:03, 5.74s/it]
19%|█▊ | 8221/43939 [13:15:51<56:52:13, 5.73s/it]
19%|█▊ | 8222/43939 [13:15:57<56:41:02, 5.71s/it]
19%|█▊ | 8223/43939 [13:16:03<56:46:59, 5.72s/it]
19%|█▊ | 8224/43939 [13:16:09<56:42:10, 5.72s/it]
19%|█▊ | 8225/43939 [13:16:14<56:44:09, 5.72s/it]
19%|█▊ | 8226/43939 [13:16:20<56:30:07, 5.70s/it]
19%|█▊ | 8227/43939 [13:16:26<56:31:40, 5.70s/it]
19%|█▊ | 8228/43939 [13:16:31<56:32:01, 5.70s/it]
19%|█▊ | 8229/43939 [13:16:37<56:36:07, 5.71s/it]
19%|█▊ | 8230/43939 [13:16:43<56:33:44, 5.70s/it]
19%|█▊ | 8231/43939 [13:16:48<56:32:24, 5.70s/it]
19%|█▊ | 8232/43939 [13:16:54<56:48:45, 5.73s/it]
19%|█▊ | 8233/43939 [13:17:00<56:39:55, 5.71s/it]
19%|█▊ | 8234/43939 [13:17:06<56:47:59, 5.73s/it]
19%|█▊ | 8235/43939 [13:17:11<56:48:53, 5.73s/it]
19%|█▊ | 8236/43939 [13:17:17<56:43:24, 5.72s/it]
19%|█▊ | 8237/43939 [13:17:23<56:40:43, 5.72s/it]
19%|█▊ | 8238/43939 [13:17:29<56:54:00, 5.74s/it]
19%|█▉ | 8239/43939 [13:17:34<56:46:53, 5.73s/it]
19%|█▉ | 8240/43939 [13:17:40<56:36:27, 5.71s/it]
19%|█▉ | 8241/43939 [13:17:46<57:15:31, 5.77s/it]
19%|█▉ | 8242/43939 [13:17:52<57:00:20, 5.75s/it]
19%|█▉ | 8243/43939 [13:17:57<56:52:05, 5.74s/it]
19%|█▉ | 8244/43939 [13:18:03<56:50:11, 5.73s/it]
19%|█▉ | 8245/43939 [13:18:09<56:39:10, 5.71s/it]
19%|█▉ | 8246/43939 [13:18:14<56:35:39, 5.71s/it]
19%|█▉ | 8247/43939 [13:18:20<56:32:22, 5.70s/it]
19%|█▉ | 8248/43939 [13:18:26<56:32:04, 5.70s/it]
19%|█▉ | 8249/43939 [13:18:31<56:30:20, 5.70s/it]
19%|█▉ | 8250/43939 [13:18:37<56:34:19, 5.71s/it]
{'loss': 1.047, 'grad_norm': 1.0669974088668823, 'learning_rate': 4.8836195603329335e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.85, 'epoch': 0.75}
19%|█▉ | 8250/43939 [13:18:37<56:34:19, 5.71s/it]
19%|█▉ | 8251/43939 [13:18:43<56:38:00, 5.71s/it]
19%|█▉ | 8252/43939 [13:18:49<56:31:27, 5.70s/it]
19%|█▉ | 8253/43939 [13:18:54<56:30:03, 5.70s/it]
19%|█▉ | 8254/43939 [13:19:00<56:29:48, 5.70s/it]
19%|█▉ | 8255/43939 [13:19:06<56:25:04, 5.69s/it]
19%|█▉ | 8256/43939 [13:19:11<56:30:40, 5.70s/it]
19%|█▉ | 8257/43939 [13:19:17<56:28:48, 5.70s/it]
19%|█▉ | 8258/43939 [13:19:23<56:29:50, 5.70s/it]
19%|█▉ | 8259/43939 [13:19:29<56:31:30, 5.70s/it]
19%|█▉ | 8260/43939 [13:19:34<56:28:46, 5.70s/it]
19%|█▉ | 8261/43939 [13:19:40<57:07:03, 5.76s/it]
19%|█▉ | 8262/43939 [13:19:46<56:52:50, 5.74s/it]
19%|█▉ | 8263/43939 [13:19:52<56:51:11, 5.74s/it]
19%|█▉ | 8264/43939 [13:19:57<56:40:56, 5.72s/it]
19%|█▉ | 8265/43939 [13:20:03<56:40:22, 5.72s/it]
19%|█▉ | 8266/43939 [13:20:09<56:33:42, 5.71s/it]
19%|█▉ | 8267/43939 [13:20:14<56:44:50, 5.73s/it]
19%|█▉ | 8268/43939 [13:20:20<56:38:29, 5.72s/it]
19%|█▉ | 8269/43939 [13:20:26<56:29:54, 5.70s/it]
19%|█▉ | 8270/43939 [13:20:31<56:24:22, 5.69s/it]
19%|█▉ | 8271/43939 [13:20:37<56:24:57, 5.69s/it]
19%|█▉ | 8272/43939 [13:20:43<56:23:44, 5.69s/it]
19%|█▉ | 8273/43939 [13:20:49<56:27:33, 5.70s/it]
19%|█▉ | 8274/43939 [13:20:54<56:27:25, 5.70s/it]
19%|█▉ | 8275/43939 [13:21:00<56:25:27, 5.70s/it]
19%|█▉ | 8276/43939 [13:21:06<56:24:10, 5.69s/it]
19%|█▉ | 8277/43939 [13:21:11<56:26:50, 5.70s/it]
19%|█▉ | 8278/43939 [13:21:17<56:24:20, 5.69s/it]
19%|█▉ | 8279/43939 [13:21:23<56:23:05, 5.69s/it]
19%|█▉ | 8280/43939 [13:21:28<56:23:00, 5.69s/it]
19%|█▉ | 8281/43939 [13:21:34<56:32:14, 5.71s/it]
19%|█▉ | 8282/43939 [13:21:40<57:11:38, 5.77s/it]
19%|█▉ | 8283/43939 [13:21:46<56:59:23, 5.75s/it]
19%|█▉ | 8284/43939 [13:21:51<56:45:37, 5.73s/it]
19%|█▉ | 8285/43939 [13:21:57<56:47:51, 5.73s/it]
19%|█▉ | 8286/43939 [13:22:03<56:33:39, 5.71s/it]
19%|█▉ | 8287/43939 [13:22:08<56:26:38, 5.70s/it]
19%|█▉ | 8288/43939 [13:22:14<56:25:28, 5.70s/it]
19%|█▉ | 8289/43939 [13:22:20<56:29:02, 5.70s/it]
19%|█▉ | 8290/43939 [13:22:26<56:28:13, 5.70s/it]
19%|█▉ | 8291/43939 [13:22:31<56:25:47, 5.70s/it]
19%|█▉ | 8292/43939 [13:22:37<56:23:28, 5.69s/it]
19%|█▉ | 8293/43939 [13:22:43<56:34:29, 5.71s/it]
19%|█▉ | 8294/43939 [13:22:48<56:27:19, 5.70s/it]
19%|█▉ | 8295/43939 [13:22:54<56:24:33, 5.70s/it]
19%|█▉ | 8296/43939 [13:23:00<56:19:02, 5.69s/it]
19%|█▉ | 8297/43939 [13:23:05<56:15:21, 5.68s/it]
19%|█▉ | 8298/43939 [13:23:11<56:19:51, 5.69s/it]
19%|█▉ | 8299/43939 [13:23:17<56:20:16, 5.69s/it]
19%|█▉ | 8300/43939 [13:23:23<56:19:34, 5.69s/it]
{'loss': 1.0282, 'grad_norm': 1.1232683658599854, 'learning_rate': 4.880606237670491e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 456.41, 'epoch': 0.76}
19%|█▉ | 8300/43939 [13:23:23<56:19:34, 5.69s/it]
19%|█▉ | 8301/43939 [13:23:28<56:19:38, 5.69s/it]
19%|█▉ | 8302/43939 [13:23:34<56:21:39, 5.69s/it]
19%|█▉ | 8303/43939 [13:23:40<57:03:06, 5.76s/it]
19%|█▉ | 8304/43939 [13:23:46<56:45:14, 5.73s/it]
19%|█▉ | 8305/43939 [13:23:51<56:41:06, 5.73s/it]
19%|█▉ | 8306/43939 [13:23:57<56:35:18, 5.72s/it]
19%|█▉ | 8307/43939 [13:24:03<56:28:24, 5.71s/it]
19%|█▉ | 8308/43939 [13:24:08<56:26:09, 5.70s/it]
19%|█▉ | 8309/43939 [13:24:14<56:26:48, 5.70s/it]
19%|█▉ | 8310/43939 [13:24:20<56:34:00, 5.72s/it]
19%|█▉ | 8311/43939 [13:24:25<56:35:24, 5.72s/it]
19%|█▉ | 8312/43939 [13:24:31<56:26:18, 5.70s/it]
19%|█▉ | 8313/43939 [13:24:37<56:23:49, 5.70s/it]
19%|█▉ | 8314/43939 [13:24:42<56:17:46, 5.69s/it]
19%|█▉ | 8315/43939 [13:24:48<56:38:50, 5.72s/it]
19%|█▉ | 8316/43939 [13:24:54<56:32:21, 5.71s/it]
19%|█▉ | 8317/43939 [13:25:00<56:33:40, 5.72s/it]
19%|█▉ | 8318/43939 [13:25:05<56:23:55, 5.70s/it]
19%|█▉ | 8319/43939 [13:25:11<56:19:19, 5.69s/it]
19%|█▉ | 8320/43939 [13:25:17<56:18:07, 5.69s/it]
19%|█▉ | 8321/43939 [13:25:22<56:26:01, 5.70s/it]
19%|█▉ | 8322/43939 [13:25:28<56:23:16, 5.70s/it]
19%|█▉ | 8323/43939 [13:25:34<56:26:51, 5.71s/it]
19%|█▉ | 8324/43939 [13:25:40<56:24:53, 5.70s/it]
19%|█▉ | 8325/43939 [13:25:45<57:02:50, 5.77s/it]
19%|█▉ | 8326/43939 [13:25:51<56:53:12, 5.75s/it]
19%|█▉ | 8327/43939 [13:25:57<56:39:10, 5.73s/it]
19%|█▉ | 8328/43939 [13:26:03<56:31:33, 5.71s/it]
19%|█▉ | 8329/43939 [13:26:08<56:23:28, 5.70s/it]
19%|█▉ | 8330/43939 [13:26:14<56:24:48, 5.70s/it]
19%|█▉ | 8331/43939 [13:26:20<56:27:51, 5.71s/it]
19%|█▉ | 8332/43939 [13:26:25<56:25:42, 5.71s/it]
19%|█▉ | 8333/43939 [13:26:31<56:20:04, 5.70s/it]
19%|█▉ | 8334/43939 [13:26:37<56:21:09, 5.70s/it]
19%|█▉ | 8335/43939 [13:26:42<56:22:12, 5.70s/it]
19%|█▉ | 8336/43939 [13:26:48<56:32:42, 5.72s/it]
19%|█▉ | 8337/43939 [13:26:54<56:38:50, 5.73s/it]
19%|█▉ | 8338/43939 [13:27:00<56:34:16, 5.72s/it]
19%|█▉ | 8339/43939 [13:27:05<56:27:38, 5.71s/it]
19%|█▉ | 8340/43939 [13:27:11<56:27:05, 5.71s/it]
19%|█▉ | 8341/43939 [13:27:17<56:39:04, 5.73s/it]
19%|█▉ | 8342/43939 [13:27:23<56:38:51, 5.73s/it]
19%|█▉ | 8343/43939 [13:27:28<56:31:05, 5.72s/it]
19%|█▉ | 8344/43939 [13:27:34<56:29:15, 5.71s/it]
19%|█▉ | 8345/43939 [13:27:40<56:18:24, 5.69s/it]
19%|█▉ | 8346/43939 [13:27:46<57:00:33, 5.77s/it]
19%|█▉ | 8347/43939 [13:27:51<56:49:05, 5.75s/it]
19%|█▉ | 8348/43939 [13:27:57<56:50:47, 5.75s/it]
19%|█▉ | 8349/43939 [13:28:03<56:49:38, 5.75s/it]
19%|█▉ | 8350/43939 [13:28:08<56:41:38, 5.73s/it]
{'loss': 1.0415, 'grad_norm': 1.0202189683914185, 'learning_rate': 4.877555355350301e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.26, 'epoch': 0.76}
19%|█▉ | 8350/43939 [13:28:08<56:41:38, 5.73s/it]
19%|█▉ | 8351/43939 [13:28:14<56:33:57, 5.72s/it]
19%|█▉ | 8352/43939 [13:28:20<56:23:49, 5.71s/it]
19%|█▉ | 8353/43939 [13:28:25<56:24:10, 5.71s/it]
19%|█▉ | 8354/43939 [13:28:31<56:21:46, 5.70s/it]
19%|█▉ | 8355/43939 [13:28:37<56:26:41, 5.71s/it]
19%|█▉ | 8356/43939 [13:28:43<56:19:44, 5.70s/it]
19%|█▉ | 8357/43939 [13:28:48<56:24:13, 5.71s/it]
19%|█▉ | 8358/43939 [13:28:54<56:24:22, 5.71s/it]
19%|█▉ | 8359/43939 [13:29:00<56:27:07, 5.71s/it]
19%|█▉ | 8360/43939 [13:29:05<56:25:25, 5.71s/it]
19%|█▉ | 8361/43939 [13:29:11<56:23:03, 5.71s/it]
19%|█▉ | 8362/43939 [13:29:17<56:22:39, 5.70s/it]
19%|█▉ | 8363/43939 [13:29:23<56:19:01, 5.70s/it]
19%|█▉ | 8364/43939 [13:29:28<56:13:33, 5.69s/it]
19%|█▉ | 8365/43939 [13:29:34<56:20:35, 5.70s/it]
19%|█▉ | 8366/43939 [13:29:40<57:00:47, 5.77s/it]
19%|█▉ | 8367/43939 [13:29:46<56:47:56, 5.75s/it]
19%|█▉ | 8368/43939 [13:29:51<56:31:15, 5.72s/it]
19%|█▉ | 8369/43939 [13:29:57<56:26:21, 5.71s/it]
19%|█▉ | 8370/43939 [13:30:03<56:27:59, 5.72s/it]
19%|█▉ | 8371/43939 [13:30:08<56:33:47, 5.73s/it]
19%|█▉ | 8372/43939 [13:30:14<56:30:23, 5.72s/it]
19%|█▉ | 8373/43939 [13:30:20<56:22:17, 5.71s/it]
19%|█▉ | 8374/43939 [13:30:25<56:19:36, 5.70s/it]
19%|█▉ | 8375/43939 [13:30:31<56:13:10, 5.69s/it]
19%|█▉ | 8376/43939 [13:30:37<56:12:45, 5.69s/it]
19%|█▉ | 8377/43939 [13:30:42<56:10:02, 5.69s/it]
19%|█▉ | 8378/43939 [13:30:48<56:21:34, 5.71s/it]
19%|█▉ | 8379/43939 [13:30:54<56:28:39, 5.72s/it]
19%|█▉ | 8380/43939 [13:31:00<56:22:06, 5.71s/it]
19%|█▉ | 8381/43939 [13:31:05<56:12:15, 5.69s/it]
19%|█▉ | 8382/43939 [13:31:11<56:13:43, 5.69s/it]
19%|█▉ | 8383/43939 [13:31:17<56:14:18, 5.69s/it]
19%|█▉ | 8384/43939 [13:31:22<56:15:40, 5.70s/it]
19%|█▉ | 8385/43939 [13:31:28<56:17:59, 5.70s/it]
19%|█▉ | 8386/43939 [13:31:34<56:10:01, 5.69s/it]
19%|█▉ | 8387/43939 [13:31:39<56:14:54, 5.70s/it]
19%|█▉ | 8388/43939 [13:31:45<56:54:11, 5.76s/it]
19%|█▉ | 8389/43939 [13:31:51<56:43:41, 5.74s/it]
19%|█▉ | 8390/43939 [13:31:57<56:36:29, 5.73s/it]
19%|█▉ | 8391/43939 [13:32:03<56:37:36, 5.73s/it]
19%|█▉ | 8392/43939 [13:32:08<56:30:20, 5.72s/it]
19%|█▉ | 8393/43939 [13:32:14<56:21:16, 5.71s/it]
19%|█▉ | 8394/43939 [13:32:20<56:16:00, 5.70s/it]
19%|█▉ | 8395/43939 [13:32:25<56:17:49, 5.70s/it]
19%|█▉ | 8396/43939 [13:32:31<56:16:27, 5.70s/it]
19%|█▉ | 8397/43939 [13:32:37<56:21:10, 5.71s/it]
19%|█▉ | 8398/43939 [13:32:42<56:20:48, 5.71s/it]
19%|█▉ | 8399/43939 [13:32:48<56:18:05, 5.70s/it]
19%|█▉ | 8400/43939 [13:32:54<56:17:25, 5.70s/it]
{'loss': 1.0322, 'grad_norm': 1.2844234704971313, 'learning_rate': 4.8744669615072015e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.13, 'epoch': 0.76}
19%|█▉ | 8400/43939 [13:32:54<56:17:25, 5.70s/it]
19%|█▉ | 8401/43939 [13:33:00<56:16:29, 5.70s/it]
19%|█▉ | 8402/43939 [13:33:05<56:13:17, 5.70s/it]
19%|█▉ | 8403/43939 [13:33:11<56:12:53, 5.69s/it]
19%|█▉ | 8404/43939 [13:33:17<56:12:21, 5.69s/it]
19%|█▉ | 8405/43939 [13:33:22<56:14:07, 5.70s/it]
19%|█▉ | 8406/43939 [13:33:28<56:18:42, 5.71s/it]
19%|█▉ | 8407/43939 [13:33:34<56:15:19, 5.70s/it]
19%|█▉ | 8408/43939 [13:33:39<56:17:37, 5.70s/it]
19%|█▉ | 8409/43939 [13:33:45<57:00:02, 5.78s/it]
19%|█▉ | 8410/43939 [13:33:51<56:46:02, 5.75s/it]
19%|█▉ | 8411/43939 [13:33:57<56:46:22, 5.75s/it]
19%|█▉ | 8412/43939 [13:34:03<56:38:55, 5.74s/it]
19%|█▉ | 8413/43939 [13:34:08<56:29:06, 5.72s/it]
19%|█▉ | 8414/43939 [13:34:14<56:21:03, 5.71s/it]
19%|█▉ | 8415/43939 [13:34:20<56:16:58, 5.70s/it]
19%|█▉ | 8416/43939 [13:34:25<56:14:39, 5.70s/it]
19%|█▉ | 8417/43939 [13:34:31<56:15:02, 5.70s/it]
19%|█▉ | 8418/43939 [13:34:37<56:18:35, 5.71s/it]
19%|█▉ | 8419/43939 [13:34:42<56:27:23, 5.72s/it]
19%|█▉ | 8420/43939 [13:34:48<56:15:08, 5.70s/it]
19%|█▉ | 8421/43939 [13:34:54<56:09:08, 5.69s/it]
19%|█▉ | 8422/43939 [13:34:59<56:09:18, 5.69s/it]
19%|█▉ | 8423/43939 [13:35:05<56:06:57, 5.69s/it]
19%|█▉ | 8424/43939 [13:35:11<56:05:07, 5.69s/it]
19%|█▉ | 8425/43939 [13:35:17<56:09:31, 5.69s/it]
19%|█▉ | 8426/43939 [13:35:22<56:15:37, 5.70s/it]
19%|█▉ | 8427/43939 [13:35:28<56:09:01, 5.69s/it]
19%|█▉ | 8428/43939 [13:35:34<56:07:19, 5.69s/it]
19%|█▉ | 8429/43939 [13:35:39<56:06:51, 5.69s/it]
19%|█▉ | 8430/43939 [13:35:45<56:48:19, 5.76s/it]
19%|█▉ | 8431/43939 [13:35:51<56:39:29, 5.74s/it]
19%|█▉ | 8432/43939 [13:35:57<56:25:24, 5.72s/it]
19%|█▉ | 8433/43939 [13:36:02<56:27:45, 5.72s/it]
19%|█▉ | 8434/43939 [13:36:08<56:26:06, 5.72s/it]
19%|█▉ | 8435/43939 [13:36:14<56:15:20, 5.70s/it]
19%|█▉ | 8436/43939 [13:36:19<56:14:20, 5.70s/it]
19%|█▉ | 8437/43939 [13:36:25<56:24:34, 5.72s/it]
19%|█▉ | 8438/43939 [13:36:31<56:18:01, 5.71s/it]
19%|█▉ | 8439/43939 [13:36:37<56:10:26, 5.70s/it]
19%|█▉ | 8440/43939 [13:36:42<56:15:29, 5.71s/it]
19%|█▉ | 8441/43939 [13:36:48<56:11:05, 5.70s/it]
19%|█▉ | 8442/43939 [13:36:54<56:09:12, 5.69s/it]
19%|█▉ | 8443/43939 [13:36:59<56:16:37, 5.71s/it]
19%|█▉ | 8444/43939 [13:37:05<56:15:09, 5.71s/it]
19%|█▉ | 8445/43939 [13:37:11<56:14:54, 5.71s/it]
19%|█▉ | 8446/43939 [13:37:16<56:18:24, 5.71s/it]
19%|█▉ | 8447/43939 [13:37:22<56:22:47, 5.72s/it]
19%|█▉ | 8448/43939 [13:37:28<56:22:15, 5.72s/it]
19%|█▉ | 8449/43939 [13:37:34<56:15:06, 5.71s/it]
19%|█▉ | 8450/43939 [13:37:39<56:09:30, 5.70s/it]
{'loss': 1.0353, 'grad_norm': 1.028136134147644, 'learning_rate': 4.8713411048678635e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.77, 'epoch': 0.77}
19%|█▉ | 8450/43939 [13:37:39<56:09:30, 5.70s/it]
19%|█▉ | 8451/43939 [13:37:45<56:14:11, 5.70s/it]
19%|█▉ | 8452/43939 [13:37:51<56:53:40, 5.77s/it]
19%|█▉ | 8453/43939 [13:37:57<56:47:31, 5.76s/it]
19%|█▉ | 8454/43939 [13:38:02<56:41:11, 5.75s/it]
19%|█▉ | 8455/43939 [13:38:08<56:29:44, 5.73s/it]
19%|█▉ | 8456/43939 [13:38:14<56:31:48, 5.74s/it]
19%|█▉ | 8457/43939 [13:38:20<56:30:58, 5.73s/it]
19%|█▉ | 8458/43939 [13:38:25<56:19:54, 5.72s/it]
19%|█▉ | 8459/43939 [13:38:31<56:23:17, 5.72s/it]
19%|█▉ | 8460/43939 [13:38:37<56:17:01, 5.71s/it]
19%|█▉ | 8461/43939 [13:38:42<56:10:27, 5.70s/it]
19%|█▉ | 8462/43939 [13:38:48<56:06:38, 5.69s/it]
19%|█▉ | 8463/43939 [13:38:54<56:04:03, 5.69s/it]
19%|█▉ | 8464/43939 [13:38:59<56:08:48, 5.70s/it]
19%|█▉ | 8465/43939 [13:39:05<56:07:33, 5.70s/it]
19%|█▉ | 8466/43939 [13:39:11<56:07:47, 5.70s/it]
19%|█▉ | 8467/43939 [13:39:16<56:07:35, 5.70s/it]
19%|█▉ | 8468/43939 [13:39:22<56:04:48, 5.69s/it]
19%|█▉ | 8469/43939 [13:39:28<56:14:48, 5.71s/it]
19%|█▉ | 8470/43939 [13:39:34<56:12:10, 5.70s/it]
19%|█▉ | 8471/43939 [13:39:39<56:10:07, 5.70s/it]
19%|█▉ | 8472/43939 [13:39:45<56:13:03, 5.71s/it]
19%|█▉ | 8473/43939 [13:39:51<56:47:35, 5.76s/it]
19%|█▉ | 8474/43939 [13:39:57<56:35:47, 5.75s/it]
19%|█▉ | 8475/43939 [13:40:02<56:28:37, 5.73s/it]
19%|█▉ | 8476/43939 [13:40:08<56:20:31, 5.72s/it]
19%|█▉ | 8477/43939 [13:40:14<56:13:46, 5.71s/it]
19%|█▉ | 8478/43939 [13:40:19<56:12:30, 5.71s/it]
19%|█▉ | 8479/43939 [13:40:25<56:12:48, 5.71s/it]
19%|█▉ | 8480/43939 [13:40:31<56:11:58, 5.71s/it]
19%|█▉ | 8481/43939 [13:40:37<56:16:25, 5.71s/it]
19%|█▉ | 8482/43939 [13:40:42<56:06:35, 5.70s/it]
19%|█▉ | 8483/43939 [13:40:48<56:03:19, 5.69s/it]
19%|█▉ | 8484/43939 [13:40:54<56:01:42, 5.69s/it]
19%|█▉ | 8485/43939 [13:40:59<56:06:06, 5.70s/it]
19%|█▉ | 8486/43939 [13:41:05<56:15:39, 5.71s/it]
19%|█▉ | 8487/43939 [13:41:11<56:12:24, 5.71s/it]
19%|█▉ | 8488/43939 [13:41:16<56:15:49, 5.71s/it]
19%|█▉ | 8489/43939 [13:41:22<56:14:41, 5.71s/it]
19%|█▉ | 8490/43939 [13:41:28<56:19:36, 5.72s/it]
19%|█▉ | 8491/43939 [13:41:34<56:13:21, 5.71s/it]
19%|█▉ | 8492/43939 [13:41:39<56:12:16, 5.71s/it]
19%|█▉ | 8493/43939 [13:41:45<56:51:56, 5.78s/it]
19%|█▉ | 8494/43939 [13:41:51<56:40:30, 5.76s/it]
19%|█▉ | 8495/43939 [13:41:57<56:27:57, 5.74s/it]
19%|█▉ | 8496/43939 [13:42:02<56:25:03, 5.73s/it]
19%|█▉ | 8497/43939 [13:42:08<56:27:03, 5.73s/it]
19%|█▉ | 8498/43939 [13:42:14<56:18:10, 5.72s/it]
19%|█▉ | 8499/43939 [13:42:19<56:14:29, 5.71s/it]
19%|█▉ | 8500/43939 [13:42:25<56:04:41, 5.70s/it]
{'loss': 1.0397, 'grad_norm': 0.9930169582366943, 'learning_rate': 4.868177834750022e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.0, 'epoch': 0.77}
19%|█▉ | 8500/43939 [13:42:25<56:04:41, 5.70s/it]
19%|█▉ | 8501/43939 [13:42:31<56:05:22, 5.70s/it]
19%|█▉ | 8502/43939 [13:42:37<56:04:26, 5.70s/it]
19%|█▉ | 8503/43939 [13:42:42<56:10:07, 5.71s/it]
19%|█▉ | 8504/43939 [13:42:48<56:17:18, 5.72s/it]
19%|█▉ | 8505/43939 [13:42:54<56:11:00, 5.71s/it]
19%|█▉ | 8506/43939 [13:42:59<56:14:12, 5.71s/it]
19%|█▉ | 8507/43939 [13:43:05<56:19:10, 5.72s/it]
19%|█▉ | 8508/43939 [13:43:11<56:18:43, 5.72s/it]
19%|█▉ | 8509/43939 [13:43:17<56:25:08, 5.73s/it]
19%|█▉ | 8510/43939 [13:43:22<56:19:45, 5.72s/it]
19%|█▉ | 8511/43939 [13:43:28<56:22:03, 5.73s/it]
19%|█▉ | 8512/43939 [13:43:34<56:14:44, 5.72s/it]
19%|█▉ | 8513/43939 [13:43:39<56:06:14, 5.70s/it]
19%|█▉ | 8514/43939 [13:43:45<56:07:02, 5.70s/it]
19%|█▉ | 8515/43939 [13:43:51<56:54:35, 5.78s/it]
19%|█▉ | 8516/43939 [13:43:57<56:38:53, 5.76s/it]
19%|█▉ | 8517/43939 [13:44:02<56:27:18, 5.74s/it]
19%|█▉ | 8518/43939 [13:44:08<56:23:31, 5.73s/it]
19%|█▉ | 8519/43939 [13:44:14<56:32:26, 5.75s/it]
19%|█▉ | 8520/43939 [13:44:20<56:32:15, 5.75s/it]
19%|█▉ | 8521/43939 [13:44:25<56:25:10, 5.73s/it]
19%|█▉ | 8522/43939 [13:44:31<56:20:15, 5.73s/it]
19%|█▉ | 8523/43939 [13:44:37<56:10:55, 5.71s/it]
19%|█▉ | 8524/43939 [13:44:43<56:14:05, 5.72s/it]
19%|█▉ | 8525/43939 [13:44:48<56:13:58, 5.72s/it]
19%|█▉ | 8526/43939 [13:44:54<56:08:14, 5.71s/it]
19%|█▉ | 8527/43939 [13:45:00<56:06:26, 5.70s/it]
19%|█▉ | 8528/43939 [13:45:05<56:10:31, 5.71s/it]
19%|█▉ | 8529/43939 [13:45:11<56:10:13, 5.71s/it]
19%|█▉ | 8530/43939 [13:45:17<56:17:01, 5.72s/it]
19%|█▉ | 8531/43939 [13:45:23<56:15:47, 5.72s/it]
19%|█▉ | 8532/43939 [13:45:28<56:12:29, 5.71s/it]
19%|█▉ | 8533/43939 [13:45:34<56:20:33, 5.73s/it]
19%|█▉ | 8534/43939 [13:45:40<56:19:53, 5.73s/it]
19%|█▉ | 8535/43939 [13:45:45<56:12:36, 5.72s/it]
19%|█▉ | 8536/43939 [13:45:51<56:48:46, 5.78s/it]
19%|█▉ | 8537/43939 [13:45:57<56:39:39, 5.76s/it]
19%|█▉ | 8538/43939 [13:46:03<56:25:50, 5.74s/it]
19%|█▉ | 8539/43939 [13:46:08<56:24:15, 5.74s/it]
19%|█▉ | 8540/43939 [13:46:14<56:14:07, 5.72s/it]
19%|█▉ | 8541/43939 [13:46:20<56:16:34, 5.72s/it]
19%|█▉ | 8542/43939 [13:46:26<56:18:12, 5.73s/it]
19%|█▉ | 8543/43939 [13:46:31<56:13:18, 5.72s/it]
19%|█▉ | 8544/43939 [13:46:37<56:08:12, 5.71s/it]
19%|█▉ | 8545/43939 [13:46:43<56:15:20, 5.72s/it]
19%|█▉ | 8546/43939 [13:46:49<56:23:13, 5.74s/it]
19%|█▉ | 8547/43939 [13:46:54<56:15:40, 5.72s/it]
19%|█▉ | 8548/43939 [13:47:00<56:26:05, 5.74s/it]
19%|█▉ | 8549/43939 [13:47:06<56:20:54, 5.73s/it]
19%|█▉ | 8550/43939 [13:47:11<56:22:26, 5.73s/it]
{'loss': 1.043, 'grad_norm': 1.0289788246154785, 'learning_rate': 4.8649772010616955e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 526.03, 'epoch': 0.78}
19%|█▉ | 8550/43939 [13:47:11<56:22:26, 5.73s/it]
19%|█▉ | 8551/43939 [13:47:17<56:12:20, 5.72s/it]
19%|█▉ | 8552/43939 [13:47:23<56:20:20, 5.73s/it]
19%|█▉ | 8553/43939 [13:47:29<56:13:56, 5.72s/it]
19%|█▉ | 8554/43939 [13:47:34<56:03:03, 5.70s/it]
19%|█▉ | 8555/43939 [13:47:40<55:57:13, 5.69s/it]
19%|█▉ | 8556/43939 [13:47:46<55:48:45, 5.68s/it]
19%|█▉ | 8557/43939 [13:47:52<56:38:01, 5.76s/it]
19%|█▉ | 8558/43939 [13:47:57<56:25:35, 5.74s/it]
19%|█▉ | 8559/43939 [13:48:03<56:12:53, 5.72s/it]
19%|█▉ | 8560/43939 [13:48:09<56:07:52, 5.71s/it]
19%|█▉ | 8561/43939 [13:48:14<56:06:50, 5.71s/it]
19%|█▉ | 8562/43939 [13:48:20<55:58:42, 5.70s/it]
19%|█▉ | 8563/43939 [13:48:26<56:07:05, 5.71s/it]
19%|█▉ | 8564/43939 [13:48:31<56:07:47, 5.71s/it]
19%|█▉ | 8565/43939 [13:48:37<56:07:48, 5.71s/it]
19%|█▉ | 8566/43939 [13:48:43<56:05:17, 5.71s/it]
19%|█▉ | 8567/43939 [13:48:49<56:03:26, 5.71s/it]
19%|█▉ | 8568/43939 [13:48:54<55:58:58, 5.70s/it]
20%|█▉ | 8569/43939 [13:49:00<56:05:55, 5.71s/it]
20%|█▉ | 8570/43939 [13:49:06<56:07:47, 5.71s/it]
20%|█▉ | 8571/43939 [13:49:11<56:08:35, 5.71s/it]
20%|█▉ | 8572/43939 [13:49:17<56:05:01, 5.71s/it]
20%|█▉ | 8573/43939 [13:49:23<56:02:01, 5.70s/it]
20%|█▉ | 8574/43939 [13:49:29<56:11:23, 5.72s/it]
20%|█▉ | 8575/43939 [13:49:34<56:01:01, 5.70s/it]
20%|█▉ | 8576/43939 [13:49:40<55:57:34, 5.70s/it]
20%|█▉ | 8577/43939 [13:49:46<55:57:57, 5.70s/it]
20%|█▉ | 8578/43939 [13:49:52<56:43:41, 5.78s/it]
20%|█▉ | 8579/43939 [13:49:57<56:36:22, 5.76s/it]
20%|█▉ | 8580/43939 [13:50:03<56:20:30, 5.74s/it]
20%|█▉ | 8581/43939 [13:50:09<56:07:28, 5.71s/it]
20%|█▉ | 8582/43939 [13:50:14<56:00:50, 5.70s/it]
20%|█▉ | 8583/43939 [13:50:20<56:04:09, 5.71s/it]
20%|█▉ | 8584/43939 [13:50:26<56:14:22, 5.73s/it]
20%|█▉ | 8585/43939 [13:50:31<56:06:08, 5.71s/it]
20%|█▉ | 8586/43939 [13:50:37<56:00:48, 5.70s/it]
20%|█▉ | 8587/43939 [13:50:43<55:51:30, 5.69s/it]
20%|█▉ | 8588/43939 [13:50:49<55:55:15, 5.69s/it]
20%|█▉ | 8589/43939 [13:50:54<55:57:52, 5.70s/it]
20%|█▉ | 8590/43939 [13:51:00<55:56:23, 5.70s/it]
20%|█▉ | 8591/43939 [13:51:06<56:02:47, 5.71s/it]
20%|█▉ | 8592/43939 [13:51:11<55:59:00, 5.70s/it]
20%|█▉ | 8593/43939 [13:51:17<55:58:59, 5.70s/it]
20%|█▉ | 8594/43939 [13:51:23<55:58:33, 5.70s/it]
20%|█▉ | 8595/43939 [13:51:28<56:10:07, 5.72s/it]
20%|█▉ | 8596/43939 [13:51:34<56:02:36, 5.71s/it]
20%|█▉ | 8597/43939 [13:51:40<56:11:22, 5.72s/it]
20%|█▉ | 8598/43939 [13:51:46<55:59:50, 5.70s/it]
20%|█▉ | 8599/43939 [13:51:51<56:02:31, 5.71s/it]
20%|█▉ | 8600/43939 [13:51:57<56:44:34, 5.78s/it]
{'loss': 1.0374, 'grad_norm': 0.923629879951477, 'learning_rate': 4.8617392543004014e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.04, 'epoch': 0.78}
20%|█▉ | 8600/43939 [13:51:57<56:44:34, 5.78s/it]
20%|█▉ | 8601/43939 [13:52:03<56:27:49, 5.75s/it]
20%|█▉ | 8602/43939 [13:52:09<56:25:17, 5.75s/it]
20%|█▉ | 8603/43939 [13:52:14<56:23:51, 5.75s/it]
20%|█▉ | 8604/43939 [13:52:20<56:11:21, 5.72s/it]
20%|█▉ | 8605/43939 [13:52:26<56:13:00, 5.73s/it]
20%|█▉ | 8606/43939 [13:52:31<56:01:47, 5.71s/it]
20%|█▉ | 8607/43939 [13:52:37<56:00:23, 5.71s/it]
20%|█▉ | 8608/43939 [13:52:43<55:55:11, 5.70s/it]
20%|█▉ | 8609/43939 [13:52:49<55:55:20, 5.70s/it]
20%|█▉ | 8610/43939 [13:52:54<55:49:08, 5.69s/it]
20%|█▉ | 8611/43939 [13:53:00<55:45:33, 5.68s/it]
20%|█▉ | 8612/43939 [13:53:06<55:49:52, 5.69s/it]
20%|█▉ | 8613/43939 [13:53:11<55:52:42, 5.69s/it]
20%|█▉ | 8614/43939 [13:53:17<55:55:09, 5.70s/it]
20%|█▉ | 8615/43939 [13:53:23<55:49:01, 5.69s/it]
20%|█▉ | 8616/43939 [13:53:28<55:47:40, 5.69s/it]
20%|█▉ | 8617/43939 [13:53:34<55:47:44, 5.69s/it]
20%|█▉ | 8618/43939 [13:53:40<55:47:10, 5.69s/it]
20%|█▉ | 8619/43939 [13:53:45<55:52:51, 5.70s/it]
20%|█▉ | 8620/43939 [13:53:51<56:22:40, 5.75s/it]
20%|█▉ | 8621/43939 [13:53:57<56:12:50, 5.73s/it]
20%|█▉ | 8622/43939 [13:54:03<56:01:29, 5.71s/it]
20%|█▉ | 8623/43939 [13:54:08<55:52:58, 5.70s/it]
20%|█▉ | 8624/43939 [13:54:14<55:45:43, 5.68s/it]
20%|█▉ | 8625/43939 [13:54:20<55:47:30, 5.69s/it]
20%|█▉ | 8626/43939 [13:54:25<55:50:56, 5.69s/it]
20%|█▉ | 8627/43939 [13:54:31<55:53:08, 5.70s/it]
20%|█▉ | 8628/43939 [13:54:37<55:58:47, 5.71s/it]
20%|█▉ | 8629/43939 [13:54:43<55:54:33, 5.70s/it]
20%|█▉ | 8630/43939 [13:54:48<55:57:26, 5.71s/it]
20%|█▉ | 8631/43939 [13:54:54<55:55:52, 5.70s/it]
20%|█▉ | 8632/43939 [13:55:00<55:57:21, 5.71s/it]
20%|█▉ | 8633/43939 [13:55:05<55:56:22, 5.70s/it]
20%|█▉ | 8634/43939 [13:55:11<55:57:43, 5.71s/it]
20%|█▉ | 8635/43939 [13:55:17<55:57:35, 5.71s/it]
20%|█▉ | 8636/43939 [13:55:23<56:01:26, 5.71s/it]
20%|█▉ | 8637/43939 [13:55:28<55:58:19, 5.71s/it]
20%|█▉ | 8638/43939 [13:55:34<56:00:48, 5.71s/it]
20%|█▉ | 8639/43939 [13:55:40<55:58:29, 5.71s/it]
20%|█▉ | 8640/43939 [13:55:45<55:54:43, 5.70s/it]
20%|█▉ | 8641/43939 [13:55:51<56:32:53, 5.77s/it]
20%|█▉ | 8642/43939 [13:55:57<56:24:50, 5.75s/it]
20%|█▉ | 8643/43939 [13:56:03<56:04:25, 5.72s/it]
20%|█▉ | 8644/43939 [13:56:08<55:59:40, 5.71s/it]
20%|█▉ | 8645/43939 [13:56:14<56:02:30, 5.72s/it]
20%|█▉ | 8646/43939 [13:56:20<55:55:59, 5.71s/it]
20%|█▉ | 8647/43939 [13:56:25<56:04:52, 5.72s/it]
20%|█▉ | 8648/43939 [13:56:31<55:56:13, 5.71s/it]
20%|█▉ | 8649/43939 [13:56:37<55:59:40, 5.71s/it]
20%|█▉ | 8650/43939 [13:56:43<55:52:02, 5.70s/it]
{'loss': 1.0222, 'grad_norm': 1.0967282056808472, 'learning_rate': 4.8584640455523604e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.9, 'epoch': 0.79}
20%|█▉ | 8650/43939 [13:56:43<55:52:02, 5.70s/it]
20%|█▉ | 8651/43939 [13:56:48<55:56:08, 5.71s/it]
20%|█▉ | 8652/43939 [13:56:54<55:53:49, 5.70s/it]
20%|█▉ | 8653/43939 [13:57:00<55:46:56, 5.69s/it]
20%|█▉ | 8654/43939 [13:57:05<55:44:48, 5.69s/it]
20%|█▉ | 8655/43939 [13:57:11<55:45:48, 5.69s/it]
20%|█▉ | 8656/43939 [13:57:17<55:47:46, 5.69s/it]
20%|█▉ | 8657/43939 [13:57:22<55:50:09, 5.70s/it]
20%|█▉ | 8658/43939 [13:57:28<55:55:30, 5.71s/it]
20%|█▉ | 8659/43939 [13:57:34<55:54:57, 5.71s/it]
20%|█▉ | 8660/43939 [13:57:39<55:50:45, 5.70s/it]
20%|█▉ | 8661/43939 [13:57:45<55:50:46, 5.70s/it]
20%|█▉ | 8662/43939 [13:57:51<55:53:30, 5.70s/it]
20%|█▉ | 8663/43939 [13:57:57<56:36:09, 5.78s/it]
20%|█▉ | 8664/43939 [13:58:03<56:22:05, 5.75s/it]
20%|█▉ | 8665/43939 [13:58:08<56:05:48, 5.73s/it]
20%|█▉ | 8666/43939 [13:58:14<56:00:08, 5.72s/it]
20%|█▉ | 8667/43939 [13:58:20<55:57:32, 5.71s/it]
20%|█▉ | 8668/43939 [13:58:25<56:02:59, 5.72s/it]
20%|█▉ | 8669/43939 [13:58:31<56:02:31, 5.72s/it]
20%|█▉ | 8670/43939 [13:58:37<56:04:06, 5.72s/it]
20%|█▉ | 8671/43939 [13:58:42<55:54:44, 5.71s/it]
20%|█▉ | 8672/43939 [13:58:48<56:04:02, 5.72s/it]
20%|█▉ | 8673/43939 [13:58:54<55:52:48, 5.70s/it]
20%|█▉ | 8674/43939 [13:59:00<55:54:27, 5.71s/it]
20%|█▉ | 8675/43939 [13:59:05<55:52:40, 5.70s/it]
20%|█▉ | 8676/43939 [13:59:11<56:06:16, 5.73s/it]
20%|█▉ | 8677/43939 [13:59:17<56:00:59, 5.72s/it]
20%|█▉ | 8678/43939 [13:59:22<55:55:29, 5.71s/it]
20%|█▉ | 8679/43939 [13:59:28<55:49:15, 5.70s/it]
20%|█▉ | 8680/43939 [13:59:34<55:50:42, 5.70s/it]
20%|█▉ | 8681/43939 [13:59:40<55:56:26, 5.71s/it]
20%|█▉ | 8682/43939 [13:59:45<55:52:19, 5.70s/it]
20%|█▉ | 8683/43939 [13:59:51<55:49:32, 5.70s/it]
20%|█▉ | 8684/43939 [13:59:57<56:39:41, 5.79s/it]
20%|█▉ | 8685/43939 [14:00:03<56:23:23, 5.76s/it]
20%|█▉ | 8686/43939 [14:00:08<56:15:51, 5.75s/it]
20%|█▉ | 8687/43939 [14:00:14<56:00:30, 5.72s/it]
20%|█▉ | 8688/43939 [14:00:20<55:53:54, 5.71s/it]
20%|█▉ | 8689/43939 [14:00:25<55:47:29, 5.70s/it]
20%|█▉ | 8690/43939 [14:00:31<55:48:59, 5.70s/it]
20%|█▉ | 8691/43939 [14:00:37<55:51:46, 5.71s/it]
20%|█▉ | 8692/43939 [14:00:43<55:52:04, 5.71s/it]
20%|█▉ | 8693/43939 [14:00:48<55:48:22, 5.70s/it]
20%|█▉ | 8694/43939 [14:00:54<55:51:01, 5.70s/it]
20%|█▉ | 8695/43939 [14:01:00<55:50:19, 5.70s/it]
20%|█▉ | 8696/43939 [14:01:05<55:47:42, 5.70s/it]
20%|█▉ | 8697/43939 [14:01:11<55:43:10, 5.69s/it]
20%|█▉ | 8698/43939 [14:01:17<55:49:52, 5.70s/it]
20%|█▉ | 8699/43939 [14:01:22<55:55:08, 5.71s/it]
20%|█▉ | 8700/43939 [14:01:28<55:49:12, 5.70s/it]
{'loss': 1.0443, 'grad_norm': 1.5993574857711792, 'learning_rate': 4.855151626491686e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.03, 'epoch': 0.79}
20%|█▉ | 8700/43939 [14:01:28<55:49:12, 5.70s/it]
20%|█▉ | 8701/43939 [14:01:34<55:50:13, 5.70s/it]
20%|█▉ | 8702/43939 [14:01:40<55:49:21, 5.70s/it]
20%|█▉ | 8703/43939 [14:01:45<55:45:24, 5.70s/it]
20%|█▉ | 8704/43939 [14:01:51<56:22:21, 5.76s/it]
20%|█▉ | 8705/43939 [14:01:57<56:19:02, 5.75s/it]
20%|█▉ | 8706/43939 [14:02:03<56:08:43, 5.74s/it]
20%|█▉ | 8707/43939 [14:02:08<55:56:08, 5.72s/it]
20%|█▉ | 8708/43939 [14:02:14<55:52:20, 5.71s/it]
20%|█▉ | 8709/43939 [14:02:20<55:59:43, 5.72s/it]
20%|█▉ | 8710/43939 [14:02:25<55:57:34, 5.72s/it]
20%|█▉ | 8711/43939 [14:02:31<55:50:21, 5.71s/it]
20%|█▉ | 8712/43939 [14:02:37<55:44:04, 5.70s/it]
20%|█▉ | 8713/43939 [14:02:42<55:51:24, 5.71s/it]
20%|█▉ | 8714/43939 [14:02:48<55:46:45, 5.70s/it]
20%|█▉ | 8715/43939 [14:02:54<55:35:36, 5.68s/it]
20%|█▉ | 8716/43939 [14:02:59<55:31:29, 5.67s/it]
20%|█▉ | 8717/43939 [14:03:05<55:30:07, 5.67s/it]
20%|█▉ | 8718/43939 [14:03:11<55:34:20, 5.68s/it]
20%|█▉ | 8719/43939 [14:03:17<55:40:24, 5.69s/it]
20%|█▉ | 8720/43939 [14:03:22<55:40:53, 5.69s/it]
20%|█▉ | 8721/43939 [14:03:28<55:48:03, 5.70s/it]
20%|█▉ | 8722/43939 [14:03:34<55:39:28, 5.69s/it]
20%|█▉ | 8723/43939 [14:03:39<55:32:56, 5.68s/it]
20%|█▉ | 8724/43939 [14:03:45<55:38:19, 5.69s/it]
20%|█▉ | 8725/43939 [14:03:51<55:40:12, 5.69s/it]
20%|█▉ | 8726/43939 [14:03:57<56:24:25, 5.77s/it]
20%|█▉ | 8727/43939 [14:04:02<56:08:23, 5.74s/it]
20%|█▉ | 8728/43939 [14:04:08<55:58:36, 5.72s/it]
20%|█▉ | 8729/43939 [14:04:14<55:49:51, 5.71s/it]
20%|█▉ | 8730/43939 [14:04:19<55:46:44, 5.70s/it]
20%|█▉ | 8731/43939 [14:04:25<55:43:58, 5.70s/it]
20%|█▉ | 8732/43939 [14:04:31<55:48:39, 5.71s/it]
20%|█▉ | 8733/43939 [14:04:36<55:54:25, 5.72s/it]
20%|█▉ | 8734/43939 [14:04:42<55:46:14, 5.70s/it]
20%|█▉ | 8735/43939 [14:04:48<55:44:40, 5.70s/it]
20%|█▉ | 8736/43939 [14:04:54<55:41:12, 5.69s/it]
20%|█▉ | 8737/43939 [14:04:59<55:51:25, 5.71s/it]
20%|█▉ | 8738/43939 [14:05:05<56:02:42, 5.73s/it]
20%|█▉ | 8739/43939 [14:05:11<56:02:54, 5.73s/it]
20%|█▉ | 8740/43939 [14:05:17<56:03:26, 5.73s/it]
20%|█▉ | 8741/43939 [14:05:22<56:02:09, 5.73s/it]
20%|█▉ | 8742/43939 [14:05:28<55:55:29, 5.72s/it]
20%|█▉ | 8743/43939 [14:05:34<55:56:56, 5.72s/it]
20%|█▉ | 8744/43939 [14:05:39<55:49:40, 5.71s/it]
20%|█▉ | 8745/43939 [14:05:45<55:43:37, 5.70s/it]
20%|█▉ | 8746/43939 [14:05:51<55:39:48, 5.69s/it]
20%|█▉ | 8747/43939 [14:05:57<56:21:40, 5.77s/it]
20%|█▉ | 8748/43939 [14:06:02<56:17:03, 5.76s/it]
20%|█▉ | 8749/43939 [14:06:08<56:11:30, 5.75s/it]
20%|█▉ | 8750/43939 [14:06:14<55:58:12, 5.73s/it]
{'loss': 1.0213, 'grad_norm': 1.1585549116134644, 'learning_rate': 4.851802049379574e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.88, 'epoch': 0.8}
20%|█▉ | 8750/43939 [14:06:14<55:58:12, 5.73s/it]
20%|█▉ | 8751/43939 [14:06:19<55:53:41, 5.72s/it]
20%|█▉ | 8752/43939 [14:06:25<55:53:30, 5.72s/it]
20%|█▉ | 8753/43939 [14:06:31<55:51:06, 5.71s/it]
20%|█▉ | 8754/43939 [14:06:37<55:57:56, 5.73s/it]
20%|█▉ | 8755/43939 [14:06:42<55:51:06, 5.71s/it]
20%|█▉ | 8756/43939 [14:06:48<55:50:46, 5.71s/it]
20%|█▉ | 8757/43939 [14:06:54<55:51:59, 5.72s/it]
20%|█▉ | 8758/43939 [14:06:59<55:38:24, 5.69s/it]
20%|█▉ | 8759/43939 [14:07:05<55:31:45, 5.68s/it]
20%|█▉ | 8760/43939 [14:07:11<55:47:10, 5.71s/it]
20%|█▉ | 8761/43939 [14:07:17<55:46:41, 5.71s/it]
20%|█▉ | 8762/43939 [14:07:22<55:47:43, 5.71s/it]
20%|█▉ | 8763/43939 [14:07:28<55:48:47, 5.71s/it]
20%|█▉ | 8764/43939 [14:07:34<55:47:55, 5.71s/it]
20%|█▉ | 8765/43939 [14:07:39<55:43:11, 5.70s/it]
20%|█▉ | 8766/43939 [14:07:45<55:40:46, 5.70s/it]
20%|█▉ | 8767/43939 [14:07:51<56:24:19, 5.77s/it]
20%|█▉ | 8768/43939 [14:07:57<56:13:22, 5.75s/it]
20%|█▉ | 8769/43939 [14:08:02<56:04:04, 5.74s/it]
20%|█▉ | 8770/43939 [14:08:08<55:54:36, 5.72s/it]
20%|█▉ | 8771/43939 [14:08:14<55:56:51, 5.73s/it]
20%|█▉ | 8772/43939 [14:08:20<55:54:54, 5.72s/it]
20%|█▉ | 8773/43939 [14:08:25<55:49:27, 5.71s/it]
20%|█▉ | 8774/43939 [14:08:31<55:42:10, 5.70s/it]
20%|█▉ | 8775/43939 [14:08:37<55:41:23, 5.70s/it]
20%|█▉ | 8776/43939 [14:08:42<55:39:00, 5.70s/it]
20%|█▉ | 8777/43939 [14:08:48<55:39:12, 5.70s/it]
20%|█▉ | 8778/43939 [14:08:54<55:45:07, 5.71s/it]
20%|█▉ | 8779/43939 [14:08:59<55:40:44, 5.70s/it]
20%|█▉ | 8780/43939 [14:09:05<55:35:54, 5.69s/it]
20%|█▉ | 8781/43939 [14:09:11<55:42:43, 5.70s/it]
20%|█▉ | 8782/43939 [14:09:17<55:48:03, 5.71s/it]
20%|█▉ | 8783/43939 [14:09:22<55:42:01, 5.70s/it]
20%|█▉ | 8784/43939 [14:09:28<55:42:57, 5.71s/it]
20%|█▉ | 8785/43939 [14:09:34<55:40:58, 5.70s/it]
20%|█▉ | 8786/43939 [14:09:39<55:33:34, 5.69s/it]
20%|█▉ | 8787/43939 [14:09:45<55:33:01, 5.69s/it]
20%|██ | 8788/43939 [14:09:51<55:34:00, 5.69s/it][2025-11-29 08:14:04,537] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.38s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:48, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.69s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.72s/it][A
19%|█▉ | 14/72 [00:35<02:38, 2.73s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:30, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.74s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.74s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.74s/it][A
38%|███▊ | 27/72 [01:11<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:55, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.74s/it][A
49%|████▊ | 35/72 [01:32<01:41, 2.74s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:32, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:52, 2.74s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.74s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.74s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 1.0147367715835571, 'eval_runtime': 195.7565, 'eval_samples_per_second': 5.103, 'eval_steps_per_second': 0.368, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 0.8}
20%|██ | 8788/43939 [14:13:06<55:34:00, 5.69s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
20%|██ | 8789/43939 [14:13:13<630:05:16, 64.53s/it]
20%|██ | 8790/43939 [14:13:18<457:52:53, 46.90s/it]
20%|██ | 8791/43939 [14:13:24<337:04:17, 34.52s/it]
20%|██ | 8792/43939 [14:13:30<252:31:15, 25.86s/it]
20%|██ | 8793/43939 [14:13:35<193:25:46, 19.81s/it]
20%|██ | 8794/43939 [14:13:41<152:05:44, 15.58s/it]
20%|██ | 8795/43939 [14:13:47<123:05:06, 12.61s/it]
20%|██ | 8796/43939 [14:13:52<102:47:28, 10.53s/it]
20%|██ | 8797/43939 [14:13:58<88:34:28, 9.07s/it]
20%|██ | 8798/43939 [14:14:04<78:35:41, 8.05s/it]
20%|██ | 8799/43939 [14:14:09<71:35:48, 7.33s/it]
20%|██ | 8800/43939 [14:14:15<66:44:40, 6.84s/it]
{'loss': 1.0558, 'grad_norm': 1.144241213798523, 'learning_rate': 4.8484153670634735e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 109.97, 'epoch': 0.8}
20%|██ | 8800/43939 [14:14:15<66:44:40, 6.84s/it]
20%|██ | 8801/43939 [14:14:21<63:27:52, 6.50s/it]
20%|██ | 8802/43939 [14:14:26<61:01:09, 6.25s/it]
20%|██ | 8803/43939 [14:14:32<59:24:06, 6.09s/it]
20%|██ | 8804/43939 [14:14:38<58:18:45, 5.97s/it]
20%|██ | 8805/43939 [14:14:44<57:37:34, 5.90s/it]
20%|██ | 8806/43939 [14:14:49<56:56:38, 5.83s/it]
20%|██ | 8807/43939 [14:14:55<56:24:37, 5.78s/it]
20%|██ | 8808/43939 [14:15:01<56:09:08, 5.75s/it]
20%|██ | 8809/43939 [14:15:06<56:31:58, 5.79s/it]
20%|██ | 8810/43939 [14:15:12<56:15:44, 5.77s/it]
20%|██ | 8811/43939 [14:15:18<56:10:45, 5.76s/it]
20%|██ | 8812/43939 [14:15:24<55:57:57, 5.74s/it]
20%|██ | 8813/43939 [14:15:29<55:53:39, 5.73s/it]
20%|██ | 8814/43939 [14:15:35<55:48:45, 5.72s/it]
20%|██ | 8815/43939 [14:15:41<55:44:44, 5.71s/it]
20%|██ | 8816/43939 [14:15:46<55:49:24, 5.72s/it]
20%|██ | 8817/43939 [14:15:52<55:36:44, 5.70s/it]
20%|██ | 8818/43939 [14:15:58<55:29:35, 5.69s/it]
20%|██ | 8819/43939 [14:16:04<55:42:16, 5.71s/it]
20%|██ | 8820/43939 [14:16:09<55:38:59, 5.70s/it]
20%|██ | 8821/43939 [14:16:15<55:43:40, 5.71s/it]
20%|██ | 8822/43939 [14:16:21<55:44:06, 5.71s/it]
20%|██ | 8823/43939 [14:16:26<55:40:57, 5.71s/it]
20%|██ | 8824/43939 [14:16:32<55:47:47, 5.72s/it]
20%|██ | 8825/43939 [14:16:38<55:47:30, 5.72s/it]
20%|██ | 8826/43939 [14:16:44<55:45:01, 5.72s/it]
20%|██ | 8827/43939 [14:16:49<55:40:55, 5.71s/it]
20%|██ | 8828/43939 [14:16:55<55:44:02, 5.71s/it]
20%|██ | 8829/43939 [14:17:01<55:50:00, 5.72s/it]
20%|██ | 8830/43939 [14:17:07<56:24:32, 5.78s/it]
20%|██ | 8831/43939 [14:17:12<56:14:54, 5.77s/it]
20%|██ | 8832/43939 [14:17:18<56:02:01, 5.75s/it]
20%|██ | 8833/43939 [14:17:24<55:53:15, 5.73s/it]
20%|██ | 8834/43939 [14:17:29<55:46:14, 5.72s/it]
20%|██ | 8835/43939 [14:17:35<55:43:16, 5.71s/it]
20%|██ | 8836/43939 [14:17:41<55:42:51, 5.71s/it]
20%|██ | 8837/43939 [14:17:47<55:34:31, 5.70s/it]
20%|██ | 8838/43939 [14:17:52<55:35:45, 5.70s/it]
20%|██ | 8839/43939 [14:17:58<55:36:31, 5.70s/it]
20%|██ | 8840/43939 [14:18:04<55:27:31, 5.69s/it]
20%|██ | 8841/43939 [14:18:09<55:21:12, 5.68s/it]
20%|██ | 8842/43939 [14:18:15<55:31:20, 5.70s/it]
20%|██ | 8843/43939 [14:18:21<55:38:18, 5.71s/it]
20%|██ | 8844/43939 [14:18:26<55:42:26, 5.71s/it]
20%|██ | 8845/43939 [14:18:32<55:47:57, 5.72s/it]
20%|██ | 8846/43939 [14:18:38<55:39:29, 5.71s/it]
20%|██ | 8847/43939 [14:18:44<55:32:48, 5.70s/it]
20%|██ | 8848/43939 [14:18:49<55:33:51, 5.70s/it]
20%|██ | 8849/43939 [14:18:55<55:32:36, 5.70s/it]
20%|██ | 8850/43939 [14:19:01<55:34:56, 5.70s/it]
{'loss': 1.033, 'grad_norm': 0.8619866371154785, 'learning_rate': 4.844991632976257e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 518.75, 'epoch': 0.81}
20%|██ | 8850/43939 [14:19:01<55:34:56, 5.70s/it]
20%|██ | 8851/43939 [14:19:07<56:18:31, 5.78s/it]
20%|██ | 8852/43939 [14:19:12<55:57:34, 5.74s/it]
20%|██ | 8853/43939 [14:19:18<55:55:12, 5.74s/it]
20%|██ | 8854/43939 [14:19:24<55:48:44, 5.73s/it]
20%|██ | 8855/43939 [14:19:29<55:50:17, 5.73s/it]
20%|██ | 8856/43939 [14:19:35<55:50:02, 5.73s/it]
20%|██ | 8857/43939 [14:19:41<55:48:44, 5.73s/it]
20%|██ | 8858/43939 [14:19:47<55:41:09, 5.71s/it]
20%|██ | 8859/43939 [14:19:52<55:46:52, 5.72s/it]
20%|██ | 8860/43939 [14:19:58<55:44:52, 5.72s/it]
20%|██ | 8861/43939 [14:20:04<55:43:53, 5.72s/it]
20%|██ | 8862/43939 [14:20:09<55:39:40, 5.71s/it]
20%|██ | 8863/43939 [14:20:15<55:32:39, 5.70s/it]
20%|██ | 8864/43939 [14:20:21<55:35:05, 5.71s/it]
20%|██ | 8865/43939 [14:20:27<55:52:40, 5.74s/it]
20%|██ | 8866/43939 [14:20:32<55:44:46, 5.72s/it]
20%|██ | 8867/43939 [14:20:38<55:41:34, 5.72s/it]
20%|██ | 8868/43939 [14:20:44<55:44:25, 5.72s/it]
20%|██ | 8869/43939 [14:20:49<55:40:04, 5.71s/it]
20%|██ | 8870/43939 [14:20:55<55:43:24, 5.72s/it]
20%|██ | 8871/43939 [14:21:01<56:14:22, 5.77s/it]
20%|██ | 8872/43939 [14:21:07<55:55:46, 5.74s/it]
20%|██ | 8873/43939 [14:21:12<55:51:07, 5.73s/it]
20%|██ | 8874/43939 [14:21:18<55:45:38, 5.72s/it]
20%|██ | 8875/43939 [14:21:24<55:39:03, 5.71s/it]
20%|██ | 8876/43939 [14:21:30<55:36:37, 5.71s/it]
20%|██ | 8877/43939 [14:21:35<55:31:33, 5.70s/it]
20%|██ | 8878/43939 [14:21:41<55:29:55, 5.70s/it]
20%|██ | 8879/43939 [14:21:47<55:35:18, 5.71s/it]
20%|██ | 8880/43939 [14:21:52<55:24:47, 5.69s/it]
20%|██ | 8881/43939 [14:21:58<55:37:50, 5.71s/it]
20%|██ | 8882/43939 [14:22:04<55:34:22, 5.71s/it]
20%|██ | 8883/43939 [14:22:09<55:36:28, 5.71s/it]
20%|██ | 8884/43939 [14:22:15<55:35:03, 5.71s/it]
20%|██ | 8885/43939 [14:22:21<55:36:52, 5.71s/it]
20%|██ | 8886/43939 [14:22:27<55:33:15, 5.71s/it]
20%|██ | 8887/43939 [14:22:32<55:40:29, 5.72s/it]
20%|██ | 8888/43939 [14:22:38<55:36:44, 5.71s/it]
20%|██ | 8889/43939 [14:22:44<55:34:06, 5.71s/it]
20%|██ | 8890/43939 [14:22:49<55:30:55, 5.70s/it]
20%|██ | 8891/43939 [14:22:55<55:24:46, 5.69s/it]
20%|██ | 8892/43939 [14:23:01<56:07:30, 5.77s/it]
20%|██ | 8893/43939 [14:23:07<55:57:23, 5.75s/it]
20%|██ | 8894/43939 [14:23:12<55:47:54, 5.73s/it]
20%|██ | 8895/43939 [14:23:18<55:37:43, 5.71s/it]
20%|██ | 8896/43939 [14:23:24<55:38:44, 5.72s/it]
20%|██ | 8897/43939 [14:23:30<55:39:24, 5.72s/it]
20%|██ | 8898/43939 [14:23:35<55:33:30, 5.71s/it]
20%|██ | 8899/43939 [14:23:41<55:24:13, 5.69s/it]
20%|██ | 8900/43939 [14:23:47<55:26:48, 5.70s/it]
{'loss': 1.0518, 'grad_norm': 1.1444193124771118, 'learning_rate': 4.841530901135376e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 518.64, 'epoch': 0.81}
20%|██ | 8900/43939 [14:23:47<55:26:48, 5.70s/it]
20%|██ | 8901/43939 [14:23:52<55:27:57, 5.70s/it]
20%|██ | 8902/43939 [14:23:58<55:27:24, 5.70s/it]
20%|██ | 8903/43939 [14:24:04<55:33:04, 5.71s/it]
20%|██ | 8904/43939 [14:24:10<55:45:03, 5.73s/it]
20%|██ | 8905/43939 [14:24:15<55:41:37, 5.72s/it]
20%|██ | 8906/43939 [14:24:21<55:46:06, 5.73s/it]
20%|██ | 8907/43939 [14:24:27<55:38:57, 5.72s/it]
20%|██ | 8908/43939 [14:24:32<55:32:21, 5.71s/it]
20%|██ | 8909/43939 [14:24:38<55:29:23, 5.70s/it]
20%|██ | 8910/43939 [14:24:44<55:30:21, 5.70s/it]
20%|██ | 8911/43939 [14:24:49<55:29:10, 5.70s/it]
20%|██ | 8912/43939 [14:24:55<55:35:55, 5.71s/it]
20%|██ | 8913/43939 [14:25:01<56:12:03, 5.78s/it]
20%|██ | 8914/43939 [14:25:07<56:06:20, 5.77s/it]
20%|██ | 8915/43939 [14:25:13<55:53:07, 5.74s/it]
20%|██ | 8916/43939 [14:25:18<55:48:31, 5.74s/it]
20%|██ | 8917/43939 [14:25:24<55:41:13, 5.72s/it]
20%|██ | 8918/43939 [14:25:30<55:34:06, 5.71s/it]
20%|██ | 8919/43939 [14:25:35<55:36:00, 5.72s/it]
20%|██ | 8920/43939 [14:25:41<55:38:52, 5.72s/it]
20%|██ | 8921/43939 [14:25:47<55:41:55, 5.73s/it]
20%|██ | 8922/43939 [14:25:53<55:32:18, 5.71s/it]
20%|██ | 8923/43939 [14:25:58<55:32:17, 5.71s/it]
20%|██ | 8924/43939 [14:26:04<55:27:44, 5.70s/it]
20%|██ | 8925/43939 [14:26:10<55:30:50, 5.71s/it]
20%|██ | 8926/43939 [14:26:15<55:28:04, 5.70s/it]
20%|██ | 8927/43939 [14:26:21<55:31:54, 5.71s/it]
20%|██ | 8928/43939 [14:26:27<55:25:42, 5.70s/it]
20%|██ | 8929/43939 [14:26:32<55:34:10, 5.71s/it]
20%|██ | 8930/43939 [14:26:38<55:36:33, 5.72s/it]
20%|██ | 8931/43939 [14:26:44<55:35:19, 5.72s/it]
20%|██ | 8932/43939 [14:26:50<55:24:56, 5.70s/it]
20%|██ | 8933/43939 [14:26:55<55:24:51, 5.70s/it]
20%|██ | 8934/43939 [14:27:01<56:01:39, 5.76s/it]
20%|██ | 8935/43939 [14:27:07<55:47:17, 5.74s/it]
20%|██ | 8936/43939 [14:27:13<55:36:42, 5.72s/it]
20%|██ | 8937/43939 [14:27:18<55:32:03, 5.71s/it]
20%|██ | 8938/43939 [14:27:24<55:28:18, 5.71s/it]
20%|██ | 8939/43939 [14:27:30<55:34:14, 5.72s/it]
20%|██ | 8940/43939 [14:27:35<55:28:40, 5.71s/it]
20%|██ | 8941/43939 [14:27:41<55:29:07, 5.71s/it]
20%|██ | 8942/43939 [14:27:47<55:22:31, 5.70s/it]
20%|██ | 8943/43939 [14:27:52<55:29:02, 5.71s/it]
20%|██ | 8944/43939 [14:27:58<55:21:37, 5.70s/it]
20%|██ | 8945/43939 [14:28:04<55:26:03, 5.70s/it]
20%|██ | 8946/43939 [14:28:10<55:32:35, 5.71s/it]
20%|██ | 8947/43939 [14:28:15<55:30:11, 5.71s/it]
20%|██ | 8948/43939 [14:28:21<55:26:50, 5.70s/it]
20%|██ | 8949/43939 [14:28:27<55:24:44, 5.70s/it]
20%|██ | 8950/43939 [14:28:32<55:23:31, 5.70s/it]
{'loss': 1.0305, 'grad_norm': 0.937384843826294, 'learning_rate': 4.838033226142008e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.59, 'epoch': 0.81}
20%|██ | 8950/43939 [14:28:32<55:23:31, 5.70s/it]
20%|██ | 8951/43939 [14:28:38<55:22:56, 5.70s/it]
20%|██ | 8952/43939 [14:28:44<55:18:58, 5.69s/it]
20%|██ | 8953/43939 [14:28:49<55:29:38, 5.71s/it]
20%|██ | 8954/43939 [14:28:55<55:34:13, 5.72s/it]
20%|██ | 8955/43939 [14:29:01<56:05:48, 5.77s/it]
20%|██ | 8956/43939 [14:29:07<55:51:00, 5.75s/it]
20%|██ | 8957/43939 [14:29:13<55:40:36, 5.73s/it]
20%|██ | 8958/43939 [14:29:18<55:27:29, 5.71s/it]
20%|██ | 8959/43939 [14:29:24<55:32:08, 5.72s/it]
20%|██ | 8960/43939 [14:29:30<55:27:29, 5.71s/it]
20%|██ | 8961/43939 [14:29:35<55:24:50, 5.70s/it]
20%|██ | 8962/43939 [14:29:41<55:21:40, 5.70s/it]
20%|██ | 8963/43939 [14:29:47<55:23:25, 5.70s/it]
20%|██ | 8964/43939 [14:29:52<55:18:43, 5.69s/it]
20%|██ | 8965/43939 [14:29:58<55:29:03, 5.71s/it]
20%|██ | 8966/43939 [14:30:04<55:27:22, 5.71s/it]
20%|██ | 8967/43939 [14:30:10<55:28:04, 5.71s/it]
20%|██ | 8968/43939 [14:30:15<55:28:06, 5.71s/it]
20%|██ | 8969/43939 [14:30:21<55:29:55, 5.71s/it]
20%|██ | 8970/43939 [14:30:27<55:17:40, 5.69s/it]
20%|██ | 8971/43939 [14:30:32<55:15:57, 5.69s/it]
20%|██ | 8972/43939 [14:30:38<55:18:38, 5.69s/it]
20%|██ | 8973/43939 [14:30:44<55:25:37, 5.71s/it]
20%|██ | 8974/43939 [14:30:49<55:28:57, 5.71s/it]
20%|██ | 8975/43939 [14:30:55<55:28:13, 5.71s/it]
20%|██ | 8976/43939 [14:31:01<55:20:01, 5.70s/it]
20%|██ | 8977/43939 [14:31:07<56:11:57, 5.79s/it]
20%|██ | 8978/43939 [14:31:13<55:57:46, 5.76s/it]
20%|██ | 8979/43939 [14:31:18<55:44:25, 5.74s/it]
20%|██ | 8980/43939 [14:31:24<55:36:51, 5.73s/it]
20%|██ | 8981/43939 [14:31:30<55:35:52, 5.73s/it]
20%|██ | 8982/43939 [14:31:35<55:28:34, 5.71s/it]
20%|██ | 8983/43939 [14:31:41<55:34:34, 5.72s/it]
20%|██ | 8984/43939 [14:31:47<55:26:20, 5.71s/it]
20%|██ | 8985/43939 [14:31:52<55:21:37, 5.70s/it]
20%|██ | 8986/43939 [14:31:58<55:30:13, 5.72s/it]
20%|██ | 8987/43939 [14:32:04<55:26:18, 5.71s/it]
20%|██ | 8988/43939 [14:32:10<55:27:29, 5.71s/it]
20%|██ | 8989/43939 [14:32:15<55:28:05, 5.71s/it]
20%|██ | 8990/43939 [14:32:21<55:25:39, 5.71s/it]
20%|██ | 8991/43939 [14:32:27<55:27:09, 5.71s/it]
20%|██ | 8992/43939 [14:32:32<55:31:37, 5.72s/it]
20%|██ | 8993/43939 [14:32:38<55:38:44, 5.73s/it]
20%|██ | 8994/43939 [14:32:44<55:37:59, 5.73s/it]
20%|██ | 8995/43939 [14:32:50<55:26:39, 5.71s/it]
20%|██ | 8996/43939 [14:32:55<55:31:10, 5.72s/it]
20%|██ | 8997/43939 [14:33:01<55:33:53, 5.72s/it]
20%|██ | 8998/43939 [14:33:07<56:12:08, 5.79s/it]
20%|██ | 8999/43939 [14:33:13<55:58:51, 5.77s/it]
20%|██ | 9000/43939 [14:33:18<55:41:13, 5.74s/it]
{'loss': 1.0417, 'grad_norm': 0.9430196285247803, 'learning_rate': 4.8344986631801944e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.31, 'epoch': 0.82}
20%|██ | 9000/43939 [14:33:18<55:41:13, 5.74s/it]
20%|██ | 9001/43939 [14:33:24<55:31:13, 5.72s/it]
20%|██ | 9002/43939 [14:33:30<55:30:44, 5.72s/it]
20%|██ | 9003/43939 [14:33:36<55:34:10, 5.73s/it]
20%|██ | 9004/43939 [14:33:41<55:33:16, 5.72s/it]
20%|██ | 9005/43939 [14:33:47<55:30:17, 5.72s/it]
20%|██ | 9006/43939 [14:33:53<55:21:03, 5.70s/it]
20%|██ | 9007/43939 [14:33:58<55:24:11, 5.71s/it]
21%|██ | 9008/43939 [14:34:04<55:25:41, 5.71s/it]
21%|██ | 9009/43939 [14:34:10<55:25:41, 5.71s/it]
21%|██ | 9010/43939 [14:34:16<55:26:54, 5.71s/it]
21%|██ | 9011/43939 [14:34:21<55:24:28, 5.71s/it]
21%|██ | 9012/43939 [14:34:27<55:25:32, 5.71s/it]
21%|██ | 9013/43939 [14:34:33<55:23:32, 5.71s/it]
21%|██ | 9014/43939 [14:34:38<55:22:29, 5.71s/it]
21%|██ | 9015/43939 [14:34:44<55:21:35, 5.71s/it]
21%|██ | 9016/43939 [14:34:50<55:15:52, 5.70s/it]
21%|██ | 9017/43939 [14:34:55<55:21:10, 5.71s/it]
21%|██ | 9018/43939 [14:35:01<55:24:07, 5.71s/it]
21%|██ | 9019/43939 [14:35:07<55:59:14, 5.77s/it]
21%|██ | 9020/43939 [14:35:13<55:51:20, 5.76s/it]
21%|██ | 9021/43939 [14:35:19<55:43:57, 5.75s/it]
21%|██ | 9022/43939 [14:35:24<55:33:43, 5.73s/it]
21%|██ | 9023/43939 [14:35:30<55:40:01, 5.74s/it]
21%|██ | 9024/43939 [14:35:36<55:38:00, 5.74s/it]
21%|██ | 9025/43939 [14:35:41<55:36:57, 5.73s/it]
21%|██ | 9026/43939 [14:35:47<55:30:35, 5.72s/it]
21%|██ | 9027/43939 [14:35:53<55:33:14, 5.73s/it]
21%|██ | 9028/43939 [14:35:59<55:27:42, 5.72s/it]
21%|██ | 9029/43939 [14:36:04<55:31:42, 5.73s/it]
21%|██ | 9030/43939 [14:36:10<55:23:03, 5.71s/it]
21%|██ | 9031/43939 [14:36:16<55:19:56, 5.71s/it]
21%|██ | 9032/43939 [14:36:21<55:22:41, 5.71s/it]
21%|██ | 9033/43939 [14:36:27<55:25:23, 5.72s/it]
21%|██ | 9034/43939 [14:36:33<55:28:27, 5.72s/it]
21%|██ | 9035/43939 [14:36:39<55:21:23, 5.71s/it]
21%|██ | 9036/43939 [14:36:44<55:18:03, 5.70s/it]
21%|██ | 9037/43939 [14:36:50<55:20:43, 5.71s/it]
21%|██ | 9038/43939 [14:36:56<55:14:30, 5.70s/it]
21%|██ | 9039/43939 [14:37:01<55:14:36, 5.70s/it]
21%|██ | 9040/43939 [14:37:07<56:00:00, 5.78s/it]
21%|██ | 9041/43939 [14:37:13<55:41:06, 5.74s/it]
21%|██ | 9042/43939 [14:37:19<55:36:44, 5.74s/it]
21%|██ | 9043/43939 [14:37:24<55:29:47, 5.73s/it]
21%|██ | 9044/43939 [14:37:30<55:36:47, 5.74s/it]
21%|██ | 9045/43939 [14:37:36<55:23:36, 5.71s/it]
21%|██ | 9046/43939 [14:37:42<55:18:36, 5.71s/it]
21%|██ | 9047/43939 [14:37:47<55:19:33, 5.71s/it]
21%|██ | 9048/43939 [14:37:53<55:23:41, 5.72s/it]
21%|██ | 9049/43939 [14:37:59<55:26:07, 5.72s/it]
21%|██ | 9050/43939 [14:38:04<55:20:41, 5.71s/it]
{'loss': 1.0416, 'grad_norm': 1.0032052993774414, 'learning_rate': 4.8309272680159723e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 546.53, 'epoch': 0.82}
21%|██ | 9050/43939 [14:38:04<55:20:41, 5.71s/it]
21%|██ | 9051/43939 [14:38:10<55:19:56, 5.71s/it]
21%|██ | 9052/43939 [14:38:16<55:22:20, 5.71s/it]
21%|██ | 9053/43939 [14:38:21<55:18:53, 5.71s/it]
21%|██ | 9054/43939 [14:38:27<55:21:03, 5.71s/it]
21%|██ | 9055/43939 [14:38:33<55:21:03, 5.71s/it]
21%|██ | 9056/43939 [14:38:39<55:17:46, 5.71s/it]
21%|██ | 9057/43939 [14:38:44<55:17:42, 5.71s/it]
21%|██ | 9058/43939 [14:38:50<55:12:09, 5.70s/it]
21%|██ | 9059/43939 [14:38:56<55:12:31, 5.70s/it]
21%|██ | 9060/43939 [14:39:01<55:13:14, 5.70s/it]
21%|██ | 9061/43939 [14:39:07<55:14:12, 5.70s/it]
21%|██ | 9062/43939 [14:39:13<55:52:16, 5.77s/it]
21%|██ | 9063/43939 [14:39:19<55:41:29, 5.75s/it]
21%|██ | 9064/43939 [14:39:24<55:34:58, 5.74s/it]
21%|██ | 9065/43939 [14:39:30<55:38:58, 5.74s/it]
21%|██ | 9066/43939 [14:39:36<55:34:13, 5.74s/it]
21%|██ | 9067/43939 [14:39:42<55:24:13, 5.72s/it]
21%|██ | 9068/43939 [14:39:47<55:18:58, 5.71s/it]
21%|██ | 9069/43939 [14:39:53<55:24:29, 5.72s/it]
21%|██ | 9070/43939 [14:39:59<55:34:50, 5.74s/it]
21%|██ | 9071/43939 [14:40:05<55:30:39, 5.73s/it]
21%|██ | 9072/43939 [14:40:10<55:24:14, 5.72s/it]
21%|██ | 9073/43939 [14:40:16<55:17:48, 5.71s/it]
21%|██ | 9074/43939 [14:40:22<55:17:34, 5.71s/it]
21%|██ | 9075/43939 [14:40:27<55:18:24, 5.71s/it]
21%|██ | 9076/43939 [14:40:33<55:18:01, 5.71s/it]
21%|██ | 9077/43939 [14:40:39<55:16:24, 5.71s/it]
21%|██ | 9078/43939 [14:40:44<55:09:30, 5.70s/it]
21%|██ | 9079/43939 [14:40:50<55:09:00, 5.70s/it]
21%|██ | 9080/43939 [14:40:56<55:03:00, 5.69s/it]
21%|██ | 9081/43939 [14:41:01<54:59:25, 5.68s/it]
21%|██ | 9082/43939 [14:41:07<55:38:19, 5.75s/it]
21%|██ | 9083/43939 [14:41:13<55:33:10, 5.74s/it]
21%|██ | 9084/43939 [14:41:19<55:30:09, 5.73s/it]
21%|██ | 9085/43939 [14:41:24<55:26:26, 5.73s/it]
21%|██ | 9086/43939 [14:41:30<55:15:54, 5.71s/it]
21%|██ | 9087/43939 [14:41:36<55:20:09, 5.72s/it]
21%|██ | 9088/43939 [14:41:42<55:17:59, 5.71s/it]
21%|██ | 9089/43939 [14:41:47<55:11:07, 5.70s/it]
21%|██ | 9090/43939 [14:41:53<55:13:41, 5.71s/it]
21%|██ | 9091/43939 [14:41:59<55:15:05, 5.71s/it]
21%|██ | 9092/43939 [14:42:04<55:17:56, 5.71s/it]
21%|██ | 9093/43939 [14:42:10<55:20:31, 5.72s/it]
21%|██ | 9094/43939 [14:42:16<55:09:21, 5.70s/it]
21%|██ | 9095/43939 [14:42:21<55:05:06, 5.69s/it]
21%|██ | 9096/43939 [14:42:27<55:08:24, 5.70s/it]
21%|██ | 9097/43939 [14:42:33<55:08:42, 5.70s/it]
21%|██ | 9098/43939 [14:42:39<55:09:10, 5.70s/it]
21%|██ | 9099/43939 [14:42:44<55:09:07, 5.70s/it]
21%|██ | 9100/43939 [14:42:50<55:08:25, 5.70s/it]
{'loss': 1.0054, 'grad_norm': 1.0585726499557495, 'learning_rate': 4.8273190969964936e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.0, 'epoch': 0.83}
21%|██ | 9100/43939 [14:42:50<55:08:25, 5.70s/it]
21%|██ | 9101/43939 [14:42:56<55:15:13, 5.71s/it]
21%|██ | 9102/43939 [14:43:01<55:12:51, 5.71s/it]
21%|██ | 9103/43939 [14:43:07<55:47:32, 5.77s/it]
21%|██ | 9104/43939 [14:43:13<55:48:11, 5.77s/it]
21%|██ | 9105/43939 [14:43:19<55:41:49, 5.76s/it]
21%|██ | 9106/43939 [14:43:25<55:52:46, 5.78s/it]
21%|██ | 9107/43939 [14:43:30<55:36:19, 5.75s/it]
21%|██ | 9108/43939 [14:43:36<55:31:13, 5.74s/it]
21%|██ | 9109/43939 [14:43:42<55:22:09, 5.72s/it]
21%|██ | 9110/43939 [14:43:47<55:20:57, 5.72s/it]
21%|██ | 9111/43939 [14:43:53<55:20:18, 5.72s/it]
21%|██ | 9112/43939 [14:43:59<55:14:51, 5.71s/it]
21%|██ | 9113/43939 [14:44:05<55:21:37, 5.72s/it]
21%|██ | 9114/43939 [14:44:10<55:18:45, 5.72s/it]
21%|██ | 9115/43939 [14:44:16<55:20:43, 5.72s/it]
21%|██ | 9116/43939 [14:44:22<55:23:06, 5.73s/it]
21%|██ | 9117/43939 [14:44:27<55:21:01, 5.72s/it]
21%|██ | 9118/43939 [14:44:33<55:23:39, 5.73s/it]
21%|██ | 9119/43939 [14:44:39<55:17:48, 5.72s/it]
21%|██ | 9120/43939 [14:44:45<55:21:10, 5.72s/it]
21%|██ | 9121/43939 [14:44:50<55:20:00, 5.72s/it]
21%|██ | 9122/43939 [14:44:56<55:30:46, 5.74s/it]
21%|██ | 9123/43939 [14:45:02<55:23:20, 5.73s/it]
21%|██ | 9124/43939 [14:45:08<55:12:19, 5.71s/it]
21%|██ | 9125/43939 [14:45:14<56:00:21, 5.79s/it]
21%|██ | 9126/43939 [14:45:19<55:45:58, 5.77s/it]
21%|██ | 9127/43939 [14:45:25<55:37:47, 5.75s/it]
21%|██ | 9128/43939 [14:45:31<55:25:38, 5.73s/it]
21%|██ | 9129/43939 [14:45:36<55:21:40, 5.73s/it]
21%|██ | 9130/43939 [14:45:42<55:19:34, 5.72s/it]
21%|██ | 9131/43939 [14:45:48<55:15:14, 5.71s/it]
21%|██ | 9132/43939 [14:45:53<55:10:38, 5.71s/it]
21%|██ | 9133/43939 [14:45:59<55:07:13, 5.70s/it]
21%|██ | 9134/43939 [14:46:05<55:05:55, 5.70s/it]
21%|██ | 9135/43939 [14:46:11<55:07:18, 5.70s/it]
21%|██ | 9136/43939 [14:46:16<55:15:49, 5.72s/it]
21%|██ | 9137/43939 [14:46:22<55:25:04, 5.73s/it]
21%|██ | 9138/43939 [14:46:28<55:20:35, 5.72s/it]
21%|██ | 9139/43939 [14:46:33<55:20:33, 5.73s/it]
21%|██ | 9140/43939 [14:46:39<55:07:51, 5.70s/it]
21%|██ | 9141/43939 [14:46:45<55:04:31, 5.70s/it]
21%|██ | 9142/43939 [14:46:51<55:16:27, 5.72s/it]
21%|██ | 9143/43939 [14:46:56<55:22:29, 5.73s/it]
21%|██ | 9144/43939 [14:47:02<55:19:24, 5.72s/it]
21%|██ | 9145/43939 [14:47:08<55:49:14, 5.78s/it]
21%|██ | 9146/43939 [14:47:14<55:39:22, 5.76s/it]
21%|██ | 9147/43939 [14:47:19<55:26:32, 5.74s/it]
21%|██ | 9148/43939 [14:47:25<55:23:29, 5.73s/it]
21%|██ | 9149/43939 [14:47:31<55:15:17, 5.72s/it]
21%|██ | 9150/43939 [14:47:36<55:04:21, 5.70s/it]
{'loss': 1.0343, 'grad_norm': 1.1669644117355347, 'learning_rate': 4.8236742070491356e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 534.53, 'epoch': 0.83}
21%|██ | 9150/43939 [14:47:36<55:04:21, 5.70s/it]
21%|██ | 9151/43939 [14:47:42<55:00:39, 5.69s/it]
21%|██ | 9152/43939 [14:47:48<55:01:49, 5.69s/it]
21%|██ | 9153/43939 [14:47:54<55:09:13, 5.71s/it]
21%|██ | 9154/43939 [14:47:59<55:19:50, 5.73s/it]
21%|██ | 9155/43939 [14:48:05<55:17:38, 5.72s/it]
21%|██ | 9156/43939 [14:48:11<55:13:00, 5.71s/it]
21%|██ | 9157/43939 [14:48:16<55:11:57, 5.71s/it]
21%|██ | 9158/43939 [14:48:22<55:23:20, 5.73s/it]
21%|██ | 9159/43939 [14:48:28<55:13:40, 5.72s/it]
21%|██ | 9160/43939 [14:48:34<55:07:34, 5.71s/it]
21%|██ | 9161/43939 [14:48:39<55:01:02, 5.70s/it]
21%|██ | 9162/43939 [14:48:45<55:06:04, 5.70s/it]
21%|██ | 9163/43939 [14:48:51<55:05:21, 5.70s/it]
21%|██ | 9164/43939 [14:48:56<55:04:31, 5.70s/it]
21%|██ | 9165/43939 [14:49:02<55:04:29, 5.70s/it]
21%|██ | 9166/43939 [14:49:08<55:43:20, 5.77s/it]
21%|██ | 9167/43939 [14:49:14<55:27:56, 5.74s/it]
21%|██ | 9168/43939 [14:49:19<55:31:00, 5.75s/it]
21%|██ | 9169/43939 [14:49:25<55:22:16, 5.73s/it]
21%|██ | 9170/43939 [14:49:31<55:15:49, 5.72s/it]
21%|██ | 9171/43939 [14:49:37<55:12:38, 5.72s/it]
21%|██ | 9172/43939 [14:49:42<55:11:51, 5.72s/it]
21%|██ | 9173/43939 [14:49:48<55:07:24, 5.71s/it]
21%|██ | 9174/43939 [14:49:54<55:19:57, 5.73s/it]
21%|██ | 9175/43939 [14:49:59<55:16:59, 5.72s/it]
21%|██ | 9176/43939 [14:50:05<55:14:03, 5.72s/it]
21%|██ | 9177/43939 [14:50:11<55:13:48, 5.72s/it]
21%|██ | 9178/43939 [14:50:17<55:15:25, 5.72s/it]
21%|██ | 9179/43939 [14:50:22<55:16:01, 5.72s/it]
21%|██ | 9180/43939 [14:50:28<55:14:04, 5.72s/it]
21%|██ | 9181/43939 [14:50:34<55:11:54, 5.72s/it]
21%|██ | 9182/43939 [14:50:39<55:13:53, 5.72s/it]
21%|██ | 9183/43939 [14:50:45<55:09:14, 5.71s/it]
21%|██ | 9184/43939 [14:50:51<55:08:00, 5.71s/it]
21%|██ | 9185/43939 [14:50:57<55:01:59, 5.70s/it]
21%|██ | 9186/43939 [14:51:02<54:55:32, 5.69s/it]
21%|██ | 9187/43939 [14:51:08<54:55:31, 5.69s/it]
21%|██ | 9188/43939 [14:51:14<55:37:30, 5.76s/it]
21%|██ | 9189/43939 [14:51:20<55:34:07, 5.76s/it]
21%|██ | 9190/43939 [14:51:25<55:37:09, 5.76s/it]
21%|██ | 9191/43939 [14:51:31<55:31:21, 5.75s/it]
21%|██ | 9192/43939 [14:51:37<55:20:04, 5.73s/it]
21%|██ | 9193/43939 [14:51:42<55:13:43, 5.72s/it]
21%|██ | 9194/43939 [14:51:48<55:12:25, 5.72s/it]
21%|██ | 9195/43939 [14:51:54<55:03:30, 5.70s/it]
21%|██ | 9196/43939 [14:52:00<55:11:23, 5.72s/it]
21%|██ | 9197/43939 [14:52:05<55:11:25, 5.72s/it]
21%|██ | 9198/43939 [14:52:11<55:12:36, 5.72s/it]
21%|██ | 9199/43939 [14:52:17<55:16:14, 5.73s/it]
21%|██ | 9200/43939 [14:52:22<55:02:40, 5.70s/it]
{'loss': 1.0371, 'grad_norm': 1.120493769645691, 'learning_rate': 4.819992655680601e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.88, 'epoch': 0.84}
21%|██ | 9200/43939 [14:52:22<55:02:40, 5.70s/it]
21%|██ | 9201/43939 [14:52:28<54:56:54, 5.69s/it]
21%|██ | 9202/43939 [14:52:34<55:01:18, 5.70s/it]
21%|██ | 9203/43939 [14:52:40<55:02:07, 5.70s/it]
21%|██ | 9204/43939 [14:52:45<55:06:30, 5.71s/it]
21%|██ | 9205/43939 [14:52:51<55:01:36, 5.70s/it]
21%|██ | 9206/43939 [14:52:57<55:00:24, 5.70s/it]
21%|██ | 9207/43939 [14:53:02<55:08:10, 5.71s/it]
21%|██ | 9208/43939 [14:53:08<55:09:47, 5.72s/it]
21%|██ | 9209/43939 [14:53:14<55:43:14, 5.78s/it]
21%|██ | 9210/43939 [14:53:20<55:28:48, 5.75s/it]
21%|██ | 9211/43939 [14:53:25<55:14:02, 5.73s/it]
21%|██ | 9212/43939 [14:53:31<55:09:14, 5.72s/it]
21%|██ | 9213/43939 [14:53:37<55:07:29, 5.71s/it]
21%|██ | 9214/43939 [14:53:43<55:13:41, 5.73s/it]
21%|██ | 9215/43939 [14:53:48<55:13:22, 5.73s/it]
21%|██ | 9216/43939 [14:53:54<55:12:34, 5.72s/it]
21%|██ | 9217/43939 [14:54:00<55:08:18, 5.72s/it]
21%|██ | 9218/43939 [14:54:05<55:04:49, 5.71s/it]
21%|██ | 9219/43939 [14:54:11<55:06:04, 5.71s/it]
21%|██ | 9220/43939 [14:54:17<55:12:24, 5.72s/it]
21%|██ | 9221/43939 [14:54:23<55:13:15, 5.73s/it]
21%|██ | 9222/43939 [14:54:28<55:12:44, 5.73s/it]
21%|██ | 9223/43939 [14:54:34<55:08:45, 5.72s/it]
21%|██ | 9224/43939 [14:54:40<55:08:14, 5.72s/it]
21%|██ | 9225/43939 [14:54:45<55:04:31, 5.71s/it]
21%|██ | 9226/43939 [14:54:51<55:04:51, 5.71s/it]
21%|██ | 9227/43939 [14:54:57<54:57:05, 5.70s/it]
21%|██ | 9228/43939 [14:55:03<55:04:33, 5.71s/it]
21%|██ | 9229/43939 [14:55:08<55:01:45, 5.71s/it]
21%|██ | 9230/43939 [14:55:14<55:39:55, 5.77s/it]
21%|██ | 9231/43939 [14:55:20<55:25:40, 5.75s/it]
21%|██ | 9232/43939 [14:55:26<55:14:16, 5.73s/it]
21%|██ | 9233/43939 [14:55:31<55:03:54, 5.71s/it]
21%|██ | 9234/43939 [14:55:37<54:59:35, 5.70s/it]
21%|██ | 9235/43939 [14:55:43<55:03:15, 5.71s/it]
21%|██ | 9236/43939 [14:55:48<55:09:05, 5.72s/it]
21%|██ | 9237/43939 [14:55:54<55:06:32, 5.72s/it]
21%|██ | 9238/43939 [14:56:00<55:01:11, 5.71s/it]
21%|██ | 9239/43939 [14:56:05<55:00:05, 5.71s/it]
21%|██ | 9240/43939 [14:56:11<55:05:29, 5.72s/it]
21%|██ | 9241/43939 [14:56:17<55:02:28, 5.71s/it]
21%|██ | 9242/43939 [14:56:23<54:55:37, 5.70s/it]
21%|██ | 9243/43939 [14:56:28<54:50:12, 5.69s/it]
21%|██ | 9244/43939 [14:56:34<54:50:35, 5.69s/it]
21%|██ | 9245/43939 [14:56:40<54:47:43, 5.69s/it]
21%|██ | 9246/43939 [14:56:45<54:51:23, 5.69s/it]
21%|██ | 9247/43939 [14:56:51<54:53:41, 5.70s/it]
21%|██ | 9248/43939 [14:56:57<54:51:17, 5.69s/it]
21%|██ | 9249/43939 [14:57:02<54:46:11, 5.68s/it]
21%|██ | 9250/43939 [14:57:08<54:47:37, 5.69s/it]
{'loss': 1.0289, 'grad_norm': 0.9619265198707581, 'learning_rate': 4.816274500976016e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.97, 'epoch': 0.84}
21%|██ | 9250/43939 [14:57:08<54:47:37, 5.69s/it]
21%|██ | 9251/43939 [14:57:14<55:35:37, 5.77s/it]
21%|██ | 9252/43939 [14:57:20<55:26:18, 5.75s/it]
21%|██ | 9253/43939 [14:57:25<55:15:55, 5.74s/it]
21%|██ | 9254/43939 [14:57:31<55:10:23, 5.73s/it]
21%|██ | 9255/43939 [14:57:37<55:06:58, 5.72s/it]
21%|██ | 9256/43939 [14:57:43<55:02:17, 5.71s/it]
21%|██ | 9257/43939 [14:57:48<55:01:14, 5.71s/it]
21%|██ | 9258/43939 [14:57:54<55:07:08, 5.72s/it]
21%|██ | 9259/43939 [14:58:00<55:04:31, 5.72s/it]
21%|██ | 9260/43939 [14:58:05<54:54:57, 5.70s/it]
21%|██ | 9261/43939 [14:58:11<54:49:22, 5.69s/it]
21%|██ | 9262/43939 [14:58:17<54:49:49, 5.69s/it]
21%|██ | 9263/43939 [14:58:22<54:51:57, 5.70s/it]
21%|██ | 9264/43939 [14:58:28<54:51:31, 5.70s/it]
21%|██ | 9265/43939 [14:58:34<54:46:16, 5.69s/it]
21%|██ | 9266/43939 [14:58:39<54:45:24, 5.69s/it]
21%|██ | 9267/43939 [14:58:45<54:50:35, 5.69s/it]
21%|██ | 9268/43939 [14:58:51<54:45:39, 5.69s/it]
21%|██ | 9269/43939 [14:58:57<54:52:53, 5.70s/it]
21%|██ | 9270/43939 [14:59:02<54:53:58, 5.70s/it]
21%|██ | 9271/43939 [14:59:08<54:49:33, 5.69s/it]
21%|██ | 9272/43939 [14:59:14<55:35:49, 5.77s/it]
21%|██ | 9273/43939 [14:59:20<55:38:15, 5.78s/it]
21%|██ | 9274/43939 [14:59:25<55:23:34, 5.75s/it]
21%|██ | 9275/43939 [14:59:31<55:20:22, 5.75s/it]
21%|██ | 9276/43939 [14:59:37<55:23:48, 5.75s/it]
21%|██ | 9277/43939 [14:59:43<55:09:23, 5.73s/it]
21%|██ | 9278/43939 [14:59:48<55:05:46, 5.72s/it]
21%|██ | 9279/43939 [14:59:54<54:56:57, 5.71s/it]
21%|██ | 9280/43939 [15:00:00<55:04:38, 5.72s/it]
21%|██ | 9281/43939 [15:00:05<55:03:38, 5.72s/it]
21%|██ | 9282/43939 [15:00:11<54:55:27, 5.71s/it]
21%|██ | 9283/43939 [15:00:17<54:51:39, 5.70s/it]
21%|██ | 9284/43939 [15:00:22<54:47:18, 5.69s/it]
21%|██ | 9285/43939 [15:00:28<54:52:53, 5.70s/it]
21%|██ | 9286/43939 [15:00:34<54:51:01, 5.70s/it]
21%|██ | 9287/43939 [15:00:40<54:51:16, 5.70s/it]
21%|██ | 9288/43939 [15:00:45<54:48:14, 5.69s/it]
21%|██ | 9289/43939 [15:00:51<54:49:02, 5.70s/it]
21%|██ | 9290/43939 [15:00:57<54:48:12, 5.69s/it]
21%|██ | 9291/43939 [15:01:02<54:48:30, 5.69s/it]
21%|██ | 9292/43939 [15:01:08<54:43:36, 5.69s/it]
21%|██ | 9293/43939 [15:01:14<55:28:30, 5.76s/it]
21%|██ | 9294/43939 [15:01:20<55:26:26, 5.76s/it]
21%|██ | 9295/43939 [15:01:25<55:25:00, 5.76s/it]
21%|██ | 9296/43939 [15:01:31<55:22:06, 5.75s/it]
21%|██ | 9297/43939 [15:01:37<55:14:40, 5.74s/it]
21%|██ | 9298/43939 [15:01:43<55:08:20, 5.73s/it]
21%|██ | 9299/43939 [15:01:48<55:05:29, 5.73s/it]
21%|██ | 9300/43939 [15:01:54<55:04:14, 5.72s/it]
{'loss': 1.029, 'grad_norm': 1.0850692987442017, 'learning_rate': 4.812519801598007e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.02, 'epoch': 0.85}
21%|██ | 9300/43939 [15:01:54<55:04:14, 5.72s/it]
21%|██ | 9301/43939 [15:02:00<54:57:52, 5.71s/it]
21%|██ | 9302/43939 [15:02:05<54:54:16, 5.71s/it]
21%|██ | 9303/43939 [15:02:11<54:53:06, 5.70s/it]
21%|██ | 9304/43939 [15:02:17<54:52:22, 5.70s/it]
21%|██ | 9305/43939 [15:02:23<54:51:50, 5.70s/it]
21%|██ | 9306/43939 [15:02:28<54:43:51, 5.69s/it]
21%|██ | 9307/43939 [15:02:34<54:39:44, 5.68s/it]
21%|██ | 9308/43939 [15:02:40<54:42:00, 5.69s/it]
21%|██ | 9309/43939 [15:02:45<54:35:58, 5.68s/it]
21%|██ | 9310/43939 [15:02:51<54:38:09, 5.68s/it]
21%|██ | 9311/43939 [15:02:57<54:41:22, 5.69s/it]
21%|██ | 9312/43939 [15:03:02<54:48:33, 5.70s/it]
21%|██ | 9313/43939 [15:03:08<54:53:25, 5.71s/it]
21%|██ | 9314/43939 [15:03:14<55:28:56, 5.77s/it]
21%|██ | 9315/43939 [15:03:20<55:15:40, 5.75s/it]
21%|██ | 9316/43939 [15:03:25<55:04:22, 5.73s/it]
21%|██ | 9317/43939 [15:03:31<54:59:38, 5.72s/it]
21%|██ | 9318/43939 [15:03:37<55:00:06, 5.72s/it]
21%|██ | 9319/43939 [15:03:42<55:02:53, 5.72s/it]
21%|██ | 9320/43939 [15:03:48<54:56:06, 5.71s/it]
21%|██ | 9321/43939 [15:03:54<55:02:20, 5.72s/it]
21%|██ | 9322/43939 [15:04:00<55:07:42, 5.73s/it]
21%|██ | 9323/43939 [15:04:05<55:04:01, 5.73s/it]
21%|██ | 9324/43939 [15:04:11<55:00:49, 5.72s/it]
21%|██ | 9325/43939 [15:04:17<54:50:28, 5.70s/it]
21%|██ | 9326/43939 [15:04:22<54:45:42, 5.70s/it]
21%|██ | 9327/43939 [15:04:28<54:41:48, 5.69s/it]
21%|██ | 9328/43939 [15:04:34<54:44:47, 5.69s/it]
21%|██ | 9329/43939 [15:04:40<54:45:39, 5.70s/it]
21%|██ | 9330/43939 [15:04:45<54:48:21, 5.70s/it]
21%|██ | 9331/43939 [15:04:51<54:41:18, 5.69s/it]
21%|██ | 9332/43939 [15:04:57<54:39:08, 5.69s/it]
21%|██ | 9333/43939 [15:05:02<54:43:13, 5.69s/it]
21%|██ | 9334/43939 [15:05:08<54:40:58, 5.69s/it]
21%|██ | 9335/43939 [15:05:14<54:39:11, 5.69s/it]
21%|██ | 9336/43939 [15:05:20<55:18:30, 5.75s/it]
21%|██ | 9337/43939 [15:05:25<55:18:04, 5.75s/it]
21%|██▏ | 9338/43939 [15:05:31<55:08:51, 5.74s/it]
21%|██▏ | 9339/43939 [15:05:37<54:57:50, 5.72s/it]
21%|██▏ | 9340/43939 [15:05:42<54:57:11, 5.72s/it]
21%|██▏ | 9341/43939 [15:05:48<54:53:51, 5.71s/it]
21%|██▏ | 9342/43939 [15:05:54<54:47:56, 5.70s/it]
21%|██▏ | 9343/43939 [15:05:59<54:47:58, 5.70s/it]
21%|██▏ | 9344/43939 [15:06:05<54:50:48, 5.71s/it]
21%|██▏ | 9345/43939 [15:06:11<54:55:16, 5.72s/it]
21%|██▏ | 9346/43939 [15:06:17<54:48:24, 5.70s/it]
21%|██▏ | 9347/43939 [15:06:22<54:49:35, 5.71s/it]
21%|██▏ | 9348/43939 [15:06:28<54:48:18, 5.70s/it]
21%|██▏ | 9349/43939 [15:06:34<54:48:52, 5.70s/it]
21%|██▏ | 9350/43939 [15:06:39<54:42:29, 5.69s/it]
{'loss': 1.0358, 'grad_norm': 1.7269978523254395, 'learning_rate': 4.808728616785781e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.95, 'epoch': 0.85}
21%|██▏ | 9350/43939 [15:06:39<54:42:29, 5.69s/it]
21%|██▏ | 9351/43939 [15:06:45<54:47:06, 5.70s/it]
21%|██▏ | 9352/43939 [15:06:51<54:46:42, 5.70s/it]
21%|██▏ | 9353/43939 [15:06:56<54:44:08, 5.70s/it]
21%|██▏ | 9354/43939 [15:07:02<54:51:00, 5.71s/it]
21%|██▏ | 9355/43939 [15:07:08<54:48:18, 5.70s/it]
21%|██▏ | 9356/43939 [15:07:14<54:48:54, 5.71s/it]
21%|██▏ | 9357/43939 [15:07:20<55:35:21, 5.79s/it]
21%|██▏ | 9358/43939 [15:07:25<55:17:27, 5.76s/it]
21%|██▏ | 9359/43939 [15:07:31<55:04:35, 5.73s/it]
21%|██▏ | 9360/43939 [15:07:37<54:56:00, 5.72s/it]
21%|██▏ | 9361/43939 [15:07:42<55:02:31, 5.73s/it]
21%|██▏ | 9362/43939 [15:07:48<55:01:15, 5.73s/it]
21%|██▏ | 9363/43939 [15:07:54<54:58:47, 5.72s/it]
21%|██▏ | 9364/43939 [15:08:00<54:59:01, 5.73s/it]
21%|██▏ | 9365/43939 [15:08:05<55:01:13, 5.73s/it]
21%|██▏ | 9366/43939 [15:08:11<54:57:23, 5.72s/it]
21%|██▏ | 9367/43939 [15:08:17<54:54:08, 5.72s/it]
21%|██▏ | 9368/43939 [15:08:22<54:53:01, 5.72s/it]
21%|██▏ | 9369/43939 [15:08:28<54:45:02, 5.70s/it]
21%|██▏ | 9370/43939 [15:08:34<54:45:13, 5.70s/it]
21%|██▏ | 9371/43939 [15:08:39<54:38:25, 5.69s/it]
21%|██▏ | 9372/43939 [15:08:45<54:32:39, 5.68s/it]
21%|██▏ | 9373/43939 [15:08:51<54:36:31, 5.69s/it]
21%|██▏ | 9374/43939 [15:08:57<54:43:22, 5.70s/it]
21%|██▏ | 9375/43939 [15:09:02<54:41:52, 5.70s/it]
21%|██▏ | 9376/43939 [15:09:08<54:40:16, 5.69s/it]
21%|██▏ | 9377/43939 [15:09:14<54:49:30, 5.71s/it]
21%|██▏ | 9378/43939 [15:09:20<55:27:56, 5.78s/it]
21%|██▏ | 9379/43939 [15:09:25<55:16:28, 5.76s/it]
21%|██▏ | 9380/43939 [15:09:31<55:07:15, 5.74s/it]
21%|██▏ | 9381/43939 [15:09:37<55:00:02, 5.73s/it]
21%|██▏ | 9382/43939 [15:09:42<54:53:55, 5.72s/it]
21%|██▏ | 9383/43939 [15:09:48<54:52:19, 5.72s/it]
21%|██▏ | 9384/43939 [15:09:54<54:45:42, 5.71s/it]
21%|██▏ | 9385/43939 [15:10:00<54:43:02, 5.70s/it]
21%|██▏ | 9386/43939 [15:10:05<54:51:41, 5.72s/it]
21%|██▏ | 9387/43939 [15:10:11<54:45:10, 5.70s/it]
21%|██▏ | 9388/43939 [15:10:17<54:43:54, 5.70s/it]
21%|██▏ | 9389/43939 [15:10:22<54:47:20, 5.71s/it]
21%|██▏ | 9390/43939 [15:10:28<54:38:45, 5.69s/it]
21%|██▏ | 9391/43939 [15:10:34<54:39:13, 5.70s/it]
21%|██▏ | 9392/43939 [15:10:39<54:40:01, 5.70s/it]
21%|██▏ | 9393/43939 [15:10:45<54:37:50, 5.69s/it]
21%|██▏ | 9394/43939 [15:10:51<54:33:38, 5.69s/it]
21%|██▏ | 9395/43939 [15:10:56<54:27:34, 5.68s/it]
21%|██▏ | 9396/43939 [15:11:02<54:37:14, 5.69s/it]
21%|██▏ | 9397/43939 [15:11:08<54:30:54, 5.68s/it]
21%|██▏ | 9398/43939 [15:11:14<54:39:46, 5.70s/it]
21%|██▏ | 9399/43939 [15:11:19<54:38:18, 5.69s/it]
21%|██▏ | 9400/43939 [15:11:25<55:24:24, 5.78s/it]
{'loss': 1.0154, 'grad_norm': 1.0903490781784058, 'learning_rate': 4.804901006354188e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.61, 'epoch': 0.86}
21%|██▏ | 9400/43939 [15:11:25<55:24:24, 5.78s/it]
21%|██▏ | 9401/43939 [15:11:31<55:16:37, 5.76s/it]
21%|██▏ | 9402/43939 [15:11:37<55:07:40, 5.75s/it]
21%|██▏ | 9403/43939 [15:11:42<54:58:41, 5.73s/it]
21%|██▏ | 9404/43939 [15:11:48<54:55:17, 5.73s/it]
21%|██▏ | 9405/43939 [15:11:54<54:44:10, 5.71s/it]
21%|██▏ | 9406/43939 [15:11:59<54:44:18, 5.71s/it]
21%|██▏ | 9407/43939 [15:12:05<54:36:23, 5.69s/it]
21%|██▏ | 9408/43939 [15:12:11<54:35:51, 5.69s/it]
21%|██▏ | 9409/43939 [15:12:16<54:36:45, 5.69s/it]
21%|██▏ | 9410/43939 [15:12:22<54:36:56, 5.69s/it]
21%|██▏ | 9411/43939 [15:12:28<54:37:27, 5.70s/it]
21%|██▏ | 9412/43939 [15:12:34<54:39:11, 5.70s/it]
21%|██▏ | 9413/43939 [15:12:39<54:43:44, 5.71s/it]
21%|██▏ | 9414/43939 [15:12:45<54:42:17, 5.70s/it]
21%|██▏ | 9415/43939 [15:12:51<54:43:39, 5.71s/it]
21%|██▏ | 9416/43939 [15:12:56<54:40:42, 5.70s/it]
21%|██▏ | 9417/43939 [15:13:02<54:40:39, 5.70s/it]
21%|██▏ | 9418/43939 [15:13:08<54:50:21, 5.72s/it]
21%|██▏ | 9419/43939 [15:13:14<54:42:04, 5.70s/it]
21%|██▏ | 9420/43939 [15:13:19<54:33:36, 5.69s/it]
21%|██▏ | 9421/43939 [15:13:25<55:19:34, 5.77s/it]
21%|██▏ | 9422/43939 [15:13:31<55:06:59, 5.75s/it]
21%|██▏ | 9423/43939 [15:13:37<54:58:26, 5.73s/it]
21%|██▏ | 9424/43939 [15:13:42<54:50:35, 5.72s/it]
21%|██▏ | 9425/43939 [15:13:48<54:50:57, 5.72s/it]
21%|██▏ | 9426/43939 [15:13:54<54:45:59, 5.71s/it]
21%|██▏ | 9427/43939 [15:13:59<54:43:10, 5.71s/it]
21%|██▏ | 9428/43939 [15:14:05<54:49:48, 5.72s/it]
21%|██▏ | 9429/43939 [15:14:11<54:45:55, 5.71s/it]
21%|██▏ | 9430/43939 [15:14:17<54:47:17, 5.72s/it]
21%|██▏ | 9431/43939 [15:14:22<54:40:37, 5.70s/it]
21%|██▏ | 9432/43939 [15:14:28<54:39:01, 5.70s/it]
21%|██▏ | 9433/43939 [15:14:34<54:40:01, 5.70s/it]
21%|██▏ | 9434/43939 [15:14:39<54:35:49, 5.70s/it]
21%|██▏ | 9435/43939 [15:14:45<54:44:16, 5.71s/it]
21%|██▏ | 9436/43939 [15:14:51<54:37:54, 5.70s/it]
21%|██▏ | 9437/43939 [15:14:56<54:36:30, 5.70s/it]
21%|██▏ | 9438/43939 [15:15:02<54:29:57, 5.69s/it]
21%|██▏ | 9439/43939 [15:15:08<54:37:55, 5.70s/it]
21%|██▏ | 9440/43939 [15:15:14<54:43:00, 5.71s/it]
21%|██▏ | 9441/43939 [15:15:19<55:15:55, 5.77s/it]
21%|██▏ | 9442/43939 [15:15:25<55:03:04, 5.74s/it]
21%|██▏ | 9443/43939 [15:15:31<54:52:56, 5.73s/it]
21%|██▏ | 9444/43939 [15:15:37<54:53:47, 5.73s/it]
21%|██▏ | 9445/43939 [15:15:42<54:44:13, 5.71s/it]
21%|██▏ | 9446/43939 [15:15:48<54:41:56, 5.71s/it]
22%|██▏ | 9447/43939 [15:15:54<54:40:24, 5.71s/it]
22%|██▏ | 9448/43939 [15:15:59<54:36:34, 5.70s/it]
22%|██▏ | 9449/43939 [15:16:05<54:35:30, 5.70s/it]
22%|██▏ | 9450/43939 [15:16:11<54:32:08, 5.69s/it]
{'loss': 1.0186, 'grad_norm': 1.1368602514266968, 'learning_rate': 4.801037030692776e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.83, 'epoch': 0.86}
22%|██▏ | 9450/43939 [15:16:11<54:32:08, 5.69s/it]
22%|██▏ | 9451/43939 [15:16:16<54:29:44, 5.69s/it]
22%|██▏ | 9452/43939 [15:16:22<54:34:02, 5.70s/it]
22%|██▏ | 9453/43939 [15:16:28<54:36:24, 5.70s/it]
22%|██▏ | 9454/43939 [15:16:33<54:32:47, 5.69s/it]
22%|██▏ | 9455/43939 [15:16:39<54:34:31, 5.70s/it]
22%|██▏ | 9456/43939 [15:16:45<54:34:32, 5.70s/it]
22%|██▏ | 9457/43939 [15:16:51<54:40:20, 5.71s/it]
22%|██▏ | 9458/43939 [15:16:56<54:34:15, 5.70s/it]
22%|██▏ | 9459/43939 [15:17:02<54:33:03, 5.70s/it]
22%|██▏ | 9460/43939 [15:17:08<54:36:17, 5.70s/it]
22%|██▏ | 9461/43939 [15:17:13<54:33:12, 5.70s/it]
22%|██▏ | 9462/43939 [15:17:19<55:15:18, 5.77s/it]
22%|██▏ | 9463/43939 [15:17:25<55:11:22, 5.76s/it]
22%|██▏ | 9464/43939 [15:17:31<54:58:06, 5.74s/it]
22%|██▏ | 9465/43939 [15:17:36<54:56:20, 5.74s/it]
22%|██▏ | 9466/43939 [15:17:42<54:47:49, 5.72s/it]
22%|██▏ | 9467/43939 [15:17:48<54:40:59, 5.71s/it]
22%|██▏ | 9468/43939 [15:17:54<54:40:47, 5.71s/it]
22%|██▏ | 9469/43939 [15:17:59<54:30:45, 5.69s/it]
22%|██▏ | 9470/43939 [15:18:05<54:39:25, 5.71s/it]
22%|██▏ | 9471/43939 [15:18:11<54:43:10, 5.72s/it]
22%|██▏ | 9472/43939 [15:18:16<54:38:27, 5.71s/it]
22%|██▏ | 9473/43939 [15:18:22<54:32:40, 5.70s/it]
22%|██▏ | 9474/43939 [15:18:28<54:33:12, 5.70s/it]
22%|██▏ | 9475/43939 [15:18:33<54:32:27, 5.70s/it]
22%|██▏ | 9476/43939 [15:18:39<54:36:06, 5.70s/it]
22%|██▏ | 9477/43939 [15:18:45<54:44:30, 5.72s/it]
22%|██▏ | 9478/43939 [15:18:51<54:33:38, 5.70s/it]
22%|██▏ | 9479/43939 [15:18:56<54:43:00, 5.72s/it]
22%|██▏ | 9480/43939 [15:19:02<54:46:34, 5.72s/it]
22%|██▏ | 9481/43939 [15:19:08<54:42:09, 5.72s/it]
22%|██▏ | 9482/43939 [15:19:13<54:40:27, 5.71s/it]
22%|██▏ | 9483/43939 [15:19:19<54:40:57, 5.71s/it]
22%|██▏ | 9484/43939 [15:19:25<55:25:03, 5.79s/it]
22%|██▏ | 9485/43939 [15:19:31<55:15:21, 5.77s/it]
22%|██▏ | 9486/43939 [15:19:37<55:05:51, 5.76s/it]
22%|██▏ | 9487/43939 [15:19:42<55:01:52, 5.75s/it]
22%|██▏ | 9488/43939 [15:19:48<54:49:21, 5.73s/it]
22%|██▏ | 9489/43939 [15:19:54<54:47:59, 5.73s/it]
22%|██▏ | 9490/43939 [15:19:59<54:39:49, 5.71s/it]
22%|██▏ | 9491/43939 [15:20:05<54:32:26, 5.70s/it]
22%|██▏ | 9492/43939 [15:20:11<54:42:27, 5.72s/it]
22%|██▏ | 9493/43939 [15:20:17<54:37:23, 5.71s/it]
22%|██▏ | 9494/43939 [15:20:22<54:31:16, 5.70s/it]
22%|██▏ | 9495/43939 [15:20:28<54:27:38, 5.69s/it]
22%|██▏ | 9496/43939 [15:20:34<54:29:32, 5.70s/it]
22%|██▏ | 9497/43939 [15:20:39<54:37:24, 5.71s/it]
22%|██▏ | 9498/43939 [15:20:45<54:31:43, 5.70s/it]
22%|██▏ | 9499/43939 [15:20:51<54:24:58, 5.69s/it]
22%|██▏ | 9500/43939 [15:20:56<54:22:54, 5.68s/it]
{'loss': 0.9895, 'grad_norm': 1.1895757913589478, 'learning_rate': 4.7971367507648424e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.56, 'epoch': 0.86}
22%|██▏ | 9500/43939 [15:20:56<54:22:54, 5.68s/it]
22%|██▏ | 9501/43939 [15:21:02<54:25:53, 5.69s/it]
22%|██▏ | 9502/43939 [15:21:08<54:28:15, 5.69s/it]
22%|██▏ | 9503/43939 [15:21:13<54:39:57, 5.71s/it]
22%|██▏ | 9504/43939 [15:21:19<55:17:33, 5.78s/it]
22%|██▏ | 9505/43939 [15:21:25<55:02:45, 5.75s/it]
22%|██▏ | 9506/43939 [15:21:31<54:54:57, 5.74s/it]
22%|██▏ | 9507/43939 [15:21:37<54:48:52, 5.73s/it]
22%|██▏ | 9508/43939 [15:21:42<54:49:00, 5.73s/it]
22%|██▏ | 9509/43939 [15:21:48<54:42:37, 5.72s/it]
22%|██▏ | 9510/43939 [15:21:54<54:39:08, 5.71s/it]
22%|██▏ | 9511/43939 [15:21:59<54:36:48, 5.71s/it]
22%|██▏ | 9512/43939 [15:22:05<54:28:50, 5.70s/it]
22%|██▏ | 9513/43939 [15:22:11<54:28:18, 5.70s/it]
22%|██▏ | 9514/43939 [15:22:16<54:29:01, 5.70s/it]
22%|██▏ | 9515/43939 [15:22:22<54:36:17, 5.71s/it]
22%|██▏ | 9516/43939 [15:22:28<54:34:05, 5.71s/it]
22%|██▏ | 9517/43939 [15:22:34<54:34:58, 5.71s/it]
22%|██▏ | 9518/43939 [15:22:39<54:35:40, 5.71s/it]
22%|██▏ | 9519/43939 [15:22:45<54:40:43, 5.72s/it]
22%|██▏ | 9520/43939 [15:22:51<54:39:57, 5.72s/it]
22%|██▏ | 9521/43939 [15:22:56<54:35:51, 5.71s/it]
22%|██▏ | 9522/43939 [15:23:02<54:35:53, 5.71s/it]
22%|██▏ | 9523/43939 [15:23:08<54:34:59, 5.71s/it]
22%|██▏ | 9524/43939 [15:23:14<54:29:51, 5.70s/it]
22%|██▏ | 9525/43939 [15:23:19<55:03:48, 5.76s/it]
22%|██▏ | 9526/43939 [15:23:25<55:00:14, 5.75s/it]
22%|██▏ | 9527/43939 [15:23:31<54:55:38, 5.75s/it]
22%|██▏ | 9528/43939 [15:23:37<54:48:27, 5.73s/it]
22%|██▏ | 9529/43939 [15:23:42<54:37:55, 5.72s/it]
22%|██▏ | 9530/43939 [15:23:48<54:34:22, 5.71s/it]
22%|██▏ | 9531/43939 [15:23:54<54:21:12, 5.69s/it]
22%|██▏ | 9532/43939 [15:23:59<54:25:25, 5.69s/it]
22%|██▏ | 9533/43939 [15:24:05<54:26:30, 5.70s/it]
22%|██▏ | 9534/43939 [15:24:11<54:36:17, 5.71s/it]
22%|██▏ | 9535/43939 [15:24:16<54:34:07, 5.71s/it]
22%|██▏ | 9536/43939 [15:24:22<54:34:12, 5.71s/it]
22%|██▏ | 9537/43939 [15:24:28<54:33:58, 5.71s/it]
22%|██▏ | 9538/43939 [15:24:34<54:39:42, 5.72s/it]
22%|██▏ | 9539/43939 [15:24:39<54:40:13, 5.72s/it]
22%|██▏ | 9540/43939 [15:24:45<54:34:31, 5.71s/it]
22%|██▏ | 9541/43939 [15:24:51<54:35:01, 5.71s/it]
22%|██▏ | 9542/43939 [15:24:56<54:32:57, 5.71s/it]
22%|██▏ | 9543/43939 [15:25:02<54:28:50, 5.70s/it]
22%|██▏ | 9544/43939 [15:25:08<54:29:08, 5.70s/it]
22%|██▏ | 9545/43939 [15:25:14<54:27:56, 5.70s/it]
22%|██▏ | 9546/43939 [15:25:19<54:31:52, 5.71s/it]
22%|██▏ | 9547/43939 [15:25:25<55:03:58, 5.76s/it]
22%|██▏ | 9548/43939 [15:25:31<54:50:22, 5.74s/it]
22%|██▏ | 9549/43939 [15:25:37<54:45:55, 5.73s/it]
22%|██▏ | 9550/43939 [15:25:42<54:46:48, 5.73s/it]
{'loss': 1.0405, 'grad_norm': 0.9769192934036255, 'learning_rate': 4.793200228106468e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.92, 'epoch': 0.87}
22%|██▏ | 9550/43939 [15:25:42<54:46:48, 5.73s/it]
22%|██▏ | 9551/43939 [15:25:48<54:40:22, 5.72s/it]
22%|██▏ | 9552/43939 [15:25:54<54:38:45, 5.72s/it]
22%|██▏ | 9553/43939 [15:25:59<54:37:30, 5.72s/it]
22%|██▏ | 9554/43939 [15:26:05<54:32:27, 5.71s/it]
22%|██▏ | 9555/43939 [15:26:11<54:30:15, 5.71s/it]
22%|██▏ | 9556/43939 [15:26:16<54:21:34, 5.69s/it]
22%|██▏ | 9557/43939 [15:26:22<54:17:26, 5.68s/it]
22%|██▏ | 9558/43939 [15:26:28<54:14:54, 5.68s/it]
22%|██▏ | 9559/43939 [15:26:34<54:14:41, 5.68s/it]
22%|██▏ | 9560/43939 [15:26:39<54:16:40, 5.68s/it]
22%|██▏ | 9561/43939 [15:26:45<54:25:36, 5.70s/it]
22%|██▏ | 9562/43939 [15:26:51<54:26:18, 5.70s/it]
22%|██▏ | 9563/43939 [15:26:56<54:25:56, 5.70s/it]
22%|██▏ | 9564/43939 [15:27:02<54:27:06, 5.70s/it]
22%|██▏ | 9565/43939 [15:27:08<54:29:22, 5.71s/it]
22%|██▏ | 9566/43939 [15:27:13<54:28:51, 5.71s/it]
22%|██▏ | 9567/43939 [15:27:19<54:27:13, 5.70s/it]
22%|██▏ | 9568/43939 [15:27:25<55:08:32, 5.78s/it]
22%|██▏ | 9569/43939 [15:27:31<54:59:29, 5.76s/it]
22%|██▏ | 9570/43939 [15:27:37<54:44:12, 5.73s/it]
22%|██▏ | 9571/43939 [15:27:42<54:38:10, 5.72s/it]
22%|██▏ | 9572/43939 [15:27:48<54:38:27, 5.72s/it]
22%|██▏ | 9573/43939 [15:27:54<54:31:27, 5.71s/it]
22%|██▏ | 9574/43939 [15:27:59<54:42:27, 5.73s/it]
22%|██▏ | 9575/43939 [15:28:05<54:41:24, 5.73s/it]
22%|██▏ | 9576/43939 [15:28:11<54:30:17, 5.71s/it]
22%|██▏ | 9577/43939 [15:28:16<54:28:59, 5.71s/it]
22%|██▏ | 9578/43939 [15:28:22<54:27:00, 5.70s/it]
22%|██▏ | 9579/43939 [15:28:28<54:23:44, 5.70s/it]
22%|██▏ | 9580/43939 [15:28:34<54:28:38, 5.71s/it]
22%|██▏ | 9581/43939 [15:28:39<54:30:46, 5.71s/it]
22%|██▏ | 9582/43939 [15:28:45<54:26:09, 5.70s/it]
22%|██▏ | 9583/43939 [15:28:51<54:30:08, 5.71s/it]
22%|██▏ | 9584/43939 [15:28:56<54:25:35, 5.70s/it]
22%|██▏ | 9585/43939 [15:29:02<54:23:15, 5.70s/it]
22%|██▏ | 9586/43939 [15:29:08<54:22:10, 5.70s/it]
22%|██▏ | 9587/43939 [15:29:13<54:23:04, 5.70s/it]
22%|██▏ | 9588/43939 [15:29:19<54:27:34, 5.71s/it]
22%|██▏ | 9589/43939 [15:29:25<55:06:23, 5.78s/it]
22%|██▏ | 9590/43939 [15:29:31<54:49:53, 5.75s/it]
22%|██▏ | 9591/43939 [15:29:37<54:38:14, 5.73s/it]
22%|██▏ | 9592/43939 [15:29:42<54:38:15, 5.73s/it]
22%|██▏ | 9593/43939 [15:29:48<54:33:42, 5.72s/it]
22%|██▏ | 9594/43939 [15:29:54<54:30:13, 5.71s/it]
22%|██▏ | 9595/43939 [15:29:59<54:26:35, 5.71s/it]
22%|██▏ | 9596/43939 [15:30:05<54:22:01, 5.70s/it]
22%|██▏ | 9597/43939 [15:30:11<54:16:20, 5.69s/it]
22%|██▏ | 9598/43939 [15:30:16<54:16:38, 5.69s/it]
22%|██▏ | 9599/43939 [15:30:22<54:23:40, 5.70s/it]
22%|██▏ | 9600/43939 [15:30:28<54:17:02, 5.69s/it]
{'loss': 1.0499, 'grad_norm': 1.0490390062332153, 'learning_rate': 4.789227524825549e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.91, 'epoch': 0.87}
22%|██▏ | 9600/43939 [15:30:28<54:17:02, 5.69s/it]
22%|██▏ | 9601/43939 [15:30:33<54:19:24, 5.70s/it]
22%|██▏ | 9602/43939 [15:30:39<54:13:12, 5.68s/it]
22%|██▏ | 9603/43939 [15:30:45<54:19:29, 5.70s/it]
22%|██▏ | 9604/43939 [15:30:51<54:27:38, 5.71s/it]
22%|██▏ | 9605/43939 [15:30:56<54:22:16, 5.70s/it]
22%|██▏ | 9606/43939 [15:31:02<54:21:47, 5.70s/it]
22%|██▏ | 9607/43939 [15:31:08<54:18:36, 5.69s/it]
22%|██▏ | 9608/43939 [15:31:13<54:25:09, 5.71s/it]
22%|██▏ | 9609/43939 [15:31:19<54:36:21, 5.73s/it]
22%|██▏ | 9610/43939 [15:31:25<54:31:22, 5.72s/it]
22%|██▏ | 9611/43939 [15:31:31<55:12:59, 5.79s/it]
22%|██▏ | 9612/43939 [15:31:37<54:59:20, 5.77s/it]
22%|██▏ | 9613/43939 [15:31:42<54:50:30, 5.75s/it]
22%|██▏ | 9614/43939 [15:31:48<54:51:18, 5.75s/it]
22%|██▏ | 9615/43939 [15:31:54<54:40:43, 5.73s/it]
22%|██▏ | 9616/43939 [15:31:59<54:28:08, 5.71s/it]
22%|██▏ | 9617/43939 [15:32:05<54:18:52, 5.70s/it]
22%|██▏ | 9618/43939 [15:32:11<54:29:02, 5.71s/it]
22%|██▏ | 9619/43939 [15:32:16<54:28:09, 5.71s/it]
22%|██▏ | 9620/43939 [15:32:22<54:20:23, 5.70s/it]
22%|██▏ | 9621/43939 [15:32:28<54:18:29, 5.70s/it]
22%|██▏ | 9622/43939 [15:32:34<54:27:54, 5.71s/it]
22%|██▏ | 9623/43939 [15:32:39<54:31:38, 5.72s/it]
22%|██▏ | 9624/43939 [15:32:45<54:29:15, 5.72s/it]
22%|██▏ | 9625/43939 [15:32:51<54:26:14, 5.71s/it]
22%|██▏ | 9626/43939 [15:32:56<54:25:12, 5.71s/it]
22%|██▏ | 9627/43939 [15:33:02<54:24:27, 5.71s/it]
22%|██▏ | 9628/43939 [15:33:08<54:21:41, 5.70s/it]
22%|██▏ | 9629/43939 [15:33:14<54:22:52, 5.71s/it]
22%|██▏ | 9630/43939 [15:33:19<54:25:01, 5.71s/it]
22%|██▏ | 9631/43939 [15:33:25<54:29:36, 5.72s/it]
22%|██▏ | 9632/43939 [15:33:31<55:07:03, 5.78s/it]
22%|██▏ | 9633/43939 [15:33:37<54:51:43, 5.76s/it]
22%|██▏ | 9634/43939 [15:33:42<54:41:13, 5.74s/it]
22%|██▏ | 9635/43939 [15:33:48<54:36:34, 5.73s/it]
22%|██▏ | 9636/43939 [15:33:54<54:26:35, 5.71s/it]
22%|██▏ | 9637/43939 [15:33:59<54:25:47, 5.71s/it]
22%|██▏ | 9638/43939 [15:34:05<54:32:34, 5.72s/it]
22%|██▏ | 9639/43939 [15:34:11<54:31:53, 5.72s/it]
22%|██▏ | 9640/43939 [15:34:17<54:27:48, 5.72s/it]
22%|██▏ | 9641/43939 [15:34:22<54:24:37, 5.71s/it]
22%|██▏ | 9642/43939 [15:34:28<54:20:22, 5.70s/it]
22%|██▏ | 9643/43939 [15:34:34<54:22:22, 5.71s/it]
22%|██▏ | 9644/43939 [15:34:39<54:22:46, 5.71s/it]
22%|██▏ | 9645/43939 [15:34:45<54:20:52, 5.71s/it]
22%|██▏ | 9646/43939 [15:34:51<54:18:53, 5.70s/it]
22%|██▏ | 9647/43939 [15:34:57<54:21:14, 5.71s/it]
22%|██▏ | 9648/43939 [15:35:02<54:19:28, 5.70s/it]
22%|██▏ | 9649/43939 [15:35:08<54:18:10, 5.70s/it]
22%|██▏ | 9650/43939 [15:35:14<54:16:37, 5.70s/it]
{'loss': 1.0322, 'grad_norm': 1.029001235961914, 'learning_rate': 4.785218703600815e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.91, 'epoch': 0.88}
22%|██▏ | 9650/43939 [15:35:14<54:16:37, 5.70s/it]
22%|██▏ | 9651/43939 [15:35:19<54:14:55, 5.70s/it]
22%|██▏ | 9652/43939 [15:35:25<54:07:12, 5.68s/it]
22%|██▏ | 9653/43939 [15:35:31<54:46:38, 5.75s/it]
22%|██▏ | 9654/43939 [15:35:37<54:37:12, 5.74s/it]
22%|██▏ | 9655/43939 [15:35:42<54:28:41, 5.72s/it]
22%|██▏ | 9656/43939 [15:35:48<54:21:10, 5.71s/it]
22%|██▏ | 9657/43939 [15:35:54<54:15:47, 5.70s/it]
22%|██▏ | 9658/43939 [15:35:59<54:20:32, 5.71s/it]
22%|██▏ | 9659/43939 [15:36:05<54:14:24, 5.70s/it]
22%|██▏ | 9660/43939 [15:36:11<54:12:29, 5.69s/it]
22%|██▏ | 9661/43939 [15:36:16<54:14:14, 5.70s/it]
22%|██▏ | 9662/43939 [15:36:22<54:13:17, 5.69s/it]
22%|██▏ | 9663/43939 [15:36:28<54:13:32, 5.70s/it]
22%|██▏ | 9664/43939 [15:36:33<54:15:28, 5.70s/it]
22%|██▏ | 9665/43939 [15:36:39<54:07:34, 5.69s/it]
22%|██▏ | 9666/43939 [15:36:45<54:19:22, 5.71s/it]
22%|██▏ | 9667/43939 [15:36:51<54:16:02, 5.70s/it]
22%|██▏ | 9668/43939 [15:36:56<54:21:32, 5.71s/it]
22%|██▏ | 9669/43939 [15:37:02<54:16:32, 5.70s/it]
22%|██▏ | 9670/43939 [15:37:08<54:17:20, 5.70s/it]
22%|██▏ | 9671/43939 [15:37:13<54:18:22, 5.71s/it]
22%|██▏ | 9672/43939 [15:37:19<54:27:41, 5.72s/it]
22%|██▏ | 9673/43939 [15:37:25<55:01:44, 5.78s/it]
22%|██▏ | 9674/43939 [15:37:31<54:46:41, 5.76s/it]
22%|██▏ | 9675/43939 [15:37:36<54:38:09, 5.74s/it]
22%|██▏ | 9676/43939 [15:37:42<54:26:45, 5.72s/it]
22%|██▏ | 9677/43939 [15:37:48<54:18:51, 5.71s/it]
22%|██▏ | 9678/43939 [15:37:54<54:24:54, 5.72s/it]
22%|██▏ | 9679/43939 [15:37:59<54:17:55, 5.71s/it]
22%|██▏ | 9680/43939 [15:38:05<54:32:18, 5.73s/it]
22%|██▏ | 9681/43939 [15:38:11<54:20:15, 5.71s/it]
22%|██▏ | 9682/43939 [15:38:16<54:16:20, 5.70s/it]
22%|██▏ | 9683/43939 [15:38:22<54:14:40, 5.70s/it]
22%|██▏ | 9684/43939 [15:38:28<54:14:15, 5.70s/it]
22%|██▏ | 9685/43939 [15:38:34<54:17:41, 5.71s/it]
22%|██▏ | 9686/43939 [15:38:39<54:16:12, 5.70s/it]
22%|██▏ | 9687/43939 [15:38:45<54:11:05, 5.70s/it]
22%|██▏ | 9688/43939 [15:38:51<54:03:49, 5.68s/it]
22%|██▏ | 9689/43939 [15:38:56<54:01:55, 5.68s/it]
22%|██▏ | 9690/43939 [15:39:02<54:13:12, 5.70s/it]
22%|██▏ | 9691/43939 [15:39:08<54:09:10, 5.69s/it]
22%|██▏ | 9692/43939 [15:39:13<54:12:11, 5.70s/it]
22%|██▏ | 9693/43939 [15:39:19<54:19:03, 5.71s/it]
22%|██▏ | 9694/43939 [15:39:25<54:17:25, 5.71s/it]
22%|██▏ | 9695/43939 [15:39:31<54:59:30, 5.78s/it]
22%|██▏ | 9696/43939 [15:39:36<54:40:22, 5.75s/it]
22%|██▏ | 9697/43939 [15:39:42<54:35:18, 5.74s/it]
22%|██▏ | 9698/43939 [15:39:48<54:30:41, 5.73s/it]
22%|██▏ | 9699/43939 [15:39:54<54:30:52, 5.73s/it]
22%|██▏ | 9700/43939 [15:39:59<54:29:42, 5.73s/it]
{'loss': 1.0196, 'grad_norm': 1.1553659439086914, 'learning_rate': 4.7811738276808415e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.05, 'epoch': 0.88}
22%|██▏ | 9700/43939 [15:39:59<54:29:42, 5.73s/it]
22%|██▏ | 9701/43939 [15:40:05<54:23:57, 5.72s/it]
22%|██▏ | 9702/43939 [15:40:11<54:18:57, 5.71s/it]
22%|██▏ | 9703/43939 [15:40:16<54:16:21, 5.71s/it]
22%|██▏ | 9704/43939 [15:40:22<54:16:56, 5.71s/it]
22%|██▏ | 9705/43939 [15:40:28<54:14:42, 5.70s/it]
22%|██▏ | 9706/43939 [15:40:34<54:16:07, 5.71s/it]
22%|██▏ | 9707/43939 [15:40:39<54:13:11, 5.70s/it]
22%|██▏ | 9708/43939 [15:40:45<54:18:52, 5.71s/it]
22%|██▏ | 9709/43939 [15:40:51<54:13:02, 5.70s/it]
22%|██▏ | 9710/43939 [15:40:56<54:17:14, 5.71s/it]
22%|██▏ | 9711/43939 [15:41:02<54:13:50, 5.70s/it]
22%|██▏ | 9712/43939 [15:41:08<54:19:41, 5.71s/it]
22%|██▏ | 9713/43939 [15:41:14<54:23:49, 5.72s/it]
22%|██▏ | 9714/43939 [15:41:19<54:11:51, 5.70s/it]
22%|██▏ | 9715/43939 [15:41:25<54:08:06, 5.69s/it]
22%|██▏ | 9716/43939 [15:41:31<54:55:43, 5.78s/it]
22%|██▏ | 9717/43939 [15:41:37<54:42:16, 5.75s/it]
22%|██▏ | 9718/43939 [15:41:42<54:30:36, 5.73s/it]
22%|██▏ | 9719/43939 [15:41:48<54:26:26, 5.73s/it]
22%|██▏ | 9720/43939 [15:41:54<54:22:12, 5.72s/it]
22%|██▏ | 9721/43939 [15:41:59<54:27:12, 5.73s/it]
22%|██▏ | 9722/43939 [15:42:05<54:20:40, 5.72s/it]
22%|██▏ | 9723/43939 [15:42:11<54:22:11, 5.72s/it]
22%|██▏ | 9724/43939 [15:42:16<54:19:42, 5.72s/it]
22%|██▏ | 9725/43939 [15:42:22<54:30:11, 5.73s/it]
22%|██▏ | 9726/43939 [15:42:28<54:19:28, 5.72s/it]
22%|██▏ | 9727/43939 [15:42:34<54:25:37, 5.73s/it]
22%|██▏ | 9728/43939 [15:42:39<54:24:48, 5.73s/it]
22%|██▏ | 9729/43939 [15:42:45<54:18:41, 5.72s/it]
22%|██▏ | 9730/43939 [15:42:51<54:14:44, 5.71s/it]
22%|██▏ | 9731/43939 [15:42:56<54:04:19, 5.69s/it]
22%|██▏ | 9732/43939 [15:43:02<54:14:06, 5.71s/it]
22%|██▏ | 9733/43939 [15:43:08<54:10:13, 5.70s/it]
22%|██▏ | 9734/43939 [15:43:14<54:11:02, 5.70s/it]
22%|██▏ | 9735/43939 [15:43:19<54:19:43, 5.72s/it]
22%|██▏ | 9736/43939 [15:43:25<55:04:13, 5.80s/it]
22%|██▏ | 9737/43939 [15:43:31<54:46:33, 5.77s/it]
22%|██▏ | 9738/43939 [15:43:37<54:38:46, 5.75s/it]
22%|██▏ | 9739/43939 [15:43:42<54:32:05, 5.74s/it]
22%|██▏ | 9740/43939 [15:43:48<54:28:24, 5.73s/it]
22%|██▏ | 9741/43939 [15:43:54<54:20:22, 5.72s/it]
22%|██▏ | 9742/43939 [15:44:00<54:13:33, 5.71s/it]
22%|██▏ | 9743/43939 [15:44:05<54:15:31, 5.71s/it]
22%|██▏ | 9744/43939 [15:44:11<54:16:43, 5.71s/it]
22%|██▏ | 9745/43939 [15:44:17<54:16:47, 5.71s/it]
22%|██▏ | 9746/43939 [15:44:22<54:25:54, 5.73s/it]
22%|██▏ | 9747/43939 [15:44:28<54:19:21, 5.72s/it]
22%|██▏ | 9748/43939 [15:44:34<54:11:58, 5.71s/it]
22%|██▏ | 9749/43939 [15:44:40<54:13:46, 5.71s/it]
22%|██▏ | 9750/43939 [15:44:45<54:16:47, 5.72s/it]
{'loss': 1.0324, 'grad_norm': 1.1080563068389893, 'learning_rate': 4.777092960883051e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.85, 'epoch': 0.89}
22%|██▏ | 9750/43939 [15:44:45<54:16:47, 5.72s/it]
22%|██▏ | 9751/43939 [15:44:51<54:17:27, 5.72s/it]
22%|██▏ | 9752/43939 [15:44:57<54:17:50, 5.72s/it]
22%|██▏ | 9753/43939 [15:45:02<54:18:11, 5.72s/it]
22%|██▏ | 9754/43939 [15:45:08<54:11:18, 5.71s/it]
22%|██▏ | 9755/43939 [15:45:14<54:08:32, 5.70s/it]
22%|██▏ | 9756/43939 [15:45:20<54:07:33, 5.70s/it]
22%|██▏ | 9757/43939 [15:45:25<54:09:24, 5.70s/it]
22%|██▏ | 9758/43939 [15:45:31<54:45:43, 5.77s/it]
22%|██▏ | 9759/43939 [15:45:37<54:37:22, 5.75s/it]
22%|██▏ | 9760/43939 [15:45:43<54:31:27, 5.74s/it]
22%|██▏ | 9761/43939 [15:45:48<54:24:00, 5.73s/it]
22%|██▏ | 9762/43939 [15:45:54<54:17:48, 5.72s/it]
22%|██▏ | 9763/43939 [15:46:00<54:16:54, 5.72s/it]
22%|██▏ | 9764/43939 [15:46:05<54:15:47, 5.72s/it]
22%|██▏ | 9765/43939 [15:46:11<54:10:10, 5.71s/it]
22%|██▏ | 9766/43939 [15:46:17<54:07:31, 5.70s/it]
22%|██▏ | 9767/43939 [15:46:22<54:11:05, 5.71s/it]
22%|██▏ | 9768/43939 [15:46:28<54:07:16, 5.70s/it]
22%|██▏ | 9769/43939 [15:46:34<54:11:11, 5.71s/it]
22%|██▏ | 9770/43939 [15:46:40<54:11:00, 5.71s/it]
22%|██▏ | 9771/43939 [15:46:45<54:11:48, 5.71s/it]
22%|██▏ | 9772/43939 [15:46:51<54:02:22, 5.69s/it]
22%|██▏ | 9773/43939 [15:46:57<54:04:55, 5.70s/it]
22%|██▏ | 9774/43939 [15:47:02<54:15:09, 5.72s/it]
22%|██▏ | 9775/43939 [15:47:08<54:12:22, 5.71s/it]
22%|██▏ | 9776/43939 [15:47:14<54:15:07, 5.72s/it]
22%|██▏ | 9777/43939 [15:47:20<54:18:06, 5.72s/it]
22%|██▏ | 9778/43939 [15:47:25<54:08:43, 5.71s/it]
22%|██▏ | 9779/43939 [15:47:31<54:45:44, 5.77s/it]
22%|██▏ | 9780/43939 [15:47:37<54:32:40, 5.75s/it]
22%|██▏ | 9781/43939 [15:47:43<54:24:38, 5.73s/it]
22%|██▏ | 9782/43939 [15:47:48<54:20:17, 5.73s/it]
22%|██▏ | 9783/43939 [15:47:54<54:24:50, 5.74s/it]
22%|██▏ | 9784/43939 [15:48:00<54:14:48, 5.72s/it]
22%|██▏ | 9785/43939 [15:48:05<54:15:31, 5.72s/it]
22%|██▏ | 9786/43939 [15:48:11<54:13:21, 5.72s/it]
22%|██▏ | 9787/43939 [15:48:17<54:03:25, 5.70s/it]
22%|██▏ | 9788/43939 [15:48:23<54:01:32, 5.70s/it]
22%|██▏ | 9789/43939 [15:48:28<54:16:44, 5.72s/it]
22%|██▏ | 9790/43939 [15:48:34<54:18:09, 5.72s/it]
22%|██▏ | 9791/43939 [15:48:40<54:07:46, 5.71s/it]
22%|██▏ | 9792/43939 [15:48:45<54:06:59, 5.71s/it]
22%|██▏ | 9793/43939 [15:48:51<54:05:32, 5.70s/it]
22%|██▏ | 9794/43939 [15:48:57<54:06:09, 5.70s/it]
22%|██▏ | 9795/43939 [15:49:03<54:07:51, 5.71s/it]
22%|██▏ | 9796/43939 [15:49:08<54:13:48, 5.72s/it]
22%|██▏ | 9797/43939 [15:49:14<54:09:50, 5.71s/it]
22%|██▏ | 9798/43939 [15:49:20<54:02:25, 5.70s/it]
22%|██▏ | 9799/43939 [15:49:25<54:11:39, 5.71s/it]
22%|██▏ | 9800/43939 [15:49:31<54:50:07, 5.78s/it]
{'loss': 1.04, 'grad_norm': 1.3158994913101196, 'learning_rate': 4.772976167592708e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.81, 'epoch': 0.89}
22%|██▏ | 9800/43939 [15:49:31<54:50:07, 5.78s/it]
22%|██▏ | 9801/43939 [15:49:37<54:49:11, 5.78s/it]
22%|██▏ | 9802/43939 [15:49:43<54:39:10, 5.76s/it]
22%|██▏ | 9803/43939 [15:49:49<54:29:08, 5.75s/it]
22%|██▏ | 9804/43939 [15:49:54<54:22:17, 5.73s/it]
22%|██▏ | 9805/43939 [15:50:00<54:12:07, 5.72s/it]
22%|██▏ | 9806/43939 [15:50:06<54:14:16, 5.72s/it]
22%|██▏ | 9807/43939 [15:50:11<54:05:05, 5.70s/it]
22%|██▏ | 9808/43939 [15:50:17<54:05:50, 5.71s/it]
22%|██▏ | 9809/43939 [15:50:23<54:14:41, 5.72s/it]
22%|██▏ | 9810/43939 [15:50:28<54:14:54, 5.72s/it]
22%|██▏ | 9811/43939 [15:50:34<54:04:35, 5.70s/it]
22%|██▏ | 9812/43939 [15:50:40<54:04:32, 5.70s/it]
22%|██▏ | 9813/43939 [15:50:46<54:06:43, 5.71s/it]
22%|██▏ | 9814/43939 [15:50:51<54:08:25, 5.71s/it]
22%|██▏ | 9815/43939 [15:50:57<54:00:46, 5.70s/it]
22%|██▏ | 9816/43939 [15:51:03<53:55:22, 5.69s/it]
22%|██▏ | 9817/43939 [15:51:08<53:56:36, 5.69s/it]
22%|██▏ | 9818/43939 [15:51:14<53:57:08, 5.69s/it]
22%|██▏ | 9819/43939 [15:51:20<53:57:35, 5.69s/it]
22%|██▏ | 9820/43939 [15:51:25<54:06:18, 5.71s/it]
22%|██▏ | 9821/43939 [15:51:31<53:58:23, 5.70s/it]
22%|██▏ | 9822/43939 [15:51:37<54:35:00, 5.76s/it]
22%|██▏ | 9823/43939 [15:51:43<54:24:40, 5.74s/it]
22%|██▏ | 9824/43939 [15:51:48<54:13:37, 5.72s/it]
22%|██▏ | 9825/43939 [15:51:54<54:26:12, 5.74s/it]
22%|██▏ | 9826/43939 [15:52:00<54:18:27, 5.73s/it]
22%|██▏ | 9827/43939 [15:52:06<54:14:58, 5.73s/it]
22%|██▏ | 9828/43939 [15:52:11<54:20:19, 5.73s/it]
22%|██▏ | 9829/43939 [15:52:17<54:16:00, 5.73s/it]
22%|██▏ | 9830/43939 [15:52:23<54:04:28, 5.71s/it]
22%|██▏ | 9831/43939 [15:52:28<54:07:58, 5.71s/it]
22%|██▏ | 9832/43939 [15:52:34<54:02:45, 5.70s/it]
22%|██▏ | 9833/43939 [15:52:40<54:04:50, 5.71s/it]
22%|██▏ | 9834/43939 [15:52:46<54:02:08, 5.70s/it]
22%|██▏ | 9835/43939 [15:52:51<54:08:40, 5.72s/it]
22%|██▏ | 9836/43939 [15:52:57<54:06:44, 5.71s/it]
22%|██▏ | 9837/43939 [15:53:03<54:03:00, 5.71s/it]
22%|██▏ | 9838/43939 [15:53:08<54:02:31, 5.71s/it]
22%|██▏ | 9839/43939 [15:53:14<53:54:43, 5.69s/it]
22%|██▏ | 9840/43939 [15:53:20<53:53:17, 5.69s/it]
22%|██▏ | 9841/43939 [15:53:25<53:54:07, 5.69s/it]
22%|██▏ | 9842/43939 [15:53:31<53:56:59, 5.70s/it]
22%|██▏ | 9843/43939 [15:53:37<54:32:34, 5.76s/it]
22%|██▏ | 9844/43939 [15:53:43<54:23:47, 5.74s/it]
22%|██▏ | 9845/43939 [15:53:48<54:18:00, 5.73s/it]
22%|██▏ | 9846/43939 [15:53:54<54:15:35, 5.73s/it]
22%|██▏ | 9847/43939 [15:54:00<54:06:10, 5.71s/it]
22%|██▏ | 9848/43939 [15:54:06<54:07:41, 5.72s/it]
22%|██▏ | 9849/43939 [15:54:11<54:07:03, 5.71s/it]
22%|██▏ | 9850/43939 [15:54:17<54:14:47, 5.73s/it]
{'loss': 1.0311, 'grad_norm': 1.4326947927474976, 'learning_rate': 4.768823512761901e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.24, 'epoch': 0.9}
22%|██▏ | 9850/43939 [15:54:17<54:14:47, 5.73s/it]
22%|██▏ | 9851/43939 [15:54:23<54:08:08, 5.72s/it]
22%|██▏ | 9852/43939 [15:54:28<53:55:52, 5.70s/it]
22%|██▏ | 9853/43939 [15:54:34<53:55:52, 5.70s/it]
22%|██▏ | 9854/43939 [15:54:40<54:06:42, 5.72s/it]
22%|██▏ | 9855/43939 [15:54:46<54:06:17, 5.71s/it]
22%|██▏ | 9856/43939 [15:54:51<54:01:00, 5.71s/it]
22%|██▏ | 9857/43939 [15:54:57<53:57:52, 5.70s/it]
22%|██▏ | 9858/43939 [15:55:03<53:56:39, 5.70s/it]
22%|██▏ | 9859/43939 [15:55:08<53:59:35, 5.70s/it]
22%|██▏ | 9860/43939 [15:55:14<53:54:01, 5.69s/it]
22%|██▏ | 9861/43939 [15:55:20<53:56:06, 5.70s/it]
22%|██▏ | 9862/43939 [15:55:26<54:07:12, 5.72s/it]
22%|██▏ | 9863/43939 [15:55:31<54:10:43, 5.72s/it]
22%|██▏ | 9864/43939 [15:55:37<54:35:29, 5.77s/it]
22%|██▏ | 9865/43939 [15:55:43<54:21:56, 5.74s/it]
22%|██▏ | 9866/43939 [15:55:48<54:10:46, 5.72s/it]
22%|██▏ | 9867/43939 [15:55:54<54:05:55, 5.72s/it]
22%|██▏ | 9868/43939 [15:56:00<54:02:30, 5.71s/it]
22%|██▏ | 9869/43939 [15:56:06<53:59:59, 5.71s/it]
22%|██▏ | 9870/43939 [15:56:11<54:01:26, 5.71s/it]
22%|██▏ | 9871/43939 [15:56:17<54:00:18, 5.71s/it]
22%|██▏ | 9872/43939 [15:56:23<53:57:52, 5.70s/it]
22%|██▏ | 9873/43939 [15:56:28<53:56:05, 5.70s/it]
22%|██▏ | 9874/43939 [15:56:34<53:56:12, 5.70s/it]
22%|██▏ | 9875/43939 [15:56:40<53:49:58, 5.69s/it]
22%|██▏ | 9876/43939 [15:56:45<53:57:21, 5.70s/it]
22%|██▏ | 9877/43939 [15:56:51<54:07:31, 5.72s/it]
22%|██▏ | 9878/43939 [15:56:57<54:11:47, 5.73s/it]
22%|██▏ | 9879/43939 [15:57:03<54:08:14, 5.72s/it]
22%|██▏ | 9880/43939 [15:57:08<54:04:47, 5.72s/it]
22%|██▏ | 9881/43939 [15:57:14<54:01:32, 5.71s/it]
22%|██▏ | 9882/43939 [15:57:20<54:00:45, 5.71s/it]
22%|██▏ | 9883/43939 [15:57:25<54:00:02, 5.71s/it]
22%|██▏ | 9884/43939 [15:57:31<53:58:45, 5.71s/it]
22%|██▏ | 9885/43939 [15:57:37<54:33:41, 5.77s/it]
22%|██▏ | 9886/43939 [15:57:43<54:25:25, 5.75s/it]
23%|██▎ | 9887/43939 [15:57:49<54:16:31, 5.74s/it]
23%|██▎ | 9888/43939 [15:57:54<54:10:07, 5.73s/it]
23%|██▎ | 9889/43939 [15:58:00<54:00:35, 5.71s/it]
23%|██▎ | 9890/43939 [15:58:06<54:00:14, 5.71s/it]
23%|██▎ | 9891/43939 [15:58:11<53:59:48, 5.71s/it]
23%|██▎ | 9892/43939 [15:58:17<53:57:52, 5.71s/it]
23%|██▎ | 9893/43939 [15:58:23<53:56:27, 5.70s/it]
23%|██▎ | 9894/43939 [15:58:28<53:54:48, 5.70s/it]
23%|██▎ | 9895/43939 [15:58:34<53:50:23, 5.69s/it]
23%|██▎ | 9896/43939 [15:58:40<53:56:10, 5.70s/it]
23%|██▎ | 9897/43939 [15:58:46<54:02:01, 5.71s/it]
23%|██▎ | 9898/43939 [15:58:51<54:04:58, 5.72s/it]
23%|██▎ | 9899/43939 [15:58:57<54:06:56, 5.72s/it]
23%|██▎ | 9900/43939 [15:59:03<54:07:04, 5.72s/it]
{'loss': 1.0129, 'grad_norm': 1.200109839439392, 'learning_rate': 4.764635061908517e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.06, 'epoch': 0.9}
23%|██▎ | 9900/43939 [15:59:03<54:07:04, 5.72s/it]
23%|██▎ | 9901/43939 [15:59:08<54:01:52, 5.71s/it]
23%|██▎ | 9902/43939 [15:59:14<53:59:43, 5.71s/it]
23%|██▎ | 9903/43939 [15:59:20<54:07:00, 5.72s/it]
23%|██▎ | 9904/43939 [15:59:26<54:01:26, 5.71s/it]
23%|██▎ | 9905/43939 [15:59:31<53:57:08, 5.71s/it]
23%|██▎ | 9906/43939 [15:59:37<54:28:27, 5.76s/it]
23%|██▎ | 9907/43939 [15:59:43<54:13:55, 5.74s/it]
23%|██▎ | 9908/43939 [15:59:49<54:11:26, 5.73s/it]
23%|██▎ | 9909/43939 [15:59:54<54:13:23, 5.74s/it]
23%|██▎ | 9910/43939 [16:00:00<54:01:47, 5.72s/it]
23%|██▎ | 9911/43939 [16:00:06<54:00:09, 5.71s/it]
23%|██▎ | 9912/43939 [16:00:11<53:57:41, 5.71s/it]
23%|██▎ | 9913/43939 [16:00:17<53:54:22, 5.70s/it]
23%|██▎ | 9914/43939 [16:00:23<53:48:22, 5.69s/it]
23%|██▎ | 9915/43939 [16:00:28<53:49:45, 5.70s/it]
23%|██▎ | 9916/43939 [16:00:34<53:55:59, 5.71s/it]
23%|██▎ | 9917/43939 [16:00:40<54:00:11, 5.71s/it]
23%|██▎ | 9918/43939 [16:00:46<54:00:51, 5.72s/it]
23%|██▎ | 9919/43939 [16:00:51<53:54:28, 5.70s/it]
23%|██▎ | 9920/43939 [16:00:57<53:51:51, 5.70s/it]
23%|██▎ | 9921/43939 [16:01:03<53:46:34, 5.69s/it]
23%|██▎ | 9922/43939 [16:01:08<53:55:09, 5.71s/it]
23%|██▎ | 9923/43939 [16:01:14<53:51:09, 5.70s/it]
23%|██▎ | 9924/43939 [16:01:20<53:51:19, 5.70s/it]
23%|██▎ | 9925/43939 [16:01:25<53:51:09, 5.70s/it]
23%|██▎ | 9926/43939 [16:01:31<54:38:16, 5.78s/it]
23%|██▎ | 9927/43939 [16:01:37<54:24:43, 5.76s/it]
23%|██▎ | 9928/43939 [16:01:43<54:11:08, 5.74s/it]
23%|██▎ | 9929/43939 [16:01:49<54:01:09, 5.72s/it]
23%|██▎ | 9930/43939 [16:01:54<53:57:00, 5.71s/it]
23%|██▎ | 9931/43939 [16:02:00<53:48:06, 5.70s/it]
23%|██▎ | 9932/43939 [16:02:06<53:47:00, 5.69s/it]
23%|██▎ | 9933/43939 [16:02:11<53:44:06, 5.69s/it]
23%|██▎ | 9934/43939 [16:02:17<53:42:35, 5.69s/it]
23%|██▎ | 9935/43939 [16:02:23<53:46:00, 5.69s/it]
23%|██▎ | 9936/43939 [16:02:28<53:45:14, 5.69s/it]
23%|██▎ | 9937/43939 [16:02:34<53:42:45, 5.69s/it]
23%|██▎ | 9938/43939 [16:02:40<53:43:42, 5.69s/it]
23%|██▎ | 9939/43939 [16:02:45<53:38:55, 5.68s/it]
23%|██▎ | 9940/43939 [16:02:51<53:37:14, 5.68s/it]
23%|██▎ | 9941/43939 [16:02:57<53:37:20, 5.68s/it]
23%|██▎ | 9942/43939 [16:03:02<53:41:05, 5.68s/it]
23%|██▎ | 9943/43939 [16:03:08<53:37:24, 5.68s/it]
23%|██▎ | 9944/43939 [16:03:14<53:44:32, 5.69s/it]
23%|██▎ | 9945/43939 [16:03:20<53:50:57, 5.70s/it]
23%|██▎ | 9946/43939 [16:03:25<53:58:10, 5.72s/it]
23%|██▎ | 9947/43939 [16:03:31<54:32:59, 5.78s/it]
23%|██▎ | 9948/43939 [16:03:37<54:14:42, 5.75s/it]
23%|██▎ | 9949/43939 [16:03:43<54:03:24, 5.73s/it]
23%|██▎ | 9950/43939 [16:03:48<54:01:47, 5.72s/it]
{'loss': 1.0137, 'grad_norm': 1.0471111536026, 'learning_rate': 4.7604108811152126e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.57, 'epoch': 0.91}
23%|██▎ | 9950/43939 [16:03:48<54:01:47, 5.72s/it]
23%|██▎ | 9951/43939 [16:03:54<54:04:40, 5.73s/it]
23%|██▎ | 9952/43939 [16:04:00<53:56:09, 5.71s/it]
23%|██▎ | 9953/43939 [16:04:05<53:46:40, 5.70s/it]
23%|██▎ | 9954/43939 [16:04:11<53:48:10, 5.70s/it]
23%|██▎ | 9955/43939 [16:04:17<53:50:04, 5.70s/it]
23%|██▎ | 9956/43939 [16:04:22<53:51:47, 5.71s/it]
23%|██▎ | 9957/43939 [16:04:28<53:49:05, 5.70s/it]
23%|██▎ | 9958/43939 [16:04:34<53:47:28, 5.70s/it]
23%|██▎ | 9959/43939 [16:04:40<53:50:51, 5.70s/it]
23%|██▎ | 9960/43939 [16:04:45<53:49:08, 5.70s/it]
23%|██▎ | 9961/43939 [16:04:51<53:45:59, 5.70s/it]
23%|██▎ | 9962/43939 [16:04:57<53:47:52, 5.70s/it]
23%|██▎ | 9963/43939 [16:05:02<53:53:46, 5.71s/it]
23%|██▎ | 9964/43939 [16:05:08<53:52:59, 5.71s/it]
23%|██▎ | 9965/43939 [16:05:14<54:01:00, 5.72s/it]
23%|██▎ | 9966/43939 [16:05:20<53:58:38, 5.72s/it]
23%|██▎ | 9967/43939 [16:05:25<53:57:43, 5.72s/it]
23%|██▎ | 9968/43939 [16:05:31<53:51:28, 5.71s/it]
23%|██▎ | 9969/43939 [16:05:37<54:23:40, 5.76s/it]
23%|██▎ | 9970/43939 [16:05:43<54:13:43, 5.75s/it]
23%|██▎ | 9971/43939 [16:05:48<54:03:12, 5.73s/it]
23%|██▎ | 9972/43939 [16:05:54<53:57:59, 5.72s/it]
23%|██▎ | 9973/43939 [16:06:00<54:04:48, 5.73s/it]
23%|██▎ | 9974/43939 [16:06:05<54:02:45, 5.73s/it]
23%|██▎ | 9975/43939 [16:06:11<53:58:44, 5.72s/it]
23%|██▎ | 9976/43939 [16:06:17<53:54:17, 5.71s/it]
23%|██▎ | 9977/43939 [16:06:23<53:53:47, 5.71s/it]
23%|██▎ | 9978/43939 [16:06:28<53:57:58, 5.72s/it]
23%|██▎ | 9979/43939 [16:06:34<53:56:16, 5.72s/it]
23%|██▎ | 9980/43939 [16:06:40<53:53:02, 5.71s/it]
23%|██▎ | 9981/43939 [16:06:45<53:44:29, 5.70s/it]
23%|██▎ | 9982/43939 [16:06:51<53:48:15, 5.70s/it]
23%|██▎ | 9983/43939 [16:06:57<53:51:34, 5.71s/it]
23%|██▎ | 9984/43939 [16:07:02<53:46:29, 5.70s/it]
23%|██▎ | 9985/43939 [16:07:08<53:45:32, 5.70s/it]
23%|██▎ | 9986/43939 [16:07:14<53:47:30, 5.70s/it]
23%|██▎ | 9987/43939 [16:07:20<53:46:07, 5.70s/it]
23%|██▎ | 9988/43939 [16:07:25<53:36:35, 5.68s/it]
23%|██▎ | 9989/43939 [16:07:31<53:31:13, 5.68s/it]
23%|██▎ | 9990/43939 [16:07:37<54:06:16, 5.74s/it]
23%|██▎ | 9991/43939 [16:07:42<54:03:21, 5.73s/it]
23%|██▎ | 9992/43939 [16:07:48<54:02:25, 5.73s/it]
23%|██▎ | 9993/43939 [16:07:54<53:52:33, 5.71s/it]
23%|██▎ | 9994/43939 [16:08:00<53:48:05, 5.71s/it]
23%|██▎ | 9995/43939 [16:08:05<53:48:49, 5.71s/it]
23%|██▎ | 9996/43939 [16:08:11<53:55:17, 5.72s/it]
23%|██▎ | 9997/43939 [16:08:17<53:55:44, 5.72s/it]
23%|██▎ | 9998/43939 [16:08:22<53:51:37, 5.71s/it]
23%|██▎ | 9999/43939 [16:08:28<53:53:42, 5.72s/it]
23%|██▎ | 10000/43939 [16:08:34<53:57:08, 5.72s/it]
{'loss': 1.0504, 'grad_norm': 1.168687343597412, 'learning_rate': 4.7561510370283654e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.16, 'epoch': 0.91}
23%|██▎ | 10000/43939 [16:08:34<53:57:08, 5.72s/it]
23%|██▎ | 10001/43939 [16:08:40<53:54:37, 5.72s/it]
23%|██▎ | 10002/43939 [16:08:45<53:50:10, 5.71s/it]
23%|██▎ | 10003/43939 [16:08:51<53:46:54, 5.71s/it]
23%|██▎ | 10004/43939 [16:08:57<53:41:31, 5.70s/it]
23%|██▎ | 10005/43939 [16:09:02<53:38:06, 5.69s/it]
23%|██▎ | 10006/43939 [16:09:08<53:36:31, 5.69s/it]
23%|██▎ | 10007/43939 [16:09:14<53:41:08, 5.70s/it]
23%|██▎ | 10008/43939 [16:09:19<53:36:52, 5.69s/it]
23%|██▎ | 10009/43939 [16:09:25<53:50:24, 5.71s/it]
23%|██▎ | 10010/43939 [16:09:31<53:47:39, 5.71s/it]
23%|██▎ | 10011/43939 [16:09:37<54:26:45, 5.78s/it]
23%|██▎ | 10012/43939 [16:09:43<54:23:28, 5.77s/it]
23%|██▎ | 10013/43939 [16:09:48<54:09:48, 5.75s/it]
23%|██▎ | 10014/43939 [16:09:54<54:00:35, 5.73s/it]
23%|██▎ | 10015/43939 [16:10:00<53:46:43, 5.71s/it]
23%|██▎ | 10016/43939 [16:10:05<53:36:27, 5.69s/it]
23%|██▎ | 10017/43939 [16:10:11<53:38:27, 5.69s/it]
23%|██▎ | 10018/43939 [16:10:17<53:34:05, 5.69s/it]
23%|██▎ | 10019/43939 [16:10:22<53:31:46, 5.68s/it]
23%|██▎ | 10020/43939 [16:10:28<53:27:55, 5.67s/it]
23%|██▎ | 10021/43939 [16:10:34<53:30:17, 5.68s/it]
23%|██▎ | 10022/43939 [16:10:39<53:34:03, 5.69s/it]
23%|██▎ | 10023/43939 [16:10:45<53:31:37, 5.68s/it]
23%|██▎ | 10024/43939 [16:10:51<53:42:40, 5.70s/it]
23%|██▎ | 10025/43939 [16:10:56<53:41:07, 5.70s/it]
23%|██▎ | 10026/43939 [16:11:02<53:36:48, 5.69s/it]
23%|██▎ | 10027/43939 [16:11:08<53:28:42, 5.68s/it]
23%|██▎ | 10028/43939 [16:11:13<53:33:18, 5.69s/it]
23%|██▎ | 10029/43939 [16:11:19<53:38:28, 5.69s/it]
23%|██▎ | 10030/43939 [16:11:25<53:45:00, 5.71s/it]
23%|██▎ | 10031/43939 [16:11:31<53:42:15, 5.70s/it]
23%|██▎ | 10032/43939 [16:11:37<54:20:49, 5.77s/it]
23%|██▎ | 10033/43939 [16:11:42<54:16:23, 5.76s/it]
23%|██▎ | 10034/43939 [16:11:48<54:12:23, 5.76s/it]
23%|██▎ | 10035/43939 [16:11:54<53:58:26, 5.73s/it]
23%|██▎ | 10036/43939 [16:11:59<54:00:51, 5.74s/it]
23%|██▎ | 10037/43939 [16:12:05<53:49:49, 5.72s/it]
23%|██▎ | 10038/43939 [16:12:11<53:38:11, 5.70s/it]
23%|██▎ | 10039/43939 [16:12:16<53:39:38, 5.70s/it]
23%|██▎ | 10040/43939 [16:12:22<53:36:53, 5.69s/it]
23%|██▎ | 10041/43939 [16:12:28<53:39:01, 5.70s/it]
23%|██▎ | 10042/43939 [16:12:34<53:36:03, 5.69s/it]
23%|██▎ | 10043/43939 [16:12:39<53:45:08, 5.71s/it]
23%|██▎ | 10044/43939 [16:12:45<53:47:18, 5.71s/it]
23%|██▎ | 10045/43939 [16:12:51<53:47:25, 5.71s/it]
23%|██▎ | 10046/43939 [16:12:56<53:51:11, 5.72s/it]
23%|██▎ | 10047/43939 [16:13:02<53:52:24, 5.72s/it]
23%|██▎ | 10048/43939 [16:13:08<53:46:15, 5.71s/it]
23%|██▎ | 10049/43939 [16:13:14<53:33:50, 5.69s/it]
23%|██▎ | 10050/43939 [16:13:19<53:34:43, 5.69s/it]
{'loss': 1.0268, 'grad_norm': 1.0843098163604736, 'learning_rate': 4.7518555968570276e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.03, 'epoch': 0.91}
23%|██▎ | 10050/43939 [16:13:19<53:34:43, 5.69s/it]
23%|██▎ | 10051/43939 [16:13:25<53:36:08, 5.69s/it]
23%|██▎ | 10052/43939 [16:13:31<53:40:31, 5.70s/it]
23%|██▎ | 10053/43939 [16:13:37<54:22:16, 5.78s/it]
23%|██▎ | 10054/43939 [16:13:42<54:14:09, 5.76s/it]
23%|██▎ | 10055/43939 [16:13:48<54:05:12, 5.75s/it]
23%|██▎ | 10056/43939 [16:13:54<53:59:41, 5.74s/it]
23%|██▎ | 10057/43939 [16:13:59<53:50:32, 5.72s/it]
23%|██▎ | 10058/43939 [16:14:05<53:49:25, 5.72s/it]
23%|██▎ | 10059/43939 [16:14:11<53:48:07, 5.72s/it]
23%|██▎ | 10060/43939 [16:14:17<53:45:14, 5.71s/it]
23%|██▎ | 10061/43939 [16:14:22<53:41:39, 5.71s/it]
23%|██▎ | 10062/43939 [16:14:28<53:39:37, 5.70s/it]
23%|██▎ | 10063/43939 [16:14:34<53:39:59, 5.70s/it]
23%|██▎ | 10064/43939 [16:14:39<53:44:51, 5.71s/it]
23%|██▎ | 10065/43939 [16:14:45<53:37:42, 5.70s/it]
23%|██▎ | 10066/43939 [16:14:51<53:42:15, 5.71s/it]
23%|██▎ | 10067/43939 [16:14:56<53:40:25, 5.70s/it]
23%|██▎ | 10068/43939 [16:15:02<53:39:08, 5.70s/it]
23%|██▎ | 10069/43939 [16:15:08<53:37:55, 5.70s/it]
23%|██▎ | 10070/43939 [16:15:14<53:38:44, 5.70s/it]
23%|██▎ | 10071/43939 [16:15:19<53:32:03, 5.69s/it]
23%|██▎ | 10072/43939 [16:15:25<53:34:52, 5.70s/it]
23%|██▎ | 10073/43939 [16:15:31<54:12:25, 5.76s/it]
23%|██▎ | 10074/43939 [16:15:37<54:07:08, 5.75s/it]
23%|██▎ | 10075/43939 [16:15:42<53:57:26, 5.74s/it]
23%|██▎ | 10076/43939 [16:15:48<53:50:10, 5.72s/it]
23%|██▎ | 10077/43939 [16:15:54<53:45:52, 5.72s/it]
23%|██▎ | 10078/43939 [16:15:59<53:32:54, 5.69s/it]
23%|██▎ | 10079/43939 [16:16:05<53:35:17, 5.70s/it]
23%|██▎ | 10080/43939 [16:16:11<53:30:59, 5.69s/it]
23%|██▎ | 10081/43939 [16:16:16<53:29:54, 5.69s/it]
23%|██▎ | 10082/43939 [16:16:22<53:28:52, 5.69s/it]
23%|██▎ | 10083/43939 [16:16:28<53:33:37, 5.70s/it]
23%|██▎ | 10084/43939 [16:16:33<53:31:46, 5.69s/it]
23%|██▎ | 10085/43939 [16:16:39<53:37:03, 5.70s/it]
23%|██▎ | 10086/43939 [16:16:45<53:39:10, 5.71s/it]
23%|██▎ | 10087/43939 [16:16:51<53:35:07, 5.70s/it]
23%|██▎ | 10088/43939 [16:16:56<53:31:50, 5.69s/it]
23%|██▎ | 10089/43939 [16:17:02<53:26:47, 5.68s/it]
23%|██▎ | 10090/43939 [16:17:08<53:34:19, 5.70s/it]
23%|██▎ | 10091/43939 [16:17:13<53:37:10, 5.70s/it]
23%|██▎ | 10092/43939 [16:17:19<53:35:06, 5.70s/it]
23%|██▎ | 10093/43939 [16:17:25<53:39:13, 5.71s/it]
23%|██▎ | 10094/43939 [16:17:31<54:11:32, 5.76s/it]
23%|██▎ | 10095/43939 [16:17:36<53:57:29, 5.74s/it]
23%|██▎ | 10096/43939 [16:17:42<53:49:01, 5.72s/it]
23%|██▎ | 10097/43939 [16:17:48<53:46:22, 5.72s/it]
23%|██▎ | 10098/43939 [16:17:53<53:39:28, 5.71s/it]
23%|██▎ | 10099/43939 [16:17:59<53:32:46, 5.70s/it]
23%|██▎ | 10100/43939 [16:18:05<53:35:42, 5.70s/it]
{'loss': 1.0117, 'grad_norm': 1.1069607734680176, 'learning_rate': 4.7475246283718614e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.0, 'epoch': 0.92}
23%|██▎ | 10100/43939 [16:18:05<53:35:42, 5.70s/it]
23%|██▎ | 10101/43939 [16:18:11<53:33:33, 5.70s/it]
23%|██▎ | 10102/43939 [16:18:16<53:34:31, 5.70s/it]
23%|██▎ | 10103/43939 [16:18:22<53:41:24, 5.71s/it]
23%|██▎ | 10104/43939 [16:18:28<53:43:08, 5.72s/it]
23%|██▎ | 10105/43939 [16:18:33<53:52:37, 5.73s/it]
23%|██▎ | 10106/43939 [16:18:39<53:45:12, 5.72s/it]
23%|██▎ | 10107/43939 [16:18:45<53:41:48, 5.71s/it]
23%|██▎ | 10108/43939 [16:18:51<53:41:28, 5.71s/it]
23%|██▎ | 10109/43939 [16:18:56<53:33:14, 5.70s/it]
23%|██▎ | 10110/43939 [16:19:02<53:33:47, 5.70s/it]
23%|██▎ | 10111/43939 [16:19:08<53:32:29, 5.70s/it]
23%|██▎ | 10112/43939 [16:19:13<53:25:15, 5.69s/it]
23%|██▎ | 10113/43939 [16:19:19<53:24:50, 5.68s/it]
23%|██▎ | 10114/43939 [16:19:25<53:30:00, 5.69s/it]
23%|██▎ | 10115/43939 [16:19:30<53:31:36, 5.70s/it]
23%|██▎ | 10116/43939 [16:19:36<54:07:15, 5.76s/it]
23%|██▎ | 10117/43939 [16:19:42<53:52:09, 5.73s/it]
23%|██▎ | 10118/43939 [16:19:48<53:43:39, 5.72s/it]
23%|██▎ | 10119/43939 [16:19:53<53:42:17, 5.72s/it]
23%|██▎ | 10120/43939 [16:19:59<53:38:58, 5.71s/it]
23%|██▎ | 10121/43939 [16:20:05<53:42:40, 5.72s/it]
23%|██▎ | 10122/43939 [16:20:11<53:43:02, 5.72s/it]
23%|██▎ | 10123/43939 [16:20:16<53:40:52, 5.71s/it]
23%|██▎ | 10124/43939 [16:20:22<53:40:19, 5.71s/it]
23%|██▎ | 10125/43939 [16:20:28<53:33:38, 5.70s/it]
23%|██▎ | 10126/43939 [16:20:33<53:42:09, 5.72s/it]
23%|██▎ | 10127/43939 [16:20:39<53:41:32, 5.72s/it]
23%|██▎ | 10128/43939 [16:20:45<53:32:03, 5.70s/it]
23%|██▎ | 10129/43939 [16:20:50<53:33:05, 5.70s/it]
23%|██▎ | 10130/43939 [16:20:56<53:38:56, 5.71s/it]
23%|██▎ | 10131/43939 [16:21:02<53:36:08, 5.71s/it]
23%|██▎ | 10132/43939 [16:21:08<53:37:36, 5.71s/it]
23%|██▎ | 10133/43939 [16:21:13<53:37:07, 5.71s/it]
23%|██▎ | 10134/43939 [16:21:19<53:39:03, 5.71s/it]
23%|██▎ | 10135/43939 [16:21:25<53:37:52, 5.71s/it]
23%|██▎ | 10136/43939 [16:21:31<54:04:45, 5.76s/it]
23%|██▎ | 10137/43939 [16:21:36<53:54:57, 5.74s/it]
23%|██▎ | 10138/43939 [16:21:42<53:53:21, 5.74s/it]
23%|██▎ | 10139/43939 [16:21:48<53:46:57, 5.73s/it]
23%|██▎ | 10140/43939 [16:21:53<53:45:39, 5.73s/it]
23%|██▎ | 10141/43939 [16:21:59<53:37:31, 5.71s/it]
23%|██▎ | 10142/43939 [16:22:05<53:39:14, 5.72s/it]
23%|██▎ | 10143/43939 [16:22:11<53:32:48, 5.70s/it]
23%|██▎ | 10144/43939 [16:22:16<53:31:15, 5.70s/it]
23%|██▎ | 10145/43939 [16:22:22<53:25:03, 5.69s/it]
23%|██▎ | 10146/43939 [16:22:28<53:31:41, 5.70s/it]
23%|██▎ | 10147/43939 [16:22:33<53:26:37, 5.69s/it]
23%|██▎ | 10148/43939 [16:22:39<53:27:43, 5.70s/it]
23%|██▎ | 10149/43939 [16:22:45<53:28:13, 5.70s/it]
23%|██▎ | 10150/43939 [16:22:50<53:22:15, 5.69s/it]
{'loss': 1.0196, 'grad_norm': 1.1321187019348145, 'learning_rate': 4.743158199904072e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.42, 'epoch': 0.92}
23%|██▎ | 10150/43939 [16:22:50<53:22:15, 5.69s/it]
23%|██▎ | 10151/43939 [16:22:56<53:20:53, 5.68s/it]
23%|██▎ | 10152/43939 [16:23:02<53:20:57, 5.68s/it]
23%|██▎ | 10153/43939 [16:23:07<53:23:29, 5.69s/it]
23%|██▎ | 10154/43939 [16:23:13<53:29:09, 5.70s/it]
23%|██▎ | 10155/43939 [16:23:19<53:23:35, 5.69s/it]
23%|██▎ | 10156/43939 [16:23:24<53:17:06, 5.68s/it]
23%|██▎ | 10157/43939 [16:23:30<53:50:18, 5.74s/it]
23%|██▎ | 10158/43939 [16:23:36<53:44:12, 5.73s/it]
23%|██▎ | 10159/43939 [16:23:42<53:41:10, 5.72s/it]
23%|██▎ | 10160/43939 [16:23:47<53:39:33, 5.72s/it]
23%|██▎ | 10161/43939 [16:23:53<53:42:28, 5.72s/it]
23%|██▎ | 10162/43939 [16:23:59<53:37:48, 5.72s/it]
23%|██▎ | 10163/43939 [16:24:05<53:35:03, 5.71s/it]
23%|██▎ | 10164/43939 [16:24:10<53:27:04, 5.70s/it]
23%|██▎ | 10165/43939 [16:24:16<53:27:02, 5.70s/it]
23%|██▎ | 10166/43939 [16:24:22<53:31:49, 5.71s/it]
23%|██▎ | 10167/43939 [16:24:27<53:33:59, 5.71s/it]
23%|██▎ | 10168/43939 [16:24:33<53:33:14, 5.71s/it]
23%|██▎ | 10169/43939 [16:24:39<53:36:38, 5.72s/it]
23%|██▎ | 10170/43939 [16:24:45<53:31:54, 5.71s/it]
23%|██▎ | 10171/43939 [16:24:50<53:32:14, 5.71s/it]
23%|██▎ | 10172/43939 [16:24:56<53:30:46, 5.71s/it]
23%|██▎ | 10173/43939 [16:25:02<53:39:48, 5.72s/it]
23%|██▎ | 10174/43939 [16:25:07<53:32:26, 5.71s/it]
23%|██▎ | 10175/43939 [16:25:13<53:26:38, 5.70s/it]
23%|██▎ | 10176/43939 [16:25:19<53:34:34, 5.71s/it]
23%|██▎ | 10177/43939 [16:25:24<53:27:38, 5.70s/it]
23%|██▎ | 10178/43939 [16:25:30<53:30:14, 5.71s/it]
23%|██▎ | 10179/43939 [16:25:36<54:01:40, 5.76s/it]
23%|██▎ | 10180/43939 [16:25:42<53:49:35, 5.74s/it]
23%|██▎ | 10181/43939 [16:25:48<53:53:43, 5.75s/it]
23%|██▎ | 10182/43939 [16:25:53<53:48:09, 5.74s/it]
23%|██▎ | 10183/43939 [16:25:59<53:46:13, 5.73s/it]
23%|██▎ | 10184/43939 [16:26:05<53:35:10, 5.72s/it]
23%|██▎ | 10185/43939 [16:26:10<53:31:45, 5.71s/it]
23%|██▎ | 10186/43939 [16:26:16<53:22:05, 5.69s/it]
23%|██▎ | 10187/43939 [16:26:22<53:13:00, 5.68s/it]
23%|██▎ | 10188/43939 [16:26:27<53:21:05, 5.69s/it]
23%|██▎ | 10189/43939 [16:26:33<53:18:55, 5.69s/it]
23%|██▎ | 10190/43939 [16:26:39<53:23:04, 5.69s/it]
23%|██▎ | 10191/43939 [16:26:44<53:26:13, 5.70s/it]
23%|██▎ | 10192/43939 [16:26:50<53:26:13, 5.70s/it]
23%|██▎ | 10193/43939 [16:26:56<53:27:41, 5.70s/it]
23%|██▎ | 10194/43939 [16:27:02<53:20:49, 5.69s/it]
23%|██▎ | 10195/43939 [16:27:07<53:25:00, 5.70s/it]
23%|██▎ | 10196/43939 [16:27:13<53:33:07, 5.71s/it]
23%|██▎ | 10197/43939 [16:27:19<53:26:35, 5.70s/it]
23%|██▎ | 10198/43939 [16:27:24<53:25:09, 5.70s/it]
23%|██▎ | 10199/43939 [16:27:30<53:19:36, 5.69s/it]
23%|██▎ | 10200/43939 [16:27:36<54:05:22, 5.77s/it]
{'loss': 1.0377, 'grad_norm': 1.0236319303512573, 'learning_rate': 4.73875638034433e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.49, 'epoch': 0.93}
23%|██▎ | 10200/43939 [16:27:36<54:05:22, 5.77s/it]
23%|██▎ | 10201/43939 [16:27:42<53:52:40, 5.75s/it]
23%|██▎ | 10202/43939 [16:27:47<53:47:49, 5.74s/it]
23%|██▎ | 10203/43939 [16:27:53<53:43:18, 5.73s/it]
23%|██▎ | 10204/43939 [16:27:59<53:34:29, 5.72s/it]
23%|██▎ | 10205/43939 [16:28:05<53:28:34, 5.71s/it]
23%|██▎ | 10206/43939 [16:28:10<53:29:06, 5.71s/it]
23%|██▎ | 10207/43939 [16:28:16<53:22:38, 5.70s/it]
23%|██▎ | 10208/43939 [16:28:22<53:23:45, 5.70s/it]
23%|██▎ | 10209/43939 [16:28:27<53:19:49, 5.69s/it]
23%|██▎ | 10210/43939 [16:28:33<53:24:38, 5.70s/it]
23%|██▎ | 10211/43939 [16:28:39<53:35:45, 5.72s/it]
23%|██▎ | 10212/43939 [16:28:44<53:31:52, 5.71s/it]
23%|██▎ | 10213/43939 [16:28:50<53:26:23, 5.70s/it]
23%|██▎ | 10214/43939 [16:28:56<53:38:19, 5.73s/it]
23%|██▎ | 10215/43939 [16:29:02<53:31:22, 5.71s/it]
23%|██▎ | 10216/43939 [16:29:07<53:35:34, 5.72s/it]
23%|██▎ | 10217/43939 [16:29:13<53:34:39, 5.72s/it]
23%|██▎ | 10218/43939 [16:29:19<53:36:51, 5.72s/it]
23%|██▎ | 10219/43939 [16:29:24<53:31:36, 5.71s/it]
23%|██▎ | 10220/43939 [16:29:30<53:34:03, 5.72s/it]
23%|██▎ | 10221/43939 [16:29:36<54:12:20, 5.79s/it]
23%|██▎ | 10222/43939 [16:29:42<54:01:53, 5.77s/it]
23%|██▎ | 10223/43939 [16:29:48<54:03:41, 5.77s/it]
23%|██▎ | 10224/43939 [16:29:53<53:51:30, 5.75s/it]
23%|██▎ | 10225/43939 [16:29:59<53:43:40, 5.74s/it]
23%|██▎ | 10226/43939 [16:30:05<53:32:07, 5.72s/it]
23%|██▎ | 10227/43939 [16:30:10<53:31:21, 5.72s/it]
23%|██▎ | 10228/43939 [16:30:16<53:29:05, 5.71s/it]
23%|██▎ | 10229/43939 [16:30:22<53:26:51, 5.71s/it]
23%|██▎ | 10230/43939 [16:30:28<53:31:40, 5.72s/it]
23%|██▎ | 10231/43939 [16:30:33<53:31:32, 5.72s/it]
23%|██▎ | 10232/43939 [16:30:39<53:23:05, 5.70s/it]
23%|██▎ | 10233/43939 [16:30:45<53:23:38, 5.70s/it]
23%|██▎ | 10234/43939 [16:30:50<53:23:15, 5.70s/it]
23%|██▎ | 10235/43939 [16:30:56<53:21:20, 5.70s/it]
23%|██▎ | 10236/43939 [16:31:02<53:21:00, 5.70s/it]
23%|██▎ | 10237/43939 [16:31:07<53:20:29, 5.70s/it]
23%|██▎ | 10238/43939 [16:31:13<53:17:57, 5.69s/it]
23%|██▎ | 10239/43939 [16:31:19<53:20:27, 5.70s/it]
23%|██▎ | 10240/43939 [16:31:25<53:24:52, 5.71s/it]
23%|██▎ | 10241/43939 [16:31:30<53:23:03, 5.70s/it]
23%|██▎ | 10242/43939 [16:31:36<53:18:32, 5.70s/it]
23%|██▎ | 10243/43939 [16:31:42<53:54:36, 5.76s/it]
23%|██▎ | 10244/43939 [16:31:48<53:46:18, 5.75s/it]
23%|██▎ | 10245/43939 [16:31:53<53:40:07, 5.73s/it]
23%|██▎ | 10246/43939 [16:31:59<53:34:02, 5.72s/it]
23%|██▎ | 10247/43939 [16:32:05<53:32:14, 5.72s/it]
23%|██▎ | 10248/43939 [16:32:10<53:28:55, 5.71s/it]
23%|██▎ | 10249/43939 [16:32:16<53:26:26, 5.71s/it]
23%|██▎ | 10250/43939 [16:32:22<53:31:30, 5.72s/it]
{'loss': 1.0076, 'grad_norm': 1.1202757358551025, 'learning_rate': 4.734319239141682e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.17, 'epoch': 0.93}
23%|██▎ | 10250/43939 [16:32:22<53:31:30, 5.72s/it]
23%|██▎ | 10251/43939 [16:32:28<53:37:31, 5.73s/it]
23%|██▎ | 10252/43939 [16:32:33<53:36:38, 5.73s/it]
23%|██▎ | 10253/43939 [16:32:39<53:43:30, 5.74s/it]
23%|██▎ | 10254/43939 [16:32:45<53:38:06, 5.73s/it]
23%|██▎ | 10255/43939 [16:32:51<53:38:59, 5.73s/it]
23%|██▎ | 10256/43939 [16:32:56<53:33:06, 5.72s/it]
23%|██▎ | 10257/43939 [16:33:02<53:27:49, 5.71s/it]
23%|██▎ | 10258/43939 [16:33:08<53:30:04, 5.72s/it]
23%|██▎ | 10259/43939 [16:33:13<53:33:50, 5.73s/it]
23%|██▎ | 10260/43939 [16:33:19<53:28:00, 5.72s/it]
23%|██▎ | 10261/43939 [16:33:25<53:24:17, 5.71s/it]
23%|██▎ | 10262/43939 [16:33:31<53:25:49, 5.71s/it]
23%|██▎ | 10263/43939 [16:33:36<53:21:50, 5.70s/it]
23%|██▎ | 10264/43939 [16:33:42<53:57:13, 5.77s/it]
23%|██▎ | 10265/43939 [16:33:48<53:40:18, 5.74s/it]
23%|██▎ | 10266/43939 [16:33:54<53:42:23, 5.74s/it]
23%|██▎ | 10267/43939 [16:33:59<53:47:24, 5.75s/it]
23%|██▎ | 10268/43939 [16:34:05<53:39:00, 5.74s/it]
23%|██▎ | 10269/43939 [16:34:11<53:27:39, 5.72s/it]
23%|██▎ | 10270/43939 [16:34:16<53:30:14, 5.72s/it]
23%|██▎ | 10271/43939 [16:34:22<53:22:51, 5.71s/it]
23%|██▎ | 10272/43939 [16:34:28<53:19:44, 5.70s/it]
23%|██▎ | 10273/43939 [16:34:34<53:24:38, 5.71s/it]
23%|██▎ | 10274/43939 [16:34:39<53:16:07, 5.70s/it]
23%|██▎ | 10275/43939 [16:34:45<53:15:37, 5.70s/it]
23%|██▎ | 10276/43939 [16:34:51<53:20:36, 5.70s/it]
23%|██▎ | 10277/43939 [16:34:56<53:19:25, 5.70s/it]
23%|██▎ | 10278/43939 [16:35:02<53:18:47, 5.70s/it]
23%|██▎ | 10279/43939 [16:35:08<53:17:37, 5.70s/it]
23%|██▎ | 10280/43939 [16:35:13<53:11:45, 5.69s/it]
23%|██▎ | 10281/43939 [16:35:19<53:17:55, 5.70s/it]
23%|██▎ | 10282/43939 [16:35:25<53:18:15, 5.70s/it]
23%|██▎ | 10283/43939 [16:35:30<53:11:32, 5.69s/it]
23%|██▎ | 10284/43939 [16:35:36<53:58:42, 5.77s/it]
23%|██▎ | 10285/43939 [16:35:42<53:45:28, 5.75s/it]
23%|██▎ | 10286/43939 [16:35:48<53:41:42, 5.74s/it]
23%|██▎ | 10287/43939 [16:35:54<53:42:33, 5.75s/it]
23%|██▎ | 10288/43939 [16:35:59<53:37:13, 5.74s/it]
23%|██▎ | 10289/43939 [16:36:05<53:36:46, 5.74s/it]
23%|██▎ | 10290/43939 [16:36:11<53:35:43, 5.73s/it]
23%|██▎ | 10291/43939 [16:36:16<53:28:36, 5.72s/it]
23%|██▎ | 10292/43939 [16:36:22<53:26:49, 5.72s/it]
23%|██▎ | 10293/43939 [16:36:28<53:22:03, 5.71s/it]
23%|██▎ | 10294/43939 [16:36:34<53:22:53, 5.71s/it]
23%|██▎ | 10295/43939 [16:36:39<53:22:05, 5.71s/it]
23%|██▎ | 10296/43939 [16:36:45<53:23:37, 5.71s/it]
23%|██▎ | 10297/43939 [16:36:51<53:24:35, 5.72s/it]
23%|██▎ | 10298/43939 [16:36:56<53:26:10, 5.72s/it]
23%|██▎ | 10299/43939 [16:37:02<53:29:43, 5.72s/it]
23%|██▎ | 10300/43939 [16:37:08<53:23:07, 5.71s/it]
{'loss': 1.0061, 'grad_norm': 1.2959187030792236, 'learning_rate': 4.72984684630246e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.26, 'epoch': 0.94}
23%|██▎ | 10300/43939 [16:37:08<53:23:07, 5.71s/it]
23%|██▎ | 10301/43939 [16:37:14<53:19:51, 5.71s/it]
23%|██▎ | 10302/43939 [16:37:19<53:21:42, 5.71s/it]
23%|██▎ | 10303/43939 [16:37:25<53:24:21, 5.72s/it]
23%|██▎ | 10304/43939 [16:37:31<53:11:59, 5.69s/it]
23%|██▎ | 10305/43939 [16:37:37<53:53:37, 5.77s/it]
23%|██▎ | 10306/43939 [16:37:42<53:36:32, 5.74s/it]
23%|██▎ | 10307/43939 [16:37:48<53:33:37, 5.73s/it]
23%|██▎ | 10308/43939 [16:37:54<53:23:17, 5.71s/it]
23%|██▎ | 10309/43939 [16:37:59<53:26:33, 5.72s/it]
23%|██▎ | 10310/43939 [16:38:05<53:25:14, 5.72s/it]
23%|██▎ | 10311/43939 [16:38:11<53:19:35, 5.71s/it]
23%|██▎ | 10312/43939 [16:38:16<53:18:13, 5.71s/it]
23%|██▎ | 10313/43939 [16:38:22<53:17:59, 5.71s/it]
23%|██▎ | 10314/43939 [16:38:28<53:13:59, 5.70s/it]
23%|██▎ | 10315/43939 [16:38:34<53:11:34, 5.70s/it]
23%|██▎ | 10316/43939 [16:38:39<53:25:03, 5.72s/it]
23%|██▎ | 10317/43939 [16:38:45<53:24:06, 5.72s/it]
23%|██▎ | 10318/43939 [16:38:51<53:23:32, 5.72s/it]
23%|██▎ | 10319/43939 [16:38:56<53:20:05, 5.71s/it]
23%|██▎ | 10320/43939 [16:39:02<53:15:03, 5.70s/it]
23%|██▎ | 10321/43939 [16:39:08<53:10:08, 5.69s/it]
23%|██▎ | 10322/43939 [16:39:14<53:22:00, 5.71s/it]
23%|██▎ | 10323/43939 [16:39:19<53:15:19, 5.70s/it]
23%|██▎ | 10324/43939 [16:39:25<53:16:41, 5.71s/it]
23%|██▎ | 10325/43939 [16:39:31<53:19:38, 5.71s/it]
24%|██▎ | 10326/43939 [16:39:36<53:11:32, 5.70s/it]
24%|██▎ | 10327/43939 [16:39:42<53:53:29, 5.77s/it]
24%|██▎ | 10328/43939 [16:39:48<53:37:40, 5.74s/it]
24%|██▎ | 10329/43939 [16:39:54<53:27:59, 5.73s/it]
24%|██▎ | 10330/43939 [16:39:59<53:24:37, 5.72s/it]
24%|██▎ | 10331/43939 [16:40:05<53:23:25, 5.72s/it]
24%|██▎ | 10332/43939 [16:40:11<53:18:44, 5.71s/it]
24%|██▎ | 10333/43939 [16:40:16<53:15:18, 5.70s/it]
24%|██▎ | 10334/43939 [16:40:22<53:08:07, 5.69s/it]
24%|██▎ | 10335/43939 [16:40:28<53:08:14, 5.69s/it]
24%|██▎ | 10336/43939 [16:40:34<53:04:44, 5.69s/it]
24%|██▎ | 10337/43939 [16:40:39<53:03:07, 5.68s/it]
24%|██▎ | 10338/43939 [16:40:45<53:10:27, 5.70s/it]
24%|██▎ | 10339/43939 [16:40:51<53:07:44, 5.69s/it]
24%|██▎ | 10340/43939 [16:40:56<53:10:30, 5.70s/it]
24%|██▎ | 10341/43939 [16:41:02<53:07:47, 5.69s/it]
24%|██▎ | 10342/43939 [16:41:08<53:09:34, 5.70s/it]
24%|██▎ | 10343/43939 [16:41:13<53:06:36, 5.69s/it]
24%|██▎ | 10344/43939 [16:41:19<53:04:52, 5.69s/it]
24%|██▎ | 10345/43939 [16:41:25<53:02:21, 5.68s/it]
24%|██▎ | 10346/43939 [16:41:30<53:08:18, 5.69s/it]
24%|██▎ | 10347/43939 [16:41:36<53:44:52, 5.76s/it]
24%|██▎ | 10348/43939 [16:41:42<53:35:50, 5.74s/it]
24%|██▎ | 10349/43939 [16:41:48<53:32:49, 5.74s/it]
24%|██▎ | 10350/43939 [16:41:54<53:29:20, 5.73s/it]
{'loss': 1.0143, 'grad_norm': 1.0744752883911133, 'learning_rate': 4.725339272389168e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.73, 'epoch': 0.94}
24%|██▎ | 10350/43939 [16:41:54<53:29:20, 5.73s/it]
24%|██▎ | 10351/43939 [16:41:59<53:20:00, 5.72s/it]
24%|██▎ | 10352/43939 [16:42:05<53:24:03, 5.72s/it]
24%|██▎ | 10353/43939 [16:42:11<53:30:29, 5.74s/it]
24%|██▎ | 10354/43939 [16:42:16<53:24:22, 5.72s/it]
24%|██▎ | 10355/43939 [16:42:22<53:24:04, 5.72s/it]
24%|██▎ | 10356/43939 [16:42:28<53:21:23, 5.72s/it]
24%|██▎ | 10357/43939 [16:42:33<53:11:11, 5.70s/it]
24%|██▎ | 10358/43939 [16:42:39<53:18:09, 5.71s/it]
24%|██▎ | 10359/43939 [16:42:45<53:17:53, 5.71s/it]
24%|██▎ | 10360/43939 [16:42:51<53:14:00, 5.71s/it]
24%|██▎ | 10361/43939 [16:42:56<53:06:03, 5.69s/it]
24%|██▎ | 10362/43939 [16:43:02<53:00:20, 5.68s/it]
24%|██▎ | 10363/43939 [16:43:08<53:02:14, 5.69s/it]
24%|██▎ | 10364/43939 [16:43:13<53:15:42, 5.71s/it]
24%|██▎ | 10365/43939 [16:43:19<53:16:33, 5.71s/it]
24%|██▎ | 10366/43939 [16:43:25<53:21:43, 5.72s/it]
24%|██▎ | 10367/43939 [16:43:31<53:29:27, 5.74s/it]
24%|██▎ | 10368/43939 [16:43:37<53:56:46, 5.78s/it]
24%|██▎ | 10369/43939 [16:43:42<53:44:57, 5.76s/it]
24%|██▎ | 10370/43939 [16:43:48<53:37:12, 5.75s/it]
24%|██▎ | 10371/43939 [16:43:54<53:27:18, 5.73s/it]
24%|██▎ | 10372/43939 [16:43:59<53:22:03, 5.72s/it]
24%|██▎ | 10373/43939 [16:44:05<53:17:26, 5.72s/it]
24%|██▎ | 10374/43939 [16:44:11<53:14:20, 5.71s/it]
24%|██▎ | 10375/43939 [16:44:16<53:07:36, 5.70s/it]
24%|██▎ | 10376/43939 [16:44:22<53:03:51, 5.69s/it]
24%|██▎ | 10377/43939 [16:44:28<53:10:58, 5.70s/it]
24%|██▎ | 10378/43939 [16:44:34<53:13:05, 5.71s/it]
24%|██▎ | 10379/43939 [16:44:39<53:20:02, 5.72s/it]
24%|██▎ | 10380/43939 [16:44:45<53:16:03, 5.71s/it]
24%|██▎ | 10381/43939 [16:44:51<53:11:05, 5.71s/it]
24%|██▎ | 10382/43939 [16:44:56<53:04:57, 5.69s/it]
24%|██▎ | 10383/43939 [16:45:02<53:06:47, 5.70s/it]
24%|██▎ | 10384/43939 [16:45:08<53:09:05, 5.70s/it]
24%|██▎ | 10385/43939 [16:45:14<53:11:41, 5.71s/it]
24%|██▎ | 10386/43939 [16:45:19<53:10:36, 5.71s/it]
24%|██▎ | 10387/43939 [16:45:25<53:01:50, 5.69s/it]
24%|██▎ | 10388/43939 [16:45:31<53:04:21, 5.69s/it]
24%|██▎ | 10389/43939 [16:45:36<53:00:27, 5.69s/it]
24%|██▎ | 10390/43939 [16:45:42<53:33:53, 5.75s/it]
24%|██▎ | 10391/43939 [16:45:48<53:28:16, 5.74s/it]
24%|██▎ | 10392/43939 [16:45:54<53:20:26, 5.72s/it]
24%|██▎ | 10393/43939 [16:45:59<53:09:56, 5.71s/it]
24%|██▎ | 10394/43939 [16:46:05<53:05:13, 5.70s/it]
24%|██▎ | 10395/43939 [16:46:11<53:04:35, 5.70s/it]
24%|██▎ | 10396/43939 [16:46:16<53:17:14, 5.72s/it]
24%|██▎ | 10397/43939 [16:46:22<53:12:08, 5.71s/it]
24%|██▎ | 10398/43939 [16:46:28<53:10:02, 5.71s/it]
24%|██▎ | 10399/43939 [16:46:33<53:11:12, 5.71s/it]
24%|██▎ | 10400/43939 [16:46:39<53:18:59, 5.72s/it]
{'loss': 1.0254, 'grad_norm': 1.035398006439209, 'learning_rate': 4.72079658851938e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.32, 'epoch': 0.95}
24%|██▎ | 10400/43939 [16:46:39<53:18:59, 5.72s/it]
24%|██▎ | 10401/43939 [16:46:45<53:27:31, 5.74s/it]
24%|██▎ | 10402/43939 [16:46:51<53:27:58, 5.74s/it]
24%|██▎ | 10403/43939 [16:46:56<53:31:41, 5.75s/it]
24%|██▎ | 10404/43939 [16:47:02<53:19:53, 5.73s/it]
24%|██▎ | 10405/43939 [16:47:08<53:24:36, 5.73s/it]
24%|██▎ | 10406/43939 [16:47:14<53:20:39, 5.73s/it]
24%|██▎ | 10407/43939 [16:47:19<53:15:49, 5.72s/it]
24%|██▎ | 10408/43939 [16:47:25<53:09:43, 5.71s/it]
24%|██▎ | 10409/43939 [16:47:31<53:10:19, 5.71s/it]
24%|██▎ | 10410/43939 [16:47:36<53:10:57, 5.71s/it]
24%|██▎ | 10411/43939 [16:47:42<53:40:08, 5.76s/it]
24%|██▎ | 10412/43939 [16:47:48<53:41:27, 5.77s/it]
24%|██▎ | 10413/43939 [16:47:54<53:38:52, 5.76s/it]
24%|██▎ | 10414/43939 [16:47:59<53:19:13, 5.73s/it]
24%|██▎ | 10415/43939 [16:48:05<53:16:30, 5.72s/it]
24%|██▎ | 10416/43939 [16:48:11<53:15:56, 5.72s/it]
24%|██▎ | 10417/43939 [16:48:17<53:12:31, 5.71s/it]
24%|██▎ | 10418/43939 [16:48:22<53:05:13, 5.70s/it]
24%|██▎ | 10419/43939 [16:48:28<53:05:38, 5.70s/it]
24%|██▎ | 10420/43939 [16:48:34<53:21:40, 5.73s/it]
24%|██▎ | 10421/43939 [16:48:39<53:16:21, 5.72s/it]
24%|██▎ | 10422/43939 [16:48:45<53:16:15, 5.72s/it]
24%|██▎ | 10423/43939 [16:48:51<53:06:42, 5.70s/it]
24%|██▎ | 10424/43939 [16:48:57<53:16:33, 5.72s/it]
24%|██▎ | 10425/43939 [16:49:02<53:14:21, 5.72s/it]
24%|██▎ | 10426/43939 [16:49:08<53:13:10, 5.72s/it]
24%|██▎ | 10427/43939 [16:49:14<53:12:19, 5.72s/it]
24%|██▎ | 10428/43939 [16:49:19<53:05:23, 5.70s/it]
24%|██▎ | 10429/43939 [16:49:25<53:02:11, 5.70s/it]
24%|██▎ | 10430/43939 [16:49:31<52:55:48, 5.69s/it]
24%|██▎ | 10431/43939 [16:49:36<53:01:38, 5.70s/it]
24%|██▎ | 10432/43939 [16:49:42<53:38:41, 5.76s/it]
24%|██▎ | 10433/43939 [16:49:48<53:29:37, 5.75s/it]
24%|██▎ | 10434/43939 [16:49:54<53:14:42, 5.72s/it]
24%|██▎ | 10435/43939 [16:49:59<53:09:41, 5.71s/it]
24%|██▍ | 10436/43939 [16:50:05<53:12:16, 5.72s/it]
24%|██▍ | 10437/43939 [16:50:11<53:11:15, 5.72s/it]
24%|██▍ | 10438/43939 [16:50:17<53:03:40, 5.70s/it]
24%|██▍ | 10439/43939 [16:50:22<53:05:42, 5.71s/it]
24%|██▍ | 10440/43939 [16:50:28<53:05:55, 5.71s/it]
24%|██▍ | 10441/43939 [16:50:34<53:03:44, 5.70s/it]
24%|██▍ | 10442/43939 [16:50:39<53:01:15, 5.70s/it]
24%|██▍ | 10443/43939 [16:50:45<53:03:09, 5.70s/it]
24%|██▍ | 10444/43939 [16:50:51<52:56:16, 5.69s/it]
24%|██▍ | 10445/43939 [16:50:57<53:07:38, 5.71s/it]
24%|██▍ | 10446/43939 [16:51:02<53:06:51, 5.71s/it]
24%|██▍ | 10447/43939 [16:51:08<53:08:11, 5.71s/it]
24%|██▍ | 10448/43939 [16:51:14<53:05:14, 5.71s/it]
24%|██▍ | 10449/43939 [16:51:19<53:07:23, 5.71s/it]
24%|██▍ | 10450/43939 [16:51:25<53:03:09, 5.70s/it]
{'loss': 0.9991, 'grad_norm': 1.0589100122451782, 'learning_rate': 4.716218866364606e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.27, 'epoch': 0.95}
24%|██▍ | 10450/43939 [16:51:25<53:03:09, 5.70s/it]
24%|██▍ | 10451/43939 [16:51:31<53:14:01, 5.72s/it]
24%|██▍ | 10452/43939 [16:51:37<53:11:04, 5.72s/it]
24%|██▍ | 10453/43939 [16:51:42<53:45:47, 5.78s/it]
24%|██▍ | 10454/43939 [16:51:48<53:30:02, 5.75s/it]
24%|██▍ | 10455/43939 [16:51:54<53:37:51, 5.77s/it]
24%|██▍ | 10456/43939 [16:52:00<53:23:11, 5.74s/it]
24%|██▍ | 10457/43939 [16:52:05<53:16:59, 5.73s/it]
24%|██▍ | 10458/43939 [16:52:11<53:12:32, 5.72s/it]
24%|██▍ | 10459/43939 [16:52:17<53:03:46, 5.71s/it]
24%|██▍ | 10460/43939 [16:52:22<53:03:54, 5.71s/it]
24%|██▍ | 10461/43939 [16:52:28<53:02:50, 5.70s/it]
24%|██▍ | 10462/43939 [16:52:34<53:02:58, 5.70s/it]
24%|██▍ | 10463/43939 [16:52:40<53:02:36, 5.70s/it]
24%|██▍ | 10464/43939 [16:52:45<53:02:39, 5.70s/it]
24%|██▍ | 10465/43939 [16:52:51<53:05:50, 5.71s/it]
24%|██▍ | 10466/43939 [16:52:57<52:59:42, 5.70s/it]
24%|██▍ | 10467/43939 [16:53:02<53:10:16, 5.72s/it]
24%|██▍ | 10468/43939 [16:53:08<53:12:01, 5.72s/it]
24%|██▍ | 10469/43939 [16:53:14<53:08:39, 5.72s/it]
24%|██▍ | 10470/43939 [16:53:20<53:10:56, 5.72s/it]
24%|██▍ | 10471/43939 [16:53:25<53:08:04, 5.72s/it]
24%|██▍ | 10472/43939 [16:53:31<53:02:08, 5.70s/it]
24%|██▍ | 10473/43939 [16:53:37<52:58:45, 5.70s/it]
24%|██▍ | 10474/43939 [16:53:42<53:31:06, 5.76s/it]
24%|██▍ | 10475/43939 [16:53:48<53:17:53, 5.73s/it]
24%|██▍ | 10476/43939 [16:53:54<53:08:04, 5.72s/it]
24%|██▍ | 10477/43939 [16:54:00<53:05:36, 5.71s/it]
24%|██▍ | 10478/43939 [16:54:05<53:15:18, 5.73s/it]
24%|██▍ | 10479/43939 [16:54:11<53:21:16, 5.74s/it]
24%|██▍ | 10480/43939 [16:54:17<53:12:59, 5.73s/it]
24%|██▍ | 10481/43939 [16:54:22<53:10:51, 5.72s/it]
24%|██▍ | 10482/43939 [16:54:28<53:05:38, 5.71s/it]
24%|██▍ | 10483/43939 [16:54:34<53:03:33, 5.71s/it]
24%|██▍ | 10484/43939 [16:54:40<53:05:17, 5.71s/it]
24%|██▍ | 10485/43939 [16:54:45<53:02:55, 5.71s/it]
24%|██▍ | 10486/43939 [16:54:51<52:55:21, 5.70s/it]
24%|██▍ | 10487/43939 [16:54:57<52:52:08, 5.69s/it]
24%|██▍ | 10488/43939 [16:55:02<52:51:36, 5.69s/it]
24%|██▍ | 10489/43939 [16:55:08<52:50:11, 5.69s/it]
24%|██▍ | 10490/43939 [16:55:14<52:46:14, 5.68s/it]
24%|██▍ | 10491/43939 [16:55:19<52:42:53, 5.67s/it]
24%|██▍ | 10492/43939 [16:55:25<52:42:41, 5.67s/it]
24%|██▍ | 10493/43939 [16:55:31<52:47:42, 5.68s/it]
24%|██▍ | 10494/43939 [16:55:36<52:42:14, 5.67s/it]
24%|██▍ | 10495/43939 [16:55:42<53:33:54, 5.77s/it]
24%|██▍ | 10496/43939 [16:55:48<53:17:43, 5.74s/it]
24%|██▍ | 10497/43939 [16:55:54<53:08:13, 5.72s/it]
24%|██▍ | 10498/43939 [16:55:59<53:08:08, 5.72s/it]
24%|██▍ | 10499/43939 [16:56:05<52:59:07, 5.70s/it]
24%|██▍ | 10500/43939 [16:56:11<53:01:05, 5.71s/it]
{'loss': 1.034, 'grad_norm': 1.072708010673523, 'learning_rate': 4.711606178149173e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.81, 'epoch': 0.96}
24%|██▍ | 10500/43939 [16:56:11<53:01:05, 5.71s/it]
24%|██▍ | 10501/43939 [16:56:16<52:53:29, 5.69s/it]
24%|██▍ | 10502/43939 [16:56:22<52:56:38, 5.70s/it]
24%|██▍ | 10503/43939 [16:56:28<52:49:57, 5.69s/it]
24%|██▍ | 10504/43939 [16:56:34<52:54:24, 5.70s/it]
24%|██▍ | 10505/43939 [16:56:39<52:50:59, 5.69s/it]
24%|██▍ | 10506/43939 [16:56:45<53:00:52, 5.71s/it]
24%|██▍ | 10507/43939 [16:56:51<52:57:54, 5.70s/it]
24%|██▍ | 10508/43939 [16:56:56<52:55:44, 5.70s/it]
24%|██▍ | 10509/43939 [16:57:02<52:53:00, 5.69s/it]
24%|██▍ | 10510/43939 [16:57:08<52:53:05, 5.70s/it]
24%|██▍ | 10511/43939 [16:57:13<52:52:09, 5.69s/it]
24%|██▍ | 10512/43939 [16:57:19<52:51:43, 5.69s/it]
24%|██▍ | 10513/43939 [16:57:25<52:55:43, 5.70s/it]
24%|██▍ | 10514/43939 [16:57:31<52:55:54, 5.70s/it]
24%|██▍ | 10515/43939 [16:57:36<53:01:26, 5.71s/it]
24%|██▍ | 10516/43939 [16:57:42<53:54:53, 5.81s/it]
24%|██▍ | 10517/43939 [16:57:48<53:42:19, 5.78s/it]
24%|██▍ | 10518/43939 [16:57:54<53:33:24, 5.77s/it]
24%|██▍ | 10519/43939 [16:57:59<53:23:46, 5.75s/it]
24%|██▍ | 10520/43939 [16:58:05<53:15:01, 5.74s/it]
24%|██▍ | 10521/43939 [16:58:11<53:12:49, 5.73s/it]
24%|██▍ | 10522/43939 [16:58:17<53:09:10, 5.73s/it]
24%|██▍ | 10523/43939 [16:58:22<53:13:38, 5.73s/it]
24%|██▍ | 10524/43939 [16:58:28<53:21:58, 5.75s/it]
24%|██▍ | 10525/43939 [16:58:34<53:10:37, 5.73s/it]
24%|██▍ | 10526/43939 [16:58:40<53:12:19, 5.73s/it]
24%|██▍ | 10527/43939 [16:58:45<53:06:58, 5.72s/it]
24%|██▍ | 10528/43939 [16:58:51<53:03:38, 5.72s/it]
24%|██▍ | 10529/43939 [16:58:57<53:01:42, 5.71s/it]
24%|██▍ | 10530/43939 [16:59:02<52:51:08, 5.70s/it]
24%|██▍ | 10531/43939 [16:59:08<53:00:44, 5.71s/it]
24%|██▍ | 10532/43939 [16:59:14<52:59:32, 5.71s/it]
24%|██▍ | 10533/43939 [16:59:20<53:01:49, 5.71s/it]
24%|██▍ | 10534/43939 [16:59:25<53:04:08, 5.72s/it]
24%|██▍ | 10535/43939 [16:59:31<52:58:00, 5.71s/it]
24%|██▍ | 10536/43939 [16:59:37<52:55:30, 5.70s/it]
24%|██▍ | 10537/43939 [16:59:42<52:59:15, 5.71s/it]
24%|██▍ | 10538/43939 [16:59:48<53:35:53, 5.78s/it]
24%|██▍ | 10539/43939 [16:59:54<53:28:46, 5.76s/it]
24%|██▍ | 10540/43939 [17:00:00<53:20:30, 5.75s/it]
24%|██▍ | 10541/43939 [17:00:05<53:16:02, 5.74s/it]
24%|██▍ | 10542/43939 [17:00:11<53:07:20, 5.73s/it]
24%|██▍ | 10543/43939 [17:00:17<53:09:29, 5.73s/it]
24%|██▍ | 10544/43939 [17:00:23<52:56:29, 5.71s/it]
24%|██▍ | 10545/43939 [17:00:28<52:57:58, 5.71s/it]
24%|██▍ | 10546/43939 [17:00:34<52:52:48, 5.70s/it]
24%|██▍ | 10547/43939 [17:00:40<52:55:14, 5.71s/it]
24%|██▍ | 10548/43939 [17:00:45<52:49:39, 5.70s/it]
24%|██▍ | 10549/43939 [17:00:51<52:48:12, 5.69s/it]
24%|██▍ | 10550/43939 [17:00:57<52:47:07, 5.69s/it]
{'loss': 1.0254, 'grad_norm': 1.2344636917114258, 'learning_rate': 4.706958596649076e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.35, 'epoch': 0.96}
24%|██▍ | 10550/43939 [17:00:57<52:47:07, 5.69s/it]
24%|██▍ | 10551/43939 [17:01:02<52:47:15, 5.69s/it]
24%|██▍ | 10552/43939 [17:01:08<52:45:22, 5.69s/it]
24%|██▍ | 10553/43939 [17:01:14<52:51:13, 5.70s/it]
24%|██▍ | 10554/43939 [17:01:19<52:46:49, 5.69s/it]
24%|██▍ | 10555/43939 [17:01:25<52:48:53, 5.70s/it]
24%|██▍ | 10556/43939 [17:01:31<52:51:27, 5.70s/it]
24%|██▍ | 10557/43939 [17:01:37<52:47:33, 5.69s/it]
24%|██▍ | 10558/43939 [17:01:42<52:47:26, 5.69s/it]
24%|██▍ | 10559/43939 [17:01:48<53:27:36, 5.77s/it]
24%|██▍ | 10560/43939 [17:01:54<53:16:55, 5.75s/it]
24%|██▍ | 10561/43939 [17:02:00<53:12:28, 5.74s/it]
24%|██▍ | 10562/43939 [17:02:05<53:04:08, 5.72s/it]
24%|██▍ | 10563/43939 [17:02:11<52:57:25, 5.71s/it]
24%|██▍ | 10564/43939 [17:02:17<52:54:54, 5.71s/it]
24%|██▍ | 10565/43939 [17:02:22<52:53:51, 5.71s/it]
24%|██▍ | 10566/43939 [17:02:28<52:45:25, 5.69s/it]
24%|██▍ | 10567/43939 [17:02:34<52:42:50, 5.69s/it]
24%|██▍ | 10568/43939 [17:02:39<52:49:49, 5.70s/it]
24%|██▍ | 10569/43939 [17:02:45<52:47:25, 5.70s/it]
24%|██▍ | 10570/43939 [17:02:51<52:45:00, 5.69s/it]
24%|██▍ | 10571/43939 [17:02:57<52:43:30, 5.69s/it]
24%|██▍ | 10572/43939 [17:03:02<52:41:59, 5.69s/it]
24%|██▍ | 10573/43939 [17:03:08<52:40:54, 5.68s/it]
24%|██▍ | 10574/43939 [17:03:14<52:45:03, 5.69s/it]
24%|██▍ | 10575/43939 [17:03:19<52:43:32, 5.69s/it]
24%|██▍ | 10576/43939 [17:03:25<52:44:40, 5.69s/it]
24%|██▍ | 10577/43939 [17:03:31<52:41:36, 5.69s/it]
24%|██▍ | 10578/43939 [17:03:36<52:47:50, 5.70s/it]
24%|██▍ | 10579/43939 [17:03:42<52:47:34, 5.70s/it]
24%|██▍ | 10580/43939 [17:03:48<53:32:21, 5.78s/it]
24%|██▍ | 10581/43939 [17:03:54<53:17:02, 5.75s/it]
24%|██▍ | 10582/43939 [17:03:59<53:11:42, 5.74s/it]
24%|██▍ | 10583/43939 [17:04:05<53:04:44, 5.73s/it]
24%|██▍ | 10584/43939 [17:04:11<52:59:13, 5.72s/it]
24%|██▍ | 10585/43939 [17:04:17<52:56:11, 5.71s/it]
24%|██▍ | 10586/43939 [17:04:22<52:49:35, 5.70s/it]
24%|██▍ | 10587/43939 [17:04:28<52:42:38, 5.69s/it]
24%|██▍ | 10588/43939 [17:04:34<52:47:04, 5.70s/it]
24%|██▍ | 10589/43939 [17:04:39<52:52:59, 5.71s/it]
24%|██▍ | 10590/43939 [17:04:45<52:50:16, 5.70s/it]
24%|██▍ | 10591/43939 [17:04:51<52:49:07, 5.70s/it]
24%|██▍ | 10592/43939 [17:04:56<52:47:00, 5.70s/it]
24%|██▍ | 10593/43939 [17:05:02<52:49:29, 5.70s/it]
24%|██▍ | 10594/43939 [17:05:08<52:46:34, 5.70s/it]
24%|██▍ | 10595/43939 [17:05:13<52:42:36, 5.69s/it]
24%|██▍ | 10596/43939 [17:05:19<52:50:55, 5.71s/it]
24%|██▍ | 10597/43939 [17:05:25<52:47:37, 5.70s/it]
24%|██▍ | 10598/43939 [17:05:31<52:44:48, 5.70s/it]
24%|██▍ | 10599/43939 [17:05:36<52:55:52, 5.72s/it]
24%|██▍ | 10600/43939 [17:05:42<53:02:00, 5.73s/it]
{'loss': 1.0171, 'grad_norm': 0.9406333565711975, 'learning_rate': 4.7022761951908365e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.51, 'epoch': 0.96}
24%|██▍ | 10600/43939 [17:05:42<53:02:00, 5.73s/it]
24%|██▍ | 10601/43939 [17:05:48<53:41:35, 5.80s/it]
24%|██▍ | 10602/43939 [17:05:54<53:31:45, 5.78s/it]
24%|██▍ | 10603/43939 [17:05:59<53:13:30, 5.75s/it]
24%|██▍ | 10604/43939 [17:06:05<53:05:23, 5.73s/it]
24%|██▍ | 10605/43939 [17:06:11<52:58:45, 5.72s/it]
24%|██▍ | 10606/43939 [17:06:17<52:54:36, 5.71s/it]
24%|██▍ | 10607/43939 [17:06:22<52:57:41, 5.72s/it]
24%|██▍ | 10608/43939 [17:06:28<52:58:19, 5.72s/it]
24%|██▍ | 10609/43939 [17:06:34<52:53:35, 5.71s/it]
24%|██▍ | 10610/43939 [17:06:39<52:52:06, 5.71s/it]
24%|██▍ | 10611/43939 [17:06:45<52:48:24, 5.70s/it]
24%|██▍ | 10612/43939 [17:06:51<52:47:41, 5.70s/it]
24%|██▍ | 10613/43939 [17:06:57<52:48:28, 5.70s/it]
24%|██▍ | 10614/43939 [17:07:02<52:46:49, 5.70s/it]
24%|██▍ | 10615/43939 [17:07:08<52:47:50, 5.70s/it]
24%|██▍ | 10616/43939 [17:07:14<52:43:38, 5.70s/it]
24%|██▍ | 10617/43939 [17:07:19<52:47:04, 5.70s/it]
24%|██▍ | 10618/43939 [17:07:25<52:47:32, 5.70s/it]
24%|██▍ | 10619/43939 [17:07:31<52:48:07, 5.70s/it]
24%|██▍ | 10620/43939 [17:07:36<52:42:45, 5.70s/it]
24%|██▍ | 10621/43939 [17:07:42<52:41:40, 5.69s/it]
24%|██▍ | 10622/43939 [17:07:48<52:49:53, 5.71s/it]
24%|██▍ | 10623/43939 [17:07:54<53:16:01, 5.76s/it]
24%|██▍ | 10624/43939 [17:07:59<53:02:38, 5.73s/it]
24%|██▍ | 10625/43939 [17:08:05<53:00:34, 5.73s/it]
24%|██▍ | 10626/43939 [17:08:11<52:55:59, 5.72s/it]
24%|██▍ | 10627/43939 [17:08:16<52:49:25, 5.71s/it]
24%|██▍ | 10628/43939 [17:08:22<52:47:24, 5.71s/it]
24%|██▍ | 10629/43939 [17:08:28<52:48:58, 5.71s/it]
24%|██▍ | 10630/43939 [17:08:34<52:47:35, 5.71s/it]
24%|██▍ | 10631/43939 [17:08:39<52:47:28, 5.71s/it]
24%|██▍ | 10632/43939 [17:08:45<52:48:07, 5.71s/it]
24%|██▍ | 10633/43939 [17:08:51<52:40:38, 5.69s/it]
24%|██▍ | 10634/43939 [17:08:56<52:39:29, 5.69s/it]
24%|██▍ | 10635/43939 [17:09:02<52:44:56, 5.70s/it]
24%|██▍ | 10636/43939 [17:09:08<52:51:47, 5.71s/it]
24%|██▍ | 10637/43939 [17:09:13<52:45:11, 5.70s/it]
24%|██▍ | 10638/43939 [17:09:19<52:48:59, 5.71s/it]
24%|██▍ | 10639/43939 [17:09:25<52:55:57, 5.72s/it]
24%|██▍ | 10640/43939 [17:09:31<52:50:58, 5.71s/it]
24%|██▍ | 10641/43939 [17:09:36<52:50:40, 5.71s/it]
24%|██▍ | 10642/43939 [17:09:42<52:54:35, 5.72s/it]
24%|██▍ | 10643/43939 [17:09:48<52:46:55, 5.71s/it]
24%|██▍ | 10644/43939 [17:09:54<53:23:18, 5.77s/it]
24%|██▍ | 10645/43939 [17:09:59<53:09:38, 5.75s/it]
24%|██▍ | 10646/43939 [17:10:05<52:54:37, 5.72s/it]
24%|██▍ | 10647/43939 [17:10:11<52:45:00, 5.70s/it]
24%|██▍ | 10648/43939 [17:10:16<52:44:38, 5.70s/it]
24%|██▍ | 10649/43939 [17:10:22<52:48:34, 5.71s/it]
24%|██▍ | 10650/43939 [17:10:28<52:49:12, 5.71s/it]
{'loss': 1.017, 'grad_norm': 1.1612303256988525, 'learning_rate': 4.6975590476503396e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.48, 'epoch': 0.97}
24%|██▍ | 10650/43939 [17:10:28<52:49:12, 5.71s/it]
24%|██▍ | 10651/43939 [17:10:34<52:48:26, 5.71s/it]
24%|██▍ | 10652/43939 [17:10:39<52:49:50, 5.71s/it]
24%|██▍ | 10653/43939 [17:10:45<52:41:20, 5.70s/it]
24%|██▍ | 10654/43939 [17:10:51<52:41:53, 5.70s/it]
24%|██▍ | 10655/43939 [17:10:56<52:43:01, 5.70s/it]
24%|██▍ | 10656/43939 [17:11:02<52:43:00, 5.70s/it]
24%|██▍ | 10657/43939 [17:11:08<52:38:59, 5.69s/it]
24%|██▍ | 10658/43939 [17:11:13<52:36:06, 5.69s/it]
24%|██▍ | 10659/43939 [17:11:19<52:45:41, 5.71s/it]
24%|██▍ | 10660/43939 [17:11:25<52:38:25, 5.69s/it]
24%|██▍ | 10661/43939 [17:11:31<52:34:30, 5.69s/it]
24%|██▍ | 10662/43939 [17:11:36<52:35:54, 5.69s/it]
24%|██▍ | 10663/43939 [17:11:42<52:31:59, 5.68s/it]
24%|██▍ | 10664/43939 [17:11:48<52:40:20, 5.70s/it]
24%|██▍ | 10665/43939 [17:11:54<53:22:26, 5.77s/it]
24%|██▍ | 10666/43939 [17:11:59<53:08:32, 5.75s/it]
24%|██▍ | 10667/43939 [17:12:05<52:59:48, 5.73s/it]
24%|██▍ | 10668/43939 [17:12:11<52:53:29, 5.72s/it]
24%|██▍ | 10669/43939 [17:12:16<52:48:00, 5.71s/it]
24%|██▍ | 10670/43939 [17:12:22<52:55:32, 5.73s/it]
24%|██▍ | 10671/43939 [17:12:28<52:59:51, 5.73s/it]
24%|██▍ | 10672/43939 [17:12:34<52:54:46, 5.73s/it]
24%|██▍ | 10673/43939 [17:12:39<52:49:08, 5.72s/it]
24%|██▍ | 10674/43939 [17:12:45<52:44:02, 5.71s/it]
24%|██▍ | 10675/43939 [17:12:51<52:42:05, 5.70s/it]
24%|██▍ | 10676/43939 [17:12:56<52:36:41, 5.69s/it]
24%|██▍ | 10677/43939 [17:13:02<52:38:51, 5.70s/it]
24%|██▍ | 10678/43939 [17:13:08<52:35:13, 5.69s/it]
24%|██▍ | 10679/43939 [17:13:13<52:33:28, 5.69s/it]
24%|██▍ | 10680/43939 [17:13:19<52:41:59, 5.70s/it]
24%|██▍ | 10681/43939 [17:13:25<52:42:47, 5.71s/it]
24%|██▍ | 10682/43939 [17:13:31<52:42:53, 5.71s/it]
24%|██▍ | 10683/43939 [17:13:36<52:44:44, 5.71s/it]
24%|██▍ | 10684/43939 [17:13:42<52:40:21, 5.70s/it]
24%|██▍ | 10685/43939 [17:13:48<52:47:52, 5.72s/it]
24%|██▍ | 10686/43939 [17:13:54<53:25:20, 5.78s/it]
24%|██▍ | 10687/43939 [17:13:59<53:08:48, 5.75s/it]
24%|██▍ | 10688/43939 [17:14:05<53:01:56, 5.74s/it]
24%|██▍ | 10689/43939 [17:14:11<52:48:46, 5.72s/it]
24%|██▍ | 10690/43939 [17:14:16<52:46:39, 5.71s/it]
24%|██▍ | 10691/43939 [17:14:22<52:42:52, 5.71s/it]
24%|██▍ | 10692/43939 [17:14:28<52:34:47, 5.69s/it]
24%|██▍ | 10693/43939 [17:14:33<52:34:17, 5.69s/it]
24%|██▍ | 10694/43939 [17:14:39<52:31:41, 5.69s/it]
24%|██▍ | 10695/43939 [17:14:45<52:28:12, 5.68s/it]
24%|██▍ | 10696/43939 [17:14:50<52:28:55, 5.68s/it]
24%|██▍ | 10697/43939 [17:14:56<52:34:47, 5.69s/it]
24%|██▍ | 10698/43939 [17:15:02<52:30:36, 5.69s/it]
24%|██▍ | 10699/43939 [17:15:08<52:32:06, 5.69s/it]
24%|██▍ | 10700/43939 [17:15:13<52:34:05, 5.69s/it]
{'loss': 1.0212, 'grad_norm': 1.5133899450302124, 'learning_rate': 4.692807228451672e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.52, 'epoch': 0.97}
24%|██▍ | 10700/43939 [17:15:13<52:34:05, 5.69s/it]
24%|██▍ | 10701/43939 [17:15:19<52:34:29, 5.69s/it]
24%|██▍ | 10702/43939 [17:15:25<52:36:06, 5.70s/it]
24%|██▍ | 10703/43939 [17:15:30<52:33:37, 5.69s/it]
24%|██▍ | 10704/43939 [17:15:36<52:40:05, 5.70s/it]
24%|██▍ | 10705/43939 [17:15:42<52:44:17, 5.71s/it]
24%|██▍ | 10706/43939 [17:15:48<52:45:05, 5.71s/it]
24%|██▍ | 10707/43939 [17:15:53<52:44:31, 5.71s/it]
24%|██▍ | 10708/43939 [17:15:59<53:17:37, 5.77s/it]
24%|██▍ | 10709/43939 [17:16:05<53:06:14, 5.75s/it]
24%|██▍ | 10710/43939 [17:16:11<52:54:29, 5.73s/it]
24%|██▍ | 10711/43939 [17:16:16<52:48:00, 5.72s/it]
24%|██▍ | 10712/43939 [17:16:22<52:46:19, 5.72s/it]
24%|██▍ | 10713/43939 [17:16:28<52:47:38, 5.72s/it]
24%|██▍ | 10714/43939 [17:16:33<52:51:26, 5.73s/it]
24%|██▍ | 10715/43939 [17:16:39<52:46:06, 5.72s/it]
24%|██▍ | 10716/43939 [17:16:45<52:51:43, 5.73s/it]
24%|██▍ | 10717/43939 [17:16:51<52:55:33, 5.74s/it]
24%|██▍ | 10718/43939 [17:16:56<52:56:34, 5.74s/it]
24%|██▍ | 10719/43939 [17:17:02<52:53:57, 5.73s/it]
24%|██▍ | 10720/43939 [17:17:08<52:58:30, 5.74s/it]
24%|██▍ | 10721/43939 [17:17:14<52:51:23, 5.73s/it]
24%|██▍ | 10722/43939 [17:17:19<52:41:42, 5.71s/it]
24%|██▍ | 10723/43939 [17:17:25<52:38:03, 5.70s/it]
24%|██▍ | 10724/43939 [17:17:31<52:39:06, 5.71s/it]
24%|██▍ | 10725/43939 [17:17:36<52:35:54, 5.70s/it]
24%|██▍ | 10726/43939 [17:17:42<52:39:33, 5.71s/it]
24%|██▍ | 10727/43939 [17:17:48<52:42:33, 5.71s/it]
24%|██▍ | 10728/43939 [17:17:54<53:16:21, 5.77s/it]
24%|██▍ | 10729/43939 [17:17:59<53:05:22, 5.75s/it]
24%|██▍ | 10730/43939 [17:18:05<52:57:46, 5.74s/it]
24%|██▍ | 10731/43939 [17:18:11<52:44:44, 5.72s/it]
24%|██▍ | 10732/43939 [17:18:16<52:43:47, 5.72s/it]
24%|██▍ | 10733/43939 [17:18:22<52:38:25, 5.71s/it]
24%|██▍ | 10734/43939 [17:18:28<52:37:14, 5.70s/it]
24%|██▍ | 10735/43939 [17:18:34<52:38:55, 5.71s/it]
24%|██▍ | 10736/43939 [17:18:39<52:34:51, 5.70s/it]
24%|██▍ | 10737/43939 [17:18:45<52:39:11, 5.71s/it]
24%|██▍ | 10738/43939 [17:18:51<52:29:56, 5.69s/it]
24%|██▍ | 10739/43939 [17:18:56<52:34:57, 5.70s/it]
24%|██▍ | 10740/43939 [17:19:02<52:38:40, 5.71s/it]
24%|██▍ | 10741/43939 [17:19:08<52:47:03, 5.72s/it]
24%|██▍ | 10742/43939 [17:19:14<52:47:43, 5.73s/it]
24%|██▍ | 10743/43939 [17:19:19<52:41:44, 5.71s/it]
24%|██▍ | 10744/43939 [17:19:25<52:43:51, 5.72s/it]
24%|██▍ | 10745/43939 [17:19:31<52:44:27, 5.72s/it]
24%|██▍ | 10746/43939 [17:19:36<52:38:13, 5.71s/it]
24%|██▍ | 10747/43939 [17:19:42<52:39:55, 5.71s/it]
24%|██▍ | 10748/43939 [17:19:48<52:34:55, 5.70s/it]
24%|██▍ | 10749/43939 [17:19:54<53:10:36, 5.77s/it]
24%|██▍ | 10750/43939 [17:19:59<52:53:54, 5.74s/it]
{'loss': 1.009, 'grad_norm': 1.1254411935806274, 'learning_rate': 4.68802081256595e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.76, 'epoch': 0.98}
24%|██▍ | 10750/43939 [17:19:59<52:53:54, 5.74s/it]
24%|██▍ | 10751/43939 [17:20:05<52:51:03, 5.73s/it]
24%|██▍ | 10752/43939 [17:20:11<52:43:09, 5.72s/it]
24%|██▍ | 10753/43939 [17:20:16<52:43:58, 5.72s/it]
24%|██▍ | 10754/43939 [17:20:22<52:38:53, 5.71s/it]
24%|██▍ | 10755/43939 [17:20:28<52:35:56, 5.71s/it]
24%|██▍ | 10756/43939 [17:20:34<52:37:20, 5.71s/it]
24%|██▍ | 10757/43939 [17:20:39<52:33:19, 5.70s/it]
24%|██▍ | 10758/43939 [17:20:45<52:41:33, 5.72s/it]
24%|██▍ | 10759/43939 [17:20:51<52:37:43, 5.71s/it]
24%|██▍ | 10760/43939 [17:20:56<52:34:10, 5.70s/it]
24%|██▍ | 10761/43939 [17:21:02<52:32:57, 5.70s/it]
24%|██▍ | 10762/43939 [17:21:08<52:31:54, 5.70s/it]
24%|██▍ | 10763/43939 [17:21:13<52:25:15, 5.69s/it]
24%|██▍ | 10764/43939 [17:21:19<52:24:09, 5.69s/it]
24%|██▍ | 10765/43939 [17:21:25<52:25:41, 5.69s/it]
25%|██▍ | 10766/43939 [17:21:31<52:22:20, 5.68s/it]
25%|██▍ | 10767/43939 [17:21:36<52:17:57, 5.68s/it]
25%|██▍ | 10768/43939 [17:21:42<52:26:48, 5.69s/it]
25%|██▍ | 10769/43939 [17:21:48<52:25:07, 5.69s/it]
25%|██▍ | 10770/43939 [17:21:53<52:20:55, 5.68s/it]
25%|██▍ | 10771/43939 [17:21:59<53:03:15, 5.76s/it]
25%|██▍ | 10772/43939 [17:22:05<52:53:06, 5.74s/it]
25%|██▍ | 10773/43939 [17:22:11<52:48:42, 5.73s/it]
25%|██▍ | 10774/43939 [17:22:16<52:44:26, 5.72s/it]
25%|██▍ | 10775/43939 [17:22:22<52:45:57, 5.73s/it]
25%|██▍ | 10776/43939 [17:22:28<52:40:26, 5.72s/it]
25%|██▍ | 10777/43939 [17:22:33<52:33:17, 5.71s/it]
25%|██▍ | 10778/43939 [17:22:39<52:25:30, 5.69s/it]
25%|██▍ | 10779/43939 [17:22:45<52:24:44, 5.69s/it]
25%|██▍ | 10780/43939 [17:22:50<52:24:35, 5.69s/it]
25%|██▍ | 10781/43939 [17:22:56<52:32:49, 5.71s/it]
25%|██▍ | 10782/43939 [17:23:02<52:31:33, 5.70s/it]
25%|██▍ | 10783/43939 [17:23:08<52:26:07, 5.69s/it]
25%|██▍ | 10784/43939 [17:23:13<52:31:49, 5.70s/it]
25%|██▍ | 10785/43939 [17:23:19<52:25:14, 5.69s/it]
25%|██▍ | 10786/43939 [17:23:25<52:23:41, 5.69s/it]
25%|██▍ | 10787/43939 [17:23:30<52:23:56, 5.69s/it]
25%|██▍ | 10788/43939 [17:23:36<52:22:13, 5.69s/it]
25%|██▍ | 10789/43939 [17:23:42<52:26:08, 5.69s/it]
25%|██▍ | 10790/43939 [17:23:47<52:22:56, 5.69s/it]
25%|██▍ | 10791/43939 [17:23:53<53:00:53, 5.76s/it]
25%|██▍ | 10792/43939 [17:23:59<52:58:49, 5.75s/it]
25%|██▍ | 10793/43939 [17:24:05<52:58:07, 5.75s/it]
25%|██▍ | 10794/43939 [17:24:11<52:50:37, 5.74s/it]
25%|██▍ | 10795/43939 [17:24:16<52:43:45, 5.73s/it]
25%|██▍ | 10796/43939 [17:24:22<52:32:45, 5.71s/it]
25%|██▍ | 10797/43939 [17:24:28<52:36:29, 5.71s/it]
25%|██▍ | 10798/43939 [17:24:33<52:33:12, 5.71s/it]
25%|██▍ | 10799/43939 [17:24:39<52:28:44, 5.70s/it]
25%|██▍ | 10800/43939 [17:24:45<52:26:57, 5.70s/it]
{'loss': 1.0262, 'grad_norm': 1.0411556959152222, 'learning_rate': 4.683199875510132e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.64, 'epoch': 0.98}
25%|██▍ | 10800/43939 [17:24:45<52:26:57, 5.70s/it]
25%|██▍ | 10801/43939 [17:24:50<52:25:16, 5.69s/it]
25%|██▍ | 10802/43939 [17:24:56<52:28:06, 5.70s/it]
25%|██▍ | 10803/43939 [17:25:02<52:23:41, 5.69s/it]
25%|██▍ | 10804/43939 [17:25:07<52:24:54, 5.69s/it]
25%|██▍ | 10805/43939 [17:25:13<52:18:34, 5.68s/it]
25%|██▍ | 10806/43939 [17:25:19<52:25:10, 5.70s/it]
25%|██▍ | 10807/43939 [17:25:24<52:19:39, 5.69s/it]
25%|██▍ | 10808/43939 [17:25:30<52:21:21, 5.69s/it]
25%|██▍ | 10809/43939 [17:25:36<52:21:26, 5.69s/it]
25%|██▍ | 10810/43939 [17:25:42<52:19:52, 5.69s/it]
25%|██▍ | 10811/43939 [17:25:47<52:29:59, 5.71s/it]
25%|██▍ | 10812/43939 [17:25:53<53:02:10, 5.76s/it]
25%|██▍ | 10813/43939 [17:25:59<52:58:58, 5.76s/it]
25%|██▍ | 10814/43939 [17:26:05<52:52:08, 5.75s/it]
25%|██▍ | 10815/43939 [17:26:10<52:47:11, 5.74s/it]
25%|██▍ | 10816/43939 [17:26:16<52:43:24, 5.73s/it]
25%|██▍ | 10817/43939 [17:26:22<52:38:05, 5.72s/it]
25%|██▍ | 10818/43939 [17:26:28<52:36:59, 5.72s/it]
25%|██▍ | 10819/43939 [17:26:33<52:42:03, 5.73s/it]
25%|██▍ | 10820/43939 [17:26:39<52:41:05, 5.73s/it]
25%|██▍ | 10821/43939 [17:26:45<52:30:08, 5.71s/it]
25%|██▍ | 10822/43939 [17:26:50<52:27:04, 5.70s/it]
25%|██▍ | 10823/43939 [17:26:56<52:28:30, 5.70s/it]
25%|██▍ | 10824/43939 [17:27:02<52:33:03, 5.71s/it]
25%|██▍ | 10825/43939 [17:27:07<52:30:48, 5.71s/it]
25%|██▍ | 10826/43939 [17:27:13<52:26:27, 5.70s/it]
25%|██▍ | 10827/43939 [17:27:19<52:18:43, 5.69s/it]
25%|██▍ | 10828/43939 [17:27:25<52:29:17, 5.71s/it]
25%|██▍ | 10829/43939 [17:27:30<52:32:45, 5.71s/it]
25%|██▍ | 10830/43939 [17:27:36<52:25:28, 5.70s/it]
25%|██▍ | 10831/43939 [17:27:42<52:22:34, 5.70s/it]
25%|██▍ | 10832/43939 [17:27:47<52:30:45, 5.71s/it]
25%|██▍ | 10833/43939 [17:27:53<52:24:49, 5.70s/it]
25%|██▍ | 10834/43939 [17:27:59<53:09:21, 5.78s/it]
25%|██▍ | 10835/43939 [17:28:05<52:51:21, 5.75s/it]
25%|██▍ | 10836/43939 [17:28:10<52:39:42, 5.73s/it]
25%|██▍ | 10837/43939 [17:28:16<52:43:51, 5.73s/it]
25%|██▍ | 10838/43939 [17:28:22<52:34:44, 5.72s/it]
25%|██▍ | 10839/43939 [17:28:28<52:27:20, 5.71s/it]
25%|██▍ | 10840/43939 [17:28:33<52:32:27, 5.71s/it]
25%|██▍ | 10841/43939 [17:28:39<52:37:51, 5.72s/it]
25%|██▍ | 10842/43939 [17:28:45<52:35:26, 5.72s/it]
25%|██▍ | 10843/43939 [17:28:50<52:34:15, 5.72s/it]
25%|██▍ | 10844/43939 [17:28:56<52:35:58, 5.72s/it]
25%|██▍ | 10845/43939 [17:29:02<52:34:57, 5.72s/it]
25%|██▍ | 10846/43939 [17:29:08<52:39:13, 5.73s/it]
25%|██▍ | 10847/43939 [17:29:13<52:34:37, 5.72s/it]
25%|██▍ | 10848/43939 [17:29:19<52:33:57, 5.72s/it]
25%|██▍ | 10849/43939 [17:29:25<52:32:27, 5.72s/it]
25%|██▍ | 10850/43939 [17:29:30<52:26:02, 5.70s/it]
{'loss': 1.0246, 'grad_norm': 0.979972243309021, 'learning_rate': 4.6783444933458284e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.13, 'epoch': 0.99}
25%|██▍ | 10850/43939 [17:29:30<52:26:02, 5.70s/it]
25%|██▍ | 10851/43939 [17:29:36<52:42:23, 5.73s/it]
25%|██▍ | 10852/43939 [17:29:42<52:36:37, 5.72s/it]
25%|██▍ | 10853/43939 [17:29:48<52:26:50, 5.71s/it]
25%|██▍ | 10854/43939 [17:29:54<53:04:14, 5.77s/it]
25%|██▍ | 10855/43939 [17:29:59<52:46:11, 5.74s/it]
25%|██▍ | 10856/43939 [17:30:05<52:46:20, 5.74s/it]
25%|██▍ | 10857/43939 [17:30:11<52:39:26, 5.73s/it]
25%|██▍ | 10858/43939 [17:30:16<52:33:47, 5.72s/it]
25%|██▍ | 10859/43939 [17:30:22<52:29:54, 5.71s/it]
25%|██▍ | 10860/43939 [17:30:28<52:19:59, 5.70s/it]
25%|██▍ | 10861/43939 [17:30:33<52:16:17, 5.69s/it]
25%|██▍ | 10862/43939 [17:30:39<52:18:22, 5.69s/it]
25%|██▍ | 10863/43939 [17:30:45<52:20:29, 5.70s/it]
25%|██▍ | 10864/43939 [17:30:50<52:18:37, 5.69s/it]
25%|██▍ | 10865/43939 [17:30:56<52:18:14, 5.69s/it]
25%|██▍ | 10866/43939 [17:31:02<52:22:56, 5.70s/it]
25%|██▍ | 10867/43939 [17:31:08<52:23:24, 5.70s/it]
25%|██▍ | 10868/43939 [17:31:13<52:19:29, 5.70s/it]
25%|██▍ | 10869/43939 [17:31:19<52:26:07, 5.71s/it]
25%|██▍ | 10870/43939 [17:31:25<52:21:53, 5.70s/it]
25%|██▍ | 10871/43939 [17:31:30<52:22:58, 5.70s/it]
25%|██▍ | 10872/43939 [17:31:36<52:18:38, 5.70s/it]
25%|██▍ | 10873/43939 [17:31:42<52:16:13, 5.69s/it]
25%|██▍ | 10874/43939 [17:31:47<52:17:50, 5.69s/it]
25%|██▍ | 10875/43939 [17:31:53<53:05:09, 5.78s/it]
25%|██▍ | 10876/43939 [17:31:59<52:54:20, 5.76s/it]
25%|██▍ | 10877/43939 [17:32:05<52:53:18, 5.76s/it]
25%|██▍ | 10878/43939 [17:32:11<52:48:29, 5.75s/it]
25%|██▍ | 10879/43939 [17:32:16<52:46:15, 5.75s/it]
25%|██▍ | 10880/43939 [17:32:22<52:33:17, 5.72s/it]
25%|██▍ | 10881/43939 [17:32:28<52:31:34, 5.72s/it]
25%|██▍ | 10882/43939 [17:32:33<52:28:19, 5.71s/it]
25%|██▍ | 10883/43939 [17:32:39<52:26:06, 5.71s/it]
25%|██▍ | 10884/43939 [17:32:45<52:26:03, 5.71s/it]
25%|██▍ | 10885/43939 [17:32:51<52:33:39, 5.72s/it]
25%|██▍ | 10886/43939 [17:32:56<52:26:22, 5.71s/it]
25%|██▍ | 10887/43939 [17:33:02<52:28:47, 5.72s/it]
25%|██▍ | 10888/43939 [17:33:08<52:19:27, 5.70s/it]
25%|██▍ | 10889/43939 [17:33:13<52:21:46, 5.70s/it]
25%|██▍ | 10890/43939 [17:33:19<52:18:02, 5.70s/it]
25%|██▍ | 10891/43939 [17:33:25<52:12:34, 5.69s/it]
25%|██▍ | 10892/43939 [17:33:30<52:16:46, 5.70s/it]
25%|██▍ | 10893/43939 [17:33:36<52:16:36, 5.69s/it]
25%|██▍ | 10894/43939 [17:33:42<52:12:22, 5.69s/it]
25%|██▍ | 10895/43939 [17:33:48<52:14:22, 5.69s/it]
25%|██▍ | 10896/43939 [17:33:53<52:23:31, 5.71s/it]
25%|██▍ | 10897/43939 [17:33:59<53:01:55, 5.78s/it]
25%|██▍ | 10898/43939 [17:34:05<52:44:46, 5.75s/it]
25%|██▍ | 10899/43939 [17:34:11<52:44:55, 5.75s/it]
25%|██▍ | 10900/43939 [17:34:16<52:38:16, 5.74s/it]
{'loss': 1.0532, 'grad_norm': 1.1155357360839844, 'learning_rate': 4.673454742678102e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 505.17, 'epoch': 0.99}
25%|██▍ | 10900/43939 [17:34:16<52:38:16, 5.74s/it]
25%|██▍ | 10901/43939 [17:34:22<52:34:04, 5.73s/it]
25%|██▍ | 10902/43939 [17:34:28<52:25:25, 5.71s/it]
25%|██▍ | 10903/43939 [17:34:33<52:29:49, 5.72s/it]
25%|██▍ | 10904/43939 [17:34:39<52:22:09, 5.71s/it]
25%|██▍ | 10905/43939 [17:34:45<52:16:30, 5.70s/it]
25%|██▍ | 10906/43939 [17:34:50<52:13:25, 5.69s/it]
25%|██▍ | 10907/43939 [17:34:56<52:15:31, 5.70s/it]
25%|██▍ | 10908/43939 [17:35:02<52:22:02, 5.71s/it]
25%|██▍ | 10909/43939 [17:35:08<52:23:43, 5.71s/it]
25%|██▍ | 10910/43939 [17:35:13<52:22:54, 5.71s/it]
25%|██▍ | 10911/43939 [17:35:19<52:18:52, 5.70s/it]
25%|██▍ | 10912/43939 [17:35:25<52:17:04, 5.70s/it]
25%|██▍ | 10913/43939 [17:35:30<52:19:00, 5.70s/it]
25%|██▍ | 10914/43939 [17:35:36<52:15:32, 5.70s/it]
25%|██▍ | 10915/43939 [17:35:42<52:15:34, 5.70s/it]
25%|██▍ | 10916/43939 [17:35:48<52:18:25, 5.70s/it]
25%|██▍ | 10917/43939 [17:35:53<52:53:16, 5.77s/it]
25%|██▍ | 10918/43939 [17:35:59<52:42:56, 5.75s/it]
25%|██▍ | 10919/43939 [17:36:05<52:38:00, 5.74s/it]
25%|██▍ | 10920/43939 [17:36:11<52:31:50, 5.73s/it]
25%|██▍ | 10921/43939 [17:36:16<52:28:03, 5.72s/it]
25%|██▍ | 10922/43939 [17:36:22<52:26:03, 5.72s/it]
25%|██▍ | 10923/43939 [17:36:28<52:18:16, 5.70s/it]
25%|██▍ | 10924/43939 [17:36:33<52:19:13, 5.71s/it]
25%|██▍ | 10925/43939 [17:36:39<52:25:44, 5.72s/it]
25%|██▍ | 10926/43939 [17:36:45<52:23:10, 5.71s/it]
25%|██▍ | 10927/43939 [17:36:51<52:20:27, 5.71s/it]
25%|██▍ | 10928/43939 [17:36:56<52:18:38, 5.70s/it]
25%|██▍ | 10929/43939 [17:37:02<52:14:34, 5.70s/it]
25%|██▍ | 10930/43939 [17:37:08<52:18:01, 5.70s/it]
25%|██▍ | 10931/43939 [17:37:13<52:20:28, 5.71s/it]
25%|██▍ | 10932/43939 [17:37:19<52:17:04, 5.70s/it]
25%|██▍ | 10933/43939 [17:37:25<52:11:56, 5.69s/it]
25%|██▍ | 10934/43939 [17:37:30<52:12:40, 5.69s/it]
25%|██▍ | 10935/43939 [17:37:36<52:17:40, 5.70s/it]
25%|██▍ | 10936/43939 [17:37:42<52:19:40, 5.71s/it]
25%|██▍ | 10937/43939 [17:37:48<52:22:36, 5.71s/it]
25%|██▍ | 10938/43939 [17:37:53<52:55:56, 5.77s/it]
25%|██▍ | 10939/43939 [17:37:59<52:47:11, 5.76s/it]
25%|██▍ | 10940/43939 [17:38:05<52:33:34, 5.73s/it]
25%|██▍ | 10941/43939 [17:38:11<52:32:37, 5.73s/it]
25%|██▍ | 10942/43939 [17:38:16<52:32:05, 5.73s/it]
25%|██▍ | 10943/43939 [17:38:22<52:24:32, 5.72s/it]
25%|██▍ | 10944/43939 [17:38:28<52:19:56, 5.71s/it]
25%|██▍ | 10945/43939 [17:38:33<52:15:53, 5.70s/it]
25%|██▍ | 10946/43939 [17:38:39<52:21:21, 5.71s/it]
25%|██▍ | 10947/43939 [17:38:45<52:20:04, 5.71s/it]
25%|██▍ | 10948/43939 [17:38:51<52:20:08, 5.71s/it]
25%|██▍ | 10949/43939 [17:38:56<52:16:21, 5.70s/it]
25%|██▍ | 10950/43939 [17:39:02<52:15:18, 5.70s/it]
{'loss': 1.0041, 'grad_norm': 1.1054545640945435, 'learning_rate': 4.668530700654262e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.33, 'epoch': 1.0}
25%|██▍ | 10950/43939 [17:39:02<52:15:18, 5.70s/it]
25%|██▍ | 10951/43939 [17:39:08<52:19:51, 5.71s/it]
25%|██▍ | 10952/43939 [17:39:13<52:21:52, 5.71s/it]
25%|██▍ | 10953/43939 [17:39:19<52:12:59, 5.70s/it]
25%|██▍ | 10954/43939 [17:39:25<52:16:30, 5.71s/it]
25%|██▍ | 10955/43939 [17:39:30<52:15:34, 5.70s/it]
25%|██▍ | 10956/43939 [17:39:36<52:16:24, 5.71s/it]
25%|██▍ | 10957/43939 [17:39:42<52:17:05, 5.71s/it]
25%|██▍ | 10958/43939 [17:39:48<52:10:02, 5.69s/it]
25%|██▍ | 10959/43939 [17:39:53<52:08:26, 5.69s/it]
25%|██▍ | 10960/43939 [17:39:59<52:45:31, 5.76s/it]
25%|██▍ | 10961/43939 [17:40:05<52:34:08, 5.74s/it]
25%|██▍ | 10962/43939 [17:40:11<52:23:40, 5.72s/it]
25%|██▍ | 10963/43939 [17:40:16<52:17:04, 5.71s/it]
25%|██▍ | 10964/43939 [17:40:22<52:15:49, 5.71s/it]
25%|██▍ | 10965/43939 [17:40:28<52:20:18, 5.71s/it]
25%|██▍ | 10966/43939 [17:40:33<52:18:51, 5.71s/it]
25%|██▍ | 10967/43939 [17:40:39<52:19:18, 5.71s/it]
25%|██▍ | 10968/43939 [17:40:45<52:25:53, 5.72s/it]
25%|██▍ | 10969/43939 [17:40:50<52:19:58, 5.71s/it]
25%|██▍ | 10970/43939 [17:40:56<52:18:51, 5.71s/it]
25%|██▍ | 10971/43939 [17:41:02<52:16:45, 5.71s/it]
25%|██▍ | 10972/43939 [17:41:08<52:12:39, 5.70s/it]
25%|██▍ | 10973/43939 [17:41:13<52:14:16, 5.70s/it]
25%|██▍ | 10974/43939 [17:41:19<52:07:15, 5.69s/it]
25%|██▍ | 10975/43939 [17:41:25<52:10:42, 5.70s/it]
25%|██▍ | 10976/43939 [17:41:30<52:11:49, 5.70s/it]
25%|██▍ | 10977/43939 [17:41:36<52:05:34, 5.69s/it]
25%|██▍ | 10978/43939 [17:41:42<52:17:07, 5.71s/it]
25%|██▍ | 10979/43939 [17:41:48<52:18:41, 5.71s/it]
25%|██▍ | 10980/43939 [17:41:53<52:46:06, 5.76s/it]
25%|██▍ | 10981/43939 [17:41:59<52:29:43, 5.73s/it]
25%|██▍ | 10982/43939 [17:42:05<52:26:39, 5.73s/it]
25%|██▍ | 10983/43939 [17:42:11<52:38:55, 5.75s/it]
25%|██▍ | 10984/43939 [17:42:16<52:31:28, 5.74s/it]
25%|██▌ | 10985/43939 [17:42:20<47:33:02, 5.19s/it][2025-11-29 11:46:34,039] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.69s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.72s/it][A
19%|█▉ | 14/72 [00:35<02:38, 2.73s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.74s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:11<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:32<01:41, 2.74s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:36, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.74s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.74s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.74s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.74s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:22<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.74s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.74s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9990127086639404, 'eval_runtime': 196.2105, 'eval_samples_per_second': 5.091, 'eval_steps_per_second': 0.367, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 1.0}
25%|██▌ | 10985/43939 [17:45:36<47:33:02, 5.19s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
25%|██▌ | 10986/43939 [17:45:43<589:30:32, 64.40s/it][2025-11-29 11:49:56,593] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-10986
25%|██▌ | 10987/43939 [17:45:50<433:37:57, 47.37s/it]
25%|██▌ | 10988/43939 [17:45:56<319:05:08, 34.86s/it]
25%|██▌ | 10989/43939 [17:46:02<239:02:38, 26.12s/it]
25%|██▌ | 10990/43939 [17:46:08<183:09:45, 20.01s/it]
25%|██▌ | 10991/43939 [17:46:13<143:55:41, 15.73s/it]
25%|██▌ | 10992/43939 [17:46:19<116:33:12, 12.74s/it]
25%|██▌ | 10993/43939 [17:46:25<97:07:54, 10.61s/it]
25%|██▌ | 10994/43939 [17:46:30<83:44:10, 9.15s/it]
25%|██▌ | 10995/43939 [17:46:36<74:17:21, 8.12s/it]
25%|██▌ | 10996/43939 [17:46:42<67:39:12, 7.39s/it]
25%|██▌ | 10997/43939 [17:46:48<63:05:59, 6.90s/it]
25%|██▌ | 10998/43939 [17:46:53<59:44:15, 6.53s/it]
25%|██▌ | 10999/43939 [17:46:59<57:30:29, 6.29s/it]
25%|██▌ | 11000/43939 [17:47:05<55:52:59, 6.11s/it]
{'loss': 0.9915, 'grad_norm': 1.1925528049468994, 'learning_rate': 4.663572444962643e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 156.98, 'epoch': 1.0}
25%|██▌ | 11000/43939 [17:47:05<55:52:59, 6.11s/it]
25%|██▌ | 11001/43939 [17:47:11<55:34:55, 6.07s/it]
25%|██▌ | 11002/43939 [17:47:16<54:39:39, 5.97s/it]
25%|██▌ | 11003/43939 [17:47:22<53:57:29, 5.90s/it]
25%|██▌ | 11004/43939 [17:47:28<53:30:50, 5.85s/it]
25%|██▌ | 11005/43939 [17:47:34<53:06:39, 5.81s/it]
25%|██▌ | 11006/43939 [17:47:39<52:50:52, 5.78s/it]
25%|██▌ | 11007/43939 [17:47:45<52:39:47, 5.76s/it]
25%|██▌ | 11008/43939 [17:47:51<52:34:57, 5.75s/it]
25%|██▌ | 11009/43939 [17:47:56<52:25:31, 5.73s/it]
25%|██▌ | 11010/43939 [17:48:02<52:17:39, 5.72s/it]
25%|██▌ | 11011/43939 [17:48:08<52:13:20, 5.71s/it]
25%|██▌ | 11012/43939 [17:48:13<52:09:23, 5.70s/it]
25%|██▌ | 11013/43939 [17:48:19<52:11:46, 5.71s/it]
25%|██▌ | 11014/43939 [17:48:25<52:12:51, 5.71s/it]
25%|██▌ | 11015/43939 [17:48:31<52:12:18, 5.71s/it]
25%|██▌ | 11016/43939 [17:48:36<52:07:25, 5.70s/it]
25%|██▌ | 11017/43939 [17:48:42<52:09:12, 5.70s/it]
25%|██▌ | 11018/43939 [17:48:48<52:13:16, 5.71s/it]
25%|██▌ | 11019/43939 [17:48:53<52:12:11, 5.71s/it]
25%|██▌ | 11020/43939 [17:48:59<52:13:00, 5.71s/it]
25%|██▌ | 11021/43939 [17:49:05<52:07:28, 5.70s/it]
25%|██▌ | 11022/43939 [17:49:11<52:42:38, 5.76s/it]
25%|██▌ | 11023/43939 [17:49:16<52:31:15, 5.74s/it]
25%|██▌ | 11024/43939 [17:49:22<52:16:55, 5.72s/it]
25%|██▌ | 11025/43939 [17:49:28<52:15:57, 5.72s/it]
25%|██▌ | 11026/43939 [17:49:34<52:17:19, 5.72s/it]
25%|██▌ | 11027/43939 [17:49:39<52:12:26, 5.71s/it]
25%|██▌ | 11028/43939 [17:49:45<52:18:39, 5.72s/it]
25%|██▌ | 11029/43939 [17:49:51<52:24:14, 5.73s/it]
25%|██▌ | 11030/43939 [17:49:56<52:16:51, 5.72s/it]
25%|██▌ | 11031/43939 [17:50:02<52:13:13, 5.71s/it]
25%|██▌ | 11032/43939 [17:50:08<52:10:50, 5.71s/it]
25%|██▌ | 11033/43939 [17:50:14<52:12:31, 5.71s/it]
25%|██▌ | 11034/43939 [17:50:19<52:10:44, 5.71s/it]
25%|██▌ | 11035/43939 [17:50:25<52:14:43, 5.72s/it]
25%|██▌ | 11036/43939 [17:50:31<52:18:31, 5.72s/it]
25%|██▌ | 11037/43939 [17:50:36<52:25:01, 5.74s/it]
25%|██▌ | 11038/43939 [17:50:42<52:22:10, 5.73s/it]
25%|██▌ | 11039/43939 [17:50:48<52:12:04, 5.71s/it]
25%|██▌ | 11040/43939 [17:50:54<52:14:04, 5.72s/it]
25%|██▌ | 11041/43939 [17:50:59<52:15:41, 5.72s/it]
25%|██▌ | 11042/43939 [17:51:05<52:15:04, 5.72s/it]
25%|██▌ | 11043/43939 [17:51:11<52:54:55, 5.79s/it]
25%|██▌ | 11044/43939 [17:51:17<52:35:02, 5.75s/it]
25%|██▌ | 11045/43939 [17:51:22<52:21:55, 5.73s/it]
25%|██▌ | 11046/43939 [17:51:28<52:09:41, 5.71s/it]
25%|██▌ | 11047/43939 [17:51:34<52:07:46, 5.71s/it]
25%|██▌ | 11048/43939 [17:51:39<52:08:27, 5.71s/it]
25%|██▌ | 11049/43939 [17:51:45<52:10:14, 5.71s/it]
25%|██▌ | 11050/43939 [17:51:51<52:05:19, 5.70s/it]
{'loss': 0.9236, 'grad_norm': 0.992103099822998, 'learning_rate': 4.6585800538313775e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.01, 'epoch': 1.01}
25%|██▌ | 11050/43939 [17:51:51<52:05:19, 5.70s/it]
25%|██▌ | 11051/43939 [17:51:57<52:06:40, 5.70s/it]
25%|██▌ | 11052/43939 [17:52:02<52:06:29, 5.70s/it]
25%|██▌ | 11053/43939 [17:52:08<52:07:14, 5.71s/it]
25%|██▌ | 11054/43939 [17:52:14<52:09:18, 5.71s/it]
25%|██▌ | 11055/43939 [17:52:19<52:11:07, 5.71s/it]
25%|██▌ | 11056/43939 [17:52:25<52:13:26, 5.72s/it]
25%|██▌ | 11057/43939 [17:52:31<52:08:04, 5.71s/it]
25%|██▌ | 11058/43939 [17:52:37<52:14:41, 5.72s/it]
25%|██▌ | 11059/43939 [17:52:42<52:08:30, 5.71s/it]
25%|██▌ | 11060/43939 [17:52:48<52:16:36, 5.72s/it]
25%|██▌ | 11061/43939 [17:52:54<52:18:16, 5.73s/it]
25%|██▌ | 11062/43939 [17:52:59<52:18:47, 5.73s/it]
25%|██▌ | 11063/43939 [17:53:05<52:08:01, 5.71s/it]
25%|██▌ | 11064/43939 [17:53:11<52:49:28, 5.78s/it]
25%|██▌ | 11065/43939 [17:53:17<52:36:29, 5.76s/it]
25%|██▌ | 11066/43939 [17:53:22<52:30:14, 5.75s/it]
25%|██▌ | 11067/43939 [17:53:28<52:23:44, 5.74s/it]
25%|██▌ | 11068/43939 [17:53:34<52:17:53, 5.73s/it]
25%|██▌ | 11069/43939 [17:53:40<52:14:29, 5.72s/it]
25%|██▌ | 11070/43939 [17:53:45<52:15:11, 5.72s/it]
25%|██▌ | 11071/43939 [17:53:51<52:13:47, 5.72s/it]
25%|██▌ | 11072/43939 [17:53:57<52:12:21, 5.72s/it]
25%|██▌ | 11073/43939 [17:54:02<52:05:21, 5.71s/it]
25%|██▌ | 11074/43939 [17:54:08<52:03:55, 5.70s/it]
25%|██▌ | 11075/43939 [17:54:14<52:08:23, 5.71s/it]
25%|██▌ | 11076/43939 [17:54:20<52:05:31, 5.71s/it]
25%|██▌ | 11077/43939 [17:54:25<51:57:27, 5.69s/it]
25%|██▌ | 11078/43939 [17:54:31<52:02:52, 5.70s/it]
25%|██▌ | 11079/43939 [17:54:37<52:02:53, 5.70s/it]
25%|██▌ | 11080/43939 [17:54:42<52:06:53, 5.71s/it]
25%|██▌ | 11081/43939 [17:54:48<52:03:22, 5.70s/it]
25%|██▌ | 11082/43939 [17:54:54<52:10:11, 5.72s/it]
25%|██▌ | 11083/43939 [17:54:59<52:05:28, 5.71s/it]
25%|██▌ | 11084/43939 [17:55:05<52:41:34, 5.77s/it]
25%|██▌ | 11085/43939 [17:55:11<52:24:37, 5.74s/it]
25%|██▌ | 11086/43939 [17:55:17<52:12:05, 5.72s/it]
25%|██▌ | 11087/43939 [17:55:22<52:09:14, 5.72s/it]
25%|██▌ | 11088/43939 [17:55:28<52:05:46, 5.71s/it]
25%|██▌ | 11089/43939 [17:55:34<52:01:04, 5.70s/it]
25%|██▌ | 11090/43939 [17:55:40<51:58:27, 5.70s/it]
25%|██▌ | 11091/43939 [17:55:45<52:01:38, 5.70s/it]
25%|██▌ | 11092/43939 [17:55:51<51:58:30, 5.70s/it]
25%|██▌ | 11093/43939 [17:55:57<51:58:36, 5.70s/it]
25%|██▌ | 11094/43939 [17:56:02<51:57:20, 5.69s/it]
25%|██▌ | 11095/43939 [17:56:08<52:02:08, 5.70s/it]
25%|██▌ | 11096/43939 [17:56:14<51:57:38, 5.70s/it]
25%|██▌ | 11097/43939 [17:56:19<51:55:07, 5.69s/it]
25%|██▌ | 11098/43939 [17:56:25<51:55:32, 5.69s/it]
25%|██▌ | 11099/43939 [17:56:31<51:53:45, 5.69s/it]
25%|██▌ | 11100/43939 [17:56:36<51:57:17, 5.70s/it]
{'loss': 0.9327, 'grad_norm': 1.2271318435668945, 'learning_rate': 4.65355360602717e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.63, 'epoch': 1.01}
25%|██▌ | 11100/43939 [17:56:36<51:57:17, 5.70s/it]
25%|██▌ | 11101/43939 [17:56:42<51:55:18, 5.69s/it]
25%|██▌ | 11102/43939 [17:56:48<51:52:47, 5.69s/it]
25%|██▌ | 11103/43939 [17:56:54<51:57:53, 5.70s/it]
25%|██▌ | 11104/43939 [17:56:59<51:54:33, 5.69s/it]
25%|██▌ | 11105/43939 [17:57:05<52:29:47, 5.76s/it]
25%|██▌ | 11106/43939 [17:57:11<52:26:19, 5.75s/it]
25%|██▌ | 11107/43939 [17:57:17<52:21:22, 5.74s/it]
25%|██▌ | 11108/43939 [17:57:22<52:11:22, 5.72s/it]
25%|██▌ | 11109/43939 [17:57:28<52:15:37, 5.73s/it]
25%|██▌ | 11110/43939 [17:57:34<52:08:32, 5.72s/it]
25%|██▌ | 11111/43939 [17:57:39<52:05:52, 5.71s/it]
25%|██▌ | 11112/43939 [17:57:45<52:04:45, 5.71s/it]
25%|██▌ | 11113/43939 [17:57:51<52:01:29, 5.71s/it]
25%|██▌ | 11114/43939 [17:57:57<51:59:32, 5.70s/it]
25%|██▌ | 11115/43939 [17:58:02<51:58:10, 5.70s/it]
25%|██▌ | 11116/43939 [17:58:08<52:02:22, 5.71s/it]
25%|██▌ | 11117/43939 [17:58:14<52:15:16, 5.73s/it]
25%|██▌ | 11118/43939 [17:58:19<52:08:58, 5.72s/it]
25%|██▌ | 11119/43939 [17:58:25<52:04:42, 5.71s/it]
25%|██▌ | 11120/43939 [17:58:31<52:04:53, 5.71s/it]
25%|██▌ | 11121/43939 [17:58:37<52:06:39, 5.72s/it]
25%|██▌ | 11122/43939 [17:58:42<52:11:46, 5.73s/it]
25%|██▌ | 11123/43939 [17:58:48<52:08:09, 5.72s/it]
25%|██▌ | 11124/43939 [17:58:54<51:59:53, 5.70s/it]
25%|██▌ | 11125/43939 [17:58:59<51:54:50, 5.70s/it]
25%|██▌ | 11126/43939 [17:59:05<51:54:42, 5.70s/it]
25%|██▌ | 11127/43939 [17:59:11<52:36:12, 5.77s/it]
25%|██▌ | 11128/43939 [17:59:17<52:25:14, 5.75s/it]
25%|██▌ | 11129/43939 [17:59:22<52:14:05, 5.73s/it]
25%|██▌ | 11130/43939 [17:59:28<52:11:53, 5.73s/it]
25%|██▌ | 11131/43939 [17:59:34<52:01:23, 5.71s/it]
25%|██▌ | 11132/43939 [17:59:39<51:56:28, 5.70s/it]
25%|██▌ | 11133/43939 [17:59:45<51:55:24, 5.70s/it]
25%|██▌ | 11134/43939 [17:59:51<51:54:40, 5.70s/it]
25%|██▌ | 11135/43939 [17:59:57<52:00:22, 5.71s/it]
25%|██▌ | 11136/43939 [18:00:02<52:04:51, 5.72s/it]
25%|██▌ | 11137/43939 [18:00:08<52:09:29, 5.72s/it]
25%|██▌ | 11138/43939 [18:00:14<52:03:51, 5.71s/it]
25%|██▌ | 11139/43939 [18:00:19<52:07:21, 5.72s/it]
25%|██▌ | 11140/43939 [18:00:25<52:06:29, 5.72s/it]
25%|██▌ | 11141/43939 [18:00:31<52:01:06, 5.71s/it]
25%|██▌ | 11142/43939 [18:00:37<52:00:20, 5.71s/it]
25%|██▌ | 11143/43939 [18:00:42<51:57:43, 5.70s/it]
25%|██▌ | 11144/43939 [18:00:48<52:00:42, 5.71s/it]
25%|██▌ | 11145/43939 [18:00:54<51:56:19, 5.70s/it]
25%|██▌ | 11146/43939 [18:00:59<51:55:22, 5.70s/it]
25%|██▌ | 11147/43939 [18:01:05<52:28:35, 5.76s/it]
25%|██▌ | 11148/43939 [18:01:11<52:21:03, 5.75s/it]
25%|██▌ | 11149/43939 [18:01:17<52:17:40, 5.74s/it]
25%|██▌ | 11150/43939 [18:01:22<52:16:29, 5.74s/it]
{'loss': 0.899, 'grad_norm': 1.0395036935806274, 'learning_rate': 4.648493180854047e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 449.05, 'epoch': 1.02}
25%|██▌ | 11150/43939 [18:01:22<52:16:29, 5.74s/it]
25%|██▌ | 11151/43939 [18:01:28<52:22:51, 5.75s/it]
25%|██▌ | 11152/43939 [18:01:34<52:21:39, 5.75s/it]
25%|██▌ | 11153/43939 [18:01:40<52:11:02, 5.73s/it]
25%|██▌ | 11154/43939 [18:01:45<52:15:08, 5.74s/it]
25%|██▌ | 11155/43939 [18:01:51<52:07:37, 5.72s/it]
25%|██▌ | 11156/43939 [18:01:57<52:03:56, 5.72s/it]
25%|██▌ | 11157/43939 [18:02:03<52:05:18, 5.72s/it]
25%|██▌ | 11158/43939 [18:02:08<51:53:44, 5.70s/it]
25%|██▌ | 11159/43939 [18:02:14<51:50:25, 5.69s/it]
25%|██▌ | 11160/43939 [18:02:20<51:47:07, 5.69s/it]
25%|██▌ | 11161/43939 [18:02:25<51:43:17, 5.68s/it]
25%|██▌ | 11162/43939 [18:02:31<51:51:58, 5.70s/it]
25%|██▌ | 11163/43939 [18:02:37<52:02:01, 5.72s/it]
25%|██▌ | 11164/43939 [18:02:42<51:56:45, 5.71s/it]
25%|██▌ | 11165/43939 [18:02:48<51:51:54, 5.70s/it]
25%|██▌ | 11166/43939 [18:02:54<51:54:18, 5.70s/it]
25%|██▌ | 11167/43939 [18:02:59<51:54:41, 5.70s/it]
25%|██▌ | 11168/43939 [18:03:05<52:34:07, 5.77s/it]
25%|██▌ | 11169/43939 [18:03:11<52:20:43, 5.75s/it]
25%|██▌ | 11170/43939 [18:03:17<52:21:30, 5.75s/it]
25%|██▌ | 11171/43939 [18:03:23<52:13:36, 5.74s/it]
25%|██▌ | 11172/43939 [18:03:28<52:06:43, 5.73s/it]
25%|██▌ | 11173/43939 [18:03:34<52:07:43, 5.73s/it]
25%|██▌ | 11174/43939 [18:03:40<52:04:25, 5.72s/it]
25%|██▌ | 11175/43939 [18:03:45<52:13:53, 5.74s/it]
25%|██▌ | 11176/43939 [18:03:51<52:02:51, 5.72s/it]
25%|██▌ | 11177/43939 [18:03:57<52:11:18, 5.73s/it]
25%|██▌ | 11178/43939 [18:04:03<52:12:00, 5.74s/it]
25%|██▌ | 11179/43939 [18:04:08<52:06:56, 5.73s/it]
25%|██▌ | 11180/43939 [18:04:14<51:59:24, 5.71s/it]
25%|██▌ | 11181/43939 [18:04:20<51:52:03, 5.70s/it]
25%|██▌ | 11182/43939 [18:04:25<51:43:35, 5.68s/it]
25%|██▌ | 11183/43939 [18:04:31<51:50:44, 5.70s/it]
25%|██▌ | 11184/43939 [18:04:37<51:48:08, 5.69s/it]
25%|██▌ | 11185/43939 [18:04:42<51:50:27, 5.70s/it]
25%|██▌ | 11186/43939 [18:04:48<51:49:57, 5.70s/it]
25%|██▌ | 11187/43939 [18:04:54<51:48:44, 5.70s/it]
25%|██▌ | 11188/43939 [18:05:00<51:47:34, 5.69s/it]
25%|██▌ | 11189/43939 [18:05:05<51:50:37, 5.70s/it]
25%|██▌ | 11190/43939 [18:05:11<52:26:31, 5.76s/it]
25%|██▌ | 11191/43939 [18:05:17<52:18:21, 5.75s/it]
25%|██▌ | 11192/43939 [18:05:23<52:08:58, 5.73s/it]
25%|██▌ | 11193/43939 [18:05:28<52:06:08, 5.73s/it]
25%|██▌ | 11194/43939 [18:05:34<52:05:22, 5.73s/it]
25%|██▌ | 11195/43939 [18:05:40<52:02:10, 5.72s/it]
25%|██▌ | 11196/43939 [18:05:45<51:55:58, 5.71s/it]
25%|██▌ | 11197/43939 [18:05:51<51:55:34, 5.71s/it]
25%|██▌ | 11198/43939 [18:05:57<51:49:24, 5.70s/it]
25%|██▌ | 11199/43939 [18:06:03<51:49:03, 5.70s/it]
25%|██▌ | 11200/43939 [18:06:08<51:44:15, 5.69s/it]
{'loss': 0.937, 'grad_norm': 1.3653130531311035, 'learning_rate': 4.643398858152105e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.45, 'epoch': 1.02}
25%|██▌ | 11200/43939 [18:06:08<51:44:15, 5.69s/it]
25%|██▌ | 11201/43939 [18:06:14<51:48:38, 5.70s/it]
25%|██▌ | 11202/43939 [18:06:20<51:52:13, 5.70s/it]
25%|██▌ | 11203/43939 [18:06:25<51:49:03, 5.70s/it]
25%|██▌ | 11204/43939 [18:06:31<51:50:22, 5.70s/it]
26%|██▌ | 11205/43939 [18:06:37<51:57:32, 5.71s/it]
26%|██▌ | 11206/43939 [18:06:42<51:54:11, 5.71s/it]
26%|██▌ | 11207/43939 [18:06:48<51:53:01, 5.71s/it]
26%|██▌ | 11208/43939 [18:06:54<51:51:35, 5.70s/it]
26%|██▌ | 11209/43939 [18:07:00<51:45:05, 5.69s/it]
26%|██▌ | 11210/43939 [18:07:05<51:44:26, 5.69s/it]
26%|██▌ | 11211/43939 [18:07:11<52:23:43, 5.76s/it]
26%|██▌ | 11212/43939 [18:07:17<52:11:20, 5.74s/it]
26%|██▌ | 11213/43939 [18:07:23<52:02:14, 5.72s/it]
26%|██▌ | 11214/43939 [18:07:28<52:07:28, 5.73s/it]
26%|██▌ | 11215/43939 [18:07:34<51:57:53, 5.72s/it]
26%|██▌ | 11216/43939 [18:07:40<52:01:13, 5.72s/it]
26%|██▌ | 11217/43939 [18:07:45<51:53:47, 5.71s/it]
26%|██▌ | 11218/43939 [18:07:51<51:47:05, 5.70s/it]
26%|██▌ | 11219/43939 [18:07:57<51:48:14, 5.70s/it]
26%|██▌ | 11220/43939 [18:08:02<51:44:10, 5.69s/it]
26%|██▌ | 11221/43939 [18:08:08<51:48:52, 5.70s/it]
26%|██▌ | 11222/43939 [18:08:14<51:52:16, 5.71s/it]
26%|██▌ | 11223/43939 [18:08:20<51:48:31, 5.70s/it]
26%|██▌ | 11224/43939 [18:08:25<51:45:54, 5.70s/it]
26%|██▌ | 11225/43939 [18:08:31<51:50:40, 5.71s/it]
26%|██▌ | 11226/43939 [18:08:37<51:47:53, 5.70s/it]
26%|██▌ | 11227/43939 [18:08:42<51:41:03, 5.69s/it]
26%|██▌ | 11228/43939 [18:08:48<51:42:58, 5.69s/it]
26%|██▌ | 11229/43939 [18:08:54<51:48:04, 5.70s/it]
26%|██▌ | 11230/43939 [18:08:59<51:50:49, 5.71s/it]
26%|██▌ | 11231/43939 [18:09:05<51:48:35, 5.70s/it]
26%|██▌ | 11232/43939 [18:09:11<52:30:37, 5.78s/it]
26%|██▌ | 11233/43939 [18:09:17<52:22:32, 5.77s/it]
26%|██▌ | 11234/43939 [18:09:23<52:16:40, 5.75s/it]
26%|██▌ | 11235/43939 [18:09:28<52:02:18, 5.73s/it]
26%|██▌ | 11236/43939 [18:09:34<51:51:37, 5.71s/it]
26%|██▌ | 11237/43939 [18:09:40<51:47:12, 5.70s/it]
26%|██▌ | 11238/43939 [18:09:45<51:50:53, 5.71s/it]
26%|██▌ | 11239/43939 [18:09:51<51:58:11, 5.72s/it]
26%|██▌ | 11240/43939 [18:09:57<51:52:13, 5.71s/it]
26%|██▌ | 11241/43939 [18:10:02<51:56:59, 5.72s/it]
26%|██▌ | 11242/43939 [18:10:08<52:00:35, 5.73s/it]
26%|██▌ | 11243/43939 [18:10:14<51:59:47, 5.73s/it]
26%|██▌ | 11244/43939 [18:10:20<52:00:00, 5.73s/it]
26%|██▌ | 11245/43939 [18:10:25<51:55:25, 5.72s/it]
26%|██▌ | 11246/43939 [18:10:31<51:54:48, 5.72s/it]
26%|██▌ | 11247/43939 [18:10:37<51:58:21, 5.72s/it]
26%|██▌ | 11248/43939 [18:10:43<51:53:48, 5.71s/it]
26%|██▌ | 11249/43939 [18:10:48<51:45:27, 5.70s/it]
26%|██▌ | 11250/43939 [18:10:54<51:45:16, 5.70s/it]
{'loss': 0.9242, 'grad_norm': 1.3725383281707764, 'learning_rate': 4.638270718296258e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.43, 'epoch': 1.02}
26%|██▌ | 11250/43939 [18:10:54<51:45:16, 5.70s/it]
26%|██▌ | 11251/43939 [18:11:00<51:44:58, 5.70s/it]
26%|██▌ | 11252/43939 [18:11:05<51:45:16, 5.70s/it]
26%|██▌ | 11253/43939 [18:11:11<51:47:51, 5.70s/it]
26%|██▌ | 11254/43939 [18:11:17<52:32:34, 5.79s/it]
26%|██▌ | 11255/43939 [18:11:23<52:19:47, 5.76s/it]
26%|██▌ | 11256/43939 [18:11:28<52:08:18, 5.74s/it]
26%|██▌ | 11257/43939 [18:11:34<52:01:21, 5.73s/it]
26%|██▌ | 11258/43939 [18:11:40<51:51:47, 5.71s/it]
26%|██▌ | 11259/43939 [18:11:45<51:43:52, 5.70s/it]
26%|██▌ | 11260/43939 [18:11:51<51:47:45, 5.71s/it]
26%|██▌ | 11261/43939 [18:11:57<51:46:49, 5.70s/it]
26%|██▌ | 11262/43939 [18:12:03<51:51:03, 5.71s/it]
26%|██▌ | 11263/43939 [18:12:08<51:51:38, 5.71s/it]
26%|██▌ | 11264/43939 [18:12:14<51:50:57, 5.71s/it]
26%|██▌ | 11265/43939 [18:12:20<51:45:51, 5.70s/it]
26%|██▌ | 11266/43939 [18:12:25<51:43:00, 5.70s/it]
26%|██▌ | 11267/43939 [18:12:31<51:41:18, 5.70s/it]
26%|██▌ | 11268/43939 [18:12:37<51:40:09, 5.69s/it]
26%|██▌ | 11269/43939 [18:12:42<51:40:33, 5.69s/it]
26%|██▌ | 11270/43939 [18:12:48<51:48:22, 5.71s/it]
26%|██▌ | 11271/43939 [18:12:54<51:53:29, 5.72s/it]
26%|██▌ | 11272/43939 [18:13:00<51:49:43, 5.71s/it]
26%|██▌ | 11273/43939 [18:13:05<51:42:33, 5.70s/it]
26%|██▌ | 11274/43939 [18:13:11<51:45:13, 5.70s/it]
26%|██▌ | 11275/43939 [18:13:17<52:20:21, 5.77s/it]
26%|██▌ | 11276/43939 [18:13:23<52:13:41, 5.76s/it]
26%|██▌ | 11277/43939 [18:13:28<52:06:04, 5.74s/it]
26%|██▌ | 11278/43939 [18:13:34<52:03:07, 5.74s/it]
26%|██▌ | 11279/43939 [18:13:40<51:54:37, 5.72s/it]
26%|██▌ | 11280/43939 [18:13:45<51:52:28, 5.72s/it]
26%|██▌ | 11281/43939 [18:13:51<51:49:54, 5.71s/it]
26%|██▌ | 11282/43939 [18:13:57<51:49:39, 5.71s/it]
26%|██▌ | 11283/43939 [18:14:03<51:48:30, 5.71s/it]
26%|██▌ | 11284/43939 [18:14:08<51:50:21, 5.71s/it]
26%|██▌ | 11285/43939 [18:14:14<51:48:24, 5.71s/it]
26%|██▌ | 11286/43939 [18:14:20<51:44:26, 5.70s/it]
26%|██▌ | 11287/43939 [18:14:25<51:49:16, 5.71s/it]
26%|██▌ | 11288/43939 [18:14:31<51:41:58, 5.70s/it]
26%|██▌ | 11289/43939 [18:14:37<51:42:33, 5.70s/it]
26%|██▌ | 11290/43939 [18:14:43<51:41:10, 5.70s/it]
26%|██▌ | 11291/43939 [18:14:48<51:35:04, 5.69s/it]
26%|██▌ | 11292/43939 [18:14:54<51:34:17, 5.69s/it]
26%|██▌ | 11293/43939 [18:15:00<51:32:43, 5.68s/it]
26%|██▌ | 11294/43939 [18:15:05<51:38:23, 5.69s/it]
26%|██▌ | 11295/43939 [18:15:11<51:38:11, 5.69s/it]
26%|██▌ | 11296/43939 [18:15:17<52:12:06, 5.76s/it]
26%|██▌ | 11297/43939 [18:15:23<52:03:36, 5.74s/it]
26%|██▌ | 11298/43939 [18:15:28<52:00:10, 5.74s/it]
26%|██▌ | 11299/43939 [18:15:34<51:53:39, 5.72s/it]
26%|██▌ | 11300/43939 [18:15:40<52:07:51, 5.75s/it]
{'loss': 0.9418, 'grad_norm': 1.0027776956558228, 'learning_rate': 4.633108842194962e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.63, 'epoch': 1.03}
26%|██▌ | 11300/43939 [18:15:40<52:07:51, 5.75s/it]
26%|██▌ | 11301/43939 [18:15:46<52:11:25, 5.76s/it]
26%|██▌ | 11302/43939 [18:15:51<52:04:23, 5.74s/it]
26%|██▌ | 11303/43939 [18:15:57<52:01:52, 5.74s/it]
26%|██▌ | 11304/43939 [18:16:03<51:54:37, 5.73s/it]
26%|██▌ | 11305/43939 [18:16:08<51:55:17, 5.73s/it]
26%|██▌ | 11306/43939 [18:16:14<51:50:14, 5.72s/it]
26%|██▌ | 11307/43939 [18:16:20<51:45:42, 5.71s/it]
26%|██▌ | 11308/43939 [18:16:25<51:37:02, 5.69s/it]
26%|██▌ | 11309/43939 [18:16:31<51:39:50, 5.70s/it]
26%|██▌ | 11310/43939 [18:16:37<51:37:11, 5.70s/it]
26%|██▌ | 11311/43939 [18:16:43<51:39:27, 5.70s/it]
26%|██▌ | 11312/43939 [18:16:48<51:41:03, 5.70s/it]
26%|██▌ | 11313/43939 [18:16:54<51:42:39, 5.71s/it]
26%|██▌ | 11314/43939 [18:17:00<51:40:33, 5.70s/it]
26%|██▌ | 11315/43939 [18:17:05<51:39:31, 5.70s/it]
26%|██▌ | 11316/43939 [18:17:11<52:17:33, 5.77s/it]
26%|██▌ | 11317/43939 [18:17:17<52:05:20, 5.75s/it]
26%|██▌ | 11318/43939 [18:17:23<51:53:24, 5.73s/it]
26%|██▌ | 11319/43939 [18:17:28<51:42:37, 5.71s/it]
26%|██▌ | 11320/43939 [18:17:34<51:40:56, 5.70s/it]
26%|██▌ | 11321/43939 [18:17:40<51:35:29, 5.69s/it]
26%|██▌ | 11322/43939 [18:17:45<51:33:30, 5.69s/it]
26%|██▌ | 11323/43939 [18:17:51<51:39:11, 5.70s/it]
26%|██▌ | 11324/43939 [18:17:57<51:34:03, 5.69s/it]
26%|██▌ | 11325/43939 [18:18:03<51:34:43, 5.69s/it]
26%|██▌ | 11326/43939 [18:18:08<51:33:35, 5.69s/it]
26%|██▌ | 11327/43939 [18:18:14<51:31:17, 5.69s/it]
26%|██▌ | 11328/43939 [18:18:20<51:31:04, 5.69s/it]
26%|██▌ | 11329/43939 [18:18:25<51:33:30, 5.69s/it]
26%|██▌ | 11330/43939 [18:18:31<51:39:51, 5.70s/it]
26%|██▌ | 11331/43939 [18:18:37<51:38:11, 5.70s/it]
26%|██▌ | 11332/43939 [18:18:42<51:44:43, 5.71s/it]
26%|██▌ | 11333/43939 [18:18:48<51:43:41, 5.71s/it]
26%|██▌ | 11334/43939 [18:18:54<51:38:19, 5.70s/it]
26%|██▌ | 11335/43939 [18:19:00<51:36:59, 5.70s/it]
26%|██▌ | 11336/43939 [18:19:05<51:36:42, 5.70s/it]
26%|██▌ | 11337/43939 [18:19:11<51:42:57, 5.71s/it]
26%|██▌ | 11338/43939 [18:19:17<52:18:29, 5.78s/it]
26%|██▌ | 11339/43939 [18:19:23<52:09:27, 5.76s/it]
26%|██▌ | 11340/43939 [18:19:28<51:58:36, 5.74s/it]
26%|██▌ | 11341/43939 [18:19:34<51:55:17, 5.73s/it]
26%|██▌ | 11342/43939 [18:19:40<51:48:58, 5.72s/it]
26%|██▌ | 11343/43939 [18:19:45<51:43:53, 5.71s/it]
26%|██▌ | 11344/43939 [18:19:51<51:43:14, 5.71s/it]
26%|██▌ | 11345/43939 [18:19:57<51:37:37, 5.70s/it]
26%|██▌ | 11346/43939 [18:20:02<51:37:25, 5.70s/it]
26%|██▌ | 11347/43939 [18:20:08<51:39:47, 5.71s/it]
26%|██▌ | 11348/43939 [18:20:14<51:40:11, 5.71s/it]
26%|██▌ | 11349/43939 [18:20:20<51:34:24, 5.70s/it]
26%|██▌ | 11350/43939 [18:20:25<51:37:31, 5.70s/it]
{'loss': 0.9295, 'grad_norm': 1.1604491472244263, 'learning_rate': 4.627913311288944e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.69, 'epoch': 1.03}
26%|██▌ | 11350/43939 [18:20:25<51:37:31, 5.70s/it]
26%|██▌ | 11351/43939 [18:20:31<51:36:38, 5.70s/it]
26%|██▌ | 11352/43939 [18:20:37<51:34:11, 5.70s/it]
26%|██▌ | 11353/43939 [18:20:42<51:39:00, 5.71s/it]
26%|██▌ | 11354/43939 [18:20:48<51:38:14, 5.70s/it]
26%|██▌ | 11355/43939 [18:20:54<51:33:02, 5.70s/it]
26%|██▌ | 11356/43939 [18:21:00<51:35:31, 5.70s/it]
26%|██▌ | 11357/43939 [18:21:05<51:39:37, 5.71s/it]
26%|██▌ | 11358/43939 [18:21:11<51:30:42, 5.69s/it]
26%|██▌ | 11359/43939 [18:21:17<52:14:10, 5.77s/it]
26%|██▌ | 11360/43939 [18:21:23<52:04:48, 5.75s/it]
26%|██▌ | 11361/43939 [18:21:28<51:58:20, 5.74s/it]
26%|██▌ | 11362/43939 [18:21:34<51:52:24, 5.73s/it]
26%|██▌ | 11363/43939 [18:21:40<51:44:20, 5.72s/it]
26%|██▌ | 11364/43939 [18:21:45<51:37:27, 5.71s/it]
26%|██▌ | 11365/43939 [18:21:51<51:39:42, 5.71s/it]
26%|██▌ | 11366/43939 [18:21:57<51:33:43, 5.70s/it]
26%|██▌ | 11367/43939 [18:22:02<51:37:12, 5.71s/it]
26%|██▌ | 11368/43939 [18:22:08<51:36:53, 5.70s/it]
26%|██▌ | 11369/43939 [18:22:14<51:32:10, 5.70s/it]
26%|██▌ | 11370/43939 [18:22:20<51:27:19, 5.69s/it]
26%|██▌ | 11371/43939 [18:22:25<51:31:39, 5.70s/it]
26%|██▌ | 11372/43939 [18:22:31<51:27:56, 5.69s/it]
26%|██▌ | 11373/43939 [18:22:37<51:28:22, 5.69s/it]
26%|██▌ | 11374/43939 [18:22:42<51:28:01, 5.69s/it]
26%|██▌ | 11375/43939 [18:22:48<51:39:37, 5.71s/it]
26%|██▌ | 11376/43939 [18:22:54<51:38:26, 5.71s/it]
26%|██▌ | 11377/43939 [18:22:59<51:34:46, 5.70s/it]
26%|██▌ | 11378/43939 [18:23:05<51:32:24, 5.70s/it]
26%|██▌ | 11379/43939 [18:23:11<51:25:06, 5.69s/it]
26%|██▌ | 11380/43939 [18:23:17<52:02:43, 5.75s/it]
26%|██▌ | 11381/43939 [18:23:22<51:51:47, 5.73s/it]
26%|██▌ | 11382/43939 [18:23:28<51:46:41, 5.73s/it]
26%|██▌ | 11383/43939 [18:23:34<51:46:27, 5.73s/it]
26%|██▌ | 11384/43939 [18:23:39<51:41:52, 5.72s/it]
26%|██▌ | 11385/43939 [18:23:45<51:40:09, 5.71s/it]
26%|██▌ | 11386/43939 [18:23:51<51:39:43, 5.71s/it]
26%|██▌ | 11387/43939 [18:23:57<51:33:42, 5.70s/it]
26%|██▌ | 11388/43939 [18:24:02<51:34:53, 5.70s/it]
26%|██▌ | 11389/43939 [18:24:08<51:31:25, 5.70s/it]
26%|██▌ | 11390/43939 [18:24:14<51:25:06, 5.69s/it]
26%|██▌ | 11391/43939 [18:24:19<51:25:12, 5.69s/it]
26%|██▌ | 11392/43939 [18:24:25<51:26:43, 5.69s/it]
26%|██▌ | 11393/43939 [18:24:31<51:29:35, 5.70s/it]
26%|██▌ | 11394/43939 [18:24:36<51:29:38, 5.70s/it]
26%|██▌ | 11395/43939 [18:24:42<51:30:54, 5.70s/it]
26%|██▌ | 11396/43939 [18:24:48<51:25:31, 5.69s/it]
26%|██▌ | 11397/43939 [18:24:54<51:31:48, 5.70s/it]
26%|██▌ | 11398/43939 [18:24:59<51:26:06, 5.69s/it]
26%|██▌ | 11399/43939 [18:25:05<51:26:53, 5.69s/it]
26%|██▌ | 11400/43939 [18:25:11<51:25:42, 5.69s/it]
{'loss': 0.9296, 'grad_norm': 1.3307735919952393, 'learning_rate': 4.622684207549913e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 448.98, 'epoch': 1.04}
26%|██▌ | 11400/43939 [18:25:11<51:25:42, 5.69s/it]
26%|██▌ | 11401/43939 [18:25:16<51:23:00, 5.69s/it]
26%|██▌ | 11402/43939 [18:25:22<52:00:14, 5.75s/it]
26%|██▌ | 11403/43939 [18:25:28<51:43:22, 5.72s/it]
26%|██▌ | 11404/43939 [18:25:33<51:33:09, 5.70s/it]
26%|██▌ | 11405/43939 [18:25:39<51:28:55, 5.70s/it]
26%|██▌ | 11406/43939 [18:25:45<51:30:53, 5.70s/it]
26%|██▌ | 11407/43939 [18:25:51<51:31:44, 5.70s/it]
26%|██▌ | 11408/43939 [18:25:56<51:34:59, 5.71s/it]
26%|██▌ | 11409/43939 [18:26:02<51:32:21, 5.70s/it]
26%|██▌ | 11410/43939 [18:26:08<51:33:56, 5.71s/it]
26%|██▌ | 11411/43939 [18:26:13<51:29:45, 5.70s/it]
26%|██▌ | 11412/43939 [18:26:19<51:40:14, 5.72s/it]
26%|██▌ | 11413/43939 [18:26:25<51:33:24, 5.71s/it]
26%|██▌ | 11414/43939 [18:26:31<51:32:26, 5.70s/it]
26%|██▌ | 11415/43939 [18:26:36<51:27:07, 5.70s/it]
26%|██▌ | 11416/43939 [18:26:42<51:32:06, 5.70s/it]
26%|██▌ | 11417/43939 [18:26:48<51:31:11, 5.70s/it]
26%|██▌ | 11418/43939 [18:26:53<51:31:00, 5.70s/it]
26%|██▌ | 11419/43939 [18:26:59<51:28:08, 5.70s/it]
26%|██▌ | 11420/43939 [18:27:05<51:43:33, 5.73s/it]
26%|██▌ | 11421/43939 [18:27:11<51:42:27, 5.72s/it]
26%|██▌ | 11422/43939 [18:27:16<51:41:40, 5.72s/it]
26%|██▌ | 11423/43939 [18:27:22<52:07:47, 5.77s/it]
26%|██▌ | 11424/43939 [18:27:28<52:01:14, 5.76s/it]
26%|██▌ | 11425/43939 [18:27:34<51:50:28, 5.74s/it]
26%|██▌ | 11426/43939 [18:27:39<51:59:38, 5.76s/it]
26%|██▌ | 11427/43939 [18:27:45<51:49:17, 5.74s/it]
26%|██▌ | 11428/43939 [18:27:51<51:42:54, 5.73s/it]
26%|██▌ | 11429/43939 [18:27:56<51:43:30, 5.73s/it]
26%|██▌ | 11430/43939 [18:28:02<51:33:00, 5.71s/it]
26%|██▌ | 11431/43939 [18:28:08<51:27:59, 5.70s/it]
26%|██▌ | 11432/43939 [18:28:14<51:28:41, 5.70s/it]
26%|██▌ | 11433/43939 [18:28:19<51:22:22, 5.69s/it]
26%|██▌ | 11434/43939 [18:28:25<51:20:52, 5.69s/it]
26%|██▌ | 11435/43939 [18:28:31<51:23:39, 5.69s/it]
26%|██▌ | 11436/43939 [18:28:36<51:25:30, 5.70s/it]
26%|██▌ | 11437/43939 [18:28:42<51:22:05, 5.69s/it]
26%|██▌ | 11438/43939 [18:28:48<51:35:45, 5.72s/it]
26%|██▌ | 11439/43939 [18:28:53<51:31:53, 5.71s/it]
26%|██▌ | 11440/43939 [18:28:59<51:32:00, 5.71s/it]
26%|██▌ | 11441/43939 [18:29:05<51:32:21, 5.71s/it]
26%|██▌ | 11442/43939 [18:29:11<51:25:19, 5.70s/it]
26%|██▌ | 11443/43939 [18:29:16<51:23:56, 5.69s/it]
26%|██▌ | 11444/43939 [18:29:22<52:08:01, 5.78s/it]
26%|██▌ | 11445/43939 [18:29:28<51:57:49, 5.76s/it]
26%|██▌ | 11446/43939 [18:29:34<51:42:20, 5.73s/it]
26%|██▌ | 11447/43939 [18:29:39<51:33:19, 5.71s/it]
26%|██▌ | 11448/43939 [18:29:45<51:26:29, 5.70s/it]
26%|██▌ | 11449/43939 [18:29:51<51:20:28, 5.69s/it]
26%|██▌ | 11450/43939 [18:29:56<51:17:34, 5.68s/it]
{'loss': 0.9336, 'grad_norm': 1.0178179740905762, 'learning_rate': 4.617421613479268e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.76, 'epoch': 1.04}
26%|██▌ | 11450/43939 [18:29:56<51:17:34, 5.68s/it]
26%|██▌ | 11451/43939 [18:30:02<51:21:55, 5.69s/it]
26%|██▌ | 11452/43939 [18:30:08<51:19:48, 5.69s/it]
26%|██▌ | 11453/43939 [18:30:13<51:21:57, 5.69s/it]
26%|██▌ | 11454/43939 [18:30:19<51:21:25, 5.69s/it]
26%|██▌ | 11455/43939 [18:30:25<51:28:52, 5.71s/it]
26%|██▌ | 11456/43939 [18:30:30<51:27:38, 5.70s/it]
26%|██▌ | 11457/43939 [18:30:36<51:26:45, 5.70s/it]
26%|██▌ | 11458/43939 [18:30:42<51:19:21, 5.69s/it]
26%|██▌ | 11459/43939 [18:30:47<51:17:55, 5.69s/it]
26%|██▌ | 11460/43939 [18:30:53<51:29:58, 5.71s/it]
26%|██▌ | 11461/43939 [18:30:59<51:29:49, 5.71s/it]
26%|██▌ | 11462/43939 [18:31:05<51:37:17, 5.72s/it]
26%|██▌ | 11463/43939 [18:31:10<51:27:19, 5.70s/it]
26%|██▌ | 11464/43939 [18:31:16<51:24:18, 5.70s/it]
26%|██▌ | 11465/43939 [18:31:22<51:59:23, 5.76s/it]
26%|██▌ | 11466/43939 [18:31:28<51:44:52, 5.74s/it]
26%|██▌ | 11467/43939 [18:31:33<51:36:54, 5.72s/it]
26%|██▌ | 11468/43939 [18:31:39<51:32:43, 5.71s/it]
26%|██▌ | 11469/43939 [18:31:45<51:33:03, 5.72s/it]
26%|██▌ | 11470/43939 [18:31:50<51:32:19, 5.71s/it]
26%|██▌ | 11471/43939 [18:31:56<51:34:06, 5.72s/it]
26%|██▌ | 11472/43939 [18:32:02<51:29:46, 5.71s/it]
26%|██▌ | 11473/43939 [18:32:08<51:26:53, 5.70s/it]
26%|██▌ | 11474/43939 [18:32:13<51:23:54, 5.70s/it]
26%|██▌ | 11475/43939 [18:32:19<51:21:36, 5.70s/it]
26%|██▌ | 11476/43939 [18:32:25<51:22:05, 5.70s/it]
26%|██▌ | 11477/43939 [18:32:30<51:21:58, 5.70s/it]
26%|██▌ | 11478/43939 [18:32:36<51:21:01, 5.69s/it]
26%|██▌ | 11479/43939 [18:32:42<51:21:02, 5.70s/it]
26%|██▌ | 11480/43939 [18:32:47<51:15:23, 5.68s/it]
26%|██▌ | 11481/43939 [18:32:53<51:17:37, 5.69s/it]
26%|██▌ | 11482/43939 [18:32:59<51:20:23, 5.69s/it]
26%|██▌ | 11483/43939 [18:33:04<51:23:19, 5.70s/it]
26%|██▌ | 11484/43939 [18:33:10<51:25:15, 5.70s/it]
26%|██▌ | 11485/43939 [18:33:16<51:21:55, 5.70s/it]
26%|██▌ | 11486/43939 [18:33:22<51:51:56, 5.75s/it]
26%|██▌ | 11487/43939 [18:33:27<51:46:18, 5.74s/it]
26%|██▌ | 11488/43939 [18:33:33<51:40:47, 5.73s/it]
26%|██▌ | 11489/43939 [18:33:39<51:34:43, 5.72s/it]
26%|██▌ | 11490/43939 [18:33:45<51:29:48, 5.71s/it]
26%|██▌ | 11491/43939 [18:33:50<51:30:17, 5.71s/it]
26%|██▌ | 11492/43939 [18:33:56<51:35:20, 5.72s/it]
26%|██▌ | 11493/43939 [18:34:02<51:29:01, 5.71s/it]
26%|██▌ | 11494/43939 [18:34:07<51:23:45, 5.70s/it]
26%|██▌ | 11495/43939 [18:34:13<51:19:50, 5.70s/it]
26%|██▌ | 11496/43939 [18:34:19<51:18:49, 5.69s/it]
26%|██▌ | 11497/43939 [18:34:24<51:20:33, 5.70s/it]
26%|██▌ | 11498/43939 [18:34:30<51:24:24, 5.70s/it]
26%|██▌ | 11499/43939 [18:34:36<51:23:27, 5.70s/it]
26%|██▌ | 11500/43939 [18:34:42<51:18:52, 5.69s/it]
{'loss': 0.9485, 'grad_norm': 1.0577303171157837, 'learning_rate': 4.612125612106798e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.24, 'epoch': 1.05}
26%|██▌ | 11500/43939 [18:34:42<51:18:52, 5.69s/it]
26%|██▌ | 11501/43939 [18:34:47<51:21:52, 5.70s/it]
26%|██▌ | 11502/43939 [18:34:53<51:22:18, 5.70s/it]
26%|██▌ | 11503/43939 [18:34:59<51:27:02, 5.71s/it]
26%|██▌ | 11504/43939 [18:35:04<51:25:04, 5.71s/it]
26%|██▌ | 11505/43939 [18:35:10<51:26:25, 5.71s/it]
26%|██▌ | 11506/43939 [18:35:16<51:28:04, 5.71s/it]
26%|██▌ | 11507/43939 [18:35:22<52:03:22, 5.78s/it]
26%|██▌ | 11508/43939 [18:35:28<51:56:14, 5.77s/it]
26%|██▌ | 11509/43939 [18:35:33<51:44:59, 5.74s/it]
26%|██▌ | 11510/43939 [18:35:39<51:29:53, 5.72s/it]
26%|██▌ | 11511/43939 [18:35:45<51:28:40, 5.71s/it]
26%|██▌ | 11512/43939 [18:35:50<51:23:42, 5.71s/it]
26%|██▌ | 11513/43939 [18:35:56<51:23:10, 5.71s/it]
26%|██▌ | 11514/43939 [18:36:02<51:24:38, 5.71s/it]
26%|██▌ | 11515/43939 [18:36:07<51:19:39, 5.70s/it]
26%|██▌ | 11516/43939 [18:36:13<51:13:17, 5.69s/it]
26%|██▌ | 11517/43939 [18:36:19<51:09:06, 5.68s/it]
26%|██▌ | 11518/43939 [18:36:24<51:12:01, 5.69s/it]
26%|██▌ | 11519/43939 [18:36:30<51:08:45, 5.68s/it]
26%|██▌ | 11520/43939 [18:36:36<51:08:03, 5.68s/it]
26%|██▌ | 11521/43939 [18:36:41<51:08:43, 5.68s/it]
26%|██▌ | 11522/43939 [18:36:47<51:08:23, 5.68s/it]
26%|██▌ | 11523/43939 [18:36:53<51:12:39, 5.69s/it]
26%|██▌ | 11524/43939 [18:36:59<51:20:15, 5.70s/it]
26%|██▌ | 11525/43939 [18:37:04<51:19:40, 5.70s/it]
26%|██▌ | 11526/43939 [18:37:10<51:21:37, 5.70s/it]
26%|██▌ | 11527/43939 [18:37:16<51:17:36, 5.70s/it]
26%|██▌ | 11528/43939 [18:37:22<52:03:31, 5.78s/it]
26%|██▌ | 11529/43939 [18:37:27<51:49:49, 5.76s/it]
26%|██▌ | 11530/43939 [18:37:33<51:39:16, 5.74s/it]
26%|██▌ | 11531/43939 [18:37:39<51:38:59, 5.74s/it]
26%|██▌ | 11532/43939 [18:37:44<51:31:24, 5.72s/it]
26%|██▌ | 11533/43939 [18:37:50<51:26:23, 5.71s/it]
26%|██▋ | 11534/43939 [18:37:56<51:20:11, 5.70s/it]
26%|██▋ | 11535/43939 [18:38:02<51:25:05, 5.71s/it]
26%|██▋ | 11536/43939 [18:38:07<51:24:33, 5.71s/it]
26%|██▋ | 11537/43939 [18:38:13<51:34:51, 5.73s/it]
26%|██▋ | 11538/43939 [18:38:19<51:30:29, 5.72s/it]
26%|██▋ | 11539/43939 [18:38:24<51:27:50, 5.72s/it]
26%|██▋ | 11540/43939 [18:38:30<51:22:57, 5.71s/it]
26%|██▋ | 11541/43939 [18:38:36<51:24:23, 5.71s/it]
26%|██▋ | 11542/43939 [18:38:42<51:21:24, 5.71s/it]
26%|██▋ | 11543/43939 [18:38:47<51:19:33, 5.70s/it]
26%|██▋ | 11544/43939 [18:38:53<51:16:17, 5.70s/it]
26%|██▋ | 11545/43939 [18:38:59<51:12:43, 5.69s/it]
26%|██▋ | 11546/43939 [18:39:04<51:15:51, 5.70s/it]
26%|██▋ | 11547/43939 [18:39:10<51:17:38, 5.70s/it]
26%|██▋ | 11548/43939 [18:39:16<51:19:08, 5.70s/it]
26%|██▋ | 11549/43939 [18:39:22<51:51:22, 5.76s/it]
26%|██▋ | 11550/43939 [18:39:27<51:42:04, 5.75s/it]
{'loss': 0.9341, 'grad_norm': 1.1296515464782715, 'learning_rate': 4.60679628698937e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.07, 'epoch': 1.05}
26%|██▋ | 11550/43939 [18:39:27<51:42:04, 5.75s/it]
26%|██▋ | 11551/43939 [18:39:33<51:28:25, 5.72s/it]
26%|██▋ | 11552/43939 [18:39:39<51:24:00, 5.71s/it]
26%|██▋ | 11553/43939 [18:39:44<51:16:42, 5.70s/it]
26%|██▋ | 11554/43939 [18:39:50<51:16:51, 5.70s/it]
26%|██▋ | 11555/43939 [18:39:56<51:11:27, 5.69s/it]
26%|██▋ | 11556/43939 [18:40:01<51:11:18, 5.69s/it]
26%|██▋ | 11557/43939 [18:40:07<51:09:36, 5.69s/it]
26%|██▋ | 11558/43939 [18:40:13<51:12:32, 5.69s/it]
26%|██▋ | 11559/43939 [18:40:19<51:25:15, 5.72s/it]
26%|██▋ | 11560/43939 [18:40:24<51:29:35, 5.73s/it]
26%|██▋ | 11561/43939 [18:40:30<51:20:52, 5.71s/it]
26%|██▋ | 11562/43939 [18:40:36<51:19:16, 5.71s/it]
26%|██▋ | 11563/43939 [18:40:41<51:22:49, 5.71s/it]
26%|██▋ | 11564/43939 [18:40:47<51:18:17, 5.70s/it]
26%|██▋ | 11565/43939 [18:40:53<51:20:39, 5.71s/it]
26%|██▋ | 11566/43939 [18:40:59<51:15:53, 5.70s/it]
26%|██▋ | 11567/43939 [18:41:04<51:11:44, 5.69s/it]
26%|██▋ | 11568/43939 [18:41:10<51:13:52, 5.70s/it]
26%|██▋ | 11569/43939 [18:41:16<51:13:08, 5.70s/it]
26%|██▋ | 11570/43939 [18:41:22<51:51:52, 5.77s/it]
26%|██▋ | 11571/43939 [18:41:27<51:41:25, 5.75s/it]
26%|██▋ | 11572/43939 [18:41:33<51:37:03, 5.74s/it]
26%|██▋ | 11573/43939 [18:41:39<51:31:25, 5.73s/it]
26%|██▋ | 11574/43939 [18:41:44<51:26:21, 5.72s/it]
26%|██▋ | 11575/43939 [18:41:50<51:19:32, 5.71s/it]
26%|██▋ | 11576/43939 [18:41:56<51:25:01, 5.72s/it]
26%|██▋ | 11577/43939 [18:42:01<51:18:27, 5.71s/it]
26%|██▋ | 11578/43939 [18:42:07<51:28:22, 5.73s/it]
26%|██▋ | 11579/43939 [18:42:13<51:16:59, 5.71s/it]
26%|██▋ | 11580/43939 [18:42:19<51:10:52, 5.69s/it]
26%|██▋ | 11581/43939 [18:42:24<51:14:07, 5.70s/it]
26%|██▋ | 11582/43939 [18:42:30<51:20:32, 5.71s/it]
26%|██▋ | 11583/43939 [18:42:36<51:20:38, 5.71s/it]
26%|██▋ | 11584/43939 [18:42:41<51:22:15, 5.72s/it]
26%|██▋ | 11585/43939 [18:42:47<51:12:49, 5.70s/it]
26%|██▋ | 11586/43939 [18:42:53<51:08:46, 5.69s/it]
26%|██▋ | 11587/43939 [18:42:59<51:14:13, 5.70s/it]
26%|██▋ | 11588/43939 [18:43:04<51:16:41, 5.71s/it]
26%|██▋ | 11589/43939 [18:43:10<51:08:45, 5.69s/it]
26%|██▋ | 11590/43939 [18:43:16<51:13:15, 5.70s/it]
26%|██▋ | 11591/43939 [18:43:22<51:53:24, 5.77s/it]
26%|██▋ | 11592/43939 [18:43:27<51:39:54, 5.75s/it]
26%|██▋ | 11593/43939 [18:43:33<51:27:36, 5.73s/it]
26%|██▋ | 11594/43939 [18:43:39<51:18:41, 5.71s/it]
26%|██▋ | 11595/43939 [18:43:44<51:15:21, 5.70s/it]
26%|██▋ | 11596/43939 [18:43:50<51:14:59, 5.70s/it]
26%|██▋ | 11597/43939 [18:43:56<51:13:39, 5.70s/it]
26%|██▋ | 11598/43939 [18:44:01<51:09:57, 5.70s/it]
26%|██▋ | 11599/43939 [18:44:07<51:21:09, 5.72s/it]
26%|██▋ | 11600/43939 [18:44:13<51:14:23, 5.70s/it]
{'loss': 0.9302, 'grad_norm': 1.32448410987854, 'learning_rate': 4.601433722209612e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.75, 'epoch': 1.06}
26%|██▋ | 11600/43939 [18:44:13<51:14:23, 5.70s/it]
26%|██▋ | 11601/43939 [18:44:18<51:12:59, 5.70s/it]
26%|██▋ | 11602/43939 [18:44:24<51:09:57, 5.70s/it]
26%|██▋ | 11603/43939 [18:44:30<51:06:25, 5.69s/it]
26%|██▋ | 11604/43939 [18:44:36<51:03:47, 5.69s/it]
26%|██▋ | 11605/43939 [18:44:41<51:00:53, 5.68s/it]
26%|██▋ | 11606/43939 [18:44:47<51:18:31, 5.71s/it]
26%|██▋ | 11607/43939 [18:44:53<51:18:58, 5.71s/it]
26%|██▋ | 11608/43939 [18:44:58<51:18:03, 5.71s/it]
26%|██▋ | 11609/43939 [18:45:04<51:17:26, 5.71s/it]
26%|██▋ | 11610/43939 [18:45:10<51:14:54, 5.71s/it]
26%|██▋ | 11611/43939 [18:45:16<51:14:50, 5.71s/it]
26%|██▋ | 11612/43939 [18:45:21<51:06:37, 5.69s/it]
26%|██▋ | 11613/43939 [18:45:27<51:42:49, 5.76s/it]
26%|██▋ | 11614/43939 [18:45:33<51:34:18, 5.74s/it]
26%|██▋ | 11615/43939 [18:45:39<51:30:42, 5.74s/it]
26%|██▋ | 11616/43939 [18:45:44<51:24:30, 5.73s/it]
26%|██▋ | 11617/43939 [18:45:50<51:19:32, 5.72s/it]
26%|██▋ | 11618/43939 [18:45:56<51:09:00, 5.70s/it]
26%|██▋ | 11619/43939 [18:46:01<51:10:28, 5.70s/it]
26%|██▋ | 11620/43939 [18:46:07<51:07:44, 5.70s/it]
26%|██▋ | 11621/43939 [18:46:13<50:58:37, 5.68s/it]
26%|██▋ | 11622/43939 [18:46:18<50:55:44, 5.67s/it]
26%|██▋ | 11623/43939 [18:46:24<50:54:55, 5.67s/it]
26%|██▋ | 11624/43939 [18:46:30<50:56:25, 5.67s/it]
26%|██▋ | 11625/43939 [18:46:35<51:01:01, 5.68s/it]
26%|██▋ | 11626/43939 [18:46:41<51:03:33, 5.69s/it]
26%|██▋ | 11627/43939 [18:46:47<51:05:20, 5.69s/it]
26%|██▋ | 11628/43939 [18:46:52<51:03:04, 5.69s/it]
26%|██▋ | 11629/43939 [18:46:58<51:05:03, 5.69s/it]
26%|██▋ | 11630/43939 [18:47:04<51:06:15, 5.69s/it]
26%|██▋ | 11631/43939 [18:47:10<51:10:58, 5.70s/it]
26%|██▋ | 11632/43939 [18:47:15<51:08:29, 5.70s/it]
26%|██▋ | 11633/43939 [18:47:21<51:09:47, 5.70s/it]
26%|██▋ | 11634/43939 [18:47:27<51:50:07, 5.78s/it]
26%|██▋ | 11635/43939 [18:47:33<51:31:45, 5.74s/it]
26%|██▋ | 11636/43939 [18:47:38<51:24:55, 5.73s/it]
26%|██▋ | 11637/43939 [18:47:44<51:27:12, 5.73s/it]
26%|██▋ | 11638/43939 [18:47:50<51:20:36, 5.72s/it]
26%|██▋ | 11639/43939 [18:47:55<51:19:27, 5.72s/it]
26%|██▋ | 11640/43939 [18:48:01<51:24:51, 5.73s/it]
26%|██▋ | 11641/43939 [18:48:07<51:22:29, 5.73s/it]
26%|██▋ | 11642/43939 [18:48:13<51:23:18, 5.73s/it]
26%|██▋ | 11643/43939 [18:48:18<51:17:32, 5.72s/it]
27%|██▋ | 11644/43939 [18:48:24<51:10:32, 5.70s/it]
27%|██▋ | 11645/43939 [18:48:30<51:13:59, 5.71s/it]
27%|██▋ | 11646/43939 [18:48:35<51:10:53, 5.71s/it]
27%|██▋ | 11647/43939 [18:48:41<51:13:15, 5.71s/it]
27%|██▋ | 11648/43939 [18:48:47<51:22:26, 5.73s/it]
27%|██▋ | 11649/43939 [18:48:53<51:08:11, 5.70s/it]
27%|██▋ | 11650/43939 [18:48:58<51:05:24, 5.70s/it]
{'loss': 0.9167, 'grad_norm': 1.3123842477798462, 'learning_rate': 4.5960380023745844e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.27, 'epoch': 1.06}
27%|██▋ | 11650/43939 [18:48:58<51:05:24, 5.70s/it]
27%|██▋ | 11651/43939 [18:49:04<51:15:27, 5.72s/it]
27%|██▋ | 11652/43939 [18:49:10<51:07:29, 5.70s/it]
27%|██▋ | 11653/43939 [18:49:15<51:04:29, 5.70s/it]
27%|██▋ | 11654/43939 [18:49:21<51:01:36, 5.69s/it]
27%|██▋ | 11655/43939 [18:49:27<51:48:36, 5.78s/it]
27%|██▋ | 11656/43939 [18:49:33<51:32:31, 5.75s/it]
27%|██▋ | 11657/43939 [18:49:38<51:21:50, 5.73s/it]
27%|██▋ | 11658/43939 [18:49:44<51:14:51, 5.72s/it]
27%|██▋ | 11659/43939 [18:49:50<51:18:54, 5.72s/it]
27%|██▋ | 11660/43939 [18:49:55<51:18:36, 5.72s/it]
27%|██▋ | 11661/43939 [18:50:01<51:13:22, 5.71s/it]
27%|██▋ | 11662/43939 [18:50:07<51:11:22, 5.71s/it]
27%|██▋ | 11663/43939 [18:50:13<51:21:12, 5.73s/it]
27%|██▋ | 11664/43939 [18:50:18<51:12:02, 5.71s/it]
27%|██▋ | 11665/43939 [18:50:24<51:16:09, 5.72s/it]
27%|██▋ | 11666/43939 [18:50:30<51:12:23, 5.71s/it]
27%|██▋ | 11667/43939 [18:50:35<51:11:11, 5.71s/it]
27%|██▋ | 11668/43939 [18:50:41<51:16:42, 5.72s/it]
27%|██▋ | 11669/43939 [18:50:47<51:17:04, 5.72s/it]
27%|██▋ | 11670/43939 [18:50:53<51:17:39, 5.72s/it]
27%|██▋ | 11671/43939 [18:50:58<51:11:45, 5.71s/it]
27%|██▋ | 11672/43939 [18:51:04<51:04:43, 5.70s/it]
27%|██▋ | 11673/43939 [18:51:10<51:00:17, 5.69s/it]
27%|██▋ | 11674/43939 [18:51:15<51:05:26, 5.70s/it]
27%|██▋ | 11675/43939 [18:51:21<51:07:41, 5.70s/it]
27%|██▋ | 11676/43939 [18:51:27<51:50:14, 5.78s/it]
27%|██▋ | 11677/43939 [18:51:33<51:39:33, 5.76s/it]
27%|██▋ | 11678/43939 [18:51:38<51:28:07, 5.74s/it]
27%|██▋ | 11679/43939 [18:51:44<51:25:42, 5.74s/it]
27%|██▋ | 11680/43939 [18:51:50<51:13:11, 5.72s/it]
27%|██▋ | 11681/43939 [18:51:56<51:05:22, 5.70s/it]
27%|██▋ | 11682/43939 [18:52:01<50:59:50, 5.69s/it]
27%|██▋ | 11683/43939 [18:52:07<50:58:45, 5.69s/it]
27%|██▋ | 11684/43939 [18:52:13<50:58:08, 5.69s/it]
27%|██▋ | 11685/43939 [18:52:18<51:03:33, 5.70s/it]
27%|██▋ | 11686/43939 [18:52:24<51:06:03, 5.70s/it]
27%|██▋ | 11687/43939 [18:52:30<51:00:36, 5.69s/it]
27%|██▋ | 11688/43939 [18:52:35<51:05:33, 5.70s/it]
27%|██▋ | 11689/43939 [18:52:41<51:01:50, 5.70s/it]
27%|██▋ | 11690/43939 [18:52:47<51:00:55, 5.69s/it]
27%|██▋ | 11691/43939 [18:52:53<51:05:40, 5.70s/it]
27%|██▋ | 11692/43939 [18:52:58<51:03:14, 5.70s/it]
27%|██▋ | 11693/43939 [18:53:04<51:10:37, 5.71s/it]
27%|██▋ | 11694/43939 [18:53:10<51:08:38, 5.71s/it]
27%|██▋ | 11695/43939 [18:53:15<51:05:49, 5.70s/it]
27%|██▋ | 11696/43939 [18:53:21<51:05:22, 5.70s/it]
27%|██▋ | 11697/43939 [18:53:27<51:01:39, 5.70s/it]
27%|██▋ | 11698/43939 [18:53:33<51:39:08, 5.77s/it]
27%|██▋ | 11699/43939 [18:53:38<51:26:36, 5.74s/it]
27%|██▋ | 11700/43939 [18:53:44<51:27:46, 5.75s/it]
{'loss': 0.9229, 'grad_norm': 0.8736046552658081, 'learning_rate': 4.5906092126144465e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.28, 'epoch': 1.07}
27%|██▋ | 11700/43939 [18:53:44<51:27:46, 5.75s/it]
27%|██▋ | 11701/43939 [18:53:50<51:20:37, 5.73s/it]
27%|██▋ | 11702/43939 [18:53:55<51:13:48, 5.72s/it]
27%|██▋ | 11703/43939 [18:54:01<51:11:30, 5.72s/it]
27%|██▋ | 11704/43939 [18:54:07<51:13:36, 5.72s/it]
27%|██▋ | 11705/43939 [18:54:13<51:06:54, 5.71s/it]
27%|██▋ | 11706/43939 [18:54:18<51:12:02, 5.72s/it]
27%|██▋ | 11707/43939 [18:54:24<51:06:07, 5.71s/it]
27%|██▋ | 11708/43939 [18:54:30<51:08:00, 5.71s/it]
27%|██▋ | 11709/43939 [18:54:35<51:03:47, 5.70s/it]
27%|██▋ | 11710/43939 [18:54:41<50:53:47, 5.69s/it]
27%|██▋ | 11711/43939 [18:54:47<50:48:01, 5.67s/it]
27%|██▋ | 11712/43939 [18:54:52<50:54:58, 5.69s/it]
27%|██▋ | 11713/43939 [18:54:58<50:59:49, 5.70s/it]
27%|██▋ | 11714/43939 [18:55:04<50:59:57, 5.70s/it]
27%|██▋ | 11715/43939 [18:55:10<50:55:49, 5.69s/it]
27%|██▋ | 11716/43939 [18:55:15<50:56:43, 5.69s/it]
27%|██▋ | 11717/43939 [18:55:21<50:55:28, 5.69s/it]
27%|██▋ | 11718/43939 [18:55:27<50:59:18, 5.70s/it]
27%|██▋ | 11719/43939 [18:55:33<51:39:11, 5.77s/it]
27%|██▋ | 11720/43939 [18:55:38<51:35:24, 5.76s/it]
27%|██▋ | 11721/43939 [18:55:44<51:26:12, 5.75s/it]
27%|██▋ | 11722/43939 [18:55:50<51:13:49, 5.72s/it]
27%|██▋ | 11723/43939 [18:55:55<51:09:18, 5.72s/it]
27%|██▋ | 11724/43939 [18:56:01<51:08:13, 5.71s/it]
27%|██▋ | 11725/43939 [18:56:07<51:07:02, 5.71s/it]
27%|██▋ | 11726/43939 [18:56:12<50:59:51, 5.70s/it]
27%|██▋ | 11727/43939 [18:56:18<51:02:18, 5.70s/it]
27%|██▋ | 11728/43939 [18:56:24<51:02:10, 5.70s/it]
27%|██▋ | 11729/43939 [18:56:30<50:57:47, 5.70s/it]
27%|██▋ | 11730/43939 [18:56:35<51:07:28, 5.71s/it]
27%|██▋ | 11731/43939 [18:56:41<51:02:45, 5.71s/it]
27%|██▋ | 11732/43939 [18:56:47<51:04:27, 5.71s/it]
27%|██▋ | 11733/43939 [18:56:52<51:03:02, 5.71s/it]
27%|██▋ | 11734/43939 [18:56:58<51:06:14, 5.71s/it]
27%|██▋ | 11735/43939 [18:57:04<51:05:23, 5.71s/it]
27%|██▋ | 11736/43939 [18:57:10<50:59:05, 5.70s/it]
27%|██▋ | 11737/43939 [18:57:15<51:04:23, 5.71s/it]
27%|██▋ | 11738/43939 [18:57:21<51:02:04, 5.71s/it]
27%|██▋ | 11739/43939 [18:57:27<50:58:50, 5.70s/it]
27%|██▋ | 11740/43939 [18:57:33<51:35:01, 5.77s/it]
27%|██▋ | 11741/43939 [18:57:38<51:17:29, 5.73s/it]
27%|██▋ | 11742/43939 [18:57:44<51:14:44, 5.73s/it]
27%|██▋ | 11743/43939 [18:57:50<51:08:31, 5.72s/it]
27%|██▋ | 11744/43939 [18:57:55<51:04:30, 5.71s/it]
27%|██▋ | 11745/43939 [18:58:01<50:59:45, 5.70s/it]
27%|██▋ | 11746/43939 [18:58:07<50:59:22, 5.70s/it]
27%|██▋ | 11747/43939 [18:58:12<50:58:39, 5.70s/it]
27%|██▋ | 11748/43939 [18:58:18<50:59:48, 5.70s/it]
27%|██▋ | 11749/43939 [18:58:24<50:59:49, 5.70s/it]
27%|██▋ | 11750/43939 [18:58:30<51:01:13, 5.71s/it]
{'loss': 0.9456, 'grad_norm': 0.9919261336326599, 'learning_rate': 4.585147438581116e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.62, 'epoch': 1.07}
27%|██▋ | 11750/43939 [18:58:30<51:01:13, 5.71s/it]
27%|██▋ | 11751/43939 [18:58:35<50:58:07, 5.70s/it]
27%|██▋ | 11752/43939 [18:58:41<50:58:11, 5.70s/it]
27%|██▋ | 11753/43939 [18:58:47<51:12:33, 5.73s/it]
27%|██▋ | 11754/43939 [18:58:52<51:13:16, 5.73s/it]
27%|██▋ | 11755/43939 [18:58:58<51:10:05, 5.72s/it]
27%|██▋ | 11756/43939 [18:59:04<51:05:08, 5.71s/it]
27%|██▋ | 11757/43939 [18:59:10<50:56:52, 5.70s/it]
27%|██▋ | 11758/43939 [18:59:15<50:49:04, 5.68s/it]
27%|██▋ | 11759/43939 [18:59:21<50:59:30, 5.70s/it]
27%|██▋ | 11760/43939 [18:59:27<50:53:53, 5.69s/it]
27%|██▋ | 11761/43939 [18:59:32<51:04:38, 5.71s/it]
27%|██▋ | 11762/43939 [18:59:38<51:37:40, 5.78s/it]
27%|██▋ | 11763/43939 [18:59:44<51:23:52, 5.75s/it]
27%|██▋ | 11764/43939 [18:59:50<51:17:29, 5.74s/it]
27%|██▋ | 11765/43939 [18:59:55<51:14:38, 5.73s/it]
27%|██▋ | 11766/43939 [19:00:01<51:08:06, 5.72s/it]
27%|██▋ | 11767/43939 [19:00:07<50:58:35, 5.70s/it]
27%|██▋ | 11768/43939 [19:00:13<51:07:07, 5.72s/it]
27%|██▋ | 11769/43939 [19:00:18<51:04:19, 5.72s/it]
27%|██▋ | 11770/43939 [19:00:24<50:59:18, 5.71s/it]
27%|██▋ | 11771/43939 [19:00:30<50:58:49, 5.71s/it]
27%|██▋ | 11772/43939 [19:00:35<50:58:48, 5.71s/it]
27%|██▋ | 11773/43939 [19:00:41<51:00:01, 5.71s/it]
27%|██▋ | 11774/43939 [19:00:47<50:56:04, 5.70s/it]
27%|██▋ | 11775/43939 [19:00:52<50:55:51, 5.70s/it]
27%|██▋ | 11776/43939 [19:00:58<50:50:09, 5.69s/it]
27%|██▋ | 11777/43939 [19:01:04<50:46:07, 5.68s/it]
27%|██▋ | 11778/43939 [19:01:09<50:48:44, 5.69s/it]
27%|██▋ | 11779/43939 [19:01:15<50:44:22, 5.68s/it]
27%|██▋ | 11780/43939 [19:01:21<50:51:22, 5.69s/it]
27%|██▋ | 11781/43939 [19:01:27<50:55:23, 5.70s/it]
27%|██▋ | 11782/43939 [19:01:32<50:57:42, 5.71s/it]
27%|██▋ | 11783/43939 [19:01:38<51:40:03, 5.78s/it]
27%|██▋ | 11784/43939 [19:01:44<51:22:56, 5.75s/it]
27%|██▋ | 11785/43939 [19:01:50<51:14:55, 5.74s/it]
27%|██▋ | 11786/43939 [19:01:55<51:09:45, 5.73s/it]
27%|██▋ | 11787/43939 [19:02:01<51:01:49, 5.71s/it]
27%|██▋ | 11788/43939 [19:02:07<50:56:21, 5.70s/it]
27%|██▋ | 11789/43939 [19:02:12<50:56:41, 5.70s/it]
27%|██▋ | 11790/43939 [19:02:18<50:58:47, 5.71s/it]
27%|██▋ | 11791/43939 [19:02:24<50:54:48, 5.70s/it]
27%|██▋ | 11792/43939 [19:02:30<51:00:22, 5.71s/it]
27%|██▋ | 11793/43939 [19:02:35<50:55:22, 5.70s/it]
27%|██▋ | 11794/43939 [19:02:41<50:53:26, 5.70s/it]
27%|██▋ | 11795/43939 [19:02:47<50:53:16, 5.70s/it]
27%|██▋ | 11796/43939 [19:02:52<50:54:36, 5.70s/it]
27%|██▋ | 11797/43939 [19:02:58<51:02:37, 5.72s/it]
27%|██▋ | 11798/43939 [19:03:04<51:00:59, 5.71s/it]
27%|██▋ | 11799/43939 [19:03:09<50:56:35, 5.71s/it]
27%|██▋ | 11800/43939 [19:03:15<50:56:42, 5.71s/it]
{'loss': 0.9131, 'grad_norm': 1.0609310865402222, 'learning_rate': 4.579652766446913e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.81, 'epoch': 1.07}
27%|██▋ | 11800/43939 [19:03:15<50:56:42, 5.71s/it]
27%|██▋ | 11801/43939 [19:03:21<50:57:44, 5.71s/it]
27%|██▋ | 11802/43939 [19:03:27<50:50:31, 5.70s/it]
27%|██▋ | 11803/43939 [19:03:32<50:55:13, 5.70s/it]
27%|██▋ | 11804/43939 [19:03:38<51:36:11, 5.78s/it]
27%|██▋ | 11805/43939 [19:03:44<51:28:01, 5.77s/it]
27%|██▋ | 11806/43939 [19:03:50<51:17:09, 5.75s/it]
27%|██▋ | 11807/43939 [19:03:55<51:03:14, 5.72s/it]
27%|██▋ | 11808/43939 [19:04:01<51:02:10, 5.72s/it]
27%|██▋ | 11809/43939 [19:04:07<51:03:13, 5.72s/it]
27%|██▋ | 11810/43939 [19:04:12<51:01:07, 5.72s/it]
27%|██▋ | 11811/43939 [19:04:18<50:56:00, 5.71s/it]
27%|██▋ | 11812/43939 [19:04:24<50:54:07, 5.70s/it]
27%|██▋ | 11813/43939 [19:04:30<50:56:01, 5.71s/it]
27%|██▋ | 11814/43939 [19:04:35<50:56:25, 5.71s/it]
27%|██▋ | 11815/43939 [19:04:41<50:55:28, 5.71s/it]
27%|██▋ | 11816/43939 [19:04:47<50:53:35, 5.70s/it]
27%|██▋ | 11817/43939 [19:04:52<50:59:26, 5.71s/it]
27%|██▋ | 11818/43939 [19:04:58<50:59:02, 5.71s/it]
27%|██▋ | 11819/43939 [19:05:04<50:54:10, 5.71s/it]
27%|██▋ | 11820/43939 [19:05:10<50:54:51, 5.71s/it]
27%|██▋ | 11821/43939 [19:05:15<50:48:45, 5.70s/it]
27%|██▋ | 11822/43939 [19:05:21<50:50:06, 5.70s/it]
27%|██▋ | 11823/43939 [19:05:27<50:51:47, 5.70s/it]
27%|██▋ | 11824/43939 [19:05:32<50:54:41, 5.71s/it]
27%|██▋ | 11825/43939 [19:05:38<50:55:03, 5.71s/it]
27%|██▋ | 11826/43939 [19:05:44<51:44:30, 5.80s/it]
27%|██▋ | 11827/43939 [19:05:50<51:32:11, 5.78s/it]
27%|██▋ | 11828/43939 [19:05:56<51:24:09, 5.76s/it]
27%|██▋ | 11829/43939 [19:06:01<51:12:28, 5.74s/it]
27%|██▋ | 11830/43939 [19:06:07<51:13:11, 5.74s/it]
27%|██▋ | 11831/43939 [19:06:13<51:10:37, 5.74s/it]
27%|██▋ | 11832/43939 [19:06:18<51:02:52, 5.72s/it]
27%|██▋ | 11833/43939 [19:06:24<50:56:13, 5.71s/it]
27%|██▋ | 11834/43939 [19:06:30<51:02:52, 5.72s/it]
27%|██▋ | 11835/43939 [19:06:36<51:03:26, 5.73s/it]
27%|██▋ | 11836/43939 [19:06:41<50:52:54, 5.71s/it]
27%|██▋ | 11837/43939 [19:06:47<50:49:56, 5.70s/it]
27%|██▋ | 11838/43939 [19:06:53<50:50:44, 5.70s/it]
27%|██▋ | 11839/43939 [19:06:58<50:53:10, 5.71s/it]
27%|██▋ | 11840/43939 [19:07:04<50:51:59, 5.70s/it]
27%|██▋ | 11841/43939 [19:07:10<50:47:39, 5.70s/it]
27%|██▋ | 11842/43939 [19:07:15<50:46:22, 5.69s/it]
27%|██▋ | 11843/43939 [19:07:21<50:49:02, 5.70s/it]
27%|██▋ | 11844/43939 [19:07:27<50:46:48, 5.70s/it]
27%|██▋ | 11845/43939 [19:07:33<50:57:17, 5.72s/it]
27%|██▋ | 11846/43939 [19:07:38<50:58:01, 5.72s/it]
27%|██▋ | 11847/43939 [19:07:44<51:33:33, 5.78s/it]
27%|██▋ | 11848/43939 [19:07:50<51:15:26, 5.75s/it]
27%|██▋ | 11849/43939 [19:07:56<51:01:45, 5.72s/it]
27%|██▋ | 11850/43939 [19:08:01<51:04:29, 5.73s/it]
{'loss': 0.9265, 'grad_norm': 1.113976001739502, 'learning_rate': 4.574125282903203e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.49, 'epoch': 1.08}
27%|██▋ | 11850/43939 [19:08:01<51:04:29, 5.73s/it]
27%|██▋ | 11851/43939 [19:08:07<51:03:39, 5.73s/it]
27%|██▋ | 11852/43939 [19:08:13<50:58:43, 5.72s/it]
27%|██▋ | 11853/43939 [19:08:18<50:58:19, 5.72s/it]
27%|██▋ | 11854/43939 [19:08:24<50:58:56, 5.72s/it]
27%|██▋ | 11855/43939 [19:08:30<50:53:03, 5.71s/it]
27%|██▋ | 11856/43939 [19:08:35<50:45:47, 5.70s/it]
27%|██▋ | 11857/43939 [19:08:41<50:45:59, 5.70s/it]
27%|██▋ | 11858/43939 [19:08:47<50:51:46, 5.71s/it]
27%|██▋ | 11859/43939 [19:08:53<50:58:19, 5.72s/it]
27%|██▋ | 11860/43939 [19:08:58<50:58:05, 5.72s/it]
27%|██▋ | 11861/43939 [19:09:04<50:54:20, 5.71s/it]
27%|██▋ | 11862/43939 [19:09:10<50:55:24, 5.72s/it]
27%|██▋ | 11863/43939 [19:09:16<50:53:59, 5.71s/it]
27%|██▋ | 11864/43939 [19:09:21<50:56:44, 5.72s/it]
27%|██▋ | 11865/43939 [19:09:27<50:55:41, 5.72s/it]
27%|██▋ | 11866/43939 [19:09:33<50:55:00, 5.72s/it]
27%|██▋ | 11867/43939 [19:09:38<50:47:25, 5.70s/it]
27%|██▋ | 11868/43939 [19:09:44<51:15:43, 5.75s/it]
27%|██▋ | 11869/43939 [19:09:50<51:05:42, 5.74s/it]
27%|██▋ | 11870/43939 [19:09:56<50:59:42, 5.72s/it]
27%|██▋ | 11871/43939 [19:10:01<50:53:00, 5.71s/it]
27%|██▋ | 11872/43939 [19:10:07<50:47:31, 5.70s/it]
27%|██▋ | 11873/43939 [19:10:13<50:52:40, 5.71s/it]
27%|██▋ | 11874/43939 [19:10:18<50:52:01, 5.71s/it]
27%|██▋ | 11875/43939 [19:10:24<50:52:25, 5.71s/it]
27%|██▋ | 11876/43939 [19:10:30<50:55:40, 5.72s/it]
27%|██▋ | 11877/43939 [19:10:36<50:52:06, 5.71s/it]
27%|██▋ | 11878/43939 [19:10:41<50:49:24, 5.71s/it]
27%|██▋ | 11879/43939 [19:10:47<50:45:07, 5.70s/it]
27%|██▋ | 11880/43939 [19:10:53<50:46:44, 5.70s/it]
27%|██▋ | 11881/43939 [19:10:58<50:45:10, 5.70s/it]
27%|██▋ | 11882/43939 [19:11:04<50:49:26, 5.71s/it]
27%|██▋ | 11883/43939 [19:11:10<50:51:24, 5.71s/it]
27%|██▋ | 11884/43939 [19:11:15<50:46:24, 5.70s/it]
27%|██▋ | 11885/43939 [19:11:21<50:44:01, 5.70s/it]
27%|██▋ | 11886/43939 [19:11:27<50:44:07, 5.70s/it]
27%|██▋ | 11887/43939 [19:11:33<50:45:35, 5.70s/it]
27%|██▋ | 11888/43939 [19:11:38<50:44:46, 5.70s/it]
27%|██▋ | 11889/43939 [19:11:44<51:14:43, 5.76s/it]
27%|██▋ | 11890/43939 [19:11:50<50:58:06, 5.73s/it]
27%|██▋ | 11891/43939 [19:11:56<50:57:06, 5.72s/it]
27%|██▋ | 11892/43939 [19:12:01<50:49:41, 5.71s/it]
27%|██▋ | 11893/43939 [19:12:07<50:47:39, 5.71s/it]
27%|██▋ | 11894/43939 [19:12:13<50:50:41, 5.71s/it]
27%|██▋ | 11895/43939 [19:12:18<50:47:53, 5.71s/it]
27%|██▋ | 11896/43939 [19:12:24<50:46:28, 5.70s/it]
27%|██▋ | 11897/43939 [19:12:30<50:48:34, 5.71s/it]
27%|██▋ | 11898/43939 [19:12:35<50:42:34, 5.70s/it]
27%|██▋ | 11899/43939 [19:12:41<50:41:13, 5.70s/it]
27%|██▋ | 11900/43939 [19:12:47<50:50:30, 5.71s/it]
{'loss': 0.9169, 'grad_norm': 1.1453485488891602, 'learning_rate': 4.568565075159027e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.74, 'epoch': 1.08}
27%|██▋ | 11900/43939 [19:12:47<50:50:30, 5.71s/it]
27%|██▋ | 11901/43939 [19:12:53<50:52:07, 5.72s/it]
27%|██▋ | 11902/43939 [19:12:58<50:47:59, 5.71s/it]
27%|██▋ | 11903/43939 [19:13:04<50:40:38, 5.69s/it]
27%|██▋ | 11904/43939 [19:13:10<50:41:17, 5.70s/it]
27%|██▋ | 11905/43939 [19:13:15<50:46:41, 5.71s/it]
27%|██▋ | 11906/43939 [19:13:21<50:42:24, 5.70s/it]
27%|██▋ | 11907/43939 [19:13:27<50:40:13, 5.69s/it]
27%|██▋ | 11908/43939 [19:13:32<50:47:01, 5.71s/it]
27%|██▋ | 11909/43939 [19:13:38<50:46:30, 5.71s/it]
27%|██▋ | 11910/43939 [19:13:44<50:45:32, 5.71s/it]
27%|██▋ | 11911/43939 [19:13:50<51:27:30, 5.78s/it]
27%|██▋ | 11912/43939 [19:13:56<51:11:46, 5.75s/it]
27%|██▋ | 11913/43939 [19:14:01<50:53:33, 5.72s/it]
27%|██▋ | 11914/43939 [19:14:07<50:48:54, 5.71s/it]
27%|██▋ | 11915/43939 [19:14:12<50:39:21, 5.69s/it]
27%|██▋ | 11916/43939 [19:14:18<50:37:39, 5.69s/it]
27%|██▋ | 11917/43939 [19:14:24<50:37:30, 5.69s/it]
27%|██▋ | 11918/43939 [19:14:30<50:41:10, 5.70s/it]
27%|██▋ | 11919/43939 [19:14:35<50:41:17, 5.70s/it]
27%|██▋ | 11920/43939 [19:14:41<50:38:19, 5.69s/it]
27%|██▋ | 11921/43939 [19:14:47<50:37:18, 5.69s/it]
27%|██▋ | 11922/43939 [19:14:52<50:32:21, 5.68s/it]
27%|██▋ | 11923/43939 [19:14:58<50:41:34, 5.70s/it]
27%|██▋ | 11924/43939 [19:15:04<50:41:54, 5.70s/it]
27%|██▋ | 11925/43939 [19:15:09<50:34:51, 5.69s/it]
27%|██▋ | 11926/43939 [19:15:15<50:35:49, 5.69s/it]
27%|██▋ | 11927/43939 [19:15:21<50:34:58, 5.69s/it]
27%|██▋ | 11928/43939 [19:15:27<50:38:13, 5.69s/it]
27%|██▋ | 11929/43939 [19:15:32<50:34:02, 5.69s/it]
27%|██▋ | 11930/43939 [19:15:38<50:30:44, 5.68s/it]
27%|██▋ | 11931/43939 [19:15:44<50:42:24, 5.70s/it]
27%|██▋ | 11932/43939 [19:15:50<51:18:45, 5.77s/it]
27%|██▋ | 11933/43939 [19:15:55<51:06:46, 5.75s/it]
27%|██▋ | 11934/43939 [19:16:01<51:03:33, 5.74s/it]
27%|██▋ | 11935/43939 [19:16:07<50:56:04, 5.73s/it]
27%|██▋ | 11936/43939 [19:16:12<50:41:04, 5.70s/it]
27%|██▋ | 11937/43939 [19:16:18<50:48:06, 5.71s/it]
27%|██▋ | 11938/43939 [19:16:24<50:49:24, 5.72s/it]
27%|██▋ | 11939/43939 [19:16:29<50:49:17, 5.72s/it]
27%|██▋ | 11940/43939 [19:16:35<50:57:31, 5.73s/it]
27%|██▋ | 11941/43939 [19:16:41<50:51:08, 5.72s/it]
27%|██▋ | 11942/43939 [19:16:47<50:46:26, 5.71s/it]
27%|██▋ | 11943/43939 [19:16:52<50:41:02, 5.70s/it]
27%|██▋ | 11944/43939 [19:16:58<50:39:26, 5.70s/it]
27%|██▋ | 11945/43939 [19:17:04<50:36:35, 5.69s/it]
27%|██▋ | 11946/43939 [19:17:09<50:37:16, 5.70s/it]
27%|██▋ | 11947/43939 [19:17:15<50:41:38, 5.70s/it]
27%|██▋ | 11948/43939 [19:17:21<50:39:26, 5.70s/it]
27%|██▋ | 11949/43939 [19:17:26<50:38:44, 5.70s/it]
27%|██▋ | 11950/43939 [19:17:32<50:34:11, 5.69s/it]
{'loss': 0.9068, 'grad_norm': 1.2608872652053833, 'learning_rate': 4.562972230939729e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.6, 'epoch': 1.09}
27%|██▋ | 11950/43939 [19:17:32<50:34:11, 5.69s/it]
27%|██▋ | 11951/43939 [19:17:38<50:40:50, 5.70s/it]
27%|██▋ | 11952/43939 [19:17:44<50:39:32, 5.70s/it]
27%|██▋ | 11953/43939 [19:17:50<51:22:38, 5.78s/it]
27%|██▋ | 11954/43939 [19:17:55<51:05:32, 5.75s/it]
27%|██▋ | 11955/43939 [19:18:01<51:05:59, 5.75s/it]
27%|██▋ | 11956/43939 [19:18:07<50:56:28, 5.73s/it]
27%|██▋ | 11957/43939 [19:18:12<50:49:43, 5.72s/it]
27%|██▋ | 11958/43939 [19:18:18<50:46:41, 5.72s/it]
27%|██▋ | 11959/43939 [19:18:24<50:41:01, 5.71s/it]
27%|██▋ | 11960/43939 [19:18:29<50:33:20, 5.69s/it]
27%|██▋ | 11961/43939 [19:18:35<50:29:42, 5.68s/it]
27%|██▋ | 11962/43939 [19:18:41<50:38:44, 5.70s/it]
27%|██▋ | 11963/43939 [19:18:47<50:39:15, 5.70s/it]
27%|██▋ | 11964/43939 [19:18:52<50:39:00, 5.70s/it]
27%|██▋ | 11965/43939 [19:18:58<50:38:52, 5.70s/it]
27%|██▋ | 11966/43939 [19:19:04<50:48:29, 5.72s/it]
27%|██▋ | 11967/43939 [19:19:09<50:46:53, 5.72s/it]
27%|██▋ | 11968/43939 [19:19:15<50:42:51, 5.71s/it]
27%|██▋ | 11969/43939 [19:19:21<50:37:55, 5.70s/it]
27%|██▋ | 11970/43939 [19:19:26<50:35:21, 5.70s/it]
27%|██▋ | 11971/43939 [19:19:32<50:33:51, 5.69s/it]
27%|██▋ | 11972/43939 [19:19:38<50:30:34, 5.69s/it]
27%|██▋ | 11973/43939 [19:19:43<50:24:37, 5.68s/it]
27%|██▋ | 11974/43939 [19:19:49<51:08:17, 5.76s/it]
27%|██▋ | 11975/43939 [19:19:55<51:04:11, 5.75s/it]
27%|██▋ | 11976/43939 [19:20:01<50:54:31, 5.73s/it]
27%|██▋ | 11977/43939 [19:20:07<50:49:10, 5.72s/it]
27%|██▋ | 11978/43939 [19:20:12<50:45:19, 5.72s/it]
27%|██▋ | 11979/43939 [19:20:18<50:44:42, 5.72s/it]
27%|██▋ | 11980/43939 [19:20:24<50:40:48, 5.71s/it]
27%|██▋ | 11981/43939 [19:20:29<50:47:54, 5.72s/it]
27%|██▋ | 11982/43939 [19:20:35<50:38:23, 5.70s/it]
27%|██▋ | 11983/43939 [19:20:41<50:34:50, 5.70s/it]
27%|██▋ | 11984/43939 [19:20:46<50:35:43, 5.70s/it]
27%|██▋ | 11985/43939 [19:20:52<50:36:06, 5.70s/it]
27%|██▋ | 11986/43939 [19:20:58<50:33:35, 5.70s/it]
27%|██▋ | 11987/43939 [19:21:04<50:36:15, 5.70s/it]
27%|██▋ | 11988/43939 [19:21:09<50:39:26, 5.71s/it]
27%|██▋ | 11989/43939 [19:21:15<50:37:49, 5.70s/it]
27%|██▋ | 11990/43939 [19:21:21<50:41:38, 5.71s/it]
27%|██▋ | 11991/43939 [19:21:26<50:48:21, 5.72s/it]
27%|██▋ | 11992/43939 [19:21:32<50:43:38, 5.72s/it]
27%|██▋ | 11993/43939 [19:21:38<50:39:27, 5.71s/it]
27%|██▋ | 11994/43939 [19:21:44<50:31:56, 5.69s/it]
27%|██▋ | 11995/43939 [19:21:49<50:32:59, 5.70s/it]
27%|██▋ | 11996/43939 [19:21:55<51:20:50, 5.79s/it]
27%|██▋ | 11997/43939 [19:22:01<51:08:55, 5.76s/it]
27%|██▋ | 11998/43939 [19:22:07<51:01:26, 5.75s/it]
27%|██▋ | 11999/43939 [19:22:12<51:05:12, 5.76s/it]
27%|██▋ | 12000/43939 [19:22:18<50:57:42, 5.74s/it]
{'loss': 0.9241, 'grad_norm': 1.3225691318511963, 'learning_rate': 4.55734683848557e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.81, 'epoch': 1.09}
27%|██▋ | 12000/43939 [19:22:18<50:57:42, 5.74s/it]
27%|██▋ | 12001/43939 [19:22:24<50:52:19, 5.73s/it]
27%|██▋ | 12002/43939 [19:22:30<50:42:18, 5.72s/it]
27%|██▋ | 12003/43939 [19:22:35<50:37:00, 5.71s/it]
27%|██▋ | 12004/43939 [19:22:41<50:38:50, 5.71s/it]
27%|██▋ | 12005/43939 [19:22:47<50:42:58, 5.72s/it]
27%|██▋ | 12006/43939 [19:22:52<50:38:46, 5.71s/it]
27%|██▋ | 12007/43939 [19:22:58<50:50:48, 5.73s/it]
27%|██▋ | 12008/43939 [19:23:04<50:47:21, 5.73s/it]
27%|██▋ | 12009/43939 [19:23:10<50:52:47, 5.74s/it]
27%|██▋ | 12010/43939 [19:23:15<50:46:04, 5.72s/it]
27%|██▋ | 12011/43939 [19:23:21<50:45:38, 5.72s/it]
27%|██▋ | 12012/43939 [19:23:27<50:41:14, 5.72s/it]
27%|██▋ | 12013/43939 [19:23:32<50:39:15, 5.71s/it]
27%|██▋ | 12014/43939 [19:23:38<50:33:16, 5.70s/it]
27%|██▋ | 12015/43939 [19:23:44<50:34:14, 5.70s/it]
27%|██▋ | 12016/43939 [19:23:50<50:31:45, 5.70s/it]
27%|██▋ | 12017/43939 [19:23:56<51:22:39, 5.79s/it]
27%|██▋ | 12018/43939 [19:24:01<51:06:08, 5.76s/it]
27%|██▋ | 12019/43939 [19:24:07<50:54:48, 5.74s/it]
27%|██▋ | 12020/43939 [19:24:13<50:49:00, 5.73s/it]
27%|██▋ | 12021/43939 [19:24:18<50:43:22, 5.72s/it]
27%|██▋ | 12022/43939 [19:24:24<50:40:22, 5.72s/it]
27%|██▋ | 12023/43939 [19:24:30<50:38:51, 5.71s/it]
27%|██▋ | 12024/43939 [19:24:35<50:35:51, 5.71s/it]
27%|██▋ | 12025/43939 [19:24:41<50:33:08, 5.70s/it]
27%|██▋ | 12026/43939 [19:24:47<50:32:58, 5.70s/it]
27%|██▋ | 12027/43939 [19:24:53<50:35:44, 5.71s/it]
27%|██▋ | 12028/43939 [19:24:58<50:33:26, 5.70s/it]
27%|██▋ | 12029/43939 [19:25:04<50:29:54, 5.70s/it]
27%|██▋ | 12030/43939 [19:25:10<50:26:15, 5.69s/it]
27%|██▋ | 12031/43939 [19:25:15<50:28:52, 5.70s/it]
27%|██▋ | 12032/43939 [19:25:21<50:35:00, 5.71s/it]
27%|██▋ | 12033/43939 [19:25:27<50:29:08, 5.70s/it]
27%|██▋ | 12034/43939 [19:25:32<50:27:43, 5.69s/it]
27%|██▋ | 12035/43939 [19:25:38<50:25:09, 5.69s/it]
27%|██▋ | 12036/43939 [19:25:44<50:27:16, 5.69s/it]
27%|██▋ | 12037/43939 [19:25:50<51:06:22, 5.77s/it]
27%|██▋ | 12038/43939 [19:25:55<50:55:56, 5.75s/it]
27%|██▋ | 12039/43939 [19:26:01<50:46:51, 5.73s/it]
27%|██▋ | 12040/43939 [19:26:07<50:46:29, 5.73s/it]
27%|██▋ | 12041/43939 [19:26:13<50:42:08, 5.72s/it]
27%|██▋ | 12042/43939 [19:26:18<50:39:00, 5.72s/it]
27%|██▋ | 12043/43939 [19:26:24<50:30:17, 5.70s/it]
27%|██▋ | 12044/43939 [19:26:30<50:35:54, 5.71s/it]
27%|██▋ | 12045/43939 [19:26:35<50:41:29, 5.72s/it]
27%|██▋ | 12046/43939 [19:26:41<50:30:01, 5.70s/it]
27%|██▋ | 12047/43939 [19:26:47<50:30:29, 5.70s/it]
27%|██▋ | 12048/43939 [19:26:53<50:40:19, 5.72s/it]
27%|██▋ | 12049/43939 [19:26:58<50:38:10, 5.72s/it]
27%|██▋ | 12050/43939 [19:27:04<50:38:58, 5.72s/it]
{'loss': 0.9294, 'grad_norm': 1.1706981658935547, 'learning_rate': 4.5516889865503356e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.18, 'epoch': 1.1}
27%|██▋ | 12050/43939 [19:27:04<50:38:58, 5.72s/it]
27%|██▋ | 12051/43939 [19:27:10<50:33:14, 5.71s/it]
27%|██▋ | 12052/43939 [19:27:15<50:28:31, 5.70s/it]
27%|██▋ | 12053/43939 [19:27:21<50:27:16, 5.70s/it]
27%|██▋ | 12054/43939 [19:27:27<50:31:10, 5.70s/it]
27%|██▋ | 12055/43939 [19:27:32<50:33:53, 5.71s/it]
27%|██▋ | 12056/43939 [19:27:38<50:31:45, 5.71s/it]
27%|██▋ | 12057/43939 [19:27:44<50:38:38, 5.72s/it]
27%|██▋ | 12058/43939 [19:27:50<50:38:34, 5.72s/it]
27%|██▋ | 12059/43939 [19:27:56<51:18:57, 5.79s/it]
27%|██▋ | 12060/43939 [19:28:01<51:00:33, 5.76s/it]
27%|██▋ | 12061/43939 [19:28:07<50:53:12, 5.75s/it]
27%|██▋ | 12062/43939 [19:28:13<50:41:40, 5.73s/it]
27%|██▋ | 12063/43939 [19:28:18<50:36:27, 5.72s/it]
27%|██▋ | 12064/43939 [19:28:24<50:34:20, 5.71s/it]
27%|██▋ | 12065/43939 [19:28:30<50:28:31, 5.70s/it]
27%|██▋ | 12066/43939 [19:28:35<50:39:13, 5.72s/it]
27%|██▋ | 12067/43939 [19:28:41<50:32:46, 5.71s/it]
27%|██▋ | 12068/43939 [19:28:47<50:31:39, 5.71s/it]
27%|██▋ | 12069/43939 [19:28:53<50:29:18, 5.70s/it]
27%|██▋ | 12070/43939 [19:28:58<50:27:20, 5.70s/it]
27%|██▋ | 12071/43939 [19:29:04<50:27:49, 5.70s/it]
27%|██▋ | 12072/43939 [19:29:10<50:27:25, 5.70s/it]
27%|██▋ | 12073/43939 [19:29:15<50:28:13, 5.70s/it]
27%|██▋ | 12074/43939 [19:29:21<50:27:58, 5.70s/it]
27%|██▋ | 12075/43939 [19:29:27<50:29:49, 5.71s/it]
27%|██▋ | 12076/43939 [19:29:32<50:24:02, 5.69s/it]
27%|██▋ | 12077/43939 [19:29:38<50:25:42, 5.70s/it]
27%|██▋ | 12078/43939 [19:29:44<50:29:38, 5.71s/it]
27%|██▋ | 12079/43939 [19:29:50<50:57:20, 5.76s/it]
27%|██▋ | 12080/43939 [19:29:55<50:47:20, 5.74s/it]
27%|██▋ | 12081/43939 [19:30:01<50:47:15, 5.74s/it]
27%|██▋ | 12082/43939 [19:30:07<50:45:42, 5.74s/it]
27%|██▋ | 12083/43939 [19:30:13<50:47:27, 5.74s/it]
28%|██▊ | 12084/43939 [19:30:18<50:44:14, 5.73s/it]
28%|██▊ | 12085/43939 [19:30:24<50:42:16, 5.73s/it]
28%|██▊ | 12086/43939 [19:30:30<50:41:13, 5.73s/it]
28%|██▊ | 12087/43939 [19:30:35<50:32:37, 5.71s/it]
28%|██▊ | 12088/43939 [19:30:41<50:34:58, 5.72s/it]
28%|██▊ | 12089/43939 [19:30:47<50:33:03, 5.71s/it]
28%|██▊ | 12090/43939 [19:30:53<50:28:52, 5.71s/it]
28%|██▊ | 12091/43939 [19:30:58<50:35:27, 5.72s/it]
28%|██▊ | 12092/43939 [19:31:04<50:33:12, 5.71s/it]
28%|██▊ | 12093/43939 [19:31:10<50:23:22, 5.70s/it]
28%|██▊ | 12094/43939 [19:31:15<50:26:57, 5.70s/it]
28%|██▊ | 12095/43939 [19:31:21<50:24:03, 5.70s/it]
28%|██▊ | 12096/43939 [19:31:27<50:28:49, 5.71s/it]
28%|██▊ | 12097/43939 [19:31:33<50:24:41, 5.70s/it]
28%|██▊ | 12098/43939 [19:31:38<50:23:02, 5.70s/it]
28%|██▊ | 12099/43939 [19:31:44<50:23:58, 5.70s/it]
28%|██▊ | 12100/43939 [19:31:50<51:01:18, 5.77s/it]
{'loss': 0.9442, 'grad_norm': 1.0240581035614014, 'learning_rate': 4.5459987643999346e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.56, 'epoch': 1.1}
28%|██▊ | 12100/43939 [19:31:50<51:01:18, 5.77s/it]
28%|██▊ | 12101/43939 [19:31:56<50:53:15, 5.75s/it]
28%|██▊ | 12102/43939 [19:32:01<50:38:20, 5.73s/it]
28%|██▊ | 12103/43939 [19:32:07<50:41:06, 5.73s/it]
28%|██▊ | 12104/43939 [19:32:13<50:42:58, 5.74s/it]
28%|██▊ | 12105/43939 [19:32:18<50:37:08, 5.72s/it]
28%|██▊ | 12106/43939 [19:32:24<50:30:35, 5.71s/it]
28%|██▊ | 12107/43939 [19:32:30<50:27:29, 5.71s/it]
28%|██▊ | 12108/43939 [19:32:36<50:28:09, 5.71s/it]
28%|██▊ | 12109/43939 [19:32:41<50:21:46, 5.70s/it]
28%|██▊ | 12110/43939 [19:32:47<50:16:36, 5.69s/it]
28%|██▊ | 12111/43939 [19:32:53<50:14:41, 5.68s/it]
28%|██▊ | 12112/43939 [19:32:58<50:16:35, 5.69s/it]
28%|██▊ | 12113/43939 [19:33:04<50:16:57, 5.69s/it]
28%|██▊ | 12114/43939 [19:33:10<50:22:28, 5.70s/it]
28%|██▊ | 12115/43939 [19:33:15<50:27:49, 5.71s/it]
28%|██▊ | 12116/43939 [19:33:21<50:24:34, 5.70s/it]
28%|██▊ | 12117/43939 [19:33:27<50:29:58, 5.71s/it]
28%|██▊ | 12118/43939 [19:33:32<50:25:52, 5.71s/it]
28%|██▊ | 12119/43939 [19:33:38<50:26:29, 5.71s/it]
28%|██▊ | 12120/43939 [19:33:44<50:26:23, 5.71s/it]
28%|██▊ | 12121/43939 [19:33:50<50:24:42, 5.70s/it]
28%|██▊ | 12122/43939 [19:33:56<51:04:58, 5.78s/it]
28%|██▊ | 12123/43939 [19:34:01<50:50:20, 5.75s/it]
28%|██▊ | 12124/43939 [19:34:07<50:46:19, 5.75s/it]
28%|██▊ | 12125/43939 [19:34:13<50:36:46, 5.73s/it]
28%|██▊ | 12126/43939 [19:34:18<50:30:59, 5.72s/it]
28%|██▊ | 12127/43939 [19:34:24<50:31:02, 5.72s/it]
28%|██▊ | 12128/43939 [19:34:30<50:29:11, 5.71s/it]
28%|██▊ | 12129/43939 [19:34:35<50:28:16, 5.71s/it]
28%|██▊ | 12130/43939 [19:34:41<50:29:46, 5.71s/it]
28%|██▊ | 12131/43939 [19:34:47<50:35:42, 5.73s/it]
28%|██▊ | 12132/43939 [19:34:53<50:36:37, 5.73s/it]
28%|██▊ | 12133/43939 [19:34:58<50:32:19, 5.72s/it]
28%|██▊ | 12134/43939 [19:35:04<50:25:44, 5.71s/it]
28%|██▊ | 12135/43939 [19:35:10<50:20:19, 5.70s/it]
28%|██▊ | 12136/43939 [19:35:15<50:23:27, 5.70s/it]
28%|██▊ | 12137/43939 [19:35:21<50:23:34, 5.70s/it]
28%|██▊ | 12138/43939 [19:35:27<50:19:13, 5.70s/it]
28%|██▊ | 12139/43939 [19:35:33<50:30:30, 5.72s/it]
28%|██▊ | 12140/43939 [19:35:38<50:26:47, 5.71s/it]
28%|██▊ | 12141/43939 [19:35:44<50:24:19, 5.71s/it]
28%|██▊ | 12142/43939 [19:35:50<50:59:31, 5.77s/it]
28%|██▊ | 12143/43939 [19:35:56<50:50:49, 5.76s/it]
28%|██▊ | 12144/43939 [19:36:01<50:43:04, 5.74s/it]
28%|██▊ | 12145/43939 [19:36:07<50:36:28, 5.73s/it]
28%|██▊ | 12146/43939 [19:36:13<50:28:26, 5.72s/it]
28%|██▊ | 12147/43939 [19:36:18<50:21:14, 5.70s/it]
28%|██▊ | 12148/43939 [19:36:24<50:28:51, 5.72s/it]
28%|██▊ | 12149/43939 [19:36:30<50:33:28, 5.73s/it]
28%|██▊ | 12150/43939 [19:36:36<50:29:50, 5.72s/it]
{'loss': 0.9343, 'grad_norm': 1.1509757041931152, 'learning_rate': 4.540276261810995e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.25, 'epoch': 1.11}
28%|██▊ | 12150/43939 [19:36:36<50:29:50, 5.72s/it]
28%|██▊ | 12151/43939 [19:36:41<50:35:00, 5.73s/it]
28%|██▊ | 12152/43939 [19:36:47<50:29:35, 5.72s/it]
28%|██▊ | 12153/43939 [19:36:53<50:27:15, 5.71s/it]
28%|██▊ | 12154/43939 [19:36:58<50:23:01, 5.71s/it]
28%|██▊ | 12155/43939 [19:37:04<50:21:40, 5.70s/it]
28%|██▊ | 12156/43939 [19:37:10<50:23:52, 5.71s/it]
28%|██▊ | 12157/43939 [19:37:16<50:21:54, 5.70s/it]
28%|██▊ | 12158/43939 [19:37:21<50:25:40, 5.71s/it]
28%|██▊ | 12159/43939 [19:37:27<50:23:45, 5.71s/it]
28%|██▊ | 12160/43939 [19:37:33<50:26:08, 5.71s/it]
28%|██▊ | 12161/43939 [19:37:38<50:29:43, 5.72s/it]
28%|██▊ | 12162/43939 [19:37:44<50:27:59, 5.72s/it]
28%|██▊ | 12163/43939 [19:37:50<51:09:20, 5.80s/it]
28%|██▊ | 12164/43939 [19:37:56<50:52:09, 5.76s/it]
28%|██▊ | 12165/43939 [19:38:02<50:42:13, 5.74s/it]
28%|██▊ | 12166/43939 [19:38:07<50:37:59, 5.74s/it]
28%|██▊ | 12167/43939 [19:38:13<50:31:49, 5.73s/it]
28%|██▊ | 12168/43939 [19:38:19<50:22:29, 5.71s/it]
28%|██▊ | 12169/43939 [19:38:24<50:26:57, 5.72s/it]
28%|██▊ | 12170/43939 [19:38:30<50:34:25, 5.73s/it]
28%|██▊ | 12171/43939 [19:38:36<50:32:40, 5.73s/it]
28%|██▊ | 12172/43939 [19:38:41<50:19:17, 5.70s/it]
28%|██▊ | 12173/43939 [19:38:47<50:16:54, 5.70s/it]
28%|██▊ | 12174/43939 [19:38:53<50:11:28, 5.69s/it]
28%|██▊ | 12175/43939 [19:38:59<50:12:34, 5.69s/it]
28%|██▊ | 12176/43939 [19:39:04<50:12:36, 5.69s/it]
28%|██▊ | 12177/43939 [19:39:10<50:10:29, 5.69s/it]
28%|██▊ | 12178/43939 [19:39:16<50:19:52, 5.70s/it]
28%|██▊ | 12179/43939 [19:39:21<50:16:24, 5.70s/it]
28%|██▊ | 12180/43939 [19:39:27<50:22:25, 5.71s/it]
28%|██▊ | 12181/43939 [19:39:33<50:18:10, 5.70s/it]
28%|██▊ | 12182/43939 [19:39:38<50:17:24, 5.70s/it]
28%|██▊ | 12183/43939 [19:39:44<50:18:44, 5.70s/it]
28%|██▊ | 12184/43939 [19:39:50<50:16:14, 5.70s/it]
28%|██▊ | 12185/43939 [19:39:56<50:44:44, 5.75s/it]
28%|██▊ | 12186/43939 [19:40:01<50:37:02, 5.74s/it]
28%|██▊ | 12187/43939 [19:40:07<50:29:24, 5.72s/it]
28%|██▊ | 12188/43939 [19:40:13<50:22:02, 5.71s/it]
28%|██▊ | 12189/43939 [19:40:19<50:21:51, 5.71s/it]
28%|██▊ | 12190/43939 [19:40:24<50:14:00, 5.70s/it]
28%|██▊ | 12191/43939 [19:40:30<50:09:24, 5.69s/it]
28%|██▊ | 12192/43939 [19:40:36<50:10:42, 5.69s/it]
28%|██▊ | 12193/43939 [19:40:41<50:23:54, 5.72s/it]
28%|██▊ | 12194/43939 [19:40:47<50:30:18, 5.73s/it]
28%|██▊ | 12195/43939 [19:40:53<50:18:54, 5.71s/it]
28%|██▊ | 12196/43939 [19:40:58<50:16:15, 5.70s/it]
28%|██▊ | 12197/43939 [19:41:04<50:15:25, 5.70s/it]
28%|██▊ | 12198/43939 [19:41:10<50:17:38, 5.70s/it]
28%|██▊ | 12199/43939 [19:41:16<50:19:45, 5.71s/it]
28%|██▊ | 12200/43939 [19:41:21<50:12:39, 5.70s/it]
{'loss': 0.9406, 'grad_norm': 1.0813158750534058, 'learning_rate': 4.534521569069441e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.6, 'epoch': 1.11}
28%|██▊ | 12200/43939 [19:41:21<50:12:39, 5.70s/it]
28%|██▊ | 12201/43939 [19:41:27<50:18:47, 5.71s/it]
28%|██▊ | 12202/43939 [19:41:33<50:22:02, 5.71s/it]
28%|██▊ | 12203/43939 [19:41:38<50:20:42, 5.71s/it]
28%|██▊ | 12204/43939 [19:41:44<50:21:29, 5.71s/it]
28%|██▊ | 12205/43939 [19:41:50<50:24:05, 5.72s/it]
28%|██▊ | 12206/43939 [19:41:56<50:53:37, 5.77s/it]
28%|██▊ | 12207/43939 [19:42:01<50:42:53, 5.75s/it]
28%|██▊ | 12208/43939 [19:42:07<50:40:14, 5.75s/it]
28%|██▊ | 12209/43939 [19:42:13<50:37:44, 5.74s/it]
28%|██▊ | 12210/43939 [19:42:19<50:31:34, 5.73s/it]
28%|██▊ | 12211/43939 [19:42:24<50:33:06, 5.74s/it]
28%|██▊ | 12212/43939 [19:42:30<50:23:55, 5.72s/it]
28%|██▊ | 12213/43939 [19:42:36<50:26:24, 5.72s/it]
28%|██▊ | 12214/43939 [19:42:41<50:24:39, 5.72s/it]
28%|██▊ | 12215/43939 [19:42:47<50:23:19, 5.72s/it]
28%|██▊ | 12216/43939 [19:42:53<50:21:23, 5.71s/it]
28%|██▊ | 12217/43939 [19:42:59<50:23:56, 5.72s/it]
28%|██▊ | 12218/43939 [19:43:04<50:26:52, 5.73s/it]
28%|██▊ | 12219/43939 [19:43:10<50:20:12, 5.71s/it]
28%|██▊ | 12220/43939 [19:43:16<50:19:11, 5.71s/it]
28%|██▊ | 12221/43939 [19:43:21<50:14:21, 5.70s/it]
28%|██▊ | 12222/43939 [19:43:27<50:15:50, 5.71s/it]
28%|██▊ | 12223/43939 [19:43:33<50:18:44, 5.71s/it]
28%|██▊ | 12224/43939 [19:43:39<50:10:55, 5.70s/it]
28%|██▊ | 12225/43939 [19:43:44<50:12:04, 5.70s/it]
28%|██▊ | 12226/43939 [19:43:50<50:06:24, 5.69s/it]
28%|██▊ | 12227/43939 [19:43:56<50:49:43, 5.77s/it]
28%|██▊ | 12228/43939 [19:44:02<50:49:16, 5.77s/it]
28%|██▊ | 12229/43939 [19:44:07<50:46:42, 5.76s/it]
28%|██▊ | 12230/43939 [19:44:13<50:43:41, 5.76s/it]
28%|██▊ | 12231/43939 [19:44:19<50:33:17, 5.74s/it]
28%|██▊ | 12232/43939 [19:44:25<50:23:03, 5.72s/it]
28%|██▊ | 12233/43939 [19:44:30<50:16:26, 5.71s/it]
28%|██▊ | 12234/43939 [19:44:36<50:21:52, 5.72s/it]
28%|██▊ | 12235/43939 [19:44:42<50:14:39, 5.71s/it]
28%|██▊ | 12236/43939 [19:44:47<50:16:28, 5.71s/it]
28%|██▊ | 12237/43939 [19:44:53<50:14:49, 5.71s/it]
28%|██▊ | 12238/43939 [19:44:59<50:09:57, 5.70s/it]
28%|██▊ | 12239/43939 [19:45:04<50:08:56, 5.70s/it]
28%|██▊ | 12240/43939 [19:45:10<50:10:54, 5.70s/it]
28%|██▊ | 12241/43939 [19:45:16<50:12:25, 5.70s/it]
28%|██▊ | 12242/43939 [19:45:22<50:22:44, 5.72s/it]
28%|██▊ | 12243/43939 [19:45:27<50:27:33, 5.73s/it]
28%|██▊ | 12244/43939 [19:45:33<50:20:22, 5.72s/it]
28%|██▊ | 12245/43939 [19:45:39<50:12:08, 5.70s/it]
28%|██▊ | 12246/43939 [19:45:44<50:13:29, 5.71s/it]
28%|██▊ | 12247/43939 [19:45:50<50:12:09, 5.70s/it]
28%|██▊ | 12248/43939 [19:45:56<50:50:21, 5.78s/it]
28%|██▊ | 12249/43939 [19:46:02<50:39:57, 5.76s/it]
28%|██▊ | 12250/43939 [19:46:07<50:32:55, 5.74s/it]
{'loss': 0.9137, 'grad_norm': 1.1653159856796265, 'learning_rate': 4.528734776969076e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.69, 'epoch': 1.12}
28%|██▊ | 12250/43939 [19:46:07<50:32:55, 5.74s/it]
28%|██▊ | 12251/43939 [19:46:13<50:26:04, 5.73s/it]
28%|██▊ | 12252/43939 [19:46:19<50:25:09, 5.73s/it]
28%|██▊ | 12253/43939 [19:46:25<50:29:44, 5.74s/it]
28%|██▊ | 12254/43939 [19:46:30<50:25:16, 5.73s/it]
28%|██▊ | 12255/43939 [19:46:36<50:21:04, 5.72s/it]
28%|██▊ | 12256/43939 [19:46:42<50:15:59, 5.71s/it]
28%|██▊ | 12257/43939 [19:46:47<50:13:46, 5.71s/it]
28%|██▊ | 12258/43939 [19:46:53<50:08:52, 5.70s/it]
28%|██▊ | 12259/43939 [19:46:59<50:14:15, 5.71s/it]
28%|██▊ | 12260/43939 [19:47:05<50:15:14, 5.71s/it]
28%|██▊ | 12261/43939 [19:47:10<50:13:31, 5.71s/it]
28%|██▊ | 12262/43939 [19:47:16<50:21:32, 5.72s/it]
28%|██▊ | 12263/43939 [19:47:22<50:14:56, 5.71s/it]
28%|██▊ | 12264/43939 [19:47:27<50:10:01, 5.70s/it]
28%|██▊ | 12265/43939 [19:47:33<50:09:32, 5.70s/it]
28%|██▊ | 12266/43939 [19:47:39<50:04:49, 5.69s/it]
28%|██▊ | 12267/43939 [19:47:44<50:05:17, 5.69s/it]
28%|██▊ | 12268/43939 [19:47:50<50:01:33, 5.69s/it]
28%|██▊ | 12269/43939 [19:47:56<50:42:26, 5.76s/it]
28%|██▊ | 12270/43939 [19:48:02<50:33:24, 5.75s/it]
28%|██▊ | 12271/43939 [19:48:07<50:28:08, 5.74s/it]
28%|██▊ | 12272/43939 [19:48:13<50:22:53, 5.73s/it]
28%|██▊ | 12273/43939 [19:48:19<50:28:08, 5.74s/it]
28%|██▊ | 12274/43939 [19:48:25<50:21:01, 5.72s/it]
28%|██▊ | 12275/43939 [19:48:30<50:10:35, 5.70s/it]
28%|██▊ | 12276/43939 [19:48:36<50:15:44, 5.71s/it]
28%|██▊ | 12277/43939 [19:48:42<50:15:52, 5.72s/it]
28%|██▊ | 12278/43939 [19:48:48<50:26:07, 5.73s/it]
28%|██▊ | 12279/43939 [19:48:53<50:21:50, 5.73s/it]
28%|██▊ | 12280/43939 [19:48:59<50:22:00, 5.73s/it]
28%|██▊ | 12281/43939 [19:49:05<50:15:41, 5.72s/it]
28%|██▊ | 12282/43939 [19:49:10<50:06:19, 5.70s/it]
28%|██▊ | 12283/43939 [19:49:16<50:04:25, 5.69s/it]
28%|██▊ | 12284/43939 [19:49:22<50:17:09, 5.72s/it]
28%|██▊ | 12285/43939 [19:49:27<50:15:29, 5.72s/it]
28%|██▊ | 12286/43939 [19:49:33<50:15:00, 5.72s/it]
28%|██▊ | 12287/43939 [19:49:39<50:11:18, 5.71s/it]
28%|██▊ | 12288/43939 [19:49:45<50:12:31, 5.71s/it]
28%|██▊ | 12289/43939 [19:49:50<50:05:51, 5.70s/it]
28%|██▊ | 12290/43939 [19:49:56<50:45:49, 5.77s/it]
28%|██▊ | 12291/43939 [19:50:02<50:33:53, 5.75s/it]
28%|██▊ | 12292/43939 [19:50:08<50:21:20, 5.73s/it]
28%|██▊ | 12293/43939 [19:50:13<50:18:26, 5.72s/it]
28%|██▊ | 12294/43939 [19:50:19<50:15:57, 5.72s/it]
28%|██▊ | 12295/43939 [19:50:25<50:12:15, 5.71s/it]
28%|██▊ | 12296/43939 [19:50:30<50:10:54, 5.71s/it]
28%|██▊ | 12297/43939 [19:50:36<50:11:48, 5.71s/it]
28%|██▊ | 12298/43939 [19:50:42<50:07:13, 5.70s/it]
28%|██▊ | 12299/43939 [19:50:48<50:08:54, 5.71s/it]
28%|██▊ | 12300/43939 [19:50:53<50:12:03, 5.71s/it]
{'loss': 0.9442, 'grad_norm': 1.3311009407043457, 'learning_rate': 4.522915976810142e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.51, 'epoch': 1.12}
28%|██▊ | 12300/43939 [19:50:53<50:12:03, 5.71s/it]
28%|██▊ | 12301/43939 [19:50:59<50:12:04, 5.71s/it]
28%|██▊ | 12302/43939 [19:51:05<50:08:11, 5.71s/it]
28%|██▊ | 12303/43939 [19:51:10<50:13:44, 5.72s/it]
28%|██▊ | 12304/43939 [19:51:16<50:06:04, 5.70s/it]
28%|██▊ | 12305/43939 [19:51:22<50:09:57, 5.71s/it]
28%|██▊ | 12306/43939 [19:51:28<50:11:53, 5.71s/it]
28%|██▊ | 12307/43939 [19:51:33<50:15:01, 5.72s/it]
28%|██▊ | 12308/43939 [19:51:39<50:17:12, 5.72s/it]
28%|██▊ | 12309/43939 [19:51:45<50:13:50, 5.72s/it]
28%|██▊ | 12310/43939 [19:51:50<50:14:22, 5.72s/it]
28%|██▊ | 12311/43939 [19:51:56<50:54:40, 5.79s/it]
28%|██▊ | 12312/43939 [19:52:02<50:47:54, 5.78s/it]
28%|██▊ | 12313/43939 [19:52:08<50:43:49, 5.77s/it]
28%|██▊ | 12314/43939 [19:52:14<50:43:18, 5.77s/it]
28%|██▊ | 12315/43939 [19:52:19<50:33:46, 5.76s/it]
28%|██▊ | 12316/43939 [19:52:25<50:23:03, 5.74s/it]
28%|██▊ | 12317/43939 [19:52:31<50:21:09, 5.73s/it]
28%|██▊ | 12318/43939 [19:52:36<50:15:45, 5.72s/it]
28%|██▊ | 12319/43939 [19:52:42<50:17:54, 5.73s/it]
28%|██▊ | 12320/43939 [19:52:48<50:17:05, 5.73s/it]
28%|██▊ | 12321/43939 [19:52:54<50:16:21, 5.72s/it]
28%|██▊ | 12322/43939 [19:52:59<50:14:09, 5.72s/it]
28%|██▊ | 12323/43939 [19:53:05<50:20:33, 5.73s/it]
28%|██▊ | 12324/43939 [19:53:11<50:15:54, 5.72s/it]
28%|██▊ | 12325/43939 [19:53:17<50:13:34, 5.72s/it]
28%|██▊ | 12326/43939 [19:53:22<50:09:47, 5.71s/it]
28%|██▊ | 12327/43939 [19:53:28<50:14:14, 5.72s/it]
28%|██▊ | 12328/43939 [19:53:34<50:16:51, 5.73s/it]
28%|██▊ | 12329/43939 [19:53:39<50:18:46, 5.73s/it]
28%|██▊ | 12330/43939 [19:53:45<50:22:25, 5.74s/it]
28%|██▊ | 12331/43939 [19:53:51<50:20:17, 5.73s/it]
28%|██▊ | 12332/43939 [19:53:57<50:22:53, 5.74s/it]
28%|██▊ | 12333/43939 [19:54:03<51:00:24, 5.81s/it]
28%|██▊ | 12334/43939 [19:54:08<50:38:07, 5.77s/it]
28%|██▊ | 12335/43939 [19:54:14<50:23:10, 5.74s/it]
28%|██▊ | 12336/43939 [19:54:20<50:15:27, 5.72s/it]
28%|██▊ | 12337/43939 [19:54:25<50:08:09, 5.71s/it]
28%|██▊ | 12338/43939 [19:54:31<50:11:19, 5.72s/it]
28%|██▊ | 12339/43939 [19:54:37<50:08:08, 5.71s/it]
28%|██▊ | 12340/43939 [19:54:43<50:02:06, 5.70s/it]
28%|██▊ | 12341/43939 [19:54:48<50:06:53, 5.71s/it]
28%|██▊ | 12342/43939 [19:54:54<50:02:17, 5.70s/it]
28%|██▊ | 12343/43939 [19:55:00<50:10:59, 5.72s/it]
28%|██▊ | 12344/43939 [19:55:05<50:11:27, 5.72s/it]
28%|██▊ | 12345/43939 [19:55:11<50:08:06, 5.71s/it]
28%|██▊ | 12346/43939 [19:55:17<50:05:31, 5.71s/it]
28%|██▊ | 12347/43939 [19:55:22<50:04:33, 5.71s/it]
28%|██▊ | 12348/43939 [19:55:28<50:06:41, 5.71s/it]
28%|██▊ | 12349/43939 [19:55:34<50:07:00, 5.71s/it]
28%|██▊ | 12350/43939 [19:55:40<49:58:11, 5.69s/it]
{'loss': 0.9156, 'grad_norm': 1.2626206874847412, 'learning_rate': 4.517065260397887e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 532.67, 'epoch': 1.12}
28%|██▊ | 12350/43939 [19:55:40<49:58:11, 5.69s/it]
28%|██▊ | 12351/43939 [19:55:45<50:03:14, 5.70s/it]
28%|██▊ | 12352/43939 [19:55:51<49:58:10, 5.70s/it]
28%|██▊ | 12353/43939 [19:55:57<49:56:56, 5.69s/it]
28%|██▊ | 12354/43939 [19:56:03<50:31:03, 5.76s/it]
28%|██▊ | 12355/43939 [19:56:08<50:16:12, 5.73s/it]
28%|██▊ | 12356/43939 [19:56:14<50:11:25, 5.72s/it]
28%|██▊ | 12357/43939 [19:56:20<50:15:01, 5.73s/it]
28%|██▊ | 12358/43939 [19:56:25<50:10:05, 5.72s/it]
28%|██▊ | 12359/43939 [19:56:31<50:06:58, 5.71s/it]
28%|██▊ | 12360/43939 [19:56:37<50:10:12, 5.72s/it]
28%|██▊ | 12361/43939 [19:56:42<50:03:25, 5.71s/it]
28%|██▊ | 12362/43939 [19:56:48<50:10:36, 5.72s/it]
28%|██▊ | 12363/43939 [19:56:54<50:03:54, 5.71s/it]
28%|██▊ | 12364/43939 [19:57:00<50:00:48, 5.70s/it]
28%|██▊ | 12365/43939 [19:57:05<49:58:58, 5.70s/it]
28%|██▊ | 12366/43939 [19:57:11<49:53:42, 5.69s/it]
28%|██▊ | 12367/43939 [19:57:17<50:01:27, 5.70s/it]
28%|██▊ | 12368/43939 [19:57:22<50:03:24, 5.71s/it]
28%|██▊ | 12369/43939 [19:57:28<50:04:12, 5.71s/it]
28%|██▊ | 12370/43939 [19:57:34<50:05:05, 5.71s/it]
28%|██▊ | 12371/43939 [19:57:40<50:06:55, 5.72s/it]
28%|██▊ | 12372/43939 [19:57:45<50:01:24, 5.70s/it]
28%|██▊ | 12373/43939 [19:57:51<50:04:49, 5.71s/it]
28%|██▊ | 12374/43939 [19:57:57<50:02:48, 5.71s/it]
28%|██▊ | 12375/43939 [19:58:03<50:44:56, 5.79s/it]
28%|██▊ | 12376/43939 [19:58:08<50:30:58, 5.76s/it]
28%|██▊ | 12377/43939 [19:58:14<50:15:48, 5.73s/it]
28%|██▊ | 12378/43939 [19:58:20<50:13:43, 5.73s/it]
28%|██▊ | 12379/43939 [19:58:25<50:11:21, 5.73s/it]
28%|██▊ | 12380/43939 [19:58:31<50:07:53, 5.72s/it]
28%|██▊ | 12381/43939 [19:58:37<50:03:11, 5.71s/it]
28%|██▊ | 12382/43939 [19:58:43<49:53:52, 5.69s/it]
28%|██▊ | 12383/43939 [19:58:48<49:52:58, 5.69s/it]
28%|██▊ | 12384/43939 [19:58:54<49:54:22, 5.69s/it]
28%|██▊ | 12385/43939 [19:59:00<49:53:09, 5.69s/it]
28%|██▊ | 12386/43939 [19:59:05<49:52:35, 5.69s/it]
28%|██▊ | 12387/43939 [19:59:11<49:48:06, 5.68s/it]
28%|██▊ | 12388/43939 [19:59:17<49:52:10, 5.69s/it]
28%|██▊ | 12389/43939 [19:59:22<49:52:08, 5.69s/it]
28%|██▊ | 12390/43939 [19:59:28<49:56:50, 5.70s/it]
28%|██▊ | 12391/43939 [19:59:34<49:57:13, 5.70s/it]
28%|██▊ | 12392/43939 [19:59:39<49:58:11, 5.70s/it]
28%|██▊ | 12393/43939 [19:59:45<49:55:55, 5.70s/it]
28%|██▊ | 12394/43939 [19:59:51<49:54:26, 5.70s/it]
28%|██▊ | 12395/43939 [19:59:57<49:55:10, 5.70s/it]
28%|██▊ | 12396/43939 [20:00:03<50:38:08, 5.78s/it]
28%|██▊ | 12397/43939 [20:00:08<50:30:10, 5.76s/it]
28%|██▊ | 12398/43939 [20:00:14<50:15:06, 5.74s/it]
28%|██▊ | 12399/43939 [20:00:20<50:03:18, 5.71s/it]
28%|██▊ | 12400/43939 [20:00:25<50:01:22, 5.71s/it]
{'loss': 0.9288, 'grad_norm': 1.1765090227127075, 'learning_rate': 4.5111827200411103e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.97, 'epoch': 1.13}
28%|██▊ | 12400/43939 [20:00:25<50:01:22, 5.71s/it]
28%|██▊ | 12401/43939 [20:00:31<50:01:13, 5.71s/it]
28%|██▊ | 12402/43939 [20:00:37<50:07:25, 5.72s/it]
28%|██▊ | 12403/43939 [20:00:42<50:06:27, 5.72s/it]
28%|██▊ | 12404/43939 [20:00:48<50:04:34, 5.72s/it]
28%|██▊ | 12405/43939 [20:00:54<50:02:22, 5.71s/it]
28%|██▊ | 12406/43939 [20:01:00<50:06:53, 5.72s/it]
28%|██▊ | 12407/43939 [20:01:05<50:05:19, 5.72s/it]
28%|██▊ | 12408/43939 [20:01:11<50:05:58, 5.72s/it]
28%|██▊ | 12409/43939 [20:01:17<50:06:35, 5.72s/it]
28%|██▊ | 12410/43939 [20:01:22<50:05:13, 5.72s/it]
28%|██▊ | 12411/43939 [20:01:28<50:01:51, 5.71s/it]
28%|██▊ | 12412/43939 [20:01:34<50:02:22, 5.71s/it]
28%|██▊ | 12413/43939 [20:01:40<50:01:29, 5.71s/it]
28%|██▊ | 12414/43939 [20:01:45<49:57:51, 5.71s/it]
28%|██▊ | 12415/43939 [20:01:51<49:57:46, 5.71s/it]
28%|██▊ | 12416/43939 [20:01:57<49:52:55, 5.70s/it]
28%|██▊ | 12417/43939 [20:02:03<50:22:03, 5.75s/it]
28%|██▊ | 12418/43939 [20:02:08<50:09:28, 5.73s/it]
28%|██▊ | 12419/43939 [20:02:14<50:00:33, 5.71s/it]
28%|██▊ | 12420/43939 [20:02:20<49:55:52, 5.70s/it]
28%|██▊ | 12421/43939 [20:02:25<49:51:09, 5.69s/it]
28%|██▊ | 12422/43939 [20:02:31<49:59:31, 5.71s/it]
28%|██▊ | 12423/43939 [20:02:37<49:53:11, 5.70s/it]
28%|██▊ | 12424/43939 [20:02:42<49:53:55, 5.70s/it]
28%|██▊ | 12425/43939 [20:02:48<49:54:39, 5.70s/it]
28%|██▊ | 12426/43939 [20:02:54<49:52:58, 5.70s/it]
28%|██▊ | 12427/43939 [20:03:00<49:58:06, 5.71s/it]
28%|██▊ | 12428/43939 [20:03:05<49:57:16, 5.71s/it]
28%|██▊ | 12429/43939 [20:03:11<49:56:00, 5.70s/it]
28%|██▊ | 12430/43939 [20:03:17<49:54:18, 5.70s/it]
28%|██▊ | 12431/43939 [20:03:22<49:55:50, 5.70s/it]
28%|██▊ | 12432/43939 [20:03:28<49:54:54, 5.70s/it]
28%|██▊ | 12433/43939 [20:03:34<49:55:14, 5.70s/it]
28%|██▊ | 12434/43939 [20:03:39<49:52:48, 5.70s/it]
28%|██▊ | 12435/43939 [20:03:45<49:49:35, 5.69s/it]
28%|██▊ | 12436/43939 [20:03:51<49:55:42, 5.71s/it]
28%|██▊ | 12437/43939 [20:03:57<49:53:02, 5.70s/it]
28%|██▊ | 12438/43939 [20:04:02<49:47:40, 5.69s/it]
28%|██▊ | 12439/43939 [20:04:08<50:24:13, 5.76s/it]
28%|██▊ | 12440/43939 [20:04:14<50:13:17, 5.74s/it]
28%|██▊ | 12441/43939 [20:04:19<50:05:46, 5.73s/it]
28%|██▊ | 12442/43939 [20:04:25<50:01:06, 5.72s/it]
28%|██▊ | 12443/43939 [20:04:31<49:59:05, 5.71s/it]
28%|██▊ | 12444/43939 [20:04:37<49:57:01, 5.71s/it]
28%|██▊ | 12445/43939 [20:04:42<50:00:18, 5.72s/it]
28%|██▊ | 12446/43939 [20:04:48<49:58:45, 5.71s/it]
28%|██▊ | 12447/43939 [20:04:54<49:56:34, 5.71s/it]
28%|██▊ | 12448/43939 [20:04:59<49:58:24, 5.71s/it]
28%|██▊ | 12449/43939 [20:05:05<49:52:33, 5.70s/it]
28%|██▊ | 12450/43939 [20:05:11<49:51:55, 5.70s/it]
{'loss': 0.9481, 'grad_norm': 1.1673645973205566, 'learning_rate': 4.5052684485507115e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.33, 'epoch': 1.13}
28%|██▊ | 12450/43939 [20:05:11<49:51:55, 5.70s/it]
28%|██▊ | 12451/43939 [20:05:17<49:58:35, 5.71s/it]
28%|██▊ | 12452/43939 [20:05:22<49:57:16, 5.71s/it]
28%|██▊ | 12453/43939 [20:05:28<49:53:30, 5.70s/it]
28%|██▊ | 12454/43939 [20:05:34<49:54:44, 5.71s/it]
28%|██▊ | 12455/43939 [20:05:39<50:06:35, 5.73s/it]
28%|██▊ | 12456/43939 [20:05:45<50:03:25, 5.72s/it]
28%|██▊ | 12457/43939 [20:05:51<50:07:51, 5.73s/it]
28%|██▊ | 12458/43939 [20:05:57<50:08:40, 5.73s/it]
28%|██▊ | 12459/43939 [20:06:03<50:34:51, 5.78s/it]
28%|██▊ | 12460/43939 [20:06:08<50:18:02, 5.75s/it]
28%|██▊ | 12461/43939 [20:06:14<50:11:46, 5.74s/it]
28%|██▊ | 12462/43939 [20:06:20<50:02:19, 5.72s/it]
28%|██▊ | 12463/43939 [20:06:25<50:01:59, 5.72s/it]
28%|██▊ | 12464/43939 [20:06:31<49:57:34, 5.71s/it]
28%|██▊ | 12465/43939 [20:06:37<49:51:40, 5.70s/it]
28%|██▊ | 12466/43939 [20:06:42<49:51:23, 5.70s/it]
28%|██▊ | 12467/43939 [20:06:48<49:55:10, 5.71s/it]
28%|██▊ | 12468/43939 [20:06:54<49:49:24, 5.70s/it]
28%|██▊ | 12469/43939 [20:07:00<49:48:48, 5.70s/it]
28%|██▊ | 12470/43939 [20:07:05<49:54:03, 5.71s/it]
28%|██▊ | 12471/43939 [20:07:11<49:52:19, 5.71s/it]
28%|██▊ | 12472/43939 [20:07:17<50:01:41, 5.72s/it]
28%|██▊ | 12473/43939 [20:07:22<49:53:30, 5.71s/it]
28%|██▊ | 12474/43939 [20:07:28<49:55:07, 5.71s/it]
28%|██▊ | 12475/43939 [20:07:34<49:52:35, 5.71s/it]
28%|██▊ | 12476/43939 [20:07:40<49:56:05, 5.71s/it]
28%|██▊ | 12477/43939 [20:07:45<49:54:48, 5.71s/it]
28%|██▊ | 12478/43939 [20:07:51<49:52:56, 5.71s/it]
28%|██▊ | 12479/43939 [20:07:57<49:55:49, 5.71s/it]
28%|██▊ | 12480/43939 [20:08:03<50:31:15, 5.78s/it]
28%|██▊ | 12481/43939 [20:08:08<50:15:50, 5.75s/it]
28%|██▊ | 12482/43939 [20:08:14<50:04:44, 5.73s/it]
28%|██▊ | 12483/43939 [20:08:20<49:59:43, 5.72s/it]
28%|██▊ | 12484/43939 [20:08:25<50:05:20, 5.73s/it]
28%|██▊ | 12485/43939 [20:08:31<49:58:51, 5.72s/it]
28%|██▊ | 12486/43939 [20:08:37<49:52:02, 5.71s/it]
28%|██▊ | 12487/43939 [20:08:43<49:50:32, 5.70s/it]
28%|██▊ | 12488/43939 [20:08:48<49:50:42, 5.71s/it]
28%|██▊ | 12489/43939 [20:08:54<49:49:51, 5.70s/it]
28%|██▊ | 12490/43939 [20:09:00<49:50:50, 5.71s/it]
28%|██▊ | 12491/43939 [20:09:05<49:50:35, 5.71s/it]
28%|██▊ | 12492/43939 [20:09:11<49:54:54, 5.71s/it]
28%|██▊ | 12493/43939 [20:09:17<49:53:25, 5.71s/it]
28%|██▊ | 12494/43939 [20:09:22<49:45:20, 5.70s/it]
28%|██▊ | 12495/43939 [20:09:28<49:39:22, 5.69s/it]
28%|██▊ | 12496/43939 [20:09:34<49:42:18, 5.69s/it]
28%|██▊ | 12497/43939 [20:09:39<49:43:40, 5.69s/it]
28%|██▊ | 12498/43939 [20:09:45<49:42:09, 5.69s/it]
28%|██▊ | 12499/43939 [20:09:51<49:50:30, 5.71s/it]
28%|██▊ | 12500/43939 [20:09:57<49:53:59, 5.71s/it]
{'loss': 0.9308, 'grad_norm': 0.9426007270812988, 'learning_rate': 4.49932253923822e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.72, 'epoch': 1.14}
28%|██▊ | 12500/43939 [20:09:57<49:53:59, 5.71s/it]
28%|██▊ | 12501/43939 [20:10:02<49:46:52, 5.70s/it]
28%|██▊ | 12502/43939 [20:10:08<50:33:06, 5.79s/it]
28%|██▊ | 12503/43939 [20:10:14<50:16:35, 5.76s/it]
28%|██▊ | 12504/43939 [20:10:20<50:10:36, 5.75s/it]
28%|██▊ | 12505/43939 [20:10:25<50:04:30, 5.73s/it]
28%|██▊ | 12506/43939 [20:10:31<50:03:05, 5.73s/it]
28%|██▊ | 12507/43939 [20:10:37<49:53:22, 5.71s/it]
28%|██▊ | 12508/43939 [20:10:43<49:51:08, 5.71s/it]
28%|██▊ | 12509/43939 [20:10:48<49:50:32, 5.71s/it]
28%|██▊ | 12510/43939 [20:10:54<49:49:38, 5.71s/it]
28%|██▊ | 12511/43939 [20:11:00<49:43:50, 5.70s/it]
28%|██▊ | 12512/43939 [20:11:05<49:46:16, 5.70s/it]
28%|██▊ | 12513/43939 [20:11:11<49:43:19, 5.70s/it]
28%|██▊ | 12514/43939 [20:11:17<49:49:33, 5.71s/it]
28%|██▊ | 12515/43939 [20:11:22<49:47:23, 5.70s/it]
28%|██▊ | 12516/43939 [20:11:28<49:46:04, 5.70s/it]
28%|██▊ | 12517/43939 [20:11:34<49:52:14, 5.71s/it]
28%|██▊ | 12518/43939 [20:11:40<49:48:26, 5.71s/it]
28%|██▊ | 12519/43939 [20:11:45<49:41:11, 5.69s/it]
28%|██▊ | 12520/43939 [20:11:51<49:41:10, 5.69s/it]
28%|██▊ | 12521/43939 [20:11:57<49:44:12, 5.70s/it]
28%|██▊ | 12522/43939 [20:12:03<50:21:02, 5.77s/it]
29%|██▊ | 12523/43939 [20:12:08<50:05:50, 5.74s/it]
29%|██▊ | 12524/43939 [20:12:14<50:03:55, 5.74s/it]
29%|██▊ | 12525/43939 [20:12:20<50:04:19, 5.74s/it]
29%|██▊ | 12526/43939 [20:12:25<50:03:58, 5.74s/it]
29%|██▊ | 12527/43939 [20:12:31<50:01:17, 5.73s/it]
29%|██▊ | 12528/43939 [20:12:37<50:00:49, 5.73s/it]
29%|██▊ | 12529/43939 [20:12:43<49:59:03, 5.73s/it]
29%|██▊ | 12530/43939 [20:12:48<49:54:45, 5.72s/it]
29%|██▊ | 12531/43939 [20:12:54<49:53:26, 5.72s/it]
29%|██▊ | 12532/43939 [20:13:00<49:49:33, 5.71s/it]
29%|██▊ | 12533/43939 [20:13:06<50:09:22, 5.75s/it]
29%|██▊ | 12534/43939 [20:13:11<50:04:54, 5.74s/it]
29%|██▊ | 12535/43939 [20:13:17<49:59:11, 5.73s/it]
29%|██▊ | 12536/43939 [20:13:23<49:48:50, 5.71s/it]
29%|██▊ | 12537/43939 [20:13:28<49:52:07, 5.72s/it]
29%|██▊ | 12538/43939 [20:13:34<49:56:53, 5.73s/it]
29%|██▊ | 12539/43939 [20:13:40<49:52:59, 5.72s/it]
29%|██▊ | 12540/43939 [20:13:46<49:48:19, 5.71s/it]
29%|██▊ | 12541/43939 [20:13:51<49:47:37, 5.71s/it]
29%|██▊ | 12542/43939 [20:13:57<49:47:25, 5.71s/it]
29%|██▊ | 12543/43939 [20:14:03<50:27:06, 5.79s/it]
29%|██▊ | 12544/43939 [20:14:09<50:13:56, 5.76s/it]
29%|██▊ | 12545/43939 [20:14:14<50:11:48, 5.76s/it]
29%|██▊ | 12546/43939 [20:14:20<50:10:10, 5.75s/it]
29%|██▊ | 12547/43939 [20:14:26<50:09:48, 5.75s/it]
29%|██▊ | 12548/43939 [20:14:32<49:58:28, 5.73s/it]
29%|██▊ | 12549/43939 [20:14:37<49:55:21, 5.73s/it]
29%|██▊ | 12550/43939 [20:14:43<49:45:37, 5.71s/it]
{'loss': 0.9325, 'grad_norm': 1.1850651502609253, 'learning_rate': 4.4933450859143285e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.12, 'epoch': 1.14}
29%|██▊ | 12550/43939 [20:14:43<49:45:37, 5.71s/it]
29%|██▊ | 12551/43939 [20:14:49<49:45:14, 5.71s/it]
29%|██▊ | 12552/43939 [20:14:54<49:45:05, 5.71s/it]
29%|██▊ | 12553/43939 [20:15:00<49:47:30, 5.71s/it]
29%|██▊ | 12554/43939 [20:15:06<49:39:31, 5.70s/it]
29%|██▊ | 12555/43939 [20:15:11<49:39:41, 5.70s/it]
29%|██▊ | 12556/43939 [20:15:17<49:41:50, 5.70s/it]
29%|██▊ | 12557/43939 [20:15:23<49:41:49, 5.70s/it]
29%|██▊ | 12558/43939 [20:15:29<49:40:29, 5.70s/it]
29%|██▊ | 12559/43939 [20:15:34<49:47:34, 5.71s/it]
29%|██▊ | 12560/43939 [20:15:40<49:44:29, 5.71s/it]
29%|██▊ | 12561/43939 [20:15:46<49:39:44, 5.70s/it]
29%|██▊ | 12562/43939 [20:15:51<49:35:27, 5.69s/it]
29%|██▊ | 12563/43939 [20:15:57<49:39:19, 5.70s/it]
29%|██▊ | 12564/43939 [20:16:03<49:39:48, 5.70s/it]
29%|██▊ | 12565/43939 [20:16:09<50:32:17, 5.80s/it]
29%|██▊ | 12566/43939 [20:16:14<50:21:21, 5.78s/it]
29%|██▊ | 12567/43939 [20:16:20<50:16:56, 5.77s/it]
29%|██▊ | 12568/43939 [20:16:26<50:05:13, 5.75s/it]
29%|██▊ | 12569/43939 [20:16:32<50:10:09, 5.76s/it]
29%|██▊ | 12570/43939 [20:16:37<50:04:33, 5.75s/it]
29%|██▊ | 12571/43939 [20:16:43<49:52:23, 5.72s/it]
29%|██▊ | 12572/43939 [20:16:49<49:48:21, 5.72s/it]
29%|██▊ | 12573/43939 [20:16:54<49:41:49, 5.70s/it]
29%|██▊ | 12574/43939 [20:17:00<49:39:25, 5.70s/it]
29%|██▊ | 12575/43939 [20:17:06<49:39:52, 5.70s/it]
29%|██▊ | 12576/43939 [20:17:12<49:36:04, 5.69s/it]
29%|██▊ | 12577/43939 [20:17:17<49:39:16, 5.70s/it]
29%|██▊ | 12578/43939 [20:17:23<49:42:45, 5.71s/it]
29%|██▊ | 12579/43939 [20:17:29<49:47:35, 5.72s/it]
29%|██▊ | 12580/43939 [20:17:34<49:44:40, 5.71s/it]
29%|██▊ | 12581/43939 [20:17:40<49:47:04, 5.72s/it]
29%|██▊ | 12582/43939 [20:17:46<49:40:49, 5.70s/it]
29%|██▊ | 12583/43939 [20:17:51<49:35:33, 5.69s/it]
29%|██▊ | 12584/43939 [20:17:57<49:37:51, 5.70s/it]
29%|██▊ | 12585/43939 [20:18:03<50:07:58, 5.76s/it]
29%|██▊ | 12586/43939 [20:18:09<49:56:49, 5.73s/it]
29%|██▊ | 12587/43939 [20:18:14<49:49:32, 5.72s/it]
29%|██▊ | 12588/43939 [20:18:20<49:40:31, 5.70s/it]
29%|██▊ | 12589/43939 [20:18:26<49:40:21, 5.70s/it]
29%|██▊ | 12590/43939 [20:18:32<49:42:27, 5.71s/it]
29%|██▊ | 12591/43939 [20:18:37<49:42:16, 5.71s/it]
29%|██▊ | 12592/43939 [20:18:43<49:43:51, 5.71s/it]
29%|██▊ | 12593/43939 [20:18:49<49:48:20, 5.72s/it]
29%|██▊ | 12594/43939 [20:18:54<49:47:22, 5.72s/it]
29%|██▊ | 12595/43939 [20:19:00<49:39:27, 5.70s/it]
29%|██▊ | 12596/43939 [20:19:06<49:34:04, 5.69s/it]
29%|██▊ | 12597/43939 [20:19:11<49:29:51, 5.69s/it]
29%|██▊ | 12598/43939 [20:19:17<49:34:22, 5.69s/it]
29%|██▊ | 12599/43939 [20:19:23<49:33:48, 5.69s/it]
29%|██▊ | 12600/43939 [20:19:29<49:34:54, 5.70s/it]
{'loss': 0.9442, 'grad_norm': 1.0401122570037842, 'learning_rate': 4.48733618288741e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.32, 'epoch': 1.15}
29%|██▊ | 12600/43939 [20:19:29<49:34:54, 5.70s/it]
29%|██▊ | 12601/43939 [20:19:34<49:38:04, 5.70s/it]
29%|██▊ | 12602/43939 [20:19:40<49:31:30, 5.69s/it]
29%|██▊ | 12603/43939 [20:19:46<49:32:56, 5.69s/it]
29%|██▊ | 12604/43939 [20:19:51<49:29:52, 5.69s/it]
29%|██▊ | 12605/43939 [20:19:57<49:32:14, 5.69s/it]
29%|██▊ | 12606/43939 [20:20:03<50:12:08, 5.77s/it]
29%|██▊ | 12607/43939 [20:20:09<50:20:30, 5.78s/it]
29%|██▊ | 12608/43939 [20:20:14<50:05:59, 5.76s/it]
29%|██▊ | 12609/43939 [20:20:20<50:04:16, 5.75s/it]
29%|██▊ | 12610/43939 [20:20:26<49:58:01, 5.74s/it]
29%|██▊ | 12611/43939 [20:20:32<49:54:32, 5.74s/it]
29%|██▊ | 12612/43939 [20:20:37<49:55:09, 5.74s/it]
29%|██▊ | 12613/43939 [20:20:43<49:44:50, 5.72s/it]
29%|██▊ | 12614/43939 [20:20:49<49:35:10, 5.70s/it]
29%|██▊ | 12615/43939 [20:20:54<49:38:52, 5.71s/it]
29%|██▊ | 12616/43939 [20:21:00<49:35:37, 5.70s/it]
29%|██▊ | 12617/43939 [20:21:06<49:40:48, 5.71s/it]
29%|██▊ | 12618/43939 [20:21:12<49:35:11, 5.70s/it]
29%|██▊ | 12619/43939 [20:21:17<49:41:14, 5.71s/it]
29%|██▊ | 12620/43939 [20:21:23<49:35:38, 5.70s/it]
29%|██▊ | 12621/43939 [20:21:29<49:39:18, 5.71s/it]
29%|██▊ | 12622/43939 [20:21:34<49:41:07, 5.71s/it]
29%|██▊ | 12623/43939 [20:21:40<49:37:09, 5.70s/it]
29%|██▊ | 12624/43939 [20:21:46<49:31:21, 5.69s/it]
29%|██▊ | 12625/43939 [20:21:51<49:30:54, 5.69s/it]
29%|██▊ | 12626/43939 [20:21:57<49:27:26, 5.69s/it]
29%|██▊ | 12627/43939 [20:22:03<49:33:21, 5.70s/it]
29%|██▊ | 12628/43939 [20:22:09<50:06:34, 5.76s/it]
29%|██▊ | 12629/43939 [20:22:14<50:04:21, 5.76s/it]
29%|██▊ | 12630/43939 [20:22:20<50:00:43, 5.75s/it]
29%|██▊ | 12631/43939 [20:22:26<49:59:10, 5.75s/it]
29%|██▊ | 12632/43939 [20:22:32<49:47:07, 5.72s/it]
29%|██▉ | 12633/43939 [20:22:37<49:43:39, 5.72s/it]
29%|██▉ | 12634/43939 [20:22:43<49:37:39, 5.71s/it]
29%|██▉ | 12635/43939 [20:22:49<49:27:55, 5.69s/it]
29%|██▉ | 12636/43939 [20:22:54<49:35:31, 5.70s/it]
29%|██▉ | 12637/43939 [20:23:00<49:34:44, 5.70s/it]
29%|██▉ | 12638/43939 [20:23:06<49:41:03, 5.71s/it]
29%|██▉ | 12639/43939 [20:23:12<49:40:31, 5.71s/it]
29%|██▉ | 12640/43939 [20:23:17<49:37:54, 5.71s/it]
29%|██▉ | 12641/43939 [20:23:23<49:37:23, 5.71s/it]
29%|██▉ | 12642/43939 [20:23:29<49:37:21, 5.71s/it]
29%|██▉ | 12643/43939 [20:23:34<49:42:29, 5.72s/it]
29%|██▉ | 12644/43939 [20:23:40<49:35:18, 5.70s/it]
29%|██▉ | 12645/43939 [20:23:46<49:32:16, 5.70s/it]
29%|██▉ | 12646/43939 [20:23:51<49:29:54, 5.69s/it]
29%|██▉ | 12647/43939 [20:23:57<49:31:05, 5.70s/it]
29%|██▉ | 12648/43939 [20:24:03<49:33:26, 5.70s/it]
29%|██▉ | 12649/43939 [20:24:09<50:16:09, 5.78s/it]
29%|██▉ | 12650/43939 [20:24:15<50:06:59, 5.77s/it]
{'loss': 0.9757, 'grad_norm': 2.068197011947632, 'learning_rate': 4.481295924962031e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.07, 'epoch': 1.15}
29%|██▉ | 12650/43939 [20:24:15<50:06:59, 5.77s/it]
29%|██▉ | 12651/43939 [20:24:20<49:57:39, 5.75s/it]
29%|██▉ | 12652/43939 [20:24:26<49:54:10, 5.74s/it]
29%|██▉ | 12653/43939 [20:24:32<49:55:40, 5.75s/it]
29%|██▉ | 12654/43939 [20:24:37<49:49:21, 5.73s/it]
29%|██▉ | 12655/43939 [20:24:43<49:43:09, 5.72s/it]
29%|██▉ | 12656/43939 [20:24:49<49:40:34, 5.72s/it]
29%|██▉ | 12657/43939 [20:24:55<49:38:07, 5.71s/it]
29%|██▉ | 12658/43939 [20:25:00<49:36:46, 5.71s/it]
29%|██▉ | 12659/43939 [20:25:06<49:31:36, 5.70s/it]
29%|██▉ | 12660/43939 [20:25:12<49:39:38, 5.72s/it]
29%|██▉ | 12661/43939 [20:25:17<49:35:06, 5.71s/it]
29%|██▉ | 12662/43939 [20:25:23<49:33:48, 5.70s/it]
29%|██▉ | 12663/43939 [20:25:29<49:34:25, 5.71s/it]
29%|██▉ | 12664/43939 [20:25:34<49:30:13, 5.70s/it]
29%|██▉ | 12665/43939 [20:25:40<49:34:58, 5.71s/it]
29%|██▉ | 12666/43939 [20:25:46<49:40:34, 5.72s/it]
29%|██▉ | 12667/43939 [20:25:52<49:30:57, 5.70s/it]
29%|██▉ | 12668/43939 [20:25:57<49:28:24, 5.70s/it]
29%|██▉ | 12669/43939 [20:26:03<49:28:43, 5.70s/it]
29%|██▉ | 12670/43939 [20:26:09<50:05:50, 5.77s/it]
29%|██▉ | 12671/43939 [20:26:15<49:57:02, 5.75s/it]
29%|██▉ | 12672/43939 [20:26:20<49:55:19, 5.75s/it]
29%|██▉ | 12673/43939 [20:26:26<49:45:25, 5.73s/it]
29%|██▉ | 12674/43939 [20:26:32<49:55:44, 5.75s/it]
29%|██▉ | 12675/43939 [20:26:38<49:51:15, 5.74s/it]
29%|██▉ | 12676/43939 [20:26:43<49:50:23, 5.74s/it]
29%|██▉ | 12677/43939 [20:26:49<49:54:10, 5.75s/it]
29%|██▉ | 12678/43939 [20:26:55<49:47:19, 5.73s/it]
29%|██▉ | 12679/43939 [20:27:00<49:42:41, 5.72s/it]
29%|██▉ | 12680/43939 [20:27:06<49:46:25, 5.73s/it]
29%|██▉ | 12681/43939 [20:27:12<49:41:15, 5.72s/it]
29%|██▉ | 12682/43939 [20:27:18<49:40:51, 5.72s/it]
29%|██▉ | 12683/43939 [20:27:23<49:34:22, 5.71s/it]
29%|██▉ | 12684/43939 [20:27:29<49:32:17, 5.71s/it]
29%|██▉ | 12685/43939 [20:27:35<49:33:13, 5.71s/it]
29%|██▉ | 12686/43939 [20:27:40<49:30:57, 5.70s/it]
29%|██▉ | 12687/43939 [20:27:46<49:32:44, 5.71s/it]
29%|██▉ | 12688/43939 [20:27:52<49:40:11, 5.72s/it]
29%|██▉ | 12689/43939 [20:27:58<49:39:37, 5.72s/it]
29%|██▉ | 12690/43939 [20:28:03<49:31:21, 5.71s/it]
29%|██▉ | 12691/43939 [20:28:09<50:03:47, 5.77s/it]
29%|██▉ | 12692/43939 [20:28:15<49:56:26, 5.75s/it]
29%|██▉ | 12693/43939 [20:28:21<49:49:13, 5.74s/it]
29%|██▉ | 12694/43939 [20:28:26<49:42:50, 5.73s/it]
29%|██▉ | 12695/43939 [20:28:32<49:35:37, 5.71s/it]
29%|██▉ | 12696/43939 [20:28:38<49:32:54, 5.71s/it]
29%|██▉ | 12697/43939 [20:28:43<49:34:06, 5.71s/it]
29%|██▉ | 12698/43939 [20:28:49<49:36:32, 5.72s/it]
29%|██▉ | 12699/43939 [20:28:55<49:30:43, 5.71s/it]
29%|██▉ | 12700/43939 [20:29:01<49:35:34, 5.72s/it]
{'loss': 0.9359, 'grad_norm': 1.1467478275299072, 'learning_rate': 4.4752244074374536e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 511.2, 'epoch': 1.16}
29%|██▉ | 12700/43939 [20:29:01<49:35:34, 5.72s/it]
29%|██▉ | 12701/43939 [20:29:06<49:35:01, 5.71s/it]
29%|██▉ | 12702/43939 [20:29:12<49:44:38, 5.73s/it]
29%|██▉ | 12703/43939 [20:29:18<49:49:44, 5.74s/it]
29%|██▉ | 12704/43939 [20:29:24<49:47:29, 5.74s/it]
29%|██▉ | 12705/43939 [20:29:29<49:42:29, 5.73s/it]
29%|██▉ | 12706/43939 [20:29:35<49:36:06, 5.72s/it]
29%|██▉ | 12707/43939 [20:29:41<49:25:30, 5.70s/it]
29%|██▉ | 12708/43939 [20:29:46<49:18:23, 5.68s/it]
29%|██▉ | 12709/43939 [20:29:52<49:25:34, 5.70s/it]
29%|██▉ | 12710/43939 [20:29:58<49:18:51, 5.68s/it]
29%|██▉ | 12711/43939 [20:30:03<49:21:39, 5.69s/it]
29%|██▉ | 12712/43939 [20:30:09<50:00:06, 5.76s/it]
29%|██▉ | 12713/43939 [20:30:15<49:45:16, 5.74s/it]
29%|██▉ | 12714/43939 [20:30:21<49:34:54, 5.72s/it]
29%|██▉ | 12715/43939 [20:30:26<49:24:00, 5.70s/it]
29%|██▉ | 12716/43939 [20:30:32<49:25:11, 5.70s/it]
29%|██▉ | 12717/43939 [20:30:38<49:25:17, 5.70s/it]
29%|██▉ | 12718/43939 [20:30:43<49:23:13, 5.69s/it]
29%|██▉ | 12719/43939 [20:30:49<49:25:45, 5.70s/it]
29%|██▉ | 12720/43939 [20:30:55<49:23:37, 5.70s/it]
29%|██▉ | 12721/43939 [20:31:00<49:23:22, 5.70s/it]
29%|██▉ | 12722/43939 [20:31:06<49:25:54, 5.70s/it]
29%|██▉ | 12723/43939 [20:31:12<49:23:41, 5.70s/it]
29%|██▉ | 12724/43939 [20:31:18<49:24:15, 5.70s/it]
29%|██▉ | 12725/43939 [20:31:23<49:27:05, 5.70s/it]
29%|██▉ | 12726/43939 [20:31:29<49:23:15, 5.70s/it]
29%|██▉ | 12727/43939 [20:31:35<49:25:34, 5.70s/it]
29%|██▉ | 12728/43939 [20:31:40<49:26:47, 5.70s/it]
29%|██▉ | 12729/43939 [20:31:46<49:29:33, 5.71s/it]
29%|██▉ | 12730/43939 [20:31:52<49:28:18, 5.71s/it]
29%|██▉ | 12731/43939 [20:31:57<49:21:15, 5.69s/it]
29%|██▉ | 12732/43939 [20:32:03<49:32:47, 5.72s/it]
29%|██▉ | 12733/43939 [20:32:09<50:03:09, 5.77s/it]
29%|██▉ | 12734/43939 [20:32:15<49:52:26, 5.75s/it]
29%|██▉ | 12735/43939 [20:32:21<49:47:09, 5.74s/it]
29%|██▉ | 12736/43939 [20:32:26<49:36:29, 5.72s/it]
29%|██▉ | 12737/43939 [20:32:32<49:27:34, 5.71s/it]
29%|██▉ | 12738/43939 [20:32:38<49:30:28, 5.71s/it]
29%|██▉ | 12739/43939 [20:32:43<49:25:07, 5.70s/it]
29%|██▉ | 12740/43939 [20:32:49<49:27:54, 5.71s/it]
29%|██▉ | 12741/43939 [20:32:55<49:27:50, 5.71s/it]
29%|██▉ | 12742/43939 [20:33:00<49:26:53, 5.71s/it]
29%|██▉ | 12743/43939 [20:33:06<49:34:33, 5.72s/it]
29%|██▉ | 12744/43939 [20:33:12<49:30:00, 5.71s/it]
29%|██▉ | 12745/43939 [20:33:18<49:23:28, 5.70s/it]
29%|██▉ | 12746/43939 [20:33:23<49:22:32, 5.70s/it]
29%|██▉ | 12747/43939 [20:33:29<49:26:09, 5.71s/it]
29%|██▉ | 12748/43939 [20:33:35<49:26:26, 5.71s/it]
29%|██▉ | 12749/43939 [20:33:40<49:26:20, 5.71s/it]
29%|██▉ | 12750/43939 [20:33:46<49:23:06, 5.70s/it]
{'loss': 0.9311, 'grad_norm': 1.0898977518081665, 'learning_rate': 4.469121726106136e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.45, 'epoch': 1.16}
29%|██▉ | 12750/43939 [20:33:46<49:23:06, 5.70s/it]
29%|██▉ | 12751/43939 [20:33:52<49:18:09, 5.69s/it]
29%|██▉ | 12752/43939 [20:33:57<49:14:10, 5.68s/it]
29%|██▉ | 12753/43939 [20:34:03<49:17:46, 5.69s/it]
29%|██▉ | 12754/43939 [20:34:09<50:01:01, 5.77s/it]
29%|██▉ | 12755/43939 [20:34:15<49:49:07, 5.75s/it]
29%|██▉ | 12756/43939 [20:34:20<49:40:27, 5.73s/it]
29%|██▉ | 12757/43939 [20:34:26<49:27:31, 5.71s/it]
29%|██▉ | 12758/43939 [20:34:32<49:25:02, 5.71s/it]
29%|██▉ | 12759/43939 [20:34:38<49:27:04, 5.71s/it]
29%|██▉ | 12760/43939 [20:34:43<49:28:30, 5.71s/it]
29%|██▉ | 12761/43939 [20:34:49<49:32:39, 5.72s/it]
29%|██▉ | 12762/43939 [20:34:55<49:30:52, 5.72s/it]
29%|██▉ | 12763/43939 [20:35:00<49:33:11, 5.72s/it]
29%|██▉ | 12764/43939 [20:35:06<49:32:29, 5.72s/it]
29%|██▉ | 12765/43939 [20:35:12<49:28:03, 5.71s/it]
29%|██▉ | 12766/43939 [20:35:18<49:26:29, 5.71s/it]
29%|██▉ | 12767/43939 [20:35:23<49:24:21, 5.71s/it]
29%|██▉ | 12768/43939 [20:35:29<49:28:02, 5.71s/it]
29%|██▉ | 12769/43939 [20:35:35<49:23:39, 5.70s/it]
29%|██▉ | 12770/43939 [20:35:40<49:25:00, 5.71s/it]
29%|██▉ | 12771/43939 [20:35:46<49:29:16, 5.72s/it]
29%|██▉ | 12772/43939 [20:35:52<49:38:50, 5.73s/it]
29%|██▉ | 12773/43939 [20:35:58<49:31:45, 5.72s/it]
29%|██▉ | 12774/43939 [20:36:03<49:30:50, 5.72s/it]
29%|██▉ | 12775/43939 [20:36:09<49:35:10, 5.73s/it]
29%|██▉ | 12776/43939 [20:36:15<50:12:00, 5.80s/it]
29%|██▉ | 12777/43939 [20:36:21<50:06:29, 5.79s/it]
29%|██▉ | 12778/43939 [20:36:26<49:52:58, 5.76s/it]
29%|██▉ | 12779/43939 [20:36:32<49:43:45, 5.75s/it]
29%|██▉ | 12780/43939 [20:36:38<49:45:57, 5.75s/it]
29%|██▉ | 12781/43939 [20:36:44<49:43:43, 5.75s/it]
29%|██▉ | 12782/43939 [20:36:49<49:33:02, 5.73s/it]
29%|██▉ | 12783/43939 [20:36:55<49:24:44, 5.71s/it]
29%|██▉ | 12784/43939 [20:37:01<49:22:09, 5.70s/it]
29%|██▉ | 12785/43939 [20:37:06<49:23:16, 5.71s/it]
29%|██▉ | 12786/43939 [20:37:12<49:26:47, 5.71s/it]
29%|██▉ | 12787/43939 [20:37:18<49:25:11, 5.71s/it]
29%|██▉ | 12788/43939 [20:37:24<49:20:01, 5.70s/it]
29%|██▉ | 12789/43939 [20:37:29<49:21:19, 5.70s/it]
29%|██▉ | 12790/43939 [20:37:35<49:26:05, 5.71s/it]
29%|██▉ | 12791/43939 [20:37:41<49:18:37, 5.70s/it]
29%|██▉ | 12792/43939 [20:37:46<49:25:21, 5.71s/it]
29%|██▉ | 12793/43939 [20:37:52<49:26:19, 5.71s/it]
29%|██▉ | 12794/43939 [20:37:58<49:25:42, 5.71s/it]
29%|██▉ | 12795/43939 [20:38:04<49:21:55, 5.71s/it]
29%|██▉ | 12796/43939 [20:38:09<49:25:24, 5.71s/it]
29%|██▉ | 12797/43939 [20:38:15<49:54:11, 5.77s/it]
29%|██▉ | 12798/43939 [20:38:21<49:41:47, 5.75s/it]
29%|██▉ | 12799/43939 [20:38:26<49:28:39, 5.72s/it]
29%|██▉ | 12800/43939 [20:38:32<49:27:37, 5.72s/it]
{'loss': 0.9584, 'grad_norm': 1.2766635417938232, 'learning_rate': 4.462987977252216e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 517.46, 'epoch': 1.17}
29%|██▉ | 12800/43939 [20:38:32<49:27:37, 5.72s/it]
29%|██▉ | 12801/43939 [20:38:38<49:27:19, 5.72s/it]
29%|██▉ | 12802/43939 [20:38:44<49:25:20, 5.71s/it]
29%|██▉ | 12803/43939 [20:38:49<49:24:34, 5.71s/it]
29%|██▉ | 12804/43939 [20:38:55<49:11:58, 5.69s/it]
29%|██▉ | 12805/43939 [20:39:01<49:10:00, 5.69s/it]
29%|██▉ | 12806/43939 [20:39:06<49:14:16, 5.69s/it]
29%|██▉ | 12807/43939 [20:39:12<49:17:12, 5.70s/it]
29%|██▉ | 12808/43939 [20:39:18<49:23:38, 5.71s/it]
29%|██▉ | 12809/43939 [20:39:24<49:21:51, 5.71s/it]
29%|██▉ | 12810/43939 [20:39:29<49:22:29, 5.71s/it]
29%|██▉ | 12811/43939 [20:39:35<49:17:38, 5.70s/it]
29%|██▉ | 12812/43939 [20:39:41<49:18:59, 5.70s/it]
29%|██▉ | 12813/43939 [20:39:46<49:20:28, 5.71s/it]
29%|██▉ | 12814/43939 [20:39:52<49:12:15, 5.69s/it]
29%|██▉ | 12815/43939 [20:39:58<49:16:12, 5.70s/it]
29%|██▉ | 12816/43939 [20:40:03<49:20:43, 5.71s/it]
29%|██▉ | 12817/43939 [20:40:09<49:22:26, 5.71s/it]
29%|██▉ | 12818/43939 [20:40:15<49:49:35, 5.76s/it]
29%|██▉ | 12819/43939 [20:40:21<49:37:39, 5.74s/it]
29%|██▉ | 12820/43939 [20:40:26<49:30:27, 5.73s/it]
29%|██▉ | 12821/43939 [20:40:32<49:24:39, 5.72s/it]
29%|██▉ | 12822/43939 [20:40:38<49:23:48, 5.71s/it]
29%|██▉ | 12823/43939 [20:40:44<49:23:08, 5.71s/it]
29%|██▉ | 12824/43939 [20:40:49<49:16:04, 5.70s/it]
29%|██▉ | 12825/43939 [20:40:55<49:16:32, 5.70s/it]
29%|██▉ | 12826/43939 [20:41:01<49:17:02, 5.70s/it]
29%|██▉ | 12827/43939 [20:41:06<49:22:35, 5.71s/it]
29%|██▉ | 12828/43939 [20:41:12<49:16:09, 5.70s/it]
29%|██▉ | 12829/43939 [20:41:18<49:20:58, 5.71s/it]
29%|██▉ | 12830/43939 [20:41:23<49:22:24, 5.71s/it]
29%|██▉ | 12831/43939 [20:41:29<49:12:25, 5.69s/it]
29%|██▉ | 12832/43939 [20:41:35<49:14:13, 5.70s/it]
29%|██▉ | 12833/43939 [20:41:41<49:15:16, 5.70s/it]
29%|██▉ | 12834/43939 [20:41:46<49:12:47, 5.70s/it]
29%|██▉ | 12835/43939 [20:41:52<49:12:38, 5.70s/it]
29%|██▉ | 12836/43939 [20:41:58<49:17:48, 5.71s/it]
29%|██▉ | 12837/43939 [20:42:03<49:16:21, 5.70s/it]
29%|██▉ | 12838/43939 [20:42:09<49:12:14, 5.70s/it]
29%|██▉ | 12839/43939 [20:42:15<49:44:36, 5.76s/it]
29%|██▉ | 12840/43939 [20:42:21<49:38:00, 5.75s/it]
29%|██▉ | 12841/43939 [20:42:26<49:28:58, 5.73s/it]
29%|██▉ | 12842/43939 [20:42:32<49:30:32, 5.73s/it]
29%|██▉ | 12843/43939 [20:42:38<49:21:40, 5.71s/it]
29%|██▉ | 12844/43939 [20:42:43<49:16:50, 5.71s/it]
29%|██▉ | 12845/43939 [20:42:49<49:14:04, 5.70s/it]
29%|██▉ | 12846/43939 [20:42:55<49:17:48, 5.71s/it]
29%|██▉ | 12847/43939 [20:43:01<49:20:20, 5.71s/it]
29%|██▉ | 12848/43939 [20:43:06<49:15:00, 5.70s/it]
29%|██▉ | 12849/43939 [20:43:12<49:14:36, 5.70s/it]
29%|██▉ | 12850/43939 [20:43:18<49:18:30, 5.71s/it]
{'loss': 0.9324, 'grad_norm': 1.1439895629882812, 'learning_rate': 4.456823257649996e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.99, 'epoch': 1.17}
29%|██▉ | 12850/43939 [20:43:18<49:18:30, 5.71s/it]
29%|██▉ | 12851/43939 [20:43:23<49:11:57, 5.70s/it]
29%|██▉ | 12852/43939 [20:43:29<49:08:04, 5.69s/it]
29%|██▉ | 12853/43939 [20:43:35<49:07:10, 5.69s/it]
29%|██▉ | 12854/43939 [20:43:40<49:09:56, 5.69s/it]
29%|██▉ | 12855/43939 [20:43:46<49:09:50, 5.69s/it]
29%|██▉ | 12856/43939 [20:43:52<49:12:55, 5.70s/it]
29%|██▉ | 12857/43939 [20:43:58<49:18:44, 5.71s/it]
29%|██▉ | 12858/43939 [20:44:03<49:16:55, 5.71s/it]
29%|██▉ | 12859/43939 [20:44:09<49:13:25, 5.70s/it]
29%|██▉ | 12860/43939 [20:44:15<49:14:14, 5.70s/it]
29%|██▉ | 12861/43939 [20:44:21<49:48:51, 5.77s/it]
29%|██▉ | 12862/43939 [20:44:26<49:38:55, 5.75s/it]
29%|██▉ | 12863/43939 [20:44:32<49:30:45, 5.74s/it]
29%|██▉ | 12864/43939 [20:44:38<49:35:37, 5.75s/it]
29%|██▉ | 12865/43939 [20:44:43<49:30:56, 5.74s/it]
29%|██▉ | 12866/43939 [20:44:49<49:25:35, 5.73s/it]
29%|██▉ | 12867/43939 [20:44:55<49:21:33, 5.72s/it]
29%|██▉ | 12868/43939 [20:45:01<49:15:05, 5.71s/it]
29%|██▉ | 12869/43939 [20:45:06<49:10:53, 5.70s/it]
29%|██▉ | 12870/43939 [20:45:12<49:06:06, 5.69s/it]
29%|██▉ | 12871/43939 [20:45:18<49:17:41, 5.71s/it]
29%|██▉ | 12872/43939 [20:45:23<49:14:41, 5.71s/it]
29%|██▉ | 12873/43939 [20:45:29<49:12:56, 5.70s/it]
29%|██▉ | 12874/43939 [20:45:35<49:19:14, 5.72s/it]
29%|██▉ | 12875/43939 [20:45:40<49:10:10, 5.70s/it]
29%|██▉ | 12876/43939 [20:45:46<49:08:57, 5.70s/it]
29%|██▉ | 12877/43939 [20:45:52<49:06:05, 5.69s/it]
29%|██▉ | 12878/43939 [20:45:58<49:10:03, 5.70s/it]
29%|██▉ | 12879/43939 [20:46:03<49:10:01, 5.70s/it]
29%|██▉ | 12880/43939 [20:46:09<49:20:27, 5.72s/it]
29%|██▉ | 12881/43939 [20:46:15<49:19:29, 5.72s/it]
29%|██▉ | 12882/43939 [20:46:21<49:56:37, 5.79s/it]
29%|██▉ | 12883/43939 [20:46:26<49:37:42, 5.75s/it]
29%|██▉ | 12884/43939 [20:46:32<49:35:32, 5.75s/it]
29%|██▉ | 12885/43939 [20:46:38<49:32:35, 5.74s/it]
29%|██▉ | 12886/43939 [20:46:44<49:28:43, 5.74s/it]
29%|██▉ | 12887/43939 [20:46:49<49:25:59, 5.73s/it]
29%|██▉ | 12888/43939 [20:46:55<49:22:14, 5.72s/it]
29%|██▉ | 12889/43939 [20:47:01<49:15:18, 5.71s/it]
29%|██▉ | 12890/43939 [20:47:06<49:20:46, 5.72s/it]
29%|██▉ | 12891/43939 [20:47:12<49:14:17, 5.71s/it]
29%|██▉ | 12892/43939 [20:47:18<49:12:21, 5.71s/it]
29%|██▉ | 12893/43939 [20:47:23<49:07:46, 5.70s/it]
29%|██▉ | 12894/43939 [20:47:29<49:18:53, 5.72s/it]
29%|██▉ | 12895/43939 [20:47:35<49:20:22, 5.72s/it]
29%|██▉ | 12896/43939 [20:47:41<49:12:36, 5.71s/it]
29%|██▉ | 12897/43939 [20:47:46<49:11:37, 5.71s/it]
29%|██▉ | 12898/43939 [20:47:52<49:10:20, 5.70s/it]
29%|██▉ | 12899/43939 [20:47:58<49:07:06, 5.70s/it]
29%|██▉ | 12900/43939 [20:48:03<49:08:13, 5.70s/it]
{'loss': 0.9518, 'grad_norm': 1.1042931079864502, 'learning_rate': 4.4506276645624164e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.33, 'epoch': 1.17}
29%|██▉ | 12900/43939 [20:48:03<49:08:13, 5.70s/it]
29%|██▉ | 12901/43939 [20:48:09<49:07:23, 5.70s/it]
29%|██▉ | 12902/43939 [20:48:15<49:12:22, 5.71s/it]
29%|██▉ | 12903/43939 [20:48:21<49:45:01, 5.77s/it]
29%|██▉ | 12904/43939 [20:48:26<49:26:53, 5.74s/it]
29%|██▉ | 12905/43939 [20:48:32<49:19:52, 5.72s/it]
29%|██▉ | 12906/43939 [20:48:38<49:08:33, 5.70s/it]
29%|██▉ | 12907/43939 [20:48:43<49:09:51, 5.70s/it]
29%|██▉ | 12908/43939 [20:48:49<49:07:27, 5.70s/it]
29%|██▉ | 12909/43939 [20:48:55<49:10:24, 5.70s/it]
29%|██▉ | 12910/43939 [20:49:01<49:17:03, 5.72s/it]
29%|██▉ | 12911/43939 [20:49:06<49:09:27, 5.70s/it]
29%|██▉ | 12912/43939 [20:49:12<49:06:09, 5.70s/it]
29%|██▉ | 12913/43939 [20:49:18<49:01:10, 5.69s/it]
29%|██▉ | 12914/43939 [20:49:23<49:01:01, 5.69s/it]
29%|██▉ | 12915/43939 [20:49:29<49:03:48, 5.69s/it]
29%|██▉ | 12916/43939 [20:49:35<48:56:58, 5.68s/it]
29%|██▉ | 12917/43939 [20:49:40<49:04:58, 5.70s/it]
29%|██▉ | 12918/43939 [20:49:46<49:03:47, 5.69s/it]
29%|██▉ | 12919/43939 [20:49:52<49:14:52, 5.72s/it]
29%|██▉ | 12920/43939 [20:49:58<49:16:30, 5.72s/it]
29%|██▉ | 12921/43939 [20:50:03<49:17:38, 5.72s/it]
29%|██▉ | 12922/43939 [20:50:09<49:15:54, 5.72s/it]
29%|██▉ | 12923/43939 [20:50:15<49:12:22, 5.71s/it]
29%|██▉ | 12924/43939 [20:50:21<49:42:06, 5.77s/it]
29%|██▉ | 12925/43939 [20:50:26<49:28:13, 5.74s/it]
29%|██▉ | 12926/43939 [20:50:32<49:20:02, 5.73s/it]
29%|██▉ | 12927/43939 [20:50:38<49:28:27, 5.74s/it]
29%|██▉ | 12928/43939 [20:50:43<49:20:35, 5.73s/it]
29%|██▉ | 12929/43939 [20:50:49<49:26:29, 5.74s/it]
29%|██▉ | 12930/43939 [20:50:55<49:14:03, 5.72s/it]
29%|██▉ | 12931/43939 [20:51:01<49:06:41, 5.70s/it]
29%|██▉ | 12932/43939 [20:51:06<49:15:48, 5.72s/it]
29%|██▉ | 12933/43939 [20:51:12<49:14:26, 5.72s/it]
29%|██▉ | 12934/43939 [20:51:18<49:15:37, 5.72s/it]
29%|██▉ | 12935/43939 [20:51:23<49:14:30, 5.72s/it]
29%|██▉ | 12936/43939 [20:51:29<49:09:07, 5.71s/it]
29%|██▉ | 12937/43939 [20:51:35<49:03:19, 5.70s/it]
29%|██▉ | 12938/43939 [20:51:41<49:02:52, 5.70s/it]
29%|██▉ | 12939/43939 [20:51:46<49:04:40, 5.70s/it]
29%|██▉ | 12940/43939 [20:51:52<49:10:35, 5.71s/it]
29%|██▉ | 12941/43939 [20:51:58<49:09:04, 5.71s/it]
29%|██▉ | 12942/43939 [20:52:03<49:06:48, 5.70s/it]
29%|██▉ | 12943/43939 [20:52:09<49:06:52, 5.70s/it]
29%|██▉ | 12944/43939 [20:52:15<49:05:49, 5.70s/it]
29%|██▉ | 12945/43939 [20:52:20<49:04:17, 5.70s/it]
29%|██▉ | 12946/43939 [20:52:26<49:48:05, 5.78s/it]
29%|██▉ | 12947/43939 [20:52:32<49:36:54, 5.76s/it]
29%|██▉ | 12948/43939 [20:52:38<49:23:22, 5.74s/it]
29%|██▉ | 12949/43939 [20:52:44<49:17:30, 5.73s/it]
29%|██▉ | 12950/43939 [20:52:49<49:16:53, 5.73s/it]
{'loss': 0.9297, 'grad_norm': 1.3344614505767822, 'learning_rate': 4.4444012957395156e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.54, 'epoch': 1.18}
29%|██▉ | 12950/43939 [20:52:49<49:16:53, 5.73s/it]
29%|██▉ | 12951/43939 [20:52:55<49:17:53, 5.73s/it]
29%|██▉ | 12952/43939 [20:53:01<49:10:12, 5.71s/it]
29%|██▉ | 12953/43939 [20:53:06<49:14:18, 5.72s/it]
29%|██▉ | 12954/43939 [20:53:12<49:10:24, 5.71s/it]
29%|██▉ | 12955/43939 [20:53:18<49:11:36, 5.72s/it]
29%|██▉ | 12956/43939 [20:53:24<49:16:33, 5.73s/it]
29%|██▉ | 12957/43939 [20:53:29<49:10:00, 5.71s/it]
29%|██▉ | 12958/43939 [20:53:35<49:05:46, 5.70s/it]
29%|██▉ | 12959/43939 [20:53:41<49:03:37, 5.70s/it]
29%|██▉ | 12960/43939 [20:53:46<49:09:49, 5.71s/it]
29%|██▉ | 12961/43939 [20:53:52<49:06:30, 5.71s/it]
29%|██▉ | 12962/43939 [20:53:58<49:10:26, 5.71s/it]
30%|██▉ | 12963/43939 [20:54:04<49:12:15, 5.72s/it]
30%|██▉ | 12964/43939 [20:54:09<49:08:31, 5.71s/it]
30%|██▉ | 12965/43939 [20:54:15<49:10:57, 5.72s/it]
30%|██▉ | 12966/43939 [20:54:21<49:12:46, 5.72s/it]
30%|██▉ | 12967/43939 [20:54:27<49:42:06, 5.78s/it]
30%|██▉ | 12968/43939 [20:54:32<49:36:28, 5.77s/it]
30%|██▉ | 12969/43939 [20:54:38<49:21:58, 5.74s/it]
30%|██▉ | 12970/43939 [20:54:44<49:20:27, 5.74s/it]
30%|██▉ | 12971/43939 [20:54:49<49:14:05, 5.72s/it]
30%|██▉ | 12972/43939 [20:54:55<49:10:30, 5.72s/it]
30%|██▉ | 12973/43939 [20:55:01<49:15:19, 5.73s/it]
30%|██▉ | 12974/43939 [20:55:07<49:13:11, 5.72s/it]
30%|██▉ | 12975/43939 [20:55:12<49:07:46, 5.71s/it]
30%|██▉ | 12976/43939 [20:55:18<49:12:45, 5.72s/it]
30%|██▉ | 12977/43939 [20:55:24<49:11:06, 5.72s/it]
30%|██▉ | 12978/43939 [20:55:29<49:11:43, 5.72s/it]
30%|██▉ | 12979/43939 [20:55:35<49:08:11, 5.71s/it]
30%|██▉ | 12980/43939 [20:55:41<49:05:59, 5.71s/it]
30%|██▉ | 12981/43939 [20:55:47<49:02:20, 5.70s/it]
30%|██▉ | 12982/43939 [20:55:52<49:02:10, 5.70s/it]
30%|██▉ | 12983/43939 [20:55:58<48:58:15, 5.70s/it]
30%|██▉ | 12984/43939 [20:56:04<48:58:30, 5.70s/it]
30%|██▉ | 12985/43939 [20:56:09<49:02:11, 5.70s/it]
30%|██▉ | 12986/43939 [20:56:15<49:00:08, 5.70s/it]
30%|██▉ | 12987/43939 [20:56:21<49:33:55, 5.76s/it]
30%|██▉ | 12988/43939 [20:56:27<49:20:02, 5.74s/it]
30%|██▉ | 12989/43939 [20:56:32<49:17:47, 5.73s/it]
30%|██▉ | 12990/43939 [20:56:38<49:08:31, 5.72s/it]
30%|██▉ | 12991/43939 [20:56:44<49:05:57, 5.71s/it]
30%|██▉ | 12992/43939 [20:56:49<49:05:04, 5.71s/it]
30%|██▉ | 12993/43939 [20:56:55<48:57:48, 5.70s/it]
30%|██▉ | 12994/43939 [20:57:01<48:57:37, 5.70s/it]
30%|██▉ | 12995/43939 [20:57:06<48:59:31, 5.70s/it]
30%|██▉ | 12996/43939 [20:57:12<49:01:34, 5.70s/it]
30%|██▉ | 12997/43939 [20:57:18<49:14:41, 5.73s/it]
30%|██▉ | 12998/43939 [20:57:24<49:15:00, 5.73s/it]
30%|██▉ | 12999/43939 [20:57:29<49:08:31, 5.72s/it]
30%|██▉ | 13000/43939 [20:57:35<49:09:02, 5.72s/it]
{'loss': 0.9451, 'grad_norm': 1.2166317701339722, 'learning_rate': 4.438144249416893e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.63, 'epoch': 1.18}
30%|██▉ | 13000/43939 [20:57:35<49:09:02, 5.72s/it]
30%|██▉ | 13001/43939 [20:57:41<49:03:27, 5.71s/it]
30%|██▉ | 13002/43939 [20:57:47<49:02:27, 5.71s/it]
30%|██▉ | 13003/43939 [20:57:52<49:00:28, 5.70s/it]
30%|██▉ | 13004/43939 [20:57:58<49:06:27, 5.71s/it]
30%|██▉ | 13005/43939 [20:58:04<49:03:34, 5.71s/it]
30%|██▉ | 13006/43939 [20:58:09<49:02:17, 5.71s/it]
30%|██▉ | 13007/43939 [20:58:15<48:57:48, 5.70s/it]
30%|██▉ | 13008/43939 [20:58:21<48:58:08, 5.70s/it]
30%|██▉ | 13009/43939 [20:58:27<49:30:13, 5.76s/it]
30%|██▉ | 13010/43939 [20:58:32<49:21:02, 5.74s/it]
30%|██▉ | 13011/43939 [20:58:38<49:17:06, 5.74s/it]
30%|██▉ | 13012/43939 [20:58:44<49:11:51, 5.73s/it]
30%|██▉ | 13013/43939 [20:58:49<49:05:16, 5.71s/it]
30%|██▉ | 13014/43939 [20:58:55<48:59:22, 5.70s/it]
30%|██▉ | 13015/43939 [20:59:01<49:00:14, 5.70s/it]
30%|██▉ | 13016/43939 [20:59:07<49:09:13, 5.72s/it]
30%|██▉ | 13017/43939 [20:59:12<49:03:30, 5.71s/it]
30%|██▉ | 13018/43939 [20:59:18<48:59:59, 5.70s/it]
30%|██▉ | 13019/43939 [20:59:24<49:02:39, 5.71s/it]
30%|██▉ | 13020/43939 [20:59:29<49:03:53, 5.71s/it]
30%|██▉ | 13021/43939 [20:59:35<48:59:24, 5.70s/it]
30%|██▉ | 13022/43939 [20:59:41<49:00:01, 5.71s/it]
30%|██▉ | 13023/43939 [20:59:47<49:02:00, 5.71s/it]
30%|██▉ | 13024/43939 [20:59:52<48:57:12, 5.70s/it]
30%|██▉ | 13025/43939 [20:59:58<48:57:19, 5.70s/it]
30%|██▉ | 13026/43939 [21:00:04<48:57:15, 5.70s/it]
30%|██▉ | 13027/43939 [21:00:09<49:02:35, 5.71s/it]
30%|██▉ | 13028/43939 [21:00:15<49:05:33, 5.72s/it]
30%|██▉ | 13029/43939 [21:00:21<49:01:31, 5.71s/it]
30%|██▉ | 13030/43939 [21:00:27<49:43:18, 5.79s/it]
30%|██▉ | 13031/43939 [21:00:33<49:35:25, 5.78s/it]
30%|██▉ | 13032/43939 [21:00:38<49:19:58, 5.75s/it]
30%|██▉ | 13033/43939 [21:00:44<49:10:44, 5.73s/it]
30%|██▉ | 13034/43939 [21:00:50<49:12:06, 5.73s/it]
30%|██▉ | 13035/43939 [21:00:55<49:04:33, 5.72s/it]
30%|██▉ | 13036/43939 [21:01:01<49:08:45, 5.73s/it]
30%|██▉ | 13037/43939 [21:01:07<49:09:27, 5.73s/it]
30%|██▉ | 13038/43939 [21:01:12<49:06:20, 5.72s/it]
30%|██▉ | 13039/43939 [21:01:18<49:03:57, 5.72s/it]
30%|██▉ | 13040/43939 [21:01:24<49:01:01, 5.71s/it]
30%|██▉ | 13041/43939 [21:01:30<48:59:41, 5.71s/it]
30%|██▉ | 13042/43939 [21:01:35<48:56:51, 5.70s/it]
30%|██▉ | 13043/43939 [21:01:41<48:56:16, 5.70s/it]
30%|██▉ | 13044/43939 [21:01:47<48:59:05, 5.71s/it]
30%|██▉ | 13045/43939 [21:01:52<49:06:23, 5.72s/it]
30%|██▉ | 13046/43939 [21:01:58<49:01:07, 5.71s/it]
30%|██▉ | 13047/43939 [21:02:04<49:02:41, 5.72s/it]
30%|██▉ | 13048/43939 [21:02:10<48:54:38, 5.70s/it]
30%|██▉ | 13049/43939 [21:02:15<48:58:39, 5.71s/it]
30%|██▉ | 13050/43939 [21:02:21<48:57:33, 5.71s/it]
{'loss': 0.9371, 'grad_norm': 1.0666178464889526, 'learning_rate': 4.4318566243141614e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.72, 'epoch': 1.19}
30%|██▉ | 13050/43939 [21:02:21<48:57:33, 5.71s/it]
30%|██▉ | 13051/43939 [21:02:27<49:31:54, 5.77s/it]
30%|██▉ | 13052/43939 [21:02:33<49:25:32, 5.76s/it]
30%|██▉ | 13053/43939 [21:02:38<49:19:01, 5.75s/it]
30%|██▉ | 13054/43939 [21:02:44<49:15:58, 5.74s/it]
30%|██▉ | 13055/43939 [21:02:50<49:10:07, 5.73s/it]
30%|██▉ | 13056/43939 [21:02:55<49:04:59, 5.72s/it]
30%|██▉ | 13057/43939 [21:03:01<49:03:27, 5.72s/it]
30%|██▉ | 13058/43939 [21:03:07<49:00:57, 5.71s/it]
30%|██▉ | 13059/43939 [21:03:13<49:00:26, 5.71s/it]
30%|██▉ | 13060/43939 [21:03:18<48:57:32, 5.71s/it]
30%|██▉ | 13061/43939 [21:03:24<48:53:16, 5.70s/it]
30%|██▉ | 13062/43939 [21:03:30<48:55:40, 5.70s/it]
30%|██▉ | 13063/43939 [21:03:35<49:03:59, 5.72s/it]
30%|██▉ | 13064/43939 [21:03:41<49:00:39, 5.71s/it]
30%|██▉ | 13065/43939 [21:03:47<48:58:57, 5.71s/it]
30%|██▉ | 13066/43939 [21:03:53<48:56:44, 5.71s/it]
30%|██▉ | 13067/43939 [21:03:58<48:56:51, 5.71s/it]
30%|██▉ | 13068/43939 [21:04:04<49:00:32, 5.72s/it]
30%|██▉ | 13069/43939 [21:04:10<48:54:28, 5.70s/it]
30%|██▉ | 13070/43939 [21:04:15<48:51:42, 5.70s/it]
30%|██▉ | 13071/43939 [21:04:21<48:54:49, 5.70s/it]
30%|██▉ | 13072/43939 [21:04:27<48:50:13, 5.70s/it]
30%|██▉ | 13073/43939 [21:04:33<49:26:24, 5.77s/it]
30%|██▉ | 13074/43939 [21:04:38<49:16:43, 5.75s/it]
30%|██▉ | 13075/43939 [21:04:44<49:06:16, 5.73s/it]
30%|██▉ | 13076/43939 [21:04:50<49:07:24, 5.73s/it]
30%|██▉ | 13077/43939 [21:04:55<49:01:52, 5.72s/it]
30%|██▉ | 13078/43939 [21:05:01<48:58:19, 5.71s/it]
30%|██▉ | 13079/43939 [21:05:07<48:51:36, 5.70s/it]
30%|██▉ | 13080/43939 [21:05:13<48:50:50, 5.70s/it]
30%|██▉ | 13081/43939 [21:05:18<48:47:50, 5.69s/it]
30%|██▉ | 13082/43939 [21:05:24<48:48:00, 5.69s/it]
30%|██▉ | 13083/43939 [21:05:30<48:51:36, 5.70s/it]
30%|██▉ | 13084/43939 [21:05:35<48:50:22, 5.70s/it]
30%|██▉ | 13085/43939 [21:05:41<48:45:28, 5.69s/it]
30%|██▉ | 13086/43939 [21:05:47<48:47:32, 5.69s/it]
30%|██▉ | 13087/43939 [21:05:52<48:42:55, 5.68s/it]
30%|██▉ | 13088/43939 [21:05:58<48:44:22, 5.69s/it]
30%|██▉ | 13089/43939 [21:06:04<48:46:16, 5.69s/it]
30%|██▉ | 13090/43939 [21:06:09<48:53:15, 5.71s/it]
30%|██▉ | 13091/43939 [21:06:15<48:53:05, 5.70s/it]
30%|██▉ | 13092/43939 [21:06:21<48:50:10, 5.70s/it]
30%|██▉ | 13093/43939 [21:06:27<48:47:42, 5.69s/it]
30%|██▉ | 13094/43939 [21:06:32<49:19:19, 5.76s/it]
30%|██▉ | 13095/43939 [21:06:38<49:11:53, 5.74s/it]
30%|██▉ | 13096/43939 [21:06:44<49:05:39, 5.73s/it]
30%|██▉ | 13097/43939 [21:06:50<49:08:18, 5.74s/it]
30%|██▉ | 13098/43939 [21:06:55<49:05:36, 5.73s/it]
30%|██▉ | 13099/43939 [21:07:01<49:07:16, 5.73s/it]
30%|██▉ | 13100/43939 [21:07:07<49:04:59, 5.73s/it]
{'loss': 0.9164, 'grad_norm': 0.9203543066978455, 'learning_rate': 4.425538519633379e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.06, 'epoch': 1.19}
30%|██▉ | 13100/43939 [21:07:07<49:04:59, 5.73s/it]
30%|██▉ | 13101/43939 [21:07:13<49:04:51, 5.73s/it]
30%|██▉ | 13102/43939 [21:07:18<49:05:31, 5.73s/it]
30%|██▉ | 13103/43939 [21:07:24<48:58:21, 5.72s/it]
30%|██▉ | 13104/43939 [21:07:30<48:57:01, 5.71s/it]
30%|██▉ | 13105/43939 [21:07:35<49:00:36, 5.72s/it]
30%|██▉ | 13106/43939 [21:07:41<48:57:20, 5.72s/it]
30%|██▉ | 13107/43939 [21:07:47<49:02:19, 5.73s/it]
30%|██▉ | 13108/43939 [21:07:53<49:01:06, 5.72s/it]
30%|██▉ | 13109/43939 [21:07:58<48:58:51, 5.72s/it]
30%|██▉ | 13110/43939 [21:08:04<48:57:19, 5.72s/it]
30%|██▉ | 13111/43939 [21:08:10<48:52:37, 5.71s/it]
30%|██▉ | 13112/43939 [21:08:15<48:52:07, 5.71s/it]
30%|██▉ | 13113/43939 [21:08:21<49:07:59, 5.74s/it]
30%|██▉ | 13114/43939 [21:08:27<49:05:14, 5.73s/it]
30%|██▉ | 13115/43939 [21:08:33<49:30:24, 5.78s/it]
30%|██▉ | 13116/43939 [21:08:39<49:18:25, 5.76s/it]
30%|██▉ | 13117/43939 [21:08:44<49:02:31, 5.73s/it]
30%|██▉ | 13118/43939 [21:08:50<48:59:16, 5.72s/it]
30%|██▉ | 13119/43939 [21:08:56<48:55:56, 5.72s/it]
30%|██▉ | 13120/43939 [21:09:01<48:52:57, 5.71s/it]
30%|██▉ | 13121/43939 [21:09:07<48:49:08, 5.70s/it]
30%|██▉ | 13122/43939 [21:09:13<48:51:18, 5.71s/it]
30%|██▉ | 13123/43939 [21:09:18<48:49:15, 5.70s/it]
30%|██▉ | 13124/43939 [21:09:24<48:49:38, 5.70s/it]
30%|██▉ | 13125/43939 [21:09:30<48:48:24, 5.70s/it]
30%|██▉ | 13126/43939 [21:09:35<48:40:51, 5.69s/it]
30%|██▉ | 13127/43939 [21:09:41<48:44:43, 5.70s/it]
30%|██▉ | 13128/43939 [21:09:47<48:43:52, 5.69s/it]
30%|██▉ | 13129/43939 [21:09:53<48:41:19, 5.69s/it]
30%|██▉ | 13130/43939 [21:09:58<48:51:17, 5.71s/it]
30%|██▉ | 13131/43939 [21:10:04<48:47:25, 5.70s/it]
30%|██▉ | 13132/43939 [21:10:10<48:49:32, 5.71s/it]
30%|██▉ | 13133/43939 [21:10:15<48:47:14, 5.70s/it]
30%|██▉ | 13134/43939 [21:10:21<48:46:00, 5.70s/it]
30%|██▉ | 13135/43939 [21:10:27<48:41:21, 5.69s/it]
30%|██▉ | 13136/43939 [21:10:33<49:19:40, 5.77s/it]
30%|██▉ | 13137/43939 [21:10:38<49:07:59, 5.74s/it]
30%|██▉ | 13138/43939 [21:10:44<48:58:55, 5.72s/it]
30%|██▉ | 13139/43939 [21:10:50<48:52:20, 5.71s/it]
30%|██▉ | 13140/43939 [21:10:55<48:45:19, 5.70s/it]
30%|██▉ | 13141/43939 [21:11:01<48:41:47, 5.69s/it]
30%|██▉ | 13142/43939 [21:11:07<48:40:09, 5.69s/it]
30%|██▉ | 13143/43939 [21:11:12<48:48:38, 5.71s/it]
30%|██▉ | 13144/43939 [21:11:18<48:42:25, 5.69s/it]
30%|██▉ | 13145/43939 [21:11:24<48:39:33, 5.69s/it]
30%|██▉ | 13146/43939 [21:11:30<48:39:05, 5.69s/it]
30%|██▉ | 13147/43939 [21:11:35<48:41:12, 5.69s/it]
30%|██▉ | 13148/43939 [21:11:41<48:39:38, 5.69s/it]
30%|██▉ | 13149/43939 [21:11:47<48:45:07, 5.70s/it]
30%|██▉ | 13150/43939 [21:11:52<48:46:43, 5.70s/it]
{'loss': 0.9459, 'grad_norm': 1.1851335763931274, 'learning_rate': 4.419190035057497e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.24, 'epoch': 1.2}
30%|██▉ | 13150/43939 [21:11:52<48:46:43, 5.70s/it]
30%|██▉ | 13151/43939 [21:11:58<48:48:25, 5.71s/it]
30%|██▉ | 13152/43939 [21:12:04<48:48:18, 5.71s/it]
30%|██▉ | 13153/43939 [21:12:09<48:46:28, 5.70s/it]
30%|██▉ | 13154/43939 [21:12:15<48:48:25, 5.71s/it]
30%|██▉ | 13155/43939 [21:12:21<48:44:58, 5.70s/it]
30%|██▉ | 13156/43939 [21:12:27<48:42:24, 5.70s/it]
30%|██▉ | 13157/43939 [21:12:33<49:21:13, 5.77s/it]
30%|██▉ | 13158/43939 [21:12:38<49:06:29, 5.74s/it]
30%|██▉ | 13159/43939 [21:12:44<49:16:26, 5.76s/it]
30%|██▉ | 13160/43939 [21:12:50<49:05:38, 5.74s/it]
30%|██▉ | 13161/43939 [21:12:55<48:56:43, 5.72s/it]
30%|██▉ | 13162/43939 [21:13:01<48:45:24, 5.70s/it]
30%|██▉ | 13163/43939 [21:13:07<48:42:14, 5.70s/it]
30%|██▉ | 13164/43939 [21:13:12<48:40:37, 5.69s/it]
30%|██▉ | 13165/43939 [21:13:18<48:39:00, 5.69s/it]
30%|██▉ | 13166/43939 [21:13:24<48:37:56, 5.69s/it]
30%|██▉ | 13167/43939 [21:13:29<48:36:33, 5.69s/it]
30%|██▉ | 13168/43939 [21:13:35<48:35:21, 5.68s/it]
30%|██▉ | 13169/43939 [21:13:41<48:36:07, 5.69s/it]
30%|██▉ | 13170/43939 [21:13:47<48:38:46, 5.69s/it]
30%|██▉ | 13171/43939 [21:13:52<48:39:10, 5.69s/it]
30%|██▉ | 13172/43939 [21:13:58<48:37:16, 5.69s/it]
30%|██▉ | 13173/43939 [21:14:04<48:39:58, 5.69s/it]
30%|██▉ | 13174/43939 [21:14:09<48:42:06, 5.70s/it]
30%|██▉ | 13175/43939 [21:14:15<48:40:51, 5.70s/it]
30%|██▉ | 13176/43939 [21:14:21<48:40:04, 5.70s/it]
30%|██▉ | 13177/43939 [21:14:27<49:17:18, 5.77s/it]
30%|██▉ | 13178/43939 [21:14:32<49:06:18, 5.75s/it]
30%|██▉ | 13179/43939 [21:14:38<49:00:05, 5.73s/it]
30%|██▉ | 13180/43939 [21:14:44<49:03:08, 5.74s/it]
30%|██▉ | 13181/43939 [21:14:49<48:54:25, 5.72s/it]
30%|███ | 13182/43939 [21:14:55<48:47:14, 5.71s/it][2025-11-29 15:19:08,969] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.36s/it][A
4%|▍ | 3/72 [00:05<02:12, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:30, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.72s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.72s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.72s/it][A
57%|█████▋ | 41/72 [01:48<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.72s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.72s/it][A
75%|███████▌ | 54/72 [02:24<00:48, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.72s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:48<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.72s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A{'eval_loss': 0.9950478672981262, 'eval_runtime': 195.7919, 'eval_samples_per_second': 5.102, 'eval_steps_per_second': 0.368, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 1.2}
30%|███ | 13182/43939 [21:18:11<48:47:14, 5.71s/it]
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A
30%|███ | 13183/43939 [21:18:17<550:59:13, 64.49s/it]
30%|███ | 13184/43939 [21:18:22<400:12:12, 46.85s/it]
30%|███ | 13185/43939 [21:18:28<294:47:28, 34.51s/it]
30%|███ | 13186/43939 [21:18:34<220:59:48, 25.87s/it]
30%|███ | 13187/43939 [21:18:40<169:30:24, 19.84s/it]
30%|███ | 13188/43939 [21:18:45<133:17:42, 15.60s/it]
30%|███ | 13189/43939 [21:18:51<107:52:39, 12.63s/it]
30%|███ | 13190/43939 [21:18:57<90:07:24, 10.55s/it]
30%|███ | 13191/43939 [21:19:03<77:43:39, 9.10s/it]
30%|███ | 13192/43939 [21:19:08<69:00:24, 8.08s/it]
30%|███ | 13193/43939 [21:19:14<62:54:57, 7.37s/it]
30%|███ | 13194/43939 [21:19:20<58:39:20, 6.87s/it]
30%|███ | 13195/43939 [21:19:25<55:39:45, 6.52s/it]
30%|███ | 13196/43939 [21:19:31<53:33:49, 6.27s/it]
30%|███ | 13197/43939 [21:19:37<52:07:09, 6.10s/it]
30%|███ | 13198/43939 [21:19:43<51:47:42, 6.07s/it]
30%|███ | 13199/43939 [21:19:48<50:50:47, 5.95s/it]
30%|███ | 13200/43939 [21:19:54<50:17:12, 5.89s/it]
{'loss': 0.9368, 'grad_norm': 1.0488026142120361, 'learning_rate': 4.412811270748776e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 177.13, 'epoch': 1.2}
30%|███ | 13200/43939 [21:19:54<50:17:12, 5.89s/it]
30%|███ | 13201/43939 [21:20:00<49:52:46, 5.84s/it]
30%|███ | 13202/43939 [21:20:06<49:33:50, 5.81s/it]
30%|███ | 13203/43939 [21:20:11<49:18:16, 5.77s/it]
30%|███ | 13204/43939 [21:20:17<49:04:36, 5.75s/it]
30%|███ | 13205/43939 [21:20:23<48:56:04, 5.73s/it]
30%|███ | 13206/43939 [21:20:28<48:51:38, 5.72s/it]
30%|███ | 13207/43939 [21:20:34<48:59:10, 5.74s/it]
30%|███ | 13208/43939 [21:20:40<48:52:30, 5.73s/it]
30%|███ | 13209/43939 [21:20:46<48:57:32, 5.74s/it]
30%|███ | 13210/43939 [21:20:51<48:49:12, 5.72s/it]
30%|███ | 13211/43939 [21:20:57<48:45:11, 5.71s/it]
30%|███ | 13212/43939 [21:21:03<48:38:34, 5.70s/it]
30%|███ | 13213/43939 [21:21:08<48:39:53, 5.70s/it]
30%|███ | 13214/43939 [21:21:14<48:46:09, 5.71s/it]
30%|███ | 13215/43939 [21:21:20<48:49:05, 5.72s/it]
30%|███ | 13216/43939 [21:21:26<48:46:35, 5.72s/it]
30%|███ | 13217/43939 [21:21:31<48:53:52, 5.73s/it]
30%|███ | 13218/43939 [21:21:37<48:54:10, 5.73s/it]
30%|███ | 13219/43939 [21:21:43<48:41:12, 5.71s/it]
30%|███ | 13220/43939 [21:21:49<49:16:17, 5.77s/it]
30%|███ | 13221/43939 [21:21:54<49:05:22, 5.75s/it]
30%|███ | 13222/43939 [21:22:00<48:56:18, 5.74s/it]
30%|███ | 13223/43939 [21:22:06<48:55:54, 5.73s/it]
30%|███ | 13224/43939 [21:22:11<48:45:07, 5.71s/it]
30%|███ | 13225/43939 [21:22:17<48:45:11, 5.71s/it]
30%|███ | 13226/43939 [21:22:23<48:45:17, 5.71s/it]
30%|███ | 13227/43939 [21:22:29<48:51:25, 5.73s/it]
30%|███ | 13228/43939 [21:22:34<48:49:39, 5.72s/it]
30%|███ | 13229/43939 [21:22:40<48:45:37, 5.72s/it]
30%|███ | 13230/43939 [21:22:46<48:41:12, 5.71s/it]
30%|███ | 13231/43939 [21:22:51<48:45:39, 5.72s/it]
30%|███ | 13232/43939 [21:22:57<48:43:11, 5.71s/it]
30%|███ | 13233/43939 [21:23:03<48:49:03, 5.72s/it]
30%|███ | 13234/43939 [21:23:09<48:48:59, 5.72s/it]
30%|███ | 13235/43939 [21:23:14<48:47:12, 5.72s/it]
30%|███ | 13236/43939 [21:23:20<48:50:42, 5.73s/it]
30%|███ | 13237/43939 [21:23:26<48:48:04, 5.72s/it]
30%|███ | 13238/43939 [21:23:31<48:46:02, 5.72s/it]
30%|███ | 13239/43939 [21:23:37<48:37:38, 5.70s/it]
30%|███ | 13240/43939 [21:23:43<48:41:27, 5.71s/it]
30%|███ | 13241/43939 [21:23:49<49:18:58, 5.78s/it]
30%|███ | 13242/43939 [21:23:55<49:04:01, 5.75s/it]
30%|███ | 13243/43939 [21:24:00<48:50:37, 5.73s/it]
30%|███ | 13244/43939 [21:24:06<48:47:51, 5.72s/it]
30%|███ | 13245/43939 [21:24:12<48:38:34, 5.71s/it]
30%|███ | 13246/43939 [21:24:17<48:46:06, 5.72s/it]
30%|███ | 13247/43939 [21:24:23<48:48:20, 5.72s/it]
30%|███ | 13248/43939 [21:24:29<48:50:43, 5.73s/it]
30%|███ | 13249/43939 [21:24:34<48:45:25, 5.72s/it]
30%|███ | 13250/43939 [21:24:40<48:47:01, 5.72s/it]
{'loss': 0.9289, 'grad_norm': 0.9081180691719055, 'learning_rate': 4.4064023273472156e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.55, 'epoch': 1.21}
30%|███ | 13250/43939 [21:24:40<48:47:01, 5.72s/it]
30%|███ | 13251/43939 [21:24:46<48:42:35, 5.71s/it]
30%|███ | 13252/43939 [21:24:52<48:40:40, 5.71s/it]
30%|███ | 13253/43939 [21:24:57<48:39:08, 5.71s/it]
30%|███ | 13254/43939 [21:25:03<48:35:17, 5.70s/it]
30%|███ | 13255/43939 [21:25:09<48:35:52, 5.70s/it]
30%|███ | 13256/43939 [21:25:14<48:35:21, 5.70s/it]
30%|███ | 13257/43939 [21:25:20<48:37:22, 5.71s/it]
30%|███ | 13258/43939 [21:25:26<48:40:54, 5.71s/it]
30%|███ | 13259/43939 [21:25:32<48:33:14, 5.70s/it]
30%|███ | 13260/43939 [21:25:37<48:36:33, 5.70s/it]
30%|███ | 13261/43939 [21:25:43<48:37:31, 5.71s/it]
30%|███ | 13262/43939 [21:25:49<49:10:01, 5.77s/it]
30%|███ | 13263/43939 [21:25:55<49:02:27, 5.76s/it]
30%|███ | 13264/43939 [21:26:00<48:53:34, 5.74s/it]
30%|███ | 13265/43939 [21:26:06<48:42:32, 5.72s/it]
30%|███ | 13266/43939 [21:26:12<48:39:29, 5.71s/it]
30%|███ | 13267/43939 [21:26:17<48:35:15, 5.70s/it]
30%|███ | 13268/43939 [21:26:23<48:33:15, 5.70s/it]
30%|███ | 13269/43939 [21:26:29<48:35:39, 5.70s/it]
30%|███ | 13270/43939 [21:26:34<48:35:05, 5.70s/it]
30%|███ | 13271/43939 [21:26:40<48:35:30, 5.70s/it]
30%|███ | 13272/43939 [21:26:46<48:34:24, 5.70s/it]
30%|███ | 13273/43939 [21:26:52<48:30:27, 5.69s/it]
30%|███ | 13274/43939 [21:26:57<48:32:07, 5.70s/it]
30%|███ | 13275/43939 [21:27:03<48:31:50, 5.70s/it]
30%|███ | 13276/43939 [21:27:09<48:27:15, 5.69s/it]
30%|███ | 13277/43939 [21:27:14<48:31:10, 5.70s/it]
30%|███ | 13278/43939 [21:27:20<48:39:15, 5.71s/it]
30%|███ | 13279/43939 [21:27:26<48:42:00, 5.72s/it]
30%|███ | 13280/43939 [21:27:31<48:42:00, 5.72s/it]
30%|███ | 13281/43939 [21:27:37<48:46:18, 5.73s/it]
30%|███ | 13282/43939 [21:27:43<48:48:22, 5.73s/it]
30%|███ | 13283/43939 [21:27:49<49:17:15, 5.79s/it]
30%|███ | 13284/43939 [21:27:55<49:04:33, 5.76s/it]
30%|███ | 13285/43939 [21:28:00<48:50:03, 5.74s/it]
30%|███ | 13286/43939 [21:28:06<48:48:22, 5.73s/it]
30%|███ | 13287/43939 [21:28:12<48:44:21, 5.72s/it]
30%|███ | 13288/43939 [21:28:17<48:42:03, 5.72s/it]
30%|███ | 13289/43939 [21:28:23<48:34:15, 5.70s/it]
30%|███ | 13290/43939 [21:28:29<48:32:04, 5.70s/it]
30%|███ | 13291/43939 [21:28:34<48:32:12, 5.70s/it]
30%|███ | 13292/43939 [21:28:40<48:25:08, 5.69s/it]
30%|███ | 13293/43939 [21:28:46<48:26:22, 5.69s/it]
30%|███ | 13294/43939 [21:28:52<48:23:44, 5.69s/it]
30%|███ | 13295/43939 [21:28:57<48:26:38, 5.69s/it]
30%|███ | 13296/43939 [21:29:03<48:34:22, 5.71s/it]
30%|███ | 13297/43939 [21:29:09<48:35:05, 5.71s/it]
30%|███ | 13298/43939 [21:29:14<48:37:34, 5.71s/it]
30%|███ | 13299/43939 [21:29:20<48:36:53, 5.71s/it]
30%|███ | 13300/43939 [21:29:26<48:41:43, 5.72s/it]
{'loss': 0.9439, 'grad_norm': 1.1023324728012085, 'learning_rate': 4.3999633059689566e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.84, 'epoch': 1.21}
30%|███ | 13300/43939 [21:29:26<48:41:43, 5.72s/it]
30%|███ | 13301/43939 [21:29:32<48:40:42, 5.72s/it]
30%|███ | 13302/43939 [21:29:37<48:36:33, 5.71s/it]
30%|███ | 13303/43939 [21:29:43<48:29:59, 5.70s/it]
30%|███ | 13304/43939 [21:29:49<48:31:57, 5.70s/it]
30%|███ | 13305/43939 [21:29:55<49:07:41, 5.77s/it]
30%|███ | 13306/43939 [21:30:00<49:02:11, 5.76s/it]
30%|███ | 13307/43939 [21:30:06<48:48:42, 5.74s/it]
30%|███ | 13308/43939 [21:30:12<48:38:16, 5.72s/it]
30%|███ | 13309/43939 [21:30:17<48:36:46, 5.71s/it]
30%|███ | 13310/43939 [21:30:23<48:29:08, 5.70s/it]
30%|███ | 13311/43939 [21:30:29<48:33:40, 5.71s/it]
30%|███ | 13312/43939 [21:30:34<48:35:28, 5.71s/it]
30%|███ | 13313/43939 [21:30:40<48:30:47, 5.70s/it]
30%|███ | 13314/43939 [21:30:46<48:31:35, 5.70s/it]
30%|███ | 13315/43939 [21:30:52<48:28:49, 5.70s/it]
30%|███ | 13316/43939 [21:30:57<48:25:11, 5.69s/it]
30%|███ | 13317/43939 [21:31:03<48:28:11, 5.70s/it]
30%|███ | 13318/43939 [21:31:09<48:28:57, 5.70s/it]
30%|███ | 13319/43939 [21:31:14<48:24:08, 5.69s/it]
30%|███ | 13320/43939 [21:31:20<48:25:15, 5.69s/it]
30%|███ | 13321/43939 [21:31:26<48:28:47, 5.70s/it]
30%|███ | 13322/43939 [21:31:31<48:28:03, 5.70s/it]
30%|███ | 13323/43939 [21:31:37<48:28:31, 5.70s/it]
30%|███ | 13324/43939 [21:31:43<48:31:16, 5.71s/it]
30%|███ | 13325/43939 [21:31:49<49:00:05, 5.76s/it]
30%|███ | 13326/43939 [21:31:54<48:53:27, 5.75s/it]
30%|███ | 13327/43939 [21:32:00<48:47:02, 5.74s/it]
30%|███ | 13328/43939 [21:32:06<48:45:17, 5.73s/it]
30%|███ | 13329/43939 [21:32:12<48:37:18, 5.72s/it]
30%|███ | 13330/43939 [21:32:17<48:30:07, 5.70s/it]
30%|███ | 13331/43939 [21:32:23<48:35:18, 5.71s/it]
30%|███ | 13332/43939 [21:32:29<48:33:31, 5.71s/it]
30%|███ | 13333/43939 [21:32:34<48:27:25, 5.70s/it]
30%|███ | 13334/43939 [21:32:40<48:22:49, 5.69s/it]
30%|███ | 13335/43939 [21:32:46<48:23:55, 5.69s/it]
30%|███ | 13336/43939 [21:32:51<48:23:14, 5.69s/it]
30%|███ | 13337/43939 [21:32:57<48:25:09, 5.70s/it]
30%|███ | 13338/43939 [21:33:03<48:29:21, 5.70s/it]
30%|███ | 13339/43939 [21:33:09<48:34:18, 5.71s/it]
30%|███ | 13340/43939 [21:33:14<48:32:45, 5.71s/it]
30%|███ | 13341/43939 [21:33:20<48:32:46, 5.71s/it]
30%|███ | 13342/43939 [21:33:26<48:25:19, 5.70s/it]
30%|███ | 13343/43939 [21:33:31<48:24:04, 5.70s/it]
30%|███ | 13344/43939 [21:33:37<48:28:33, 5.70s/it]
30%|███ | 13345/43939 [21:33:43<48:27:10, 5.70s/it]
30%|███ | 13346/43939 [21:33:49<48:59:00, 5.76s/it]
30%|███ | 13347/43939 [21:33:54<48:50:53, 5.75s/it]
30%|███ | 13348/43939 [21:34:00<48:45:44, 5.74s/it]
30%|███ | 13349/43939 [21:34:06<48:37:19, 5.72s/it]
30%|███ | 13350/43939 [21:34:11<48:32:52, 5.71s/it]
{'loss': 0.92, 'grad_norm': 1.2254292964935303, 'learning_rate': 4.393494308204693e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 454.8, 'epoch': 1.22}
30%|███ | 13350/43939 [21:34:12<48:32:52, 5.71s/it]
30%|███ | 13351/43939 [21:34:17<48:32:42, 5.71s/it]
30%|███ | 13352/43939 [21:34:23<48:28:46, 5.71s/it]
30%|███ | 13353/43939 [21:34:29<48:29:37, 5.71s/it]
30%|███ | 13354/43939 [21:34:34<48:37:43, 5.72s/it]
30%|███ | 13355/43939 [21:34:40<48:31:27, 5.71s/it]
30%|███ | 13356/43939 [21:34:46<48:35:59, 5.72s/it]
30%|███ | 13357/43939 [21:34:52<48:35:16, 5.72s/it]
30%|███ | 13358/43939 [21:34:57<48:39:19, 5.73s/it]
30%|███ | 13359/43939 [21:35:03<48:42:08, 5.73s/it]
30%|███ | 13360/43939 [21:35:09<48:35:42, 5.72s/it]
30%|███ | 13361/43939 [21:35:14<48:35:22, 5.72s/it]
30%|███ | 13362/43939 [21:35:20<48:31:04, 5.71s/it]
30%|███ | 13363/43939 [21:35:26<48:27:20, 5.71s/it]
30%|███ | 13364/43939 [21:35:32<48:32:12, 5.71s/it]
30%|███ | 13365/43939 [21:35:37<48:31:35, 5.71s/it]
30%|███ | 13366/43939 [21:35:43<48:31:28, 5.71s/it]
30%|███ | 13367/43939 [21:35:49<48:25:41, 5.70s/it]
30%|███ | 13368/43939 [21:35:55<48:54:57, 5.76s/it]
30%|███ | 13369/43939 [21:36:00<48:46:25, 5.74s/it]
30%|███ | 13370/43939 [21:36:06<48:40:13, 5.73s/it]
30%|███ | 13371/43939 [21:36:12<48:34:47, 5.72s/it]
30%|███ | 13372/43939 [21:36:17<48:28:45, 5.71s/it]
30%|███ | 13373/43939 [21:36:23<48:29:18, 5.71s/it]
30%|███ | 13374/43939 [21:36:29<48:27:49, 5.71s/it]
30%|███ | 13375/43939 [21:36:34<48:23:15, 5.70s/it]
30%|███ | 13376/43939 [21:36:40<48:19:39, 5.69s/it]
30%|███ | 13377/43939 [21:36:46<48:19:08, 5.69s/it]
30%|███ | 13378/43939 [21:36:51<48:22:35, 5.70s/it]
30%|███ | 13379/43939 [21:36:57<48:21:30, 5.70s/it]
30%|███ | 13380/43939 [21:37:03<48:23:23, 5.70s/it]
30%|███ | 13381/43939 [21:37:09<48:22:24, 5.70s/it]
30%|███ | 13382/43939 [21:37:14<48:32:01, 5.72s/it]
30%|███ | 13383/43939 [21:37:20<48:29:15, 5.71s/it]
30%|███ | 13384/43939 [21:37:26<48:21:43, 5.70s/it]
30%|███ | 13385/43939 [21:37:31<48:19:10, 5.69s/it]
30%|███ | 13386/43939 [21:37:37<48:23:19, 5.70s/it]
30%|███ | 13387/43939 [21:37:43<48:22:10, 5.70s/it]
30%|███ | 13388/43939 [21:37:49<48:33:40, 5.72s/it]
30%|███ | 13389/43939 [21:37:54<49:00:17, 5.77s/it]
30%|███ | 13390/43939 [21:38:00<48:52:02, 5.76s/it]
30%|███ | 13391/43939 [21:38:06<48:45:05, 5.75s/it]
30%|███ | 13392/43939 [21:38:12<48:45:48, 5.75s/it]
30%|███ | 13393/43939 [21:38:17<48:40:00, 5.74s/it]
30%|███ | 13394/43939 [21:38:23<48:28:49, 5.71s/it]
30%|███ | 13395/43939 [21:38:29<48:30:36, 5.72s/it]
30%|███ | 13396/43939 [21:38:34<48:26:56, 5.71s/it]
30%|███ | 13397/43939 [21:38:40<48:25:40, 5.71s/it]
30%|███ | 13398/43939 [21:38:46<48:22:43, 5.70s/it]
30%|███ | 13399/43939 [21:38:52<48:21:40, 5.70s/it]
30%|███ | 13400/43939 [21:38:57<48:23:28, 5.70s/it]
{'loss': 0.9393, 'grad_norm': 1.0367655754089355, 'learning_rate': 4.3869954361180676e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.84, 'epoch': 1.22}
30%|███ | 13400/43939 [21:38:57<48:23:28, 5.70s/it]
30%|███ | 13401/43939 [21:39:03<48:29:21, 5.72s/it]
31%|███ | 13402/43939 [21:39:09<48:24:34, 5.71s/it]
31%|███ | 13403/43939 [21:39:14<48:20:00, 5.70s/it]
31%|███ | 13404/43939 [21:39:20<48:32:12, 5.72s/it]
31%|███ | 13405/43939 [21:39:26<48:28:39, 5.72s/it]
31%|███ | 13406/43939 [21:39:32<48:33:53, 5.73s/it]
31%|███ | 13407/43939 [21:39:37<48:27:23, 5.71s/it]
31%|███ | 13408/43939 [21:39:43<48:24:05, 5.71s/it]
31%|███ | 13409/43939 [21:39:49<48:22:54, 5.71s/it]
31%|███ | 13410/43939 [21:39:55<49:03:11, 5.78s/it]
31%|███ | 13411/43939 [21:40:00<48:49:38, 5.76s/it]
31%|███ | 13412/43939 [21:40:06<48:41:15, 5.74s/it]
31%|███ | 13413/43939 [21:40:12<48:37:31, 5.73s/it]
31%|███ | 13414/43939 [21:40:18<48:37:20, 5.73s/it]
31%|███ | 13415/43939 [21:40:23<48:28:34, 5.72s/it]
31%|███ | 13416/43939 [21:40:29<48:28:44, 5.72s/it]
31%|███ | 13417/43939 [21:40:35<48:24:56, 5.71s/it]
31%|███ | 13418/43939 [21:40:40<48:35:47, 5.73s/it]
31%|███ | 13419/43939 [21:40:46<48:29:37, 5.72s/it]
31%|███ | 13420/43939 [21:40:52<48:21:25, 5.70s/it]
31%|███ | 13421/43939 [21:40:57<48:18:24, 5.70s/it]
31%|███ | 13422/43939 [21:41:03<48:27:37, 5.72s/it]
31%|███ | 13423/43939 [21:41:09<48:24:19, 5.71s/it]
31%|███ | 13424/43939 [21:41:15<48:20:48, 5.70s/it]
31%|███ | 13425/43939 [21:41:20<48:20:52, 5.70s/it]
31%|███ | 13426/43939 [21:41:26<48:15:28, 5.69s/it]
31%|███ | 13427/43939 [21:41:32<48:17:27, 5.70s/it]
31%|███ | 13428/43939 [21:41:37<48:24:44, 5.71s/it]
31%|███ | 13429/43939 [21:41:43<48:33:30, 5.73s/it]
31%|███ | 13430/43939 [21:41:49<48:29:08, 5.72s/it]
31%|███ | 13431/43939 [21:41:55<48:20:20, 5.70s/it]
31%|███ | 13432/43939 [21:42:00<48:54:41, 5.77s/it]
31%|███ | 13433/43939 [21:42:06<48:39:39, 5.74s/it]
31%|███ | 13434/43939 [21:42:12<48:26:26, 5.72s/it]
31%|███ | 13435/43939 [21:42:18<48:25:38, 5.72s/it]
31%|███ | 13436/43939 [21:42:23<48:23:00, 5.71s/it]
31%|███ | 13437/43939 [21:42:29<48:24:30, 5.71s/it]
31%|███ | 13438/43939 [21:42:35<48:23:55, 5.71s/it]
31%|███ | 13439/43939 [21:42:40<48:21:08, 5.71s/it]
31%|███ | 13440/43939 [21:42:46<48:22:49, 5.71s/it]
31%|███ | 13441/43939 [21:42:52<48:20:15, 5.71s/it]
31%|███ | 13442/43939 [21:42:58<48:31:16, 5.73s/it]
31%|███ | 13443/43939 [21:43:03<48:33:57, 5.73s/it]
31%|███ | 13444/43939 [21:43:09<48:30:04, 5.73s/it]
31%|███ | 13445/43939 [21:43:15<48:28:30, 5.72s/it]
31%|███ | 13446/43939 [21:43:20<48:29:54, 5.73s/it]
31%|███ | 13447/43939 [21:43:26<48:21:37, 5.71s/it]
31%|███ | 13448/43939 [21:43:32<48:18:50, 5.70s/it]
31%|███ | 13449/43939 [21:43:37<48:11:58, 5.69s/it]
31%|███ | 13450/43939 [21:43:43<48:10:36, 5.69s/it]
{'loss': 0.9323, 'grad_norm': 1.5422024726867676, 'learning_rate': 4.380466792244059e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.73, 'epoch': 1.22}
31%|███ | 13450/43939 [21:43:43<48:10:36, 5.69s/it]
31%|███ | 13451/43939 [21:43:49<48:12:02, 5.69s/it]
31%|███ | 13452/43939 [21:43:55<48:20:05, 5.71s/it]
31%|███ | 13453/43939 [21:44:00<48:48:52, 5.76s/it]
31%|███ | 13454/43939 [21:44:06<48:38:44, 5.74s/it]
31%|███ | 13455/43939 [21:44:12<48:38:07, 5.74s/it]
31%|███ | 13456/43939 [21:44:18<48:35:41, 5.74s/it]
31%|███ | 13457/43939 [21:44:23<48:29:51, 5.73s/it]
31%|███ | 13458/43939 [21:44:29<48:26:36, 5.72s/it]
31%|███ | 13459/43939 [21:44:35<48:25:18, 5.72s/it]
31%|███ | 13460/43939 [21:44:40<48:20:01, 5.71s/it]
31%|███ | 13461/43939 [21:44:46<48:18:21, 5.71s/it]
31%|███ | 13462/43939 [21:44:52<48:18:21, 5.71s/it]
31%|███ | 13463/43939 [21:44:58<48:13:24, 5.70s/it]
31%|███ | 13464/43939 [21:45:03<48:26:32, 5.72s/it]
31%|███ | 13465/43939 [21:45:09<48:23:15, 5.72s/it]
31%|███ | 13466/43939 [21:45:15<48:18:02, 5.71s/it]
31%|███ | 13467/43939 [21:45:20<48:27:06, 5.72s/it]
31%|███ | 13468/43939 [21:45:26<48:23:32, 5.72s/it]
31%|███ | 13469/43939 [21:45:32<48:26:02, 5.72s/it]
31%|███ | 13470/43939 [21:45:38<48:21:11, 5.71s/it]
31%|███ | 13471/43939 [21:45:43<48:18:33, 5.71s/it]
31%|███ | 13472/43939 [21:45:49<48:18:58, 5.71s/it]
31%|███ | 13473/43939 [21:45:55<48:16:48, 5.70s/it]
31%|███ | 13474/43939 [21:46:01<48:46:18, 5.76s/it]
31%|███ | 13475/43939 [21:46:06<48:41:41, 5.75s/it]
31%|███ | 13476/43939 [21:46:12<48:36:46, 5.74s/it]
31%|███ | 13477/43939 [21:46:18<48:31:00, 5.73s/it]
31%|███ | 13478/43939 [21:46:23<48:24:38, 5.72s/it]
31%|███ | 13479/43939 [21:46:29<48:24:34, 5.72s/it]
31%|███ | 13480/43939 [21:46:35<48:30:47, 5.73s/it]
31%|███ | 13481/43939 [21:46:41<48:24:02, 5.72s/it]
31%|███ | 13482/43939 [21:46:46<48:26:57, 5.73s/it]
31%|███ | 13483/43939 [21:46:52<48:23:18, 5.72s/it]
31%|███ | 13484/43939 [21:46:58<48:13:07, 5.70s/it]
31%|███ | 13485/43939 [21:47:03<48:15:56, 5.71s/it]
31%|███ | 13486/43939 [21:47:09<48:15:12, 5.70s/it]
31%|███ | 13487/43939 [21:47:15<48:14:50, 5.70s/it]
31%|███ | 13488/43939 [21:47:21<48:18:55, 5.71s/it]
31%|███ | 13489/43939 [21:47:26<48:18:00, 5.71s/it]
31%|███ | 13490/43939 [21:47:32<48:12:23, 5.70s/it]
31%|███ | 13491/43939 [21:47:38<48:13:40, 5.70s/it]
31%|███ | 13492/43939 [21:47:43<48:12:14, 5.70s/it]
31%|███ | 13493/43939 [21:47:49<48:11:55, 5.70s/it]
31%|███ | 13494/43939 [21:47:55<48:08:10, 5.69s/it]
31%|███ | 13495/43939 [21:48:01<48:40:44, 5.76s/it]
31%|███ | 13496/43939 [21:48:06<48:29:41, 5.73s/it]
31%|███ | 13497/43939 [21:48:12<48:23:14, 5.72s/it]
31%|███ | 13498/43939 [21:48:18<48:24:40, 5.73s/it]
31%|███ | 13499/43939 [21:48:23<48:18:18, 5.71s/it]
31%|███ | 13500/43939 [21:48:29<48:10:35, 5.70s/it]
{'loss': 0.9177, 'grad_norm': 1.0496447086334229, 'learning_rate': 4.373908479587366e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.53, 'epoch': 1.23}
31%|███ | 13500/43939 [21:48:29<48:10:35, 5.70s/it]
31%|███ | 13501/43939 [21:48:35<48:16:19, 5.71s/it]
31%|███ | 13502/43939 [21:48:40<48:08:22, 5.69s/it]
31%|███ | 13503/43939 [21:48:46<48:16:29, 5.71s/it]
31%|███ | 13504/43939 [21:48:52<48:15:39, 5.71s/it]
31%|███ | 13505/43939 [21:48:58<48:08:49, 5.70s/it]
31%|███ | 13506/43939 [21:49:03<48:04:40, 5.69s/it]
31%|███ | 13507/43939 [21:49:09<48:14:41, 5.71s/it]
31%|███ | 13508/43939 [21:49:15<48:14:43, 5.71s/it]
31%|███ | 13509/43939 [21:49:20<48:18:14, 5.71s/it]
31%|███ | 13510/43939 [21:49:26<48:14:57, 5.71s/it]
31%|███ | 13511/43939 [21:49:32<48:12:33, 5.70s/it]
31%|███ | 13512/43939 [21:49:38<48:11:08, 5.70s/it]
31%|███ | 13513/43939 [21:49:43<48:11:09, 5.70s/it]
31%|███ | 13514/43939 [21:49:49<48:07:16, 5.69s/it]
31%|███ | 13515/43939 [21:49:55<48:08:25, 5.70s/it]
31%|███ | 13516/43939 [21:50:01<48:42:34, 5.76s/it]
31%|███ | 13517/43939 [21:50:06<48:36:11, 5.75s/it]
31%|███ | 13518/43939 [21:50:12<48:29:53, 5.74s/it]
31%|███ | 13519/43939 [21:50:18<48:26:45, 5.73s/it]
31%|███ | 13520/43939 [21:50:23<48:18:30, 5.72s/it]
31%|███ | 13521/43939 [21:50:29<48:15:45, 5.71s/it]
31%|███ | 13522/43939 [21:50:35<48:19:22, 5.72s/it]
31%|███ | 13523/43939 [21:50:41<48:24:19, 5.73s/it]
31%|███ | 13524/43939 [21:50:46<48:24:50, 5.73s/it]
31%|███ | 13525/43939 [21:50:52<48:15:05, 5.71s/it]
31%|███ | 13526/43939 [21:50:58<48:17:18, 5.72s/it]
31%|███ | 13527/43939 [21:51:03<48:14:05, 5.71s/it]
31%|███ | 13528/43939 [21:51:09<48:06:08, 5.69s/it]
31%|███ | 13529/43939 [21:51:15<48:06:42, 5.70s/it]
31%|███ | 13530/43939 [21:51:20<48:07:15, 5.70s/it]
31%|███ | 13531/43939 [21:51:26<48:09:28, 5.70s/it]
31%|███ | 13532/43939 [21:51:32<48:03:44, 5.69s/it]
31%|███ | 13533/43939 [21:51:38<48:07:49, 5.70s/it]
31%|███ | 13534/43939 [21:51:43<48:08:15, 5.70s/it]
31%|███ | 13535/43939 [21:51:49<48:03:42, 5.69s/it]
31%|███ | 13536/43939 [21:51:55<48:06:15, 5.70s/it]
31%|███ | 13537/43939 [21:52:01<48:42:52, 5.77s/it]
31%|███ | 13538/43939 [21:52:06<48:33:45, 5.75s/it]
31%|███ | 13539/43939 [21:52:12<48:24:39, 5.73s/it]
31%|███ | 13540/43939 [21:52:18<48:20:48, 5.73s/it]
31%|███ | 13541/43939 [21:52:23<48:21:29, 5.73s/it]
31%|███ | 13542/43939 [21:52:29<48:24:42, 5.73s/it]
31%|███ | 13543/43939 [21:52:35<48:22:41, 5.73s/it]
31%|███ | 13544/43939 [21:52:41<48:18:59, 5.72s/it]
31%|███ | 13545/43939 [21:52:46<48:23:27, 5.73s/it]
31%|███ | 13546/43939 [21:52:52<48:18:52, 5.72s/it]
31%|███ | 13547/43939 [21:52:58<48:14:47, 5.71s/it]
31%|███ | 13548/43939 [21:53:03<48:11:43, 5.71s/it]
31%|███ | 13549/43939 [21:53:09<48:14:47, 5.72s/it]
31%|███ | 13550/43939 [21:53:15<48:11:45, 5.71s/it]
{'loss': 0.9251, 'grad_norm': 1.2163076400756836, 'learning_rate': 4.367320601620784e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.1, 'epoch': 1.23}
31%|███ | 13550/43939 [21:53:15<48:11:45, 5.71s/it]
31%|███ | 13551/43939 [21:53:21<48:10:17, 5.71s/it]
31%|███ | 13552/43939 [21:53:26<48:13:13, 5.71s/it]
31%|███ | 13553/43939 [21:53:32<48:14:39, 5.72s/it]
31%|███ | 13554/43939 [21:53:38<48:10:07, 5.71s/it]
31%|███ | 13555/43939 [21:53:43<48:05:48, 5.70s/it]
31%|███ | 13556/43939 [21:53:49<47:59:39, 5.69s/it]
31%|███ | 13557/43939 [21:53:55<48:00:31, 5.69s/it]
31%|███ | 13558/43939 [21:54:01<48:40:22, 5.77s/it]
31%|███ | 13559/43939 [21:54:06<48:33:42, 5.75s/it]
31%|███ | 13560/43939 [21:54:12<48:31:33, 5.75s/it]
31%|███ | 13561/43939 [21:54:18<48:29:23, 5.75s/it]
31%|███ | 13562/43939 [21:54:24<48:18:34, 5.73s/it]
31%|███ | 13563/43939 [21:54:29<48:20:18, 5.73s/it]
31%|███ | 13564/43939 [21:54:35<48:15:59, 5.72s/it]
31%|███ | 13565/43939 [21:54:41<48:15:00, 5.72s/it]
31%|███ | 13566/43939 [21:54:46<48:17:44, 5.72s/it]
31%|███ | 13567/43939 [21:54:52<48:16:17, 5.72s/it]
31%|███ | 13568/43939 [21:54:58<48:09:28, 5.71s/it]
31%|███ | 13569/43939 [21:55:03<48:00:31, 5.69s/it]
31%|███ | 13570/43939 [21:55:09<48:01:20, 5.69s/it]
31%|███ | 13571/43939 [21:55:15<47:59:31, 5.69s/it]
31%|███ | 13572/43939 [21:55:21<48:03:58, 5.70s/it]
31%|███ | 13573/43939 [21:55:26<47:58:24, 5.69s/it]
31%|███ | 13574/43939 [21:55:32<47:57:23, 5.69s/it]
31%|███ | 13575/43939 [21:55:38<47:59:02, 5.69s/it]
31%|███ | 13576/43939 [21:55:43<47:54:15, 5.68s/it]
31%|███ | 13577/43939 [21:55:49<47:51:10, 5.67s/it]
31%|███ | 13578/43939 [21:55:55<47:53:19, 5.68s/it]
31%|███ | 13579/43939 [21:56:01<48:26:13, 5.74s/it]
31%|███ | 13580/43939 [21:56:06<48:27:00, 5.75s/it]
31%|███ | 13581/43939 [21:56:12<48:21:28, 5.73s/it]
31%|███ | 13582/43939 [21:56:18<48:19:23, 5.73s/it]
31%|███ | 13583/43939 [21:56:23<48:15:42, 5.72s/it]
31%|███ | 13584/43939 [21:56:29<48:08:03, 5.71s/it]
31%|███ | 13585/43939 [21:56:35<48:02:32, 5.70s/it]
31%|███ | 13586/43939 [21:56:40<48:03:27, 5.70s/it]
31%|███ | 13587/43939 [21:56:46<47:57:44, 5.69s/it]
31%|███ | 13588/43939 [21:56:52<47:57:48, 5.69s/it]
31%|███ | 13589/43939 [21:56:58<48:01:34, 5.70s/it]
31%|███ | 13590/43939 [21:57:03<47:58:47, 5.69s/it]
31%|███ | 13591/43939 [21:57:09<47:58:51, 5.69s/it]
31%|███ | 13592/43939 [21:57:15<48:01:28, 5.70s/it]
31%|███ | 13593/43939 [21:57:20<48:06:44, 5.71s/it]
31%|███ | 13594/43939 [21:57:26<48:04:29, 5.70s/it]
31%|███ | 13595/43939 [21:57:32<48:12:04, 5.72s/it]
31%|███ | 13596/43939 [21:57:38<48:13:22, 5.72s/it]
31%|███ | 13597/43939 [21:57:43<48:11:48, 5.72s/it]
31%|███ | 13598/43939 [21:57:49<48:11:05, 5.72s/it]
31%|███ | 13599/43939 [21:57:55<48:12:15, 5.72s/it]
31%|███ | 13600/43939 [21:58:01<48:44:39, 5.78s/it]
{'loss': 0.9002, 'grad_norm': 1.3128591775894165, 'learning_rate': 4.360703262283568e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 433.67, 'epoch': 1.24}
31%|███ | 13600/43939 [21:58:01<48:44:39, 5.78s/it]
31%|███ | 13601/43939 [21:58:06<48:31:48, 5.76s/it]
31%|███ | 13602/43939 [21:58:12<48:20:51, 5.74s/it]
31%|███ | 13603/43939 [21:58:18<48:13:46, 5.72s/it]
31%|███ | 13604/43939 [21:58:23<48:04:40, 5.71s/it]
31%|███ | 13605/43939 [21:58:29<48:00:31, 5.70s/it]
31%|███ | 13606/43939 [21:58:35<48:11:56, 5.72s/it]
31%|███ | 13607/43939 [21:58:40<48:09:32, 5.72s/it]
31%|███ | 13608/43939 [21:58:46<48:07:04, 5.71s/it]
31%|███ | 13609/43939 [21:58:52<48:11:43, 5.72s/it]
31%|███ | 13610/43939 [21:58:58<48:12:41, 5.72s/it]
31%|███ | 13611/43939 [21:59:03<48:08:39, 5.71s/it]
31%|███ | 13612/43939 [21:59:09<48:02:29, 5.70s/it]
31%|███ | 13613/43939 [21:59:15<48:06:05, 5.71s/it]
31%|███ | 13614/43939 [21:59:20<48:04:50, 5.71s/it]
31%|███ | 13615/43939 [21:59:26<47:58:48, 5.70s/it]
31%|███ | 13616/43939 [21:59:32<48:00:15, 5.70s/it]
31%|███ | 13617/43939 [21:59:38<48:01:24, 5.70s/it]
31%|███ | 13618/43939 [21:59:43<48:00:17, 5.70s/it]
31%|███ | 13619/43939 [21:59:49<48:00:33, 5.70s/it]
31%|███ | 13620/43939 [21:59:55<48:00:45, 5.70s/it]
31%|███ | 13621/43939 [22:00:01<48:33:37, 5.77s/it]
31%|███ | 13622/43939 [22:00:06<48:24:06, 5.75s/it]
31%|███ | 13623/43939 [22:00:12<48:20:44, 5.74s/it]
31%|███ | 13624/43939 [22:00:18<48:13:11, 5.73s/it]
31%|███ | 13625/43939 [22:00:23<48:03:41, 5.71s/it]
31%|███ | 13626/43939 [22:00:29<48:01:47, 5.70s/it]
31%|███ | 13627/43939 [22:00:35<48:04:07, 5.71s/it]
31%|███ | 13628/43939 [22:00:40<47:56:19, 5.69s/it]
31%|███ | 13629/43939 [22:00:46<47:57:12, 5.70s/it]
31%|███ | 13630/43939 [22:00:52<48:00:05, 5.70s/it]
31%|███ | 13631/43939 [22:00:58<48:01:40, 5.70s/it]
31%|███ | 13632/43939 [22:01:03<47:56:04, 5.69s/it]
31%|███ | 13633/43939 [22:01:09<47:57:00, 5.70s/it]
31%|███ | 13634/43939 [22:01:15<47:58:12, 5.70s/it]
31%|███ | 13635/43939 [22:01:20<47:59:41, 5.70s/it]
31%|███ | 13636/43939 [22:01:26<48:00:55, 5.70s/it]
31%|███ | 13637/43939 [22:01:32<47:56:25, 5.70s/it]
31%|███ | 13638/43939 [22:01:37<47:54:25, 5.69s/it]
31%|███ | 13639/43939 [22:01:43<48:04:07, 5.71s/it]
31%|███ | 13640/43939 [22:01:49<47:59:36, 5.70s/it]
31%|███ | 13641/43939 [22:01:55<47:59:01, 5.70s/it]
31%|███ | 13642/43939 [22:02:00<47:57:14, 5.70s/it]
31%|███ | 13643/43939 [22:02:06<48:33:20, 5.77s/it]
31%|███ | 13644/43939 [22:02:12<48:27:29, 5.76s/it]
31%|███ | 13645/43939 [22:02:18<48:19:48, 5.74s/it]
31%|███ | 13646/43939 [22:02:23<48:09:34, 5.72s/it]
31%|███ | 13647/43939 [22:02:29<48:07:14, 5.72s/it]
31%|███ | 13648/43939 [22:02:35<48:09:40, 5.72s/it]
31%|███ | 13649/43939 [22:02:40<48:09:19, 5.72s/it]
31%|███ | 13650/43939 [22:02:46<48:14:31, 5.73s/it]
{'loss': 0.9312, 'grad_norm': 1.1230911016464233, 'learning_rate': 4.354056565979796e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 454.13, 'epoch': 1.24}
31%|███ | 13650/43939 [22:02:46<48:14:31, 5.73s/it]
31%|███ | 13651/43939 [22:02:52<48:08:42, 5.72s/it]
31%|███ | 13652/43939 [22:02:58<48:07:03, 5.72s/it]
31%|███ | 13653/43939 [22:03:03<48:03:09, 5.71s/it]
31%|███ | 13654/43939 [22:03:09<48:02:36, 5.71s/it]
31%|███ | 13655/43939 [22:03:15<47:59:50, 5.71s/it]
31%|███ | 13656/43939 [22:03:20<47:59:59, 5.71s/it]
31%|███ | 13657/43939 [22:03:26<47:56:43, 5.70s/it]
31%|███ | 13658/43939 [22:03:32<48:00:13, 5.71s/it]
31%|███ | 13659/43939 [22:03:37<47:54:45, 5.70s/it]
31%|███ | 13660/43939 [22:03:43<47:55:31, 5.70s/it]
31%|███ | 13661/43939 [22:03:49<47:56:35, 5.70s/it]
31%|███ | 13662/43939 [22:03:55<47:56:17, 5.70s/it]
31%|███ | 13663/43939 [22:04:00<47:58:43, 5.70s/it]
31%|███ | 13664/43939 [22:04:06<48:24:18, 5.76s/it]
31%|███ | 13665/43939 [22:04:12<48:16:17, 5.74s/it]
31%|███ | 13666/43939 [22:04:18<48:11:47, 5.73s/it]
31%|███ | 13667/43939 [22:04:23<48:13:47, 5.74s/it]
31%|███ | 13668/43939 [22:04:29<48:15:03, 5.74s/it]
31%|███ | 13669/43939 [22:04:35<48:10:53, 5.73s/it]
31%|███ | 13670/43939 [22:04:40<48:03:49, 5.72s/it]
31%|███ | 13671/43939 [22:04:46<47:59:06, 5.71s/it]
31%|███ | 13672/43939 [22:04:52<47:53:19, 5.70s/it]
31%|███ | 13673/43939 [22:04:58<47:53:24, 5.70s/it]
31%|███ | 13674/43939 [22:05:03<47:51:12, 5.69s/it]
31%|███ | 13675/43939 [22:05:09<47:54:20, 5.70s/it]
31%|███ | 13676/43939 [22:05:15<48:00:07, 5.71s/it]
31%|███ | 13677/43939 [22:05:20<48:01:13, 5.71s/it]
31%|███ | 13678/43939 [22:05:26<47:56:46, 5.70s/it]
31%|███ | 13679/43939 [22:05:32<47:57:13, 5.70s/it]
31%|███ | 13680/43939 [22:05:38<47:58:43, 5.71s/it]
31%|███ | 13681/43939 [22:05:43<47:50:55, 5.69s/it]
31%|███ | 13682/43939 [22:05:49<47:44:58, 5.68s/it]
31%|███ | 13683/43939 [22:05:54<47:40:40, 5.67s/it]
31%|███ | 13684/43939 [22:06:00<47:42:22, 5.68s/it]
31%|███ | 13685/43939 [22:06:06<48:15:36, 5.74s/it]
31%|███ | 13686/43939 [22:06:12<48:10:02, 5.73s/it]
31%|███ | 13687/43939 [22:06:17<48:01:17, 5.71s/it]
31%|███ | 13688/43939 [22:06:23<47:58:29, 5.71s/it]
31%|███ | 13689/43939 [22:06:29<47:57:48, 5.71s/it]
31%|███ | 13690/43939 [22:06:35<47:54:30, 5.70s/it]
31%|███ | 13691/43939 [22:06:40<47:52:07, 5.70s/it]
31%|███ | 13692/43939 [22:06:46<47:45:06, 5.68s/it]
31%|███ | 13693/43939 [22:06:52<47:41:47, 5.68s/it]
31%|███ | 13694/43939 [22:06:57<47:44:54, 5.68s/it]
31%|███ | 13695/43939 [22:07:03<47:45:56, 5.69s/it]
31%|███ | 13696/43939 [22:07:09<47:51:58, 5.70s/it]
31%|███ | 13697/43939 [22:07:14<47:43:01, 5.68s/it]
31%|███ | 13698/43939 [22:07:20<47:47:14, 5.69s/it]
31%|███ | 13699/43939 [22:07:26<47:54:40, 5.70s/it]
31%|███ | 13700/43939 [22:07:31<47:57:05, 5.71s/it]
{'loss': 0.9268, 'grad_norm': 1.0309466123580933, 'learning_rate': 4.347380617576723e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.99, 'epoch': 1.25}
31%|███ | 13700/43939 [22:07:31<47:57:05, 5.71s/it]
31%|███ | 13701/43939 [22:07:37<48:03:49, 5.72s/it]
31%|███ | 13702/43939 [22:07:43<48:01:21, 5.72s/it]
31%|███ | 13703/43939 [22:07:49<48:01:56, 5.72s/it]
31%|███ | 13704/43939 [22:07:54<47:58:20, 5.71s/it]
31%|███ | 13705/43939 [22:08:00<48:00:56, 5.72s/it]
31%|███ | 13706/43939 [22:08:06<48:30:41, 5.78s/it]
31%|███ | 13707/43939 [22:08:12<48:19:28, 5.75s/it]
31%|███ | 13708/43939 [22:08:17<48:10:16, 5.74s/it]
31%|███ | 13709/43939 [22:08:23<48:02:36, 5.72s/it]
31%|███ | 13710/43939 [22:08:29<47:57:07, 5.71s/it]
31%|███ | 13711/43939 [22:08:34<47:59:56, 5.72s/it]
31%|███ | 13712/43939 [22:08:40<47:59:12, 5.72s/it]
31%|███ | 13713/43939 [22:08:46<47:55:25, 5.71s/it]
31%|███ | 13714/43939 [22:08:52<47:53:06, 5.70s/it]
31%|███ | 13715/43939 [22:08:57<48:02:59, 5.72s/it]
31%|███ | 13716/43939 [22:09:03<47:54:56, 5.71s/it]
31%|███ | 13717/43939 [22:09:09<48:01:51, 5.72s/it]
31%|███ | 13718/43939 [22:09:14<48:02:04, 5.72s/it]
31%|███ | 13719/43939 [22:09:20<47:53:00, 5.70s/it]
31%|███ | 13720/43939 [22:09:26<47:48:29, 5.70s/it]
31%|███ | 13721/43939 [22:09:32<47:48:01, 5.69s/it]
31%|███ | 13722/43939 [22:09:37<47:55:01, 5.71s/it]
31%|███ | 13723/43939 [22:09:43<47:51:16, 5.70s/it]
31%|███ | 13724/43939 [22:09:49<47:59:03, 5.72s/it]
31%|███ | 13725/43939 [22:09:54<47:57:52, 5.71s/it]
31%|███ | 13726/43939 [22:10:00<47:47:26, 5.69s/it]
31%|███ | 13727/43939 [22:10:06<48:17:54, 5.76s/it]
31%|███ | 13728/43939 [22:10:12<48:15:16, 5.75s/it]
31%|███ | 13729/43939 [22:10:17<48:05:50, 5.73s/it]
31%|███ | 13730/43939 [22:10:23<48:04:30, 5.73s/it]
31%|███▏ | 13731/43939 [22:10:29<47:58:32, 5.72s/it]
31%|███▏ | 13732/43939 [22:10:34<47:55:02, 5.71s/it]
31%|███▏ | 13733/43939 [22:10:40<47:43:37, 5.69s/it]
31%|███▏ | 13734/43939 [22:10:46<47:47:55, 5.70s/it]
31%|███▏ | 13735/43939 [22:10:52<47:44:18, 5.69s/it]
31%|███▏ | 13736/43939 [22:10:57<47:51:55, 5.71s/it]
31%|███▏ | 13737/43939 [22:11:03<47:53:20, 5.71s/it]
31%|███▏ | 13738/43939 [22:11:09<47:55:35, 5.71s/it]
31%|███▏ | 13739/43939 [22:11:14<47:49:39, 5.70s/it]
31%|███▏ | 13740/43939 [22:11:20<47:49:59, 5.70s/it]
31%|███▏ | 13741/43939 [22:11:26<47:46:38, 5.70s/it]
31%|███▏ | 13742/43939 [22:11:31<47:47:15, 5.70s/it]
31%|███▏ | 13743/43939 [22:11:37<47:47:52, 5.70s/it]
31%|███▏ | 13744/43939 [22:11:43<47:55:58, 5.71s/it]
31%|███▏ | 13745/43939 [22:11:49<47:59:03, 5.72s/it]
31%|███▏ | 13746/43939 [22:11:54<47:56:15, 5.72s/it]
31%|███▏ | 13747/43939 [22:12:00<48:02:28, 5.73s/it]
31%|███▏ | 13748/43939 [22:12:06<48:28:16, 5.78s/it]
31%|███▏ | 13749/43939 [22:12:12<48:16:46, 5.76s/it]
31%|███▏ | 13750/43939 [22:12:17<48:11:57, 5.75s/it]
{'loss': 0.9517, 'grad_norm': 0.8729075193405151, 'learning_rate': 4.3406755224031225e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.03, 'epoch': 1.25}
31%|███▏ | 13750/43939 [22:12:17<48:11:57, 5.75s/it]
31%|███▏ | 13751/43939 [22:12:23<47:58:08, 5.72s/it]
31%|███▏ | 13752/43939 [22:12:29<47:48:12, 5.70s/it]
31%|███▏ | 13753/43939 [22:12:34<47:45:52, 5.70s/it]
31%|███▏ | 13754/43939 [22:12:40<47:49:46, 5.70s/it]
31%|███▏ | 13755/43939 [22:12:46<47:53:49, 5.71s/it]
31%|███▏ | 13756/43939 [22:12:52<47:49:45, 5.70s/it]
31%|███▏ | 13757/43939 [22:12:57<47:53:03, 5.71s/it]
31%|███▏ | 13758/43939 [22:13:03<47:52:29, 5.71s/it]
31%|███▏ | 13759/43939 [22:13:09<47:51:37, 5.71s/it]
31%|███▏ | 13760/43939 [22:13:14<47:41:44, 5.69s/it]
31%|███▏ | 13761/43939 [22:13:20<47:38:22, 5.68s/it]
31%|███▏ | 13762/43939 [22:13:26<47:39:57, 5.69s/it]
31%|███▏ | 13763/43939 [22:13:31<47:47:35, 5.70s/it]
31%|███▏ | 13764/43939 [22:13:37<47:43:24, 5.69s/it]
31%|███▏ | 13765/43939 [22:13:43<47:38:35, 5.68s/it]
31%|███▏ | 13766/43939 [22:13:48<47:39:51, 5.69s/it]
31%|███▏ | 13767/43939 [22:13:54<47:42:46, 5.69s/it]
31%|███▏ | 13768/43939 [22:14:00<48:25:57, 5.78s/it]
31%|███▏ | 13769/43939 [22:14:06<48:08:36, 5.74s/it]
31%|███▏ | 13770/43939 [22:14:11<47:56:26, 5.72s/it]
31%|███▏ | 13771/43939 [22:14:17<47:59:08, 5.73s/it]
31%|███▏ | 13772/43939 [22:14:23<47:57:36, 5.72s/it]
31%|███▏ | 13773/43939 [22:14:29<47:52:55, 5.71s/it]
31%|███▏ | 13774/43939 [22:14:34<47:58:57, 5.73s/it]
31%|███▏ | 13775/43939 [22:14:40<47:49:56, 5.71s/it]
31%|███▏ | 13776/43939 [22:14:46<47:46:06, 5.70s/it]
31%|███▏ | 13777/43939 [22:14:52<47:57:27, 5.72s/it]
31%|███▏ | 13778/43939 [22:14:57<47:54:35, 5.72s/it]
31%|███▏ | 13779/43939 [22:15:03<47:50:29, 5.71s/it]
31%|███▏ | 13780/43939 [22:15:09<47:46:37, 5.70s/it]
31%|███▏ | 13781/43939 [22:15:14<48:01:44, 5.73s/it]
31%|███▏ | 13782/43939 [22:15:20<47:49:26, 5.71s/it]
31%|███▏ | 13783/43939 [22:15:26<47:52:06, 5.71s/it]
31%|███▏ | 13784/43939 [22:15:31<47:42:23, 5.70s/it]
31%|███▏ | 13785/43939 [22:15:37<47:39:15, 5.69s/it]
31%|███▏ | 13786/43939 [22:15:43<47:43:29, 5.70s/it]
31%|███▏ | 13787/43939 [22:15:49<47:49:21, 5.71s/it]
31%|███▏ | 13788/43939 [22:15:54<47:47:26, 5.71s/it]
31%|███▏ | 13789/43939 [22:16:00<47:38:31, 5.69s/it]
31%|███▏ | 13790/43939 [22:16:06<48:11:57, 5.76s/it]
31%|███▏ | 13791/43939 [22:16:12<48:04:13, 5.74s/it]
31%|███▏ | 13792/43939 [22:16:17<47:56:31, 5.72s/it]
31%|███▏ | 13793/43939 [22:16:23<48:01:10, 5.73s/it]
31%|███▏ | 13794/43939 [22:16:29<47:53:29, 5.72s/it]
31%|███▏ | 13795/43939 [22:16:34<47:49:09, 5.71s/it]
31%|███▏ | 13796/43939 [22:16:40<47:55:20, 5.72s/it]
31%|███▏ | 13797/43939 [22:16:46<47:51:20, 5.72s/it]
31%|███▏ | 13798/43939 [22:16:52<47:48:54, 5.71s/it]
31%|███▏ | 13799/43939 [22:16:57<47:47:58, 5.71s/it]
31%|███▏ | 13800/43939 [22:17:03<47:46:47, 5.71s/it]
{'loss': 0.9236, 'grad_norm': 1.2097883224487305, 'learning_rate': 4.3339413862476285e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.31, 'epoch': 1.26}
31%|███▏ | 13800/43939 [22:17:03<47:46:47, 5.71s/it]
31%|███▏ | 13801/43939 [22:17:09<47:41:35, 5.70s/it]
31%|███▏ | 13802/43939 [22:17:14<47:40:05, 5.69s/it]
31%|███▏ | 13803/43939 [22:17:20<47:49:20, 5.71s/it]
31%|███▏ | 13804/43939 [22:17:26<47:53:25, 5.72s/it]
31%|███▏ | 13805/43939 [22:17:32<47:54:17, 5.72s/it]
31%|███▏ | 13806/43939 [22:17:37<47:52:58, 5.72s/it]
31%|███▏ | 13807/43939 [22:17:43<47:51:50, 5.72s/it]
31%|███▏ | 13808/43939 [22:17:49<47:53:46, 5.72s/it]
31%|███▏ | 13809/43939 [22:17:54<47:45:58, 5.71s/it]
31%|███▏ | 13810/43939 [22:18:00<48:14:11, 5.76s/it]
31%|███▏ | 13811/43939 [22:18:06<48:03:59, 5.74s/it]
31%|███▏ | 13812/43939 [22:18:12<47:55:07, 5.73s/it]
31%|███▏ | 13813/43939 [22:18:17<47:59:53, 5.74s/it]
31%|███▏ | 13814/43939 [22:18:23<47:51:17, 5.72s/it]
31%|███▏ | 13815/43939 [22:18:29<47:52:32, 5.72s/it]
31%|███▏ | 13816/43939 [22:18:34<47:50:16, 5.72s/it]
31%|███▏ | 13817/43939 [22:18:40<47:42:58, 5.70s/it]
31%|███▏ | 13818/43939 [22:18:46<47:44:34, 5.71s/it]
31%|███▏ | 13819/43939 [22:18:52<47:44:00, 5.71s/it]
31%|███▏ | 13820/43939 [22:18:57<47:41:05, 5.70s/it]
31%|███▏ | 13821/43939 [22:19:03<47:43:57, 5.71s/it]
31%|███▏ | 13822/43939 [22:19:09<47:44:13, 5.71s/it]
31%|███▏ | 13823/43939 [22:19:14<47:45:52, 5.71s/it]
31%|███▏ | 13824/43939 [22:19:20<47:42:39, 5.70s/it]
31%|███▏ | 13825/43939 [22:19:26<47:38:42, 5.70s/it]
31%|███▏ | 13826/43939 [22:19:31<47:36:47, 5.69s/it]
31%|███▏ | 13827/43939 [22:19:37<47:42:37, 5.70s/it]
31%|███▏ | 13828/43939 [22:19:43<47:50:44, 5.72s/it]
31%|███▏ | 13829/43939 [22:19:49<47:49:09, 5.72s/it]
31%|███▏ | 13830/43939 [22:19:54<47:44:17, 5.71s/it]
31%|███▏ | 13831/43939 [22:20:00<48:23:44, 5.79s/it]
31%|███▏ | 13832/43939 [22:20:06<48:13:45, 5.77s/it]
31%|███▏ | 13833/43939 [22:20:12<48:05:07, 5.75s/it]
31%|███▏ | 13834/43939 [22:20:17<47:59:30, 5.74s/it]
31%|███▏ | 13835/43939 [22:20:23<47:56:23, 5.73s/it]
31%|███▏ | 13836/43939 [22:20:29<47:56:04, 5.73s/it]
31%|███▏ | 13837/43939 [22:20:35<47:51:14, 5.72s/it]
31%|███▏ | 13838/43939 [22:20:40<47:41:27, 5.70s/it]
31%|███▏ | 13839/43939 [22:20:46<47:35:38, 5.69s/it]
31%|███▏ | 13840/43939 [22:20:52<47:34:46, 5.69s/it]
32%|███▏ | 13841/43939 [22:20:57<47:33:46, 5.69s/it]
32%|███▏ | 13842/43939 [22:21:03<47:43:57, 5.71s/it]
32%|███▏ | 13843/43939 [22:21:09<47:41:06, 5.70s/it]
32%|███▏ | 13844/43939 [22:21:14<47:45:08, 5.71s/it]
32%|███▏ | 13845/43939 [22:21:20<47:43:51, 5.71s/it]
32%|███▏ | 13846/43939 [22:21:26<47:40:16, 5.70s/it]
32%|███▏ | 13847/43939 [22:21:32<47:46:00, 5.71s/it]
32%|███▏ | 13848/43939 [22:21:37<47:45:11, 5.71s/it]
32%|███▏ | 13849/43939 [22:21:43<47:38:04, 5.70s/it]
32%|███▏ | 13850/43939 [22:21:49<47:37:20, 5.70s/it]
{'loss': 0.9261, 'grad_norm': 1.1136751174926758, 'learning_rate': 4.327178315357063e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.2, 'epoch': 1.26}
32%|███▏ | 13850/43939 [22:21:49<47:37:20, 5.70s/it]
32%|███▏ | 13851/43939 [22:21:54<47:36:03, 5.70s/it]
32%|███▏ | 13852/43939 [22:22:00<48:11:27, 5.77s/it]
32%|███▏ | 13853/43939 [22:22:06<47:53:22, 5.73s/it]
32%|███▏ | 13854/43939 [22:22:12<47:50:30, 5.72s/it]
32%|███▏ | 13855/43939 [22:22:17<47:46:34, 5.72s/it]
32%|███▏ | 13856/43939 [22:22:23<47:36:59, 5.70s/it]
32%|███▏ | 13857/43939 [22:22:29<47:29:49, 5.68s/it]
32%|███▏ | 13858/43939 [22:22:34<47:37:52, 5.70s/it]
32%|███▏ | 13859/43939 [22:22:40<47:39:47, 5.70s/it]
32%|███▏ | 13860/43939 [22:22:46<47:41:10, 5.71s/it]
32%|███▏ | 13861/43939 [22:22:52<47:39:25, 5.70s/it]
32%|███▏ | 13862/43939 [22:22:57<47:38:31, 5.70s/it]
32%|███▏ | 13863/43939 [22:23:03<47:40:30, 5.71s/it]
32%|███▏ | 13864/43939 [22:23:09<47:42:04, 5.71s/it]
32%|███▏ | 13865/43939 [22:23:14<47:43:00, 5.71s/it]
32%|███▏ | 13866/43939 [22:23:20<47:43:04, 5.71s/it]
32%|███▏ | 13867/43939 [22:23:26<47:37:51, 5.70s/it]
32%|███▏ | 13868/43939 [22:23:31<47:40:55, 5.71s/it]
32%|███▏ | 13869/43939 [22:23:37<47:39:06, 5.70s/it]
32%|███▏ | 13870/43939 [22:23:43<47:40:54, 5.71s/it]
32%|███▏ | 13871/43939 [22:23:49<47:37:44, 5.70s/it]
32%|███▏ | 13872/43939 [22:23:54<47:45:24, 5.72s/it]
32%|███▏ | 13873/43939 [22:24:00<48:14:37, 5.78s/it]
32%|███▏ | 13874/43939 [22:24:06<48:08:05, 5.76s/it]
32%|███▏ | 13875/43939 [22:24:12<47:53:23, 5.73s/it]
32%|███▏ | 13876/43939 [22:24:17<48:00:25, 5.75s/it]
32%|███▏ | 13877/43939 [22:24:23<47:58:54, 5.75s/it]
32%|███▏ | 13878/43939 [22:24:29<47:51:51, 5.73s/it]
32%|███▏ | 13879/43939 [22:24:35<47:48:56, 5.73s/it]
32%|███▏ | 13880/43939 [22:24:40<47:44:33, 5.72s/it]
32%|███▏ | 13881/43939 [22:24:46<47:36:21, 5.70s/it]
32%|███▏ | 13882/43939 [22:24:52<47:36:54, 5.70s/it]
32%|███▏ | 13883/43939 [22:24:57<47:41:38, 5.71s/it]
32%|███▏ | 13884/43939 [22:25:03<47:41:22, 5.71s/it]
32%|███▏ | 13885/43939 [22:25:09<47:38:02, 5.71s/it]
32%|███▏ | 13886/43939 [22:25:15<47:39:07, 5.71s/it]
32%|███▏ | 13887/43939 [22:25:20<47:45:58, 5.72s/it]
32%|███▏ | 13888/43939 [22:25:26<47:47:12, 5.72s/it]
32%|███▏ | 13889/43939 [22:25:32<47:42:20, 5.72s/it]
32%|███▏ | 13890/43939 [22:25:37<47:32:09, 5.70s/it]
32%|███▏ | 13891/43939 [22:25:43<47:33:12, 5.70s/it]
32%|███▏ | 13892/43939 [22:25:49<47:31:47, 5.69s/it]
32%|███▏ | 13893/43939 [22:25:54<47:31:43, 5.69s/it]
32%|███▏ | 13894/43939 [22:26:00<48:15:06, 5.78s/it]
32%|███▏ | 13895/43939 [22:26:06<48:04:33, 5.76s/it]
32%|███▏ | 13896/43939 [22:26:12<48:01:51, 5.76s/it]
32%|███▏ | 13897/43939 [22:26:18<47:50:33, 5.73s/it]
32%|███▏ | 13898/43939 [22:26:23<47:52:33, 5.74s/it]
32%|███▏ | 13899/43939 [22:26:29<47:43:33, 5.72s/it]
32%|███▏ | 13900/43939 [22:26:35<47:45:05, 5.72s/it]
{'loss': 0.9285, 'grad_norm': 1.1039016246795654, 'learning_rate': 4.320386416434764e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.68, 'epoch': 1.27}
32%|███▏ | 13900/43939 [22:26:35<47:45:05, 5.72s/it]
32%|███▏ | 13901/43939 [22:26:40<47:38:19, 5.71s/it]
32%|███▏ | 13902/43939 [22:26:46<47:42:38, 5.72s/it]
32%|███▏ | 13903/43939 [22:26:52<47:44:54, 5.72s/it]
32%|███▏ | 13904/43939 [22:26:58<47:45:45, 5.72s/it]
32%|███▏ | 13905/43939 [22:27:03<47:41:52, 5.72s/it]
32%|███▏ | 13906/43939 [22:27:09<47:44:41, 5.72s/it]
32%|███▏ | 13907/43939 [22:27:15<47:34:44, 5.70s/it]
32%|███▏ | 13908/43939 [22:27:20<47:36:14, 5.71s/it]
32%|███▏ | 13909/43939 [22:27:26<47:37:49, 5.71s/it]
32%|███▏ | 13910/43939 [22:27:32<47:31:36, 5.70s/it]
32%|███▏ | 13911/43939 [22:27:38<47:34:44, 5.70s/it]
32%|███▏ | 13912/43939 [22:27:43<47:29:57, 5.69s/it]
32%|███▏ | 13913/43939 [22:27:49<47:28:02, 5.69s/it]
32%|███▏ | 13914/43939 [22:27:55<47:28:44, 5.69s/it]
32%|███▏ | 13915/43939 [22:28:00<47:30:17, 5.70s/it]
32%|███▏ | 13916/43939 [22:28:06<48:02:47, 5.76s/it]
32%|███▏ | 13917/43939 [22:28:12<48:00:26, 5.76s/it]
32%|███▏ | 13918/43939 [22:28:18<47:50:18, 5.74s/it]
32%|███▏ | 13919/43939 [22:28:23<47:49:09, 5.73s/it]
32%|███▏ | 13920/43939 [22:28:29<47:39:58, 5.72s/it]
32%|███▏ | 13921/43939 [22:28:35<47:40:53, 5.72s/it]
32%|███▏ | 13922/43939 [22:28:40<47:45:35, 5.73s/it]
32%|███▏ | 13923/43939 [22:28:46<47:40:43, 5.72s/it]
32%|███▏ | 13924/43939 [22:28:52<47:35:35, 5.71s/it]
32%|███▏ | 13925/43939 [22:28:58<47:28:51, 5.70s/it]
32%|███▏ | 13926/43939 [22:29:03<47:33:32, 5.70s/it]
32%|███▏ | 13927/43939 [22:29:09<47:31:34, 5.70s/it]
32%|███▏ | 13928/43939 [22:29:15<47:33:05, 5.70s/it]
32%|███▏ | 13929/43939 [22:29:20<47:38:24, 5.71s/it]
32%|███▏ | 13930/43939 [22:29:26<47:41:36, 5.72s/it]
32%|███▏ | 13931/43939 [22:29:32<47:37:54, 5.71s/it]
32%|███▏ | 13932/43939 [22:29:38<47:45:11, 5.73s/it]
32%|███▏ | 13933/43939 [22:29:43<47:39:09, 5.72s/it]
32%|███▏ | 13934/43939 [22:29:49<47:41:24, 5.72s/it]
32%|███▏ | 13935/43939 [22:29:55<47:39:29, 5.72s/it]
32%|███▏ | 13936/43939 [22:30:00<47:35:55, 5.71s/it]
32%|███▏ | 13937/43939 [22:30:06<48:12:06, 5.78s/it]
32%|███▏ | 13938/43939 [22:30:12<48:03:57, 5.77s/it]
32%|███▏ | 13939/43939 [22:30:18<47:58:08, 5.76s/it]
32%|███▏ | 13940/43939 [22:30:24<47:49:34, 5.74s/it]
32%|███▏ | 13941/43939 [22:30:29<47:43:28, 5.73s/it]
32%|███▏ | 13942/43939 [22:30:35<47:41:47, 5.72s/it]
32%|███▏ | 13943/43939 [22:30:41<47:36:48, 5.71s/it]
32%|███▏ | 13944/43939 [22:30:46<47:35:00, 5.71s/it]
32%|███▏ | 13945/43939 [22:30:52<47:34:01, 5.71s/it]
32%|███▏ | 13946/43939 [22:30:58<47:35:03, 5.71s/it]
32%|███▏ | 13947/43939 [22:31:03<47:26:25, 5.69s/it]
32%|███▏ | 13948/43939 [22:31:09<47:24:43, 5.69s/it]
32%|███▏ | 13949/43939 [22:31:15<47:22:04, 5.69s/it]
32%|███▏ | 13950/43939 [22:31:20<47:24:41, 5.69s/it]
{'loss': 0.9201, 'grad_norm': 1.1589677333831787, 'learning_rate': 4.3135657966388946e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.49, 'epoch': 1.27}
32%|███▏ | 13950/43939 [22:31:20<47:24:41, 5.69s/it]
32%|███▏ | 13951/43939 [22:31:26<47:22:52, 5.69s/it]
32%|███▏ | 13952/43939 [22:31:32<47:31:51, 5.71s/it]
32%|███▏ | 13953/43939 [22:31:38<47:30:17, 5.70s/it]
32%|███▏ | 13954/43939 [22:31:43<47:34:31, 5.71s/it]
32%|███▏ | 13955/43939 [22:31:49<47:43:14, 5.73s/it]
32%|███▏ | 13956/43939 [22:31:55<47:45:10, 5.73s/it]
32%|███▏ | 13957/43939 [22:32:01<47:41:02, 5.73s/it]
32%|███▏ | 13958/43939 [22:32:06<48:12:08, 5.79s/it]
32%|███▏ | 13959/43939 [22:32:12<47:57:16, 5.76s/it]
32%|███▏ | 13960/43939 [22:32:18<47:46:02, 5.74s/it]
32%|███▏ | 13961/43939 [22:32:24<47:45:13, 5.73s/it]
32%|███▏ | 13962/43939 [22:32:29<47:39:25, 5.72s/it]
32%|███▏ | 13963/43939 [22:32:35<47:35:17, 5.72s/it]
32%|███▏ | 13964/43939 [22:32:41<47:32:24, 5.71s/it]
32%|███▏ | 13965/43939 [22:32:46<47:32:15, 5.71s/it]
32%|███▏ | 13966/43939 [22:32:52<47:32:42, 5.71s/it]
32%|███▏ | 13967/43939 [22:32:58<47:35:52, 5.72s/it]
32%|███▏ | 13968/43939 [22:33:04<47:35:06, 5.72s/it]
32%|███▏ | 13969/43939 [22:33:09<47:43:18, 5.73s/it]
32%|███▏ | 13970/43939 [22:33:15<47:40:24, 5.73s/it]
32%|███▏ | 13971/43939 [22:33:21<47:37:23, 5.72s/it]
32%|███▏ | 13972/43939 [22:33:26<47:40:06, 5.73s/it]
32%|███▏ | 13973/43939 [22:33:32<47:36:21, 5.72s/it]
32%|███▏ | 13974/43939 [22:33:38<47:29:54, 5.71s/it]
32%|███▏ | 13975/43939 [22:33:44<47:30:22, 5.71s/it]
32%|███▏ | 13976/43939 [22:33:49<47:35:29, 5.72s/it]
32%|███▏ | 13977/43939 [22:33:55<47:36:12, 5.72s/it]
32%|███▏ | 13978/43939 [22:34:01<47:37:20, 5.72s/it]
32%|███▏ | 13979/43939 [22:34:06<47:34:16, 5.72s/it]
32%|███▏ | 13980/43939 [22:34:12<48:09:54, 5.79s/it]
32%|███▏ | 13981/43939 [22:34:18<47:54:01, 5.76s/it]
32%|███▏ | 13982/43939 [22:34:24<47:45:10, 5.74s/it]
32%|███▏ | 13983/43939 [22:34:29<47:35:25, 5.72s/it]
32%|███▏ | 13984/43939 [22:34:35<47:32:26, 5.71s/it]
32%|███▏ | 13985/43939 [22:34:41<47:39:50, 5.73s/it]
32%|███▏ | 13986/43939 [22:34:47<47:40:47, 5.73s/it]
32%|███▏ | 13987/43939 [22:34:52<47:32:50, 5.71s/it]
32%|███▏ | 13988/43939 [22:34:58<47:28:56, 5.71s/it]
32%|███▏ | 13989/43939 [22:35:04<47:34:41, 5.72s/it]
32%|███▏ | 13990/43939 [22:35:09<47:30:10, 5.71s/it]
32%|███▏ | 13991/43939 [22:35:15<47:28:12, 5.71s/it]
32%|███▏ | 13992/43939 [22:35:21<47:27:38, 5.71s/it]
32%|███▏ | 13993/43939 [22:35:27<47:26:19, 5.70s/it]
32%|███▏ | 13994/43939 [22:35:32<47:24:20, 5.70s/it]
32%|███▏ | 13995/43939 [22:35:38<47:25:31, 5.70s/it]
32%|███▏ | 13996/43939 [22:35:44<47:24:01, 5.70s/it]
32%|███▏ | 13997/43939 [22:35:49<47:29:45, 5.71s/it]
32%|███▏ | 13998/43939 [22:35:55<47:25:44, 5.70s/it]
32%|███▏ | 13999/43939 [22:36:01<47:27:40, 5.71s/it]
32%|███▏ | 14000/43939 [22:36:07<47:29:17, 5.71s/it]
{'loss': 0.9389, 'grad_norm': 1.0915837287902832, 'learning_rate': 4.306716563580763e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.74, 'epoch': 1.27}
32%|███▏ | 14000/43939 [22:36:07<47:29:17, 5.71s/it]
32%|███▏ | 14001/43939 [22:36:13<48:09:32, 5.79s/it]
32%|███▏ | 14002/43939 [22:36:18<47:56:45, 5.77s/it]
32%|███▏ | 14003/43939 [22:36:24<47:43:55, 5.74s/it]
32%|███▏ | 14004/43939 [22:36:30<47:42:19, 5.74s/it]
32%|███▏ | 14005/43939 [22:36:35<47:34:10, 5.72s/it]
32%|███▏ | 14006/43939 [22:36:41<47:38:17, 5.73s/it]
32%|███▏ | 14007/43939 [22:36:47<47:34:03, 5.72s/it]
32%|███▏ | 14008/43939 [22:36:52<47:29:28, 5.71s/it]
32%|███▏ | 14009/43939 [22:36:58<47:32:21, 5.72s/it]
32%|███▏ | 14010/43939 [22:37:04<47:28:36, 5.71s/it]
32%|███▏ | 14011/43939 [22:37:10<47:28:43, 5.71s/it]
32%|███▏ | 14012/43939 [22:37:15<47:22:40, 5.70s/it]
32%|███▏ | 14013/43939 [22:37:21<47:26:17, 5.71s/it]
32%|███▏ | 14014/43939 [22:37:27<47:37:19, 5.73s/it]
32%|███▏ | 14015/43939 [22:37:32<47:28:19, 5.71s/it]
32%|███▏ | 14016/43939 [22:37:38<47:33:48, 5.72s/it]
32%|███▏ | 14017/43939 [22:37:44<47:33:23, 5.72s/it]
32%|███▏ | 14018/43939 [22:37:50<47:28:53, 5.71s/it]
32%|███▏ | 14019/43939 [22:37:55<47:24:12, 5.70s/it]
32%|███▏ | 14020/43939 [22:38:01<47:26:20, 5.71s/it]
32%|███▏ | 14021/43939 [22:38:07<48:05:03, 5.79s/it]
32%|███▏ | 14022/43939 [22:38:13<47:50:53, 5.76s/it]
32%|███▏ | 14023/43939 [22:38:18<47:40:08, 5.74s/it]
32%|███▏ | 14024/43939 [22:38:24<47:38:16, 5.73s/it]
32%|███▏ | 14025/43939 [22:38:30<47:32:32, 5.72s/it]
32%|███▏ | 14026/43939 [22:38:35<47:30:04, 5.72s/it]
32%|███▏ | 14027/43939 [22:38:41<47:23:42, 5.70s/it]
32%|███▏ | 14028/43939 [22:38:47<47:22:02, 5.70s/it]
32%|███▏ | 14029/43939 [22:38:53<47:27:13, 5.71s/it]
32%|███▏ | 14030/43939 [22:38:58<47:22:22, 5.70s/it]
32%|███▏ | 14031/43939 [22:39:04<47:22:28, 5.70s/it]
32%|███▏ | 14032/43939 [22:39:10<47:20:12, 5.70s/it]
32%|███▏ | 14033/43939 [22:39:15<47:17:20, 5.69s/it]
32%|███▏ | 14034/43939 [22:39:21<47:23:02, 5.70s/it]
32%|███▏ | 14035/43939 [22:39:27<47:21:24, 5.70s/it]
32%|███▏ | 14036/43939 [22:39:32<47:13:59, 5.69s/it]
32%|███▏ | 14037/43939 [22:39:38<47:11:48, 5.68s/it]
32%|███▏ | 14038/43939 [22:39:44<47:15:05, 5.69s/it]
32%|███▏ | 14039/43939 [22:39:49<47:15:16, 5.69s/it]
32%|███▏ | 14040/43939 [22:39:55<47:17:30, 5.69s/it]
32%|███▏ | 14041/43939 [22:40:01<47:20:52, 5.70s/it]
32%|███▏ | 14042/43939 [22:40:07<47:56:38, 5.77s/it]
32%|███▏ | 14043/43939 [22:40:13<47:42:04, 5.74s/it]
32%|███▏ | 14044/43939 [22:40:18<47:29:37, 5.72s/it]
32%|███▏ | 14045/43939 [22:40:24<47:26:40, 5.71s/it]
32%|███▏ | 14046/43939 [22:40:30<47:19:52, 5.70s/it]
32%|███▏ | 14047/43939 [22:40:35<47:19:23, 5.70s/it]
32%|███▏ | 14048/43939 [22:40:41<47:21:23, 5.70s/it]
32%|███▏ | 14049/43939 [22:40:47<47:22:45, 5.71s/it]
32%|███▏ | 14050/43939 [22:40:52<47:19:54, 5.70s/it]
{'loss': 0.9502, 'grad_norm': 1.0671683549880981, 'learning_rate': 4.299838825323115e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.69, 'epoch': 1.28}
32%|███▏ | 14050/43939 [22:40:52<47:19:54, 5.70s/it]
32%|███▏ | 14051/43939 [22:40:58<47:23:51, 5.71s/it]
32%|███▏ | 14052/43939 [22:41:04<47:18:22, 5.70s/it]
32%|███▏ | 14053/43939 [22:41:09<47:12:04, 5.69s/it]
32%|███▏ | 14054/43939 [22:41:15<47:12:27, 5.69s/it]
32%|███▏ | 14055/43939 [22:41:21<47:12:18, 5.69s/it]
32%|███▏ | 14056/43939 [22:41:26<47:15:12, 5.69s/it]
32%|███▏ | 14057/43939 [22:41:32<47:15:15, 5.69s/it]
32%|███▏ | 14058/43939 [22:41:38<47:16:23, 5.70s/it]
32%|███▏ | 14059/43939 [22:41:44<47:22:21, 5.71s/it]
32%|███▏ | 14060/43939 [22:41:49<47:23:26, 5.71s/it]
32%|███▏ | 14061/43939 [22:41:55<47:32:06, 5.73s/it]
32%|███▏ | 14062/43939 [22:42:01<47:30:04, 5.72s/it]
32%|███▏ | 14063/43939 [22:42:07<47:29:29, 5.72s/it]
32%|███▏ | 14064/43939 [22:42:12<47:53:47, 5.77s/it]
32%|███▏ | 14065/43939 [22:42:18<47:42:45, 5.75s/it]
32%|███▏ | 14066/43939 [22:42:24<47:36:24, 5.74s/it]
32%|███▏ | 14067/43939 [22:42:30<47:27:21, 5.72s/it]
32%|███▏ | 14068/43939 [22:42:35<47:26:33, 5.72s/it]
32%|███▏ | 14069/43939 [22:42:41<47:23:33, 5.71s/it]
32%|███▏ | 14070/43939 [22:42:47<47:18:19, 5.70s/it]
32%|███▏ | 14071/43939 [22:42:52<47:17:36, 5.70s/it]
32%|███▏ | 14072/43939 [22:42:58<47:12:48, 5.69s/it]
32%|███▏ | 14073/43939 [22:43:04<47:11:22, 5.69s/it]
32%|███▏ | 14074/43939 [22:43:09<47:11:38, 5.69s/it]
32%|███▏ | 14075/43939 [22:43:15<47:10:03, 5.69s/it]
32%|███▏ | 14076/43939 [22:43:21<47:13:39, 5.69s/it]
32%|███▏ | 14077/43939 [22:43:26<47:10:07, 5.69s/it]
32%|███▏ | 14078/43939 [22:43:32<47:13:57, 5.69s/it]
32%|███▏ | 14079/43939 [22:43:38<47:14:14, 5.70s/it]
32%|███▏ | 14080/43939 [22:43:44<47:14:58, 5.70s/it]
32%|███▏ | 14081/43939 [22:43:49<47:17:26, 5.70s/it]
32%|███▏ | 14082/43939 [22:43:55<47:10:26, 5.69s/it]
32%|███▏ | 14083/43939 [22:44:01<47:14:24, 5.70s/it]
32%|███▏ | 14084/43939 [22:44:07<47:51:50, 5.77s/it]
32%|███▏ | 14085/43939 [22:44:12<47:49:00, 5.77s/it]
32%|███▏ | 14086/43939 [22:44:18<47:40:11, 5.75s/it]
32%|███▏ | 14087/43939 [22:44:24<47:26:48, 5.72s/it]
32%|███▏ | 14088/43939 [22:44:29<47:15:16, 5.70s/it]
32%|███▏ | 14089/43939 [22:44:35<47:11:41, 5.69s/it]
32%|███▏ | 14090/43939 [22:44:41<47:11:28, 5.69s/it]
32%|███▏ | 14091/43939 [22:44:46<47:17:51, 5.70s/it]
32%|███▏ | 14092/43939 [22:44:52<47:10:18, 5.69s/it]
32%|███▏ | 14093/43939 [22:44:58<47:11:02, 5.69s/it]
32%|███▏ | 14094/43939 [22:45:03<47:12:50, 5.70s/it]
32%|███▏ | 14095/43939 [22:45:09<47:11:47, 5.69s/it]
32%|███▏ | 14096/43939 [22:45:15<47:10:20, 5.69s/it]
32%|███▏ | 14097/43939 [22:45:21<47:13:57, 5.70s/it]
32%|███▏ | 14098/43939 [22:45:26<47:23:21, 5.72s/it]
32%|███▏ | 14099/43939 [22:45:32<47:15:43, 5.70s/it]
32%|███▏ | 14100/43939 [22:45:38<47:15:36, 5.70s/it]
{'loss': 0.9269, 'grad_norm': 1.0630742311477661, 'learning_rate': 4.2929326903784345e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.24, 'epoch': 1.28}
32%|███▏ | 14100/43939 [22:45:38<47:15:36, 5.70s/it]
32%|███▏ | 14101/43939 [22:45:43<47:18:18, 5.71s/it]
32%|███▏ | 14102/43939 [22:45:49<47:16:53, 5.70s/it]
32%|███▏ | 14103/43939 [22:45:55<47:12:36, 5.70s/it]
32%|███▏ | 14104/43939 [22:46:00<47:08:14, 5.69s/it]
32%|███▏ | 14105/43939 [22:46:06<47:12:00, 5.70s/it]
32%|███▏ | 14106/43939 [22:46:12<47:53:02, 5.78s/it]
32%|███▏ | 14107/43939 [22:46:18<47:39:58, 5.75s/it]
32%|███▏ | 14108/43939 [22:46:24<47:36:32, 5.75s/it]
32%|███▏ | 14109/43939 [22:46:29<47:35:32, 5.74s/it]
32%|███▏ | 14110/43939 [22:46:35<47:23:31, 5.72s/it]
32%|███▏ | 14111/43939 [22:46:41<47:19:25, 5.71s/it]
32%|███▏ | 14112/43939 [22:46:46<47:22:20, 5.72s/it]
32%|███▏ | 14113/43939 [22:46:52<47:19:02, 5.71s/it]
32%|███▏ | 14114/43939 [22:46:58<47:16:15, 5.71s/it]
32%|███▏ | 14115/43939 [22:47:04<47:20:47, 5.72s/it]
32%|███▏ | 14116/43939 [22:47:09<47:12:53, 5.70s/it]
32%|███▏ | 14117/43939 [22:47:15<47:14:00, 5.70s/it]
32%|███▏ | 14118/43939 [22:47:21<47:14:49, 5.70s/it]
32%|███▏ | 14119/43939 [22:47:26<47:24:12, 5.72s/it]
32%|███▏ | 14120/43939 [22:47:32<47:20:32, 5.72s/it]
32%|███▏ | 14121/43939 [22:47:38<47:14:40, 5.70s/it]
32%|███▏ | 14122/43939 [22:47:43<47:14:25, 5.70s/it]
32%|███▏ | 14123/43939 [22:47:49<47:14:38, 5.70s/it]
32%|███▏ | 14124/43939 [22:47:55<47:16:21, 5.71s/it]
32%|███▏ | 14125/43939 [22:48:01<47:30:57, 5.74s/it]
32%|███▏ | 14126/43939 [22:48:06<47:22:35, 5.72s/it]
32%|███▏ | 14127/43939 [22:48:12<47:50:43, 5.78s/it]
32%|███▏ | 14128/43939 [22:48:18<47:51:19, 5.78s/it]
32%|███▏ | 14129/43939 [22:48:24<47:39:06, 5.75s/it]
32%|███▏ | 14130/43939 [22:48:29<47:37:14, 5.75s/it]
32%|███▏ | 14131/43939 [22:48:35<47:36:57, 5.75s/it]
32%|███▏ | 14132/43939 [22:48:41<47:24:37, 5.73s/it]
32%|███▏ | 14133/43939 [22:48:47<47:26:36, 5.73s/it]
32%|███▏ | 14134/43939 [22:48:52<47:25:36, 5.73s/it]
32%|███▏ | 14135/43939 [22:48:58<47:20:28, 5.72s/it]
32%|███▏ | 14136/43939 [22:49:04<47:26:26, 5.73s/it]
32%|███▏ | 14137/43939 [22:49:09<47:16:35, 5.71s/it]
32%|███▏ | 14138/43939 [22:49:15<47:11:43, 5.70s/it]
32%|███▏ | 14139/43939 [22:49:21<47:20:24, 5.72s/it]
32%|███▏ | 14140/43939 [22:49:27<47:16:25, 5.71s/it]
32%|███▏ | 14141/43939 [22:49:32<47:13:50, 5.71s/it]
32%|███▏ | 14142/43939 [22:49:38<47:15:13, 5.71s/it]
32%|███▏ | 14143/43939 [22:49:44<47:16:00, 5.71s/it]
32%|███▏ | 14144/43939 [22:49:49<47:14:44, 5.71s/it]
32%|███▏ | 14145/43939 [22:49:55<47:08:12, 5.70s/it]
32%|███▏ | 14146/43939 [22:50:01<47:07:10, 5.69s/it]
32%|███▏ | 14147/43939 [22:50:07<47:13:24, 5.71s/it]
32%|███▏ | 14148/43939 [22:50:12<47:53:26, 5.79s/it]
32%|███▏ | 14149/43939 [22:50:18<47:36:44, 5.75s/it]
32%|███▏ | 14150/43939 [22:50:24<47:31:33, 5.74s/it]
{'loss': 0.924, 'grad_norm': 1.12111496925354, 'learning_rate': 4.285998267707229e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.36, 'epoch': 1.29}
32%|███▏ | 14150/43939 [22:50:24<47:31:33, 5.74s/it]
32%|███▏ | 14151/43939 [22:50:30<47:24:51, 5.73s/it]
32%|███▏ | 14152/43939 [22:50:35<47:18:02, 5.72s/it]
32%|███▏ | 14153/43939 [22:50:41<47:18:43, 5.72s/it]
32%|███▏ | 14154/43939 [22:50:47<47:13:36, 5.71s/it]
32%|███▏ | 14155/43939 [22:50:52<47:17:06, 5.72s/it]
32%|███▏ | 14156/43939 [22:50:58<47:15:29, 5.71s/it]
32%|███▏ | 14157/43939 [22:51:04<47:09:59, 5.70s/it]
32%|███▏ | 14158/43939 [22:51:09<47:09:54, 5.70s/it]
32%|███▏ | 14159/43939 [22:51:15<47:11:15, 5.70s/it]
32%|███▏ | 14160/43939 [22:51:21<47:10:47, 5.70s/it]
32%|███▏ | 14161/43939 [22:51:27<47:18:35, 5.72s/it]
32%|███▏ | 14162/43939 [22:51:32<47:10:39, 5.70s/it]
32%|███▏ | 14163/43939 [22:51:38<47:13:20, 5.71s/it]
32%|███▏ | 14164/43939 [22:51:44<47:07:23, 5.70s/it]
32%|███▏ | 14165/43939 [22:51:49<47:02:36, 5.69s/it]
32%|███▏ | 14166/43939 [22:51:55<47:02:36, 5.69s/it]
32%|███▏ | 14167/43939 [22:52:01<47:07:06, 5.70s/it]
32%|███▏ | 14168/43939 [22:52:06<47:04:21, 5.69s/it]
32%|███▏ | 14169/43939 [22:52:12<47:41:20, 5.77s/it]
32%|███▏ | 14170/43939 [22:52:18<47:32:03, 5.75s/it]
32%|███▏ | 14171/43939 [22:52:24<47:18:54, 5.72s/it]
32%|███▏ | 14172/43939 [22:52:29<47:10:19, 5.70s/it]
32%|███▏ | 14173/43939 [22:52:35<47:14:37, 5.71s/it]
32%|███▏ | 14174/43939 [22:52:41<47:09:42, 5.70s/it]
32%|███▏ | 14175/43939 [22:52:47<47:16:32, 5.72s/it]
32%|███▏ | 14176/43939 [22:52:52<47:22:24, 5.73s/it]
32%|███▏ | 14177/43939 [22:52:58<47:16:54, 5.72s/it]
32%|███▏ | 14178/43939 [22:53:04<47:14:32, 5.71s/it]
32%|███▏ | 14179/43939 [22:53:09<47:12:57, 5.71s/it]
32%|███▏ | 14180/43939 [22:53:15<47:19:35, 5.73s/it]
32%|███▏ | 14181/43939 [22:53:21<47:13:41, 5.71s/it]
32%|███▏ | 14182/43939 [22:53:27<47:05:17, 5.70s/it]
32%|███▏ | 14183/43939 [22:53:32<47:11:31, 5.71s/it]
32%|███▏ | 14184/43939 [22:53:38<47:06:18, 5.70s/it]
32%|███▏ | 14185/43939 [22:53:44<47:17:06, 5.72s/it]
32%|███▏ | 14186/43939 [22:53:49<47:11:37, 5.71s/it]
32%|███▏ | 14187/43939 [22:53:55<47:20:19, 5.73s/it]
32%|███▏ | 14188/43939 [22:54:01<47:10:24, 5.71s/it]
32%|███▏ | 14189/43939 [22:54:07<47:02:48, 5.69s/it]
32%|███▏ | 14190/43939 [22:54:12<47:03:50, 5.70s/it]
32%|███▏ | 14191/43939 [22:54:18<47:47:23, 5.78s/it]
32%|███▏ | 14192/43939 [22:54:24<47:39:37, 5.77s/it]
32%|███▏ | 14193/43939 [22:54:30<47:29:59, 5.75s/it]
32%|███▏ | 14194/43939 [22:54:35<47:18:52, 5.73s/it]
32%|███▏ | 14195/43939 [22:54:41<47:08:19, 5.71s/it]
32%|███▏ | 14196/43939 [22:54:47<47:05:43, 5.70s/it]
32%|███▏ | 14197/43939 [22:54:52<47:07:26, 5.70s/it]
32%|███▏ | 14198/43939 [22:54:58<47:08:53, 5.71s/it]
32%|███▏ | 14199/43939 [22:55:04<47:03:52, 5.70s/it]
32%|███▏ | 14200/43939 [22:55:10<47:07:36, 5.70s/it]
{'loss': 0.9401, 'grad_norm': 1.1524324417114258, 'learning_rate': 4.2790356667163125e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.08, 'epoch': 1.29}
32%|███▏ | 14200/43939 [22:55:10<47:07:36, 5.70s/it]
32%|███▏ | 14201/43939 [22:55:15<47:09:13, 5.71s/it]
32%|███▏ | 14202/43939 [22:55:21<47:08:21, 5.71s/it]
32%|███▏ | 14203/43939 [22:55:27<47:05:05, 5.70s/it]
32%|███▏ | 14204/43939 [22:55:32<47:00:58, 5.69s/it]
32%|███▏ | 14205/43939 [22:55:38<46:58:14, 5.69s/it]
32%|███▏ | 14206/43939 [22:55:44<46:59:49, 5.69s/it]
32%|███▏ | 14207/43939 [22:55:49<47:06:21, 5.70s/it]
32%|███▏ | 14208/43939 [22:55:55<47:05:51, 5.70s/it]
32%|███▏ | 14209/43939 [22:56:01<47:10:02, 5.71s/it]
32%|███▏ | 14210/43939 [22:56:06<47:03:55, 5.70s/it]
32%|███▏ | 14211/43939 [22:56:12<47:04:17, 5.70s/it]
32%|███▏ | 14212/43939 [22:56:18<47:42:38, 5.78s/it]
32%|███▏ | 14213/43939 [22:56:24<47:34:54, 5.76s/it]
32%|███▏ | 14214/43939 [22:56:30<47:23:58, 5.74s/it]
32%|███▏ | 14215/43939 [22:56:35<47:18:04, 5.73s/it]
32%|███▏ | 14216/43939 [22:56:41<47:11:56, 5.72s/it]
32%|███▏ | 14217/43939 [22:56:47<47:08:16, 5.71s/it]
32%|███▏ | 14218/43939 [22:56:52<47:08:36, 5.71s/it]
32%|███▏ | 14219/43939 [22:56:58<47:13:39, 5.72s/it]
32%|███▏ | 14220/43939 [22:57:04<47:09:45, 5.71s/it]
32%|███▏ | 14221/43939 [22:57:09<47:02:34, 5.70s/it]
32%|███▏ | 14222/43939 [22:57:15<46:59:50, 5.69s/it]
32%|███▏ | 14223/43939 [22:57:21<47:00:14, 5.69s/it]
32%|███▏ | 14224/43939 [22:57:27<47:04:15, 5.70s/it]
32%|███▏ | 14225/43939 [22:57:32<47:02:54, 5.70s/it]
32%|███▏ | 14226/43939 [22:57:38<46:58:42, 5.69s/it]
32%|███▏ | 14227/43939 [22:57:44<46:58:13, 5.69s/it]
32%|███▏ | 14228/43939 [22:57:49<47:01:44, 5.70s/it]
32%|███▏ | 14229/43939 [22:57:55<46:59:21, 5.69s/it]
32%|███▏ | 14230/43939 [22:58:01<46:53:22, 5.68s/it]
32%|███▏ | 14231/43939 [22:58:06<47:00:02, 5.70s/it]
32%|███▏ | 14232/43939 [22:58:12<47:04:59, 5.71s/it]
32%|███▏ | 14233/43939 [22:58:18<47:35:17, 5.77s/it]
32%|███▏ | 14234/43939 [22:58:24<47:28:29, 5.75s/it]
32%|███▏ | 14235/43939 [22:58:29<47:19:47, 5.74s/it]
32%|███▏ | 14236/43939 [22:58:35<47:09:44, 5.72s/it]
32%|███▏ | 14237/43939 [22:58:41<47:03:43, 5.70s/it]
32%|███▏ | 14238/43939 [22:58:47<47:04:29, 5.71s/it]
32%|███▏ | 14239/43939 [22:58:52<47:00:41, 5.70s/it]
32%|███▏ | 14240/43939 [22:58:58<46:58:50, 5.69s/it]
32%|███▏ | 14241/43939 [22:59:04<47:04:24, 5.71s/it]
32%|███▏ | 14242/43939 [22:59:09<47:05:51, 5.71s/it]
32%|███▏ | 14243/43939 [22:59:15<47:03:42, 5.71s/it]
32%|███▏ | 14244/43939 [22:59:21<47:03:37, 5.71s/it]
32%|███▏ | 14245/43939 [22:59:26<47:06:23, 5.71s/it]
32%|███▏ | 14246/43939 [22:59:32<47:05:56, 5.71s/it]
32%|███▏ | 14247/43939 [22:59:38<47:07:33, 5.71s/it]
32%|███▏ | 14248/43939 [22:59:44<47:12:31, 5.72s/it]
32%|███▏ | 14249/43939 [22:59:49<47:14:53, 5.73s/it]
32%|███▏ | 14250/43939 [22:59:55<47:09:40, 5.72s/it]
{'loss': 0.9413, 'grad_norm': 1.3202351331710815, 'learning_rate': 4.272044997257077e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.27, 'epoch': 1.3}
32%|███▏ | 14250/43939 [22:59:55<47:09:40, 5.72s/it]
32%|███▏ | 14251/43939 [23:00:01<47:10:07, 5.72s/it]
32%|███▏ | 14252/43939 [23:00:07<47:10:22, 5.72s/it]
32%|███▏ | 14253/43939 [23:00:12<47:07:03, 5.71s/it]
32%|███▏ | 14254/43939 [23:00:18<47:37:27, 5.78s/it]
32%|███▏ | 14255/43939 [23:00:24<47:22:25, 5.75s/it]
32%|███▏ | 14256/43939 [23:00:29<47:11:57, 5.72s/it]
32%|███▏ | 14257/43939 [23:00:35<47:13:19, 5.73s/it]
32%|███▏ | 14258/43939 [23:00:41<47:10:37, 5.72s/it]
32%|███▏ | 14259/43939 [23:00:47<47:07:04, 5.72s/it]
32%|███▏ | 14260/43939 [23:00:52<47:07:56, 5.72s/it]
32%|███▏ | 14261/43939 [23:00:58<47:05:02, 5.71s/it]
32%|███▏ | 14262/43939 [23:01:04<47:04:24, 5.71s/it]
32%|███▏ | 14263/43939 [23:01:09<47:03:48, 5.71s/it]
32%|███▏ | 14264/43939 [23:01:15<47:00:26, 5.70s/it]
32%|███▏ | 14265/43939 [23:01:21<47:02:08, 5.71s/it]
32%|███▏ | 14266/43939 [23:01:27<47:00:26, 5.70s/it]
32%|███▏ | 14267/43939 [23:01:32<47:00:12, 5.70s/it]
32%|███▏ | 14268/43939 [23:01:38<47:03:04, 5.71s/it]
32%|███▏ | 14269/43939 [23:01:44<47:03:11, 5.71s/it]
32%|███▏ | 14270/43939 [23:01:49<46:54:32, 5.69s/it]
32%|███▏ | 14271/43939 [23:01:55<46:55:14, 5.69s/it]
32%|███▏ | 14272/43939 [23:02:01<47:04:38, 5.71s/it]
32%|███▏ | 14273/43939 [23:02:07<47:05:30, 5.71s/it]
32%|███▏ | 14274/43939 [23:02:12<47:03:12, 5.71s/it]
32%|███▏ | 14275/43939 [23:02:18<47:31:34, 5.77s/it]
32%|███▏ | 14276/43939 [23:02:24<47:16:00, 5.74s/it]
32%|███▏ | 14277/43939 [23:02:29<47:08:22, 5.72s/it]
32%|███▏ | 14278/43939 [23:02:35<47:02:39, 5.71s/it]
32%|███▏ | 14279/43939 [23:02:41<47:01:22, 5.71s/it]
32%|███▏ | 14280/43939 [23:02:47<46:57:00, 5.70s/it]
33%|███▎ | 14281/43939 [23:02:52<46:58:32, 5.70s/it]
33%|███▎ | 14282/43939 [23:02:58<46:58:47, 5.70s/it]
33%|███▎ | 14283/43939 [23:03:04<46:57:03, 5.70s/it]
33%|███▎ | 14284/43939 [23:03:09<46:52:07, 5.69s/it]
33%|███▎ | 14285/43939 [23:03:15<46:52:45, 5.69s/it]
33%|███▎ | 14286/43939 [23:03:21<46:49:40, 5.69s/it]
33%|███▎ | 14287/43939 [23:03:26<46:55:26, 5.70s/it]
33%|███▎ | 14288/43939 [23:03:32<47:03:15, 5.71s/it]
33%|███▎ | 14289/43939 [23:03:38<47:01:09, 5.71s/it]
33%|███▎ | 14290/43939 [23:03:44<47:06:19, 5.72s/it]
33%|███▎ | 14291/43939 [23:03:49<47:02:51, 5.71s/it]
33%|███▎ | 14292/43939 [23:03:55<46:59:45, 5.71s/it]
33%|███▎ | 14293/43939 [23:04:01<47:03:48, 5.72s/it]
33%|███▎ | 14294/43939 [23:04:06<47:01:29, 5.71s/it]
33%|███▎ | 14295/43939 [23:04:12<46:58:44, 5.71s/it]
33%|███▎ | 14296/43939 [23:04:18<47:37:58, 5.78s/it]
33%|███▎ | 14297/43939 [23:04:24<47:22:18, 5.75s/it]
33%|███▎ | 14298/43939 [23:04:29<47:15:55, 5.74s/it]
33%|███▎ | 14299/43939 [23:04:35<47:10:44, 5.73s/it]
33%|███▎ | 14300/43939 [23:04:41<47:07:22, 5.72s/it]
{'loss': 0.908, 'grad_norm': 1.1498830318450928, 'learning_rate': 4.265026369623761e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.5, 'epoch': 1.3}
33%|███▎ | 14300/43939 [23:04:41<47:07:22, 5.72s/it]
33%|███▎ | 14301/43939 [23:04:47<47:03:01, 5.72s/it]
33%|███▎ | 14302/43939 [23:04:52<47:06:49, 5.72s/it]
33%|███▎ | 14303/43939 [23:04:58<47:06:44, 5.72s/it]
33%|███▎ | 14304/43939 [23:05:04<47:05:18, 5.72s/it]
33%|███▎ | 14305/43939 [23:05:09<47:03:20, 5.72s/it]
33%|███▎ | 14306/43939 [23:05:15<47:05:32, 5.72s/it]
33%|███▎ | 14307/43939 [23:05:21<47:04:26, 5.72s/it]
33%|███▎ | 14308/43939 [23:05:27<46:58:25, 5.71s/it]
33%|███▎ | 14309/43939 [23:05:32<46:56:08, 5.70s/it]
33%|███▎ | 14310/43939 [23:05:38<46:55:54, 5.70s/it]
33%|███▎ | 14311/43939 [23:05:44<46:51:56, 5.69s/it]
33%|███▎ | 14312/43939 [23:05:49<46:49:22, 5.69s/it]
33%|███▎ | 14313/43939 [23:05:55<46:50:33, 5.69s/it]
33%|███▎ | 14314/43939 [23:06:01<46:50:40, 5.69s/it]
33%|███▎ | 14315/43939 [23:06:06<46:51:41, 5.69s/it]
33%|███▎ | 14316/43939 [23:06:12<47:41:47, 5.80s/it]
33%|███▎ | 14317/43939 [23:06:18<47:25:21, 5.76s/it]
33%|███▎ | 14318/43939 [23:06:24<47:16:01, 5.74s/it]
33%|███▎ | 14319/43939 [23:06:30<47:11:14, 5.74s/it]
33%|███▎ | 14320/43939 [23:06:35<47:08:59, 5.73s/it]
33%|███▎ | 14321/43939 [23:06:41<47:08:28, 5.73s/it]
33%|███▎ | 14322/43939 [23:06:47<47:04:35, 5.72s/it]
33%|███▎ | 14323/43939 [23:06:52<46:58:55, 5.71s/it]
33%|███▎ | 14324/43939 [23:06:58<47:01:19, 5.72s/it]
33%|███▎ | 14325/43939 [23:07:04<47:08:45, 5.73s/it]
33%|███▎ | 14326/43939 [23:07:10<47:03:26, 5.72s/it]
33%|███▎ | 14327/43939 [23:07:15<47:05:26, 5.72s/it]
33%|███▎ | 14328/43939 [23:07:21<47:04:11, 5.72s/it]
33%|███▎ | 14329/43939 [23:07:27<47:01:49, 5.72s/it]
33%|███▎ | 14330/43939 [23:07:32<46:57:48, 5.71s/it]
33%|███▎ | 14331/43939 [23:07:38<46:55:19, 5.71s/it]
33%|███▎ | 14332/43939 [23:07:44<46:55:59, 5.71s/it]
33%|███▎ | 14333/43939 [23:07:50<46:59:38, 5.71s/it]
33%|███▎ | 14334/43939 [23:07:55<46:57:07, 5.71s/it]
33%|███▎ | 14335/43939 [23:08:01<47:02:04, 5.72s/it]
33%|███▎ | 14336/43939 [23:08:07<47:04:38, 5.73s/it]
33%|███▎ | 14337/43939 [23:08:12<46:58:05, 5.71s/it]
33%|███▎ | 14338/43939 [23:08:18<47:28:39, 5.77s/it]
33%|███▎ | 14339/43939 [23:08:24<47:21:16, 5.76s/it]
33%|███▎ | 14340/43939 [23:08:30<47:11:03, 5.74s/it]
33%|███▎ | 14341/43939 [23:08:35<47:01:52, 5.72s/it]
33%|███▎ | 14342/43939 [23:08:41<47:00:36, 5.72s/it]
33%|███▎ | 14343/43939 [23:08:47<47:02:11, 5.72s/it]
33%|███▎ | 14344/43939 [23:08:53<46:59:38, 5.72s/it]
33%|███▎ | 14345/43939 [23:08:58<46:56:11, 5.71s/it]
33%|███▎ | 14346/43939 [23:09:04<46:51:22, 5.70s/it]
33%|███▎ | 14347/43939 [23:09:10<46:52:22, 5.70s/it]
33%|███▎ | 14348/43939 [23:09:15<46:53:21, 5.70s/it]
33%|███▎ | 14349/43939 [23:09:21<46:53:16, 5.70s/it]
33%|███▎ | 14350/43939 [23:09:27<46:55:03, 5.71s/it]
{'loss': 0.9448, 'grad_norm': 1.057232141494751, 'learning_rate': 4.257979894551709e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.65, 'epoch': 1.31}
33%|███▎ | 14350/43939 [23:09:27<46:55:03, 5.71s/it]
33%|███▎ | 14351/43939 [23:09:33<47:00:28, 5.72s/it]
33%|███▎ | 14352/43939 [23:09:38<47:02:03, 5.72s/it]
33%|███▎ | 14353/43939 [23:09:44<47:01:31, 5.72s/it]
33%|███▎ | 14354/43939 [23:09:50<46:56:51, 5.71s/it]
33%|███▎ | 14355/43939 [23:09:55<46:54:01, 5.71s/it]
33%|███▎ | 14356/43939 [23:10:01<46:51:15, 5.70s/it]
33%|███▎ | 14357/43939 [23:10:07<46:52:36, 5.70s/it]
33%|███▎ | 14358/43939 [23:10:13<46:57:15, 5.71s/it]
33%|███▎ | 14359/43939 [23:10:18<47:31:05, 5.78s/it]
33%|███▎ | 14360/43939 [23:10:24<47:19:31, 5.76s/it]
33%|███▎ | 14361/43939 [23:10:30<47:06:38, 5.73s/it]
33%|███▎ | 14362/43939 [23:10:36<47:01:52, 5.72s/it]
33%|███▎ | 14363/43939 [23:10:41<47:08:19, 5.74s/it]
33%|███▎ | 14364/43939 [23:10:47<47:02:13, 5.73s/it]
33%|███▎ | 14365/43939 [23:10:53<46:54:37, 5.71s/it]
33%|███▎ | 14366/43939 [23:10:58<46:47:00, 5.70s/it]
33%|███▎ | 14367/43939 [23:11:04<46:42:16, 5.69s/it]
33%|███▎ | 14368/43939 [23:11:10<46:46:52, 5.70s/it]
33%|███▎ | 14369/43939 [23:11:15<46:44:15, 5.69s/it]
33%|███▎ | 14370/43939 [23:11:21<46:44:52, 5.69s/it]
33%|███▎ | 14371/43939 [23:11:27<46:45:58, 5.69s/it]
33%|███▎ | 14372/43939 [23:11:33<46:47:57, 5.70s/it]
33%|███▎ | 14373/43939 [23:11:38<46:41:22, 5.68s/it]
33%|███▎ | 14374/43939 [23:11:44<46:50:14, 5.70s/it]
33%|███▎ | 14375/43939 [23:11:50<46:51:40, 5.71s/it]
33%|███▎ | 14376/43939 [23:11:55<46:51:05, 5.71s/it]
33%|███▎ | 14377/43939 [23:12:01<46:48:09, 5.70s/it]
33%|███▎ | 14378/43939 [23:12:07<46:49:09, 5.70s/it]
33%|███▎ | 14379/43939 [23:12:12<46:43:24, 5.69s/it]
33%|███▎ | 14380/43939 [23:12:18<47:09:04, 5.74s/it]
33%|███▎ | 14381/43939 [23:12:24<47:02:20, 5.73s/it]
33%|███▎ | 14382/43939 [23:12:30<46:57:19, 5.72s/it]
33%|███▎ | 14383/43939 [23:12:35<47:06:08, 5.74s/it]
33%|███▎ | 14384/43939 [23:12:41<46:55:40, 5.72s/it]
33%|███▎ | 14385/43939 [23:12:47<47:07:58, 5.74s/it]
33%|███▎ | 14386/43939 [23:12:53<47:02:54, 5.73s/it]
33%|███▎ | 14387/43939 [23:12:58<46:59:16, 5.72s/it]
33%|███▎ | 14388/43939 [23:13:04<47:00:36, 5.73s/it]
33%|███▎ | 14389/43939 [23:13:10<47:01:18, 5.73s/it]
33%|███▎ | 14390/43939 [23:13:15<46:54:15, 5.71s/it]
33%|███▎ | 14391/43939 [23:13:21<46:51:19, 5.71s/it]
33%|███▎ | 14392/43939 [23:13:27<46:50:40, 5.71s/it]
33%|███▎ | 14393/43939 [23:13:33<46:50:11, 5.71s/it]
33%|███▎ | 14394/43939 [23:13:38<46:46:41, 5.70s/it]
33%|███▎ | 14395/43939 [23:13:44<46:41:04, 5.69s/it]
33%|███▎ | 14396/43939 [23:13:50<46:50:48, 5.71s/it]
33%|███▎ | 14397/43939 [23:13:55<46:46:58, 5.70s/it]
33%|███▎ | 14398/43939 [23:14:01<46:49:23, 5.71s/it]
33%|███▎ | 14399/43939 [23:14:07<46:45:38, 5.70s/it]
33%|███▎ | 14400/43939 [23:14:12<46:48:04, 5.70s/it]
{'loss': 0.9111, 'grad_norm': 1.2352851629257202, 'learning_rate': 4.2509056832156246e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.8, 'epoch': 1.31}
33%|███▎ | 14400/43939 [23:14:12<46:48:04, 5.70s/it]
33%|███▎ | 14401/43939 [23:14:18<46:49:14, 5.71s/it]
33%|███▎ | 14402/43939 [23:14:24<47:21:56, 5.77s/it]
33%|███▎ | 14403/43939 [23:14:30<47:12:58, 5.75s/it]
33%|███▎ | 14404/43939 [23:14:36<47:02:38, 5.73s/it]
33%|███▎ | 14405/43939 [23:14:41<47:05:46, 5.74s/it]
33%|███▎ | 14406/43939 [23:14:47<47:01:36, 5.73s/it]
33%|███▎ | 14407/43939 [23:14:53<46:59:22, 5.73s/it]
33%|███▎ | 14408/43939 [23:14:58<46:50:39, 5.71s/it]
33%|███▎ | 14409/43939 [23:15:04<46:53:09, 5.72s/it]
33%|███▎ | 14410/43939 [23:15:10<46:52:06, 5.71s/it]
33%|███▎ | 14411/43939 [23:15:16<46:56:31, 5.72s/it]
33%|███▎ | 14412/43939 [23:15:21<46:52:19, 5.71s/it]
33%|███▎ | 14413/43939 [23:15:27<46:42:59, 5.70s/it]
33%|███▎ | 14414/43939 [23:15:33<46:44:40, 5.70s/it]
33%|███▎ | 14415/43939 [23:15:38<46:44:35, 5.70s/it]
33%|███▎ | 14416/43939 [23:15:44<46:45:43, 5.70s/it]
33%|███▎ | 14417/43939 [23:15:50<46:47:19, 5.71s/it]
33%|███▎ | 14418/43939 [23:15:55<46:46:13, 5.70s/it]
33%|███▎ | 14419/43939 [23:16:01<46:45:52, 5.70s/it]
33%|███▎ | 14420/43939 [23:16:07<46:42:13, 5.70s/it]
33%|███▎ | 14421/43939 [23:16:13<46:43:01, 5.70s/it]
33%|███▎ | 14422/43939 [23:16:18<46:40:36, 5.69s/it]
33%|███▎ | 14423/43939 [23:16:24<47:08:15, 5.75s/it]
33%|███▎ | 14424/43939 [23:16:30<47:00:37, 5.73s/it]
33%|███▎ | 14425/43939 [23:16:35<46:54:51, 5.72s/it]
33%|███▎ | 14426/43939 [23:16:41<46:50:20, 5.71s/it]
33%|███▎ | 14427/43939 [23:16:47<46:49:43, 5.71s/it]
33%|███▎ | 14428/43939 [23:16:53<46:40:48, 5.69s/it]
33%|███▎ | 14429/43939 [23:16:58<46:44:35, 5.70s/it]
33%|███▎ | 14430/43939 [23:17:04<46:42:15, 5.70s/it]
33%|███▎ | 14431/43939 [23:17:10<46:40:49, 5.70s/it]
33%|███▎ | 14432/43939 [23:17:15<46:40:07, 5.69s/it]
33%|███▎ | 14433/43939 [23:17:21<46:38:04, 5.69s/it]
33%|███▎ | 14434/43939 [23:17:27<46:40:16, 5.69s/it]
33%|███▎ | 14435/43939 [23:17:32<46:38:24, 5.69s/it]
33%|███▎ | 14436/43939 [23:17:38<46:39:02, 5.69s/it]
33%|███▎ | 14437/43939 [23:17:44<46:44:47, 5.70s/it]
33%|███▎ | 14438/43939 [23:17:50<46:45:33, 5.71s/it]
33%|███▎ | 14439/43939 [23:17:55<46:44:30, 5.70s/it]
33%|███▎ | 14440/43939 [23:18:01<46:38:34, 5.69s/it]
33%|███▎ | 14441/43939 [23:18:07<46:38:33, 5.69s/it]
33%|███▎ | 14442/43939 [23:18:12<46:40:45, 5.70s/it]
33%|███▎ | 14443/43939 [23:18:18<46:43:32, 5.70s/it]
33%|███▎ | 14444/43939 [23:18:24<47:13:35, 5.76s/it]
33%|███▎ | 14445/43939 [23:18:30<46:57:45, 5.73s/it]
33%|███▎ | 14446/43939 [23:18:35<46:52:49, 5.72s/it]
33%|███▎ | 14447/43939 [23:18:41<46:51:35, 5.72s/it]
33%|███▎ | 14448/43939 [23:18:47<46:53:00, 5.72s/it]
33%|███▎ | 14449/43939 [23:18:52<46:52:47, 5.72s/it]
33%|███▎ | 14450/43939 [23:18:58<46:53:16, 5.72s/it]
{'loss': 0.9255, 'grad_norm': 1.4061874151229858, 'learning_rate': 4.2438038472278154e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.92, 'epoch': 1.32}
33%|███▎ | 14450/43939 [23:18:58<46:53:16, 5.72s/it]
33%|███▎ | 14451/43939 [23:19:04<46:52:53, 5.72s/it]
33%|███▎ | 14452/43939 [23:19:10<46:55:18, 5.73s/it]
33%|███▎ | 14453/43939 [23:19:15<46:57:50, 5.73s/it]
33%|███▎ | 14454/43939 [23:19:21<46:49:30, 5.72s/it]
33%|███▎ | 14455/43939 [23:19:27<46:47:53, 5.71s/it]
33%|███▎ | 14456/43939 [23:19:32<46:42:20, 5.70s/it]
33%|███▎ | 14457/43939 [23:19:38<46:39:17, 5.70s/it]
33%|███▎ | 14458/43939 [23:19:44<46:33:53, 5.69s/it]
33%|███▎ | 14459/43939 [23:19:49<46:37:09, 5.69s/it]
33%|███▎ | 14460/43939 [23:19:55<46:36:04, 5.69s/it]
33%|███▎ | 14461/43939 [23:20:01<46:41:24, 5.70s/it]
33%|███▎ | 14462/43939 [23:20:07<46:45:24, 5.71s/it]
33%|███▎ | 14463/43939 [23:20:12<46:40:23, 5.70s/it]
33%|███▎ | 14464/43939 [23:20:18<46:44:59, 5.71s/it]
33%|███▎ | 14465/43939 [23:20:24<47:13:31, 5.77s/it]
33%|███▎ | 14466/43939 [23:20:30<47:01:38, 5.74s/it]
33%|███▎ | 14467/43939 [23:20:35<46:50:02, 5.72s/it]
33%|███▎ | 14468/43939 [23:20:41<46:56:59, 5.74s/it]
33%|███▎ | 14469/43939 [23:20:47<46:49:20, 5.72s/it]
33%|███▎ | 14470/43939 [23:20:52<46:46:15, 5.71s/it]
33%|███▎ | 14471/43939 [23:20:58<46:42:00, 5.71s/it]
33%|███▎ | 14472/43939 [23:21:04<46:45:38, 5.71s/it]
33%|███▎ | 14473/43939 [23:21:10<46:40:24, 5.70s/it]
33%|███▎ | 14474/43939 [23:21:15<46:38:59, 5.70s/it]
33%|███▎ | 14475/43939 [23:21:21<46:37:23, 5.70s/it]
33%|███▎ | 14476/43939 [23:21:27<46:37:36, 5.70s/it]
33%|███▎ | 14477/43939 [23:21:32<46:39:38, 5.70s/it]
33%|███▎ | 14478/43939 [23:21:38<46:41:12, 5.70s/it]
33%|███▎ | 14479/43939 [23:21:44<46:40:08, 5.70s/it]
33%|███▎ | 14480/43939 [23:21:49<46:43:05, 5.71s/it]
33%|███▎ | 14481/43939 [23:21:55<46:37:28, 5.70s/it]
33%|███▎ | 14482/43939 [23:22:01<46:35:45, 5.69s/it]
33%|███▎ | 14483/43939 [23:22:07<46:37:23, 5.70s/it]
33%|███▎ | 14484/43939 [23:22:12<46:39:46, 5.70s/it]
33%|███▎ | 14485/43939 [23:22:18<46:32:57, 5.69s/it]
33%|███▎ | 14486/43939 [23:22:24<47:02:21, 5.75s/it]
33%|███▎ | 14487/43939 [23:22:29<46:55:18, 5.74s/it]
33%|███▎ | 14488/43939 [23:22:35<46:52:22, 5.73s/it]
33%|███▎ | 14489/43939 [23:22:41<46:45:41, 5.72s/it]
33%|███▎ | 14490/43939 [23:22:47<46:41:17, 5.71s/it]
33%|███▎ | 14491/43939 [23:22:52<46:38:19, 5.70s/it]
33%|███▎ | 14492/43939 [23:22:58<46:37:00, 5.70s/it]
33%|███▎ | 14493/43939 [23:23:04<46:36:21, 5.70s/it]
33%|███▎ | 14494/43939 [23:23:09<46:40:47, 5.71s/it]
33%|███▎ | 14495/43939 [23:23:15<46:47:28, 5.72s/it]
33%|███▎ | 14496/43939 [23:23:21<46:46:41, 5.72s/it]
33%|███▎ | 14497/43939 [23:23:27<46:41:01, 5.71s/it]
33%|███▎ | 14498/43939 [23:23:32<46:43:37, 5.71s/it]
33%|███▎ | 14499/43939 [23:23:38<46:36:29, 5.70s/it]
33%|███▎ | 14500/43939 [23:23:44<46:38:58, 5.70s/it]
{'loss': 0.9517, 'grad_norm': 1.0230389833450317, 'learning_rate': 4.236674498636434e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.74, 'epoch': 1.32}
33%|███▎ | 14500/43939 [23:23:44<46:38:58, 5.70s/it]
33%|███▎ | 14501/43939 [23:23:49<46:35:43, 5.70s/it]
33%|███▎ | 14502/43939 [23:23:55<46:39:37, 5.71s/it]
33%|███▎ | 14503/43939 [23:24:01<46:41:41, 5.71s/it]
33%|███▎ | 14504/43939 [23:24:06<46:37:09, 5.70s/it]
33%|███▎ | 14505/43939 [23:24:12<46:36:51, 5.70s/it]
33%|███▎ | 14506/43939 [23:24:18<46:41:10, 5.71s/it]
33%|███▎ | 14507/43939 [23:24:24<47:17:24, 5.78s/it]
33%|███▎ | 14508/43939 [23:24:30<47:05:40, 5.76s/it]
33%|███▎ | 14509/43939 [23:24:35<46:57:41, 5.74s/it]
33%|███▎ | 14510/43939 [23:24:41<46:46:59, 5.72s/it]
33%|███▎ | 14511/43939 [23:24:47<46:45:49, 5.72s/it]
33%|███▎ | 14512/43939 [23:24:52<46:42:09, 5.71s/it]
33%|███▎ | 14513/43939 [23:24:58<46:40:13, 5.71s/it]
33%|███▎ | 14514/43939 [23:25:04<46:46:06, 5.72s/it]
33%|███▎ | 14515/43939 [23:25:09<46:44:04, 5.72s/it]
33%|███▎ | 14516/43939 [23:25:15<46:38:30, 5.71s/it]
33%|███▎ | 14517/43939 [23:25:21<46:41:22, 5.71s/it]
33%|███▎ | 14518/43939 [23:25:27<46:36:27, 5.70s/it]
33%|███▎ | 14519/43939 [23:25:32<46:42:54, 5.72s/it]
33%|███▎ | 14520/43939 [23:25:38<46:35:53, 5.70s/it]
33%|███▎ | 14521/43939 [23:25:44<46:32:53, 5.70s/it]
33%|███▎ | 14522/43939 [23:25:49<46:44:01, 5.72s/it]
33%|███▎ | 14523/43939 [23:25:55<46:40:05, 5.71s/it]
33%|███▎ | 14524/43939 [23:26:01<46:37:46, 5.71s/it]
33%|███▎ | 14525/43939 [23:26:07<46:44:16, 5.72s/it]
33%|███▎ | 14526/43939 [23:26:12<46:36:48, 5.71s/it]
33%|███▎ | 14527/43939 [23:26:18<46:43:07, 5.72s/it]
33%|███▎ | 14528/43939 [23:26:24<47:13:40, 5.78s/it]
33%|███▎ | 14529/43939 [23:26:30<47:01:30, 5.76s/it]
33%|███▎ | 14530/43939 [23:26:35<46:55:59, 5.75s/it]
33%|███▎ | 14531/43939 [23:26:41<46:43:40, 5.72s/it]
33%|███▎ | 14532/43939 [23:26:47<46:43:17, 5.72s/it]
33%|███▎ | 14533/43939 [23:26:52<46:36:26, 5.71s/it]
33%|███▎ | 14534/43939 [23:26:58<46:41:08, 5.72s/it]
33%|███▎ | 14535/43939 [23:27:04<46:35:14, 5.70s/it]
33%|███▎ | 14536/43939 [23:27:10<46:38:41, 5.71s/it]
33%|███▎ | 14537/43939 [23:27:15<46:41:33, 5.72s/it]
33%|███▎ | 14538/43939 [23:27:21<46:42:30, 5.72s/it]
33%|███▎ | 14539/43939 [23:27:27<46:42:43, 5.72s/it]
33%|███▎ | 14540/43939 [23:27:32<46:35:16, 5.70s/it]
33%|███▎ | 14541/43939 [23:27:38<46:42:38, 5.72s/it]
33%|███▎ | 14542/43939 [23:27:44<46:47:02, 5.73s/it]
33%|███▎ | 14543/43939 [23:27:50<46:48:01, 5.73s/it]
33%|███▎ | 14544/43939 [23:27:55<46:44:13, 5.72s/it]
33%|███▎ | 14545/43939 [23:28:01<46:40:43, 5.72s/it]
33%|███▎ | 14546/43939 [23:28:07<46:45:46, 5.73s/it]
33%|███▎ | 14547/43939 [23:28:12<46:39:53, 5.72s/it]
33%|███▎ | 14548/43939 [23:28:18<46:38:11, 5.71s/it]
33%|███▎ | 14549/43939 [23:28:24<46:37:18, 5.71s/it]
33%|███▎ | 14550/43939 [23:28:30<47:10:47, 5.78s/it]
{'loss': 0.9185, 'grad_norm': 1.123724102973938, 'learning_rate': 4.229517749923708e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.22, 'epoch': 1.32}
33%|███▎ | 14550/43939 [23:28:30<47:10:47, 5.78s/it]
33%|███▎ | 14551/43939 [23:28:36<46:56:52, 5.75s/it]
33%|███▎ | 14552/43939 [23:28:41<46:56:35, 5.75s/it]
33%|███▎ | 14553/43939 [23:28:47<46:48:40, 5.73s/it]
33%|███▎ | 14554/43939 [23:28:53<46:38:36, 5.71s/it]
33%|███▎ | 14555/43939 [23:28:58<46:31:50, 5.70s/it]
33%|███▎ | 14556/43939 [23:29:04<46:29:11, 5.70s/it]
33%|███▎ | 14557/43939 [23:29:10<46:31:54, 5.70s/it]
33%|███▎ | 14558/43939 [23:29:15<46:32:08, 5.70s/it]
33%|███▎ | 14559/43939 [23:29:21<46:27:02, 5.69s/it]
33%|███▎ | 14560/43939 [23:29:27<46:40:07, 5.72s/it]
33%|███▎ | 14561/43939 [23:29:33<46:44:24, 5.73s/it]
33%|███▎ | 14562/43939 [23:29:38<46:34:06, 5.71s/it]
33%|███▎ | 14563/43939 [23:29:44<46:34:14, 5.71s/it]
33%|███▎ | 14564/43939 [23:29:50<46:35:38, 5.71s/it]
33%|███▎ | 14565/43939 [23:29:55<46:34:23, 5.71s/it]
33%|███▎ | 14566/43939 [23:30:01<46:32:48, 5.70s/it]
33%|███▎ | 14567/43939 [23:30:07<46:31:33, 5.70s/it]
33%|███▎ | 14568/43939 [23:30:12<46:30:16, 5.70s/it]
33%|███▎ | 14569/43939 [23:30:18<46:29:51, 5.70s/it]
33%|███▎ | 14570/43939 [23:30:24<46:34:30, 5.71s/it]
33%|███▎ | 14571/43939 [23:30:30<47:06:16, 5.77s/it]
33%|███▎ | 14572/43939 [23:30:36<47:00:52, 5.76s/it]
33%|███▎ | 14573/43939 [23:30:41<46:55:23, 5.75s/it]
33%|███▎ | 14574/43939 [23:30:47<46:46:11, 5.73s/it]
33%|███▎ | 14575/43939 [23:30:53<46:44:12, 5.73s/it]
33%|███▎ | 14576/43939 [23:30:58<46:43:47, 5.73s/it]
33%|███▎ | 14577/43939 [23:31:04<46:39:20, 5.72s/it]
33%|███▎ | 14578/43939 [23:31:10<46:36:33, 5.71s/it]
33%|███▎ | 14579/43939 [23:31:16<46:39:43, 5.72s/it]
33%|███▎ | 14580/43939 [23:31:21<46:34:02, 5.71s/it]
33%|███▎ | 14581/43939 [23:31:27<46:25:14, 5.69s/it]
33%|███▎ | 14582/43939 [23:31:33<46:22:39, 5.69s/it]
33%|███▎ | 14583/43939 [23:31:38<46:21:46, 5.69s/it]
33%|███▎ | 14584/43939 [23:31:44<46:22:18, 5.69s/it]
33%|███▎ | 14585/43939 [23:31:50<46:21:29, 5.69s/it]
33%|███▎ | 14586/43939 [23:31:55<46:25:13, 5.69s/it]
33%|███▎ | 14587/43939 [23:32:01<46:25:55, 5.69s/it]
33%|███▎ | 14588/43939 [23:32:07<46:24:37, 5.69s/it]
33%|███▎ | 14589/43939 [23:32:12<46:26:43, 5.70s/it]
33%|███▎ | 14590/43939 [23:32:18<46:26:58, 5.70s/it]
33%|███▎ | 14591/43939 [23:32:24<46:34:15, 5.71s/it]
33%|███▎ | 14592/43939 [23:32:30<47:09:41, 5.79s/it]
33%|███▎ | 14593/43939 [23:32:36<46:54:00, 5.75s/it]
33%|███▎ | 14594/43939 [23:32:41<46:50:54, 5.75s/it]
33%|███▎ | 14595/43939 [23:32:47<46:42:34, 5.73s/it]
33%|███▎ | 14596/43939 [23:32:53<46:46:28, 5.74s/it]
33%|███▎ | 14597/43939 [23:32:58<46:40:39, 5.73s/it]
33%|███▎ | 14598/43939 [23:33:04<46:34:37, 5.71s/it]
33%|███▎ | 14599/43939 [23:33:10<46:28:40, 5.70s/it]
33%|███▎ | 14600/43939 [23:33:15<46:26:37, 5.70s/it]
{'loss': 0.9355, 'grad_norm': 1.2315624952316284, 'learning_rate': 4.2223337140041645e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.01, 'epoch': 1.33}
33%|███▎ | 14600/43939 [23:33:15<46:26:37, 5.70s/it]
33%|███▎ | 14601/43939 [23:33:21<46:24:30, 5.69s/it]
33%|███▎ | 14602/43939 [23:33:27<46:26:48, 5.70s/it]
33%|███▎ | 14603/43939 [23:33:33<46:30:11, 5.71s/it]
33%|███▎ | 14604/43939 [23:33:38<46:30:17, 5.71s/it]
33%|███▎ | 14605/43939 [23:33:44<46:32:03, 5.71s/it]
33%|███▎ | 14606/43939 [23:33:50<46:29:19, 5.71s/it]
33%|███▎ | 14607/43939 [23:33:55<46:27:25, 5.70s/it]
33%|███▎ | 14608/43939 [23:34:01<46:33:44, 5.71s/it]
33%|███▎ | 14609/43939 [23:34:07<46:26:09, 5.70s/it]
33%|███▎ | 14610/43939 [23:34:13<46:34:37, 5.72s/it]
33%|███▎ | 14611/43939 [23:34:18<46:32:09, 5.71s/it]
33%|███▎ | 14612/43939 [23:34:24<46:26:44, 5.70s/it]
33%|███▎ | 14613/43939 [23:34:30<46:57:14, 5.76s/it]
33%|███▎ | 14614/43939 [23:34:36<46:48:35, 5.75s/it]
33%|███▎ | 14615/43939 [23:34:41<46:42:51, 5.73s/it]
33%|███▎ | 14616/43939 [23:34:47<46:36:44, 5.72s/it]
33%|███▎ | 14617/43939 [23:34:53<46:36:09, 5.72s/it]
33%|███▎ | 14618/43939 [23:34:58<46:36:10, 5.72s/it]
33%|███▎ | 14619/43939 [23:35:04<46:31:11, 5.71s/it]
33%|███▎ | 14620/43939 [23:35:10<46:23:56, 5.70s/it]
33%|███▎ | 14621/43939 [23:35:15<46:24:58, 5.70s/it]
33%|███▎ | 14622/43939 [23:35:21<46:22:13, 5.69s/it]
33%|███▎ | 14623/43939 [23:35:27<46:23:04, 5.70s/it]
33%|███▎ | 14624/43939 [23:35:33<46:23:41, 5.70s/it]
33%|███▎ | 14625/43939 [23:35:38<46:20:17, 5.69s/it]
33%|███▎ | 14626/43939 [23:35:44<46:17:45, 5.69s/it]
33%|███▎ | 14627/43939 [23:35:50<46:17:16, 5.68s/it]
33%|███▎ | 14628/43939 [23:35:55<46:18:19, 5.69s/it]
33%|███▎ | 14629/43939 [23:36:01<46:22:20, 5.70s/it]
33%|███▎ | 14630/43939 [23:36:07<46:14:59, 5.68s/it]
33%|███▎ | 14631/43939 [23:36:12<46:17:59, 5.69s/it]
33%|███▎ | 14632/43939 [23:36:18<46:15:57, 5.68s/it]
33%|███▎ | 14633/43939 [23:36:24<46:15:49, 5.68s/it]
33%|███▎ | 14634/43939 [23:36:29<46:16:09, 5.68s/it]
33%|███▎ | 14635/43939 [23:36:35<46:49:37, 5.75s/it]
33%|███▎ | 14636/43939 [23:36:41<46:42:56, 5.74s/it]
33%|███▎ | 14637/43939 [23:36:47<46:38:12, 5.73s/it]
33%|███▎ | 14638/43939 [23:36:52<46:34:35, 5.72s/it]
33%|███▎ | 14639/43939 [23:36:58<46:40:07, 5.73s/it]
33%|███▎ | 14640/43939 [23:37:04<46:38:00, 5.73s/it]
33%|███▎ | 14641/43939 [23:37:10<46:34:38, 5.72s/it]
33%|███▎ | 14642/43939 [23:37:15<46:35:54, 5.73s/it]
33%|███▎ | 14643/43939 [23:37:21<46:37:39, 5.73s/it]
33%|███▎ | 14644/43939 [23:37:27<46:35:57, 5.73s/it]
33%|███▎ | 14645/43939 [23:37:32<46:29:33, 5.71s/it]
33%|███▎ | 14646/43939 [23:37:38<46:29:30, 5.71s/it]
33%|███▎ | 14647/43939 [23:37:44<46:28:04, 5.71s/it]
33%|███▎ | 14648/43939 [23:37:50<46:35:22, 5.73s/it]
33%|███▎ | 14649/43939 [23:37:55<46:29:33, 5.71s/it]
33%|███▎ | 14650/43939 [23:38:01<46:27:11, 5.71s/it]
{'loss': 0.9217, 'grad_norm': 1.0510320663452148, 'learning_rate': 4.215122504222855e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.41, 'epoch': 1.33}
33%|███▎ | 14650/43939 [23:38:01<46:27:11, 5.71s/it]
33%|███▎ | 14651/43939 [23:38:07<46:21:56, 5.70s/it]
33%|███▎ | 14652/43939 [23:38:12<46:29:55, 5.72s/it]
33%|███▎ | 14653/43939 [23:38:18<46:22:01, 5.70s/it]
33%|███▎ | 14654/43939 [23:38:24<46:23:29, 5.70s/it]
33%|███▎ | 14655/43939 [23:38:29<46:16:57, 5.69s/it]
33%|███▎ | 14656/43939 [23:38:35<46:52:03, 5.76s/it]
33%|███▎ | 14657/43939 [23:38:41<46:44:08, 5.75s/it]
33%|███▎ | 14658/43939 [23:38:47<46:39:04, 5.74s/it]
33%|███▎ | 14659/43939 [23:38:53<46:34:45, 5.73s/it]
33%|███▎ | 14660/43939 [23:38:58<46:25:09, 5.71s/it]
33%|███▎ | 14661/43939 [23:39:04<46:25:01, 5.71s/it]
33%|███▎ | 14662/43939 [23:39:10<46:30:50, 5.72s/it]
33%|███▎ | 14663/43939 [23:39:15<46:29:57, 5.72s/it]
33%|███▎ | 14664/43939 [23:39:21<46:24:55, 5.71s/it]
33%|███▎ | 14665/43939 [23:39:27<46:21:45, 5.70s/it]
33%|███▎ | 14666/43939 [23:39:33<46:29:07, 5.72s/it]
33%|███▎ | 14667/43939 [23:39:38<46:20:13, 5.70s/it]
33%|███▎ | 14668/43939 [23:39:44<46:22:50, 5.70s/it]
33%|███▎ | 14669/43939 [23:39:50<46:26:15, 5.71s/it]
33%|███▎ | 14670/43939 [23:39:55<46:29:46, 5.72s/it]
33%|███▎ | 14671/43939 [23:40:01<46:30:46, 5.72s/it]
33%|███▎ | 14672/43939 [23:40:07<46:21:19, 5.70s/it]
33%|███▎ | 14673/43939 [23:40:12<46:26:49, 5.71s/it]
33%|███▎ | 14674/43939 [23:40:18<46:22:27, 5.70s/it]
33%|███▎ | 14675/43939 [23:40:24<46:30:36, 5.72s/it]
33%|███▎ | 14676/43939 [23:40:30<46:31:36, 5.72s/it]
33%|███▎ | 14677/43939 [23:40:36<47:02:30, 5.79s/it]
33%|███▎ | 14678/43939 [23:40:41<46:50:49, 5.76s/it]
33%|███▎ | 14679/43939 [23:40:47<46:42:45, 5.75s/it]
33%|███▎ | 14680/43939 [23:40:53<46:32:00, 5.73s/it]
33%|███▎ | 14681/43939 [23:40:58<46:31:51, 5.73s/it]
33%|███▎ | 14682/43939 [23:41:04<46:25:40, 5.71s/it]
33%|███▎ | 14683/43939 [23:41:10<46:31:36, 5.73s/it]
33%|███▎ | 14684/43939 [23:41:16<46:31:40, 5.73s/it]
33%|███▎ | 14685/43939 [23:41:21<46:28:33, 5.72s/it]
33%|███▎ | 14686/43939 [23:41:27<46:27:19, 5.72s/it]
33%|███▎ | 14687/43939 [23:41:33<46:32:08, 5.73s/it]
33%|███▎ | 14688/43939 [23:41:38<46:25:57, 5.71s/it]
33%|███▎ | 14689/43939 [23:41:44<46:32:28, 5.73s/it]
33%|███▎ | 14690/43939 [23:41:50<46:31:14, 5.73s/it]
33%|███▎ | 14691/43939 [23:41:56<46:26:44, 5.72s/it]
33%|███▎ | 14692/43939 [23:42:01<46:19:37, 5.70s/it]
33%|███▎ | 14693/43939 [23:42:07<46:13:18, 5.69s/it]
33%|███▎ | 14694/43939 [23:42:13<46:18:19, 5.70s/it]
33%|███▎ | 14695/43939 [23:42:18<46:17:59, 5.70s/it]
33%|███▎ | 14696/43939 [23:42:24<46:15:49, 5.70s/it]
33%|███▎ | 14697/43939 [23:42:30<46:17:52, 5.70s/it]
33%|███▎ | 14698/43939 [23:42:36<46:47:54, 5.76s/it]
33%|███▎ | 14699/43939 [23:42:41<46:35:27, 5.74s/it]
33%|███▎ | 14700/43939 [23:42:47<46:30:36, 5.73s/it]
{'loss': 0.9199, 'grad_norm': 1.1847153902053833, 'learning_rate': 4.207884234353559e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.71, 'epoch': 1.34}
33%|███▎ | 14700/43939 [23:42:47<46:30:36, 5.73s/it]
33%|███▎ | 14701/43939 [23:42:53<46:34:29, 5.73s/it]
33%|███▎ | 14702/43939 [23:42:59<46:34:59, 5.74s/it]
33%|███▎ | 14703/43939 [23:43:04<46:22:52, 5.71s/it]
33%|███▎ | 14704/43939 [23:43:10<46:28:28, 5.72s/it]
33%|███▎ | 14705/43939 [23:43:16<46:27:58, 5.72s/it]
33%|███▎ | 14706/43939 [23:43:21<46:24:18, 5.71s/it]
33%|███▎ | 14707/43939 [23:43:27<46:17:45, 5.70s/it]
33%|███▎ | 14708/43939 [23:43:33<46:14:15, 5.69s/it]
33%|███▎ | 14709/43939 [23:43:38<46:17:45, 5.70s/it]
33%|███▎ | 14710/43939 [23:43:44<46:17:54, 5.70s/it]
33%|███▎ | 14711/43939 [23:43:50<46:17:43, 5.70s/it]
33%|███▎ | 14712/43939 [23:43:56<46:18:08, 5.70s/it]
33%|███▎ | 14713/43939 [23:44:01<46:23:45, 5.71s/it]
33%|███▎ | 14714/43939 [23:44:07<46:31:14, 5.73s/it]
33%|███▎ | 14715/43939 [23:44:13<46:22:54, 5.71s/it]
33%|███▎ | 14716/43939 [23:44:18<46:17:11, 5.70s/it]
33%|███▎ | 14717/43939 [23:44:24<46:23:39, 5.72s/it]
33%|███▎ | 14718/43939 [23:44:30<46:20:21, 5.71s/it]
33%|███▎ | 14719/43939 [23:44:35<46:16:40, 5.70s/it]
34%|███▎ | 14720/43939 [23:44:41<46:54:10, 5.78s/it]
34%|███▎ | 14721/43939 [23:44:47<46:38:04, 5.75s/it]
34%|███▎ | 14722/43939 [23:44:53<46:33:39, 5.74s/it]
34%|███▎ | 14723/43939 [23:44:59<46:26:05, 5.72s/it]
34%|███▎ | 14724/43939 [23:45:04<46:17:25, 5.70s/it]
34%|███▎ | 14725/43939 [23:45:10<46:17:08, 5.70s/it]
34%|███▎ | 14726/43939 [23:45:16<46:14:17, 5.70s/it]
34%|███▎ | 14727/43939 [23:45:21<46:12:31, 5.69s/it]
34%|███▎ | 14728/43939 [23:45:27<46:15:54, 5.70s/it]
34%|███▎ | 14729/43939 [23:45:33<46:18:01, 5.71s/it]
34%|███▎ | 14730/43939 [23:45:38<46:15:36, 5.70s/it]
34%|███▎ | 14731/43939 [23:45:44<46:13:15, 5.70s/it]
34%|███▎ | 14732/43939 [23:45:50<46:17:34, 5.71s/it]
34%|███▎ | 14733/43939 [23:45:55<46:12:41, 5.70s/it]
34%|███▎ | 14734/43939 [23:46:01<46:10:19, 5.69s/it]
34%|███▎ | 14735/43939 [23:46:07<46:15:08, 5.70s/it]
34%|███▎ | 14736/43939 [23:46:13<46:11:57, 5.70s/it]
34%|███▎ | 14737/43939 [23:46:18<46:15:25, 5.70s/it]
34%|███▎ | 14738/43939 [23:46:24<46:13:33, 5.70s/it]
34%|███▎ | 14739/43939 [23:46:30<46:12:57, 5.70s/it]
34%|███▎ | 14740/43939 [23:46:35<46:20:59, 5.71s/it]
34%|███▎ | 14741/43939 [23:46:41<46:51:16, 5.78s/it]
34%|███▎ | 14742/43939 [23:46:47<46:39:24, 5.75s/it]
34%|███▎ | 14743/43939 [23:46:53<46:33:30, 5.74s/it]
34%|███▎ | 14744/43939 [23:46:58<46:32:06, 5.74s/it]
34%|███▎ | 14745/43939 [23:47:04<46:26:01, 5.73s/it]
34%|███▎ | 14746/43939 [23:47:10<46:22:37, 5.72s/it]
34%|███▎ | 14747/43939 [23:47:16<46:17:34, 5.71s/it]
34%|███▎ | 14748/43939 [23:47:21<46:11:07, 5.70s/it]
34%|███▎ | 14749/43939 [23:47:27<46:12:32, 5.70s/it]
34%|███▎ | 14750/43939 [23:47:33<46:14:33, 5.70s/it]
{'loss': 0.9372, 'grad_norm': 1.2461003065109253, 'learning_rate': 4.2006190185969916e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.24, 'epoch': 1.34}
34%|███▎ | 14750/43939 [23:47:33<46:14:33, 5.70s/it]
34%|███▎ | 14751/43939 [23:47:38<46:18:46, 5.71s/it]
34%|███▎ | 14752/43939 [23:47:44<46:11:40, 5.70s/it]
34%|███▎ | 14753/43939 [23:47:50<46:08:14, 5.69s/it]
34%|███▎ | 14754/43939 [23:47:55<46:08:18, 5.69s/it]
34%|███▎ | 14755/43939 [23:48:01<46:13:39, 5.70s/it]
34%|███▎ | 14756/43939 [23:48:07<46:10:38, 5.70s/it]
34%|███▎ | 14757/43939 [23:48:13<46:12:39, 5.70s/it]
34%|███▎ | 14758/43939 [23:48:18<46:11:29, 5.70s/it]
34%|███▎ | 14759/43939 [23:48:24<46:10:28, 5.70s/it]
34%|███▎ | 14760/43939 [23:48:30<46:13:19, 5.70s/it]
34%|███▎ | 14761/43939 [23:48:35<46:13:26, 5.70s/it]
34%|███▎ | 14762/43939 [23:48:41<46:41:14, 5.76s/it]
34%|███▎ | 14763/43939 [23:48:47<46:30:02, 5.74s/it]
34%|███▎ | 14764/43939 [23:48:53<46:23:06, 5.72s/it]
34%|███▎ | 14765/43939 [23:48:58<46:19:21, 5.72s/it]
34%|███▎ | 14766/43939 [23:49:04<46:16:58, 5.71s/it]
34%|███▎ | 14767/43939 [23:49:10<46:16:26, 5.71s/it]
34%|███▎ | 14768/43939 [23:49:15<46:17:48, 5.71s/it]
34%|███▎ | 14769/43939 [23:49:21<46:10:47, 5.70s/it]
34%|███▎ | 14770/43939 [23:49:27<46:10:29, 5.70s/it]
34%|███▎ | 14771/43939 [23:49:32<46:03:18, 5.68s/it]
34%|███▎ | 14772/43939 [23:49:38<46:08:10, 5.69s/it]
34%|███▎ | 14773/43939 [23:49:44<46:04:08, 5.69s/it]
34%|███▎ | 14774/43939 [23:49:50<46:08:27, 5.70s/it]
34%|███▎ | 14775/43939 [23:49:55<46:05:38, 5.69s/it]
34%|███▎ | 14776/43939 [23:50:01<45:59:55, 5.68s/it]
34%|███▎ | 14777/43939 [23:50:07<45:57:27, 5.67s/it]
34%|███▎ | 14778/43939 [23:50:12<46:02:42, 5.68s/it]
34%|███▎ | 14779/43939 [23:50:18<46:17:47, 5.72s/it]
34%|███▎ | 14780/43939 [23:50:24<46:19:34, 5.72s/it]
34%|███▎ | 14781/43939 [23:50:29<46:14:57, 5.71s/it]
34%|███▎ | 14782/43939 [23:50:35<46:16:47, 5.71s/it]
34%|███▎ | 14783/43939 [23:50:41<46:50:33, 5.78s/it]
34%|███▎ | 14784/43939 [23:50:47<46:34:39, 5.75s/it]
34%|███▎ | 14785/43939 [23:50:53<46:28:06, 5.74s/it]
34%|███▎ | 14786/43939 [23:50:58<46:28:47, 5.74s/it]
34%|███▎ | 14787/43939 [23:51:04<46:27:07, 5.74s/it]
34%|███▎ | 14788/43939 [23:51:10<46:25:53, 5.73s/it]
34%|███▎ | 14789/43939 [23:51:15<46:19:44, 5.72s/it]
34%|███▎ | 14790/43939 [23:51:21<46:15:47, 5.71s/it]
34%|███▎ | 14791/43939 [23:51:27<46:14:46, 5.71s/it]
34%|███▎ | 14792/43939 [23:51:33<46:21:41, 5.73s/it]
34%|███▎ | 14793/43939 [23:51:38<46:18:12, 5.72s/it]
34%|███▎ | 14794/43939 [23:51:44<46:14:35, 5.71s/it]
34%|███▎ | 14795/43939 [23:51:50<46:11:12, 5.71s/it]
34%|███▎ | 14796/43939 [23:51:55<46:11:05, 5.71s/it]
34%|███▎ | 14797/43939 [23:52:01<46:13:51, 5.71s/it]
34%|███▎ | 14798/43939 [23:52:07<46:13:35, 5.71s/it]
34%|███▎ | 14799/43939 [23:52:12<46:06:16, 5.70s/it]
34%|███▎ | 14800/43939 [23:52:18<46:10:44, 5.71s/it]
{'loss': 0.945, 'grad_norm': 1.0908567905426025, 'learning_rate': 4.193326971579008e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.48, 'epoch': 1.35}
34%|███▎ | 14800/43939 [23:52:18<46:10:44, 5.71s/it]
34%|███▎ | 14801/43939 [23:52:24<46:18:50, 5.72s/it]
34%|███▎ | 14802/43939 [23:52:30<46:18:09, 5.72s/it]
34%|███▎ | 14803/43939 [23:52:35<46:17:20, 5.72s/it]
34%|███▎ | 14804/43939 [23:52:41<46:49:15, 5.79s/it]
34%|███▎ | 14805/43939 [23:52:47<46:38:26, 5.76s/it]
34%|███▎ | 14806/43939 [23:52:53<46:36:58, 5.76s/it]
34%|███▎ | 14807/43939 [23:52:59<46:29:12, 5.74s/it]
34%|███▎ | 14808/43939 [23:53:04<46:21:03, 5.73s/it]
34%|███▎ | 14809/43939 [23:53:10<46:20:26, 5.73s/it]
34%|███▎ | 14810/43939 [23:53:16<46:15:50, 5.72s/it]
34%|███▎ | 14811/43939 [23:53:21<46:16:17, 5.72s/it]
34%|███▎ | 14812/43939 [23:53:27<46:09:24, 5.70s/it]
34%|███▎ | 14813/43939 [23:53:33<46:08:08, 5.70s/it]
34%|███▎ | 14814/43939 [23:53:38<46:02:17, 5.69s/it]
34%|███▎ | 14815/43939 [23:53:44<45:59:16, 5.68s/it]
34%|███▎ | 14816/43939 [23:53:50<46:01:58, 5.69s/it]
34%|███▎ | 14817/43939 [23:53:55<46:04:40, 5.70s/it]
34%|███▎ | 14818/43939 [23:54:01<46:01:17, 5.69s/it]
34%|███▎ | 14819/43939 [23:54:07<45:58:54, 5.68s/it]
34%|███▎ | 14820/43939 [23:54:12<46:00:00, 5.69s/it]
34%|███▎ | 14821/43939 [23:54:18<45:58:40, 5.68s/it]
34%|███▎ | 14822/43939 [23:54:24<45:58:08, 5.68s/it]
34%|███▎ | 14823/43939 [23:54:30<46:03:08, 5.69s/it]
34%|███▎ | 14824/43939 [23:54:35<46:30:57, 5.75s/it]
34%|███▎ | 14825/43939 [23:54:41<46:23:08, 5.74s/it]
34%|███▎ | 14826/43939 [23:54:47<46:22:19, 5.73s/it]
34%|███▎ | 14827/43939 [23:54:53<46:12:33, 5.71s/it]
34%|███▎ | 14828/43939 [23:54:58<46:08:56, 5.71s/it]
34%|███▎ | 14829/43939 [23:55:04<46:08:25, 5.71s/it]
34%|███▍ | 14830/43939 [23:55:10<46:07:10, 5.70s/it]
34%|███▍ | 14831/43939 [23:55:15<46:03:07, 5.70s/it]
34%|███▍ | 14832/43939 [23:55:21<46:03:46, 5.70s/it]
34%|███▍ | 14833/43939 [23:55:27<46:04:12, 5.70s/it]
34%|███▍ | 14834/43939 [23:55:32<46:07:30, 5.71s/it]
34%|███▍ | 14835/43939 [23:55:38<46:08:31, 5.71s/it]
34%|███▍ | 14836/43939 [23:55:44<46:08:08, 5.71s/it]
34%|███▍ | 14837/43939 [23:55:50<46:17:25, 5.73s/it]
34%|███▍ | 14838/43939 [23:55:55<46:14:29, 5.72s/it]
34%|███▍ | 14839/43939 [23:56:01<46:08:01, 5.71s/it]
34%|███▍ | 14840/43939 [23:56:07<46:05:14, 5.70s/it]
34%|███▍ | 14841/43939 [23:56:12<46:03:32, 5.70s/it]
34%|███▍ | 14842/43939 [23:56:18<46:02:15, 5.70s/it]
34%|███▍ | 14843/43939 [23:56:24<46:01:44, 5.70s/it]
34%|███▍ | 14844/43939 [23:56:29<46:01:26, 5.69s/it]
34%|███▍ | 14845/43939 [23:56:35<46:29:18, 5.75s/it]
34%|███▍ | 14846/43939 [23:56:41<46:22:10, 5.74s/it]
34%|███▍ | 14847/43939 [23:56:47<46:12:28, 5.72s/it]
34%|███▍ | 14848/43939 [23:56:52<46:08:09, 5.71s/it]
34%|███▍ | 14849/43939 [23:56:58<46:06:22, 5.71s/it]
34%|███▍ | 14850/43939 [23:57:04<46:03:22, 5.70s/it]
{'loss': 0.8984, 'grad_norm': 0.9096114039421082, 'learning_rate': 4.186008208348783e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.45, 'epoch': 1.35}
34%|███▍ | 14850/43939 [23:57:04<46:03:22, 5.70s/it]
34%|███▍ | 14851/43939 [23:57:10<46:11:19, 5.72s/it]
34%|███▍ | 14852/43939 [23:57:15<46:07:51, 5.71s/it]
34%|███▍ | 14853/43939 [23:57:21<46:02:27, 5.70s/it]
34%|███▍ | 14854/43939 [23:57:27<46:03:59, 5.70s/it]
34%|███▍ | 14855/43939 [23:57:32<46:01:30, 5.70s/it]
34%|███▍ | 14856/43939 [23:57:38<46:03:40, 5.70s/it]
34%|███▍ | 14857/43939 [23:57:44<46:04:40, 5.70s/it]
34%|███▍ | 14858/43939 [23:57:49<46:10:21, 5.72s/it]
34%|███▍ | 14859/43939 [23:57:55<46:09:38, 5.71s/it]
34%|███▍ | 14860/43939 [23:58:01<46:07:37, 5.71s/it]
34%|███▍ | 14861/43939 [23:58:07<46:08:08, 5.71s/it]
34%|███▍ | 14862/43939 [23:58:12<46:10:23, 5.72s/it]
34%|███▍ | 14863/43939 [23:58:18<46:06:03, 5.71s/it]
34%|███▍ | 14864/43939 [23:58:24<46:11:22, 5.72s/it]
34%|███▍ | 14865/43939 [23:58:29<46:05:58, 5.71s/it]
34%|███▍ | 14866/43939 [23:58:35<46:00:59, 5.70s/it]
34%|███▍ | 14867/43939 [23:58:41<46:28:52, 5.76s/it]
34%|███▍ | 14868/43939 [23:58:47<46:23:00, 5.74s/it]
34%|███▍ | 14869/43939 [23:58:52<46:14:12, 5.73s/it]
34%|███▍ | 14870/43939 [23:58:58<46:09:03, 5.72s/it]
34%|███▍ | 14871/43939 [23:59:04<46:04:49, 5.71s/it]
34%|███▍ | 14872/43939 [23:59:10<46:04:45, 5.71s/it]
34%|███▍ | 14873/43939 [23:59:15<46:07:31, 5.71s/it]
34%|███▍ | 14874/43939 [23:59:21<46:08:35, 5.72s/it]
34%|███▍ | 14875/43939 [23:59:27<46:05:23, 5.71s/it]
34%|███▍ | 14876/43939 [23:59:32<46:03:23, 5.70s/it]
34%|███▍ | 14877/43939 [23:59:38<46:02:17, 5.70s/it]
34%|███▍ | 14878/43939 [23:59:44<46:02:03, 5.70s/it]
34%|███▍ | 14879/43939 [23:59:49<46:00:16, 5.70s/it]
34%|███▍ | 14880/43939 [23:59:55<46:05:47, 5.71s/it]
34%|███▍ | 14881/43939 [24:00:01<46:09:17, 5.72s/it]
34%|███▍ | 14882/43939 [24:00:07<46:09:45, 5.72s/it]
34%|███▍ | 14883/43939 [24:00:12<46:05:20, 5.71s/it]
34%|███▍ | 14884/43939 [24:00:18<46:00:10, 5.70s/it]
34%|███▍ | 14885/43939 [24:00:24<45:54:59, 5.69s/it]
34%|███▍ | 14886/43939 [24:00:29<46:07:55, 5.72s/it]
34%|███▍ | 14887/43939 [24:00:35<46:04:38, 5.71s/it]
34%|███▍ | 14888/43939 [24:00:41<46:34:13, 5.77s/it]
34%|███▍ | 14889/43939 [24:00:47<46:23:57, 5.75s/it]
34%|███▍ | 14890/43939 [24:00:52<46:23:14, 5.75s/it]
34%|███▍ | 14891/43939 [24:00:58<46:16:25, 5.73s/it]
34%|███▍ | 14892/43939 [24:01:04<46:11:45, 5.73s/it]
34%|███▍ | 14893/43939 [24:01:10<46:18:26, 5.74s/it]
34%|███▍ | 14894/43939 [24:01:15<46:13:37, 5.73s/it]
34%|███▍ | 14895/43939 [24:01:21<46:10:40, 5.72s/it]
34%|███▍ | 14896/43939 [24:01:27<46:07:22, 5.72s/it]
34%|███▍ | 14897/43939 [24:01:32<46:02:59, 5.71s/it]
34%|███▍ | 14898/43939 [24:01:38<46:02:33, 5.71s/it]
34%|███▍ | 14899/43939 [24:01:44<46:00:28, 5.70s/it]
34%|███▍ | 14900/43939 [24:01:50<45:58:06, 5.70s/it]
{'loss': 0.9285, 'grad_norm': 1.0902049541473389, 'learning_rate': 4.178662844377006e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.1, 'epoch': 1.36}
34%|███▍ | 14900/43939 [24:01:50<45:58:06, 5.70s/it]
34%|███▍ | 14901/43939 [24:01:55<46:00:40, 5.70s/it]
34%|███▍ | 14902/43939 [24:02:01<46:02:18, 5.71s/it]
34%|███▍ | 14903/43939 [24:02:07<45:59:50, 5.70s/it]
34%|███▍ | 14904/43939 [24:02:12<45:56:29, 5.70s/it]
34%|███▍ | 14905/43939 [24:02:18<45:57:11, 5.70s/it]
34%|███▍ | 14906/43939 [24:02:24<45:56:34, 5.70s/it]
34%|███▍ | 14907/43939 [24:02:29<45:57:46, 5.70s/it]
34%|███▍ | 14908/43939 [24:02:35<45:56:22, 5.70s/it]
34%|███▍ | 14909/43939 [24:02:41<46:32:40, 5.77s/it]
34%|███▍ | 14910/43939 [24:02:47<46:21:49, 5.75s/it]
34%|███▍ | 14911/43939 [24:02:53<46:21:07, 5.75s/it]
34%|███▍ | 14912/43939 [24:02:58<46:10:46, 5.73s/it]
34%|███▍ | 14913/43939 [24:03:04<46:00:30, 5.71s/it]
34%|███▍ | 14914/43939 [24:03:10<45:52:04, 5.69s/it]
34%|███▍ | 14915/43939 [24:03:15<46:04:10, 5.71s/it]
34%|███▍ | 14916/43939 [24:03:21<45:57:47, 5.70s/it]
34%|███▍ | 14917/43939 [24:03:27<45:58:16, 5.70s/it]
34%|███▍ | 14918/43939 [24:03:32<45:51:36, 5.69s/it]
34%|███▍ | 14919/43939 [24:03:38<45:58:58, 5.70s/it]
34%|███▍ | 14920/43939 [24:03:44<46:04:22, 5.72s/it]
34%|███▍ | 14921/43939 [24:03:50<45:58:10, 5.70s/it]
34%|███▍ | 14922/43939 [24:03:55<45:55:12, 5.70s/it]
34%|███▍ | 14923/43939 [24:04:01<45:57:35, 5.70s/it]
34%|███▍ | 14924/43939 [24:04:07<46:08:19, 5.72s/it]
34%|███▍ | 14925/43939 [24:04:12<46:04:30, 5.72s/it]
34%|███▍ | 14926/43939 [24:04:18<46:18:22, 5.75s/it]
34%|███▍ | 14927/43939 [24:04:24<46:09:28, 5.73s/it]
34%|███▍ | 14928/43939 [24:04:30<46:08:48, 5.73s/it]
34%|███▍ | 14929/43939 [24:04:35<46:06:42, 5.72s/it]
34%|███▍ | 14930/43939 [24:04:41<46:33:10, 5.78s/it]
34%|███▍ | 14931/43939 [24:04:47<46:24:32, 5.76s/it]
34%|███▍ | 14932/43939 [24:04:53<46:13:39, 5.74s/it]
34%|███▍ | 14933/43939 [24:04:58<46:07:46, 5.73s/it]
34%|███▍ | 14934/43939 [24:05:04<46:02:44, 5.72s/it]
34%|███▍ | 14935/43939 [24:05:10<46:00:50, 5.71s/it]
34%|███▍ | 14936/43939 [24:05:15<46:00:05, 5.71s/it]
34%|███▍ | 14937/43939 [24:05:21<45:51:06, 5.69s/it]
34%|███▍ | 14938/43939 [24:05:27<45:53:47, 5.70s/it]
34%|███▍ | 14939/43939 [24:05:32<45:51:43, 5.69s/it]
34%|███▍ | 14940/43939 [24:05:38<45:50:31, 5.69s/it]
34%|███▍ | 14941/43939 [24:05:44<45:49:00, 5.69s/it]
34%|███▍ | 14942/43939 [24:05:50<45:51:57, 5.69s/it]
34%|███▍ | 14943/43939 [24:05:55<45:50:17, 5.69s/it]
34%|███▍ | 14944/43939 [24:06:01<45:49:38, 5.69s/it]
34%|███▍ | 14945/43939 [24:06:07<45:54:54, 5.70s/it]
34%|███▍ | 14946/43939 [24:06:12<45:55:08, 5.70s/it]
34%|███▍ | 14947/43939 [24:06:18<45:53:25, 5.70s/it]
34%|███▍ | 14948/43939 [24:06:24<45:49:20, 5.69s/it]
34%|███▍ | 14949/43939 [24:06:29<45:50:08, 5.69s/it]
34%|███▍ | 14950/43939 [24:06:35<45:49:09, 5.69s/it]
{'loss': 0.9319, 'grad_norm': 1.561834692955017, 'learning_rate': 4.171290995554056e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.48, 'epoch': 1.36}
34%|███▍ | 14950/43939 [24:06:35<45:49:09, 5.69s/it]
34%|███▍ | 14951/43939 [24:06:41<46:22:38, 5.76s/it]
34%|███▍ | 14952/43939 [24:06:47<46:15:30, 5.75s/it]
34%|███▍ | 14953/43939 [24:06:52<46:08:09, 5.73s/it]
34%|███▍ | 14954/43939 [24:06:58<45:58:48, 5.71s/it]
34%|███▍ | 14955/43939 [24:07:04<45:53:04, 5.70s/it]
34%|███▍ | 14956/43939 [24:07:09<45:52:24, 5.70s/it]
34%|███▍ | 14957/43939 [24:07:15<45:56:17, 5.71s/it]
34%|███▍ | 14958/43939 [24:07:21<46:05:53, 5.73s/it]
34%|███▍ | 14959/43939 [24:07:27<45:57:17, 5.71s/it]
34%|███▍ | 14960/43939 [24:07:32<45:54:30, 5.70s/it]
34%|███▍ | 14961/43939 [24:07:38<45:54:38, 5.70s/it]
34%|███▍ | 14962/43939 [24:07:44<45:47:18, 5.69s/it]
34%|███▍ | 14963/43939 [24:07:49<45:54:22, 5.70s/it]
34%|███▍ | 14964/43939 [24:07:55<45:48:32, 5.69s/it]
34%|███▍ | 14965/43939 [24:08:01<45:45:18, 5.69s/it]
34%|███▍ | 14966/43939 [24:08:06<45:46:33, 5.69s/it]
34%|███▍ | 14967/43939 [24:08:12<45:47:17, 5.69s/it]
34%|███▍ | 14968/43939 [24:08:18<45:45:19, 5.69s/it]
34%|███▍ | 14969/43939 [24:08:23<45:42:05, 5.68s/it]
34%|███▍ | 14970/43939 [24:08:29<45:45:06, 5.69s/it]
34%|███▍ | 14971/43939 [24:08:35<45:47:00, 5.69s/it]
34%|███▍ | 14972/43939 [24:08:41<46:26:20, 5.77s/it]
34%|███▍ | 14973/43939 [24:08:47<46:21:26, 5.76s/it]
34%|███▍ | 14974/43939 [24:08:52<46:14:39, 5.75s/it]
34%|███▍ | 14975/43939 [24:08:58<46:11:22, 5.74s/it]
34%|███▍ | 14976/43939 [24:09:04<46:06:34, 5.73s/it]
34%|███▍ | 14977/43939 [24:09:09<46:01:32, 5.72s/it]
34%|███▍ | 14978/43939 [24:09:15<45:55:55, 5.71s/it]
34%|███▍ | 14979/43939 [24:09:21<45:54:56, 5.71s/it]
34%|███▍ | 14980/43939 [24:09:26<45:52:57, 5.70s/it]
34%|███▍ | 14981/43939 [24:09:32<45:48:26, 5.69s/it]
34%|███▍ | 14982/43939 [24:09:38<45:49:34, 5.70s/it]
34%|███▍ | 14983/43939 [24:09:44<45:49:35, 5.70s/it]
34%|███▍ | 14984/43939 [24:09:49<45:54:55, 5.71s/it]
34%|███▍ | 14985/43939 [24:09:55<45:51:37, 5.70s/it]
34%|███▍ | 14986/43939 [24:10:01<45:48:36, 5.70s/it]
34%|███▍ | 14987/43939 [24:10:06<45:49:07, 5.70s/it]
34%|███▍ | 14988/43939 [24:10:12<45:59:36, 5.72s/it]
34%|███▍ | 14989/43939 [24:10:18<45:59:36, 5.72s/it]
34%|███▍ | 14990/43939 [24:10:24<45:54:24, 5.71s/it]
34%|███▍ | 14991/43939 [24:10:29<45:52:13, 5.70s/it]
34%|███▍ | 14992/43939 [24:10:35<45:49:26, 5.70s/it]
34%|███▍ | 14993/43939 [24:10:41<46:27:46, 5.78s/it]
34%|███▍ | 14994/43939 [24:10:47<46:14:45, 5.75s/it]
34%|███▍ | 14995/43939 [24:10:52<46:05:47, 5.73s/it]
34%|███▍ | 14996/43939 [24:10:58<46:03:30, 5.73s/it]
34%|███▍ | 14997/43939 [24:11:04<45:59:42, 5.72s/it]
34%|███▍ | 14998/43939 [24:11:09<45:58:51, 5.72s/it]
34%|███▍ | 14999/43939 [24:11:15<45:58:41, 5.72s/it]
34%|███▍ | 15000/43939 [24:11:21<45:55:56, 5.71s/it]
{'loss': 0.9382, 'grad_norm': 1.1353206634521484, 'learning_rate': 4.163892778188171e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.98, 'epoch': 1.37}
34%|███▍ | 15000/43939 [24:11:21<45:55:56, 5.71s/it]
34%|███▍ | 15001/43939 [24:11:27<46:03:55, 5.73s/it]
34%|███▍ | 15002/43939 [24:11:32<45:58:51, 5.72s/it]
34%|███▍ | 15003/43939 [24:11:38<45:55:27, 5.71s/it]
34%|███▍ | 15004/43939 [24:11:44<45:56:06, 5.72s/it]
34%|███▍ | 15005/43939 [24:11:49<45:55:26, 5.71s/it]
34%|███▍ | 15006/43939 [24:11:55<45:49:29, 5.70s/it]
34%|███▍ | 15007/43939 [24:12:01<45:58:12, 5.72s/it]
34%|███▍ | 15008/43939 [24:12:07<45:58:32, 5.72s/it]
34%|███▍ | 15009/43939 [24:12:12<45:57:21, 5.72s/it]
34%|███▍ | 15010/43939 [24:12:18<45:58:15, 5.72s/it]
34%|███▍ | 15011/43939 [24:12:24<45:52:46, 5.71s/it]
34%|███▍ | 15012/43939 [24:12:29<45:54:59, 5.71s/it]
34%|███▍ | 15013/43939 [24:12:35<45:55:26, 5.72s/it]
34%|███▍ | 15014/43939 [24:12:41<46:29:34, 5.79s/it]
34%|███▍ | 15015/43939 [24:12:47<46:20:28, 5.77s/it]
34%|███▍ | 15016/43939 [24:12:53<46:12:32, 5.75s/it]
34%|███▍ | 15017/43939 [24:12:58<46:08:19, 5.74s/it]
34%|███▍ | 15018/43939 [24:13:04<46:09:01, 5.74s/it]
34%|███▍ | 15019/43939 [24:13:10<46:05:10, 5.74s/it]
34%|███▍ | 15020/43939 [24:13:15<46:05:48, 5.74s/it]
34%|███▍ | 15021/43939 [24:13:21<45:56:52, 5.72s/it]
34%|███▍ | 15022/43939 [24:13:27<45:50:56, 5.71s/it]
34%|███▍ | 15023/43939 [24:13:33<45:52:56, 5.71s/it]
34%|███▍ | 15024/43939 [24:13:38<45:48:57, 5.70s/it]
34%|███▍ | 15025/43939 [24:13:44<45:50:08, 5.71s/it]
34%|███▍ | 15026/43939 [24:13:50<45:48:12, 5.70s/it]
34%|███▍ | 15027/43939 [24:13:55<45:49:28, 5.71s/it]
34%|███▍ | 15028/43939 [24:14:01<45:53:17, 5.71s/it]
34%|███▍ | 15029/43939 [24:14:07<45:47:04, 5.70s/it]
34%|███▍ | 15030/43939 [24:14:12<45:49:53, 5.71s/it]
34%|███▍ | 15031/43939 [24:14:18<45:42:32, 5.69s/it]
34%|███▍ | 15032/43939 [24:14:24<45:46:19, 5.70s/it]
34%|███▍ | 15033/43939 [24:14:30<45:43:49, 5.70s/it]
34%|███▍ | 15034/43939 [24:14:35<46:20:43, 5.77s/it]
34%|███▍ | 15035/43939 [24:14:41<46:18:36, 5.77s/it]
34%|███▍ | 15036/43939 [24:14:47<46:04:35, 5.74s/it]
34%|███▍ | 15037/43939 [24:14:53<45:59:38, 5.73s/it]
34%|███▍ | 15038/43939 [24:14:58<46:00:12, 5.73s/it]
34%|███▍ | 15039/43939 [24:15:04<45:58:47, 5.73s/it]
34%|███▍ | 15040/43939 [24:15:10<45:54:22, 5.72s/it]
34%|███▍ | 15041/43939 [24:15:16<45:59:32, 5.73s/it]
34%|███▍ | 15042/43939 [24:15:21<45:57:14, 5.72s/it]
34%|███▍ | 15043/43939 [24:15:27<45:55:00, 5.72s/it]
34%|███▍ | 15044/43939 [24:15:33<45:49:43, 5.71s/it]
34%|███▍ | 15045/43939 [24:15:38<45:49:06, 5.71s/it]
34%|███▍ | 15046/43939 [24:15:44<45:44:28, 5.70s/it]
34%|███▍ | 15047/43939 [24:15:50<45:46:57, 5.70s/it]
34%|███▍ | 15048/43939 [24:15:55<45:53:25, 5.72s/it]
34%|███▍ | 15049/43939 [24:16:01<45:53:51, 5.72s/it]
34%|███▍ | 15050/43939 [24:16:07<45:51:12, 5.71s/it]
{'loss': 0.9328, 'grad_norm': 0.9608362913131714, 'learning_rate': 4.156468309003616e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.52, 'epoch': 1.37}
34%|███▍ | 15050/43939 [24:16:07<45:51:12, 5.71s/it]
34%|███▍ | 15051/43939 [24:16:13<45:49:50, 5.71s/it]
34%|███▍ | 15052/43939 [24:16:18<45:46:48, 5.71s/it]
34%|███▍ | 15053/43939 [24:16:24<45:43:29, 5.70s/it]
34%|███▍ | 15054/43939 [24:16:30<45:35:53, 5.68s/it]
34%|███▍ | 15055/43939 [24:16:36<46:11:50, 5.76s/it]
34%|███▍ | 15056/43939 [24:16:41<46:07:32, 5.75s/it]
34%|███▍ | 15057/43939 [24:16:47<46:02:40, 5.74s/it]
34%|███▍ | 15058/43939 [24:16:53<45:55:57, 5.73s/it]
34%|███▍ | 15059/43939 [24:16:58<45:47:26, 5.71s/it]
34%|███▍ | 15060/43939 [24:17:04<45:49:16, 5.71s/it]
34%|███▍ | 15061/43939 [24:17:10<45:42:38, 5.70s/it]
34%|███▍ | 15062/43939 [24:17:15<45:47:39, 5.71s/it]
34%|███▍ | 15063/43939 [24:17:21<45:54:26, 5.72s/it]
34%|███▍ | 15064/43939 [24:17:27<45:55:50, 5.73s/it]
34%|███▍ | 15065/43939 [24:17:33<45:47:08, 5.71s/it]
34%|███▍ | 15066/43939 [24:17:38<45:45:29, 5.71s/it]
34%|███▍ | 15067/43939 [24:17:44<45:47:29, 5.71s/it]
34%|███▍ | 15068/43939 [24:17:50<45:46:17, 5.71s/it]
34%|███▍ | 15069/43939 [24:17:55<45:45:40, 5.71s/it]
34%|███▍ | 15070/43939 [24:18:01<45:48:37, 5.71s/it]
34%|███▍ | 15071/43939 [24:18:07<45:47:16, 5.71s/it]
34%|███▍ | 15072/43939 [24:18:13<45:43:54, 5.70s/it]
34%|███▍ | 15073/43939 [24:18:18<45:44:38, 5.70s/it]
34%|███▍ | 15074/43939 [24:18:24<45:46:33, 5.71s/it]
34%|███▍ | 15075/43939 [24:18:30<45:50:19, 5.72s/it]
34%|███▍ | 15076/43939 [24:18:35<45:53:30, 5.72s/it]
34%|███▍ | 15077/43939 [24:18:41<46:24:19, 5.79s/it]
34%|███▍ | 15078/43939 [24:18:47<46:13:24, 5.77s/it]
34%|███▍ | 15079/43939 [24:18:53<46:06:01, 5.75s/it]
34%|███▍ | 15080/43939 [24:18:59<45:56:09, 5.73s/it]
34%|███▍ | 15081/43939 [24:19:04<45:47:21, 5.71s/it]
34%|███▍ | 15082/43939 [24:19:10<45:51:44, 5.72s/it]
34%|███▍ | 15083/43939 [24:19:16<45:43:22, 5.70s/it]
34%|███▍ | 15084/43939 [24:19:21<45:41:56, 5.70s/it]
34%|███▍ | 15085/43939 [24:19:27<45:48:35, 5.72s/it]
34%|███▍ | 15086/43939 [24:19:33<45:48:23, 5.72s/it]
34%|███▍ | 15087/43939 [24:19:38<45:44:07, 5.71s/it]
34%|███▍ | 15088/43939 [24:19:44<45:36:59, 5.69s/it]
34%|███▍ | 15089/43939 [24:19:50<45:42:28, 5.70s/it]
34%|███▍ | 15090/43939 [24:19:56<45:42:28, 5.70s/it]
34%|███▍ | 15091/43939 [24:20:01<45:46:00, 5.71s/it]
34%|███▍ | 15092/43939 [24:20:07<45:46:57, 5.71s/it]
34%|███▍ | 15093/43939 [24:20:13<45:50:04, 5.72s/it]
34%|███▍ | 15094/43939 [24:20:18<45:48:13, 5.72s/it]
34%|███▍ | 15095/43939 [24:20:24<45:47:59, 5.72s/it]
34%|███▍ | 15096/43939 [24:20:30<45:44:23, 5.71s/it]
34%|███▍ | 15097/43939 [24:20:36<46:15:31, 5.77s/it]
34%|███▍ | 15098/43939 [24:20:41<46:05:59, 5.75s/it]
34%|███▍ | 15099/43939 [24:20:47<45:56:43, 5.74s/it]
34%|███▍ | 15100/43939 [24:20:53<45:48:18, 5.72s/it]
{'loss': 0.9274, 'grad_norm': 1.2019129991531372, 'learning_rate': 4.149017705138839e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.49, 'epoch': 1.37}
34%|███▍ | 15100/43939 [24:20:53<45:48:18, 5.72s/it]
34%|███▍ | 15101/43939 [24:20:59<45:47:40, 5.72s/it]
34%|███▍ | 15102/43939 [24:21:04<45:46:33, 5.71s/it]
34%|███▍ | 15103/43939 [24:21:10<45:44:28, 5.71s/it]
34%|███▍ | 15104/43939 [24:21:16<45:40:04, 5.70s/it]
34%|███▍ | 15105/43939 [24:21:21<45:39:50, 5.70s/it]
34%|███▍ | 15106/43939 [24:21:27<45:41:42, 5.71s/it]
34%|███▍ | 15107/43939 [24:21:33<45:38:30, 5.70s/it]
34%|███▍ | 15108/43939 [24:21:38<45:38:20, 5.70s/it]
34%|███▍ | 15109/43939 [24:21:44<45:35:51, 5.69s/it]
34%|███▍ | 15110/43939 [24:21:50<45:30:12, 5.68s/it]
34%|███▍ | 15111/43939 [24:21:55<45:28:51, 5.68s/it]
34%|███▍ | 15112/43939 [24:22:01<45:24:18, 5.67s/it]
34%|███▍ | 15113/43939 [24:22:07<45:29:08, 5.68s/it]
34%|███▍ | 15114/43939 [24:22:13<45:29:09, 5.68s/it]
34%|███▍ | 15115/43939 [24:22:18<45:26:17, 5.68s/it]
34%|███▍ | 15116/43939 [24:22:24<45:36:14, 5.70s/it]
34%|███▍ | 15117/43939 [24:22:30<45:41:21, 5.71s/it]
34%|███▍ | 15118/43939 [24:22:36<46:10:42, 5.77s/it]
34%|███▍ | 15119/43939 [24:22:41<46:04:06, 5.75s/it]
34%|███▍ | 15120/43939 [24:22:47<46:04:29, 5.76s/it]
34%|███▍ | 15121/43939 [24:22:53<45:53:04, 5.73s/it]
34%|███▍ | 15122/43939 [24:22:58<45:45:53, 5.72s/it]
34%|███▍ | 15123/43939 [24:23:04<45:37:47, 5.70s/it]
34%|███▍ | 15124/43939 [24:23:10<45:39:08, 5.70s/it]
34%|███▍ | 15125/43939 [24:23:15<45:35:58, 5.70s/it]
34%|███▍ | 15126/43939 [24:23:21<45:39:36, 5.70s/it]
34%|███▍ | 15127/43939 [24:23:27<45:36:57, 5.70s/it]
34%|███▍ | 15128/43939 [24:23:33<45:34:05, 5.69s/it]
34%|███▍ | 15129/43939 [24:23:38<45:32:11, 5.69s/it]
34%|███▍ | 15130/43939 [24:23:44<45:31:41, 5.69s/it]
34%|███▍ | 15131/43939 [24:23:50<45:37:36, 5.70s/it]
34%|███▍ | 15132/43939 [24:23:55<45:33:30, 5.69s/it]
34%|███▍ | 15133/43939 [24:24:01<45:33:41, 5.69s/it]
34%|███▍ | 15134/43939 [24:24:07<45:33:41, 5.69s/it]
34%|███▍ | 15135/43939 [24:24:12<45:36:08, 5.70s/it]
34%|███▍ | 15136/43939 [24:24:18<45:33:53, 5.70s/it]
34%|███▍ | 15137/43939 [24:24:24<45:42:39, 5.71s/it]
34%|███▍ | 15138/43939 [24:24:30<45:39:22, 5.71s/it]
34%|███▍ | 15139/43939 [24:24:35<45:43:47, 5.72s/it]
34%|███▍ | 15140/43939 [24:24:41<46:13:58, 5.78s/it]
34%|███▍ | 15141/43939 [24:24:47<46:01:53, 5.75s/it]
34%|███▍ | 15142/43939 [24:24:53<45:55:10, 5.74s/it]
34%|███▍ | 15143/43939 [24:24:58<45:49:14, 5.73s/it]
34%|███▍ | 15144/43939 [24:25:04<45:42:03, 5.71s/it]
34%|███▍ | 15145/43939 [24:25:10<45:34:12, 5.70s/it]
34%|███▍ | 15146/43939 [24:25:15<45:33:35, 5.70s/it]
34%|███▍ | 15147/43939 [24:25:21<45:35:59, 5.70s/it]
34%|███▍ | 15148/43939 [24:25:27<45:35:51, 5.70s/it]
34%|███▍ | 15149/43939 [24:25:32<45:38:03, 5.71s/it]
34%|███▍ | 15150/43939 [24:25:38<45:43:43, 5.72s/it]
{'loss': 0.9357, 'grad_norm': 1.0083372592926025, 'learning_rate': 4.1415410841446254e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.77, 'epoch': 1.38}
34%|███▍ | 15150/43939 [24:25:38<45:43:43, 5.72s/it]
34%|███▍ | 15151/43939 [24:25:44<45:38:38, 5.71s/it]
34%|███▍ | 15152/43939 [24:25:50<45:34:07, 5.70s/it]
34%|███▍ | 15153/43939 [24:25:55<45:35:32, 5.70s/it]
34%|███▍ | 15154/43939 [24:26:01<45:35:30, 5.70s/it]
34%|███▍ | 15155/43939 [24:26:07<45:36:24, 5.70s/it]
34%|███▍ | 15156/43939 [24:26:12<45:32:27, 5.70s/it]
34%|███▍ | 15157/43939 [24:26:18<45:28:50, 5.69s/it]
34%|███▍ | 15158/43939 [24:26:24<45:27:05, 5.69s/it]
35%|███▍ | 15159/43939 [24:26:29<45:34:05, 5.70s/it]
35%|███▍ | 15160/43939 [24:26:35<45:38:40, 5.71s/it]
35%|███▍ | 15161/43939 [24:26:41<45:39:46, 5.71s/it]
35%|███▍ | 15162/43939 [24:26:47<46:23:03, 5.80s/it]
35%|███▍ | 15163/43939 [24:26:53<46:05:13, 5.77s/it]
35%|███▍ | 15164/43939 [24:26:58<46:03:27, 5.76s/it]
35%|███▍ | 15165/43939 [24:27:04<45:56:45, 5.75s/it]
35%|███▍ | 15166/43939 [24:27:10<45:49:45, 5.73s/it]
35%|███▍ | 15167/43939 [24:27:15<45:41:36, 5.72s/it]
35%|███▍ | 15168/43939 [24:27:21<45:44:47, 5.72s/it]
35%|███▍ | 15169/43939 [24:27:27<45:55:51, 5.75s/it]
35%|███▍ | 15170/43939 [24:27:33<45:44:05, 5.72s/it]
35%|███▍ | 15171/43939 [24:27:38<45:44:51, 5.72s/it]
35%|███▍ | 15172/43939 [24:27:44<45:37:33, 5.71s/it]
35%|███▍ | 15173/43939 [24:27:50<45:36:28, 5.71s/it]
35%|███▍ | 15174/43939 [24:27:56<45:38:23, 5.71s/it]
35%|███▍ | 15175/43939 [24:28:01<45:36:14, 5.71s/it]
35%|███▍ | 15176/43939 [24:28:07<45:37:10, 5.71s/it]
35%|███▍ | 15177/43939 [24:28:13<45:38:31, 5.71s/it]
35%|███▍ | 15178/43939 [24:28:18<45:39:48, 5.72s/it]
35%|███▍ | 15179/43939 [24:28:24<45:31:37, 5.70s/it]
35%|███▍ | 15180/43939 [24:28:30<45:40:01, 5.72s/it]
35%|███▍ | 15181/43939 [24:28:35<45:37:35, 5.71s/it]
35%|███▍ | 15182/43939 [24:28:41<46:14:45, 5.79s/it]
35%|███▍ | 15183/43939 [24:28:47<46:02:31, 5.76s/it]
35%|███▍ | 15184/43939 [24:28:53<45:51:12, 5.74s/it]
35%|███▍ | 15185/43939 [24:28:59<45:49:03, 5.74s/it]
35%|███▍ | 15186/43939 [24:29:04<45:40:08, 5.72s/it]
35%|███▍ | 15187/43939 [24:29:10<45:35:48, 5.71s/it]
35%|███▍ | 15188/43939 [24:29:16<45:33:07, 5.70s/it]
35%|███▍ | 15189/43939 [24:29:21<45:32:15, 5.70s/it]
35%|███▍ | 15190/43939 [24:29:27<45:31:00, 5.70s/it]
35%|███▍ | 15191/43939 [24:29:33<45:30:35, 5.70s/it]
35%|███▍ | 15192/43939 [24:29:38<45:28:14, 5.69s/it]
35%|███▍ | 15193/43939 [24:29:44<45:27:51, 5.69s/it]
35%|███▍ | 15194/43939 [24:29:50<45:35:26, 5.71s/it]
35%|███▍ | 15195/43939 [24:29:56<45:30:59, 5.70s/it]
35%|███▍ | 15196/43939 [24:30:01<45:30:24, 5.70s/it]
35%|███▍ | 15197/43939 [24:30:07<45:30:52, 5.70s/it]
35%|███▍ | 15198/43939 [24:30:13<45:32:48, 5.71s/it]
35%|███▍ | 15199/43939 [24:30:18<45:35:59, 5.71s/it]
35%|███▍ | 15200/43939 [24:30:24<45:35:33, 5.71s/it]
{'loss': 0.9472, 'grad_norm': 0.9793724417686462, 'learning_rate': 4.134038563982241e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.95, 'epoch': 1.38}
35%|███▍ | 15200/43939 [24:30:24<45:35:33, 5.71s/it]
35%|███▍ | 15201/43939 [24:30:30<45:36:45, 5.71s/it]
35%|███▍ | 15202/43939 [24:30:35<45:35:29, 5.71s/it]
35%|███▍ | 15203/43939 [24:30:41<46:02:04, 5.77s/it]
35%|███▍ | 15204/43939 [24:30:47<45:53:39, 5.75s/it]
35%|███▍ | 15205/43939 [24:30:53<45:52:09, 5.75s/it]
35%|███▍ | 15206/43939 [24:30:59<45:43:31, 5.73s/it]
35%|███▍ | 15207/43939 [24:31:04<45:39:25, 5.72s/it]
35%|███▍ | 15208/43939 [24:31:10<45:33:18, 5.71s/it]
35%|███▍ | 15209/43939 [24:31:16<45:33:58, 5.71s/it]
35%|███▍ | 15210/43939 [24:31:21<45:30:53, 5.70s/it]
35%|███▍ | 15211/43939 [24:31:27<45:29:02, 5.70s/it]
35%|███▍ | 15212/43939 [24:31:33<45:27:26, 5.70s/it]
35%|███▍ | 15213/43939 [24:31:38<45:27:55, 5.70s/it]
35%|███▍ | 15214/43939 [24:31:44<45:27:33, 5.70s/it]
35%|███▍ | 15215/43939 [24:31:50<45:37:21, 5.72s/it]
35%|███▍ | 15216/43939 [24:31:56<45:31:14, 5.71s/it]
35%|███▍ | 15217/43939 [24:32:01<45:34:26, 5.71s/it]
35%|███▍ | 15218/43939 [24:32:07<45:32:20, 5.71s/it]
35%|███▍ | 15219/43939 [24:32:13<45:27:42, 5.70s/it]
35%|███▍ | 15220/43939 [24:32:18<45:27:40, 5.70s/it]
35%|███▍ | 15221/43939 [24:32:24<45:31:59, 5.71s/it]
35%|███▍ | 15222/43939 [24:32:30<45:33:55, 5.71s/it]
35%|███▍ | 15223/43939 [24:32:36<45:40:24, 5.73s/it]
35%|███▍ | 15224/43939 [24:32:41<46:10:09, 5.79s/it]
35%|███▍ | 15225/43939 [24:32:47<46:03:58, 5.78s/it]
35%|███▍ | 15226/43939 [24:32:53<45:52:21, 5.75s/it]
35%|███▍ | 15227/43939 [24:32:59<45:53:25, 5.75s/it]
35%|███▍ | 15228/43939 [24:33:04<45:48:58, 5.74s/it]
35%|███▍ | 15229/43939 [24:33:10<45:42:48, 5.73s/it]
35%|███▍ | 15230/43939 [24:33:16<45:30:33, 5.71s/it]
35%|███▍ | 15231/43939 [24:33:21<45:24:35, 5.69s/it]
35%|███▍ | 15232/43939 [24:33:27<45:27:44, 5.70s/it]
35%|███▍ | 15233/43939 [24:33:33<45:30:20, 5.71s/it]
35%|███▍ | 15234/43939 [24:33:39<45:36:27, 5.72s/it]
35%|███▍ | 15235/43939 [24:33:44<45:33:03, 5.71s/it]
35%|███▍ | 15236/43939 [24:33:50<45:37:20, 5.72s/it]
35%|███▍ | 15237/43939 [24:33:56<45:34:13, 5.72s/it]
35%|███▍ | 15238/43939 [24:34:01<45:32:56, 5.71s/it]
35%|███▍ | 15239/43939 [24:34:07<45:30:46, 5.71s/it]
35%|███▍ | 15240/43939 [24:34:13<45:36:06, 5.72s/it]
35%|███▍ | 15241/43939 [24:34:19<45:41:55, 5.73s/it]
35%|███▍ | 15242/43939 [24:34:24<45:37:38, 5.72s/it]
35%|███▍ | 15243/43939 [24:34:30<45:26:45, 5.70s/it]
35%|███▍ | 15244/43939 [24:34:36<45:23:01, 5.69s/it]
35%|███▍ | 15245/43939 [24:34:42<45:54:43, 5.76s/it]
35%|███▍ | 15246/43939 [24:34:47<45:48:35, 5.75s/it]
35%|███▍ | 15247/43939 [24:34:53<45:35:28, 5.72s/it]
35%|███▍ | 15248/43939 [24:34:59<45:31:24, 5.71s/it]
35%|███▍ | 15249/43939 [24:35:04<45:25:33, 5.70s/it]
35%|███▍ | 15250/43939 [24:35:10<45:22:37, 5.69s/it]
{'loss': 0.9446, 'grad_norm': 1.1577225923538208, 'learning_rate': 4.126510263021572e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.93, 'epoch': 1.39}
35%|███▍ | 15250/43939 [24:35:10<45:22:37, 5.69s/it]
35%|███▍ | 15251/43939 [24:35:16<45:26:47, 5.70s/it]
35%|███▍ | 15252/43939 [24:35:21<45:22:31, 5.69s/it]
35%|███▍ | 15253/43939 [24:35:27<45:34:14, 5.72s/it]
35%|███▍ | 15254/43939 [24:35:33<45:32:28, 5.72s/it]
35%|███▍ | 15255/43939 [24:35:39<45:29:07, 5.71s/it]
35%|███▍ | 15256/43939 [24:35:44<45:24:10, 5.70s/it]
35%|███▍ | 15257/43939 [24:35:50<45:25:11, 5.70s/it]
35%|███▍ | 15258/43939 [24:35:56<45:31:37, 5.71s/it]
35%|███▍ | 15259/43939 [24:36:01<45:25:13, 5.70s/it]
35%|███▍ | 15260/43939 [24:36:07<45:23:13, 5.70s/it]
35%|███▍ | 15261/43939 [24:36:13<45:25:33, 5.70s/it]
35%|███▍ | 15262/43939 [24:36:19<45:30:34, 5.71s/it]
35%|███▍ | 15263/43939 [24:36:24<45:27:08, 5.71s/it]
35%|███▍ | 15264/43939 [24:36:30<45:22:37, 5.70s/it]
35%|███▍ | 15265/43939 [24:36:36<45:19:33, 5.69s/it]
35%|███▍ | 15266/43939 [24:36:42<46:00:16, 5.78s/it]
35%|███▍ | 15267/43939 [24:36:47<45:46:51, 5.75s/it]
35%|███▍ | 15268/43939 [24:36:53<45:38:26, 5.73s/it]
35%|███▍ | 15269/43939 [24:36:59<45:35:51, 5.73s/it]
35%|███▍ | 15270/43939 [24:37:04<45:39:00, 5.73s/it]
35%|███▍ | 15271/43939 [24:37:10<45:32:51, 5.72s/it]
35%|███▍ | 15272/43939 [24:37:16<45:33:44, 5.72s/it]
35%|███▍ | 15273/43939 [24:37:21<45:27:43, 5.71s/it]
35%|███▍ | 15274/43939 [24:37:27<45:24:08, 5.70s/it]
35%|███▍ | 15275/43939 [24:37:33<45:23:56, 5.70s/it]
35%|███▍ | 15276/43939 [24:37:39<45:31:19, 5.72s/it]
35%|███▍ | 15277/43939 [24:37:44<45:28:25, 5.71s/it]
35%|███▍ | 15278/43939 [24:37:50<45:23:50, 5.70s/it]
35%|███▍ | 15279/43939 [24:37:56<45:26:01, 5.71s/it]
35%|███▍ | 15280/43939 [24:38:01<45:24:39, 5.70s/it]
35%|███▍ | 15281/43939 [24:38:07<45:32:54, 5.72s/it]
35%|███▍ | 15282/43939 [24:38:13<45:30:21, 5.72s/it]
35%|███▍ | 15283/43939 [24:38:19<45:31:56, 5.72s/it]
35%|███▍ | 15284/43939 [24:38:24<45:28:49, 5.71s/it]
35%|███▍ | 15285/43939 [24:38:30<45:25:11, 5.71s/it]
35%|███▍ | 15286/43939 [24:38:36<45:29:46, 5.72s/it]
35%|███▍ | 15287/43939 [24:38:42<46:02:26, 5.78s/it]
35%|███▍ | 15288/43939 [24:38:47<45:50:21, 5.76s/it]
35%|███▍ | 15289/43939 [24:38:53<45:46:27, 5.75s/it]
35%|███▍ | 15290/43939 [24:38:59<45:46:39, 5.75s/it]
35%|███▍ | 15291/43939 [24:39:05<45:36:40, 5.73s/it]
35%|███▍ | 15292/43939 [24:39:10<45:31:12, 5.72s/it]
35%|███▍ | 15293/43939 [24:39:16<45:24:12, 5.71s/it]
35%|███▍ | 15294/43939 [24:39:22<45:28:23, 5.71s/it]
35%|███▍ | 15295/43939 [24:39:27<45:25:21, 5.71s/it]
35%|███▍ | 15296/43939 [24:39:33<45:25:47, 5.71s/it]
35%|███▍ | 15297/43939 [24:39:39<45:22:59, 5.70s/it]
35%|███▍ | 15298/43939 [24:39:44<45:20:26, 5.70s/it]
35%|███▍ | 15299/43939 [24:39:50<45:28:12, 5.72s/it]
35%|███▍ | 15300/43939 [24:39:56<45:27:57, 5.72s/it]
{'loss': 0.902, 'grad_norm': 1.3958821296691895, 'learning_rate': 4.1189563000392564e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 466.16, 'epoch': 1.39}
35%|███▍ | 15300/43939 [24:39:56<45:27:57, 5.72s/it]
35%|███▍ | 15301/43939 [24:40:02<45:33:56, 5.73s/it]
35%|███▍ | 15302/43939 [24:40:07<45:31:18, 5.72s/it]
35%|███▍ | 15303/43939 [24:40:13<45:28:48, 5.72s/it]
35%|███▍ | 15304/43939 [24:40:19<45:26:01, 5.71s/it]
35%|███▍ | 15305/43939 [24:40:24<45:23:09, 5.71s/it]
35%|███▍ | 15306/43939 [24:40:30<45:16:57, 5.69s/it]
35%|███▍ | 15307/43939 [24:40:36<45:12:04, 5.68s/it]
35%|███▍ | 15308/43939 [24:40:42<45:50:33, 5.76s/it]
35%|███▍ | 15309/43939 [24:40:47<45:43:00, 5.75s/it]
35%|███▍ | 15310/43939 [24:40:53<45:31:09, 5.72s/it]
35%|███▍ | 15311/43939 [24:40:59<45:27:15, 5.72s/it]
35%|███▍ | 15312/43939 [24:41:05<45:23:58, 5.71s/it]
35%|███▍ | 15313/43939 [24:41:10<45:23:15, 5.71s/it]
35%|███▍ | 15314/43939 [24:41:16<45:23:44, 5.71s/it]
35%|███▍ | 15315/43939 [24:41:22<45:22:13, 5.71s/it]
35%|███▍ | 15316/43939 [24:41:27<45:17:40, 5.70s/it]
35%|███▍ | 15317/43939 [24:41:33<45:22:38, 5.71s/it]
35%|███▍ | 15318/43939 [24:41:39<45:31:04, 5.73s/it]
35%|███▍ | 15319/43939 [24:41:45<45:28:41, 5.72s/it]
35%|███▍ | 15320/43939 [24:41:50<45:25:02, 5.71s/it]
35%|███▍ | 15321/43939 [24:41:56<45:20:12, 5.70s/it]
35%|███▍ | 15322/43939 [24:42:02<45:17:20, 5.70s/it]
35%|███▍ | 15323/43939 [24:42:07<45:14:13, 5.69s/it]
35%|███▍ | 15324/43939 [24:42:13<45:11:13, 5.68s/it]
35%|███▍ | 15325/43939 [24:42:19<45:16:05, 5.70s/it]
35%|███▍ | 15326/43939 [24:42:24<45:15:07, 5.69s/it]
35%|███▍ | 15327/43939 [24:42:30<45:17:35, 5.70s/it]
35%|███▍ | 15328/43939 [24:42:36<45:21:42, 5.71s/it]
35%|███▍ | 15329/43939 [24:42:42<45:46:34, 5.76s/it]
35%|███▍ | 15330/43939 [24:42:47<45:42:07, 5.75s/it]
35%|███▍ | 15331/43939 [24:42:53<45:32:24, 5.73s/it]
35%|███▍ | 15332/43939 [24:42:59<45:30:59, 5.73s/it]
35%|███▍ | 15333/43939 [24:43:04<45:23:58, 5.71s/it]
35%|███▍ | 15334/43939 [24:43:10<45:29:49, 5.73s/it]
35%|███▍ | 15335/43939 [24:43:16<45:22:19, 5.71s/it]
35%|███▍ | 15336/43939 [24:43:22<45:22:05, 5.71s/it]
35%|███▍ | 15337/43939 [24:43:27<45:14:43, 5.69s/it]
35%|███▍ | 15338/43939 [24:43:33<45:16:16, 5.70s/it]
35%|███▍ | 15339/43939 [24:43:39<45:17:45, 5.70s/it]
35%|███▍ | 15340/43939 [24:43:44<45:17:24, 5.70s/it]
35%|███▍ | 15341/43939 [24:43:50<45:20:59, 5.71s/it]
35%|███▍ | 15342/43939 [24:43:56<45:15:58, 5.70s/it]
35%|███▍ | 15343/43939 [24:44:02<45:20:44, 5.71s/it]
35%|███▍ | 15344/43939 [24:44:07<45:31:41, 5.73s/it]
35%|███▍ | 15345/43939 [24:44:13<45:29:08, 5.73s/it]
35%|███▍ | 15346/43939 [24:44:19<45:22:33, 5.71s/it]
35%|███▍ | 15347/43939 [24:44:24<45:27:47, 5.72s/it]
35%|███▍ | 15348/43939 [24:44:30<45:26:37, 5.72s/it]
35%|███▍ | 15349/43939 [24:44:36<45:29:20, 5.73s/it]
35%|███▍ | 15350/43939 [24:44:42<45:22:56, 5.71s/it]
{'loss': 0.9329, 'grad_norm': 1.136844277381897, 'learning_rate': 4.111376794216813e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.98, 'epoch': 1.4}
35%|███▍ | 15350/43939 [24:44:42<45:22:56, 5.71s/it]
35%|███▍ | 15351/43939 [24:44:48<46:05:00, 5.80s/it]
35%|███▍ | 15352/43939 [24:44:53<45:52:34, 5.78s/it]
35%|███▍ | 15353/43939 [24:44:59<45:37:06, 5.74s/it]
35%|███▍ | 15354/43939 [24:45:05<45:34:26, 5.74s/it]
35%|███▍ | 15355/43939 [24:45:10<45:28:47, 5.73s/it]
35%|███▍ | 15356/43939 [24:45:16<45:27:28, 5.73s/it]
35%|███▍ | 15357/43939 [24:45:22<45:26:14, 5.72s/it]
35%|███▍ | 15358/43939 [24:45:28<45:26:28, 5.72s/it]
35%|███▍ | 15359/43939 [24:45:33<45:22:03, 5.71s/it]
35%|███▍ | 15360/43939 [24:45:39<45:17:33, 5.71s/it]
35%|███▍ | 15361/43939 [24:45:45<45:14:57, 5.70s/it]
35%|███▍ | 15362/43939 [24:45:50<45:22:09, 5.72s/it]
35%|███▍ | 15363/43939 [24:45:56<45:20:15, 5.71s/it]
35%|███▍ | 15364/43939 [24:46:02<45:21:00, 5.71s/it]
35%|███▍ | 15365/43939 [24:46:08<45:25:04, 5.72s/it]
35%|███▍ | 15366/43939 [24:46:13<45:14:11, 5.70s/it]
35%|███▍ | 15367/43939 [24:46:19<45:20:32, 5.71s/it]
35%|███▍ | 15368/43939 [24:46:25<45:20:21, 5.71s/it]
35%|███▍ | 15369/43939 [24:46:30<45:20:05, 5.71s/it]
35%|███▍ | 15370/43939 [24:46:36<45:15:04, 5.70s/it]
35%|███▍ | 15371/43939 [24:46:42<45:17:31, 5.71s/it]
35%|███▍ | 15372/43939 [24:46:48<45:52:38, 5.78s/it]
35%|███▍ | 15373/43939 [24:46:53<45:40:56, 5.76s/it]
35%|███▍ | 15374/43939 [24:46:59<45:32:14, 5.74s/it]
35%|███▍ | 15375/43939 [24:47:05<45:27:50, 5.73s/it]
35%|███▍ | 15376/43939 [24:47:11<45:24:55, 5.72s/it]
35%|███▍ | 15377/43939 [24:47:16<45:21:39, 5.72s/it]
35%|███▍ | 15378/43939 [24:47:22<45:20:48, 5.72s/it]
35%|███▌ | 15379/43939 [24:47:28<45:16:51, 5.71s/it][2025-11-29 18:51:41,461] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.94s/it][A
6%|▌ | 4/72 [00:08<02:32, 2.24s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:48, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.69s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.72s/it][A
19%|█▉ | 14/72 [00:35<02:38, 2.73s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:41<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:30, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:25, 2.74s/it][A
28%|██▊ | 20/72 [00:52<02:22, 2.74s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.74s/it][A
38%|███▊ | 27/72 [01:11<02:03, 2.74s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.74s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:33<01:41, 2.74s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:44<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.74s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.74s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.74s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.74s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.74s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.74s/it][A
68%|██████▊ | 49/72 [02:11<01:03, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:52, 2.74s/it][A
75%|███████▌ | 54/72 [02:25<00:49, 2.74s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.74s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.74s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.74s/it][A
81%|████████ | 58/72 [02:36<00:38, 2.74s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.74s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:47<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.74s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.74s/it][A
92%|█████████▏| 66/72 [02:58<00:16, 2.74s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.74s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.74s/it][A
96%|█████████▌| 69/72 [03:06<00:08, 2.74s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.74s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9896542429924011, 'eval_runtime': 199.7817, 'eval_samples_per_second': 5.0, 'eval_steps_per_second': 0.36, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 1.4}
35%|███▌ | 15379/43939 [24:50:47<45:16:51, 5.71s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
35%|███▌ | 15380/43939 [24:50:53<521:01:52, 65.68s/it]
35%|███▌ | 15381/43939 [24:50:59<378:13:10, 47.68s/it]
35%|███▌ | 15382/43939 [24:51:05<278:18:28, 35.08s/it]
35%|███▌ | 15383/43939 [24:51:10<208:26:29, 26.28s/it]
35%|███▌ | 15384/43939 [24:51:16<159:29:56, 20.11s/it]
35%|███▌ | 15385/43939 [24:51:22<125:08:12, 15.78s/it]
35%|███▌ | 15386/43939 [24:51:27<101:05:08, 12.75s/it]
35%|███▌ | 15387/43939 [24:51:33<84:20:12, 10.63s/it]
35%|███▌ | 15388/43939 [24:51:39<72:34:54, 9.15s/it]
35%|███▌ | 15389/43939 [24:51:45<64:25:42, 8.12s/it]
35%|███▌ | 15390/43939 [24:51:50<58:38:20, 7.39s/it]
35%|███▌ | 15391/43939 [24:51:56<54:36:45, 6.89s/it]
35%|███▌ | 15392/43939 [24:52:02<52:17:56, 6.60s/it]
35%|███▌ | 15393/43939 [24:52:07<50:02:40, 6.31s/it]
35%|███▌ | 15394/43939 [24:52:13<48:35:09, 6.13s/it]
35%|███▌ | 15395/43939 [24:52:19<47:43:03, 6.02s/it]
35%|███▌ | 15396/43939 [24:52:25<46:54:23, 5.92s/it]
35%|███▌ | 15397/43939 [24:52:30<46:21:28, 5.85s/it]
35%|███▌ | 15398/43939 [24:52:36<45:59:37, 5.80s/it]
35%|███▌ | 15399/43939 [24:52:42<45:43:54, 5.77s/it]
35%|███▌ | 15400/43939 [24:52:47<45:40:35, 5.76s/it]
{'loss': 0.8975, 'grad_norm': 1.1986658573150635, 'learning_rate': 4.103771865138757e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 191.15, 'epoch': 1.4}
35%|███▌ | 15400/43939 [24:52:47<45:40:35, 5.76s/it]
35%|███▌ | 15401/43939 [24:52:53<45:35:02, 5.75s/it]
35%|███▌ | 15402/43939 [24:52:59<45:26:41, 5.73s/it]
35%|███▌ | 15403/43939 [24:53:05<45:24:42, 5.73s/it]
35%|███▌ | 15404/43939 [24:53:10<45:22:13, 5.72s/it]
35%|███▌ | 15405/43939 [24:53:16<45:13:51, 5.71s/it]
35%|███▌ | 15406/43939 [24:53:22<45:10:07, 5.70s/it]
35%|███▌ | 15407/43939 [24:53:27<45:09:34, 5.70s/it]
35%|███▌ | 15408/43939 [24:53:33<45:06:41, 5.69s/it]
35%|███▌ | 15409/43939 [24:53:39<45:04:20, 5.69s/it]
35%|███▌ | 15410/43939 [24:53:44<45:00:57, 5.68s/it]
35%|███▌ | 15411/43939 [24:53:50<45:07:44, 5.69s/it]
35%|███▌ | 15412/43939 [24:53:56<45:12:30, 5.71s/it]
35%|███▌ | 15413/43939 [24:54:02<45:43:33, 5.77s/it]
35%|███▌ | 15414/43939 [24:54:07<45:31:25, 5.75s/it]
35%|███▌ | 15415/43939 [24:54:13<45:27:43, 5.74s/it]
35%|███▌ | 15416/43939 [24:54:19<45:24:03, 5.73s/it]
35%|███▌ | 15417/43939 [24:54:25<45:23:47, 5.73s/it]
35%|███▌ | 15418/43939 [24:54:30<45:20:52, 5.72s/it]
35%|███▌ | 15419/43939 [24:54:36<45:24:46, 5.73s/it]
35%|███▌ | 15420/43939 [24:54:42<45:27:20, 5.74s/it]
35%|███▌ | 15421/43939 [24:54:48<45:22:03, 5.73s/it]
35%|███▌ | 15422/43939 [24:54:53<45:14:55, 5.71s/it]
35%|███▌ | 15423/43939 [24:54:59<45:10:39, 5.70s/it]
35%|███▌ | 15424/43939 [24:55:05<45:07:13, 5.70s/it]
35%|███▌ | 15425/43939 [24:55:10<45:14:41, 5.71s/it]
35%|███▌ | 15426/43939 [24:55:16<45:20:47, 5.73s/it]
35%|███▌ | 15427/43939 [24:55:22<45:29:40, 5.74s/it]
35%|███▌ | 15428/43939 [24:55:28<45:22:40, 5.73s/it]
35%|███▌ | 15429/43939 [24:55:33<45:19:10, 5.72s/it]
35%|███▌ | 15430/43939 [24:55:39<45:15:03, 5.71s/it]
35%|███▌ | 15431/43939 [24:55:45<45:11:48, 5.71s/it]
35%|███▌ | 15432/43939 [24:55:50<45:13:39, 5.71s/it]
35%|███▌ | 15433/43939 [24:55:56<45:12:59, 5.71s/it]
35%|███▌ | 15434/43939 [24:56:02<45:12:29, 5.71s/it]
35%|███▌ | 15435/43939 [24:56:08<45:36:15, 5.76s/it]
35%|███▌ | 15436/43939 [24:56:13<45:22:52, 5.73s/it]
35%|███▌ | 15437/43939 [24:56:19<45:15:02, 5.72s/it]
35%|███▌ | 15438/43939 [24:56:25<45:16:52, 5.72s/it]
35%|███▌ | 15439/43939 [24:56:30<45:05:13, 5.70s/it]
35%|███▌ | 15440/43939 [24:56:36<45:02:33, 5.69s/it]
35%|███▌ | 15441/43939 [24:56:42<44:58:28, 5.68s/it]
35%|███▌ | 15442/43939 [24:56:47<44:55:09, 5.67s/it]
35%|███▌ | 15443/43939 [24:56:53<44:53:23, 5.67s/it]
35%|███▌ | 15444/43939 [24:56:59<44:58:31, 5.68s/it]
35%|███▌ | 15445/43939 [24:57:04<45:03:24, 5.69s/it]
35%|███▌ | 15446/43939 [24:57:10<45:07:29, 5.70s/it]
35%|███▌ | 15447/43939 [24:57:16<45:11:06, 5.71s/it]
35%|███▌ | 15448/43939 [24:57:22<45:13:49, 5.72s/it]
35%|███▌ | 15449/43939 [24:57:27<45:16:04, 5.72s/it]
35%|███▌ | 15450/43939 [24:57:33<45:09:12, 5.71s/it]
{'loss': 0.9352, 'grad_norm': 1.2396776676177979, 'learning_rate': 4.096141632790717e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.13, 'epoch': 1.41}
35%|███▌ | 15450/43939 [24:57:33<45:09:12, 5.71s/it]
35%|███▌ | 15451/43939 [24:57:39<45:15:28, 5.72s/it]
35%|███▌ | 15452/43939 [24:57:44<45:13:27, 5.72s/it]
35%|███▌ | 15453/43939 [24:57:50<45:10:03, 5.71s/it]
35%|███▌ | 15454/43939 [24:57:56<45:05:02, 5.70s/it]
35%|███▌ | 15455/43939 [24:58:02<45:12:28, 5.71s/it]
35%|███▌ | 15456/43939 [24:58:07<45:38:11, 5.77s/it]
35%|███▌ | 15457/43939 [24:58:13<45:34:02, 5.76s/it]
35%|███▌ | 15458/43939 [24:58:19<45:22:13, 5.73s/it]
35%|███▌ | 15459/43939 [24:58:25<45:17:43, 5.73s/it]
35%|███▌ | 15460/43939 [24:58:30<45:17:44, 5.73s/it]
35%|███▌ | 15461/43939 [24:58:36<45:10:02, 5.71s/it]
35%|███▌ | 15462/43939 [24:58:42<45:09:15, 5.71s/it]
35%|███▌ | 15463/43939 [24:58:47<45:16:47, 5.72s/it]
35%|███▌ | 15464/43939 [24:58:53<45:16:37, 5.72s/it]
35%|███▌ | 15465/43939 [24:58:59<45:18:23, 5.73s/it]
35%|███▌ | 15466/43939 [24:59:05<45:16:51, 5.73s/it]
35%|███▌ | 15467/43939 [24:59:10<45:14:47, 5.72s/it]
35%|███▌ | 15468/43939 [24:59:16<45:08:51, 5.71s/it]
35%|███▌ | 15469/43939 [24:59:22<45:13:43, 5.72s/it]
35%|███▌ | 15470/43939 [24:59:28<45:14:03, 5.72s/it]
35%|███▌ | 15471/43939 [24:59:33<45:13:57, 5.72s/it]
35%|███▌ | 15472/43939 [24:59:39<45:10:54, 5.71s/it]
35%|███▌ | 15473/43939 [24:59:45<45:07:45, 5.71s/it]
35%|███▌ | 15474/43939 [24:59:50<45:08:53, 5.71s/it]
35%|███▌ | 15475/43939 [24:59:56<45:06:11, 5.70s/it]
35%|███▌ | 15476/43939 [25:00:02<45:06:44, 5.71s/it]
35%|███▌ | 15477/43939 [25:00:08<45:39:49, 5.78s/it]
35%|███▌ | 15478/43939 [25:00:13<45:28:41, 5.75s/it]
35%|███▌ | 15479/43939 [25:00:19<45:22:19, 5.74s/it]
35%|███▌ | 15480/43939 [25:00:25<45:23:54, 5.74s/it]
35%|███▌ | 15481/43939 [25:00:31<45:24:42, 5.74s/it]
35%|███▌ | 15482/43939 [25:00:36<45:22:02, 5.74s/it]
35%|███▌ | 15483/43939 [25:00:42<45:21:35, 5.74s/it]
35%|███▌ | 15484/43939 [25:00:48<45:17:08, 5.73s/it]
35%|███▌ | 15485/43939 [25:00:53<45:10:06, 5.71s/it]
35%|███▌ | 15486/43939 [25:00:59<45:12:02, 5.72s/it]
35%|███▌ | 15487/43939 [25:01:05<45:09:09, 5.71s/it]
35%|███▌ | 15488/43939 [25:01:11<45:07:54, 5.71s/it]
35%|███▌ | 15489/43939 [25:01:16<45:12:57, 5.72s/it]
35%|███▌ | 15490/43939 [25:01:22<45:10:49, 5.72s/it]
35%|███▌ | 15491/43939 [25:01:28<45:07:16, 5.71s/it]
35%|███▌ | 15492/43939 [25:01:33<45:10:36, 5.72s/it]
35%|███▌ | 15493/43939 [25:01:39<45:00:50, 5.70s/it]
35%|███▌ | 15494/43939 [25:01:45<45:03:35, 5.70s/it]
35%|███▌ | 15495/43939 [25:01:50<44:55:58, 5.69s/it]
35%|███▌ | 15496/43939 [25:01:56<44:55:31, 5.69s/it]
35%|███▌ | 15497/43939 [25:02:02<45:08:40, 5.71s/it]
35%|███▌ | 15498/43939 [25:02:08<45:38:51, 5.78s/it]
35%|███▌ | 15499/43939 [25:02:14<45:27:36, 5.75s/it]
35%|███▌ | 15500/43939 [25:02:19<45:14:45, 5.73s/it]
{'loss': 0.9341, 'grad_norm': 1.1720110177993774, 'learning_rate': 4.0884862175575376e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 511.5, 'epoch': 1.41}
35%|███▌ | 15500/43939 [25:02:19<45:14:45, 5.73s/it]
35%|███▌ | 15501/43939 [25:02:25<45:06:58, 5.71s/it]
35%|███▌ | 15502/43939 [25:02:31<45:09:01, 5.72s/it]
35%|███▌ | 15503/43939 [25:02:36<45:07:12, 5.71s/it]
35%|███▌ | 15504/43939 [25:02:42<45:08:28, 5.72s/it]
35%|███▌ | 15505/43939 [25:02:48<45:15:52, 5.73s/it]
35%|███▌ | 15506/43939 [25:02:54<45:15:34, 5.73s/it]
35%|███▌ | 15507/43939 [25:02:59<45:10:17, 5.72s/it]
35%|███▌ | 15508/43939 [25:03:05<45:03:03, 5.70s/it]
35%|███▌ | 15509/43939 [25:03:11<45:04:47, 5.71s/it]
35%|███▌ | 15510/43939 [25:03:16<44:58:46, 5.70s/it]
35%|███▌ | 15511/43939 [25:03:22<45:02:27, 5.70s/it]
35%|███▌ | 15512/43939 [25:03:28<45:02:14, 5.70s/it]
35%|███▌ | 15513/43939 [25:03:33<44:58:42, 5.70s/it]
35%|███▌ | 15514/43939 [25:03:39<45:06:45, 5.71s/it]
35%|███▌ | 15515/43939 [25:03:45<44:57:52, 5.69s/it]
35%|███▌ | 15516/43939 [25:03:51<45:03:53, 5.71s/it]
35%|███▌ | 15517/43939 [25:03:56<44:58:00, 5.70s/it]
35%|███▌ | 15518/43939 [25:04:02<45:05:01, 5.71s/it]
35%|███▌ | 15519/43939 [25:04:08<45:35:01, 5.77s/it]
35%|███▌ | 15520/43939 [25:04:14<45:21:06, 5.74s/it]
35%|███▌ | 15521/43939 [25:04:19<45:15:53, 5.73s/it]
35%|███▌ | 15522/43939 [25:04:25<45:17:42, 5.74s/it]
35%|███▌ | 15523/43939 [25:04:31<45:17:56, 5.74s/it]
35%|███▌ | 15524/43939 [25:04:36<45:13:49, 5.73s/it]
35%|███▌ | 15525/43939 [25:04:42<45:05:16, 5.71s/it]
35%|███▌ | 15526/43939 [25:04:48<45:02:20, 5.71s/it]
35%|███▌ | 15527/43939 [25:04:54<45:03:28, 5.71s/it]
35%|███▌ | 15528/43939 [25:04:59<45:09:44, 5.72s/it]
35%|███▌ | 15529/43939 [25:05:05<45:05:21, 5.71s/it]
35%|███▌ | 15530/43939 [25:05:11<45:02:06, 5.71s/it]
35%|███▌ | 15531/43939 [25:05:16<44:59:30, 5.70s/it]
35%|███▌ | 15532/43939 [25:05:22<45:00:49, 5.70s/it]
35%|███▌ | 15533/43939 [25:05:28<44:59:21, 5.70s/it]
35%|███▌ | 15534/43939 [25:05:33<45:01:55, 5.71s/it]
35%|███▌ | 15535/43939 [25:05:39<44:54:22, 5.69s/it]
35%|███▌ | 15536/43939 [25:05:45<44:48:25, 5.68s/it]
35%|███▌ | 15537/43939 [25:05:51<44:56:52, 5.70s/it]
35%|███▌ | 15538/43939 [25:05:56<44:53:22, 5.69s/it]
35%|███▌ | 15539/43939 [25:06:02<44:55:44, 5.70s/it]
35%|███▌ | 15540/43939 [25:06:08<45:28:28, 5.76s/it]
35%|███▌ | 15541/43939 [25:06:14<45:18:52, 5.74s/it]
35%|███▌ | 15542/43939 [25:06:19<45:09:05, 5.72s/it]
35%|███▌ | 15543/43939 [25:06:25<45:02:41, 5.71s/it]
35%|███▌ | 15544/43939 [25:06:31<45:07:17, 5.72s/it]
35%|███▌ | 15545/43939 [25:06:36<45:02:54, 5.71s/it]
35%|███▌ | 15546/43939 [25:06:42<45:07:20, 5.72s/it]
35%|███▌ | 15547/43939 [25:06:48<45:03:52, 5.71s/it]
35%|███▌ | 15548/43939 [25:06:53<44:59:54, 5.71s/it]
35%|███▌ | 15549/43939 [25:06:59<45:01:58, 5.71s/it]
35%|███▌ | 15550/43939 [25:07:05<44:57:22, 5.70s/it]
{'loss': 0.932, 'grad_norm': 1.0548924207687378, 'learning_rate': 4.0808057402213855e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.27, 'epoch': 1.42}
35%|███▌ | 15550/43939 [25:07:05<44:57:22, 5.70s/it]
35%|███▌ | 15551/43939 [25:07:11<45:00:03, 5.71s/it]
35%|███▌ | 15552/43939 [25:07:16<44:55:46, 5.70s/it]
35%|███▌ | 15553/43939 [25:07:22<44:53:18, 5.69s/it]
35%|███▌ | 15554/43939 [25:07:28<44:52:00, 5.69s/it]
35%|███▌ | 15555/43939 [25:07:33<44:54:26, 5.70s/it]
35%|███▌ | 15556/43939 [25:07:39<44:56:40, 5.70s/it]
35%|███▌ | 15557/43939 [25:07:45<44:55:17, 5.70s/it]
35%|███▌ | 15558/43939 [25:07:50<44:52:33, 5.69s/it]
35%|███▌ | 15559/43939 [25:07:56<44:53:34, 5.69s/it]
35%|███▌ | 15560/43939 [25:08:02<44:52:29, 5.69s/it]
35%|███▌ | 15561/43939 [25:08:08<45:24:29, 5.76s/it]
35%|███▌ | 15562/43939 [25:08:13<45:20:10, 5.75s/it]
35%|███▌ | 15563/43939 [25:08:19<45:13:14, 5.74s/it]
35%|███▌ | 15564/43939 [25:08:25<45:05:11, 5.72s/it]
35%|███▌ | 15565/43939 [25:08:31<45:02:01, 5.71s/it]
35%|███▌ | 15566/43939 [25:08:36<44:59:27, 5.71s/it]
35%|███▌ | 15567/43939 [25:08:42<45:01:51, 5.71s/it]
35%|███▌ | 15568/43939 [25:08:48<44:50:42, 5.69s/it]
35%|███▌ | 15569/43939 [25:08:53<44:51:00, 5.69s/it]
35%|███▌ | 15570/43939 [25:08:59<44:49:53, 5.69s/it]
35%|███▌ | 15571/43939 [25:09:05<45:03:52, 5.72s/it]
35%|███▌ | 15572/43939 [25:09:10<45:01:29, 5.71s/it]
35%|███▌ | 15573/43939 [25:09:16<45:08:20, 5.73s/it]
35%|███▌ | 15574/43939 [25:09:22<44:58:45, 5.71s/it]
35%|███▌ | 15575/43939 [25:09:28<44:58:59, 5.71s/it]
35%|███▌ | 15576/43939 [25:09:33<45:03:44, 5.72s/it]
35%|███▌ | 15577/43939 [25:09:39<45:06:58, 5.73s/it]
35%|███▌ | 15578/43939 [25:09:45<45:09:11, 5.73s/it]
35%|███▌ | 15579/43939 [25:09:50<44:57:43, 5.71s/it]
35%|███▌ | 15580/43939 [25:09:56<45:02:38, 5.72s/it]
35%|███▌ | 15581/43939 [25:10:02<45:06:46, 5.73s/it]
35%|███▌ | 15582/43939 [25:10:08<44:59:27, 5.71s/it]
35%|███▌ | 15583/43939 [25:10:14<45:29:40, 5.78s/it]
35%|███▌ | 15584/43939 [25:10:19<45:18:55, 5.75s/it]
35%|███▌ | 15585/43939 [25:10:25<45:14:58, 5.75s/it]
35%|███▌ | 15586/43939 [25:10:31<45:05:16, 5.72s/it]
35%|███▌ | 15587/43939 [25:10:36<45:01:56, 5.72s/it]
35%|███▌ | 15588/43939 [25:10:42<44:57:37, 5.71s/it]
35%|███▌ | 15589/43939 [25:10:48<44:50:05, 5.69s/it]
35%|███▌ | 15590/43939 [25:10:53<44:51:11, 5.70s/it]
35%|███▌ | 15591/43939 [25:10:59<44:51:03, 5.70s/it]
35%|███▌ | 15592/43939 [25:11:05<44:53:12, 5.70s/it]
35%|███▌ | 15593/43939 [25:11:11<44:59:50, 5.71s/it]
35%|███▌ | 15594/43939 [25:11:16<44:59:31, 5.71s/it]
35%|███▌ | 15595/43939 [25:11:22<44:53:37, 5.70s/it]
35%|███▌ | 15596/43939 [25:11:28<44:49:01, 5.69s/it]
35%|███▌ | 15597/43939 [25:11:33<44:53:47, 5.70s/it]
35%|███▌ | 15598/43939 [25:11:39<44:50:32, 5.70s/it]
36%|███▌ | 15599/43939 [25:11:45<44:49:29, 5.69s/it]
36%|███▌ | 15600/43939 [25:11:50<44:49:44, 5.69s/it]
{'loss': 0.9466, 'grad_norm': 1.3402786254882812, 'learning_rate': 4.073100321959838e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.6, 'epoch': 1.42}
36%|███▌ | 15600/43939 [25:11:50<44:49:44, 5.69s/it]
36%|███▌ | 15601/43939 [25:11:56<44:49:24, 5.69s/it]
36%|███▌ | 15602/43939 [25:12:02<44:49:24, 5.69s/it]
36%|███▌ | 15603/43939 [25:12:08<44:51:57, 5.70s/it]
36%|███▌ | 15604/43939 [25:12:13<45:18:05, 5.76s/it]
36%|███▌ | 15605/43939 [25:12:19<45:08:14, 5.73s/it]
36%|███▌ | 15606/43939 [25:12:25<45:00:46, 5.72s/it]
36%|███▌ | 15607/43939 [25:12:31<45:06:37, 5.73s/it]
36%|███▌ | 15608/43939 [25:12:36<45:07:54, 5.73s/it]
36%|███▌ | 15609/43939 [25:12:42<45:02:15, 5.72s/it]
36%|███▌ | 15610/43939 [25:12:48<44:58:41, 5.72s/it]
36%|███▌ | 15611/43939 [25:12:53<45:00:13, 5.72s/it]
36%|███▌ | 15612/43939 [25:12:59<44:59:41, 5.72s/it]
36%|███▌ | 15613/43939 [25:13:05<44:56:34, 5.71s/it]
36%|███▌ | 15614/43939 [25:13:11<44:57:55, 5.71s/it]
36%|███▌ | 15615/43939 [25:13:16<44:57:58, 5.72s/it]
36%|███▌ | 15616/43939 [25:13:22<45:03:07, 5.73s/it]
36%|███▌ | 15617/43939 [25:13:28<44:56:01, 5.71s/it]
36%|███▌ | 15618/43939 [25:13:33<44:58:54, 5.72s/it]
36%|███▌ | 15619/43939 [25:13:39<44:55:14, 5.71s/it]
36%|███▌ | 15620/43939 [25:13:45<44:57:14, 5.71s/it]
36%|███▌ | 15621/43939 [25:13:51<45:02:40, 5.73s/it]
36%|███▌ | 15622/43939 [25:13:56<45:00:26, 5.72s/it]
36%|███▌ | 15623/43939 [25:14:02<45:00:18, 5.72s/it]
36%|███▌ | 15624/43939 [25:14:08<45:00:42, 5.72s/it]
36%|███▌ | 15625/43939 [25:14:14<45:23:04, 5.77s/it]
36%|███▌ | 15626/43939 [25:14:19<45:12:21, 5.75s/it]
36%|███▌ | 15627/43939 [25:14:25<45:09:49, 5.74s/it]
36%|███▌ | 15628/43939 [25:14:31<45:04:37, 5.73s/it]
36%|███▌ | 15629/43939 [25:14:36<44:57:31, 5.72s/it]
36%|███▌ | 15630/43939 [25:14:42<44:50:19, 5.70s/it]
36%|███▌ | 15631/43939 [25:14:48<44:53:18, 5.71s/it]
36%|███▌ | 15632/43939 [25:14:54<44:49:04, 5.70s/it]
36%|███▌ | 15633/43939 [25:14:59<44:51:48, 5.71s/it]
36%|███▌ | 15634/43939 [25:15:05<44:48:08, 5.70s/it]
36%|███▌ | 15635/43939 [25:15:11<44:51:46, 5.71s/it]
36%|███▌ | 15636/43939 [25:15:16<44:55:56, 5.72s/it]
36%|███▌ | 15637/43939 [25:15:22<44:53:13, 5.71s/it]
36%|███▌ | 15638/43939 [25:15:28<44:45:06, 5.69s/it]
36%|███▌ | 15639/43939 [25:15:33<44:40:56, 5.68s/it]
36%|███▌ | 15640/43939 [25:15:39<44:42:16, 5.69s/it]
36%|███▌ | 15641/43939 [25:15:45<44:47:19, 5.70s/it]
36%|███▌ | 15642/43939 [25:15:51<44:48:39, 5.70s/it]
36%|███▌ | 15643/43939 [25:15:56<44:49:05, 5.70s/it]
36%|███▌ | 15644/43939 [25:16:02<44:49:39, 5.70s/it]
36%|███▌ | 15645/43939 [25:16:08<44:47:48, 5.70s/it]
36%|███▌ | 15646/43939 [25:16:13<44:50:27, 5.71s/it]
36%|███▌ | 15647/43939 [25:16:19<45:22:26, 5.77s/it]
36%|███▌ | 15648/43939 [25:16:25<45:12:43, 5.75s/it]
36%|███▌ | 15649/43939 [25:16:31<45:04:37, 5.74s/it]
36%|███▌ | 15650/43939 [25:16:36<44:57:49, 5.72s/it]
{'loss': 0.9325, 'grad_norm': 1.2365429401397705, 'learning_rate': 4.065370084343978e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.0, 'epoch': 1.42}
36%|███▌ | 15650/43939 [25:16:36<44:57:49, 5.72s/it]
36%|███▌ | 15651/43939 [25:16:42<44:59:01, 5.72s/it]
36%|███▌ | 15652/43939 [25:16:48<44:58:28, 5.72s/it]
36%|███▌ | 15653/43939 [25:16:54<44:56:03, 5.72s/it]
36%|███▌ | 15654/43939 [25:16:59<44:53:20, 5.71s/it]
36%|███▌ | 15655/43939 [25:17:05<44:55:32, 5.72s/it]
36%|███▌ | 15656/43939 [25:17:11<44:52:37, 5.71s/it]
36%|███▌ | 15657/43939 [25:17:16<44:51:53, 5.71s/it]
36%|███▌ | 15658/43939 [25:17:22<44:54:56, 5.72s/it]
36%|███▌ | 15659/43939 [25:17:28<44:48:17, 5.70s/it]
36%|███▌ | 15660/43939 [25:17:33<44:46:00, 5.70s/it]
36%|███▌ | 15661/43939 [25:17:39<44:43:57, 5.69s/it]
36%|███▌ | 15662/43939 [25:17:45<44:41:19, 5.69s/it]
36%|███▌ | 15663/43939 [25:17:51<44:44:09, 5.70s/it]
36%|███▌ | 15664/43939 [25:17:56<44:42:49, 5.69s/it]
36%|███▌ | 15665/43939 [25:18:02<44:42:52, 5.69s/it]
36%|███▌ | 15666/43939 [25:18:08<44:45:00, 5.70s/it]
36%|███▌ | 15667/43939 [25:18:13<44:44:00, 5.70s/it]
36%|███▌ | 15668/43939 [25:18:19<45:21:02, 5.77s/it]
36%|███▌ | 15669/43939 [25:18:25<45:08:39, 5.75s/it]
36%|███▌ | 15670/43939 [25:18:31<45:01:47, 5.73s/it]
36%|███▌ | 15671/43939 [25:18:36<44:57:48, 5.73s/it]
36%|███▌ | 15672/43939 [25:18:42<44:51:49, 5.71s/it]
36%|███▌ | 15673/43939 [25:18:48<44:53:12, 5.72s/it]
36%|███▌ | 15674/43939 [25:18:53<44:50:11, 5.71s/it]
36%|███▌ | 15675/43939 [25:18:59<44:44:35, 5.70s/it]
36%|███▌ | 15676/43939 [25:19:05<44:45:47, 5.70s/it]
36%|███▌ | 15677/43939 [25:19:11<44:46:39, 5.70s/it]
36%|███▌ | 15678/43939 [25:19:16<44:44:46, 5.70s/it]
36%|███▌ | 15679/43939 [25:19:22<44:41:28, 5.69s/it]
36%|███▌ | 15680/43939 [25:19:28<44:42:46, 5.70s/it]
36%|███▌ | 15681/43939 [25:19:33<44:48:17, 5.71s/it]
36%|███▌ | 15682/43939 [25:19:39<44:46:55, 5.71s/it]
36%|███▌ | 15683/43939 [25:19:45<44:45:14, 5.70s/it]
36%|███▌ | 15684/43939 [25:19:50<44:42:12, 5.70s/it]
36%|███▌ | 15685/43939 [25:19:56<44:45:04, 5.70s/it]
36%|███▌ | 15686/43939 [25:20:02<44:46:01, 5.70s/it]
36%|███▌ | 15687/43939 [25:20:08<44:40:39, 5.69s/it]
36%|███▌ | 15688/43939 [25:20:13<44:40:27, 5.69s/it]
36%|███▌ | 15689/43939 [25:20:19<45:23:25, 5.78s/it]
36%|███▌ | 15690/43939 [25:20:25<45:11:28, 5.76s/it]
36%|███▌ | 15691/43939 [25:20:31<45:05:27, 5.75s/it]
36%|███▌ | 15692/43939 [25:20:36<45:03:29, 5.74s/it]
36%|███▌ | 15693/43939 [25:20:42<45:03:25, 5.74s/it]
36%|███▌ | 15694/43939 [25:20:48<45:05:04, 5.75s/it]
36%|███▌ | 15695/43939 [25:20:54<44:57:26, 5.73s/it]
36%|███▌ | 15696/43939 [25:20:59<44:53:03, 5.72s/it]
36%|███▌ | 15697/43939 [25:21:05<44:50:02, 5.71s/it]
36%|███▌ | 15698/43939 [25:21:11<44:46:48, 5.71s/it]
36%|███▌ | 15699/43939 [25:21:16<44:45:00, 5.70s/it]
36%|███▌ | 15700/43939 [25:21:22<44:41:36, 5.70s/it]
{'loss': 0.9391, 'grad_norm': 1.2400281429290771, 'learning_rate': 4.057615149336468e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.35, 'epoch': 1.43}
36%|███▌ | 15700/43939 [25:21:22<44:41:36, 5.70s/it]
36%|███▌ | 15701/43939 [25:21:28<44:44:16, 5.70s/it]
36%|███▌ | 15702/43939 [25:21:33<44:36:50, 5.69s/it]
36%|███▌ | 15703/43939 [25:21:39<44:42:34, 5.70s/it]
36%|███▌ | 15704/43939 [25:21:45<44:42:50, 5.70s/it]
36%|███▌ | 15705/43939 [25:21:51<44:39:40, 5.69s/it]
36%|███▌ | 15706/43939 [25:21:56<44:44:08, 5.70s/it]
36%|███▌ | 15707/43939 [25:22:02<44:39:35, 5.69s/it]
36%|███▌ | 15708/43939 [25:22:08<44:38:45, 5.69s/it]
36%|███▌ | 15709/43939 [25:22:13<44:39:43, 5.70s/it]
36%|███▌ | 15710/43939 [25:22:19<45:19:55, 5.78s/it]
36%|███▌ | 15711/43939 [25:22:25<45:08:23, 5.76s/it]
36%|███▌ | 15712/43939 [25:22:31<45:02:24, 5.74s/it]
36%|███▌ | 15713/43939 [25:22:36<45:00:21, 5.74s/it]
36%|███▌ | 15714/43939 [25:22:42<44:58:51, 5.74s/it]
36%|███▌ | 15715/43939 [25:22:48<44:52:12, 5.72s/it]
36%|███▌ | 15716/43939 [25:22:54<44:47:13, 5.71s/it]
36%|███▌ | 15717/43939 [25:22:59<44:46:58, 5.71s/it]
36%|███▌ | 15718/43939 [25:23:05<44:43:33, 5.71s/it]
36%|███▌ | 15719/43939 [25:23:11<44:37:45, 5.69s/it]
36%|███▌ | 15720/43939 [25:23:16<44:49:38, 5.72s/it]
36%|███▌ | 15721/43939 [25:23:22<44:43:58, 5.71s/it]
36%|███▌ | 15722/43939 [25:23:28<44:45:44, 5.71s/it]
36%|███▌ | 15723/43939 [25:23:33<44:46:45, 5.71s/it]
36%|███▌ | 15724/43939 [25:23:39<44:48:04, 5.72s/it]
36%|███▌ | 15725/43939 [25:23:45<44:48:24, 5.72s/it]
36%|███▌ | 15726/43939 [25:23:51<44:46:50, 5.71s/it]
36%|███▌ | 15727/43939 [25:23:56<44:49:20, 5.72s/it]
36%|███▌ | 15728/43939 [25:24:02<44:49:37, 5.72s/it]
36%|███▌ | 15729/43939 [25:24:08<44:46:35, 5.71s/it]
36%|███▌ | 15730/43939 [25:24:13<44:38:38, 5.70s/it]
36%|███▌ | 15731/43939 [25:24:19<44:44:18, 5.71s/it]
36%|███▌ | 15732/43939 [25:24:25<45:21:21, 5.79s/it]
36%|███▌ | 15733/43939 [25:24:31<45:08:09, 5.76s/it]
36%|███▌ | 15734/43939 [25:24:37<44:58:38, 5.74s/it]
36%|███▌ | 15735/43939 [25:24:42<44:48:57, 5.72s/it]
36%|███▌ | 15736/43939 [25:24:48<44:45:50, 5.71s/it]
36%|███▌ | 15737/43939 [25:24:54<44:43:15, 5.71s/it]
36%|███▌ | 15738/43939 [25:24:59<44:43:44, 5.71s/it]
36%|███▌ | 15739/43939 [25:25:05<44:38:30, 5.70s/it]
36%|███▌ | 15740/43939 [25:25:11<44:42:58, 5.71s/it]
36%|███▌ | 15741/43939 [25:25:16<44:40:31, 5.70s/it]
36%|███▌ | 15742/43939 [25:25:22<44:36:58, 5.70s/it]
36%|███▌ | 15743/43939 [25:25:28<44:38:11, 5.70s/it]
36%|███▌ | 15744/43939 [25:25:34<44:37:51, 5.70s/it]
36%|███▌ | 15745/43939 [25:25:39<44:38:47, 5.70s/it]
36%|███▌ | 15746/43939 [25:25:45<44:37:09, 5.70s/it]
36%|███▌ | 15747/43939 [25:25:51<44:42:32, 5.71s/it]
36%|███▌ | 15748/43939 [25:25:56<44:53:06, 5.73s/it]
36%|███▌ | 15749/43939 [25:26:02<44:41:08, 5.71s/it]
36%|███▌ | 15750/43939 [25:26:08<44:53:33, 5.73s/it]
{'loss': 0.9449, 'grad_norm': 1.0249978303909302, 'learning_rate': 4.0498356392896305e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.7, 'epoch': 1.43}
36%|███▌ | 15750/43939 [25:26:08<44:53:33, 5.73s/it]
36%|███▌ | 15751/43939 [25:26:14<44:46:09, 5.72s/it]
36%|███▌ | 15752/43939 [25:26:19<44:37:38, 5.70s/it]
36%|███▌ | 15753/43939 [25:26:25<45:17:53, 5.79s/it]
36%|███▌ | 15754/43939 [25:26:31<45:13:35, 5.78s/it]
36%|███▌ | 15755/43939 [25:26:37<45:05:41, 5.76s/it]
36%|███▌ | 15756/43939 [25:26:42<45:00:54, 5.75s/it]
36%|███▌ | 15757/43939 [25:26:48<44:57:59, 5.74s/it]
36%|███▌ | 15758/43939 [25:26:54<44:55:09, 5.74s/it]
36%|███▌ | 15759/43939 [25:27:00<44:41:57, 5.71s/it]
36%|███▌ | 15760/43939 [25:27:05<44:52:36, 5.73s/it]
36%|███▌ | 15761/43939 [25:27:11<44:54:36, 5.74s/it]
36%|███▌ | 15762/43939 [25:27:17<44:54:55, 5.74s/it]
36%|███▌ | 15763/43939 [25:27:22<44:46:22, 5.72s/it]
36%|███▌ | 15764/43939 [25:27:28<44:37:49, 5.70s/it]
36%|███▌ | 15765/43939 [25:27:34<44:36:57, 5.70s/it]
36%|███▌ | 15766/43939 [25:27:40<44:39:44, 5.71s/it]
36%|███▌ | 15767/43939 [25:27:45<44:42:29, 5.71s/it]
36%|███▌ | 15768/43939 [25:27:51<44:43:10, 5.71s/it]
36%|███▌ | 15769/43939 [25:27:57<44:37:44, 5.70s/it]
36%|███▌ | 15770/43939 [25:28:02<44:48:25, 5.73s/it]
36%|███▌ | 15771/43939 [25:28:08<44:48:25, 5.73s/it]
36%|███▌ | 15772/43939 [25:28:14<44:46:50, 5.72s/it]
36%|███▌ | 15773/43939 [25:28:20<44:43:40, 5.72s/it]
36%|███▌ | 15774/43939 [25:28:26<45:14:00, 5.78s/it]
36%|███▌ | 15775/43939 [25:28:31<45:04:01, 5.76s/it]
36%|███▌ | 15776/43939 [25:28:37<45:00:16, 5.75s/it]
36%|███▌ | 15777/43939 [25:28:43<44:50:02, 5.73s/it]
36%|███▌ | 15778/43939 [25:28:48<44:44:05, 5.72s/it]
36%|███▌ | 15779/43939 [25:28:54<44:41:55, 5.71s/it]
36%|███▌ | 15780/43939 [25:29:00<44:38:12, 5.71s/it]
36%|███▌ | 15781/43939 [25:29:05<44:42:06, 5.72s/it]
36%|███▌ | 15782/43939 [25:29:11<44:43:31, 5.72s/it]
36%|███▌ | 15783/43939 [25:29:17<44:39:02, 5.71s/it]
36%|███▌ | 15784/43939 [25:29:23<44:41:01, 5.71s/it]
36%|███▌ | 15785/43939 [25:29:28<44:40:30, 5.71s/it]
36%|███▌ | 15786/43939 [25:29:34<44:39:37, 5.71s/it]
36%|███▌ | 15787/43939 [25:29:40<44:31:20, 5.69s/it]
36%|███▌ | 15788/43939 [25:29:45<44:35:51, 5.70s/it]
36%|███▌ | 15789/43939 [25:29:51<44:38:13, 5.71s/it]
36%|███▌ | 15790/43939 [25:29:57<44:39:32, 5.71s/it]
36%|███▌ | 15791/43939 [25:30:03<44:38:59, 5.71s/it]
36%|███▌ | 15792/43939 [25:30:08<44:39:50, 5.71s/it]
36%|███▌ | 15793/43939 [25:30:14<44:46:02, 5.73s/it]
36%|███▌ | 15794/43939 [25:30:20<44:43:21, 5.72s/it]
36%|███▌ | 15795/43939 [25:30:26<45:18:37, 5.80s/it]
36%|███▌ | 15796/43939 [25:30:31<45:05:31, 5.77s/it]
36%|███▌ | 15797/43939 [25:30:37<44:58:25, 5.75s/it]
36%|███▌ | 15798/43939 [25:30:43<44:49:54, 5.74s/it]
36%|███▌ | 15799/43939 [25:30:49<44:44:36, 5.72s/it]
36%|███▌ | 15800/43939 [25:30:54<44:37:34, 5.71s/it]
{'loss': 0.9456, 'grad_norm': 1.4083410501480103, 'learning_rate': 4.042031676943521e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 528.47, 'epoch': 1.44}
36%|███▌ | 15800/43939 [25:30:54<44:37:34, 5.71s/it]
36%|███▌ | 15801/43939 [25:31:00<44:40:00, 5.71s/it]
36%|███▌ | 15802/43939 [25:31:06<44:37:19, 5.71s/it]
36%|███▌ | 15803/43939 [25:31:11<44:35:50, 5.71s/it]
36%|███▌ | 15804/43939 [25:31:17<44:34:43, 5.70s/it]
36%|███▌ | 15805/43939 [25:31:23<44:37:32, 5.71s/it]
36%|███▌ | 15806/43939 [25:31:28<44:39:44, 5.72s/it]
36%|███▌ | 15807/43939 [25:31:34<44:37:36, 5.71s/it]
36%|███▌ | 15808/43939 [25:31:40<44:41:50, 5.72s/it]
36%|███▌ | 15809/43939 [25:31:46<44:39:30, 5.72s/it]
36%|███▌ | 15810/43939 [25:31:51<44:43:32, 5.72s/it]
36%|███▌ | 15811/43939 [25:31:57<44:41:31, 5.72s/it]
36%|███▌ | 15812/43939 [25:32:03<44:39:18, 5.72s/it]
36%|███▌ | 15813/43939 [25:32:09<44:42:49, 5.72s/it]
36%|███▌ | 15814/43939 [25:32:14<44:37:16, 5.71s/it]
36%|███▌ | 15815/43939 [25:32:20<44:41:53, 5.72s/it]
36%|███▌ | 15816/43939 [25:32:26<44:37:24, 5.71s/it]
36%|███▌ | 15817/43939 [25:32:32<45:01:18, 5.76s/it]
36%|███▌ | 15818/43939 [25:32:37<44:53:33, 5.75s/it]
36%|███▌ | 15819/43939 [25:32:43<44:48:00, 5.74s/it]
36%|███▌ | 15820/43939 [25:32:49<44:46:00, 5.73s/it]
36%|███▌ | 15821/43939 [25:32:54<44:44:20, 5.73s/it]
36%|███▌ | 15822/43939 [25:33:00<44:36:43, 5.71s/it]
36%|███▌ | 15823/43939 [25:33:06<44:30:33, 5.70s/it]
36%|███▌ | 15824/43939 [25:33:11<44:32:37, 5.70s/it]
36%|███▌ | 15825/43939 [25:33:17<44:29:55, 5.70s/it]
36%|███▌ | 15826/43939 [25:33:23<44:32:05, 5.70s/it]
36%|███▌ | 15827/43939 [25:33:29<44:30:09, 5.70s/it]
36%|███▌ | 15828/43939 [25:33:34<44:30:57, 5.70s/it]
36%|███▌ | 15829/43939 [25:33:40<44:32:18, 5.70s/it]
36%|███▌ | 15830/43939 [25:33:46<44:35:54, 5.71s/it]
36%|███▌ | 15831/43939 [25:33:51<44:28:21, 5.70s/it]
36%|███▌ | 15832/43939 [25:33:57<44:30:53, 5.70s/it]
36%|███▌ | 15833/43939 [25:34:03<44:32:55, 5.71s/it]
36%|███▌ | 15834/43939 [25:34:08<44:37:00, 5.72s/it]
36%|███▌ | 15835/43939 [25:34:14<44:34:34, 5.71s/it]
36%|███▌ | 15836/43939 [25:34:20<44:33:47, 5.71s/it]
36%|███▌ | 15837/43939 [25:34:26<44:34:20, 5.71s/it]
36%|███▌ | 15838/43939 [25:34:32<45:06:23, 5.78s/it]
36%|███▌ | 15839/43939 [25:34:37<45:01:20, 5.77s/it]
36%|███▌ | 15840/43939 [25:34:43<44:51:47, 5.75s/it]
36%|███▌ | 15841/43939 [25:34:49<44:50:24, 5.75s/it]
36%|███▌ | 15842/43939 [25:34:54<44:43:47, 5.73s/it]
36%|███▌ | 15843/43939 [25:35:00<44:34:15, 5.71s/it]
36%|███▌ | 15844/43939 [25:35:06<44:30:32, 5.70s/it]
36%|███▌ | 15845/43939 [25:35:11<44:23:54, 5.69s/it]
36%|███▌ | 15846/43939 [25:35:17<44:31:37, 5.71s/it]
36%|███▌ | 15847/43939 [25:35:23<44:30:30, 5.70s/it]
36%|███▌ | 15848/43939 [25:35:29<44:31:46, 5.71s/it]
36%|███▌ | 15849/43939 [25:35:34<44:26:40, 5.70s/it]
36%|███▌ | 15850/43939 [25:35:40<44:35:51, 5.72s/it]
{'loss': 0.9328, 'grad_norm': 1.1018062829971313, 'learning_rate': 4.034203385423983e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.51, 'epoch': 1.44}
36%|███▌ | 15850/43939 [25:35:40<44:35:51, 5.72s/it]
36%|███▌ | 15851/43939 [25:35:46<44:34:49, 5.71s/it]
36%|███▌ | 15852/43939 [25:35:51<44:31:16, 5.71s/it]
36%|███▌ | 15853/43939 [25:35:57<44:31:57, 5.71s/it]
36%|███▌ | 15854/43939 [25:36:03<44:29:41, 5.70s/it]
36%|███▌ | 15855/43939 [25:36:09<44:40:57, 5.73s/it]
36%|███▌ | 15856/43939 [25:36:14<44:34:40, 5.71s/it]
36%|███▌ | 15857/43939 [25:36:20<44:31:36, 5.71s/it]
36%|███▌ | 15858/43939 [25:36:26<44:32:39, 5.71s/it]
36%|███▌ | 15859/43939 [25:36:32<45:01:36, 5.77s/it]
36%|███▌ | 15860/43939 [25:36:37<44:51:48, 5.75s/it]
36%|███▌ | 15861/43939 [25:36:43<44:39:41, 5.73s/it]
36%|███▌ | 15862/43939 [25:36:49<44:36:00, 5.72s/it]
36%|███▌ | 15863/43939 [25:36:54<44:38:41, 5.72s/it]
36%|███▌ | 15864/43939 [25:37:00<44:38:27, 5.72s/it]
36%|███▌ | 15865/43939 [25:37:06<44:37:22, 5.72s/it]
36%|███▌ | 15866/43939 [25:37:12<44:42:02, 5.73s/it]
36%|███▌ | 15867/43939 [25:37:17<44:36:58, 5.72s/it]
36%|███▌ | 15868/43939 [25:37:23<44:35:37, 5.72s/it]
36%|███▌ | 15869/43939 [25:37:29<44:25:57, 5.70s/it]
36%|███▌ | 15870/43939 [25:37:34<44:24:51, 5.70s/it]
36%|███▌ | 15871/43939 [25:37:40<44:30:38, 5.71s/it]
36%|███▌ | 15872/43939 [25:37:46<44:29:23, 5.71s/it]
36%|███▌ | 15873/43939 [25:37:51<44:24:39, 5.70s/it]
36%|███▌ | 15874/43939 [25:37:57<44:24:44, 5.70s/it]
36%|███▌ | 15875/43939 [25:38:03<44:24:22, 5.70s/it]
36%|███▌ | 15876/43939 [25:38:09<44:28:32, 5.71s/it]
36%|███▌ | 15877/43939 [25:38:14<44:26:42, 5.70s/it]
36%|███▌ | 15878/43939 [25:38:20<44:30:14, 5.71s/it]
36%|███▌ | 15879/43939 [25:38:26<44:24:41, 5.70s/it]
36%|███▌ | 15880/43939 [25:38:31<44:22:03, 5.69s/it]
36%|███▌ | 15881/43939 [25:38:37<44:53:03, 5.76s/it]
36%|███▌ | 15882/43939 [25:38:43<44:42:36, 5.74s/it]
36%|███▌ | 15883/43939 [25:38:49<44:39:03, 5.73s/it]
36%|███▌ | 15884/43939 [25:38:54<44:33:15, 5.72s/it]
36%|███▌ | 15885/43939 [25:39:00<44:32:08, 5.71s/it]
36%|███▌ | 15886/43939 [25:39:06<44:33:02, 5.72s/it]
36%|███▌ | 15887/43939 [25:39:12<44:29:19, 5.71s/it]
36%|███▌ | 15888/43939 [25:39:17<44:32:53, 5.72s/it]
36%|███▌ | 15889/43939 [25:39:23<44:29:24, 5.71s/it]
36%|███▌ | 15890/43939 [25:39:29<44:28:15, 5.71s/it]
36%|███▌ | 15891/43939 [25:39:34<44:24:17, 5.70s/it]
36%|███▌ | 15892/43939 [25:39:40<44:24:21, 5.70s/it]
36%|███▌ | 15893/43939 [25:39:46<44:24:23, 5.70s/it]
36%|███▌ | 15894/43939 [25:39:51<44:22:33, 5.70s/it]
36%|███▌ | 15895/43939 [25:39:57<44:24:45, 5.70s/it]
36%|███▌ | 15896/43939 [25:40:03<44:20:34, 5.69s/it]
36%|███▌ | 15897/43939 [25:40:09<44:27:34, 5.71s/it]
36%|███▌ | 15898/43939 [25:40:14<44:23:15, 5.70s/it]
36%|███▌ | 15899/43939 [25:40:20<44:26:07, 5.70s/it]
36%|███▌ | 15900/43939 [25:40:26<44:30:38, 5.71s/it]
{'loss': 0.935, 'grad_norm': 1.071813702583313, 'learning_rate': 4.0263508882407097e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.51, 'epoch': 1.45}
36%|███▌ | 15900/43939 [25:40:26<44:30:38, 5.71s/it]
36%|███▌ | 15901/43939 [25:40:31<44:30:55, 5.72s/it]
36%|███▌ | 15902/43939 [25:40:37<44:54:33, 5.77s/it]
36%|███▌ | 15903/43939 [25:40:43<44:41:14, 5.74s/it]
36%|███▌ | 15904/43939 [25:40:49<44:36:18, 5.73s/it]
36%|███▌ | 15905/43939 [25:40:54<44:29:43, 5.71s/it]
36%|███▌ | 15906/43939 [25:41:00<44:27:28, 5.71s/it]
36%|███▌ | 15907/43939 [25:41:06<44:21:35, 5.70s/it]
36%|███▌ | 15908/43939 [25:41:11<44:21:40, 5.70s/it]
36%|███▌ | 15909/43939 [25:41:17<44:20:39, 5.70s/it]
36%|███▌ | 15910/43939 [25:41:23<44:28:25, 5.71s/it]
36%|███▌ | 15911/43939 [25:41:29<44:28:57, 5.71s/it]
36%|███▌ | 15912/43939 [25:41:34<44:26:05, 5.71s/it]
36%|███▌ | 15913/43939 [25:41:40<44:21:47, 5.70s/it]
36%|███▌ | 15914/43939 [25:41:46<44:20:39, 5.70s/it]
36%|███▌ | 15915/43939 [25:41:51<44:30:57, 5.72s/it]
36%|███▌ | 15916/43939 [25:41:57<44:36:07, 5.73s/it]
36%|███▌ | 15917/43939 [25:42:03<44:34:29, 5.73s/it]
36%|███▌ | 15918/43939 [25:42:09<44:30:53, 5.72s/it]
36%|███▌ | 15919/43939 [25:42:14<44:26:26, 5.71s/it]
36%|███▌ | 15920/43939 [25:42:20<44:26:22, 5.71s/it]
36%|███▌ | 15921/43939 [25:42:26<44:21:15, 5.70s/it]
36%|███▌ | 15922/43939 [25:42:31<44:21:01, 5.70s/it]
36%|███▌ | 15923/43939 [25:42:37<44:55:01, 5.77s/it]
36%|███▌ | 15924/43939 [25:42:43<44:38:47, 5.74s/it]
36%|███▌ | 15925/43939 [25:42:49<44:30:33, 5.72s/it]
36%|███▌ | 15926/43939 [25:42:54<44:30:23, 5.72s/it]
36%|███▌ | 15927/43939 [25:43:00<44:29:46, 5.72s/it]
36%|███▋ | 15928/43939 [25:43:06<44:25:16, 5.71s/it]
36%|███▋ | 15929/43939 [25:43:11<44:23:24, 5.71s/it]
36%|███▋ | 15930/43939 [25:43:17<44:23:28, 5.71s/it]
36%|███▋ | 15931/43939 [25:43:23<44:24:06, 5.71s/it]
36%|███▋ | 15932/43939 [25:43:29<44:25:05, 5.71s/it]
36%|███▋ | 15933/43939 [25:43:34<44:32:01, 5.72s/it]
36%|███▋ | 15934/43939 [25:43:40<44:24:44, 5.71s/it]
36%|███▋ | 15935/43939 [25:43:46<44:25:35, 5.71s/it]
36%|███▋ | 15936/43939 [25:43:51<44:26:30, 5.71s/it]
36%|███▋ | 15937/43939 [25:43:57<44:28:11, 5.72s/it]
36%|███▋ | 15938/43939 [25:44:03<44:24:45, 5.71s/it]
36%|███▋ | 15939/43939 [25:44:09<44:26:18, 5.71s/it]
36%|███▋ | 15940/43939 [25:44:14<44:31:55, 5.73s/it]
36%|███▋ | 15941/43939 [25:44:20<44:34:51, 5.73s/it]
36%|███▋ | 15942/43939 [25:44:26<44:27:50, 5.72s/it]
36%|███▋ | 15943/43939 [25:44:31<44:27:28, 5.72s/it]
36%|███▋ | 15944/43939 [25:44:37<44:55:48, 5.78s/it]
36%|███▋ | 15945/43939 [25:44:43<44:45:37, 5.76s/it]
36%|███▋ | 15946/43939 [25:44:49<44:41:27, 5.75s/it]
36%|███▋ | 15947/43939 [25:44:55<44:41:10, 5.75s/it]
36%|███▋ | 15948/43939 [25:45:00<44:29:09, 5.72s/it]
36%|███▋ | 15949/43939 [25:45:06<44:27:36, 5.72s/it]
36%|███▋ | 15950/43939 [25:45:12<44:30:01, 5.72s/it]
{'loss': 0.9373, 'grad_norm': 0.9755890369415283, 'learning_rate': 4.0184743092853005e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.57, 'epoch': 1.45}
36%|███▋ | 15950/43939 [25:45:12<44:30:01, 5.72s/it]
36%|███▋ | 15951/43939 [25:45:17<44:32:29, 5.73s/it]
36%|███▋ | 15952/43939 [25:45:23<44:29:31, 5.72s/it]
36%|███▋ | 15953/43939 [25:45:29<44:27:15, 5.72s/it]
36%|███▋ | 15954/43939 [25:45:35<44:34:30, 5.73s/it]
36%|███▋ | 15955/43939 [25:45:40<44:36:05, 5.74s/it]
36%|███▋ | 15956/43939 [25:45:46<44:26:10, 5.72s/it]
36%|███▋ | 15957/43939 [25:45:52<44:25:46, 5.72s/it]
36%|███▋ | 15958/43939 [25:45:57<44:22:11, 5.71s/it]
36%|███▋ | 15959/43939 [25:46:03<44:21:12, 5.71s/it]
36%|███▋ | 15960/43939 [25:46:09<44:24:17, 5.71s/it]
36%|███▋ | 15961/43939 [25:46:15<44:24:02, 5.71s/it]
36%|███▋ | 15962/43939 [25:46:20<44:31:55, 5.73s/it]
36%|███▋ | 15963/43939 [25:46:26<44:28:19, 5.72s/it]
36%|███▋ | 15964/43939 [25:46:32<44:26:13, 5.72s/it]
36%|███▋ | 15965/43939 [25:46:38<44:56:45, 5.78s/it]
36%|███▋ | 15966/43939 [25:46:43<44:47:23, 5.76s/it]
36%|███▋ | 15967/43939 [25:46:49<44:45:43, 5.76s/it]
36%|███▋ | 15968/43939 [25:46:55<44:45:06, 5.76s/it]
36%|███▋ | 15969/43939 [25:47:01<44:38:05, 5.74s/it]
36%|███▋ | 15970/43939 [25:47:06<44:38:36, 5.75s/it]
36%|███▋ | 15971/43939 [25:47:12<44:36:19, 5.74s/it]
36%|███▋ | 15972/43939 [25:47:18<44:33:04, 5.73s/it]
36%|███▋ | 15973/43939 [25:47:24<44:24:45, 5.72s/it]
36%|███▋ | 15974/43939 [25:47:29<44:23:32, 5.71s/it]
36%|███▋ | 15975/43939 [25:47:35<44:23:32, 5.71s/it]
36%|███▋ | 15976/43939 [25:47:41<44:17:19, 5.70s/it]
36%|███▋ | 15977/43939 [25:47:46<44:18:00, 5.70s/it]
36%|███▋ | 15978/43939 [25:47:52<44:13:39, 5.69s/it]
36%|███▋ | 15979/43939 [25:47:58<44:15:38, 5.70s/it]
36%|███▋ | 15980/43939 [25:48:03<44:23:25, 5.72s/it]
36%|███▋ | 15981/43939 [25:48:09<44:24:25, 5.72s/it]
36%|███▋ | 15982/43939 [25:48:15<44:25:50, 5.72s/it]
36%|███▋ | 15983/43939 [25:48:21<44:27:10, 5.72s/it]
36%|███▋ | 15984/43939 [25:48:26<44:24:58, 5.72s/it]
36%|███▋ | 15985/43939 [25:48:32<44:23:32, 5.72s/it]
36%|███▋ | 15986/43939 [25:48:38<45:00:22, 5.80s/it]
36%|███▋ | 15987/43939 [25:48:44<44:49:36, 5.77s/it]
36%|███▋ | 15988/43939 [25:48:49<44:38:41, 5.75s/it]
36%|███▋ | 15989/43939 [25:48:55<44:29:12, 5.73s/it]
36%|███▋ | 15990/43939 [25:49:01<44:24:54, 5.72s/it]
36%|███▋ | 15991/43939 [25:49:07<44:21:49, 5.71s/it]
36%|███▋ | 15992/43939 [25:49:12<44:25:12, 5.72s/it]
36%|███▋ | 15993/43939 [25:49:18<44:22:36, 5.72s/it]
36%|███▋ | 15994/43939 [25:49:24<44:15:32, 5.70s/it]
36%|███▋ | 15995/43939 [25:49:29<44:12:42, 5.70s/it]
36%|███▋ | 15996/43939 [25:49:35<44:12:24, 5.70s/it]
36%|███▋ | 15997/43939 [25:49:41<44:10:50, 5.69s/it]
36%|███▋ | 15998/43939 [25:49:46<44:10:25, 5.69s/it]
36%|███▋ | 15999/43939 [25:49:52<44:14:44, 5.70s/it]
36%|███▋ | 16000/43939 [25:49:58<44:12:31, 5.70s/it]
{'loss': 0.9303, 'grad_norm': 1.1645768880844116, 'learning_rate': 4.010573772829297e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.61, 'epoch': 1.46}
36%|███▋ | 16000/43939 [25:49:58<44:12:31, 5.70s/it]
36%|███▋ | 16001/43939 [25:50:04<44:13:26, 5.70s/it]
36%|███▋ | 16002/43939 [25:50:09<44:16:19, 5.70s/it]
36%|███▋ | 16003/43939 [25:50:15<44:22:31, 5.72s/it]
36%|███▋ | 16004/43939 [25:50:21<44:19:43, 5.71s/it]
36%|███▋ | 16005/43939 [25:50:26<44:23:07, 5.72s/it]
36%|███▋ | 16006/43939 [25:50:32<44:17:33, 5.71s/it]
36%|███▋ | 16007/43939 [25:50:38<44:54:00, 5.79s/it]
36%|███▋ | 16008/43939 [25:50:44<44:39:22, 5.76s/it]
36%|███▋ | 16009/43939 [25:50:49<44:31:46, 5.74s/it]
36%|███▋ | 16010/43939 [25:50:55<44:24:40, 5.72s/it]
36%|███▋ | 16011/43939 [25:51:01<44:24:32, 5.72s/it]
36%|███▋ | 16012/43939 [25:51:07<44:18:53, 5.71s/it]
36%|███▋ | 16013/43939 [25:51:12<44:23:03, 5.72s/it]
36%|███▋ | 16014/43939 [25:51:18<44:20:42, 5.72s/it]
36%|███▋ | 16015/43939 [25:51:24<44:16:16, 5.71s/it]
36%|███▋ | 16016/43939 [25:51:29<44:11:49, 5.70s/it]
36%|███▋ | 16017/43939 [25:51:35<44:13:23, 5.70s/it]
36%|███▋ | 16018/43939 [25:51:41<44:08:18, 5.69s/it]
36%|███▋ | 16019/43939 [25:51:46<44:16:09, 5.71s/it]
36%|███▋ | 16020/43939 [25:51:52<44:13:23, 5.70s/it]
36%|███▋ | 16021/43939 [25:51:58<44:10:56, 5.70s/it]
36%|███▋ | 16022/43939 [25:52:04<44:11:02, 5.70s/it]
36%|███▋ | 16023/43939 [25:52:09<44:07:48, 5.69s/it]
36%|███▋ | 16024/43939 [25:52:15<44:09:48, 5.70s/it]
36%|███▋ | 16025/43939 [25:52:21<44:15:40, 5.71s/it]
36%|███▋ | 16026/43939 [25:52:26<44:09:50, 5.70s/it]
36%|███▋ | 16027/43939 [25:52:32<44:08:42, 5.69s/it]
36%|███▋ | 16028/43939 [25:52:38<44:13:03, 5.70s/it]
36%|███▋ | 16029/43939 [25:52:44<44:46:23, 5.78s/it]
36%|███▋ | 16030/43939 [25:52:49<44:36:22, 5.75s/it]
36%|███▋ | 16031/43939 [25:52:55<44:27:01, 5.73s/it]
36%|███▋ | 16032/43939 [25:53:01<44:21:57, 5.72s/it]
36%|███▋ | 16033/43939 [25:53:06<44:17:50, 5.71s/it]
36%|███▋ | 16034/43939 [25:53:12<44:13:39, 5.71s/it]
36%|███▋ | 16035/43939 [25:53:18<44:16:43, 5.71s/it]
36%|███▋ | 16036/43939 [25:53:24<44:24:44, 5.73s/it]
36%|███▋ | 16037/43939 [25:53:29<44:17:15, 5.71s/it]
37%|███▋ | 16038/43939 [25:53:35<44:14:46, 5.71s/it]
37%|███▋ | 16039/43939 [25:53:41<44:11:44, 5.70s/it]
37%|███▋ | 16040/43939 [25:53:46<44:15:08, 5.71s/it]
37%|███▋ | 16041/43939 [25:53:52<44:10:11, 5.70s/it]
37%|███▋ | 16042/43939 [25:53:58<44:08:40, 5.70s/it]
37%|███▋ | 16043/43939 [25:54:04<44:07:53, 5.70s/it]
37%|███▋ | 16044/43939 [25:54:09<44:12:43, 5.71s/it]
37%|███▋ | 16045/43939 [25:54:15<44:13:21, 5.71s/it]
37%|███▋ | 16046/43939 [25:54:21<44:12:32, 5.71s/it]
37%|███▋ | 16047/43939 [25:54:26<44:10:36, 5.70s/it]
37%|███▋ | 16048/43939 [25:54:32<44:13:16, 5.71s/it]
37%|███▋ | 16049/43939 [25:54:38<44:13:43, 5.71s/it]
37%|███▋ | 16050/43939 [25:54:44<44:40:07, 5.77s/it]
{'loss': 0.9087, 'grad_norm': 1.2208948135375977, 'learning_rate': 4.002649403522226e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.81, 'epoch': 1.46}
37%|███▋ | 16050/43939 [25:54:44<44:40:07, 5.77s/it]
37%|███▋ | 16051/43939 [25:54:49<44:30:40, 5.75s/it]
37%|███▋ | 16052/43939 [25:54:55<44:22:27, 5.73s/it]
37%|███▋ | 16053/43939 [25:55:01<44:24:16, 5.73s/it]
37%|███▋ | 16054/43939 [25:55:07<44:20:16, 5.72s/it]
37%|███▋ | 16055/43939 [25:55:12<44:16:49, 5.72s/it]
37%|███▋ | 16056/43939 [25:55:18<44:22:14, 5.73s/it]
37%|███▋ | 16057/43939 [25:55:24<44:15:49, 5.72s/it]
37%|███▋ | 16058/43939 [25:55:29<44:14:31, 5.71s/it]
37%|███▋ | 16059/43939 [25:55:35<44:18:45, 5.72s/it]
37%|███▋ | 16060/43939 [25:55:41<44:15:16, 5.71s/it]
37%|███▋ | 16061/43939 [25:55:46<44:10:08, 5.70s/it]
37%|███▋ | 16062/43939 [25:55:52<44:04:14, 5.69s/it]
37%|███▋ | 16063/43939 [25:55:58<44:10:05, 5.70s/it]
37%|███▋ | 16064/43939 [25:56:04<44:11:48, 5.71s/it]
37%|███▋ | 16065/43939 [25:56:09<44:10:13, 5.70s/it]
37%|███▋ | 16066/43939 [25:56:15<44:12:14, 5.71s/it]
37%|███▋ | 16067/43939 [25:56:21<44:09:36, 5.70s/it]
37%|███▋ | 16068/43939 [25:56:26<44:10:53, 5.71s/it]
37%|███▋ | 16069/43939 [25:56:32<44:13:52, 5.71s/it]
37%|███▋ | 16070/43939 [25:56:38<44:18:46, 5.72s/it]
37%|███▋ | 16071/43939 [25:56:44<44:46:37, 5.78s/it]
37%|███▋ | 16072/43939 [25:56:49<44:31:32, 5.75s/it]
37%|███▋ | 16073/43939 [25:56:55<44:30:15, 5.75s/it]
37%|███▋ | 16074/43939 [25:57:01<44:19:35, 5.73s/it]
37%|███▋ | 16075/43939 [25:57:07<44:16:05, 5.72s/it]
37%|███▋ | 16076/43939 [25:57:12<44:10:27, 5.71s/it]
37%|███▋ | 16077/43939 [25:57:18<44:10:22, 5.71s/it]
37%|███▋ | 16078/43939 [25:57:24<44:10:22, 5.71s/it]
37%|███▋ | 16079/43939 [25:57:29<44:11:43, 5.71s/it]
37%|███▋ | 16080/43939 [25:57:35<44:04:44, 5.70s/it]
37%|███▋ | 16081/43939 [25:57:41<44:05:59, 5.70s/it]
37%|███▋ | 16082/43939 [25:57:47<44:08:12, 5.70s/it]
37%|███▋ | 16083/43939 [25:57:52<44:09:35, 5.71s/it]
37%|███▋ | 16084/43939 [25:57:58<44:08:57, 5.71s/it]
37%|███▋ | 16085/43939 [25:58:04<44:11:22, 5.71s/it]
37%|███▋ | 16086/43939 [25:58:09<44:08:28, 5.71s/it]
37%|███▋ | 16087/43939 [25:58:15<44:05:11, 5.70s/it]
37%|███▋ | 16088/43939 [25:58:21<44:01:46, 5.69s/it]
37%|███▋ | 16089/43939 [25:58:26<44:01:47, 5.69s/it]
37%|███▋ | 16090/43939 [25:58:32<43:56:15, 5.68s/it]
37%|███▋ | 16091/43939 [25:58:38<43:57:55, 5.68s/it]
37%|███▋ | 16092/43939 [25:58:44<44:32:24, 5.76s/it]
37%|███▋ | 16093/43939 [25:58:49<44:24:38, 5.74s/it]
37%|███▋ | 16094/43939 [25:58:55<44:21:54, 5.74s/it]
37%|███▋ | 16095/43939 [25:59:01<44:18:59, 5.73s/it]
37%|███▋ | 16096/43939 [25:59:07<44:20:56, 5.73s/it]
37%|███▋ | 16097/43939 [25:59:12<44:14:06, 5.72s/it]
37%|███▋ | 16098/43939 [25:59:18<44:12:46, 5.72s/it]
37%|███▋ | 16099/43939 [25:59:24<44:10:43, 5.71s/it]
37%|███▋ | 16100/43939 [25:59:29<44:08:14, 5.71s/it]
{'loss': 0.9268, 'grad_norm': 1.198129653930664, 'learning_rate': 3.9947013263896385e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.62, 'epoch': 1.47}
37%|███▋ | 16100/43939 [25:59:29<44:08:14, 5.71s/it]
37%|███▋ | 16101/43939 [25:59:35<44:09:04, 5.71s/it]
37%|███▋ | 16102/43939 [25:59:41<44:09:20, 5.71s/it]
37%|███▋ | 16103/43939 [25:59:46<44:06:08, 5.70s/it]
37%|███▋ | 16104/43939 [25:59:52<44:05:13, 5.70s/it]
37%|███▋ | 16105/43939 [25:59:58<44:11:45, 5.72s/it]
37%|███▋ | 16106/43939 [26:00:04<44:19:37, 5.73s/it]
37%|███▋ | 16107/43939 [26:00:09<44:13:08, 5.72s/it]
37%|███▋ | 16108/43939 [26:00:15<44:11:12, 5.72s/it]
37%|███▋ | 16109/43939 [26:00:21<44:09:14, 5.71s/it]
37%|███▋ | 16110/43939 [26:00:26<44:10:09, 5.71s/it]
37%|███▋ | 16111/43939 [26:00:32<44:04:02, 5.70s/it]
37%|███▋ | 16112/43939 [26:00:38<44:08:13, 5.71s/it]
37%|███▋ | 16113/43939 [26:00:44<44:36:18, 5.77s/it]
37%|███▋ | 16114/43939 [26:00:50<44:25:26, 5.75s/it]
37%|███▋ | 16115/43939 [26:00:55<44:16:12, 5.73s/it]
37%|███▋ | 16116/43939 [26:01:01<44:10:18, 5.72s/it]
37%|███▋ | 16117/43939 [26:01:07<44:04:30, 5.70s/it]
37%|███▋ | 16118/43939 [26:01:12<44:01:59, 5.70s/it]
37%|███▋ | 16119/43939 [26:01:18<44:07:37, 5.71s/it]
37%|███▋ | 16120/43939 [26:01:24<44:04:34, 5.70s/it]
37%|███▋ | 16121/43939 [26:01:29<44:03:01, 5.70s/it]
37%|███▋ | 16122/43939 [26:01:35<44:09:40, 5.72s/it]
37%|███▋ | 16123/43939 [26:01:41<44:09:45, 5.72s/it]
37%|███▋ | 16124/43939 [26:01:47<44:06:17, 5.71s/it]
37%|███▋ | 16125/43939 [26:01:52<44:01:37, 5.70s/it]
37%|███▋ | 16126/43939 [26:01:58<43:57:45, 5.69s/it]
37%|███▋ | 16127/43939 [26:02:04<44:02:42, 5.70s/it]
37%|███▋ | 16128/43939 [26:02:09<44:11:04, 5.72s/it]
37%|███▋ | 16129/43939 [26:02:15<44:12:11, 5.72s/it]
37%|███▋ | 16130/43939 [26:02:21<44:08:20, 5.71s/it]
37%|███▋ | 16131/43939 [26:02:26<44:06:51, 5.71s/it]
37%|███▋ | 16132/43939 [26:02:32<44:07:08, 5.71s/it]
37%|███▋ | 16133/43939 [26:02:38<44:03:17, 5.70s/it]
37%|███▋ | 16134/43939 [26:02:44<44:32:05, 5.77s/it]
37%|███▋ | 16135/43939 [26:02:50<44:26:21, 5.75s/it]
37%|███▋ | 16136/43939 [26:02:55<44:23:39, 5.75s/it]
37%|███▋ | 16137/43939 [26:03:01<44:19:52, 5.74s/it]
37%|███▋ | 16138/43939 [26:03:07<44:18:45, 5.74s/it]
37%|███▋ | 16139/43939 [26:03:12<44:16:11, 5.73s/it]
37%|███▋ | 16140/43939 [26:03:18<44:07:05, 5.71s/it]
37%|███▋ | 16141/43939 [26:03:24<44:01:46, 5.70s/it]
37%|███▋ | 16142/43939 [26:03:29<43:58:22, 5.69s/it]
37%|███▋ | 16143/43939 [26:03:35<43:53:41, 5.69s/it]
37%|███▋ | 16144/43939 [26:03:41<43:55:51, 5.69s/it]
37%|███▋ | 16145/43939 [26:03:47<43:57:34, 5.69s/it]
37%|███▋ | 16146/43939 [26:03:52<43:58:54, 5.70s/it]
37%|███▋ | 16147/43939 [26:03:58<44:00:31, 5.70s/it]
37%|███▋ | 16148/43939 [26:04:04<44:04:15, 5.71s/it]
37%|███▋ | 16149/43939 [26:04:09<44:01:10, 5.70s/it]
37%|███▋ | 16150/43939 [26:04:15<43:59:02, 5.70s/it]
{'loss': 0.9258, 'grad_norm': 1.2478697299957275, 'learning_rate': 3.986729666831126e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.22, 'epoch': 1.47}
37%|███▋ | 16150/43939 [26:04:15<43:59:02, 5.70s/it]
37%|███▋ | 16151/43939 [26:04:21<44:03:58, 5.71s/it]
37%|███▋ | 16152/43939 [26:04:27<44:13:54, 5.73s/it]
37%|███▋ | 16153/43939 [26:04:32<44:09:42, 5.72s/it]
37%|███▋ | 16154/43939 [26:04:38<44:08:30, 5.72s/it]
37%|███▋ | 16155/43939 [26:04:44<44:30:37, 5.77s/it]
37%|███▋ | 16156/43939 [26:04:50<44:19:54, 5.74s/it]
37%|███▋ | 16157/43939 [26:04:55<44:10:12, 5.72s/it]
37%|███▋ | 16158/43939 [26:05:01<44:11:26, 5.73s/it]
37%|███▋ | 16159/43939 [26:05:07<44:07:30, 5.72s/it]
37%|███▋ | 16160/43939 [26:05:12<44:02:31, 5.71s/it]
37%|███▋ | 16161/43939 [26:05:18<43:59:42, 5.70s/it]
37%|███▋ | 16162/43939 [26:05:24<44:00:59, 5.70s/it]
37%|███▋ | 16163/43939 [26:05:29<44:00:34, 5.70s/it]
37%|███▋ | 16164/43939 [26:05:35<43:59:27, 5.70s/it]
37%|███▋ | 16165/43939 [26:05:41<44:03:09, 5.71s/it]
37%|███▋ | 16166/43939 [26:05:47<44:03:08, 5.71s/it]
37%|███▋ | 16167/43939 [26:05:52<43:55:48, 5.69s/it]
37%|███▋ | 16168/43939 [26:05:58<44:01:45, 5.71s/it]
37%|███▋ | 16169/43939 [26:06:04<43:55:13, 5.69s/it]
37%|███▋ | 16170/43939 [26:06:09<43:51:58, 5.69s/it]
37%|███▋ | 16171/43939 [26:06:15<43:53:28, 5.69s/it]
37%|███▋ | 16172/43939 [26:06:21<43:49:38, 5.68s/it]
37%|███▋ | 16173/43939 [26:06:26<43:48:13, 5.68s/it]
37%|███▋ | 16174/43939 [26:06:32<43:50:56, 5.69s/it]
37%|███▋ | 16175/43939 [26:06:38<43:54:48, 5.69s/it]
37%|███▋ | 16176/43939 [26:06:44<44:20:23, 5.75s/it]
37%|███▋ | 16177/43939 [26:06:49<44:10:15, 5.73s/it]
37%|███▋ | 16178/43939 [26:06:55<44:04:45, 5.72s/it]
37%|███▋ | 16179/43939 [26:07:01<43:59:48, 5.71s/it]
37%|███▋ | 16180/43939 [26:07:06<43:56:41, 5.70s/it]
37%|███▋ | 16181/43939 [26:07:12<43:56:53, 5.70s/it]
37%|███▋ | 16182/43939 [26:07:18<44:01:17, 5.71s/it]
37%|███▋ | 16183/43939 [26:07:23<44:01:31, 5.71s/it]
37%|███▋ | 16184/43939 [26:07:29<43:55:33, 5.70s/it]
37%|███▋ | 16185/43939 [26:07:35<43:58:08, 5.70s/it]
37%|███▋ | 16186/43939 [26:07:41<44:01:17, 5.71s/it]
37%|███▋ | 16187/43939 [26:07:46<43:57:01, 5.70s/it]
37%|███▋ | 16188/43939 [26:07:52<43:54:18, 5.70s/it]
37%|███▋ | 16189/43939 [26:07:58<44:03:43, 5.72s/it]
37%|███▋ | 16190/43939 [26:08:03<44:01:53, 5.71s/it]
37%|███▋ | 16191/43939 [26:08:09<44:08:06, 5.73s/it]
37%|███▋ | 16192/43939 [26:08:15<44:04:50, 5.72s/it]
37%|███▋ | 16193/43939 [26:08:21<44:05:34, 5.72s/it]
37%|███▋ | 16194/43939 [26:08:26<44:00:58, 5.71s/it]
37%|███▋ | 16195/43939 [26:08:32<43:56:28, 5.70s/it]
37%|███▋ | 16196/43939 [26:08:38<43:53:24, 5.70s/it]
37%|███▋ | 16197/43939 [26:08:44<44:26:41, 5.77s/it]
37%|███▋ | 16198/43939 [26:08:49<44:17:20, 5.75s/it]
37%|███▋ | 16199/43939 [26:08:55<44:12:10, 5.74s/it]
37%|███▋ | 16200/43939 [26:09:01<44:05:11, 5.72s/it]
{'loss': 0.9123, 'grad_norm': 1.1567541360855103, 'learning_rate': 3.978734550618353e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.73, 'epoch': 1.47}
37%|███▋ | 16200/43939 [26:09:01<44:05:11, 5.72s/it]
37%|███▋ | 16201/43939 [26:09:06<44:04:51, 5.72s/it]
37%|███▋ | 16202/43939 [26:09:12<44:02:01, 5.72s/it]
37%|███▋ | 16203/43939 [26:09:18<44:06:28, 5.72s/it]
37%|███▋ | 16204/43939 [26:09:24<44:06:41, 5.73s/it]
37%|███▋ | 16205/43939 [26:09:29<44:03:30, 5.72s/it]
37%|███▋ | 16206/43939 [26:09:35<43:57:53, 5.71s/it]
37%|███▋ | 16207/43939 [26:09:41<43:57:04, 5.71s/it]
37%|███▋ | 16208/43939 [26:09:46<43:57:25, 5.71s/it]
37%|███▋ | 16209/43939 [26:09:52<44:03:38, 5.72s/it]
37%|███▋ | 16210/43939 [26:09:58<43:55:12, 5.70s/it]
37%|███▋ | 16211/43939 [26:10:04<44:08:40, 5.73s/it]
37%|███▋ | 16212/43939 [26:10:09<44:04:05, 5.72s/it]
37%|███▋ | 16213/43939 [26:10:15<44:03:18, 5.72s/it]
37%|███▋ | 16214/43939 [26:10:21<44:02:10, 5.72s/it]
37%|███▋ | 16215/43939 [26:10:26<44:00:04, 5.71s/it]
37%|███▋ | 16216/43939 [26:10:32<44:01:39, 5.72s/it]
37%|███▋ | 16217/43939 [26:10:38<44:42:30, 5.81s/it]
37%|███▋ | 16218/43939 [26:10:44<44:22:07, 5.76s/it]
37%|███▋ | 16219/43939 [26:10:49<44:09:09, 5.73s/it]
37%|███▋ | 16220/43939 [26:10:55<44:02:50, 5.72s/it]
37%|███▋ | 16221/43939 [26:11:01<44:07:11, 5.73s/it]
37%|███▋ | 16222/43939 [26:11:07<44:01:41, 5.72s/it]
37%|███▋ | 16223/43939 [26:11:12<44:05:02, 5.73s/it]
37%|███▋ | 16224/43939 [26:11:18<44:01:43, 5.72s/it]
37%|███▋ | 16225/43939 [26:11:24<44:01:09, 5.72s/it]
37%|███▋ | 16226/43939 [26:11:29<43:59:12, 5.71s/it]
37%|███▋ | 16227/43939 [26:11:35<43:58:32, 5.71s/it]
37%|███▋ | 16228/43939 [26:11:41<44:01:56, 5.72s/it]
37%|███▋ | 16229/43939 [26:11:47<44:07:05, 5.73s/it]
37%|███▋ | 16230/43939 [26:11:52<44:05:03, 5.73s/it]
37%|███▋ | 16231/43939 [26:11:58<44:01:00, 5.72s/it]
37%|███▋ | 16232/43939 [26:12:04<43:58:00, 5.71s/it]
37%|███▋ | 16233/43939 [26:12:10<43:55:19, 5.71s/it]
37%|███▋ | 16234/43939 [26:12:15<44:03:31, 5.73s/it]
37%|███▋ | 16235/43939 [26:12:21<43:58:21, 5.71s/it]
37%|███▋ | 16236/43939 [26:12:27<44:02:45, 5.72s/it]
37%|███▋ | 16237/43939 [26:12:32<44:05:47, 5.73s/it]
37%|███▋ | 16238/43939 [26:12:38<44:04:05, 5.73s/it]
37%|███▋ | 16239/43939 [26:12:44<44:34:39, 5.79s/it]
37%|███▋ | 16240/43939 [26:12:50<44:30:08, 5.78s/it]
37%|███▋ | 16241/43939 [26:12:56<44:13:57, 5.75s/it]
37%|███▋ | 16242/43939 [26:13:01<44:03:08, 5.73s/it]
37%|███▋ | 16243/43939 [26:13:07<43:56:58, 5.71s/it]
37%|███▋ | 16244/43939 [26:13:13<43:57:21, 5.71s/it]
37%|███▋ | 16245/43939 [26:13:18<43:59:14, 5.72s/it]
37%|███▋ | 16246/43939 [26:13:24<43:56:38, 5.71s/it]
37%|███▋ | 16247/43939 [26:13:30<43:48:09, 5.69s/it]
37%|███▋ | 16248/43939 [26:13:35<43:48:55, 5.70s/it]
37%|███▋ | 16249/43939 [26:13:41<43:52:49, 5.70s/it]
37%|███▋ | 16250/43939 [26:13:47<43:53:10, 5.71s/it]
{'loss': 0.9078, 'grad_norm': 1.2206943035125732, 'learning_rate': 3.970716103893065e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.22, 'epoch': 1.48}
37%|███▋ | 16250/43939 [26:13:47<43:53:10, 5.71s/it]
37%|███▋ | 16251/43939 [26:13:53<43:57:41, 5.72s/it]
37%|███▋ | 16252/43939 [26:13:58<43:47:31, 5.69s/it]
37%|███▋ | 16253/43939 [26:14:04<43:50:17, 5.70s/it]
37%|███▋ | 16254/43939 [26:14:10<43:43:39, 5.69s/it]
37%|███▋ | 16255/43939 [26:14:15<43:41:22, 5.68s/it]
37%|███▋ | 16256/43939 [26:14:21<43:43:15, 5.69s/it]
37%|███▋ | 16257/43939 [26:14:27<43:48:33, 5.70s/it]
37%|███▋ | 16258/43939 [26:14:32<43:54:39, 5.71s/it]
37%|███▋ | 16259/43939 [26:14:38<43:49:32, 5.70s/it]
37%|███▋ | 16260/43939 [26:14:44<44:21:23, 5.77s/it]
37%|███▋ | 16261/43939 [26:14:50<44:08:30, 5.74s/it]
37%|███▋ | 16262/43939 [26:14:55<44:01:50, 5.73s/it]
37%|███▋ | 16263/43939 [26:15:01<44:01:07, 5.73s/it]
37%|███▋ | 16264/43939 [26:15:07<43:53:39, 5.71s/it]
37%|███▋ | 16265/43939 [26:15:12<43:53:08, 5.71s/it]
37%|███▋ | 16266/43939 [26:15:18<43:54:44, 5.71s/it]
37%|███▋ | 16267/43939 [26:15:24<43:55:17, 5.71s/it]
37%|███▋ | 16268/43939 [26:15:30<43:53:57, 5.71s/it]
37%|███▋ | 16269/43939 [26:15:35<43:55:03, 5.71s/it]
37%|███▋ | 16270/43939 [26:15:41<43:59:29, 5.72s/it]
37%|███▋ | 16271/43939 [26:15:47<43:58:35, 5.72s/it]
37%|███▋ | 16272/43939 [26:15:53<43:54:07, 5.71s/it]
37%|███▋ | 16273/43939 [26:15:58<43:56:10, 5.72s/it]
37%|███▋ | 16274/43939 [26:16:04<43:54:44, 5.71s/it]
37%|███▋ | 16275/43939 [26:16:10<43:48:34, 5.70s/it]
37%|███▋ | 16276/43939 [26:16:15<43:52:17, 5.71s/it]
37%|███▋ | 16277/43939 [26:16:21<43:58:46, 5.72s/it]
37%|███▋ | 16278/43939 [26:16:27<43:54:51, 5.72s/it]
37%|███▋ | 16279/43939 [26:16:32<43:47:44, 5.70s/it]
37%|███▋ | 16280/43939 [26:16:38<43:47:55, 5.70s/it]
37%|███▋ | 16281/43939 [26:16:44<44:19:09, 5.77s/it]
37%|███▋ | 16282/43939 [26:16:50<44:05:41, 5.74s/it]
37%|███▋ | 16283/43939 [26:16:56<44:10:23, 5.75s/it]
37%|███▋ | 16284/43939 [26:17:01<44:00:50, 5.73s/it]
37%|███▋ | 16285/43939 [26:17:07<43:55:02, 5.72s/it]
37%|███▋ | 16286/43939 [26:17:13<43:56:58, 5.72s/it]
37%|███▋ | 16287/43939 [26:17:18<43:54:21, 5.72s/it]
37%|███▋ | 16288/43939 [26:17:24<43:57:09, 5.72s/it]
37%|███▋ | 16289/43939 [26:17:30<43:58:23, 5.73s/it]
37%|███▋ | 16290/43939 [26:17:36<43:53:23, 5.71s/it]
37%|███▋ | 16291/43939 [26:17:41<43:57:52, 5.72s/it]
37%|███▋ | 16292/43939 [26:17:47<43:55:03, 5.72s/it]
37%|███▋ | 16293/43939 [26:17:53<43:47:50, 5.70s/it]
37%|███▋ | 16294/43939 [26:17:58<43:51:35, 5.71s/it]
37%|███▋ | 16295/43939 [26:18:04<43:50:47, 5.71s/it]
37%|███▋ | 16296/43939 [26:18:10<43:59:42, 5.73s/it]
37%|███▋ | 16297/43939 [26:18:16<44:01:24, 5.73s/it]
37%|███▋ | 16298/43939 [26:18:21<43:59:40, 5.73s/it]
37%|███▋ | 16299/43939 [26:18:27<43:53:24, 5.72s/it]
37%|███▋ | 16300/43939 [26:18:33<43:47:01, 5.70s/it]
{'loss': 0.9334, 'grad_norm': 1.3736034631729126, 'learning_rate': 3.9626744531651036e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.59, 'epoch': 1.48}
37%|███▋ | 16300/43939 [26:18:33<43:47:01, 5.70s/it]
37%|███▋ | 16301/43939 [26:18:38<43:47:03, 5.70s/it]
37%|███▋ | 16302/43939 [26:18:44<43:51:24, 5.71s/it]
37%|███▋ | 16303/43939 [26:18:50<44:22:33, 5.78s/it]
37%|███▋ | 16304/43939 [26:18:56<44:13:04, 5.76s/it]
37%|███▋ | 16305/43939 [26:19:01<44:08:03, 5.75s/it]
37%|███▋ | 16306/43939 [26:19:07<43:58:09, 5.73s/it]
37%|███▋ | 16307/43939 [26:19:13<43:49:31, 5.71s/it]
37%|███▋ | 16308/43939 [26:19:19<43:51:00, 5.71s/it]
37%|███▋ | 16309/43939 [26:19:24<43:46:31, 5.70s/it]
37%|███▋ | 16310/43939 [26:19:30<43:46:09, 5.70s/it]
37%|███▋ | 16311/43939 [26:19:36<43:44:52, 5.70s/it]
37%|███▋ | 16312/43939 [26:19:41<43:45:36, 5.70s/it]
37%|███▋ | 16313/43939 [26:19:47<43:41:12, 5.69s/it]
37%|███▋ | 16314/43939 [26:19:53<43:44:19, 5.70s/it]
37%|███▋ | 16315/43939 [26:19:58<43:50:25, 5.71s/it]
37%|███▋ | 16316/43939 [26:20:04<43:48:29, 5.71s/it]
37%|███▋ | 16317/43939 [26:20:10<43:44:43, 5.70s/it]
37%|███▋ | 16318/43939 [26:20:16<43:46:45, 5.71s/it]
37%|███▋ | 16319/43939 [26:20:21<43:43:25, 5.70s/it]
37%|███▋ | 16320/43939 [26:20:27<43:45:25, 5.70s/it]
37%|███▋ | 16321/43939 [26:20:33<43:42:14, 5.70s/it]
37%|███▋ | 16322/43939 [26:20:38<43:39:28, 5.69s/it]
37%|███▋ | 16323/43939 [26:20:44<43:33:48, 5.68s/it]
37%|███▋ | 16324/43939 [26:20:50<44:10:32, 5.76s/it]
37%|███▋ | 16325/43939 [26:20:56<44:06:27, 5.75s/it]
37%|███▋ | 16326/43939 [26:21:01<43:56:37, 5.73s/it]
37%|███▋ | 16327/43939 [26:21:07<43:50:26, 5.72s/it]
37%|███▋ | 16328/43939 [26:21:13<43:49:16, 5.71s/it]
37%|███▋ | 16329/43939 [26:21:18<43:50:08, 5.72s/it]
37%|███▋ | 16330/43939 [26:21:24<43:50:09, 5.72s/it]
37%|███▋ | 16331/43939 [26:21:30<43:51:20, 5.72s/it]
37%|███▋ | 16332/43939 [26:21:36<43:49:53, 5.72s/it]
37%|███▋ | 16333/43939 [26:21:41<43:41:52, 5.70s/it]
37%|███▋ | 16334/43939 [26:21:47<43:41:54, 5.70s/it]
37%|███▋ | 16335/43939 [26:21:53<43:42:42, 5.70s/it]
37%|███▋ | 16336/43939 [26:21:58<43:51:02, 5.72s/it]
37%|███▋ | 16337/43939 [26:22:04<43:54:21, 5.73s/it]
37%|███▋ | 16338/43939 [26:22:10<43:50:44, 5.72s/it]
37%|███▋ | 16339/43939 [26:22:16<43:48:05, 5.71s/it]
37%|███▋ | 16340/43939 [26:22:21<43:46:33, 5.71s/it]
37%|███▋ | 16341/43939 [26:22:27<43:48:26, 5.71s/it]
37%|███▋ | 16342/43939 [26:22:33<43:53:28, 5.73s/it]
37%|███▋ | 16343/43939 [26:22:38<43:49:33, 5.72s/it]
37%|███▋ | 16344/43939 [26:22:44<43:48:17, 5.71s/it]
37%|███▋ | 16345/43939 [26:22:50<44:21:24, 5.79s/it]
37%|███▋ | 16346/43939 [26:22:56<44:07:11, 5.76s/it]
37%|███▋ | 16347/43939 [26:23:01<43:58:28, 5.74s/it]
37%|███▋ | 16348/43939 [26:23:07<43:55:29, 5.73s/it]
37%|███▋ | 16349/43939 [26:23:13<43:44:59, 5.71s/it]
37%|███▋ | 16350/43939 [26:23:19<43:44:24, 5.71s/it]
{'loss': 0.9211, 'grad_norm': 1.0454034805297852, 'learning_rate': 3.954609725310405e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.05, 'epoch': 1.49}
37%|███▋ | 16350/43939 [26:23:19<43:44:24, 5.71s/it]
37%|███▋ | 16351/43939 [26:23:24<43:42:48, 5.70s/it]
37%|███▋ | 16352/43939 [26:23:30<43:42:57, 5.70s/it]
37%|███▋ | 16353/43939 [26:23:36<43:42:21, 5.70s/it]
37%|███▋ | 16354/43939 [26:23:41<43:42:37, 5.70s/it]
37%|███▋ | 16355/43939 [26:23:47<43:37:17, 5.69s/it]
37%|███▋ | 16356/43939 [26:23:53<43:35:30, 5.69s/it]
37%|███▋ | 16357/43939 [26:23:58<43:44:18, 5.71s/it]
37%|███▋ | 16358/43939 [26:24:04<43:44:06, 5.71s/it]
37%|███▋ | 16359/43939 [26:24:10<43:51:21, 5.72s/it]
37%|███▋ | 16360/43939 [26:24:16<43:50:37, 5.72s/it]
37%|███▋ | 16361/43939 [26:24:21<43:46:37, 5.71s/it]
37%|███▋ | 16362/43939 [26:24:27<43:45:24, 5.71s/it]
37%|███▋ | 16363/43939 [26:24:33<43:41:35, 5.70s/it]
37%|███▋ | 16364/43939 [26:24:38<43:40:55, 5.70s/it]
37%|███▋ | 16365/43939 [26:24:44<43:38:18, 5.70s/it]
37%|███▋ | 16366/43939 [26:24:50<44:09:01, 5.76s/it]
37%|███▋ | 16367/43939 [26:24:56<44:01:36, 5.75s/it]
37%|███▋ | 16368/43939 [26:25:01<43:55:39, 5.74s/it]
37%|███▋ | 16369/43939 [26:25:07<43:57:34, 5.74s/it]
37%|███▋ | 16370/43939 [26:25:13<43:50:57, 5.73s/it]
37%|███▋ | 16371/43939 [26:25:19<43:45:32, 5.71s/it]
37%|███▋ | 16372/43939 [26:25:24<43:41:56, 5.71s/it]
37%|███▋ | 16373/43939 [26:25:30<43:46:48, 5.72s/it]
37%|███▋ | 16374/43939 [26:25:36<43:44:54, 5.71s/it]
37%|███▋ | 16375/43939 [26:25:41<43:47:50, 5.72s/it]
37%|███▋ | 16376/43939 [26:25:47<43:43:33, 5.71s/it]
37%|███▋ | 16377/43939 [26:25:53<43:41:15, 5.71s/it]
37%|███▋ | 16378/43939 [26:25:59<43:38:23, 5.70s/it]
37%|███▋ | 16379/43939 [26:26:04<43:36:17, 5.70s/it]
37%|███▋ | 16380/43939 [26:26:10<43:34:26, 5.69s/it]
37%|███▋ | 16381/43939 [26:26:16<43:37:55, 5.70s/it]
37%|███▋ | 16382/43939 [26:26:21<43:36:34, 5.70s/it]
37%|███▋ | 16383/43939 [26:26:27<43:32:27, 5.69s/it]
37%|███▋ | 16384/43939 [26:26:33<43:31:24, 5.69s/it]
37%|███▋ | 16385/43939 [26:26:38<43:35:25, 5.70s/it]
37%|███▋ | 16386/43939 [26:26:44<43:34:17, 5.69s/it]
37%|███▋ | 16387/43939 [26:26:50<43:40:16, 5.71s/it]
37%|███▋ | 16388/43939 [26:26:56<44:16:57, 5.79s/it]
37%|███▋ | 16389/43939 [26:27:01<44:06:42, 5.76s/it]
37%|███▋ | 16390/43939 [26:27:07<43:57:10, 5.74s/it]
37%|███▋ | 16391/43939 [26:27:13<43:52:35, 5.73s/it]
37%|███▋ | 16392/43939 [26:27:19<43:46:46, 5.72s/it]
37%|███▋ | 16393/43939 [26:27:24<43:39:31, 5.71s/it]
37%|███▋ | 16394/43939 [26:27:30<43:36:47, 5.70s/it]
37%|███▋ | 16395/43939 [26:27:36<43:36:05, 5.70s/it]
37%|███▋ | 16396/43939 [26:27:41<43:41:00, 5.71s/it]
37%|███▋ | 16397/43939 [26:27:47<43:43:40, 5.72s/it]
37%|███▋ | 16398/43939 [26:27:53<43:43:25, 5.72s/it]
37%|███▋ | 16399/43939 [26:27:59<43:39:49, 5.71s/it]
37%|███▋ | 16400/43939 [26:28:04<43:39:16, 5.71s/it]
{'loss': 0.9063, 'grad_norm': 1.0870639085769653, 'learning_rate': 3.946522047569005e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.05, 'epoch': 1.49}
37%|███▋ | 16400/43939 [26:28:04<43:39:16, 5.71s/it]
37%|███▋ | 16401/43939 [26:28:10<43:43:14, 5.72s/it]
37%|███▋ | 16402/43939 [26:28:16<43:37:44, 5.70s/it]
37%|███▋ | 16403/43939 [26:28:21<43:38:42, 5.71s/it]
37%|███▋ | 16404/43939 [26:28:27<43:34:56, 5.70s/it]
37%|███▋ | 16405/43939 [26:28:33<43:32:17, 5.69s/it]
37%|███▋ | 16406/43939 [26:28:38<43:31:01, 5.69s/it]
37%|███▋ | 16407/43939 [26:28:44<43:32:05, 5.69s/it]
37%|███▋ | 16408/43939 [26:28:50<43:33:33, 5.70s/it]
37%|███▋ | 16409/43939 [26:28:56<43:59:23, 5.75s/it]
37%|███▋ | 16410/43939 [26:29:01<43:52:36, 5.74s/it]
37%|███▋ | 16411/43939 [26:29:07<43:55:53, 5.75s/it]
37%|███▋ | 16412/43939 [26:29:13<43:53:16, 5.74s/it]
37%|███▋ | 16413/43939 [26:29:19<43:47:41, 5.73s/it]
37%|███▋ | 16414/43939 [26:29:24<43:44:05, 5.72s/it]
37%|███▋ | 16415/43939 [26:29:30<43:40:36, 5.71s/it]
37%|███▋ | 16416/43939 [26:29:36<43:37:21, 5.71s/it]
37%|███▋ | 16417/43939 [26:29:41<43:39:34, 5.71s/it]
37%|███▋ | 16418/43939 [26:29:47<43:37:11, 5.71s/it]
37%|███▋ | 16419/43939 [26:29:53<43:36:42, 5.71s/it]
37%|███▋ | 16420/43939 [26:29:58<43:35:34, 5.70s/it]
37%|███▋ | 16421/43939 [26:30:04<43:37:44, 5.71s/it]
37%|███▋ | 16422/43939 [26:30:10<43:35:44, 5.70s/it]
37%|███▋ | 16423/43939 [26:30:16<43:36:14, 5.70s/it]
37%|███▋ | 16424/43939 [26:30:21<43:37:15, 5.71s/it]
37%|███▋ | 16425/43939 [26:30:27<43:34:31, 5.70s/it]
37%|███▋ | 16426/43939 [26:30:33<43:34:01, 5.70s/it]
37%|███▋ | 16427/43939 [26:30:38<43:32:34, 5.70s/it]
37%|███▋ | 16428/43939 [26:30:44<43:33:15, 5.70s/it]
37%|███▋ | 16429/43939 [26:30:50<43:34:11, 5.70s/it]
37%|███▋ | 16430/43939 [26:30:56<44:02:11, 5.76s/it]
37%|███▋ | 16431/43939 [26:31:01<43:51:27, 5.74s/it]
37%|███▋ | 16432/43939 [26:31:07<43:54:40, 5.75s/it]
37%|███▋ | 16433/43939 [26:31:13<43:49:30, 5.74s/it]
37%|███▋ | 16434/43939 [26:31:19<43:41:04, 5.72s/it]
37%|███▋ | 16435/43939 [26:31:24<43:36:34, 5.71s/it]
37%|███▋ | 16436/43939 [26:31:30<43:30:08, 5.69s/it]
37%|███▋ | 16437/43939 [26:31:36<43:42:41, 5.72s/it]
37%|███▋ | 16438/43939 [26:31:41<43:36:53, 5.71s/it]
37%|███▋ | 16439/43939 [26:31:47<43:32:44, 5.70s/it]
37%|███▋ | 16440/43939 [26:31:53<43:36:14, 5.71s/it]
37%|███▋ | 16441/43939 [26:31:58<43:33:30, 5.70s/it]
37%|███▋ | 16442/43939 [26:32:04<43:32:29, 5.70s/it]
37%|███▋ | 16443/43939 [26:32:10<43:34:00, 5.70s/it]
37%|███▋ | 16444/43939 [26:32:16<43:33:25, 5.70s/it]
37%|███▋ | 16445/43939 [26:32:21<43:34:09, 5.70s/it]
37%|███▋ | 16446/43939 [26:32:27<43:34:52, 5.71s/it]
37%|███▋ | 16447/43939 [26:32:33<43:34:52, 5.71s/it]
37%|███▋ | 16448/43939 [26:32:38<43:41:31, 5.72s/it]
37%|███▋ | 16449/43939 [26:32:44<43:47:23, 5.73s/it]
37%|███▋ | 16450/43939 [26:32:50<43:41:37, 5.72s/it]
{'loss': 0.9153, 'grad_norm': 0.9848358631134033, 'learning_rate': 3.938411547543023e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.41, 'epoch': 1.5}
37%|███▋ | 16450/43939 [26:32:50<43:41:37, 5.72s/it]
37%|███▋ | 16451/43939 [26:32:56<44:12:03, 5.79s/it]
37%|███▋ | 16452/43939 [26:33:02<44:01:57, 5.77s/it]
37%|███▋ | 16453/43939 [26:33:07<43:52:27, 5.75s/it]
37%|███▋ | 16454/43939 [26:33:13<43:44:06, 5.73s/it]
37%|███▋ | 16455/43939 [26:33:19<43:42:06, 5.72s/it]
37%|███▋ | 16456/43939 [26:33:24<43:37:11, 5.71s/it]
37%|███▋ | 16457/43939 [26:33:30<43:34:32, 5.71s/it]
37%|███▋ | 16458/43939 [26:33:36<43:39:00, 5.72s/it]
37%|███▋ | 16459/43939 [26:33:41<43:35:43, 5.71s/it]
37%|███▋ | 16460/43939 [26:33:47<43:32:51, 5.71s/it]
37%|███▋ | 16461/43939 [26:33:53<43:32:25, 5.70s/it]
37%|███▋ | 16462/43939 [26:33:59<43:31:32, 5.70s/it]
37%|███▋ | 16463/43939 [26:34:04<43:35:21, 5.71s/it]
37%|███▋ | 16464/43939 [26:34:10<43:34:06, 5.71s/it]
37%|███▋ | 16465/43939 [26:34:16<43:35:23, 5.71s/it]
37%|███▋ | 16466/43939 [26:34:21<43:29:38, 5.70s/it]
37%|███▋ | 16467/43939 [26:34:27<43:26:19, 5.69s/it]
37%|███▋ | 16468/43939 [26:34:33<43:25:21, 5.69s/it]
37%|███▋ | 16469/43939 [26:34:38<43:28:44, 5.70s/it]
37%|███▋ | 16470/43939 [26:34:44<43:32:15, 5.71s/it]
37%|███▋ | 16471/43939 [26:34:50<43:29:30, 5.70s/it]
37%|███▋ | 16472/43939 [26:34:56<44:04:32, 5.78s/it]
37%|███▋ | 16473/43939 [26:35:02<43:54:32, 5.76s/it]
37%|███▋ | 16474/43939 [26:35:07<43:42:01, 5.73s/it]
37%|███▋ | 16475/43939 [26:35:13<43:38:16, 5.72s/it]
37%|███▋ | 16476/43939 [26:35:19<43:36:53, 5.72s/it]
37%|███▋ | 16477/43939 [26:35:24<43:30:13, 5.70s/it]
38%|███▊ | 16478/43939 [26:35:30<43:25:27, 5.69s/it]
38%|███▊ | 16479/43939 [26:35:36<43:26:12, 5.69s/it][2025-11-29 20:39:49,463] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-16479
38%|███▊ | 16480/43939 [26:35:43<48:13:09, 6.32s/it]
38%|███▊ | 16481/43939 [26:35:49<46:45:35, 6.13s/it]
38%|███▊ | 16482/43939 [26:35:55<45:49:51, 6.01s/it]
38%|███▊ | 16483/43939 [26:36:01<45:10:37, 5.92s/it]
38%|███▊ | 16484/43939 [26:36:06<44:37:23, 5.85s/it]
38%|███▊ | 16485/43939 [26:36:12<44:14:47, 5.80s/it]
38%|███▊ | 16486/43939 [26:36:18<44:05:26, 5.78s/it]
38%|███▊ | 16487/43939 [26:36:23<43:53:42, 5.76s/it]
38%|███▊ | 16488/43939 [26:36:29<43:44:55, 5.74s/it]
38%|███▊ | 16489/43939 [26:36:35<43:41:31, 5.73s/it]
38%|███▊ | 16490/43939 [26:36:41<43:41:35, 5.73s/it]
38%|███▊ | 16491/43939 [26:36:46<43:45:27, 5.74s/it]
38%|███▊ | 16492/43939 [26:36:52<44:05:41, 5.78s/it]
38%|███▊ | 16493/43939 [26:36:58<43:56:13, 5.76s/it]
38%|███▊ | 16494/43939 [26:37:04<43:48:37, 5.75s/it]
38%|███▊ | 16495/43939 [26:37:09<43:37:52, 5.72s/it]
38%|███▊ | 16496/43939 [26:37:15<43:35:15, 5.72s/it]
38%|███▊ | 16497/43939 [26:37:21<43:34:51, 5.72s/it]
38%|███▊ | 16498/43939 [26:37:26<43:36:13, 5.72s/it]
38%|███▊ | 16499/43939 [26:37:32<43:37:26, 5.72s/it]
38%|███▊ | 16500/43939 [26:37:38<43:38:21, 5.73s/it]
{'loss': 0.9268, 'grad_norm': 1.014233946800232, 'learning_rate': 3.930278353194658e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.53, 'epoch': 1.5}
38%|███▊ | 16500/43939 [26:37:38<43:38:21, 5.73s/it]
38%|███▊ | 16501/43939 [26:37:44<43:32:35, 5.71s/it]
38%|███▊ | 16502/43939 [26:37:49<43:25:59, 5.70s/it]
38%|███▊ | 16503/43939 [26:37:55<43:37:51, 5.73s/it]
38%|███▊ | 16504/43939 [26:38:01<43:38:55, 5.73s/it]
38%|███▊ | 16505/43939 [26:38:06<43:39:38, 5.73s/it]
38%|███▊ | 16506/43939 [26:38:12<43:30:12, 5.71s/it]
38%|███▊ | 16507/43939 [26:38:18<43:26:12, 5.70s/it]
38%|███▊ | 16508/43939 [26:38:23<43:25:33, 5.70s/it]
38%|███▊ | 16509/43939 [26:38:29<43:25:43, 5.70s/it]
38%|███▊ | 16510/43939 [26:38:35<43:25:29, 5.70s/it]
38%|███▊ | 16511/43939 [26:38:41<43:23:10, 5.69s/it]
38%|███▊ | 16512/43939 [26:38:46<43:23:19, 5.70s/it]
38%|███▊ | 16513/43939 [26:38:52<43:52:34, 5.76s/it]
38%|███▊ | 16514/43939 [26:38:58<43:48:00, 5.75s/it]
38%|███▊ | 16515/43939 [26:39:04<43:41:22, 5.74s/it]
38%|███▊ | 16516/43939 [26:39:09<43:31:11, 5.71s/it]
38%|███▊ | 16517/43939 [26:39:15<43:28:37, 5.71s/it]
38%|███▊ | 16518/43939 [26:39:21<43:25:34, 5.70s/it]
38%|███▊ | 16519/43939 [26:39:26<43:31:27, 5.71s/it]
38%|███▊ | 16520/43939 [26:39:32<43:26:47, 5.70s/it]
38%|███▊ | 16521/43939 [26:39:38<43:23:34, 5.70s/it]
38%|███▊ | 16522/43939 [26:39:43<43:21:46, 5.69s/it]
38%|███▊ | 16523/43939 [26:39:49<43:23:53, 5.70s/it]
38%|███▊ | 16524/43939 [26:39:55<43:26:48, 5.71s/it]
38%|███▊ | 16525/43939 [26:40:01<43:28:24, 5.71s/it]
38%|███▊ | 16526/43939 [26:40:06<43:24:35, 5.70s/it]
38%|███▊ | 16527/43939 [26:40:12<43:28:40, 5.71s/it]
38%|███▊ | 16528/43939 [26:40:18<43:24:54, 5.70s/it]
38%|███▊ | 16529/43939 [26:40:23<43:27:17, 5.71s/it]
38%|███▊ | 16530/43939 [26:40:29<43:23:33, 5.70s/it]
38%|███▊ | 16531/43939 [26:40:35<43:20:38, 5.69s/it]
38%|███▊ | 16532/43939 [26:40:41<43:27:07, 5.71s/it]
38%|███▊ | 16533/43939 [26:40:46<43:25:55, 5.71s/it]
38%|███▊ | 16534/43939 [26:40:52<43:58:48, 5.78s/it]
38%|███▊ | 16535/43939 [26:40:58<43:49:49, 5.76s/it]
38%|███▊ | 16536/43939 [26:41:04<43:47:08, 5.75s/it]
38%|███▊ | 16537/43939 [26:41:09<43:40:03, 5.74s/it]
38%|███▊ | 16538/43939 [26:41:15<43:45:01, 5.75s/it]
38%|███▊ | 16539/43939 [26:41:21<43:34:06, 5.72s/it]
38%|███▊ | 16540/43939 [26:41:26<43:28:33, 5.71s/it]
38%|███▊ | 16541/43939 [26:41:32<43:28:23, 5.71s/it]
38%|███▊ | 16542/43939 [26:41:38<43:27:15, 5.71s/it]
38%|███▊ | 16543/43939 [26:41:44<43:30:07, 5.72s/it]
38%|███▊ | 16544/43939 [26:41:49<43:25:19, 5.71s/it]
38%|███▊ | 16545/43939 [26:41:55<43:22:26, 5.70s/it]
38%|███▊ | 16546/43939 [26:42:01<43:18:52, 5.69s/it]
38%|███▊ | 16547/43939 [26:42:06<43:18:44, 5.69s/it]
38%|███▊ | 16548/43939 [26:42:12<43:18:21, 5.69s/it]
38%|███▊ | 16549/43939 [26:42:18<43:21:43, 5.70s/it]
38%|███▊ | 16550/43939 [26:42:23<43:15:25, 5.69s/it]
{'loss': 0.9417, 'grad_norm': 1.2883192300796509, 'learning_rate': 3.9221225928441634e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 455.43, 'epoch': 1.51}
38%|███▊ | 16550/43939 [26:42:23<43:15:25, 5.69s/it]
38%|███▊ | 16551/43939 [26:42:29<43:17:30, 5.69s/it]
38%|███▊ | 16552/43939 [26:42:35<43:20:20, 5.70s/it]
38%|███▊ | 16553/43939 [26:42:40<43:18:13, 5.69s/it]
38%|███▊ | 16554/43939 [26:42:46<43:51:16, 5.77s/it]
38%|███▊ | 16555/43939 [26:42:52<43:45:31, 5.75s/it]
38%|███▊ | 16556/43939 [26:42:58<43:43:11, 5.75s/it]
38%|███▊ | 16557/43939 [26:43:04<43:36:25, 5.73s/it]
38%|███▊ | 16558/43939 [26:43:09<43:28:00, 5.71s/it]
38%|███▊ | 16559/43939 [26:43:15<43:33:29, 5.73s/it]
38%|███▊ | 16560/43939 [26:43:21<43:30:22, 5.72s/it]
38%|███▊ | 16561/43939 [26:43:26<43:29:30, 5.72s/it]
38%|███▊ | 16562/43939 [26:43:32<43:31:01, 5.72s/it]
38%|███▊ | 16563/43939 [26:43:38<43:27:40, 5.72s/it]
38%|███▊ | 16564/43939 [26:43:44<43:27:26, 5.71s/it]
38%|███▊ | 16565/43939 [26:43:49<43:24:39, 5.71s/it]
38%|███▊ | 16566/43939 [26:43:55<43:28:00, 5.72s/it]
38%|███▊ | 16567/43939 [26:44:01<43:29:19, 5.72s/it]
38%|███▊ | 16568/43939 [26:44:06<43:23:48, 5.71s/it]
38%|███▊ | 16569/43939 [26:44:12<43:19:09, 5.70s/it]
38%|███▊ | 16570/43939 [26:44:18<43:19:44, 5.70s/it]
38%|███▊ | 16571/43939 [26:44:23<43:16:04, 5.69s/it]
38%|███▊ | 16572/43939 [26:44:29<43:17:13, 5.69s/it]
38%|███▊ | 16573/43939 [26:44:35<43:16:41, 5.69s/it]
38%|███▊ | 16574/43939 [26:44:41<43:18:56, 5.70s/it]
38%|███▊ | 16575/43939 [26:44:46<43:24:33, 5.71s/it]
38%|███▊ | 16576/43939 [26:44:52<43:53:10, 5.77s/it]
38%|███▊ | 16577/43939 [26:44:58<43:42:40, 5.75s/it]
38%|███▊ | 16578/43939 [26:45:04<43:33:33, 5.73s/it]
38%|███▊ | 16579/43939 [26:45:09<43:26:23, 5.72s/it]
38%|███▊ | 16580/43939 [26:45:15<43:26:01, 5.72s/it]
38%|███▊ | 16581/43939 [26:45:21<43:22:24, 5.71s/it]
38%|███▊ | 16582/43939 [26:45:26<43:21:11, 5.70s/it]
38%|███▊ | 16583/43939 [26:45:32<43:19:39, 5.70s/it]
38%|███▊ | 16584/43939 [26:45:38<43:12:54, 5.69s/it]
38%|███▊ | 16585/43939 [26:45:43<43:17:11, 5.70s/it]
38%|███▊ | 16586/43939 [26:45:49<43:23:37, 5.71s/it]
38%|███▊ | 16587/43939 [26:45:55<43:19:26, 5.70s/it]
38%|███▊ | 16588/43939 [26:46:01<43:16:07, 5.70s/it]
38%|███▊ | 16589/43939 [26:46:06<43:14:06, 5.69s/it]
38%|███▊ | 16590/43939 [26:46:12<43:18:18, 5.70s/it]
38%|███▊ | 16591/43939 [26:46:18<43:14:33, 5.69s/it]
38%|███▊ | 16592/43939 [26:46:23<43:15:08, 5.69s/it]
38%|███▊ | 16593/43939 [26:46:29<43:18:00, 5.70s/it]
38%|███▊ | 16594/43939 [26:46:35<43:18:59, 5.70s/it]
38%|███▊ | 16595/43939 [26:46:40<43:15:57, 5.70s/it]
38%|███▊ | 16596/43939 [26:46:46<43:43:12, 5.76s/it]
38%|███▊ | 16597/43939 [26:46:52<43:34:16, 5.74s/it]
38%|███▊ | 16598/43939 [26:46:58<43:29:27, 5.73s/it]
38%|███▊ | 16599/43939 [26:47:03<43:22:39, 5.71s/it]
38%|███▊ | 16600/43939 [26:47:09<43:17:35, 5.70s/it]
{'loss': 0.9091, 'grad_norm': 1.175374150276184, 'learning_rate': 3.913944395167825e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.19, 'epoch': 1.51}
38%|███▊ | 16600/43939 [26:47:09<43:17:35, 5.70s/it]
38%|███▊ | 16601/43939 [26:47:15<43:16:28, 5.70s/it]
38%|███▊ | 16602/43939 [26:47:20<43:19:19, 5.71s/it]
38%|███▊ | 16603/43939 [26:47:26<43:14:44, 5.70s/it]
38%|███▊ | 16604/43939 [26:47:32<43:11:33, 5.69s/it]
38%|███▊ | 16605/43939 [26:47:37<43:07:23, 5.68s/it]
38%|███▊ | 16606/43939 [26:47:43<43:06:36, 5.68s/it]
38%|███▊ | 16607/43939 [26:47:49<43:09:20, 5.68s/it]
38%|███▊ | 16608/43939 [26:47:55<43:09:39, 5.69s/it]
38%|███▊ | 16609/43939 [26:48:00<43:10:07, 5.69s/it]
38%|███▊ | 16610/43939 [26:48:06<43:16:20, 5.70s/it]
38%|███▊ | 16611/43939 [26:48:12<43:17:25, 5.70s/it]
38%|███▊ | 16612/43939 [26:48:17<43:25:24, 5.72s/it]
38%|███▊ | 16613/43939 [26:48:23<43:20:16, 5.71s/it]
38%|███▊ | 16614/43939 [26:48:29<43:13:10, 5.69s/it]
38%|███▊ | 16615/43939 [26:48:34<43:10:00, 5.69s/it]
38%|███▊ | 16616/43939 [26:48:40<43:13:11, 5.69s/it]
38%|███▊ | 16617/43939 [26:48:46<43:48:37, 5.77s/it]
38%|███▊ | 16618/43939 [26:48:52<43:36:22, 5.75s/it]
38%|███▊ | 16619/43939 [26:48:58<43:33:44, 5.74s/it]
38%|███▊ | 16620/43939 [26:49:03<43:34:54, 5.74s/it]
38%|███▊ | 16621/43939 [26:49:09<43:23:48, 5.72s/it]
38%|███▊ | 16622/43939 [26:49:15<43:25:32, 5.72s/it]
38%|███▊ | 16623/43939 [26:49:20<43:21:49, 5.71s/it]
38%|███▊ | 16624/43939 [26:49:26<43:24:51, 5.72s/it]
38%|███▊ | 16625/43939 [26:49:32<43:31:17, 5.74s/it]
38%|███▊ | 16626/43939 [26:49:38<43:24:45, 5.72s/it]
38%|███▊ | 16627/43939 [26:49:43<43:16:14, 5.70s/it]
38%|███▊ | 16628/43939 [26:49:49<43:14:01, 5.70s/it]
38%|███▊ | 16629/43939 [26:49:55<43:10:11, 5.69s/it]
38%|███▊ | 16630/43939 [26:50:00<43:11:39, 5.69s/it]
38%|███▊ | 16631/43939 [26:50:06<43:20:53, 5.71s/it]
38%|███▊ | 16632/43939 [26:50:12<43:20:38, 5.71s/it]
38%|███▊ | 16633/43939 [26:50:17<43:20:15, 5.71s/it]
38%|███▊ | 16634/43939 [26:50:23<43:15:12, 5.70s/it]
38%|███▊ | 16635/43939 [26:50:29<43:11:16, 5.69s/it]
38%|███▊ | 16636/43939 [26:50:35<43:15:08, 5.70s/it]
38%|███▊ | 16637/43939 [26:50:40<43:24:20, 5.72s/it]
38%|███▊ | 16638/43939 [26:50:46<43:18:50, 5.71s/it]
38%|███▊ | 16639/43939 [26:50:52<43:47:44, 5.78s/it]
38%|███▊ | 16640/43939 [26:50:58<43:31:17, 5.74s/it]
38%|███▊ | 16641/43939 [26:51:03<43:26:21, 5.73s/it]
38%|███▊ | 16642/43939 [26:51:09<43:19:22, 5.71s/it]
38%|███▊ | 16643/43939 [26:51:15<43:21:09, 5.72s/it]
38%|███▊ | 16644/43939 [26:51:20<43:15:43, 5.71s/it]
38%|███▊ | 16645/43939 [26:51:26<43:14:25, 5.70s/it]
38%|███▊ | 16646/43939 [26:51:32<43:14:42, 5.70s/it]
38%|███▊ | 16647/43939 [26:51:37<43:16:11, 5.71s/it]
38%|███▊ | 16648/43939 [26:51:43<43:15:44, 5.71s/it]
38%|███▊ | 16649/43939 [26:51:49<43:18:56, 5.71s/it]
38%|███▊ | 16650/43939 [26:51:55<43:25:54, 5.73s/it]
{'loss': 0.923, 'grad_norm': 1.1349321603775024, 'learning_rate': 3.9057438891959284e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.05, 'epoch': 1.52}
38%|███▊ | 16650/43939 [26:51:55<43:25:54, 5.73s/it]
38%|███▊ | 16651/43939 [26:52:00<43:24:58, 5.73s/it]
38%|███▊ | 16652/43939 [26:52:06<43:20:46, 5.72s/it]
38%|███▊ | 16653/43939 [26:52:12<43:26:59, 5.73s/it]
38%|███▊ | 16654/43939 [26:52:18<43:22:05, 5.72s/it]
38%|███▊ | 16655/43939 [26:52:23<43:16:57, 5.71s/it]
38%|███▊ | 16656/43939 [26:52:29<43:17:07, 5.71s/it]
38%|███▊ | 16657/43939 [26:52:35<43:22:16, 5.72s/it]
38%|███▊ | 16658/43939 [26:52:40<43:24:23, 5.73s/it]
38%|███▊ | 16659/43939 [26:52:46<43:24:56, 5.73s/it]
38%|███▊ | 16660/43939 [26:52:52<43:54:11, 5.79s/it]
38%|███▊ | 16661/43939 [26:52:58<43:42:45, 5.77s/it]
38%|███▊ | 16662/43939 [26:53:04<43:36:58, 5.76s/it]
38%|███▊ | 16663/43939 [26:53:09<43:31:40, 5.75s/it]
38%|███▊ | 16664/43939 [26:53:15<43:25:03, 5.73s/it]
38%|███▊ | 16665/43939 [26:53:21<43:24:03, 5.73s/it]
38%|███▊ | 16666/43939 [26:53:26<43:23:22, 5.73s/it]
38%|███▊ | 16667/43939 [26:53:32<43:23:29, 5.73s/it]
38%|███▊ | 16668/43939 [26:53:38<43:20:24, 5.72s/it]
38%|███▊ | 16669/43939 [26:53:44<43:14:10, 5.71s/it]
38%|███▊ | 16670/43939 [26:53:49<43:09:27, 5.70s/it]
38%|███▊ | 16671/43939 [26:53:55<43:13:02, 5.71s/it]
38%|███▊ | 16672/43939 [26:54:01<43:07:30, 5.69s/it]
38%|███▊ | 16673/43939 [26:54:06<43:05:36, 5.69s/it]
38%|███▊ | 16674/43939 [26:54:12<43:12:41, 5.71s/it]
38%|███▊ | 16675/43939 [26:54:18<43:13:15, 5.71s/it]
38%|███▊ | 16676/43939 [26:54:23<43:12:18, 5.71s/it]
38%|███▊ | 16677/43939 [26:54:29<43:13:58, 5.71s/it]
38%|███▊ | 16678/43939 [26:54:35<43:20:03, 5.72s/it]
38%|███▊ | 16679/43939 [26:54:41<43:18:15, 5.72s/it]
38%|███▊ | 16680/43939 [26:54:46<43:15:21, 5.71s/it]
38%|███▊ | 16681/43939 [26:54:52<43:44:03, 5.78s/it]
38%|███▊ | 16682/43939 [26:54:58<43:33:57, 5.75s/it]
38%|███▊ | 16683/43939 [26:55:04<43:21:40, 5.73s/it]
38%|███▊ | 16684/43939 [26:55:09<43:19:38, 5.72s/it]
38%|███▊ | 16685/43939 [26:55:15<43:19:36, 5.72s/it]
38%|███▊ | 16686/43939 [26:55:21<43:18:46, 5.72s/it]
38%|███▊ | 16687/43939 [26:55:27<43:20:13, 5.72s/it]
38%|███▊ | 16688/43939 [26:55:32<43:20:21, 5.73s/it]
38%|███▊ | 16689/43939 [26:55:38<43:15:54, 5.72s/it]
38%|███▊ | 16690/43939 [26:55:44<43:17:24, 5.72s/it]
38%|███▊ | 16691/43939 [26:55:49<43:12:14, 5.71s/it]
38%|███▊ | 16692/43939 [26:55:55<43:13:28, 5.71s/it]
38%|███▊ | 16693/43939 [26:56:01<43:11:33, 5.71s/it]
38%|███▊ | 16694/43939 [26:56:06<43:15:24, 5.72s/it]
38%|███▊ | 16695/43939 [26:56:12<43:15:41, 5.72s/it]
38%|███▊ | 16696/43939 [26:56:18<43:16:46, 5.72s/it]
38%|███▊ | 16697/43939 [26:56:24<43:14:54, 5.72s/it]
38%|███▊ | 16698/43939 [26:56:29<43:13:59, 5.71s/it]
38%|███▊ | 16699/43939 [26:56:35<43:15:09, 5.72s/it]
38%|███▊ | 16700/43939 [26:56:41<43:12:29, 5.71s/it]
{'loss': 0.94, 'grad_norm': 1.1461342573165894, 'learning_rate': 3.8975212043107274e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 526.13, 'epoch': 1.52}
38%|███▊ | 16700/43939 [26:56:41<43:12:29, 5.71s/it]
38%|███▊ | 16701/43939 [26:56:46<43:11:16, 5.71s/it]
38%|███▊ | 16702/43939 [26:56:52<43:07:33, 5.70s/it]
38%|███▊ | 16703/43939 [26:56:58<43:38:54, 5.77s/it]
38%|███▊ | 16704/43939 [26:57:04<43:29:55, 5.75s/it]
38%|███▊ | 16705/43939 [26:57:10<43:27:19, 5.74s/it]
38%|███▊ | 16706/43939 [26:57:15<43:22:42, 5.73s/it]
38%|███▊ | 16707/43939 [26:57:21<43:16:53, 5.72s/it]
38%|███▊ | 16708/43939 [26:57:27<43:11:34, 5.71s/it]
38%|███▊ | 16709/43939 [26:57:32<43:10:48, 5.71s/it]
38%|███▊ | 16710/43939 [26:57:38<43:10:52, 5.71s/it]
38%|███▊ | 16711/43939 [26:57:44<43:06:39, 5.70s/it]
38%|███▊ | 16712/43939 [26:57:49<43:11:09, 5.71s/it]
38%|███▊ | 16713/43939 [26:57:55<43:08:30, 5.70s/it]
38%|███▊ | 16714/43939 [26:58:01<43:12:01, 5.71s/it]
38%|███▊ | 16715/43939 [26:58:07<43:19:43, 5.73s/it]
38%|███▊ | 16716/43939 [26:58:12<43:19:07, 5.73s/it]
38%|███▊ | 16717/43939 [26:58:18<43:17:24, 5.72s/it]
38%|███▊ | 16718/43939 [26:58:24<43:09:23, 5.71s/it]
38%|███▊ | 16719/43939 [26:58:29<43:04:25, 5.70s/it]
38%|███▊ | 16720/43939 [26:58:35<43:01:27, 5.69s/it]
38%|███▊ | 16721/43939 [26:58:41<43:02:09, 5.69s/it]
38%|███▊ | 16722/43939 [26:58:46<43:02:12, 5.69s/it]
38%|███▊ | 16723/43939 [26:58:52<43:00:47, 5.69s/it]
38%|███▊ | 16724/43939 [26:58:58<43:37:06, 5.77s/it]
38%|███▊ | 16725/43939 [26:59:04<43:28:53, 5.75s/it]
38%|███▊ | 16726/43939 [26:59:10<43:23:33, 5.74s/it]
38%|███▊ | 16727/43939 [26:59:15<43:17:38, 5.73s/it]
38%|███▊ | 16728/43939 [26:59:21<43:10:53, 5.71s/it]
38%|███▊ | 16729/43939 [26:59:27<43:10:46, 5.71s/it]
38%|███▊ | 16730/43939 [26:59:32<43:08:33, 5.71s/it]
38%|███▊ | 16731/43939 [26:59:38<43:07:22, 5.71s/it]
38%|███▊ | 16732/43939 [26:59:44<43:12:37, 5.72s/it]
38%|███▊ | 16733/43939 [26:59:49<43:09:32, 5.71s/it]
38%|███▊ | 16734/43939 [26:59:55<43:07:35, 5.71s/it]
38%|███▊ | 16735/43939 [27:00:01<43:08:06, 5.71s/it]
38%|███▊ | 16736/43939 [27:00:07<43:04:57, 5.70s/it]
38%|███▊ | 16737/43939 [27:00:12<43:04:48, 5.70s/it]
38%|███▊ | 16738/43939 [27:00:18<43:00:30, 5.69s/it]
38%|███▊ | 16739/43939 [27:00:24<43:03:15, 5.70s/it]
38%|███▊ | 16740/43939 [27:00:29<43:08:19, 5.71s/it]
38%|███▊ | 16741/43939 [27:00:35<43:11:42, 5.72s/it]
38%|███▊ | 16742/43939 [27:00:41<43:13:45, 5.72s/it]
38%|███▊ | 16743/43939 [27:00:47<43:06:33, 5.71s/it]
38%|███▊ | 16744/43939 [27:00:52<43:06:34, 5.71s/it]
38%|███▊ | 16745/43939 [27:00:58<43:40:06, 5.78s/it]
38%|███▊ | 16746/43939 [27:01:04<43:27:58, 5.75s/it]
38%|███▊ | 16747/43939 [27:01:10<43:15:29, 5.73s/it]
38%|███▊ | 16748/43939 [27:01:15<43:15:46, 5.73s/it]
38%|███▊ | 16749/43939 [27:01:21<43:12:56, 5.72s/it]
38%|███▊ | 16750/43939 [27:01:27<43:06:53, 5.71s/it]
{'loss': 0.9189, 'grad_norm': 1.3331114053726196, 'learning_rate': 3.8892764702443986e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.65, 'epoch': 1.52}
38%|███▊ | 16750/43939 [27:01:27<43:06:53, 5.71s/it]
38%|███▊ | 16751/43939 [27:01:32<43:01:39, 5.70s/it]
38%|███▊ | 16752/43939 [27:01:38<43:02:29, 5.70s/it]
38%|███▊ | 16753/43939 [27:01:44<43:06:00, 5.71s/it]
38%|███▊ | 16754/43939 [27:01:49<43:05:47, 5.71s/it]
38%|███▊ | 16755/43939 [27:01:55<43:04:12, 5.70s/it]
38%|███▊ | 16756/43939 [27:02:01<43:03:29, 5.70s/it]
38%|███▊ | 16757/43939 [27:02:07<43:08:14, 5.71s/it]
38%|███▊ | 16758/43939 [27:02:12<43:04:32, 5.71s/it]
38%|███▊ | 16759/43939 [27:02:18<42:59:01, 5.69s/it]
38%|███▊ | 16760/43939 [27:02:24<42:59:37, 5.69s/it]
38%|███▊ | 16761/43939 [27:02:29<43:03:01, 5.70s/it]
38%|███▊ | 16762/43939 [27:02:35<43:01:10, 5.70s/it]
38%|███▊ | 16763/43939 [27:02:41<42:59:08, 5.69s/it]
38%|███▊ | 16764/43939 [27:02:46<42:59:22, 5.70s/it]
38%|███▊ | 16765/43939 [27:02:52<43:03:36, 5.70s/it]
38%|███▊ | 16766/43939 [27:02:58<43:27:44, 5.76s/it]
38%|███▊ | 16767/43939 [27:03:04<43:25:01, 5.75s/it]
38%|███▊ | 16768/43939 [27:03:09<43:15:30, 5.73s/it]
38%|███▊ | 16769/43939 [27:03:15<43:13:27, 5.73s/it]
38%|███▊ | 16770/43939 [27:03:21<43:06:18, 5.71s/it]
38%|███▊ | 16771/43939 [27:03:27<43:06:50, 5.71s/it]
38%|███▊ | 16772/43939 [27:03:32<43:08:45, 5.72s/it]
38%|███▊ | 16773/43939 [27:03:38<43:06:08, 5.71s/it]
38%|███▊ | 16774/43939 [27:03:44<43:01:37, 5.70s/it]
38%|███▊ | 16775/43939 [27:03:49<42:53:43, 5.68s/it]
38%|███▊ | 16776/43939 [27:03:55<42:56:35, 5.69s/it]
38%|███▊ | 16777/43939 [27:04:01<42:56:55, 5.69s/it]
38%|███▊ | 16778/43939 [27:04:06<42:55:27, 5.69s/it]
38%|███▊ | 16779/43939 [27:04:12<42:57:36, 5.69s/it]
38%|███▊ | 16780/43939 [27:04:18<42:56:41, 5.69s/it]
38%|███▊ | 16781/43939 [27:04:23<42:52:37, 5.68s/it]
38%|███▊ | 16782/43939 [27:04:29<42:54:55, 5.69s/it]
38%|███▊ | 16783/43939 [27:04:35<43:00:09, 5.70s/it]
38%|███▊ | 16784/43939 [27:04:41<43:00:48, 5.70s/it]
38%|███▊ | 16785/43939 [27:04:46<43:02:55, 5.71s/it]
38%|███▊ | 16786/43939 [27:04:52<42:59:59, 5.70s/it]
38%|███▊ | 16787/43939 [27:04:58<43:01:00, 5.70s/it]
38%|███▊ | 16788/43939 [27:05:04<43:33:33, 5.78s/it]
38%|███▊ | 16789/43939 [27:05:09<43:26:14, 5.76s/it]
38%|███▊ | 16790/43939 [27:05:15<43:22:34, 5.75s/it]
38%|███▊ | 16791/43939 [27:05:21<43:17:22, 5.74s/it]
38%|███▊ | 16792/43939 [27:05:27<43:15:46, 5.74s/it]
38%|███▊ | 16793/43939 [27:05:32<43:14:57, 5.74s/it]
38%|███▊ | 16794/43939 [27:05:38<43:10:21, 5.73s/it]
38%|███▊ | 16795/43939 [27:05:44<43:05:34, 5.72s/it]
38%|███▊ | 16796/43939 [27:05:49<42:58:00, 5.70s/it]
38%|███▊ | 16797/43939 [27:05:55<42:52:30, 5.69s/it]
38%|███▊ | 16798/43939 [27:06:01<42:48:52, 5.68s/it]
38%|███▊ | 16799/43939 [27:06:06<42:49:44, 5.68s/it]
38%|███▊ | 16800/43939 [27:06:12<42:50:53, 5.68s/it]
{'loss': 0.9374, 'grad_norm': 1.2893600463867188, 'learning_rate': 3.8810098170769964e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.33, 'epoch': 1.53}
38%|███▊ | 16800/43939 [27:06:12<42:50:53, 5.68s/it]
38%|███▊ | 16801/43939 [27:06:18<42:53:21, 5.69s/it]
38%|███▊ | 16802/43939 [27:06:23<42:56:32, 5.70s/it]
38%|███▊ | 16803/43939 [27:06:29<42:53:54, 5.69s/it]
38%|███▊ | 16804/43939 [27:06:35<42:53:21, 5.69s/it]
38%|███▊ | 16805/43939 [27:06:41<42:56:14, 5.70s/it]
38%|███▊ | 16806/43939 [27:06:46<42:59:16, 5.70s/it]
38%|███▊ | 16807/43939 [27:06:52<43:08:31, 5.72s/it]
38%|███▊ | 16808/43939 [27:06:58<43:05:19, 5.72s/it]
38%|███▊ | 16809/43939 [27:07:04<43:31:47, 5.78s/it]
38%|███▊ | 16810/43939 [27:07:09<43:21:14, 5.75s/it]
38%|███▊ | 16811/43939 [27:07:15<43:15:46, 5.74s/it]
38%|███▊ | 16812/43939 [27:07:21<43:17:58, 5.75s/it]
38%|███▊ | 16813/43939 [27:07:27<43:16:15, 5.74s/it]
38%|███▊ | 16814/43939 [27:07:32<43:08:05, 5.72s/it]
38%|███▊ | 16815/43939 [27:07:38<43:01:34, 5.71s/it]
38%|███▊ | 16816/43939 [27:07:44<42:56:36, 5.70s/it]
38%|███▊ | 16817/43939 [27:07:49<42:53:36, 5.69s/it]
38%|███▊ | 16818/43939 [27:07:55<42:54:50, 5.70s/it]
38%|███▊ | 16819/43939 [27:08:01<43:03:30, 5.72s/it]
38%|███▊ | 16820/43939 [27:08:06<42:54:50, 5.70s/it]
38%|███▊ | 16821/43939 [27:08:12<42:48:24, 5.68s/it]
38%|███▊ | 16822/43939 [27:08:18<42:50:06, 5.69s/it]
38%|███▊ | 16823/43939 [27:08:23<42:48:13, 5.68s/it]
38%|███▊ | 16824/43939 [27:08:29<42:46:27, 5.68s/it]
38%|███▊ | 16825/43939 [27:08:35<42:48:20, 5.68s/it]
38%|███▊ | 16826/43939 [27:08:40<42:45:30, 5.68s/it]
38%|███▊ | 16827/43939 [27:08:46<42:50:20, 5.69s/it]
38%|███▊ | 16828/43939 [27:08:52<42:59:42, 5.71s/it]
38%|███▊ | 16829/43939 [27:08:58<42:56:38, 5.70s/it]
38%|███▊ | 16830/43939 [27:09:04<43:30:31, 5.78s/it]
38%|███▊ | 16831/43939 [27:09:09<43:24:00, 5.76s/it]
38%|███▊ | 16832/43939 [27:09:15<43:13:39, 5.74s/it]
38%|███▊ | 16833/43939 [27:09:21<43:09:22, 5.73s/it]
38%|███▊ | 16834/43939 [27:09:26<43:03:02, 5.72s/it]
38%|███▊ | 16835/43939 [27:09:32<42:57:57, 5.71s/it]
38%|███▊ | 16836/43939 [27:09:38<42:57:32, 5.71s/it]
38%|███▊ | 16837/43939 [27:09:43<43:01:44, 5.72s/it]
38%|███▊ | 16838/43939 [27:09:49<43:04:46, 5.72s/it]
38%|███▊ | 16839/43939 [27:09:55<43:02:59, 5.72s/it]
38%|███▊ | 16840/43939 [27:10:01<43:00:06, 5.71s/it]
38%|███▊ | 16841/43939 [27:10:06<42:59:51, 5.71s/it]
38%|███▊ | 16842/43939 [27:10:12<43:07:56, 5.73s/it]
38%|███▊ | 16843/43939 [27:10:18<42:58:09, 5.71s/it]
38%|███▊ | 16844/43939 [27:10:23<42:59:13, 5.71s/it]
38%|███▊ | 16845/43939 [27:10:29<43:02:11, 5.72s/it]
38%|███▊ | 16846/43939 [27:10:35<42:57:29, 5.71s/it]
38%|███▊ | 16847/43939 [27:10:41<42:53:26, 5.70s/it]
38%|███▊ | 16848/43939 [27:10:46<42:54:05, 5.70s/it]
38%|███▊ | 16849/43939 [27:10:52<42:50:51, 5.69s/it]
38%|███▊ | 16850/43939 [27:10:58<42:52:44, 5.70s/it]
{'loss': 0.9344, 'grad_norm': 1.2013905048370361, 'learning_rate': 3.8727213752344015e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.3, 'epoch': 1.53}
38%|███▊ | 16850/43939 [27:10:58<42:52:44, 5.70s/it]
38%|███▊ | 16851/43939 [27:11:03<42:53:52, 5.70s/it]
38%|███▊ | 16852/43939 [27:11:09<43:27:36, 5.78s/it]
38%|███▊ | 16853/43939 [27:11:15<43:14:54, 5.75s/it]
38%|███▊ | 16854/43939 [27:11:21<43:07:32, 5.73s/it]
38%|███▊ | 16855/43939 [27:11:26<42:59:51, 5.72s/it]
38%|███▊ | 16856/43939 [27:11:32<42:55:31, 5.71s/it]
38%|███▊ | 16857/43939 [27:11:38<43:01:39, 5.72s/it]
38%|███▊ | 16858/43939 [27:11:44<43:00:22, 5.72s/it]
38%|███▊ | 16859/43939 [27:11:49<42:56:01, 5.71s/it]
38%|███▊ | 16860/43939 [27:11:55<43:00:50, 5.72s/it]
38%|███▊ | 16861/43939 [27:12:01<42:57:28, 5.71s/it]
38%|███▊ | 16862/43939 [27:12:06<42:56:08, 5.71s/it]
38%|███▊ | 16863/43939 [27:12:12<42:59:11, 5.72s/it]
38%|███▊ | 16864/43939 [27:12:18<42:57:36, 5.71s/it]
38%|███▊ | 16865/43939 [27:12:23<42:49:40, 5.69s/it]
38%|███▊ | 16866/43939 [27:12:29<42:49:09, 5.69s/it]
38%|███▊ | 16867/43939 [27:12:35<42:54:26, 5.71s/it]
38%|███▊ | 16868/43939 [27:12:41<42:49:55, 5.70s/it]
38%|███▊ | 16869/43939 [27:12:46<43:00:12, 5.72s/it]
38%|███▊ | 16870/43939 [27:12:52<42:58:22, 5.72s/it]
38%|███▊ | 16871/43939 [27:12:58<42:58:12, 5.71s/it]
38%|███▊ | 16872/43939 [27:13:03<43:02:14, 5.72s/it]
38%|███▊ | 16873/43939 [27:13:09<43:25:52, 5.78s/it]
38%|███▊ | 16874/43939 [27:13:15<43:23:20, 5.77s/it]
38%|███▊ | 16875/43939 [27:13:21<43:12:50, 5.75s/it]
38%|███▊ | 16876/43939 [27:13:27<43:09:36, 5.74s/it]
38%|███▊ | 16877/43939 [27:13:32<43:06:22, 5.73s/it]
38%|███▊ | 16878/43939 [27:13:38<42:56:25, 5.71s/it]
38%|███▊ | 16879/43939 [27:13:44<42:51:36, 5.70s/it]
38%|███▊ | 16880/43939 [27:13:49<42:57:13, 5.71s/it]
38%|███▊ | 16881/43939 [27:13:55<42:55:52, 5.71s/it]
38%|███▊ | 16882/43939 [27:14:01<42:50:20, 5.70s/it]
38%|███▊ | 16883/43939 [27:14:06<42:51:27, 5.70s/it]
38%|███▊ | 16884/43939 [27:14:12<42:51:24, 5.70s/it]
38%|███▊ | 16885/43939 [27:14:18<42:48:19, 5.70s/it]
38%|███▊ | 16886/43939 [27:14:24<42:55:12, 5.71s/it]
38%|███▊ | 16887/43939 [27:14:29<43:00:52, 5.72s/it]
38%|███▊ | 16888/43939 [27:14:35<43:02:38, 5.73s/it]
38%|███▊ | 16889/43939 [27:14:41<42:55:04, 5.71s/it]
38%|███▊ | 16890/43939 [27:14:46<42:52:43, 5.71s/it]
38%|███▊ | 16891/43939 [27:14:52<42:55:12, 5.71s/it]
38%|███▊ | 16892/43939 [27:14:58<42:58:05, 5.72s/it]
38%|███▊ | 16893/43939 [27:15:04<43:01:04, 5.73s/it]
38%|███▊ | 16894/43939 [27:15:10<43:26:51, 5.78s/it]
38%|███▊ | 16895/43939 [27:15:15<43:19:20, 5.77s/it]
38%|███▊ | 16896/43939 [27:15:21<43:11:55, 5.75s/it]
38%|███▊ | 16897/43939 [27:15:27<43:08:18, 5.74s/it]
38%|███▊ | 16898/43939 [27:15:32<43:05:58, 5.74s/it]
38%|███▊ | 16899/43939 [27:15:38<43:10:30, 5.75s/it]
38%|███▊ | 16900/43939 [27:15:44<43:05:59, 5.74s/it]
{'loss': 0.9253, 'grad_norm': 1.1550096273422241, 'learning_rate': 3.8644112754862614e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.09, 'epoch': 1.54}
38%|███▊ | 16900/43939 [27:15:44<43:05:59, 5.74s/it]
38%|███▊ | 16901/43939 [27:15:50<43:02:29, 5.73s/it]
38%|███▊ | 16902/43939 [27:15:55<42:57:26, 5.72s/it]
38%|███▊ | 16903/43939 [27:16:01<42:55:49, 5.72s/it]
38%|███▊ | 16904/43939 [27:16:07<43:02:30, 5.73s/it]
38%|███▊ | 16905/43939 [27:16:12<42:54:36, 5.71s/it]
38%|███▊ | 16906/43939 [27:16:18<42:50:23, 5.71s/it]
38%|███▊ | 16907/43939 [27:16:24<42:50:31, 5.71s/it]
38%|███▊ | 16908/43939 [27:16:30<42:49:06, 5.70s/it]
38%|███▊ | 16909/43939 [27:16:35<42:51:32, 5.71s/it]
38%|███▊ | 16910/43939 [27:16:41<42:47:05, 5.70s/it]
38%|███▊ | 16911/43939 [27:16:47<42:47:28, 5.70s/it]
38%|███▊ | 16912/43939 [27:16:52<42:50:14, 5.71s/it]
38%|███▊ | 16913/43939 [27:16:58<42:51:48, 5.71s/it]
38%|███▊ | 16914/43939 [27:17:04<42:55:05, 5.72s/it]
38%|███▊ | 16915/43939 [27:17:10<43:23:08, 5.78s/it]
38%|███▊ | 16916/43939 [27:17:16<43:16:49, 5.77s/it]
39%|███▊ | 16917/43939 [27:17:21<43:09:28, 5.75s/it]
39%|███▊ | 16918/43939 [27:17:27<43:02:06, 5.73s/it]
39%|███▊ | 16919/43939 [27:17:33<42:53:18, 5.71s/it]
39%|███▊ | 16920/43939 [27:17:38<42:53:02, 5.71s/it]
39%|███▊ | 16921/43939 [27:17:44<42:48:59, 5.71s/it]
39%|███▊ | 16922/43939 [27:17:50<42:42:17, 5.69s/it]
39%|███▊ | 16923/43939 [27:17:55<42:44:53, 5.70s/it]
39%|███▊ | 16924/43939 [27:18:01<42:47:23, 5.70s/it]
39%|███▊ | 16925/43939 [27:18:07<42:46:26, 5.70s/it]
39%|███▊ | 16926/43939 [27:18:12<42:45:13, 5.70s/it]
39%|███▊ | 16927/43939 [27:18:18<42:49:01, 5.71s/it]
39%|███▊ | 16928/43939 [27:18:24<42:48:18, 5.71s/it]
39%|███▊ | 16929/43939 [27:18:30<42:44:57, 5.70s/it]
39%|███▊ | 16930/43939 [27:18:35<42:42:38, 5.69s/it]
39%|███▊ | 16931/43939 [27:18:41<42:45:21, 5.70s/it]
39%|███▊ | 16932/43939 [27:18:47<42:43:28, 5.70s/it]
39%|███▊ | 16933/43939 [27:18:52<42:45:32, 5.70s/it]
39%|███▊ | 16934/43939 [27:18:58<42:40:24, 5.69s/it]
39%|███▊ | 16935/43939 [27:19:04<42:43:48, 5.70s/it]
39%|███▊ | 16936/43939 [27:19:10<43:25:14, 5.79s/it]
39%|███▊ | 16937/43939 [27:19:15<43:16:54, 5.77s/it]
39%|███▊ | 16938/43939 [27:19:21<43:15:18, 5.77s/it]
39%|███▊ | 16939/43939 [27:19:27<43:06:06, 5.75s/it]
39%|███▊ | 16940/43939 [27:19:33<43:09:30, 5.75s/it]
39%|███▊ | 16941/43939 [27:19:38<43:04:18, 5.74s/it]
39%|███▊ | 16942/43939 [27:19:44<42:56:00, 5.73s/it]
39%|███▊ | 16943/43939 [27:19:50<42:47:41, 5.71s/it]
39%|███▊ | 16944/43939 [27:19:55<42:49:10, 5.71s/it]
39%|███▊ | 16945/43939 [27:20:01<42:48:36, 5.71s/it]
39%|███▊ | 16946/43939 [27:20:07<42:48:13, 5.71s/it]
39%|███▊ | 16947/43939 [27:20:13<42:48:42, 5.71s/it]
39%|███▊ | 16948/43939 [27:20:18<42:46:57, 5.71s/it]
39%|███▊ | 16949/43939 [27:20:24<42:51:17, 5.72s/it]
39%|███▊ | 16950/43939 [27:20:30<42:51:06, 5.72s/it]
{'loss': 0.9166, 'grad_norm': 1.0183569192886353, 'learning_rate': 3.856079648943929e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.68, 'epoch': 1.54}
39%|███▊ | 16950/43939 [27:20:30<42:51:06, 5.72s/it]
39%|███▊ | 16951/43939 [27:20:35<42:48:20, 5.71s/it]
39%|███▊ | 16952/43939 [27:20:41<42:48:38, 5.71s/it]
39%|███▊ | 16953/43939 [27:20:47<42:44:21, 5.70s/it]
39%|███▊ | 16954/43939 [27:20:53<42:43:52, 5.70s/it]
39%|███▊ | 16955/43939 [27:20:58<42:40:07, 5.69s/it]
39%|███▊ | 16956/43939 [27:21:04<42:41:55, 5.70s/it]
39%|███▊ | 16957/43939 [27:21:10<43:18:38, 5.78s/it]
39%|███▊ | 16958/43939 [27:21:16<43:06:25, 5.75s/it]
39%|███▊ | 16959/43939 [27:21:21<43:01:28, 5.74s/it]
39%|███▊ | 16960/43939 [27:21:27<42:47:35, 5.71s/it]
39%|███▊ | 16961/43939 [27:21:33<42:45:53, 5.71s/it]
39%|███▊ | 16962/43939 [27:21:38<42:42:48, 5.70s/it]
39%|███▊ | 16963/43939 [27:21:44<42:44:10, 5.70s/it]
39%|███▊ | 16964/43939 [27:21:50<42:48:16, 5.71s/it]
39%|███▊ | 16965/43939 [27:21:55<42:46:43, 5.71s/it]
39%|███▊ | 16966/43939 [27:22:01<42:46:35, 5.71s/it]
39%|███▊ | 16967/43939 [27:22:07<42:49:02, 5.71s/it]
39%|███▊ | 16968/43939 [27:22:13<42:46:39, 5.71s/it]
39%|███▊ | 16969/43939 [27:22:18<42:47:41, 5.71s/it]
39%|███▊ | 16970/43939 [27:22:24<42:44:20, 5.71s/it]
39%|███▊ | 16971/43939 [27:22:30<42:39:42, 5.69s/it]
39%|███▊ | 16972/43939 [27:22:35<42:39:35, 5.69s/it]
39%|███▊ | 16973/43939 [27:22:41<42:40:11, 5.70s/it]
39%|███▊ | 16974/43939 [27:22:47<42:47:02, 5.71s/it]
39%|███▊ | 16975/43939 [27:22:53<42:45:16, 5.71s/it]
39%|███▊ | 16976/43939 [27:22:58<42:46:05, 5.71s/it]
39%|███▊ | 16977/43939 [27:23:04<42:46:14, 5.71s/it]
39%|███▊ | 16978/43939 [27:23:10<43:22:23, 5.79s/it]
39%|███▊ | 16979/43939 [27:23:16<43:19:04, 5.78s/it]
39%|███▊ | 16980/43939 [27:23:21<43:09:12, 5.76s/it]
39%|███▊ | 16981/43939 [27:23:27<42:58:04, 5.74s/it]
39%|███▊ | 16982/43939 [27:23:33<42:52:48, 5.73s/it]
39%|███▊ | 16983/43939 [27:23:39<42:54:58, 5.73s/it]
39%|███▊ | 16984/43939 [27:23:44<42:46:26, 5.71s/it]
39%|███▊ | 16985/43939 [27:23:50<42:43:43, 5.71s/it]
39%|███▊ | 16986/43939 [27:23:56<42:45:35, 5.71s/it]
39%|███▊ | 16987/43939 [27:24:01<42:39:55, 5.70s/it]
39%|███▊ | 16988/43939 [27:24:07<42:39:03, 5.70s/it]
39%|███▊ | 16989/43939 [27:24:13<42:45:54, 5.71s/it]
39%|███▊ | 16990/43939 [27:24:18<42:42:43, 5.71s/it]
39%|███▊ | 16991/43939 [27:24:24<42:43:50, 5.71s/it]
39%|███▊ | 16992/43939 [27:24:30<42:42:10, 5.70s/it]
39%|███▊ | 16993/43939 [27:24:36<42:38:37, 5.70s/it]
39%|███▊ | 16994/43939 [27:24:41<42:36:16, 5.69s/it]
39%|███▊ | 16995/43939 [27:24:47<42:34:50, 5.69s/it]
39%|███▊ | 16996/43939 [27:24:53<42:32:37, 5.68s/it]
39%|███▊ | 16997/43939 [27:24:58<42:34:34, 5.69s/it]
39%|███▊ | 16998/43939 [27:25:04<42:32:21, 5.68s/it]
39%|███▊ | 16999/43939 [27:25:10<42:33:30, 5.69s/it]
39%|███▊ | 17000/43939 [27:25:16<43:12:32, 5.77s/it]
{'loss': 0.942, 'grad_norm': 1.1174321174621582, 'learning_rate': 3.84772662705839e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.8, 'epoch': 1.55}
39%|███▊ | 17000/43939 [27:25:16<43:12:32, 5.77s/it]
39%|███▊ | 17001/43939 [27:25:21<43:06:44, 5.76s/it]
39%|███▊ | 17002/43939 [27:25:27<42:55:51, 5.74s/it]
39%|███▊ | 17003/43939 [27:25:33<42:45:17, 5.71s/it]
39%|███▊ | 17004/43939 [27:25:38<42:40:49, 5.70s/it]
39%|███▊ | 17005/43939 [27:25:44<42:36:41, 5.70s/it]
39%|███▊ | 17006/43939 [27:25:50<42:38:19, 5.70s/it]
39%|███▊ | 17007/43939 [27:25:55<42:34:11, 5.69s/it]
39%|███▊ | 17008/43939 [27:26:01<42:32:05, 5.69s/it]
39%|███▊ | 17009/43939 [27:26:07<42:37:08, 5.70s/it]
39%|███▊ | 17010/43939 [27:26:12<42:36:08, 5.70s/it]
39%|███▊ | 17011/43939 [27:26:18<42:33:15, 5.69s/it]
39%|███▊ | 17012/43939 [27:26:24<42:33:19, 5.69s/it]
39%|███▊ | 17013/43939 [27:26:30<42:39:04, 5.70s/it]
39%|███▊ | 17014/43939 [27:26:35<42:37:58, 5.70s/it]
39%|███▊ | 17015/43939 [27:26:41<42:38:15, 5.70s/it]
39%|███▊ | 17016/43939 [27:26:47<42:39:51, 5.70s/it]
39%|███▊ | 17017/43939 [27:26:52<42:39:48, 5.70s/it]
39%|███▊ | 17018/43939 [27:26:58<42:39:39, 5.70s/it]
39%|███▊ | 17019/43939 [27:27:04<42:42:10, 5.71s/it]
39%|███▊ | 17020/43939 [27:27:10<42:37:07, 5.70s/it]
39%|███▊ | 17021/43939 [27:27:15<43:04:20, 5.76s/it]
39%|███▊ | 17022/43939 [27:27:21<42:49:18, 5.73s/it]
39%|███▊ | 17023/43939 [27:27:27<42:42:58, 5.71s/it]
39%|███▊ | 17024/43939 [27:27:32<42:40:15, 5.71s/it]
39%|███▊ | 17025/43939 [27:27:38<42:33:37, 5.69s/it]
39%|███▊ | 17026/43939 [27:27:44<42:30:03, 5.69s/it]
39%|███▉ | 17027/43939 [27:27:49<42:31:29, 5.69s/it]
39%|███▉ | 17028/43939 [27:27:55<42:34:03, 5.69s/it]
39%|███▉ | 17029/43939 [27:28:01<42:35:52, 5.70s/it]
39%|███▉ | 17030/43939 [27:28:07<42:38:13, 5.70s/it]
39%|███▉ | 17031/43939 [27:28:12<42:37:38, 5.70s/it]
39%|███▉ | 17032/43939 [27:28:18<42:35:05, 5.70s/it]
39%|███▉ | 17033/43939 [27:28:24<42:42:17, 5.71s/it]
39%|███▉ | 17034/43939 [27:28:29<42:48:48, 5.73s/it]
39%|███▉ | 17035/43939 [27:28:35<42:43:35, 5.72s/it]
39%|███▉ | 17036/43939 [27:28:41<42:40:22, 5.71s/it]
39%|███▉ | 17037/43939 [27:28:47<42:37:28, 5.70s/it]
39%|███▉ | 17038/43939 [27:28:52<42:43:34, 5.72s/it]
39%|███▉ | 17039/43939 [27:28:58<42:43:37, 5.72s/it]
39%|███▉ | 17040/43939 [27:29:04<42:42:25, 5.72s/it]
39%|███▉ | 17041/43939 [27:29:09<42:40:35, 5.71s/it]
39%|███▉ | 17042/43939 [27:29:15<43:10:10, 5.78s/it]
39%|███▉ | 17043/43939 [27:29:21<42:58:22, 5.75s/it]
39%|███▉ | 17044/43939 [27:29:27<42:56:59, 5.75s/it]
39%|███▉ | 17045/43939 [27:29:33<42:57:16, 5.75s/it]
39%|███▉ | 17046/43939 [27:29:38<42:52:17, 5.74s/it]
39%|███▉ | 17047/43939 [27:29:44<42:47:31, 5.73s/it]
39%|███▉ | 17048/43939 [27:29:50<42:54:15, 5.74s/it]
39%|███▉ | 17049/43939 [27:29:56<42:59:53, 5.76s/it]
39%|███▉ | 17050/43939 [27:30:01<42:52:04, 5.74s/it]
{'loss': 0.9161, 'grad_norm': 1.1995723247528076, 'learning_rate': 3.8393523416181955e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.92, 'epoch': 1.55}
39%|███▉ | 17050/43939 [27:30:01<42:52:04, 5.74s/it]
39%|███▉ | 17051/43939 [27:30:07<42:45:44, 5.73s/it]
39%|███▉ | 17052/43939 [27:30:13<42:39:27, 5.71s/it]
39%|███▉ | 17053/43939 [27:30:18<42:43:48, 5.72s/it]
39%|███▉ | 17054/43939 [27:30:24<42:40:54, 5.72s/it]
39%|███▉ | 17055/43939 [27:30:30<42:36:58, 5.71s/it]
39%|███▉ | 17056/43939 [27:30:35<42:36:03, 5.70s/it]
39%|███▉ | 17057/43939 [27:30:41<42:46:09, 5.73s/it]
39%|███▉ | 17058/43939 [27:30:47<42:40:23, 5.71s/it]
39%|███▉ | 17059/43939 [27:30:53<42:38:56, 5.71s/it]
39%|███▉ | 17060/43939 [27:30:58<42:39:45, 5.71s/it]
39%|███▉ | 17061/43939 [27:31:04<42:32:29, 5.70s/it]
39%|███▉ | 17062/43939 [27:31:10<42:39:15, 5.71s/it]
39%|███▉ | 17063/43939 [27:31:16<43:13:42, 5.79s/it]
39%|███▉ | 17064/43939 [27:31:21<43:03:09, 5.77s/it]
39%|███▉ | 17065/43939 [27:31:27<42:53:55, 5.75s/it]
39%|███▉ | 17066/43939 [27:31:33<42:50:19, 5.74s/it]
39%|███▉ | 17067/43939 [27:31:39<42:44:42, 5.73s/it]
39%|███▉ | 17068/43939 [27:31:44<42:47:29, 5.73s/it]
39%|███▉ | 17069/43939 [27:31:50<42:47:42, 5.73s/it]
39%|███▉ | 17070/43939 [27:31:56<42:46:43, 5.73s/it]
39%|███▉ | 17071/43939 [27:32:02<42:49:33, 5.74s/it]
39%|███▉ | 17072/43939 [27:32:07<42:46:05, 5.73s/it]
39%|███▉ | 17073/43939 [27:32:13<42:42:23, 5.72s/it]
39%|███▉ | 17074/43939 [27:32:19<42:42:23, 5.72s/it]
39%|███▉ | 17075/43939 [27:32:24<42:44:59, 5.73s/it]
39%|███▉ | 17076/43939 [27:32:30<42:48:43, 5.74s/it]
39%|███▉ | 17077/43939 [27:32:36<42:53:06, 5.75s/it]
39%|███▉ | 17078/43939 [27:32:42<42:45:30, 5.73s/it]
39%|███▉ | 17079/43939 [27:32:47<42:40:36, 5.72s/it]
39%|███▉ | 17080/43939 [27:32:53<42:37:58, 5.71s/it]
39%|███▉ | 17081/43939 [27:32:59<42:32:12, 5.70s/it]
39%|███▉ | 17082/43939 [27:33:04<42:33:30, 5.70s/it]
39%|███▉ | 17083/43939 [27:33:10<42:33:51, 5.71s/it]
39%|███▉ | 17084/43939 [27:33:16<42:42:41, 5.73s/it]
39%|███▉ | 17085/43939 [27:33:22<43:17:05, 5.80s/it]
39%|███▉ | 17086/43939 [27:33:28<43:02:02, 5.77s/it]
39%|███▉ | 17087/43939 [27:33:33<42:50:50, 5.74s/it]
39%|███▉ | 17088/43939 [27:33:39<42:43:31, 5.73s/it]
39%|███▉ | 17089/43939 [27:33:45<42:39:28, 5.72s/it]
39%|███▉ | 17090/43939 [27:33:50<42:36:49, 5.71s/it]
39%|███▉ | 17091/43939 [27:33:56<42:32:17, 5.70s/it]
39%|███▉ | 17092/43939 [27:34:02<42:26:46, 5.69s/it]
39%|███▉ | 17093/43939 [27:34:07<42:27:55, 5.69s/it]
39%|███▉ | 17094/43939 [27:34:13<42:34:46, 5.71s/it]
39%|███▉ | 17095/43939 [27:34:19<42:34:24, 5.71s/it]
39%|███▉ | 17096/43939 [27:34:25<42:32:46, 5.71s/it]
39%|███▉ | 17097/43939 [27:34:30<42:41:14, 5.73s/it]
39%|███▉ | 17098/43939 [27:34:36<42:41:49, 5.73s/it]
39%|███▉ | 17099/43939 [27:34:42<42:39:19, 5.72s/it]
39%|███▉ | 17100/43939 [27:34:47<42:38:35, 5.72s/it]
{'loss': 0.9074, 'grad_norm': 1.2188096046447754, 'learning_rate': 3.8309569247473766e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.16, 'epoch': 1.56}
39%|███▉ | 17100/43939 [27:34:47<42:38:35, 5.72s/it]
39%|███▉ | 17101/43939 [27:34:53<42:42:42, 5.73s/it]
39%|███▉ | 17102/43939 [27:34:59<42:32:27, 5.71s/it]
39%|███▉ | 17103/43939 [27:35:05<42:35:50, 5.71s/it]
39%|███▉ | 17104/43939 [27:35:10<42:36:27, 5.72s/it]
39%|███▉ | 17105/43939 [27:35:16<42:33:04, 5.71s/it]
39%|███▉ | 17106/43939 [27:35:22<43:06:47, 5.78s/it]
39%|███▉ | 17107/43939 [27:35:28<42:56:23, 5.76s/it]
39%|███▉ | 17108/43939 [27:35:33<42:51:28, 5.75s/it]
39%|███▉ | 17109/43939 [27:35:39<42:41:44, 5.73s/it]
39%|███▉ | 17110/43939 [27:35:45<42:42:30, 5.73s/it]
39%|███▉ | 17111/43939 [27:35:51<42:47:45, 5.74s/it]
39%|███▉ | 17112/43939 [27:35:56<42:42:47, 5.73s/it]
39%|███▉ | 17113/43939 [27:36:02<42:34:07, 5.71s/it]
39%|███▉ | 17114/43939 [27:36:08<42:29:44, 5.70s/it]
39%|███▉ | 17115/43939 [27:36:13<42:37:47, 5.72s/it]
39%|███▉ | 17116/43939 [27:36:19<42:33:47, 5.71s/it]
39%|███▉ | 17117/43939 [27:36:25<42:34:29, 5.71s/it]
39%|███▉ | 17118/43939 [27:36:31<42:32:44, 5.71s/it]
39%|███▉ | 17119/43939 [27:36:36<42:25:36, 5.69s/it]
39%|███▉ | 17120/43939 [27:36:42<42:31:07, 5.71s/it]
39%|███▉ | 17121/43939 [27:36:48<42:27:04, 5.70s/it]
39%|███▉ | 17122/43939 [27:36:53<42:24:06, 5.69s/it]
39%|███▉ | 17123/43939 [27:36:59<42:20:55, 5.69s/it]
39%|███▉ | 17124/43939 [27:37:05<42:22:46, 5.69s/it]
39%|███▉ | 17125/43939 [27:37:10<42:30:05, 5.71s/it]
39%|███▉ | 17126/43939 [27:37:16<42:29:15, 5.70s/it]
39%|███▉ | 17127/43939 [27:37:22<42:59:17, 5.77s/it]
39%|███▉ | 17128/43939 [27:37:28<42:54:16, 5.76s/it]
39%|███▉ | 17129/43939 [27:37:33<42:49:53, 5.75s/it]
39%|███▉ | 17130/43939 [27:37:39<42:49:36, 5.75s/it]
39%|███▉ | 17131/43939 [27:37:45<42:38:43, 5.73s/it]
39%|███▉ | 17132/43939 [27:37:51<42:35:21, 5.72s/it]
39%|███▉ | 17133/43939 [27:37:56<42:32:32, 5.71s/it]
39%|███▉ | 17134/43939 [27:38:02<42:28:25, 5.70s/it]
39%|███▉ | 17135/43939 [27:38:08<42:25:08, 5.70s/it]
39%|███▉ | 17136/43939 [27:38:13<42:24:32, 5.70s/it]
39%|███▉ | 17137/43939 [27:38:19<42:29:32, 5.71s/it]
39%|███▉ | 17138/43939 [27:38:25<42:35:53, 5.72s/it]
39%|███▉ | 17139/43939 [27:38:31<42:32:15, 5.71s/it]
39%|███▉ | 17140/43939 [27:38:36<42:32:21, 5.71s/it]
39%|███▉ | 17141/43939 [27:38:42<42:31:19, 5.71s/it]
39%|███▉ | 17142/43939 [27:38:48<42:27:59, 5.71s/it]
39%|███▉ | 17143/43939 [27:38:53<42:30:12, 5.71s/it]
39%|███▉ | 17144/43939 [27:38:59<42:26:23, 5.70s/it]
39%|███▉ | 17145/43939 [27:39:05<42:27:59, 5.71s/it]
39%|███▉ | 17146/43939 [27:39:11<42:35:28, 5.72s/it]
39%|███▉ | 17147/43939 [27:39:16<42:29:28, 5.71s/it]
39%|███▉ | 17148/43939 [27:39:22<43:00:12, 5.78s/it]
39%|███▉ | 17149/43939 [27:39:28<42:51:16, 5.76s/it]
39%|███▉ | 17150/43939 [27:39:34<42:47:25, 5.75s/it]
{'loss': 0.9312, 'grad_norm': 1.2756145000457764, 'learning_rate': 3.822540508903364e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.17, 'epoch': 1.56}
39%|███▉ | 17150/43939 [27:39:34<42:47:25, 5.75s/it]
39%|███▉ | 17151/43939 [27:39:39<42:39:12, 5.73s/it]
39%|███▉ | 17152/43939 [27:39:45<42:36:26, 5.73s/it]
39%|███▉ | 17153/43939 [27:39:51<42:36:39, 5.73s/it]
39%|███▉ | 17154/43939 [27:39:56<42:31:52, 5.72s/it]
39%|███▉ | 17155/43939 [27:40:02<42:38:32, 5.73s/it]
39%|███▉ | 17156/43939 [27:40:08<42:30:51, 5.71s/it]
39%|███▉ | 17157/43939 [27:40:14<42:32:16, 5.72s/it]
39%|███▉ | 17158/43939 [27:40:19<42:27:48, 5.71s/it]
39%|███▉ | 17159/43939 [27:40:25<42:23:39, 5.70s/it]
39%|███▉ | 17160/43939 [27:40:31<42:27:25, 5.71s/it]
39%|███▉ | 17161/43939 [27:40:36<42:23:43, 5.70s/it]
39%|███▉ | 17162/43939 [27:40:42<42:24:38, 5.70s/it]
39%|███▉ | 17163/43939 [27:40:48<42:26:11, 5.71s/it]
39%|███▉ | 17164/43939 [27:40:54<42:30:43, 5.72s/it]
39%|███▉ | 17165/43939 [27:40:59<42:28:38, 5.71s/it]
39%|███▉ | 17166/43939 [27:41:05<42:29:59, 5.71s/it]
39%|███▉ | 17167/43939 [27:41:11<42:29:24, 5.71s/it]
39%|███▉ | 17168/43939 [27:41:16<42:20:12, 5.69s/it]
39%|███▉ | 17169/43939 [27:41:22<42:17:09, 5.69s/it]
39%|███▉ | 17170/43939 [27:41:28<42:45:40, 5.75s/it]
39%|███▉ | 17171/43939 [27:41:34<42:42:02, 5.74s/it]
39%|███▉ | 17172/43939 [27:41:39<42:33:57, 5.72s/it]
39%|███▉ | 17173/43939 [27:41:45<42:31:24, 5.72s/it]
39%|███▉ | 17174/43939 [27:41:51<42:35:07, 5.73s/it]
39%|███▉ | 17175/43939 [27:41:56<42:26:52, 5.71s/it]
39%|███▉ | 17176/43939 [27:42:02<42:27:42, 5.71s/it]
39%|███▉ | 17177/43939 [27:42:08<42:25:50, 5.71s/it]
39%|███▉ | 17178/43939 [27:42:13<42:20:26, 5.70s/it]
39%|███▉ | 17179/43939 [27:42:19<42:26:18, 5.71s/it]
39%|███▉ | 17180/43939 [27:42:25<42:27:23, 5.71s/it]
39%|███▉ | 17181/43939 [27:42:31<42:24:25, 5.71s/it]
39%|███▉ | 17182/43939 [27:42:36<42:18:21, 5.69s/it]
39%|███▉ | 17183/43939 [27:42:42<42:22:27, 5.70s/it]
39%|███▉ | 17184/43939 [27:42:48<42:23:01, 5.70s/it]
39%|███▉ | 17185/43939 [27:42:53<42:19:02, 5.69s/it]
39%|███▉ | 17186/43939 [27:42:59<42:17:19, 5.69s/it]
39%|███▉ | 17187/43939 [27:43:05<42:13:35, 5.68s/it]
39%|███▉ | 17188/43939 [27:43:10<42:15:39, 5.69s/it]
39%|███▉ | 17189/43939 [27:43:16<42:17:34, 5.69s/it]
39%|███▉ | 17190/43939 [27:43:22<42:24:34, 5.71s/it]
39%|███▉ | 17191/43939 [27:43:28<42:50:16, 5.77s/it]
39%|███▉ | 17192/43939 [27:43:33<42:42:28, 5.75s/it]
39%|███▉ | 17193/43939 [27:43:39<42:40:50, 5.74s/it]
39%|███▉ | 17194/43939 [27:43:45<42:38:18, 5.74s/it]
39%|███▉ | 17195/43939 [27:43:51<42:36:05, 5.73s/it]
39%|███▉ | 17196/43939 [27:43:56<42:26:59, 5.71s/it]
39%|███▉ | 17197/43939 [27:44:02<42:28:59, 5.72s/it]
39%|███▉ | 17198/43939 [27:44:08<42:24:36, 5.71s/it]
39%|███▉ | 17199/43939 [27:44:13<42:24:05, 5.71s/it]
39%|███▉ | 17200/43939 [27:44:19<42:27:41, 5.72s/it]
{'loss': 0.937, 'grad_norm': 1.1586005687713623, 'learning_rate': 3.814103226874896e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.34, 'epoch': 1.57}
39%|███▉ | 17200/43939 [27:44:19<42:27:41, 5.72s/it]
39%|███▉ | 17201/43939 [27:44:25<42:28:58, 5.72s/it]
39%|███▉ | 17202/43939 [27:44:31<42:30:05, 5.72s/it]
39%|███▉ | 17203/43939 [27:44:36<42:23:52, 5.71s/it]
39%|███▉ | 17204/43939 [27:44:42<42:19:33, 5.70s/it]
39%|███▉ | 17205/43939 [27:44:48<42:21:14, 5.70s/it]
39%|███▉ | 17206/43939 [27:44:53<42:22:12, 5.71s/it]
39%|███▉ | 17207/43939 [27:44:59<42:23:52, 5.71s/it]
39%|███▉ | 17208/43939 [27:45:05<42:20:49, 5.70s/it]
39%|███▉ | 17209/43939 [27:45:11<42:20:13, 5.70s/it]
39%|███▉ | 17210/43939 [27:45:16<42:17:06, 5.70s/it]
39%|███▉ | 17211/43939 [27:45:22<42:20:23, 5.70s/it]
39%|███▉ | 17212/43939 [27:45:28<42:46:35, 5.76s/it]
39%|███▉ | 17213/43939 [27:45:34<42:43:42, 5.76s/it]
39%|███▉ | 17214/43939 [27:45:39<42:39:26, 5.75s/it]
39%|███▉ | 17215/43939 [27:45:45<42:32:18, 5.73s/it]
39%|███▉ | 17216/43939 [27:45:51<42:42:09, 5.75s/it]
39%|███▉ | 17217/43939 [27:45:56<42:33:28, 5.73s/it]
39%|███▉ | 17218/43939 [27:46:02<42:30:49, 5.73s/it]
39%|███▉ | 17219/43939 [27:46:08<42:25:06, 5.72s/it]
39%|███▉ | 17220/43939 [27:46:14<42:28:07, 5.72s/it]
39%|███▉ | 17221/43939 [27:46:19<42:24:26, 5.71s/it]
39%|███▉ | 17222/43939 [27:46:25<42:20:32, 5.71s/it]
39%|███▉ | 17223/43939 [27:46:31<42:21:54, 5.71s/it]
39%|███▉ | 17224/43939 [27:46:36<42:19:43, 5.70s/it]
39%|███▉ | 17225/43939 [27:46:42<42:19:01, 5.70s/it]
39%|███▉ | 17226/43939 [27:46:48<42:15:16, 5.69s/it]
39%|███▉ | 17227/43939 [27:46:53<42:14:47, 5.69s/it]
39%|███▉ | 17228/43939 [27:46:59<42:18:58, 5.70s/it]
39%|███▉ | 17229/43939 [27:47:05<42:23:11, 5.71s/it]
39%|███▉ | 17230/43939 [27:47:11<42:16:42, 5.70s/it]
39%|███▉ | 17231/43939 [27:47:16<42:16:09, 5.70s/it]
39%|███▉ | 17232/43939 [27:47:22<42:19:55, 5.71s/it]
39%|███▉ | 17233/43939 [27:47:28<42:56:39, 5.79s/it]
39%|███▉ | 17234/43939 [27:47:34<42:45:22, 5.76s/it]
39%|███▉ | 17235/43939 [27:47:39<42:40:46, 5.75s/it]
39%|███▉ | 17236/43939 [27:47:45<42:31:48, 5.73s/it]
39%|███▉ | 17237/43939 [27:47:51<42:30:49, 5.73s/it]
39%|███▉ | 17238/43939 [27:47:57<42:27:34, 5.72s/it]
39%|███▉ | 17239/43939 [27:48:02<42:24:44, 5.72s/it]
39%|███▉ | 17240/43939 [27:48:08<42:23:26, 5.72s/it]
39%|███▉ | 17241/43939 [27:48:14<42:21:51, 5.71s/it]
39%|███▉ | 17242/43939 [27:48:19<42:23:42, 5.72s/it]
39%|███▉ | 17243/43939 [27:48:25<42:20:02, 5.71s/it]
39%|███▉ | 17244/43939 [27:48:31<42:24:53, 5.72s/it]
39%|███▉ | 17245/43939 [27:48:37<42:27:20, 5.73s/it]
39%|███▉ | 17246/43939 [27:48:42<42:23:50, 5.72s/it]
39%|███▉ | 17247/43939 [27:48:48<42:25:08, 5.72s/it]
39%|███▉ | 17248/43939 [27:48:54<42:21:51, 5.71s/it]
39%|███▉ | 17249/43939 [27:48:59<42:25:04, 5.72s/it]
39%|███▉ | 17250/43939 [27:49:05<42:21:19, 5.71s/it]
{'loss': 0.923, 'grad_norm': 1.2390333414077759, 'learning_rate': 3.805645211779924e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.31, 'epoch': 1.57}
39%|███▉ | 17250/43939 [27:49:05<42:21:19, 5.71s/it]
39%|███▉ | 17251/43939 [27:49:11<42:25:59, 5.72s/it]
39%|███▉ | 17252/43939 [27:49:17<42:30:49, 5.73s/it]
39%|███▉ | 17253/43939 [27:49:22<42:25:53, 5.72s/it]
39%|███▉ | 17254/43939 [27:49:28<42:49:02, 5.78s/it]
39%|███▉ | 17255/43939 [27:49:34<42:38:37, 5.75s/it]
39%|███▉ | 17256/43939 [27:49:40<42:29:26, 5.73s/it]
39%|███▉ | 17257/43939 [27:49:45<42:25:07, 5.72s/it]
39%|███▉ | 17258/43939 [27:49:51<42:22:15, 5.72s/it]
39%|███▉ | 17259/43939 [27:49:57<42:22:39, 5.72s/it]
39%|███▉ | 17260/43939 [27:50:02<42:21:47, 5.72s/it]
39%|███▉ | 17261/43939 [27:50:08<42:27:27, 5.73s/it]
39%|███▉ | 17262/43939 [27:50:14<42:25:09, 5.72s/it]
39%|███▉ | 17263/43939 [27:50:20<42:22:54, 5.72s/it]
39%|███▉ | 17264/43939 [27:50:25<42:18:45, 5.71s/it]
39%|███▉ | 17265/43939 [27:50:31<42:20:36, 5.71s/it]
39%|███▉ | 17266/43939 [27:50:37<42:14:37, 5.70s/it]
39%|███▉ | 17267/43939 [27:50:42<42:18:33, 5.71s/it]
39%|███▉ | 17268/43939 [27:50:48<42:13:06, 5.70s/it]
39%|███▉ | 17269/43939 [27:50:54<42:16:02, 5.71s/it]
39%|███▉ | 17270/43939 [27:51:00<42:15:00, 5.70s/it]
39%|███▉ | 17271/43939 [27:51:05<42:14:34, 5.70s/it]
39%|███▉ | 17272/43939 [27:51:11<42:17:58, 5.71s/it]
39%|███▉ | 17273/43939 [27:51:17<42:15:22, 5.70s/it]
39%|███▉ | 17274/43939 [27:51:23<42:42:54, 5.77s/it]
39%|███▉ | 17275/43939 [27:51:28<42:34:02, 5.75s/it]
39%|███▉ | 17276/43939 [27:51:34<42:36:37, 5.75s/it]
39%|███▉ | 17277/43939 [27:51:40<42:31:18, 5.74s/it]
39%|███▉ | 17278/43939 [27:51:45<42:26:05, 5.73s/it]
39%|███▉ | 17279/43939 [27:51:51<42:17:29, 5.71s/it]
39%|███▉ | 17280/43939 [27:51:57<42:16:20, 5.71s/it]
39%|███▉ | 17281/43939 [27:52:03<42:09:08, 5.69s/it]
39%|███▉ | 17282/43939 [27:52:08<42:11:45, 5.70s/it]
39%|███▉ | 17283/43939 [27:52:14<42:07:22, 5.69s/it]
39%|███▉ | 17284/43939 [27:52:20<42:06:52, 5.69s/it]
39%|███▉ | 17285/43939 [27:52:25<42:12:19, 5.70s/it]
39%|███▉ | 17286/43939 [27:52:31<42:08:19, 5.69s/it]
39%|███▉ | 17287/43939 [27:52:37<42:09:59, 5.70s/it]
39%|███▉ | 17288/43939 [27:52:42<42:08:26, 5.69s/it]
39%|███▉ | 17289/43939 [27:52:48<42:08:34, 5.69s/it]
39%|███▉ | 17290/43939 [27:52:54<42:06:37, 5.69s/it]
39%|███▉ | 17291/43939 [27:52:59<42:09:05, 5.69s/it]
39%|███▉ | 17292/43939 [27:53:05<42:16:19, 5.71s/it]
39%|███▉ | 17293/43939 [27:53:11<42:12:18, 5.70s/it]
39%|███▉ | 17294/43939 [27:53:17<42:10:03, 5.70s/it]
39%|███▉ | 17295/43939 [27:53:23<42:45:01, 5.78s/it]
39%|███▉ | 17296/43939 [27:53:28<42:38:14, 5.76s/it]
39%|███▉ | 17297/43939 [27:53:34<42:40:04, 5.77s/it]
39%|███▉ | 17298/43939 [27:53:40<42:34:02, 5.75s/it]
39%|███▉ | 17299/43939 [27:53:45<42:25:06, 5.73s/it]
39%|███▉ | 17300/43939 [27:53:51<42:22:17, 5.73s/it]
{'loss': 0.9124, 'grad_norm': 1.0773036479949951, 'learning_rate': 3.7971665970635124e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.61, 'epoch': 1.57}
39%|███▉ | 17300/43939 [27:53:51<42:22:17, 5.73s/it]
39%|███▉ | 17301/43939 [27:53:57<42:23:49, 5.73s/it]
39%|███▉ | 17302/43939 [27:54:03<42:24:08, 5.73s/it]
39%|███▉ | 17303/43939 [27:54:08<42:22:17, 5.73s/it]
39%|███▉ | 17304/43939 [27:54:14<42:19:37, 5.72s/it]
39%|███▉ | 17305/43939 [27:54:20<42:15:04, 5.71s/it]
39%|███▉ | 17306/43939 [27:54:25<42:07:43, 5.69s/it]
39%|███▉ | 17307/43939 [27:54:31<42:12:37, 5.71s/it]
39%|███▉ | 17308/43939 [27:54:37<42:06:45, 5.69s/it]
39%|███▉ | 17309/43939 [27:54:43<42:12:25, 5.71s/it]
39%|███▉ | 17310/43939 [27:54:48<42:08:51, 5.70s/it]
39%|███▉ | 17311/43939 [27:54:54<42:12:04, 5.71s/it]
39%|███▉ | 17312/43939 [27:55:00<42:09:43, 5.70s/it]
39%|███▉ | 17313/43939 [27:55:05<42:09:34, 5.70s/it]
39%|███▉ | 17314/43939 [27:55:11<42:09:53, 5.70s/it]
39%|███▉ | 17315/43939 [27:55:17<42:15:47, 5.71s/it]
39%|███▉ | 17316/43939 [27:55:22<42:14:34, 5.71s/it]
39%|███▉ | 17317/43939 [27:55:28<42:44:57, 5.78s/it]
39%|███▉ | 17318/43939 [27:55:34<42:31:57, 5.75s/it]
39%|███▉ | 17319/43939 [27:55:40<42:23:46, 5.73s/it]
39%|███▉ | 17320/43939 [27:55:45<42:18:49, 5.72s/it]
39%|███▉ | 17321/43939 [27:55:51<42:19:47, 5.72s/it]
39%|███▉ | 17322/43939 [27:55:57<42:19:48, 5.73s/it]
39%|███▉ | 17323/43939 [27:56:03<42:12:15, 5.71s/it]
39%|███▉ | 17324/43939 [27:56:08<42:09:15, 5.70s/it]
39%|███▉ | 17325/43939 [27:56:14<42:10:23, 5.70s/it]
39%|███▉ | 17326/43939 [27:56:20<42:07:52, 5.70s/it]
39%|███▉ | 17327/43939 [27:56:25<42:09:26, 5.70s/it]
39%|███▉ | 17328/43939 [27:56:31<42:17:01, 5.72s/it]
39%|███▉ | 17329/43939 [27:56:37<42:12:01, 5.71s/it]
39%|███▉ | 17330/43939 [27:56:43<42:13:55, 5.71s/it]
39%|███▉ | 17331/43939 [27:56:48<42:12:31, 5.71s/it]
39%|███▉ | 17332/43939 [27:56:54<42:12:46, 5.71s/it]
39%|███▉ | 17333/43939 [27:57:00<42:11:54, 5.71s/it]
39%|███▉ | 17334/43939 [27:57:05<42:16:15, 5.72s/it]
39%|███▉ | 17335/43939 [27:57:11<42:12:57, 5.71s/it]
39%|███▉ | 17336/43939 [27:57:17<42:12:43, 5.71s/it]
39%|███▉ | 17337/43939 [27:57:23<42:12:28, 5.71s/it]
39%|███▉ | 17338/43939 [27:57:29<42:47:30, 5.79s/it]
39%|███▉ | 17339/43939 [27:57:34<42:32:56, 5.76s/it]
39%|███▉ | 17340/43939 [27:57:40<42:25:21, 5.74s/it]
39%|███▉ | 17341/43939 [27:57:46<42:15:56, 5.72s/it]
39%|███▉ | 17342/43939 [27:57:51<42:11:48, 5.71s/it]
39%|███▉ | 17343/43939 [27:57:57<42:06:16, 5.70s/it]
39%|███▉ | 17344/43939 [27:58:03<42:10:34, 5.71s/it]
39%|███▉ | 17345/43939 [27:58:08<42:10:14, 5.71s/it]
39%|███▉ | 17346/43939 [27:58:14<42:12:43, 5.71s/it]
39%|███▉ | 17347/43939 [27:58:20<42:13:54, 5.72s/it]
39%|███▉ | 17348/43939 [27:58:26<42:09:53, 5.71s/it]
39%|███▉ | 17349/43939 [27:58:31<42:09:00, 5.71s/it]
39%|███▉ | 17350/43939 [27:58:37<42:07:34, 5.70s/it]
{'loss': 0.9285, 'grad_norm': 1.438437581062317, 'learning_rate': 3.788667516495732e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.1, 'epoch': 1.58}
39%|███▉ | 17350/43939 [27:58:37<42:07:34, 5.70s/it]
39%|███▉ | 17351/43939 [27:58:43<42:05:15, 5.70s/it]
39%|███▉ | 17352/43939 [27:58:48<42:07:01, 5.70s/it]
39%|███▉ | 17353/43939 [27:58:54<42:04:08, 5.70s/it]
39%|███▉ | 17354/43939 [27:59:00<42:03:33, 5.70s/it]
39%|███▉ | 17355/43939 [27:59:05<42:04:44, 5.70s/it]
40%|███▉ | 17356/43939 [27:59:11<42:05:15, 5.70s/it]
40%|███▉ | 17357/43939 [27:59:17<42:04:25, 5.70s/it]
40%|███▉ | 17358/43939 [27:59:23<42:06:19, 5.70s/it]
40%|███▉ | 17359/43939 [27:59:28<42:32:30, 5.76s/it]
40%|███▉ | 17360/43939 [27:59:34<42:24:01, 5.74s/it]
40%|███▉ | 17361/43939 [27:59:40<42:14:12, 5.72s/it]
40%|███▉ | 17362/43939 [27:59:45<42:14:48, 5.72s/it]
40%|███▉ | 17363/43939 [27:59:51<42:05:29, 5.70s/it]
40%|███▉ | 17364/43939 [27:59:57<42:08:24, 5.71s/it]
40%|███▉ | 17365/43939 [28:00:03<42:12:02, 5.72s/it]
40%|███▉ | 17366/43939 [28:00:08<42:11:27, 5.72s/it]
40%|███▉ | 17367/43939 [28:00:14<42:11:01, 5.72s/it]
40%|███▉ | 17368/43939 [28:00:20<42:07:49, 5.71s/it]
40%|███▉ | 17369/43939 [28:00:25<42:06:20, 5.70s/it]
40%|███▉ | 17370/43939 [28:00:31<42:02:06, 5.70s/it]
40%|███▉ | 17371/43939 [28:00:37<42:03:57, 5.70s/it]
40%|███▉ | 17372/43939 [28:00:43<42:08:10, 5.71s/it]
40%|███▉ | 17373/43939 [28:00:48<42:10:14, 5.71s/it]
40%|███▉ | 17374/43939 [28:00:54<42:13:21, 5.72s/it]
40%|███▉ | 17375/43939 [28:01:00<42:15:19, 5.73s/it]
40%|███▉ | 17376/43939 [28:01:05<42:07:15, 5.71s/it]
40%|███▉ | 17377/43939 [28:01:11<42:06:34, 5.71s/it]
40%|███▉ | 17378/43939 [28:01:17<42:07:23, 5.71s/it]
40%|███▉ | 17379/43939 [28:01:22<41:59:41, 5.69s/it]
40%|███▉ | 17380/43939 [28:01:28<42:31:43, 5.76s/it]
40%|███▉ | 17381/43939 [28:01:34<42:25:14, 5.75s/it]
40%|███▉ | 17382/43939 [28:01:40<42:19:02, 5.74s/it]
40%|███▉ | 17383/43939 [28:01:46<42:09:25, 5.71s/it]
40%|███▉ | 17384/43939 [28:01:51<42:05:49, 5.71s/it]
40%|███▉ | 17385/43939 [28:01:57<42:04:29, 5.70s/it]
40%|███▉ | 17386/43939 [28:02:03<42:03:58, 5.70s/it]
40%|███▉ | 17387/43939 [28:02:08<41:59:52, 5.69s/it]
40%|███▉ | 17388/43939 [28:02:14<42:02:46, 5.70s/it]
40%|███▉ | 17389/43939 [28:02:20<42:02:07, 5.70s/it]
40%|███▉ | 17390/43939 [28:02:25<42:04:44, 5.71s/it]
40%|███▉ | 17391/43939 [28:02:31<41:58:30, 5.69s/it]
40%|███▉ | 17392/43939 [28:02:37<42:03:56, 5.70s/it]
40%|███▉ | 17393/43939 [28:02:42<42:01:15, 5.70s/it]
40%|███▉ | 17394/43939 [28:02:48<42:05:30, 5.71s/it]
40%|███▉ | 17395/43939 [28:02:54<42:04:54, 5.71s/it]
40%|███▉ | 17396/43939 [28:03:00<41:58:23, 5.69s/it]
40%|███▉ | 17397/43939 [28:03:05<41:55:30, 5.69s/it]
40%|███▉ | 17398/43939 [28:03:11<41:58:30, 5.69s/it]
40%|███▉ | 17399/43939 [28:03:17<42:00:12, 5.70s/it]
40%|███▉ | 17400/43939 [28:03:22<42:04:26, 5.71s/it]
{'loss': 0.9061, 'grad_norm': 1.0797569751739502, 'learning_rate': 3.780148104169552e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.69, 'epoch': 1.58}
40%|███▉ | 17400/43939 [28:03:22<42:04:26, 5.71s/it]
40%|███▉ | 17401/43939 [28:03:28<42:31:19, 5.77s/it]
40%|███▉ | 17402/43939 [28:03:34<42:17:10, 5.74s/it]
40%|███▉ | 17403/43939 [28:03:40<42:09:19, 5.72s/it]
40%|███▉ | 17404/43939 [28:03:45<42:07:45, 5.72s/it]
40%|███▉ | 17405/43939 [28:03:51<42:09:57, 5.72s/it]
40%|███▉ | 17406/43939 [28:03:57<42:07:54, 5.72s/it]
40%|███▉ | 17407/43939 [28:04:02<42:03:19, 5.71s/it]
40%|███▉ | 17408/43939 [28:04:08<42:03:05, 5.71s/it]
40%|███▉ | 17409/43939 [28:04:14<41:58:08, 5.70s/it]
40%|███▉ | 17410/43939 [28:04:20<41:58:36, 5.70s/it]
40%|███▉ | 17411/43939 [28:04:25<42:02:10, 5.70s/it]
40%|███▉ | 17412/43939 [28:04:31<42:00:34, 5.70s/it]
40%|███▉ | 17413/43939 [28:04:37<41:58:58, 5.70s/it]
40%|███▉ | 17414/43939 [28:04:42<42:04:38, 5.71s/it]
40%|███▉ | 17415/43939 [28:04:48<42:01:24, 5.70s/it]
40%|███▉ | 17416/43939 [28:04:54<42:03:26, 5.71s/it]
40%|███▉ | 17417/43939 [28:04:59<41:56:34, 5.69s/it]
40%|███▉ | 17418/43939 [28:05:05<41:58:34, 5.70s/it]
40%|███▉ | 17419/43939 [28:05:11<41:56:43, 5.69s/it]
40%|███▉ | 17420/43939 [28:05:17<41:55:49, 5.69s/it]
40%|███▉ | 17421/43939 [28:05:22<41:57:38, 5.70s/it]
40%|███▉ | 17422/43939 [28:05:28<42:32:16, 5.78s/it]
40%|███▉ | 17423/43939 [28:05:34<42:15:35, 5.74s/it]
40%|███▉ | 17424/43939 [28:05:40<42:09:25, 5.72s/it]
40%|███▉ | 17425/43939 [28:05:45<42:01:07, 5.71s/it]
40%|███▉ | 17426/43939 [28:05:51<41:56:06, 5.69s/it]
40%|███▉ | 17427/43939 [28:05:57<41:57:57, 5.70s/it]
40%|███▉ | 17428/43939 [28:06:02<41:57:00, 5.70s/it]
40%|███▉ | 17429/43939 [28:06:08<41:59:47, 5.70s/it]
40%|███▉ | 17430/43939 [28:06:14<41:58:23, 5.70s/it]
40%|███▉ | 17431/43939 [28:06:19<42:05:31, 5.72s/it]
40%|███▉ | 17432/43939 [28:06:25<42:09:50, 5.73s/it]
40%|███▉ | 17433/43939 [28:06:31<42:01:58, 5.71s/it]
40%|███▉ | 17434/43939 [28:06:37<42:05:46, 5.72s/it]
40%|███▉ | 17435/43939 [28:06:42<42:01:11, 5.71s/it]
40%|███▉ | 17436/43939 [28:06:48<41:59:03, 5.70s/it]
40%|███▉ | 17437/43939 [28:06:54<41:56:00, 5.70s/it]
40%|███▉ | 17438/43939 [28:06:59<41:56:10, 5.70s/it]
40%|███▉ | 17439/43939 [28:07:05<41:50:13, 5.68s/it]
40%|███▉ | 17440/43939 [28:07:11<41:48:35, 5.68s/it]
40%|███▉ | 17441/43939 [28:07:16<41:49:51, 5.68s/it]
40%|███▉ | 17442/43939 [28:07:22<41:56:47, 5.70s/it]
40%|███▉ | 17443/43939 [28:07:28<42:32:16, 5.78s/it]
40%|███▉ | 17444/43939 [28:07:34<42:29:16, 5.77s/it]
40%|███▉ | 17445/43939 [28:07:40<42:22:17, 5.76s/it]
40%|███▉ | 17446/43939 [28:07:45<42:11:27, 5.73s/it]
40%|███▉ | 17447/43939 [28:07:51<42:10:04, 5.73s/it]
40%|███▉ | 17448/43939 [28:07:57<42:09:21, 5.73s/it]
40%|███▉ | 17449/43939 [28:08:02<42:05:04, 5.72s/it]
40%|███▉ | 17450/43939 [28:08:08<41:59:04, 5.71s/it]
{'loss': 0.9014, 'grad_norm': 2.565443992614746, 'learning_rate': 3.771608494498722e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.42, 'epoch': 1.59}
40%|███▉ | 17450/43939 [28:08:08<41:59:04, 5.71s/it]
40%|███▉ | 17451/43939 [28:08:14<42:01:33, 5.71s/it]
40%|███▉ | 17452/43939 [28:08:19<41:57:34, 5.70s/it]
40%|███▉ | 17453/43939 [28:08:25<41:55:13, 5.70s/it]
40%|███▉ | 17454/43939 [28:08:31<41:53:32, 5.69s/it]
40%|███▉ | 17455/43939 [28:08:36<41:49:13, 5.68s/it]
40%|███▉ | 17456/43939 [28:08:42<41:55:31, 5.70s/it]
40%|███▉ | 17457/43939 [28:08:48<41:56:42, 5.70s/it]
40%|███▉ | 17458/43939 [28:08:54<42:03:36, 5.72s/it]
40%|███▉ | 17459/43939 [28:08:59<42:02:18, 5.72s/it]
40%|███▉ | 17460/43939 [28:09:05<42:00:11, 5.71s/it]
40%|███▉ | 17461/43939 [28:09:11<41:58:34, 5.71s/it]
40%|███▉ | 17462/43939 [28:09:16<41:56:36, 5.70s/it]
40%|███▉ | 17463/43939 [28:09:22<41:59:51, 5.71s/it]
40%|███▉ | 17464/43939 [28:09:28<42:34:53, 5.79s/it]
40%|███▉ | 17465/43939 [28:09:34<42:26:03, 5.77s/it]
40%|███▉ | 17466/43939 [28:09:40<42:15:38, 5.75s/it]
40%|███▉ | 17467/43939 [28:09:45<42:12:48, 5.74s/it]
40%|███▉ | 17468/43939 [28:09:51<42:08:23, 5.73s/it]
40%|███▉ | 17469/43939 [28:09:57<42:01:36, 5.72s/it]
40%|███▉ | 17470/43939 [28:10:02<41:58:28, 5.71s/it]
40%|███▉ | 17471/43939 [28:10:08<42:00:12, 5.71s/it]
40%|███▉ | 17472/43939 [28:10:14<41:55:17, 5.70s/it]
40%|███▉ | 17473/43939 [28:10:20<42:06:27, 5.73s/it]
40%|███▉ | 17474/43939 [28:10:25<41:57:07, 5.71s/it]
40%|███▉ | 17475/43939 [28:10:31<41:58:14, 5.71s/it]
40%|███▉ | 17476/43939 [28:10:37<42:01:11, 5.72s/it]
40%|███▉ | 17477/43939 [28:10:42<41:58:50, 5.71s/it]
40%|███▉ | 17478/43939 [28:10:48<42:02:12, 5.72s/it]
40%|███▉ | 17479/43939 [28:10:54<42:00:10, 5.71s/it]
40%|███▉ | 17480/43939 [28:11:00<42:01:26, 5.72s/it]
40%|███▉ | 17481/43939 [28:11:05<42:00:11, 5.72s/it]
40%|███▉ | 17482/43939 [28:11:11<42:05:34, 5.73s/it]
40%|███▉ | 17483/43939 [28:11:17<42:08:35, 5.73s/it]
40%|███▉ | 17484/43939 [28:11:23<42:38:59, 5.80s/it]
40%|███▉ | 17485/43939 [28:11:28<42:30:48, 5.79s/it]
40%|███▉ | 17486/43939 [28:11:34<42:20:01, 5.76s/it]
40%|███▉ | 17487/43939 [28:11:40<42:06:09, 5.73s/it]
40%|███▉ | 17488/43939 [28:11:46<42:02:23, 5.72s/it]
40%|███▉ | 17489/43939 [28:11:51<41:59:35, 5.72s/it]
40%|███▉ | 17490/43939 [28:11:57<41:53:32, 5.70s/it]
40%|███▉ | 17491/43939 [28:12:03<41:57:57, 5.71s/it]
40%|███▉ | 17492/43939 [28:12:08<41:54:08, 5.70s/it]
40%|███▉ | 17493/43939 [28:12:14<41:56:12, 5.71s/it]
40%|███▉ | 17494/43939 [28:12:20<41:55:38, 5.71s/it]
40%|███▉ | 17495/43939 [28:12:25<41:54:50, 5.71s/it]
40%|███▉ | 17496/43939 [28:12:31<41:55:04, 5.71s/it]
40%|███▉ | 17497/43939 [28:12:37<41:49:47, 5.70s/it]
40%|███▉ | 17498/43939 [28:12:43<41:49:47, 5.70s/it]
40%|███▉ | 17499/43939 [28:12:48<41:57:44, 5.71s/it]
40%|███▉ | 17500/43939 [28:12:54<42:02:30, 5.72s/it]
{'loss': 0.9448, 'grad_norm': 1.248572826385498, 'learning_rate': 3.763048822215652e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.78, 'epoch': 1.59}
40%|███▉ | 17500/43939 [28:12:54<42:02:30, 5.72s/it]
40%|███▉ | 17501/43939 [28:13:00<41:55:41, 5.71s/it]
40%|███▉ | 17502/43939 [28:13:05<41:54:51, 5.71s/it]
40%|███▉ | 17503/43939 [28:13:11<41:53:30, 5.70s/it]
40%|███▉ | 17504/43939 [28:13:17<41:52:45, 5.70s/it]
40%|███▉ | 17505/43939 [28:13:23<42:25:50, 5.78s/it]
40%|███▉ | 17506/43939 [28:13:28<42:14:14, 5.75s/it]
40%|███▉ | 17507/43939 [28:13:34<42:13:17, 5.75s/it]
40%|███▉ | 17508/43939 [28:13:40<42:07:40, 5.74s/it]
40%|███▉ | 17509/43939 [28:13:46<41:57:59, 5.72s/it]
40%|███▉ | 17510/43939 [28:13:51<41:54:52, 5.71s/it]
40%|███▉ | 17511/43939 [28:13:57<41:56:38, 5.71s/it]
40%|███▉ | 17512/43939 [28:14:03<41:55:14, 5.71s/it]
40%|███▉ | 17513/43939 [28:14:08<41:52:55, 5.71s/it]
40%|███▉ | 17514/43939 [28:14:14<41:51:07, 5.70s/it]
40%|███▉ | 17515/43939 [28:14:20<41:50:47, 5.70s/it]
40%|███▉ | 17516/43939 [28:14:25<41:49:03, 5.70s/it]
40%|███▉ | 17517/43939 [28:14:31<41:52:39, 5.71s/it]
40%|███▉ | 17518/43939 [28:14:37<41:45:12, 5.69s/it]
40%|███▉ | 17519/43939 [28:14:43<41:47:52, 5.70s/it]
40%|███▉ | 17520/43939 [28:14:48<41:50:33, 5.70s/it]
40%|███▉ | 17521/43939 [28:14:54<41:47:43, 5.70s/it]
40%|███▉ | 17522/43939 [28:15:00<41:47:50, 5.70s/it]
40%|███▉ | 17523/43939 [28:15:05<41:49:37, 5.70s/it]
40%|███▉ | 17524/43939 [28:15:11<41:46:27, 5.69s/it]
40%|███▉ | 17525/43939 [28:15:17<41:51:37, 5.71s/it]
40%|███▉ | 17526/43939 [28:15:22<41:50:44, 5.70s/it]
40%|███▉ | 17527/43939 [28:15:28<42:23:56, 5.78s/it]
40%|███▉ | 17528/43939 [28:15:34<42:10:13, 5.75s/it]
40%|███▉ | 17529/43939 [28:15:40<42:03:04, 5.73s/it]
40%|███▉ | 17530/43939 [28:15:46<42:04:34, 5.74s/it]
40%|███▉ | 17531/43939 [28:15:51<41:54:26, 5.71s/it]
40%|███▉ | 17532/43939 [28:15:57<41:55:29, 5.72s/it]
40%|███▉ | 17533/43939 [28:16:03<41:49:46, 5.70s/it]
40%|███▉ | 17534/43939 [28:16:08<41:57:22, 5.72s/it]
40%|███▉ | 17535/43939 [28:16:14<41:51:11, 5.71s/it]
40%|███▉ | 17536/43939 [28:16:20<41:52:01, 5.71s/it]
40%|███▉ | 17537/43939 [28:16:25<41:46:12, 5.70s/it]
40%|███▉ | 17538/43939 [28:16:31<41:52:58, 5.71s/it]
40%|███▉ | 17539/43939 [28:16:37<41:50:13, 5.71s/it]
40%|███▉ | 17540/43939 [28:16:43<41:46:20, 5.70s/it]
40%|███▉ | 17541/43939 [28:16:48<41:46:41, 5.70s/it]
40%|███▉ | 17542/43939 [28:16:54<41:45:15, 5.69s/it]
40%|███▉ | 17543/43939 [28:17:00<41:44:07, 5.69s/it]
40%|███▉ | 17544/43939 [28:17:05<41:45:19, 5.69s/it]
40%|███▉ | 17545/43939 [28:17:11<41:42:20, 5.69s/it]
40%|███▉ | 17546/43939 [28:17:17<41:38:40, 5.68s/it]
40%|███▉ | 17547/43939 [28:17:23<42:05:03, 5.74s/it]
40%|███▉ | 17548/43939 [28:17:28<42:05:35, 5.74s/it]
40%|███▉ | 17549/43939 [28:17:34<42:08:52, 5.75s/it]
40%|███▉ | 17550/43939 [28:17:40<42:04:17, 5.74s/it]
{'loss': 0.9445, 'grad_norm': 1.2229149341583252, 'learning_rate': 3.754469222369288e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.23, 'epoch': 1.6}
40%|███▉ | 17550/43939 [28:17:40<42:04:17, 5.74s/it]
40%|███▉ | 17551/43939 [28:17:45<41:58:11, 5.73s/it]
40%|███▉ | 17552/43939 [28:17:51<42:01:40, 5.73s/it]
40%|███▉ | 17553/43939 [28:17:57<41:56:16, 5.72s/it]
40%|███▉ | 17554/43939 [28:18:03<41:48:38, 5.70s/it]
40%|███▉ | 17555/43939 [28:18:08<41:52:51, 5.71s/it]
40%|███▉ | 17556/43939 [28:18:14<41:53:19, 5.72s/it]
40%|███▉ | 17557/43939 [28:18:20<41:51:13, 5.71s/it]
40%|███▉ | 17558/43939 [28:18:25<41:45:47, 5.70s/it]
40%|███▉ | 17559/43939 [28:18:31<41:39:14, 5.68s/it]
40%|███▉ | 17560/43939 [28:18:37<41:44:05, 5.70s/it]
40%|███▉ | 17561/43939 [28:18:42<41:43:50, 5.70s/it]
40%|███▉ | 17562/43939 [28:18:48<41:43:29, 5.69s/it]
40%|███▉ | 17563/43939 [28:18:54<41:46:10, 5.70s/it]
40%|███▉ | 17564/43939 [28:19:00<41:46:52, 5.70s/it]
40%|███▉ | 17565/43939 [28:19:05<41:49:36, 5.71s/it]
40%|███▉ | 17566/43939 [28:19:11<41:47:38, 5.71s/it]
40%|███▉ | 17567/43939 [28:19:17<41:44:11, 5.70s/it]
40%|███▉ | 17568/43939 [28:19:23<42:17:13, 5.77s/it]
40%|███▉ | 17569/43939 [28:19:28<42:06:29, 5.75s/it]
40%|███▉ | 17570/43939 [28:19:34<41:59:59, 5.73s/it]
40%|███▉ | 17571/43939 [28:19:40<41:55:59, 5.73s/it]
40%|███▉ | 17572/43939 [28:19:45<41:49:55, 5.71s/it]
40%|███▉ | 17573/43939 [28:19:51<41:50:55, 5.71s/it]
40%|███▉ | 17574/43939 [28:19:57<41:50:15, 5.71s/it]
40%|███▉ | 17575/43939 [28:20:03<41:48:30, 5.71s/it]
40%|████ | 17576/43939 [28:20:08<41:51:07, 5.72s/it][2025-11-29 22:24:22,083] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:34, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.74s/it][A
38%|███▊ | 27/72 [01:11<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.74s/it][A
43%|████▎ | 31/72 [01:21<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:32<01:41, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<01:00, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:44, 2.75s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.75s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.74s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.74s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.74s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.74s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.74s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.978286862373352, 'eval_runtime': 196.934, 'eval_samples_per_second': 5.073, 'eval_steps_per_second': 0.366, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 1.6}
40%|████ | 17576/43939 [28:23:25<41:51:07, 5.72s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
40%|████ | 17577/43939 [28:23:31<474:42:31, 64.83s/it]
40%|████ | 17578/43939 [28:23:37<344:42:56, 47.08s/it]
40%|████ | 17579/43939 [28:23:42<253:47:32, 34.66s/it]
40%|████ | 17580/43939 [28:23:48<190:11:23, 25.98s/it]
40%|████ | 17581/43939 [28:23:54<145:41:39, 19.90s/it]
40%|████ | 17582/43939 [28:24:00<114:30:48, 15.64s/it]
40%|████ | 17583/43939 [28:24:05<92:39:35, 12.66s/it]
40%|████ | 17584/43939 [28:24:11<77:26:53, 10.58s/it]
40%|████ | 17585/43939 [28:24:17<66:48:39, 9.13s/it]
40%|████ | 17586/43939 [28:24:22<59:11:54, 8.09s/it]
40%|████ | 17587/43939 [28:24:28<53:57:32, 7.37s/it]
40%|████ | 17588/43939 [28:24:34<50:19:32, 6.88s/it]
40%|████ | 17589/43939 [28:24:39<47:50:04, 6.54s/it]
40%|████ | 17590/43939 [28:24:45<46:39:49, 6.38s/it]
40%|████ | 17591/43939 [28:24:51<45:08:47, 6.17s/it]
40%|████ | 17592/43939 [28:24:57<44:04:10, 6.02s/it]
40%|████ | 17593/43939 [28:25:03<43:17:56, 5.92s/it]
40%|████ | 17594/43939 [28:25:08<42:46:11, 5.84s/it]
40%|████ | 17595/43939 [28:25:14<42:26:22, 5.80s/it]
40%|████ | 17596/43939 [28:25:20<42:13:03, 5.77s/it]
40%|████ | 17597/43939 [28:25:25<42:01:23, 5.74s/it]
40%|████ | 17598/43939 [28:25:31<41:56:14, 5.73s/it]
40%|████ | 17599/43939 [28:25:37<41:54:14, 5.73s/it]
40%|████ | 17600/43939 [28:25:42<41:46:34, 5.71s/it]
{'loss': 0.9362, 'grad_norm': 1.1244183778762817, 'learning_rate': 3.745869830322979e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 234.94, 'epoch': 1.6}
40%|████ | 17600/43939 [28:25:42<41:46:34, 5.71s/it]
40%|████ | 17601/43939 [28:25:48<41:46:35, 5.71s/it]
40%|████ | 17602/43939 [28:25:54<41:41:48, 5.70s/it]
40%|████ | 17603/43939 [28:25:59<41:43:27, 5.70s/it]
40%|████ | 17604/43939 [28:26:05<41:46:15, 5.71s/it]
40%|████ | 17605/43939 [28:26:11<41:46:40, 5.71s/it]
40%|████ | 17606/43939 [28:26:17<41:37:12, 5.69s/it]
40%|████ | 17607/43939 [28:26:22<41:38:21, 5.69s/it]
40%|████ | 17608/43939 [28:26:28<41:36:08, 5.69s/it]
40%|████ | 17609/43939 [28:26:34<41:32:19, 5.68s/it]
40%|████ | 17610/43939 [28:26:40<42:09:06, 5.76s/it]
40%|████ | 17611/43939 [28:26:45<42:03:59, 5.75s/it]
40%|████ | 17612/43939 [28:26:51<42:00:52, 5.75s/it]
40%|████ | 17613/43939 [28:26:57<41:52:52, 5.73s/it]
40%|████ | 17614/43939 [28:27:02<41:48:25, 5.72s/it]
40%|████ | 17615/43939 [28:27:08<41:47:05, 5.71s/it]
40%|████ | 17616/43939 [28:27:14<41:43:31, 5.71s/it]
40%|████ | 17617/43939 [28:27:19<41:39:43, 5.70s/it]
40%|████ | 17618/43939 [28:27:25<41:36:36, 5.69s/it]
40%|████ | 17619/43939 [28:27:31<41:38:20, 5.70s/it]
40%|████ | 17620/43939 [28:27:37<41:37:11, 5.69s/it]
40%|████ | 17621/43939 [28:27:42<41:34:22, 5.69s/it]
40%|████ | 17622/43939 [28:27:48<41:42:02, 5.70s/it]
40%|████ | 17623/43939 [28:27:54<41:43:15, 5.71s/it]
40%|████ | 17624/43939 [28:27:59<41:42:21, 5.71s/it]
40%|████ | 17625/43939 [28:28:05<41:46:55, 5.72s/it]
40%|████ | 17626/43939 [28:28:11<41:46:09, 5.71s/it]
40%|████ | 17627/43939 [28:28:17<41:47:17, 5.72s/it]
40%|████ | 17628/43939 [28:28:22<41:44:08, 5.71s/it]
40%|████ | 17629/43939 [28:28:28<41:44:53, 5.71s/it]
40%|████ | 17630/43939 [28:28:34<41:40:24, 5.70s/it]
40%|████ | 17631/43939 [28:28:40<42:13:40, 5.78s/it]
40%|████ | 17632/43939 [28:28:45<41:59:59, 5.75s/it]
40%|████ | 17633/43939 [28:28:51<41:53:49, 5.73s/it]
40%|████ | 17634/43939 [28:28:57<41:45:28, 5.71s/it]
40%|████ | 17635/43939 [28:29:02<41:44:52, 5.71s/it]
40%|████ | 17636/43939 [28:29:08<41:36:30, 5.69s/it]
40%|████ | 17637/43939 [28:29:14<41:34:32, 5.69s/it]
40%|████ | 17638/43939 [28:29:19<41:39:21, 5.70s/it]
40%|████ | 17639/43939 [28:29:25<41:37:13, 5.70s/it]
40%|████ | 17640/43939 [28:29:31<41:38:26, 5.70s/it]
40%|████ | 17641/43939 [28:29:36<41:39:18, 5.70s/it]
40%|████ | 17642/43939 [28:29:42<41:40:16, 5.70s/it]
40%|████ | 17643/43939 [28:29:48<41:43:24, 5.71s/it]
40%|████ | 17644/43939 [28:29:54<41:42:49, 5.71s/it]
40%|████ | 17645/43939 [28:29:59<41:35:33, 5.69s/it]
40%|████ | 17646/43939 [28:30:05<41:32:17, 5.69s/it]
40%|████ | 17647/43939 [28:30:11<41:30:34, 5.68s/it]
40%|████ | 17648/43939 [28:30:16<41:29:26, 5.68s/it]
40%|████ | 17649/43939 [28:30:22<41:29:51, 5.68s/it]
40%|████ | 17650/43939 [28:30:28<41:37:09, 5.70s/it]
{'loss': 0.916, 'grad_norm': 0.8700757026672363, 'learning_rate': 3.7372507817523425e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.41, 'epoch': 1.61}
40%|████ | 17650/43939 [28:30:28<41:37:09, 5.70s/it]
40%|████ | 17651/43939 [28:30:33<41:36:49, 5.70s/it]
40%|████ | 17652/43939 [28:30:39<41:41:33, 5.71s/it]
40%|████ | 17653/43939 [28:30:45<42:10:19, 5.78s/it]
40%|████ | 17654/43939 [28:30:51<41:57:19, 5.75s/it]
40%|████ | 17655/43939 [28:30:56<41:46:46, 5.72s/it]
40%|████ | 17656/43939 [28:31:02<41:53:47, 5.74s/it]
40%|████ | 17657/43939 [28:31:08<41:47:56, 5.73s/it]
40%|████ | 17658/43939 [28:31:14<41:43:26, 5.72s/it]
40%|████ | 17659/43939 [28:31:19<41:48:45, 5.73s/it]
40%|████ | 17660/43939 [28:31:25<41:46:06, 5.72s/it]
40%|████ | 17661/43939 [28:31:31<41:43:03, 5.72s/it]
40%|████ | 17662/43939 [28:31:36<41:41:52, 5.71s/it]
40%|████ | 17663/43939 [28:31:42<41:42:14, 5.71s/it]
40%|████ | 17664/43939 [28:31:48<41:50:33, 5.73s/it]
40%|████ | 17665/43939 [28:31:54<41:48:25, 5.73s/it]
40%|████ | 17666/43939 [28:31:59<41:50:24, 5.73s/it]
40%|████ | 17667/43939 [28:32:05<41:47:54, 5.73s/it]
40%|████ | 17668/43939 [28:32:11<41:44:58, 5.72s/it]
40%|████ | 17669/43939 [28:32:17<41:44:10, 5.72s/it]
40%|████ | 17670/43939 [28:32:22<41:39:18, 5.71s/it]
40%|████ | 17671/43939 [28:32:28<41:38:02, 5.71s/it]
40%|████ | 17672/43939 [28:32:34<41:35:01, 5.70s/it]
40%|████ | 17673/43939 [28:32:39<41:30:58, 5.69s/it]
40%|████ | 17674/43939 [28:32:45<41:58:02, 5.75s/it]
40%|████ | 17675/43939 [28:32:51<41:54:19, 5.74s/it]
40%|████ | 17676/43939 [28:32:57<41:48:54, 5.73s/it]
40%|████ | 17677/43939 [28:33:02<41:52:02, 5.74s/it]
40%|████ | 17678/43939 [28:33:08<41:48:51, 5.73s/it]
40%|████ | 17679/43939 [28:33:14<41:49:56, 5.73s/it]
40%|████ | 17680/43939 [28:33:20<41:45:04, 5.72s/it]
40%|████ | 17681/43939 [28:33:25<41:40:43, 5.71s/it]
40%|████ | 17682/43939 [28:33:31<41:41:43, 5.72s/it]
40%|████ | 17683/43939 [28:33:37<41:38:39, 5.71s/it]
40%|████ | 17684/43939 [28:33:42<41:34:17, 5.70s/it]
40%|████ | 17685/43939 [28:33:48<41:32:37, 5.70s/it]
40%|████ | 17686/43939 [28:33:54<41:44:17, 5.72s/it]
40%|████ | 17687/43939 [28:33:59<41:39:58, 5.71s/it]
40%|████ | 17688/43939 [28:34:05<41:32:58, 5.70s/it]
40%|████ | 17689/43939 [28:34:11<41:33:15, 5.70s/it]
40%|████ | 17690/43939 [28:34:17<41:33:58, 5.70s/it]
40%|████ | 17691/43939 [28:34:22<41:34:04, 5.70s/it]
40%|████ | 17692/43939 [28:34:28<41:41:25, 5.72s/it]
40%|████ | 17693/43939 [28:34:34<41:42:19, 5.72s/it]
40%|████ | 17694/43939 [28:34:39<41:42:08, 5.72s/it]
40%|████ | 17695/43939 [28:34:45<42:10:42, 5.79s/it]
40%|████ | 17696/43939 [28:34:51<41:59:02, 5.76s/it]
40%|████ | 17697/43939 [28:34:57<41:56:03, 5.75s/it]
40%|████ | 17698/43939 [28:35:03<41:53:39, 5.75s/it]
40%|████ | 17699/43939 [28:35:08<41:48:08, 5.74s/it]
40%|████ | 17700/43939 [28:35:14<41:43:23, 5.72s/it]
{'loss': 0.9229, 'grad_norm': 1.0719106197357178, 'learning_rate': 3.7286122126431243e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.3, 'epoch': 1.61}
40%|████ | 17700/43939 [28:35:14<41:43:23, 5.72s/it]
40%|████ | 17701/43939 [28:35:20<41:38:13, 5.71s/it]
40%|████ | 17702/43939 [28:35:25<41:37:23, 5.71s/it]
40%|████ | 17703/43939 [28:35:31<41:38:50, 5.71s/it]
40%|████ | 17704/43939 [28:35:37<41:36:59, 5.71s/it]
40%|████ | 17705/43939 [28:35:42<41:34:58, 5.71s/it]
40%|████ | 17706/43939 [28:35:48<41:34:07, 5.70s/it]
40%|████ | 17707/43939 [28:35:54<41:31:40, 5.70s/it]
40%|████ | 17708/43939 [28:36:00<41:29:16, 5.69s/it]
40%|████ | 17709/43939 [28:36:05<41:28:41, 5.69s/it]
40%|████ | 17710/43939 [28:36:11<41:30:42, 5.70s/it]
40%|████ | 17711/43939 [28:36:17<41:31:00, 5.70s/it]
40%|████ | 17712/43939 [28:36:22<41:31:05, 5.70s/it]
40%|████ | 17713/43939 [28:36:28<41:36:26, 5.71s/it]
40%|████ | 17714/43939 [28:36:34<41:35:26, 5.71s/it]
40%|████ | 17715/43939 [28:36:40<41:37:58, 5.72s/it]
40%|████ | 17716/43939 [28:36:45<42:07:31, 5.78s/it]
40%|████ | 17717/43939 [28:36:51<42:01:44, 5.77s/it]
40%|████ | 17718/43939 [28:36:57<41:52:46, 5.75s/it]
40%|████ | 17719/43939 [28:37:03<41:48:32, 5.74s/it]
40%|████ | 17720/43939 [28:37:08<41:45:20, 5.73s/it]
40%|████ | 17721/43939 [28:37:14<41:42:04, 5.73s/it]
40%|████ | 17722/43939 [28:37:20<41:35:56, 5.71s/it]
40%|████ | 17723/43939 [28:37:25<41:35:06, 5.71s/it]
40%|████ | 17724/43939 [28:37:31<41:40:26, 5.72s/it]
40%|████ | 17725/43939 [28:37:37<41:33:46, 5.71s/it]
40%|████ | 17726/43939 [28:37:43<41:30:59, 5.70s/it]
40%|████ | 17727/43939 [28:37:48<41:31:28, 5.70s/it]
40%|████ | 17728/43939 [28:37:54<41:31:43, 5.70s/it]
40%|████ | 17729/43939 [28:38:00<41:35:04, 5.71s/it]
40%|████ | 17730/43939 [28:38:05<41:33:56, 5.71s/it]
40%|████ | 17731/43939 [28:38:11<41:33:49, 5.71s/it]
40%|████ | 17732/43939 [28:38:17<41:30:50, 5.70s/it]
40%|████ | 17733/43939 [28:38:23<41:34:42, 5.71s/it]
40%|████ | 17734/43939 [28:38:28<41:35:27, 5.71s/it]
40%|████ | 17735/43939 [28:38:34<41:32:47, 5.71s/it]
40%|████ | 17736/43939 [28:38:40<41:30:21, 5.70s/it]
40%|████ | 17737/43939 [28:38:45<41:30:14, 5.70s/it]
40%|████ | 17738/43939 [28:38:51<42:02:10, 5.78s/it]
40%|████ | 17739/43939 [28:38:57<41:53:05, 5.76s/it]
40%|████ | 17740/43939 [28:39:03<41:54:50, 5.76s/it]
40%|████ | 17741/43939 [28:39:08<41:42:54, 5.73s/it]
40%|████ | 17742/43939 [28:39:14<41:43:47, 5.73s/it]
40%|████ | 17743/43939 [28:39:20<41:34:23, 5.71s/it]
40%|████ | 17744/43939 [28:39:26<41:34:59, 5.71s/it]
40%|████ | 17745/43939 [28:39:31<41:34:14, 5.71s/it]
40%|████ | 17746/43939 [28:39:37<41:32:00, 5.71s/it]
40%|████ | 17747/43939 [28:39:43<41:42:48, 5.73s/it]
40%|████ | 17748/43939 [28:39:48<41:42:49, 5.73s/it]
40%|████ | 17749/43939 [28:39:54<41:37:35, 5.72s/it]
40%|████ | 17750/43939 [28:40:00<41:33:12, 5.71s/it]
{'loss': 0.9109, 'grad_norm': 1.4649990797042847, 'learning_rate': 3.719954259289053e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.75, 'epoch': 1.62}
40%|████ | 17750/43939 [28:40:00<41:33:12, 5.71s/it]
40%|████ | 17751/43939 [28:40:06<41:26:21, 5.70s/it]
40%|████ | 17752/43939 [28:40:11<41:22:27, 5.69s/it]
40%|████ | 17753/43939 [28:40:17<41:19:34, 5.68s/it]
40%|████ | 17754/43939 [28:40:23<41:27:27, 5.70s/it]
40%|████ | 17755/43939 [28:40:28<41:24:48, 5.69s/it]
40%|████ | 17756/43939 [28:40:34<41:28:24, 5.70s/it]
40%|████ | 17757/43939 [28:40:40<41:36:53, 5.72s/it]
40%|████ | 17758/43939 [28:40:46<42:00:56, 5.78s/it]
40%|████ | 17759/43939 [28:40:51<41:47:54, 5.75s/it]
40%|████ | 17760/43939 [28:40:57<41:41:20, 5.73s/it]
40%|████ | 17761/43939 [28:41:03<41:43:56, 5.74s/it]
40%|████ | 17762/43939 [28:41:09<41:41:38, 5.73s/it]
40%|████ | 17763/43939 [28:41:14<41:41:11, 5.73s/it]
40%|████ | 17764/43939 [28:41:20<41:36:28, 5.72s/it]
40%|████ | 17765/43939 [28:41:26<41:32:35, 5.71s/it]
40%|████ | 17766/43939 [28:41:31<41:25:59, 5.70s/it]
40%|████ | 17767/43939 [28:41:37<41:24:24, 5.70s/it]
40%|████ | 17768/43939 [28:41:43<41:17:48, 5.68s/it]
40%|████ | 17769/43939 [28:41:48<41:19:02, 5.68s/it]
40%|████ | 17770/43939 [28:41:54<41:31:22, 5.71s/it]
40%|████ | 17771/43939 [28:42:00<41:35:42, 5.72s/it]
40%|████ | 17772/43939 [28:42:06<41:32:00, 5.71s/it]
40%|████ | 17773/43939 [28:42:11<41:29:18, 5.71s/it]
40%|████ | 17774/43939 [28:42:17<41:34:58, 5.72s/it]
40%|████ | 17775/43939 [28:42:23<41:32:16, 5.72s/it]
40%|████ | 17776/43939 [28:42:28<41:25:39, 5.70s/it]
40%|████ | 17777/43939 [28:42:34<41:27:23, 5.70s/it]
40%|████ | 17778/43939 [28:42:40<41:28:17, 5.71s/it]
40%|████ | 17779/43939 [28:42:46<41:56:57, 5.77s/it]
40%|████ | 17780/43939 [28:42:51<41:44:56, 5.75s/it]
40%|████ | 17781/43939 [28:42:57<41:37:50, 5.73s/it]
40%|████ | 17782/43939 [28:43:03<41:40:39, 5.74s/it]
40%|████ | 17783/43939 [28:43:09<41:36:05, 5.73s/it]
40%|████ | 17784/43939 [28:43:14<41:30:55, 5.71s/it]
40%|████ | 17785/43939 [28:43:20<41:28:09, 5.71s/it]
40%|████ | 17786/43939 [28:43:26<41:30:48, 5.71s/it]
40%|████ | 17787/43939 [28:43:31<41:28:26, 5.71s/it]
40%|████ | 17788/43939 [28:43:37<41:27:21, 5.71s/it]
40%|████ | 17789/43939 [28:43:43<41:25:55, 5.70s/it]
40%|████ | 17790/43939 [28:43:49<41:32:24, 5.72s/it]
40%|████ | 17791/43939 [28:43:54<41:29:55, 5.71s/it]
40%|████ | 17792/43939 [28:44:00<41:32:46, 5.72s/it]
40%|████ | 17793/43939 [28:44:06<41:24:08, 5.70s/it]
40%|████ | 17794/43939 [28:44:11<41:29:20, 5.71s/it]
40%|████ | 17795/43939 [28:44:17<41:27:47, 5.71s/it]
41%|████ | 17796/43939 [28:44:23<41:30:03, 5.71s/it]
41%|████ | 17797/43939 [28:44:28<41:25:11, 5.70s/it]
41%|████ | 17798/43939 [28:44:34<41:25:45, 5.71s/it]
41%|████ | 17799/43939 [28:44:40<41:23:54, 5.70s/it]
41%|████ | 17800/43939 [28:44:46<41:23:58, 5.70s/it]
{'loss': 0.9168, 'grad_norm': 1.1497403383255005, 'learning_rate': 3.711277058289687e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.72, 'epoch': 1.62}
41%|████ | 17800/43939 [28:44:46<41:23:58, 5.70s/it]
41%|████ | 17801/43939 [28:44:51<41:52:24, 5.77s/it]
41%|████ | 17802/43939 [28:44:57<41:50:11, 5.76s/it]
41%|████ | 17803/43939 [28:45:03<41:37:58, 5.73s/it]
41%|████ | 17804/43939 [28:45:09<41:29:35, 5.72s/it]
41%|████ | 17805/43939 [28:45:14<41:28:28, 5.71s/it]
41%|████ | 17806/43939 [28:45:20<41:31:32, 5.72s/it]
41%|████ | 17807/43939 [28:45:26<41:24:24, 5.70s/it]
41%|████ | 17808/43939 [28:45:31<41:23:44, 5.70s/it]
41%|████ | 17809/43939 [28:45:37<41:23:35, 5.70s/it]
41%|████ | 17810/43939 [28:45:43<41:24:01, 5.70s/it]
41%|████ | 17811/43939 [28:45:49<41:26:23, 5.71s/it]
41%|████ | 17812/43939 [28:45:54<41:24:21, 5.71s/it]
41%|████ | 17813/43939 [28:46:00<41:18:45, 5.69s/it]
41%|████ | 17814/43939 [28:46:06<41:18:55, 5.69s/it]
41%|████ | 17815/43939 [28:46:11<41:19:34, 5.69s/it]
41%|████ | 17816/43939 [28:46:17<41:17:06, 5.69s/it]
41%|████ | 17817/43939 [28:46:23<41:13:38, 5.68s/it]
41%|████ | 17818/43939 [28:46:28<41:15:39, 5.69s/it]
41%|████ | 17819/43939 [28:46:34<41:22:23, 5.70s/it]
41%|████ | 17820/43939 [28:46:40<41:22:52, 5.70s/it]
41%|████ | 17821/43939 [28:46:46<41:28:53, 5.72s/it]
41%|████ | 17822/43939 [28:46:51<41:55:33, 5.78s/it]
41%|████ | 17823/43939 [28:46:57<41:47:14, 5.76s/it]
41%|████ | 17824/43939 [28:47:03<41:41:38, 5.75s/it]
41%|████ | 17825/43939 [28:47:09<41:37:08, 5.74s/it]
41%|████ | 17826/43939 [28:47:14<41:38:10, 5.74s/it]
41%|████ | 17827/43939 [28:47:20<41:38:11, 5.74s/it]
41%|████ | 17828/43939 [28:47:26<41:29:14, 5.72s/it]
41%|████ | 17829/43939 [28:47:31<41:25:33, 5.71s/it]
41%|████ | 17830/43939 [28:47:37<41:20:09, 5.70s/it]
41%|████ | 17831/43939 [28:47:43<41:19:12, 5.70s/it]
41%|████ | 17832/43939 [28:47:49<41:24:32, 5.71s/it]
41%|████ | 17833/43939 [28:47:54<41:21:03, 5.70s/it]
41%|████ | 17834/43939 [28:48:00<41:21:02, 5.70s/it]
41%|████ | 17835/43939 [28:48:06<41:27:54, 5.72s/it]
41%|████ | 17836/43939 [28:48:11<41:23:12, 5.71s/it]
41%|████ | 17837/43939 [28:48:17<41:19:41, 5.70s/it]
41%|████ | 17838/43939 [28:48:23<41:19:28, 5.70s/it]
41%|████ | 17839/43939 [28:48:29<41:28:13, 5.72s/it]
41%|████ | 17840/43939 [28:48:34<41:20:10, 5.70s/it]
41%|████ | 17841/43939 [28:48:40<41:20:05, 5.70s/it]
41%|████ | 17842/43939 [28:48:46<41:24:56, 5.71s/it]
41%|████ | 17843/43939 [28:48:52<42:04:44, 5.80s/it]
41%|████ | 17844/43939 [28:48:57<41:44:44, 5.76s/it]
41%|████ | 17845/43939 [28:49:03<41:31:23, 5.73s/it]
41%|████ | 17846/43939 [28:49:09<41:25:46, 5.72s/it]
41%|████ | 17847/43939 [28:49:14<41:25:49, 5.72s/it]
41%|████ | 17848/43939 [28:49:20<41:23:56, 5.71s/it]
41%|████ | 17849/43939 [28:49:26<41:24:41, 5.71s/it]
41%|████ | 17850/43939 [28:49:31<41:18:54, 5.70s/it]
{'loss': 0.9309, 'grad_norm': 1.1114615201950073, 'learning_rate': 3.702580746548266e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.84, 'epoch': 1.62}
41%|████ | 17850/43939 [28:49:31<41:18:54, 5.70s/it]
41%|████ | 17851/43939 [28:49:37<41:14:10, 5.69s/it]
41%|████ | 17852/43939 [28:49:43<41:13:29, 5.69s/it]
41%|████ | 17853/43939 [28:49:49<41:16:53, 5.70s/it]
41%|████ | 17854/43939 [28:49:54<41:19:50, 5.70s/it]
41%|████ | 17855/43939 [28:50:00<41:18:23, 5.70s/it]
41%|████ | 17856/43939 [28:50:06<41:12:33, 5.69s/it]
41%|████ | 17857/43939 [28:50:11<41:13:42, 5.69s/it]
41%|████ | 17858/43939 [28:50:17<41:12:45, 5.69s/it]
41%|████ | 17859/43939 [28:50:23<41:14:57, 5.69s/it]
41%|████ | 17860/43939 [28:50:28<41:24:57, 5.72s/it]
41%|████ | 17861/43939 [28:50:34<41:30:00, 5.73s/it]
41%|████ | 17862/43939 [28:50:40<41:22:09, 5.71s/it]
41%|████ | 17863/43939 [28:50:46<41:19:53, 5.71s/it]
41%|████ | 17864/43939 [28:50:51<41:19:50, 5.71s/it]
41%|████ | 17865/43939 [28:50:57<41:57:42, 5.79s/it]
41%|████ | 17866/43939 [28:51:03<41:45:13, 5.77s/it]
41%|████ | 17867/43939 [28:51:09<41:35:21, 5.74s/it]
41%|████ | 17868/43939 [28:51:14<41:36:18, 5.75s/it]
41%|████ | 17869/43939 [28:51:20<41:30:52, 5.73s/it]
41%|████ | 17870/43939 [28:51:26<41:25:53, 5.72s/it]
41%|████ | 17871/43939 [28:51:32<41:23:46, 5.72s/it]
41%|████ | 17872/43939 [28:51:37<41:25:27, 5.72s/it]
41%|████ | 17873/43939 [28:51:43<41:23:02, 5.72s/it]
41%|████ | 17874/43939 [28:51:49<41:24:31, 5.72s/it]
41%|████ | 17875/43939 [28:51:54<41:22:40, 5.72s/it]
41%|████ | 17876/43939 [28:52:00<41:18:08, 5.70s/it]
41%|████ | 17877/43939 [28:52:06<41:20:25, 5.71s/it]
41%|████ | 17878/43939 [28:52:12<41:23:11, 5.72s/it]
41%|████ | 17879/43939 [28:52:17<41:20:14, 5.71s/it]
41%|████ | 17880/43939 [28:52:23<41:23:02, 5.72s/it]
41%|████ | 17881/43939 [28:52:29<41:20:22, 5.71s/it]
41%|████ | 17882/43939 [28:52:34<41:13:16, 5.70s/it]
41%|████ | 17883/43939 [28:52:40<41:17:42, 5.71s/it]
41%|████ | 17884/43939 [28:52:46<41:15:48, 5.70s/it]
41%|████ | 17885/43939 [28:52:51<41:12:23, 5.69s/it]
41%|████ | 17886/43939 [28:52:57<41:51:08, 5.78s/it]
41%|████ | 17887/43939 [28:53:03<41:38:23, 5.75s/it]
41%|████ | 17888/43939 [28:53:09<41:31:05, 5.74s/it]
41%|████ | 17889/43939 [28:53:14<41:28:16, 5.73s/it]
41%|████ | 17890/43939 [28:53:20<41:27:03, 5.73s/it]
41%|████ | 17891/43939 [28:53:26<41:27:13, 5.73s/it]
41%|████ | 17892/43939 [28:53:32<41:23:03, 5.72s/it]
41%|████ | 17893/43939 [28:53:37<41:20:34, 5.71s/it]
41%|████ | 17894/43939 [28:53:43<41:22:02, 5.72s/it]
41%|████ | 17895/43939 [28:53:49<41:25:54, 5.73s/it]
41%|████ | 17896/43939 [28:53:55<41:24:23, 5.72s/it]
41%|████ | 17897/43939 [28:54:00<41:17:22, 5.71s/it]
41%|████ | 17898/43939 [28:54:06<41:17:51, 5.71s/it]
41%|████ | 17899/43939 [28:54:12<41:11:00, 5.69s/it]
41%|████ | 17900/43939 [28:54:17<41:17:32, 5.71s/it]
{'loss': 0.8966, 'grad_norm': 1.0297075510025024, 'learning_rate': 3.6938654612695414e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.83, 'epoch': 1.63}
41%|████ | 17900/43939 [28:54:17<41:17:32, 5.71s/it]
41%|████ | 17901/43939 [28:54:23<41:27:56, 5.73s/it]
41%|████ | 17902/43939 [28:54:29<41:30:41, 5.74s/it]
41%|████ | 17903/43939 [28:54:35<41:25:11, 5.73s/it]
41%|████ | 17904/43939 [28:54:40<41:21:29, 5.72s/it]
41%|████ | 17905/43939 [28:54:46<41:15:21, 5.70s/it]
41%|████ | 17906/43939 [28:54:52<41:23:35, 5.72s/it]
41%|████ | 17907/43939 [28:54:58<41:54:18, 5.80s/it]
41%|████ | 17908/43939 [28:55:03<41:41:52, 5.77s/it]
41%|████ | 17909/43939 [28:55:09<41:39:05, 5.76s/it]
41%|████ | 17910/43939 [28:55:15<41:34:37, 5.75s/it]
41%|████ | 17911/43939 [28:55:20<41:22:36, 5.72s/it]
41%|████ | 17912/43939 [28:55:26<41:20:38, 5.72s/it]
41%|████ | 17913/43939 [28:55:32<41:17:35, 5.71s/it]
41%|████ | 17914/43939 [28:55:38<41:12:51, 5.70s/it]
41%|████ | 17915/43939 [28:55:43<41:11:00, 5.70s/it]
41%|████ | 17916/43939 [28:55:49<41:17:01, 5.71s/it]
41%|████ | 17917/43939 [28:55:55<41:16:38, 5.71s/it]
41%|████ | 17918/43939 [28:56:00<41:15:08, 5.71s/it]
41%|████ | 17919/43939 [28:56:06<41:21:35, 5.72s/it]
41%|████ | 17920/43939 [28:56:12<41:19:34, 5.72s/it]
41%|████ | 17921/43939 [28:56:18<41:16:21, 5.71s/it]
41%|████ | 17922/43939 [28:56:23<41:18:54, 5.72s/it]
41%|████ | 17923/43939 [28:56:29<41:12:50, 5.70s/it]
41%|████ | 17924/43939 [28:56:35<41:09:01, 5.69s/it]
41%|████ | 17925/43939 [28:56:40<41:15:09, 5.71s/it]
41%|████ | 17926/43939 [28:56:46<41:18:10, 5.72s/it]
41%|████ | 17927/43939 [28:56:52<41:15:36, 5.71s/it]
41%|████ | 17928/43939 [28:56:58<41:48:10, 5.79s/it]
41%|████ | 17929/43939 [28:57:03<41:34:09, 5.75s/it]
41%|████ | 17930/43939 [28:57:09<41:33:25, 5.75s/it]
41%|████ | 17931/43939 [28:57:15<41:27:07, 5.74s/it]
41%|████ | 17932/43939 [28:57:21<41:25:36, 5.73s/it]
41%|████ | 17933/43939 [28:57:26<41:18:59, 5.72s/it]
41%|████ | 17934/43939 [28:57:32<41:11:59, 5.70s/it]
41%|████ | 17935/43939 [28:57:38<41:06:21, 5.69s/it]
41%|████ | 17936/43939 [28:57:43<41:07:49, 5.69s/it]
41%|████ | 17937/43939 [28:57:49<41:11:29, 5.70s/it]
41%|████ | 17938/43939 [28:57:55<41:14:08, 5.71s/it]
41%|████ | 17939/43939 [28:58:00<41:11:52, 5.70s/it]
41%|████ | 17940/43939 [28:58:06<41:07:17, 5.69s/it]
41%|████ | 17941/43939 [28:58:12<41:19:10, 5.72s/it]
41%|████ | 17942/43939 [28:58:18<41:17:52, 5.72s/it]
41%|████ | 17943/43939 [28:58:23<41:14:00, 5.71s/it]
41%|████ | 17944/43939 [28:58:29<41:18:37, 5.72s/it]
41%|████ | 17945/43939 [28:58:35<41:10:22, 5.70s/it]
41%|████ | 17946/43939 [28:58:40<41:06:39, 5.69s/it]
41%|████ | 17947/43939 [28:58:46<41:02:37, 5.68s/it]
41%|████ | 17948/43939 [28:58:52<41:09:01, 5.70s/it]
41%|████ | 17949/43939 [28:58:57<41:05:15, 5.69s/it]
41%|████ | 17950/43939 [28:59:03<41:40:32, 5.77s/it]
{'loss': 0.9375, 'grad_norm': 0.9728580713272095, 'learning_rate': 3.6851313399576205e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.73, 'epoch': 1.63}
41%|████ | 17950/43939 [28:59:03<41:40:32, 5.77s/it]
41%|████ | 17951/43939 [28:59:09<41:36:19, 5.76s/it]
41%|████ | 17952/43939 [28:59:15<41:35:30, 5.76s/it]
41%|████ | 17953/43939 [28:59:21<41:25:09, 5.74s/it]
41%|████ | 17954/43939 [28:59:26<41:17:56, 5.72s/it]
41%|████ | 17955/43939 [28:59:32<41:10:14, 5.70s/it]
41%|████ | 17956/43939 [28:59:38<41:08:31, 5.70s/it]
41%|████ | 17957/43939 [28:59:43<41:05:06, 5.69s/it]
41%|████ | 17958/43939 [28:59:49<41:02:53, 5.69s/it]
41%|████ | 17959/43939 [28:59:55<41:08:45, 5.70s/it]
41%|████ | 17960/43939 [29:00:00<41:06:18, 5.70s/it]
41%|████ | 17961/43939 [29:00:06<41:04:15, 5.69s/it]
41%|████ | 17962/43939 [29:00:12<41:03:24, 5.69s/it]
41%|████ | 17963/43939 [29:00:18<41:05:08, 5.69s/it]
41%|████ | 17964/43939 [29:00:23<41:02:44, 5.69s/it]
41%|████ | 17965/43939 [29:00:29<41:05:59, 5.70s/it]
41%|████ | 17966/43939 [29:00:35<41:06:28, 5.70s/it]
41%|████ | 17967/43939 [29:00:40<41:08:19, 5.70s/it]
41%|████ | 17968/43939 [29:00:46<41:09:40, 5.71s/it]
41%|████ | 17969/43939 [29:00:52<41:07:18, 5.70s/it]
41%|████ | 17970/43939 [29:00:57<41:05:27, 5.70s/it]
41%|████ | 17971/43939 [29:01:03<41:43:04, 5.78s/it]
41%|████ | 17972/43939 [29:01:09<41:30:26, 5.75s/it]
41%|████ | 17973/43939 [29:01:15<41:33:05, 5.76s/it]
41%|████ | 17974/43939 [29:01:21<41:20:57, 5.73s/it]
41%|████ | 17975/43939 [29:01:26<41:20:56, 5.73s/it]
41%|████ | 17976/43939 [29:01:32<41:22:41, 5.74s/it]
41%|████ | 17977/43939 [29:01:38<41:16:57, 5.72s/it]
41%|████ | 17978/43939 [29:01:43<41:12:49, 5.72s/it]
41%|████ | 17979/43939 [29:01:49<41:10:19, 5.71s/it]
41%|████ | 17980/43939 [29:01:55<41:15:09, 5.72s/it]
41%|████ | 17981/43939 [29:02:01<41:14:09, 5.72s/it]
41%|████ | 17982/43939 [29:02:06<41:13:31, 5.72s/it]
41%|████ | 17983/43939 [29:02:12<41:10:21, 5.71s/it]
41%|████ | 17984/43939 [29:02:18<41:12:38, 5.72s/it]
41%|████ | 17985/43939 [29:02:23<41:09:33, 5.71s/it]
41%|████ | 17986/43939 [29:02:29<41:05:48, 5.70s/it]
41%|████ | 17987/43939 [29:02:35<41:09:37, 5.71s/it]
41%|████ | 17988/43939 [29:02:40<41:05:19, 5.70s/it]
41%|████ | 17989/43939 [29:02:46<41:08:12, 5.71s/it]
41%|████ | 17990/43939 [29:02:52<41:09:24, 5.71s/it]
41%|████ | 17991/43939 [29:02:58<41:09:55, 5.71s/it]
41%|████ | 17992/43939 [29:03:04<41:41:40, 5.78s/it]
41%|████ | 17993/43939 [29:03:09<41:28:42, 5.76s/it]
41%|████ | 17994/43939 [29:03:15<41:19:04, 5.73s/it]
41%|████ | 17995/43939 [29:03:21<41:08:47, 5.71s/it]
41%|████ | 17996/43939 [29:03:26<41:08:01, 5.71s/it]
41%|████ | 17997/43939 [29:03:32<41:14:38, 5.72s/it]
41%|████ | 17998/43939 [29:03:38<41:09:03, 5.71s/it]
41%|████ | 17999/43939 [29:03:43<41:08:35, 5.71s/it]
41%|████ | 18000/43939 [29:03:49<41:11:48, 5.72s/it]
{'loss': 0.9432, 'grad_norm': 1.536992073059082, 'learning_rate': 3.6763785204137916e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.98, 'epoch': 1.64}
41%|████ | 18000/43939 [29:03:49<41:11:48, 5.72s/it]
41%|████ | 18001/43939 [29:03:55<41:10:44, 5.72s/it]
41%|████ | 18002/43939 [29:04:01<41:07:49, 5.71s/it]
41%|████ | 18003/43939 [29:04:06<41:08:14, 5.71s/it]
41%|████ | 18004/43939 [29:04:12<41:08:18, 5.71s/it]
41%|████ | 18005/43939 [29:04:18<41:07:03, 5.71s/it]
41%|████ | 18006/43939 [29:04:23<41:04:42, 5.70s/it]
41%|████ | 18007/43939 [29:04:29<41:03:35, 5.70s/it]
41%|████ | 18008/43939 [29:04:35<40:59:44, 5.69s/it]
41%|████ | 18009/43939 [29:04:40<41:01:15, 5.70s/it]
41%|████ | 18010/43939 [29:04:46<41:04:21, 5.70s/it]
41%|████ | 18011/43939 [29:04:52<41:01:59, 5.70s/it]
41%|████ | 18012/43939 [29:04:58<41:04:56, 5.70s/it]
41%|████ | 18013/43939 [29:05:04<41:30:58, 5.76s/it]
41%|████ | 18014/43939 [29:05:09<41:25:34, 5.75s/it]
41%|████ | 18015/43939 [29:05:15<41:14:25, 5.73s/it]
41%|████ | 18016/43939 [29:05:21<41:14:12, 5.73s/it]
41%|████ | 18017/43939 [29:05:26<41:07:27, 5.71s/it]
41%|████ | 18018/43939 [29:05:32<41:05:19, 5.71s/it]
41%|████ | 18019/43939 [29:05:38<40:58:30, 5.69s/it]
41%|████ | 18020/43939 [29:05:43<40:59:35, 5.69s/it]
41%|████ | 18021/43939 [29:05:49<41:02:54, 5.70s/it]
41%|████ | 18022/43939 [29:05:55<41:04:03, 5.70s/it]
41%|████ | 18023/43939 [29:06:00<41:02:40, 5.70s/it]
41%|████ | 18024/43939 [29:06:06<41:00:12, 5.70s/it]
41%|████ | 18025/43939 [29:06:12<41:01:11, 5.70s/it]
41%|████ | 18026/43939 [29:06:18<40:56:11, 5.69s/it]
41%|████ | 18027/43939 [29:06:23<41:03:52, 5.71s/it]
41%|████ | 18028/43939 [29:06:29<41:02:52, 5.70s/it]
41%|████ | 18029/43939 [29:06:35<41:01:12, 5.70s/it]
41%|████ | 18030/43939 [29:06:40<40:57:05, 5.69s/it]
41%|████ | 18031/43939 [29:06:46<40:53:23, 5.68s/it]
41%|████ | 18032/43939 [29:06:52<41:00:42, 5.70s/it]
41%|████ | 18033/43939 [29:06:57<41:00:37, 5.70s/it]
41%|████ | 18034/43939 [29:07:03<41:00:21, 5.70s/it]
41%|████ | 18035/43939 [29:07:09<41:25:39, 5.76s/it]
41%|████ | 18036/43939 [29:07:15<41:15:33, 5.73s/it]
41%|████ | 18037/43939 [29:07:20<41:20:25, 5.75s/it]
41%|████ | 18038/43939 [29:07:26<41:18:16, 5.74s/it]
41%|████ | 18039/43939 [29:07:32<41:13:48, 5.73s/it]
41%|████ | 18040/43939 [29:07:38<41:18:09, 5.74s/it]
41%|████ | 18041/43939 [29:07:43<41:16:20, 5.74s/it]
41%|████ | 18042/43939 [29:07:49<41:09:29, 5.72s/it]
41%|████ | 18043/43939 [29:07:55<41:12:45, 5.73s/it]
41%|████ | 18044/43939 [29:08:01<41:14:07, 5.73s/it]
41%|████ | 18045/43939 [29:08:06<41:11:20, 5.73s/it]
41%|████ | 18046/43939 [29:08:12<41:08:44, 5.72s/it]
41%|████ | 18047/43939 [29:08:18<41:06:09, 5.71s/it]
41%|████ | 18048/43939 [29:08:23<41:03:43, 5.71s/it]
41%|████ | 18049/43939 [29:08:29<41:02:57, 5.71s/it]
41%|████ | 18050/43939 [29:08:35<41:03:39, 5.71s/it]
{'loss': 0.9226, 'grad_norm': 1.3161475658416748, 'learning_rate': 3.6676071407343504e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.59, 'epoch': 1.64}
41%|████ | 18050/43939 [29:08:35<41:03:39, 5.71s/it]
41%|████ | 18051/43939 [29:08:41<41:05:30, 5.71s/it]
41%|████ | 18052/43939 [29:08:46<41:02:46, 5.71s/it]
41%|████ | 18053/43939 [29:08:52<41:06:11, 5.72s/it]
41%|████ | 18054/43939 [29:08:58<41:03:33, 5.71s/it]
41%|████ | 18055/43939 [29:09:03<41:04:07, 5.71s/it]
41%|████ | 18056/43939 [29:09:09<41:32:39, 5.78s/it]
41%|████ | 18057/43939 [29:09:15<41:22:06, 5.75s/it]
41%|████ | 18058/43939 [29:09:21<41:13:13, 5.73s/it]
41%|████ | 18059/43939 [29:09:26<41:07:50, 5.72s/it]
41%|████ | 18060/43939 [29:09:32<41:05:54, 5.72s/it]
41%|████ | 18061/43939 [29:09:38<41:02:40, 5.71s/it]
41%|████ | 18062/43939 [29:09:44<41:03:16, 5.71s/it]
41%|████ | 18063/43939 [29:09:49<41:02:32, 5.71s/it]
41%|████ | 18064/43939 [29:09:55<41:02:09, 5.71s/it]
41%|████ | 18065/43939 [29:10:01<40:56:31, 5.70s/it]
41%|████ | 18066/43939 [29:10:06<41:01:03, 5.71s/it]
41%|████ | 18067/43939 [29:10:12<41:04:54, 5.72s/it]
41%|████ | 18068/43939 [29:10:18<41:03:04, 5.71s/it]
41%|████ | 18069/43939 [29:10:24<41:07:39, 5.72s/it]
41%|████ | 18070/43939 [29:10:29<41:04:30, 5.72s/it]
41%|████ | 18071/43939 [29:10:35<41:03:00, 5.71s/it]
41%|████ | 18072/43939 [29:10:41<40:55:09, 5.69s/it]
41%|████ | 18073/43939 [29:10:46<40:59:14, 5.70s/it]
41%|████ | 18074/43939 [29:10:52<40:58:09, 5.70s/it]
41%|████ | 18075/43939 [29:10:58<41:03:00, 5.71s/it]
41%|████ | 18076/43939 [29:11:03<41:04:58, 5.72s/it]
41%|████ | 18077/43939 [29:11:09<41:36:43, 5.79s/it]
41%|████ | 18078/43939 [29:11:15<41:25:08, 5.77s/it]
41%|████ | 18079/43939 [29:11:21<41:17:46, 5.75s/it]
41%|████ | 18080/43939 [29:11:27<41:07:26, 5.73s/it]
41%|████ | 18081/43939 [29:11:32<41:10:08, 5.73s/it]
41%|████ | 18082/43939 [29:11:38<41:09:20, 5.73s/it]
41%|████ | 18083/43939 [29:11:44<41:08:53, 5.73s/it]
41%|████ | 18084/43939 [29:11:49<41:04:45, 5.72s/it]
41%|████ | 18085/43939 [29:11:55<41:08:22, 5.73s/it]
41%|████ | 18086/43939 [29:12:01<41:05:52, 5.72s/it]
41%|████ | 18087/43939 [29:12:07<41:03:13, 5.72s/it]
41%|████ | 18088/43939 [29:12:12<41:05:11, 5.72s/it]
41%|████ | 18089/43939 [29:12:18<41:03:53, 5.72s/it]
41%|████ | 18090/43939 [29:12:24<41:04:14, 5.72s/it]
41%|████ | 18091/43939 [29:12:29<41:02:26, 5.72s/it]
41%|████ | 18092/43939 [29:12:35<41:01:31, 5.71s/it]
41%|████ | 18093/43939 [29:12:41<41:08:09, 5.73s/it]
41%|████ | 18094/43939 [29:12:47<40:57:42, 5.71s/it]
41%|████ | 18095/43939 [29:12:52<40:56:55, 5.70s/it]
41%|████ | 18096/43939 [29:12:58<40:56:29, 5.70s/it]
41%|████ | 18097/43939 [29:13:04<40:56:02, 5.70s/it]
41%|████ | 18098/43939 [29:13:09<40:58:46, 5.71s/it]
41%|████ | 18099/43939 [29:13:15<41:26:56, 5.77s/it]
41%|████ | 18100/43939 [29:13:21<41:12:22, 5.74s/it]
{'loss': 0.9435, 'grad_norm': 1.2327815294265747, 'learning_rate': 3.6588173393084256e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.17, 'epoch': 1.65}
41%|████ | 18100/43939 [29:13:21<41:12:22, 5.74s/it]
41%|████ | 18101/43939 [29:13:27<41:11:37, 5.74s/it]
41%|████ | 18102/43939 [29:13:32<41:10:09, 5.74s/it]
41%|████ | 18103/43939 [29:13:38<41:06:54, 5.73s/it]
41%|████ | 18104/43939 [29:13:44<41:04:52, 5.72s/it]
41%|████ | 18105/43939 [29:13:50<41:02:14, 5.72s/it]
41%|████ | 18106/43939 [29:13:55<41:02:43, 5.72s/it]
41%|████ | 18107/43939 [29:14:01<40:59:22, 5.71s/it]
41%|████ | 18108/43939 [29:14:07<41:01:16, 5.72s/it]
41%|████ | 18109/43939 [29:14:12<40:54:48, 5.70s/it]
41%|████ | 18110/43939 [29:14:18<40:58:03, 5.71s/it]
41%|████ | 18111/43939 [29:14:24<41:06:41, 5.73s/it]
41%|████ | 18112/43939 [29:14:30<41:02:23, 5.72s/it]
41%|████ | 18113/43939 [29:14:35<41:01:13, 5.72s/it]
41%|████ | 18114/43939 [29:14:41<40:56:21, 5.71s/it]
41%|████ | 18115/43939 [29:14:47<40:56:40, 5.71s/it]
41%|████ | 18116/43939 [29:14:52<40:53:54, 5.70s/it]
41%|████ | 18117/43939 [29:14:58<40:57:07, 5.71s/it]
41%|████ | 18118/43939 [29:15:04<41:00:05, 5.72s/it]
41%|████ | 18119/43939 [29:15:10<40:56:43, 5.71s/it]
41%|████ | 18120/43939 [29:15:16<41:29:07, 5.78s/it]
41%|████ | 18121/43939 [29:15:21<41:17:12, 5.76s/it]
41%|████ | 18122/43939 [29:15:27<41:09:50, 5.74s/it]
41%|████ | 18123/43939 [29:15:33<41:01:38, 5.72s/it]
41%|████ | 18124/43939 [29:15:38<41:01:53, 5.72s/it]
41%|████▏ | 18125/43939 [29:15:44<41:05:49, 5.73s/it]
41%|████▏ | 18126/43939 [29:15:50<40:58:51, 5.72s/it]
41%|████▏ | 18127/43939 [29:15:55<41:03:00, 5.73s/it]
41%|████▏ | 18128/43939 [29:16:01<41:06:06, 5.73s/it]
41%|████▏ | 18129/43939 [29:16:07<40:57:28, 5.71s/it]
41%|████▏ | 18130/43939 [29:16:13<40:53:47, 5.70s/it]
41%|████▏ | 18131/43939 [29:16:18<40:55:53, 5.71s/it]
41%|████▏ | 18132/43939 [29:16:24<40:55:44, 5.71s/it]
41%|████▏ | 18133/43939 [29:16:30<40:57:21, 5.71s/it]
41%|████▏ | 18134/43939 [29:16:35<40:59:31, 5.72s/it]
41%|████▏ | 18135/43939 [29:16:41<41:02:23, 5.73s/it]
41%|████▏ | 18136/43939 [29:16:47<41:00:44, 5.72s/it]
41%|████▏ | 18137/43939 [29:16:53<40:52:08, 5.70s/it]
41%|████▏ | 18138/43939 [29:16:58<40:53:15, 5.71s/it]
41%|████▏ | 18139/43939 [29:17:04<40:53:06, 5.70s/it]
41%|████▏ | 18140/43939 [29:17:10<40:54:03, 5.71s/it]
41%|████▏ | 18141/43939 [29:17:16<41:27:29, 5.79s/it]
41%|████▏ | 18142/43939 [29:17:21<41:15:39, 5.76s/it]
41%|████▏ | 18143/43939 [29:17:27<41:07:47, 5.74s/it]
41%|████▏ | 18144/43939 [29:17:33<41:09:02, 5.74s/it]
41%|████▏ | 18145/43939 [29:17:39<41:05:09, 5.73s/it]
41%|████▏ | 18146/43939 [29:17:44<40:53:35, 5.71s/it]
41%|████▏ | 18147/43939 [29:17:50<40:57:56, 5.72s/it]
41%|████▏ | 18148/43939 [29:17:56<40:58:07, 5.72s/it]
41%|████▏ | 18149/43939 [29:18:01<40:52:32, 5.71s/it]
41%|████▏ | 18150/43939 [29:18:07<40:50:05, 5.70s/it]
{'loss': 0.9309, 'grad_norm': 1.203054666519165, 'learning_rate': 3.6500092548157905e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.55, 'epoch': 1.65}
41%|████▏ | 18150/43939 [29:18:07<40:50:05, 5.70s/it]
41%|████▏ | 18151/43939 [29:18:13<40:52:26, 5.71s/it]
41%|████▏ | 18152/43939 [29:18:18<40:53:18, 5.71s/it]
41%|████▏ | 18153/43939 [29:18:24<40:49:19, 5.70s/it]
41%|████▏ | 18154/43939 [29:18:30<40:49:34, 5.70s/it]
41%|████▏ | 18155/43939 [29:18:36<40:49:35, 5.70s/it]
41%|████▏ | 18156/43939 [29:18:41<40:46:35, 5.69s/it]
41%|████▏ | 18157/43939 [29:18:47<40:42:16, 5.68s/it]
41%|████▏ | 18158/43939 [29:18:53<40:41:09, 5.68s/it]
41%|████▏ | 18159/43939 [29:18:58<40:43:24, 5.69s/it]
41%|████▏ | 18160/43939 [29:19:04<40:51:06, 5.70s/it]
41%|████▏ | 18161/43939 [29:19:10<40:44:17, 5.69s/it]
41%|████▏ | 18162/43939 [29:19:16<41:13:28, 5.76s/it]
41%|████▏ | 18163/43939 [29:19:21<41:09:18, 5.75s/it]
41%|████▏ | 18164/43939 [29:19:27<41:02:32, 5.73s/it]
41%|████▏ | 18165/43939 [29:19:33<40:59:37, 5.73s/it]
41%|████▏ | 18166/43939 [29:19:38<41:01:03, 5.73s/it]
41%|████▏ | 18167/43939 [29:19:44<40:59:55, 5.73s/it]
41%|████▏ | 18168/43939 [29:19:50<40:57:09, 5.72s/it]
41%|████▏ | 18169/43939 [29:19:56<40:50:15, 5.70s/it]
41%|████▏ | 18170/43939 [29:20:01<40:47:09, 5.70s/it]
41%|████▏ | 18171/43939 [29:20:07<40:50:29, 5.71s/it]
41%|████▏ | 18172/43939 [29:20:13<40:49:23, 5.70s/it]
41%|████▏ | 18173/43939 [29:20:18<40:46:15, 5.70s/it]
41%|████▏ | 18174/43939 [29:20:24<40:42:46, 5.69s/it]
41%|████▏ | 18175/43939 [29:20:30<40:39:55, 5.68s/it]
41%|████▏ | 18176/43939 [29:20:35<40:48:53, 5.70s/it]
41%|████▏ | 18177/43939 [29:20:41<40:52:04, 5.71s/it]
41%|████▏ | 18178/43939 [29:20:47<40:50:28, 5.71s/it]
41%|████▏ | 18179/43939 [29:20:53<40:51:30, 5.71s/it]
41%|████▏ | 18180/43939 [29:20:58<40:54:15, 5.72s/it]
41%|████▏ | 18181/43939 [29:21:04<40:52:02, 5.71s/it]
41%|████▏ | 18182/43939 [29:21:10<40:57:52, 5.73s/it]
41%|████▏ | 18183/43939 [29:21:15<40:52:01, 5.71s/it]
41%|████▏ | 18184/43939 [29:21:21<41:15:22, 5.77s/it]
41%|████▏ | 18185/43939 [29:21:27<41:03:39, 5.74s/it]
41%|████▏ | 18186/43939 [29:21:33<40:56:19, 5.72s/it]
41%|████▏ | 18187/43939 [29:21:38<40:56:06, 5.72s/it]
41%|████▏ | 18188/43939 [29:21:44<40:59:28, 5.73s/it]
41%|████▏ | 18189/43939 [29:21:50<40:55:59, 5.72s/it]
41%|████▏ | 18190/43939 [29:21:55<40:48:15, 5.70s/it]
41%|████▏ | 18191/43939 [29:22:01<40:55:37, 5.72s/it]
41%|████▏ | 18192/43939 [29:22:07<40:53:44, 5.72s/it]
41%|████▏ | 18193/43939 [29:22:13<40:57:43, 5.73s/it]
41%|████▏ | 18194/43939 [29:22:18<40:54:34, 5.72s/it]
41%|████▏ | 18195/43939 [29:22:24<40:53:18, 5.72s/it]
41%|████▏ | 18196/43939 [29:22:30<40:54:56, 5.72s/it]
41%|████▏ | 18197/43939 [29:22:36<40:53:13, 5.72s/it]
41%|████▏ | 18198/43939 [29:22:41<40:53:55, 5.72s/it]
41%|████▏ | 18199/43939 [29:22:47<40:54:35, 5.72s/it]
41%|████▏ | 18200/43939 [29:22:53<40:55:39, 5.72s/it]
{'loss': 0.9039, 'grad_norm': 1.238101840019226, 'learning_rate': 3.641183026224675e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.61, 'epoch': 1.66}
41%|████▏ | 18200/43939 [29:22:53<40:55:39, 5.72s/it]
41%|████▏ | 18201/43939 [29:22:58<40:58:38, 5.73s/it]
41%|████▏ | 18202/43939 [29:23:04<40:57:40, 5.73s/it]
41%|████▏ | 18203/43939 [29:23:10<40:51:56, 5.72s/it]
41%|████▏ | 18204/43939 [29:23:16<40:48:31, 5.71s/it]
41%|████▏ | 18205/43939 [29:23:22<41:16:26, 5.77s/it]
41%|████▏ | 18206/43939 [29:23:27<41:07:15, 5.75s/it]
41%|████▏ | 18207/43939 [29:23:33<41:00:29, 5.74s/it]
41%|████▏ | 18208/43939 [29:23:39<40:54:32, 5.72s/it]
41%|████▏ | 18209/43939 [29:23:44<40:50:26, 5.71s/it]
41%|████▏ | 18210/43939 [29:23:50<40:51:04, 5.72s/it]
41%|████▏ | 18211/43939 [29:23:56<40:48:59, 5.71s/it]
41%|████▏ | 18212/43939 [29:24:01<40:51:12, 5.72s/it]
41%|████▏ | 18213/43939 [29:24:07<40:49:36, 5.71s/it]
41%|████▏ | 18214/43939 [29:24:13<40:51:22, 5.72s/it]
41%|████▏ | 18215/43939 [29:24:19<40:47:53, 5.71s/it]
41%|████▏ | 18216/43939 [29:24:24<40:41:54, 5.70s/it]
41%|████▏ | 18217/43939 [29:24:30<40:49:25, 5.71s/it]
41%|████▏ | 18218/43939 [29:24:36<40:48:15, 5.71s/it]
41%|████▏ | 18219/43939 [29:24:41<40:49:52, 5.72s/it]
41%|████▏ | 18220/43939 [29:24:47<40:46:55, 5.71s/it]
41%|████▏ | 18221/43939 [29:24:53<40:51:14, 5.72s/it]
41%|████▏ | 18222/43939 [29:24:59<40:53:07, 5.72s/it]
41%|████▏ | 18223/43939 [29:25:04<40:51:23, 5.72s/it]
41%|████▏ | 18224/43939 [29:25:10<40:52:47, 5.72s/it]
41%|████▏ | 18225/43939 [29:25:16<40:50:07, 5.72s/it]
41%|████▏ | 18226/43939 [29:25:22<41:15:00, 5.78s/it]
41%|████▏ | 18227/43939 [29:25:27<41:02:37, 5.75s/it]
41%|████▏ | 18228/43939 [29:25:33<40:59:16, 5.74s/it]
41%|████▏ | 18229/43939 [29:25:39<40:50:57, 5.72s/it]
41%|████▏ | 18230/43939 [29:25:44<40:42:38, 5.70s/it]
41%|████▏ | 18231/43939 [29:25:50<40:42:59, 5.70s/it]
41%|████▏ | 18232/43939 [29:25:56<40:42:02, 5.70s/it]
41%|████▏ | 18233/43939 [29:26:02<40:48:00, 5.71s/it]
41%|████▏ | 18234/43939 [29:26:07<40:42:23, 5.70s/it]
42%|████▏ | 18235/43939 [29:26:13<40:42:15, 5.70s/it]
42%|████▏ | 18236/43939 [29:26:19<40:39:38, 5.70s/it]
42%|████▏ | 18237/43939 [29:26:24<40:34:46, 5.68s/it]
42%|████▏ | 18238/43939 [29:26:30<40:36:36, 5.69s/it]
42%|████▏ | 18239/43939 [29:26:36<40:40:47, 5.70s/it]
42%|████▏ | 18240/43939 [29:26:41<40:37:53, 5.69s/it]
42%|████▏ | 18241/43939 [29:26:47<40:31:59, 5.68s/it]
42%|████▏ | 18242/43939 [29:26:53<40:32:47, 5.68s/it]
42%|████▏ | 18243/43939 [29:26:58<40:33:14, 5.68s/it]
42%|████▏ | 18244/43939 [29:27:04<40:38:09, 5.69s/it]
42%|████▏ | 18245/43939 [29:27:10<40:34:12, 5.68s/it]
42%|████▏ | 18246/43939 [29:27:15<40:33:44, 5.68s/it]
42%|████▏ | 18247/43939 [29:27:21<41:03:47, 5.75s/it]
42%|████▏ | 18248/43939 [29:27:27<40:56:12, 5.74s/it]
42%|████▏ | 18249/43939 [29:27:33<40:50:37, 5.72s/it]
42%|████▏ | 18250/43939 [29:27:38<40:50:15, 5.72s/it]
{'loss': 0.9035, 'grad_norm': 1.1978271007537842, 'learning_rate': 3.6323387927895784e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.23, 'epoch': 1.66}
42%|████▏ | 18250/43939 [29:27:38<40:50:15, 5.72s/it]
42%|████▏ | 18251/43939 [29:27:44<40:45:12, 5.71s/it]
42%|████▏ | 18252/43939 [29:27:50<40:46:01, 5.71s/it]
42%|████▏ | 18253/43939 [29:27:56<40:44:54, 5.71s/it]
42%|████▏ | 18254/43939 [29:28:01<40:47:26, 5.72s/it]
42%|████▏ | 18255/43939 [29:28:07<40:47:37, 5.72s/it]
42%|████▏ | 18256/43939 [29:28:13<40:47:03, 5.72s/it]
42%|████▏ | 18257/43939 [29:28:18<40:47:18, 5.72s/it]
42%|████▏ | 18258/43939 [29:28:24<40:50:08, 5.72s/it]
42%|████▏ | 18259/43939 [29:28:30<40:42:57, 5.71s/it]
42%|████▏ | 18260/43939 [29:28:36<40:45:00, 5.71s/it]
42%|████▏ | 18261/43939 [29:28:41<40:45:53, 5.72s/it]
42%|████▏ | 18262/43939 [29:28:47<40:42:28, 5.71s/it]
42%|████▏ | 18263/43939 [29:28:53<40:43:32, 5.71s/it]
42%|████▏ | 18264/43939 [29:28:58<40:42:53, 5.71s/it]
42%|████▏ | 18265/43939 [29:29:04<40:45:04, 5.71s/it]
42%|████▏ | 18266/43939 [29:29:10<40:42:13, 5.71s/it]
42%|████▏ | 18267/43939 [29:29:16<40:39:20, 5.70s/it]
42%|████▏ | 18268/43939 [29:29:21<40:39:03, 5.70s/it]
42%|████▏ | 18269/43939 [29:29:27<41:04:48, 5.76s/it]
42%|████▏ | 18270/43939 [29:29:33<40:56:40, 5.74s/it]
42%|████▏ | 18271/43939 [29:29:38<40:45:38, 5.72s/it]
42%|████▏ | 18272/43939 [29:29:44<40:41:12, 5.71s/it]
42%|████▏ | 18273/43939 [29:29:50<40:44:25, 5.71s/it]
42%|████▏ | 18274/43939 [29:29:56<40:47:54, 5.72s/it]
42%|████▏ | 18275/43939 [29:30:01<40:46:53, 5.72s/it]
42%|████▏ | 18276/43939 [29:30:07<40:44:39, 5.72s/it]
42%|████▏ | 18277/43939 [29:30:13<40:42:41, 5.71s/it]
42%|████▏ | 18278/43939 [29:30:18<40:40:07, 5.71s/it]
42%|████▏ | 18279/43939 [29:30:24<40:37:37, 5.70s/it]
42%|████▏ | 18280/43939 [29:30:30<40:36:57, 5.70s/it]
42%|████▏ | 18281/43939 [29:30:36<40:41:01, 5.71s/it]
42%|████▏ | 18282/43939 [29:30:41<40:43:09, 5.71s/it]
42%|████▏ | 18283/43939 [29:30:47<40:39:08, 5.70s/it]
42%|████▏ | 18284/43939 [29:30:53<40:34:17, 5.69s/it]
42%|████▏ | 18285/43939 [29:30:58<40:34:59, 5.69s/it]
42%|████▏ | 18286/43939 [29:31:04<40:36:39, 5.70s/it]
42%|████▏ | 18287/43939 [29:31:10<40:45:46, 5.72s/it]
42%|████▏ | 18288/43939 [29:31:16<40:46:56, 5.72s/it]
42%|████▏ | 18289/43939 [29:31:21<40:47:10, 5.72s/it]
42%|████▏ | 18290/43939 [29:31:27<41:08:07, 5.77s/it]
42%|████▏ | 18291/43939 [29:31:33<41:01:38, 5.76s/it]
42%|████▏ | 18292/43939 [29:31:39<40:58:40, 5.75s/it]
42%|████▏ | 18293/43939 [29:31:44<40:49:35, 5.73s/it]
42%|████▏ | 18294/43939 [29:31:50<40:42:26, 5.71s/it]
42%|████▏ | 18295/43939 [29:31:56<40:48:20, 5.73s/it]
42%|████▏ | 18296/43939 [29:32:01<40:44:05, 5.72s/it]
42%|████▏ | 18297/43939 [29:32:07<40:40:33, 5.71s/it]
42%|████▏ | 18298/43939 [29:32:13<40:35:44, 5.70s/it]
42%|████▏ | 18299/43939 [29:32:18<40:32:13, 5.69s/it]
42%|████▏ | 18300/43939 [29:32:24<40:34:36, 5.70s/it]
{'loss': 0.9249, 'grad_norm': 1.2342469692230225, 'learning_rate': 3.6234766940490654e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.71, 'epoch': 1.67}
42%|████▏ | 18300/43939 [29:32:24<40:34:36, 5.70s/it]
42%|████▏ | 18301/43939 [29:32:30<40:36:09, 5.70s/it]
42%|████▏ | 18302/43939 [29:32:36<40:37:43, 5.71s/it]
42%|████▏ | 18303/43939 [29:32:41<40:35:15, 5.70s/it]
42%|████▏ | 18304/43939 [29:32:47<40:33:27, 5.70s/it]
42%|████▏ | 18305/43939 [29:32:53<40:40:52, 5.71s/it]
42%|████▏ | 18306/43939 [29:32:58<40:35:14, 5.70s/it]
42%|████▏ | 18307/43939 [29:33:04<40:35:32, 5.70s/it]
42%|████▏ | 18308/43939 [29:33:10<40:40:50, 5.71s/it]
42%|████▏ | 18309/43939 [29:33:16<40:43:02, 5.72s/it]
42%|████▏ | 18310/43939 [29:33:21<40:42:31, 5.72s/it]
42%|████▏ | 18311/43939 [29:33:27<41:04:40, 5.77s/it]
42%|████▏ | 18312/43939 [29:33:33<40:57:57, 5.75s/it]
42%|████▏ | 18313/43939 [29:33:39<40:49:33, 5.74s/it]
42%|████▏ | 18314/43939 [29:33:44<40:41:24, 5.72s/it]
42%|████▏ | 18315/43939 [29:33:50<40:41:05, 5.72s/it]
42%|████▏ | 18316/43939 [29:33:56<40:44:21, 5.72s/it]
42%|████▏ | 18317/43939 [29:34:01<40:45:41, 5.73s/it]
42%|████▏ | 18318/43939 [29:34:07<40:46:38, 5.73s/it]
42%|████▏ | 18319/43939 [29:34:13<40:49:20, 5.74s/it]
42%|████▏ | 18320/43939 [29:34:19<40:42:04, 5.72s/it]
42%|████▏ | 18321/43939 [29:34:24<40:38:07, 5.71s/it]
42%|████▏ | 18322/43939 [29:34:30<40:32:22, 5.70s/it]
42%|████▏ | 18323/43939 [29:34:36<40:33:27, 5.70s/it]
42%|████▏ | 18324/43939 [29:34:41<40:32:43, 5.70s/it]
42%|████▏ | 18325/43939 [29:34:47<40:32:53, 5.70s/it]
42%|████▏ | 18326/43939 [29:34:53<40:29:28, 5.69s/it]
42%|████▏ | 18327/43939 [29:34:58<40:32:15, 5.70s/it]
42%|████▏ | 18328/43939 [29:35:04<40:32:31, 5.70s/it]
42%|████▏ | 18329/43939 [29:35:10<40:31:29, 5.70s/it]
42%|████▏ | 18330/43939 [29:35:16<40:29:44, 5.69s/it]
42%|████▏ | 18331/43939 [29:35:21<40:25:12, 5.68s/it]
42%|████▏ | 18332/43939 [29:35:27<40:24:38, 5.68s/it]
42%|████▏ | 18333/43939 [29:35:33<40:52:54, 5.75s/it]
42%|████▏ | 18334/43939 [29:35:39<40:48:21, 5.74s/it]
42%|████▏ | 18335/43939 [29:35:44<40:46:50, 5.73s/it]
42%|████▏ | 18336/43939 [29:35:50<40:50:17, 5.74s/it]
42%|████▏ | 18337/43939 [29:35:56<40:46:17, 5.73s/it]
42%|████▏ | 18338/43939 [29:36:01<40:38:47, 5.72s/it]
42%|████▏ | 18339/43939 [29:36:07<40:38:23, 5.71s/it]
42%|████▏ | 18340/43939 [29:36:13<40:41:47, 5.72s/it]
42%|████▏ | 18341/43939 [29:36:19<40:46:55, 5.74s/it]
42%|████▏ | 18342/43939 [29:36:24<40:54:58, 5.75s/it]
42%|████▏ | 18343/43939 [29:36:30<40:48:29, 5.74s/it]
42%|████▏ | 18344/43939 [29:36:36<40:44:11, 5.73s/it]
42%|████▏ | 18345/43939 [29:36:42<40:39:16, 5.72s/it]
42%|████▏ | 18346/43939 [29:36:47<40:33:16, 5.70s/it]
42%|████▏ | 18347/43939 [29:36:53<40:35:46, 5.71s/it]
42%|████▏ | 18348/43939 [29:36:59<40:35:01, 5.71s/it]
42%|████▏ | 18349/43939 [29:37:04<40:32:57, 5.70s/it]
42%|████▏ | 18350/43939 [29:37:10<40:33:37, 5.71s/it]
{'loss': 0.927, 'grad_norm': 1.2014917135238647, 'learning_rate': 3.61459686982357e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.19, 'epoch': 1.67}
42%|████▏ | 18350/43939 [29:37:10<40:33:37, 5.71s/it]
42%|████▏ | 18351/43939 [29:37:16<40:35:45, 5.71s/it]
42%|████▏ | 18352/43939 [29:37:21<40:37:17, 5.72s/it]
42%|████▏ | 18353/43939 [29:37:27<40:39:11, 5.72s/it]
42%|████▏ | 18354/43939 [29:37:33<41:15:52, 5.81s/it]
42%|████▏ | 18355/43939 [29:37:39<41:05:30, 5.78s/it]
42%|████▏ | 18356/43939 [29:37:45<40:53:56, 5.76s/it]
42%|████▏ | 18357/43939 [29:37:50<40:47:52, 5.74s/it]
42%|████▏ | 18358/43939 [29:37:56<40:37:09, 5.72s/it]
42%|████▏ | 18359/43939 [29:38:02<40:26:59, 5.69s/it]
42%|████▏ | 18360/43939 [29:38:07<40:31:04, 5.70s/it]
42%|████▏ | 18361/43939 [29:38:13<40:28:20, 5.70s/it]
42%|████▏ | 18362/43939 [29:38:19<40:26:32, 5.69s/it]
42%|████▏ | 18363/43939 [29:38:24<40:31:49, 5.70s/it]
42%|████▏ | 18364/43939 [29:38:30<40:33:06, 5.71s/it]
42%|████▏ | 18365/43939 [29:38:36<40:31:04, 5.70s/it]
42%|████▏ | 18366/43939 [29:38:42<40:42:00, 5.73s/it]
42%|████▏ | 18367/43939 [29:38:47<40:33:30, 5.71s/it]
42%|████▏ | 18368/43939 [29:38:53<40:39:19, 5.72s/it]
42%|████▏ | 18369/43939 [29:38:59<40:41:46, 5.73s/it]
42%|████▏ | 18370/43939 [29:39:05<40:46:07, 5.74s/it]
42%|████▏ | 18371/43939 [29:39:10<40:38:27, 5.72s/it]
42%|████▏ | 18372/43939 [29:39:16<40:35:07, 5.71s/it]
42%|████▏ | 18373/43939 [29:39:22<40:36:06, 5.72s/it]
42%|████▏ | 18374/43939 [29:39:28<41:00:44, 5.78s/it]
42%|████▏ | 18375/43939 [29:39:33<40:46:25, 5.74s/it]
42%|████▏ | 18376/43939 [29:39:39<40:42:39, 5.73s/it]
42%|████▏ | 18377/43939 [29:39:45<40:39:39, 5.73s/it]
42%|████▏ | 18378/43939 [29:39:50<40:35:02, 5.72s/it]
42%|████▏ | 18379/43939 [29:39:56<40:33:35, 5.71s/it]
42%|████▏ | 18380/43939 [29:40:02<40:41:18, 5.73s/it]
42%|████▏ | 18381/43939 [29:40:08<40:39:50, 5.73s/it]
42%|████▏ | 18382/43939 [29:40:13<40:41:56, 5.73s/it]
42%|████▏ | 18383/43939 [29:40:19<40:40:29, 5.73s/it]
42%|████▏ | 18384/43939 [29:40:25<40:32:54, 5.71s/it]
42%|████▏ | 18385/43939 [29:40:30<40:30:26, 5.71s/it]
42%|████▏ | 18386/43939 [29:40:36<40:26:49, 5.70s/it]
42%|████▏ | 18387/43939 [29:40:42<40:29:31, 5.70s/it]
42%|████▏ | 18388/43939 [29:40:48<40:30:58, 5.71s/it]
42%|████▏ | 18389/43939 [29:40:53<40:33:17, 5.71s/it]
42%|████▏ | 18390/43939 [29:40:59<40:34:51, 5.72s/it]
42%|████▏ | 18391/43939 [29:41:05<40:32:07, 5.71s/it]
42%|████▏ | 18392/43939 [29:41:10<40:32:09, 5.71s/it]
42%|████▏ | 18393/43939 [29:41:16<40:29:56, 5.71s/it]
42%|████▏ | 18394/43939 [29:41:22<40:25:11, 5.70s/it]
42%|████▏ | 18395/43939 [29:41:28<40:57:48, 5.77s/it]
42%|████▏ | 18396/43939 [29:41:33<40:48:51, 5.75s/it]
42%|████▏ | 18397/43939 [29:41:39<40:48:50, 5.75s/it]
42%|████▏ | 18398/43939 [29:41:45<40:40:03, 5.73s/it]
42%|████▏ | 18399/43939 [29:41:51<40:34:35, 5.72s/it]
42%|████▏ | 18400/43939 [29:41:56<40:34:13, 5.72s/it]
{'loss': 0.9165, 'grad_norm': 1.1069117784500122, 'learning_rate': 3.6056994602131864e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.47, 'epoch': 1.68}
42%|████▏ | 18400/43939 [29:41:56<40:34:13, 5.72s/it]
42%|████▏ | 18401/43939 [29:42:02<40:33:55, 5.72s/it]
42%|████▏ | 18402/43939 [29:42:08<40:28:02, 5.70s/it]
42%|████▏ | 18403/43939 [29:42:13<40:26:40, 5.70s/it]
42%|████▏ | 18404/43939 [29:42:19<40:24:50, 5.70s/it]
42%|████▏ | 18405/43939 [29:42:25<40:29:47, 5.71s/it]
42%|████▏ | 18406/43939 [29:42:30<40:25:38, 5.70s/it]
42%|████▏ | 18407/43939 [29:42:36<40:24:44, 5.70s/it]
42%|████▏ | 18408/43939 [29:42:42<40:22:49, 5.69s/it]
42%|████▏ | 18409/43939 [29:42:47<40:20:58, 5.69s/it]
42%|████▏ | 18410/43939 [29:42:53<40:24:30, 5.70s/it]
42%|████▏ | 18411/43939 [29:42:59<40:29:09, 5.71s/it]
42%|████▏ | 18412/43939 [29:43:05<40:30:48, 5.71s/it]
42%|████▏ | 18413/43939 [29:43:10<40:29:54, 5.71s/it]
42%|████▏ | 18414/43939 [29:43:16<40:22:58, 5.70s/it]
42%|████▏ | 18415/43939 [29:43:22<40:26:45, 5.70s/it]
42%|████▏ | 18416/43939 [29:43:27<40:21:57, 5.69s/it]
42%|████▏ | 18417/43939 [29:43:33<40:51:41, 5.76s/it]
42%|████▏ | 18418/43939 [29:43:39<40:43:08, 5.74s/it]
42%|████▏ | 18419/43939 [29:43:45<40:41:36, 5.74s/it]
42%|████▏ | 18420/43939 [29:43:51<40:40:49, 5.74s/it]
42%|████▏ | 18421/43939 [29:43:56<40:36:35, 5.73s/it]
42%|████▏ | 18422/43939 [29:44:02<40:35:50, 5.73s/it]
42%|████▏ | 18423/43939 [29:44:08<40:31:34, 5.72s/it]
42%|████▏ | 18424/43939 [29:44:13<40:27:42, 5.71s/it]
42%|████▏ | 18425/43939 [29:44:19<40:21:57, 5.70s/it]
42%|████▏ | 18426/43939 [29:44:25<40:25:47, 5.70s/it]
42%|████▏ | 18427/43939 [29:44:30<40:23:22, 5.70s/it]
42%|████▏ | 18428/43939 [29:44:36<40:34:50, 5.73s/it]
42%|████▏ | 18429/43939 [29:44:42<40:28:36, 5.71s/it]
42%|████▏ | 18430/43939 [29:44:48<40:30:28, 5.72s/it]
42%|████▏ | 18431/43939 [29:44:53<40:28:18, 5.71s/it]
42%|████▏ | 18432/43939 [29:44:59<40:28:49, 5.71s/it]
42%|████▏ | 18433/43939 [29:45:05<40:21:07, 5.70s/it]
42%|████▏ | 18434/43939 [29:45:10<40:21:08, 5.70s/it]
42%|████▏ | 18435/43939 [29:45:16<40:24:24, 5.70s/it]
42%|████▏ | 18436/43939 [29:45:22<40:22:45, 5.70s/it]
42%|████▏ | 18437/43939 [29:45:27<40:21:13, 5.70s/it]
42%|████▏ | 18438/43939 [29:45:33<40:50:02, 5.76s/it]
42%|████▏ | 18439/43939 [29:45:39<40:42:06, 5.75s/it]
42%|████▏ | 18440/43939 [29:45:45<40:37:39, 5.74s/it]
42%|████▏ | 18441/43939 [29:45:51<40:35:09, 5.73s/it]
42%|████▏ | 18442/43939 [29:45:56<40:31:27, 5.72s/it]
42%|████▏ | 18443/43939 [29:46:02<40:26:41, 5.71s/it]
42%|████▏ | 18444/43939 [29:46:08<40:27:27, 5.71s/it]
42%|████▏ | 18445/43939 [29:46:13<40:24:30, 5.71s/it]
42%|████▏ | 18446/43939 [29:46:19<40:23:08, 5.70s/it]
42%|████▏ | 18447/43939 [29:46:25<40:18:23, 5.69s/it]
42%|████▏ | 18448/43939 [29:46:30<40:16:50, 5.69s/it]
42%|████▏ | 18449/43939 [29:46:36<40:15:19, 5.69s/it]
42%|████▏ | 18450/43939 [29:46:42<40:16:39, 5.69s/it]
{'loss': 0.9243, 'grad_norm': 1.2550182342529297, 'learning_rate': 3.5967846055954594e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.09, 'epoch': 1.68}
42%|████▏ | 18450/43939 [29:46:42<40:16:39, 5.69s/it]
42%|████▏ | 18451/43939 [29:46:47<40:17:39, 5.69s/it]
42%|████▏ | 18452/43939 [29:46:53<40:21:21, 5.70s/it]
42%|████▏ | 18453/43939 [29:46:59<40:17:13, 5.69s/it]
42%|████▏ | 18454/43939 [29:47:04<40:14:50, 5.69s/it]
42%|████▏ | 18455/43939 [29:47:10<40:16:01, 5.69s/it]
42%|████▏ | 18456/43939 [29:47:16<40:13:20, 5.68s/it]
42%|████▏ | 18457/43939 [29:47:22<40:17:16, 5.69s/it]
42%|████▏ | 18458/43939 [29:47:27<40:17:15, 5.69s/it]
42%|████▏ | 18459/43939 [29:47:33<40:50:44, 5.77s/it]
42%|████▏ | 18460/43939 [29:47:39<40:41:06, 5.75s/it]
42%|████▏ | 18461/43939 [29:47:45<40:31:26, 5.73s/it]
42%|████▏ | 18462/43939 [29:47:50<40:29:06, 5.72s/it]
42%|████▏ | 18463/43939 [29:47:56<40:28:21, 5.72s/it]
42%|████▏ | 18464/43939 [29:48:02<40:28:01, 5.72s/it]
42%|████▏ | 18465/43939 [29:48:07<40:26:45, 5.72s/it]
42%|████▏ | 18466/43939 [29:48:13<40:24:33, 5.71s/it]
42%|████▏ | 18467/43939 [29:48:19<40:18:26, 5.70s/it]
42%|████▏ | 18468/43939 [29:48:25<40:21:58, 5.71s/it]
42%|████▏ | 18469/43939 [29:48:30<40:18:59, 5.70s/it]
42%|████▏ | 18470/43939 [29:48:36<40:20:16, 5.70s/it]
42%|████▏ | 18471/43939 [29:48:42<40:26:17, 5.72s/it]
42%|████▏ | 18472/43939 [29:48:47<40:26:32, 5.72s/it]
42%|████▏ | 18473/43939 [29:48:53<40:19:16, 5.70s/it]
42%|████▏ | 18474/43939 [29:48:59<40:15:13, 5.69s/it]
42%|████▏ | 18475/43939 [29:49:04<40:13:26, 5.69s/it]
42%|████▏ | 18476/43939 [29:49:10<40:19:55, 5.70s/it]
42%|████▏ | 18477/43939 [29:49:16<40:19:10, 5.70s/it]
42%|████▏ | 18478/43939 [29:49:22<40:18:35, 5.70s/it]
42%|████▏ | 18479/43939 [29:49:27<40:14:15, 5.69s/it]
42%|████▏ | 18480/43939 [29:49:33<40:46:12, 5.77s/it]
42%|████▏ | 18481/43939 [29:49:39<40:37:10, 5.74s/it]
42%|████▏ | 18482/43939 [29:49:45<40:30:29, 5.73s/it]
42%|████▏ | 18483/43939 [29:49:50<40:25:39, 5.72s/it]
42%|████▏ | 18484/43939 [29:49:56<40:28:41, 5.72s/it]
42%|████▏ | 18485/43939 [29:50:02<40:32:55, 5.73s/it]
42%|████▏ | 18486/43939 [29:50:07<40:26:12, 5.72s/it]
42%|████▏ | 18487/43939 [29:50:13<40:26:49, 5.72s/it]
42%|████▏ | 18488/43939 [29:50:19<40:22:44, 5.71s/it]
42%|████▏ | 18489/43939 [29:50:25<40:19:33, 5.70s/it]
42%|████▏ | 18490/43939 [29:50:30<40:17:39, 5.70s/it]
42%|████▏ | 18491/43939 [29:50:36<40:22:57, 5.71s/it]
42%|████▏ | 18492/43939 [29:50:42<40:20:36, 5.71s/it]
42%|████▏ | 18493/43939 [29:50:47<40:19:59, 5.71s/it]
42%|████▏ | 18494/43939 [29:50:53<40:17:57, 5.70s/it]
42%|████▏ | 18495/43939 [29:50:59<40:24:28, 5.72s/it]
42%|████▏ | 18496/43939 [29:51:05<40:25:32, 5.72s/it]
42%|████▏ | 18497/43939 [29:51:10<40:26:38, 5.72s/it]
42%|████▏ | 18498/43939 [29:51:16<40:25:18, 5.72s/it]
42%|████▏ | 18499/43939 [29:51:22<40:19:32, 5.71s/it]
42%|████▏ | 18500/43939 [29:51:27<40:21:30, 5.71s/it]
{'loss': 0.9316, 'grad_norm': 1.1848024129867554, 'learning_rate': 3.587852446623171e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.94, 'epoch': 1.68}
42%|████▏ | 18500/43939 [29:51:27<40:21:30, 5.71s/it]
42%|████▏ | 18501/43939 [29:51:33<40:53:20, 5.79s/it]
42%|████▏ | 18502/43939 [29:51:39<40:40:37, 5.76s/it]
42%|████▏ | 18503/43939 [29:51:45<40:32:05, 5.74s/it]
42%|████▏ | 18504/43939 [29:51:50<40:26:52, 5.72s/it]
42%|████▏ | 18505/43939 [29:51:56<40:22:37, 5.72s/it]
42%|████▏ | 18506/43939 [29:52:02<40:17:27, 5.70s/it]
42%|████▏ | 18507/43939 [29:52:07<40:16:43, 5.70s/it]
42%|████▏ | 18508/43939 [29:52:13<40:19:34, 5.71s/it]
42%|████▏ | 18509/43939 [29:52:19<40:12:04, 5.69s/it]
42%|████▏ | 18510/43939 [29:52:25<40:12:06, 5.69s/it]
42%|████▏ | 18511/43939 [29:52:30<40:11:21, 5.69s/it]
42%|████▏ | 18512/43939 [29:52:36<40:12:10, 5.69s/it]
42%|████▏ | 18513/43939 [29:52:42<40:13:38, 5.70s/it]
42%|████▏ | 18514/43939 [29:52:47<40:12:19, 5.69s/it]
42%|████▏ | 18515/43939 [29:52:53<40:14:18, 5.70s/it]
42%|████▏ | 18516/43939 [29:52:59<40:17:53, 5.71s/it]
42%|████▏ | 18517/43939 [29:53:04<40:14:00, 5.70s/it]
42%|████▏ | 18518/43939 [29:53:10<40:12:54, 5.70s/it]
42%|████▏ | 18519/43939 [29:53:16<40:19:44, 5.71s/it]
42%|████▏ | 18520/43939 [29:53:22<40:15:03, 5.70s/it]
42%|████▏ | 18521/43939 [29:53:27<40:14:09, 5.70s/it]
42%|████▏ | 18522/43939 [29:53:33<40:42:34, 5.77s/it]
42%|████▏ | 18523/43939 [29:53:39<40:37:57, 5.76s/it]
42%|████▏ | 18524/43939 [29:53:45<40:35:10, 5.75s/it]
42%|████▏ | 18525/43939 [29:53:50<40:29:13, 5.74s/it]
42%|████▏ | 18526/43939 [29:53:56<40:34:08, 5.75s/it]
42%|████▏ | 18527/43939 [29:54:02<40:35:24, 5.75s/it]
42%|████▏ | 18528/43939 [29:54:08<40:32:35, 5.74s/it]
42%|████▏ | 18529/43939 [29:54:13<40:30:21, 5.74s/it]
42%|████▏ | 18530/43939 [29:54:19<40:25:05, 5.73s/it]
42%|████▏ | 18531/43939 [29:54:25<40:22:58, 5.72s/it]
42%|████▏ | 18532/43939 [29:54:30<40:22:57, 5.72s/it]
42%|████▏ | 18533/43939 [29:54:36<40:22:05, 5.72s/it]
42%|████▏ | 18534/43939 [29:54:42<40:26:40, 5.73s/it]
42%|████▏ | 18535/43939 [29:54:48<40:21:31, 5.72s/it]
42%|████▏ | 18536/43939 [29:54:53<40:13:53, 5.70s/it]
42%|████▏ | 18537/43939 [29:54:59<40:06:42, 5.68s/it]
42%|████▏ | 18538/43939 [29:55:05<40:09:03, 5.69s/it]
42%|████▏ | 18539/43939 [29:55:10<40:10:05, 5.69s/it]
42%|████▏ | 18540/43939 [29:55:16<40:08:24, 5.69s/it]
42%|████▏ | 18541/43939 [29:55:22<40:08:23, 5.69s/it]
42%|████▏ | 18542/43939 [29:55:27<40:09:53, 5.69s/it]
42%|████▏ | 18543/43939 [29:55:33<40:37:47, 5.76s/it]
42%|████▏ | 18544/43939 [29:55:39<40:28:11, 5.74s/it]
42%|████▏ | 18545/43939 [29:55:45<40:23:28, 5.73s/it]
42%|████▏ | 18546/43939 [29:55:50<40:21:32, 5.72s/it]
42%|████▏ | 18547/43939 [29:55:56<40:20:58, 5.72s/it]
42%|████▏ | 18548/43939 [29:56:02<40:22:23, 5.72s/it]
42%|████▏ | 18549/43939 [29:56:08<40:19:10, 5.72s/it]
42%|████▏ | 18550/43939 [29:56:13<40:20:40, 5.72s/it]
{'loss': 0.9118, 'grad_norm': 1.1091415882110596, 'learning_rate': 3.578903124222119e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 466.38, 'epoch': 1.69}
42%|████▏ | 18550/43939 [29:56:13<40:20:40, 5.72s/it]
42%|████▏ | 18551/43939 [29:56:19<40:17:53, 5.71s/it]
42%|████▏ | 18552/43939 [29:56:25<40:13:52, 5.70s/it]
42%|████▏ | 18553/43939 [29:56:30<40:12:20, 5.70s/it]
42%|████▏ | 18554/43939 [29:56:36<40:10:32, 5.70s/it]
42%|████▏ | 18555/43939 [29:56:42<40:10:24, 5.70s/it]
42%|████▏ | 18556/43939 [29:56:47<40:10:10, 5.70s/it]
42%|████▏ | 18557/43939 [29:56:53<40:08:30, 5.69s/it]
42%|████▏ | 18558/43939 [29:56:59<40:17:28, 5.71s/it]
42%|████▏ | 18559/43939 [29:57:05<40:14:26, 5.71s/it]
42%|████▏ | 18560/43939 [29:57:10<40:13:12, 5.71s/it]
42%|████▏ | 18561/43939 [29:57:16<40:06:15, 5.69s/it]
42%|████▏ | 18562/43939 [29:57:22<40:08:44, 5.70s/it]
42%|████▏ | 18563/43939 [29:57:27<40:11:55, 5.70s/it]
42%|████▏ | 18564/43939 [29:57:33<40:40:40, 5.77s/it]
42%|████▏ | 18565/43939 [29:57:39<40:29:05, 5.74s/it]
42%|████▏ | 18566/43939 [29:57:45<40:17:05, 5.72s/it]
42%|████▏ | 18567/43939 [29:57:50<40:20:57, 5.73s/it]
42%|████▏ | 18568/43939 [29:57:56<40:16:18, 5.71s/it]
42%|████▏ | 18569/43939 [29:58:02<40:12:49, 5.71s/it]
42%|████▏ | 18570/43939 [29:58:07<40:14:36, 5.71s/it]
42%|████▏ | 18571/43939 [29:58:13<40:15:34, 5.71s/it]
42%|████▏ | 18572/43939 [29:58:19<40:12:56, 5.71s/it]
42%|████▏ | 18573/43939 [29:58:25<40:18:59, 5.72s/it]
42%|████▏ | 18574/43939 [29:58:30<40:15:35, 5.71s/it]
42%|████▏ | 18575/43939 [29:58:36<40:15:13, 5.71s/it]
42%|████▏ | 18576/43939 [29:58:42<40:17:41, 5.72s/it]
42%|████▏ | 18577/43939 [29:58:47<40:15:15, 5.71s/it]
42%|████▏ | 18578/43939 [29:58:53<40:13:10, 5.71s/it]
42%|████▏ | 18579/43939 [29:58:59<40:10:45, 5.70s/it]
42%|████▏ | 18580/43939 [29:59:05<40:06:33, 5.69s/it]
42%|████▏ | 18581/43939 [29:59:10<40:05:45, 5.69s/it]
42%|████▏ | 18582/43939 [29:59:16<40:01:23, 5.68s/it]
42%|████▏ | 18583/43939 [29:59:22<40:04:10, 5.69s/it]
42%|████▏ | 18584/43939 [29:59:27<40:34:51, 5.76s/it]
42%|████▏ | 18585/43939 [29:59:33<40:26:50, 5.74s/it]
42%|████▏ | 18586/43939 [29:59:39<40:23:36, 5.74s/it]
42%|████▏ | 18587/43939 [29:59:45<40:19:05, 5.73s/it]
42%|████▏ | 18588/43939 [29:59:50<40:16:35, 5.72s/it]
42%|████▏ | 18589/43939 [29:59:56<40:11:46, 5.71s/it]
42%|████▏ | 18590/43939 [30:00:02<40:07:39, 5.70s/it]
42%|████▏ | 18591/43939 [30:00:07<40:09:30, 5.70s/it]
42%|████▏ | 18592/43939 [30:00:13<40:06:30, 5.70s/it]
42%|████▏ | 18593/43939 [30:00:19<40:06:49, 5.70s/it]
42%|████▏ | 18594/43939 [30:00:25<40:10:53, 5.71s/it]
42%|████▏ | 18595/43939 [30:00:30<40:12:02, 5.71s/it]
42%|████▏ | 18596/43939 [30:00:36<40:14:36, 5.72s/it]
42%|████▏ | 18597/43939 [30:00:42<40:16:47, 5.72s/it]
42%|████▏ | 18598/43939 [30:00:47<40:15:59, 5.72s/it]
42%|████▏ | 18599/43939 [30:00:53<40:16:46, 5.72s/it]
42%|████▏ | 18600/43939 [30:00:59<40:12:54, 5.71s/it]
{'loss': 0.9016, 'grad_norm': 1.416194200515747, 'learning_rate': 3.569936779588896e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.76, 'epoch': 1.69}
42%|████▏ | 18600/43939 [30:00:59<40:12:54, 5.71s/it]
42%|████▏ | 18601/43939 [30:01:05<40:12:23, 5.71s/it]
42%|████▏ | 18602/43939 [30:01:10<40:09:58, 5.71s/it]
42%|████▏ | 18603/43939 [30:01:16<40:10:02, 5.71s/it]
42%|████▏ | 18604/43939 [30:01:22<40:09:54, 5.71s/it]
42%|████▏ | 18605/43939 [30:01:28<40:40:59, 5.78s/it]
42%|████▏ | 18606/43939 [30:01:33<40:28:06, 5.75s/it]
42%|████▏ | 18607/43939 [30:01:39<40:22:18, 5.74s/it]
42%|████▏ | 18608/43939 [30:01:45<40:17:13, 5.73s/it]
42%|████▏ | 18609/43939 [30:01:50<40:14:27, 5.72s/it]
42%|████▏ | 18610/43939 [30:01:56<40:17:22, 5.73s/it]
42%|████▏ | 18611/43939 [30:02:02<40:18:50, 5.73s/it]
42%|████▏ | 18612/43939 [30:02:08<40:19:10, 5.73s/it]
42%|████▏ | 18613/43939 [30:02:13<40:19:36, 5.73s/it]
42%|████▏ | 18614/43939 [30:02:19<40:12:32, 5.72s/it]
42%|████▏ | 18615/43939 [30:02:25<40:05:21, 5.70s/it]
42%|████▏ | 18616/43939 [30:02:30<40:04:38, 5.70s/it]
42%|████▏ | 18617/43939 [30:02:36<40:10:23, 5.71s/it]
42%|████▏ | 18618/43939 [30:02:42<40:02:10, 5.69s/it]
42%|████▏ | 18619/43939 [30:02:47<40:00:03, 5.69s/it]
42%|████▏ | 18620/43939 [30:02:53<40:00:28, 5.69s/it]
42%|████▏ | 18621/43939 [30:02:59<40:01:16, 5.69s/it]
42%|████▏ | 18622/43939 [30:03:05<40:07:54, 5.71s/it]
42%|████▏ | 18623/43939 [30:03:10<40:06:48, 5.70s/it]
42%|████▏ | 18624/43939 [30:03:16<40:08:34, 5.71s/it]
42%|████▏ | 18625/43939 [30:03:22<40:11:53, 5.72s/it]
42%|████▏ | 18626/43939 [30:03:27<40:09:56, 5.71s/it]
42%|████▏ | 18627/43939 [30:03:33<40:33:48, 5.77s/it]
42%|████▏ | 18628/43939 [30:03:39<40:24:11, 5.75s/it]
42%|████▏ | 18629/43939 [30:03:45<40:19:30, 5.74s/it]
42%|████▏ | 18630/43939 [30:03:50<40:18:18, 5.73s/it]
42%|████▏ | 18631/43939 [30:03:56<40:11:02, 5.72s/it]
42%|████▏ | 18632/43939 [30:04:02<40:10:27, 5.71s/it]
42%|████▏ | 18633/43939 [30:04:08<40:03:43, 5.70s/it]
42%|████▏ | 18634/43939 [30:04:13<39:59:23, 5.69s/it]
42%|████▏ | 18635/43939 [30:04:19<39:59:02, 5.69s/it]
42%|████▏ | 18636/43939 [30:04:25<40:04:50, 5.70s/it]
42%|████▏ | 18637/43939 [30:04:30<40:04:09, 5.70s/it]
42%|████▏ | 18638/43939 [30:04:36<40:01:07, 5.69s/it]
42%|████▏ | 18639/43939 [30:04:42<40:00:59, 5.69s/it]
42%|████▏ | 18640/43939 [30:04:47<40:04:37, 5.70s/it]
42%|████▏ | 18641/43939 [30:04:53<40:00:55, 5.69s/it]
42%|████▏ | 18642/43939 [30:04:59<40:02:49, 5.70s/it]
42%|████▏ | 18643/43939 [30:05:04<40:05:25, 5.71s/it]
42%|████▏ | 18644/43939 [30:05:10<40:02:49, 5.70s/it]
42%|████▏ | 18645/43939 [30:05:16<40:03:37, 5.70s/it]
42%|████▏ | 18646/43939 [30:05:22<40:02:32, 5.70s/it]
42%|████▏ | 18647/43939 [30:05:27<40:27:03, 5.76s/it]
42%|████▏ | 18648/43939 [30:05:33<40:20:12, 5.74s/it]
42%|████▏ | 18649/43939 [30:05:39<40:13:06, 5.73s/it]
42%|████▏ | 18650/43939 [30:05:45<40:08:25, 5.71s/it]
{'loss': 0.8922, 'grad_norm': 1.7901129722595215, 'learning_rate': 3.5609535541886564e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.89, 'epoch': 1.7}
42%|████▏ | 18650/43939 [30:05:45<40:08:25, 5.71s/it]
42%|████▏ | 18651/43939 [30:05:50<40:05:11, 5.71s/it]
42%|████▏ | 18652/43939 [30:05:56<40:03:03, 5.70s/it]
42%|████▏ | 18653/43939 [30:06:02<40:03:42, 5.70s/it]
42%|████▏ | 18654/43939 [30:06:07<40:01:09, 5.70s/it]
42%|████▏ | 18655/43939 [30:06:13<40:04:53, 5.71s/it]
42%|████▏ | 18656/43939 [30:06:19<40:07:48, 5.71s/it]
42%|████▏ | 18657/43939 [30:06:24<40:06:18, 5.71s/it]
42%|████▏ | 18658/43939 [30:06:30<40:00:48, 5.70s/it]
42%|████▏ | 18659/43939 [30:06:36<39:56:57, 5.69s/it]
42%|████▏ | 18660/43939 [30:06:42<39:58:49, 5.69s/it]
42%|████▏ | 18661/43939 [30:06:47<39:53:25, 5.68s/it]
42%|████▏ | 18662/43939 [30:06:53<39:59:11, 5.69s/it]
42%|████▏ | 18663/43939 [30:06:59<40:04:04, 5.71s/it]
42%|████▏ | 18664/43939 [30:07:04<40:03:56, 5.71s/it]
42%|████▏ | 18665/43939 [30:07:10<40:02:45, 5.70s/it]
42%|████▏ | 18666/43939 [30:07:16<40:01:40, 5.70s/it]
42%|████▏ | 18667/43939 [30:07:21<39:58:01, 5.69s/it]
42%|████▏ | 18668/43939 [30:07:27<40:24:51, 5.76s/it]
42%|████▏ | 18669/43939 [30:07:33<40:21:49, 5.75s/it]
42%|████▏ | 18670/43939 [30:07:39<40:19:50, 5.75s/it]
42%|████▏ | 18671/43939 [30:07:45<40:20:19, 5.75s/it]
42%|████▏ | 18672/43939 [30:07:50<40:09:43, 5.72s/it]
42%|████▏ | 18673/43939 [30:07:56<40:09:47, 5.72s/it]
42%|████▏ | 18674/43939 [30:08:02<40:03:24, 5.71s/it]
43%|████▎ | 18675/43939 [30:08:07<39:58:37, 5.70s/it]
43%|████▎ | 18676/43939 [30:08:13<39:56:39, 5.69s/it]
43%|████▎ | 18677/43939 [30:08:19<39:57:25, 5.69s/it]
43%|████▎ | 18678/43939 [30:08:24<39:55:40, 5.69s/it]
43%|████▎ | 18679/43939 [30:08:30<39:56:04, 5.69s/it]
43%|████▎ | 18680/43939 [30:08:36<39:54:01, 5.69s/it]
43%|████▎ | 18681/43939 [30:08:41<39:58:13, 5.70s/it]
43%|████▎ | 18682/43939 [30:08:47<39:55:27, 5.69s/it]
43%|████▎ | 18683/43939 [30:08:53<39:56:53, 5.69s/it]
43%|████▎ | 18684/43939 [30:08:59<39:59:25, 5.70s/it]
43%|████▎ | 18685/43939 [30:09:04<39:59:43, 5.70s/it]
43%|████▎ | 18686/43939 [30:09:10<39:57:05, 5.70s/it]
43%|████▎ | 18687/43939 [30:09:16<39:58:12, 5.70s/it]
43%|████▎ | 18688/43939 [30:09:21<39:58:06, 5.70s/it]
43%|████▎ | 18689/43939 [30:09:27<40:06:37, 5.72s/it]
43%|████▎ | 18690/43939 [30:09:33<40:27:37, 5.77s/it]
43%|████▎ | 18691/43939 [30:09:39<40:20:08, 5.75s/it]
43%|████▎ | 18692/43939 [30:09:44<40:11:27, 5.73s/it]
43%|████▎ | 18693/43939 [30:09:50<40:05:53, 5.72s/it]
43%|████▎ | 18694/43939 [30:09:56<40:02:52, 5.71s/it]
43%|████▎ | 18695/43939 [30:10:01<40:01:50, 5.71s/it]
43%|████▎ | 18696/43939 [30:10:07<39:56:40, 5.70s/it]
43%|████▎ | 18697/43939 [30:10:13<39:53:44, 5.69s/it]
43%|████▎ | 18698/43939 [30:10:18<39:53:15, 5.69s/it]
43%|████▎ | 18699/43939 [30:10:24<39:59:00, 5.70s/it]
43%|████▎ | 18700/43939 [30:10:30<40:08:34, 5.73s/it]
{'loss': 0.9334, 'grad_norm': 1.1177266836166382, 'learning_rate': 3.551953589752893e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.15, 'epoch': 1.7}
43%|████▎ | 18700/43939 [30:10:30<40:08:34, 5.73s/it]
43%|████▎ | 18701/43939 [30:10:36<40:06:48, 5.72s/it]
43%|████▎ | 18702/43939 [30:10:41<40:07:35, 5.72s/it]
43%|████▎ | 18703/43939 [30:10:47<40:07:28, 5.72s/it]
43%|████▎ | 18704/43939 [30:10:53<40:07:01, 5.72s/it]
43%|████▎ | 18705/43939 [30:10:59<40:02:19, 5.71s/it]
43%|████▎ | 18706/43939 [30:11:04<39:58:08, 5.70s/it]
43%|████▎ | 18707/43939 [30:11:10<39:54:28, 5.69s/it]
43%|████▎ | 18708/43939 [30:11:16<39:54:05, 5.69s/it]
43%|████▎ | 18709/43939 [30:11:21<40:01:01, 5.71s/it]
43%|████▎ | 18710/43939 [30:11:27<39:57:41, 5.70s/it]
43%|████▎ | 18711/43939 [30:11:33<39:56:34, 5.70s/it]
43%|████▎ | 18712/43939 [30:11:39<40:30:00, 5.78s/it]
43%|████▎ | 18713/43939 [30:11:44<40:19:53, 5.76s/it]
43%|████▎ | 18714/43939 [30:11:50<40:14:44, 5.74s/it]
43%|████▎ | 18715/43939 [30:11:56<40:11:55, 5.74s/it]
43%|████▎ | 18716/43939 [30:12:02<40:10:29, 5.73s/it]
43%|████▎ | 18717/43939 [30:12:07<40:07:37, 5.73s/it]
43%|████▎ | 18718/43939 [30:12:13<40:05:50, 5.72s/it]
43%|████▎ | 18719/43939 [30:12:19<40:08:00, 5.73s/it]
43%|████▎ | 18720/43939 [30:12:24<40:06:30, 5.73s/it]
43%|████▎ | 18721/43939 [30:12:30<40:02:37, 5.72s/it]
43%|████▎ | 18722/43939 [30:12:36<39:59:14, 5.71s/it]
43%|████▎ | 18723/43939 [30:12:42<39:58:07, 5.71s/it]
43%|████▎ | 18724/43939 [30:12:47<40:07:01, 5.73s/it]
43%|████▎ | 18725/43939 [30:12:53<39:59:58, 5.71s/it]
43%|████▎ | 18726/43939 [30:12:59<39:58:04, 5.71s/it]
43%|████▎ | 18727/43939 [30:13:04<39:59:45, 5.71s/it]
43%|████▎ | 18728/43939 [30:13:10<40:02:54, 5.72s/it]
43%|████▎ | 18729/43939 [30:13:16<39:58:36, 5.71s/it]
43%|████▎ | 18730/43939 [30:13:21<39:54:22, 5.70s/it]
43%|████▎ | 18731/43939 [30:13:27<39:54:14, 5.70s/it]
43%|████▎ | 18732/43939 [30:13:33<40:18:37, 5.76s/it]
43%|████▎ | 18733/43939 [30:13:39<40:10:50, 5.74s/it]
43%|████▎ | 18734/43939 [30:13:44<40:03:49, 5.72s/it]
43%|████▎ | 18735/43939 [30:13:50<40:05:30, 5.73s/it]
43%|████▎ | 18736/43939 [30:13:56<40:06:58, 5.73s/it]
43%|████▎ | 18737/43939 [30:14:02<40:02:20, 5.72s/it]
43%|████▎ | 18738/43939 [30:14:07<40:02:50, 5.72s/it]
43%|████▎ | 18739/43939 [30:14:13<40:00:35, 5.72s/it]
43%|████▎ | 18740/43939 [30:14:19<39:59:03, 5.71s/it]
43%|████▎ | 18741/43939 [30:14:24<40:01:10, 5.72s/it]
43%|████▎ | 18742/43939 [30:14:30<40:00:58, 5.72s/it]
43%|████▎ | 18743/43939 [30:14:36<39:58:09, 5.71s/it]
43%|████▎ | 18744/43939 [30:14:42<39:52:11, 5.70s/it]
43%|████▎ | 18745/43939 [30:14:47<39:58:54, 5.71s/it]
43%|████▎ | 18746/43939 [30:14:53<39:53:14, 5.70s/it]
43%|████▎ | 18747/43939 [30:14:59<39:51:03, 5.69s/it]
43%|████▎ | 18748/43939 [30:15:04<39:55:15, 5.71s/it]
43%|████▎ | 18749/43939 [30:15:10<39:55:42, 5.71s/it]
43%|████▎ | 18750/43939 [30:15:16<39:53:00, 5.70s/it]
{'loss': 0.9083, 'grad_norm': 1.293908953666687, 'learning_rate': 3.542937028277194e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.77, 'epoch': 1.71}
43%|████▎ | 18750/43939 [30:15:16<39:53:00, 5.70s/it]
43%|████▎ | 18751/43939 [30:15:22<39:54:28, 5.70s/it]
43%|████▎ | 18752/43939 [30:15:27<39:49:03, 5.69s/it]
43%|████▎ | 18753/43939 [30:15:33<40:21:47, 5.77s/it]
43%|████▎ | 18754/43939 [30:15:39<40:11:26, 5.74s/it]
43%|████▎ | 18755/43939 [30:15:44<40:04:30, 5.73s/it]
43%|████▎ | 18756/43939 [30:15:50<40:01:51, 5.72s/it]
43%|████▎ | 18757/43939 [30:15:56<39:58:44, 5.72s/it]
43%|████▎ | 18758/43939 [30:16:02<39:52:51, 5.70s/it]
43%|████▎ | 18759/43939 [30:16:07<39:47:52, 5.69s/it]
43%|████▎ | 18760/43939 [30:16:13<39:50:50, 5.70s/it]
43%|████▎ | 18761/43939 [30:16:19<39:52:14, 5.70s/it]
43%|████▎ | 18762/43939 [30:16:24<39:48:07, 5.69s/it]
43%|████▎ | 18763/43939 [30:16:30<39:51:41, 5.70s/it]
43%|████▎ | 18764/43939 [30:16:36<39:52:46, 5.70s/it]
43%|████▎ | 18765/43939 [30:16:41<39:54:13, 5.71s/it]
43%|████▎ | 18766/43939 [30:16:47<39:55:20, 5.71s/it]
43%|████▎ | 18767/43939 [30:16:53<39:58:43, 5.72s/it]
43%|████▎ | 18768/43939 [30:16:59<39:58:32, 5.72s/it]
43%|████▎ | 18769/43939 [30:17:04<40:05:26, 5.73s/it]
43%|████▎ | 18770/43939 [30:17:10<40:02:11, 5.73s/it]
43%|████▎ | 18771/43939 [30:17:16<39:56:41, 5.71s/it]
43%|████▎ | 18772/43939 [30:17:22<39:57:40, 5.72s/it]
43%|████▎ | 18773/43939 [30:17:27<39:56:58, 5.71s/it]
43%|████▎ | 18774/43939 [30:17:33<40:21:33, 5.77s/it]
43%|████▎ | 18775/43939 [30:17:39<40:08:33, 5.74s/it]
43%|████▎ | 18776/43939 [30:17:45<40:01:00, 5.73s/it]
43%|████▎ | 18777/43939 [30:17:50<39:51:12, 5.70s/it]
43%|████▎ | 18778/43939 [30:17:56<39:46:01, 5.69s/it]
43%|████▎ | 18779/43939 [30:18:02<39:54:45, 5.71s/it]
43%|████▎ | 18780/43939 [30:18:07<39:53:51, 5.71s/it]
43%|████▎ | 18781/43939 [30:18:13<39:49:59, 5.70s/it]
43%|████▎ | 18782/43939 [30:18:19<39:46:54, 5.69s/it]
43%|████▎ | 18783/43939 [30:18:24<39:46:46, 5.69s/it]
43%|████▎ | 18784/43939 [30:18:30<39:45:52, 5.69s/it]
43%|████▎ | 18785/43939 [30:18:36<39:56:31, 5.72s/it]
43%|████▎ | 18786/43939 [30:18:42<39:59:31, 5.72s/it]
43%|████▎ | 18787/43939 [30:18:47<39:50:31, 5.70s/it]
43%|████▎ | 18788/43939 [30:18:53<39:52:50, 5.71s/it]
43%|████▎ | 18789/43939 [30:18:59<39:53:32, 5.71s/it]
43%|████▎ | 18790/43939 [30:19:04<39:51:01, 5.70s/it]
43%|████▎ | 18791/43939 [30:19:10<39:56:54, 5.72s/it]
43%|████▎ | 18792/43939 [30:19:16<39:52:31, 5.71s/it]
43%|████▎ | 18793/43939 [30:19:21<39:53:11, 5.71s/it]
43%|████▎ | 18794/43939 [30:19:27<39:48:48, 5.70s/it]
43%|████▎ | 18795/43939 [30:19:33<40:22:20, 5.78s/it]
43%|████▎ | 18796/43939 [30:19:39<40:09:15, 5.75s/it]
43%|████▎ | 18797/43939 [30:19:44<40:04:19, 5.74s/it]
43%|████▎ | 18798/43939 [30:19:50<39:59:25, 5.73s/it]
43%|████▎ | 18799/43939 [30:19:56<39:52:41, 5.71s/it]
43%|████▎ | 18800/43939 [30:20:02<39:49:06, 5.70s/it]
{'loss': 0.918, 'grad_norm': 1.327156901359558, 'learning_rate': 3.533904012019007e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.53, 'epoch': 1.71}
43%|████▎ | 18800/43939 [30:20:02<39:49:06, 5.70s/it]
43%|████▎ | 18801/43939 [30:20:07<39:46:04, 5.70s/it]
43%|████▎ | 18802/43939 [30:20:13<39:45:05, 5.69s/it]
43%|████▎ | 18803/43939 [30:20:19<39:47:15, 5.70s/it]
43%|████▎ | 18804/43939 [30:20:24<39:47:32, 5.70s/it]
43%|████▎ | 18805/43939 [30:20:30<39:42:08, 5.69s/it]
43%|████▎ | 18806/43939 [30:20:36<39:42:41, 5.69s/it]
43%|████▎ | 18807/43939 [30:20:41<39:41:33, 5.69s/it]
43%|████▎ | 18808/43939 [30:20:47<39:43:49, 5.69s/it]
43%|████▎ | 18809/43939 [30:20:53<39:44:15, 5.69s/it]
43%|████▎ | 18810/43939 [30:20:58<39:43:27, 5.69s/it]
43%|████▎ | 18811/43939 [30:21:04<39:44:19, 5.69s/it]
43%|████▎ | 18812/43939 [30:21:10<39:44:24, 5.69s/it]
43%|████▎ | 18813/43939 [30:21:16<39:43:03, 5.69s/it]
43%|████▎ | 18814/43939 [30:21:21<39:44:00, 5.69s/it]
43%|████▎ | 18815/43939 [30:21:27<39:43:54, 5.69s/it]
43%|████▎ | 18816/43939 [30:21:33<40:14:45, 5.77s/it]
43%|████▎ | 18817/43939 [30:21:39<40:06:28, 5.75s/it]
43%|████▎ | 18818/43939 [30:21:44<40:04:37, 5.74s/it]
43%|████▎ | 18819/43939 [30:21:50<40:02:20, 5.74s/it]
43%|████▎ | 18820/43939 [30:21:56<39:52:51, 5.72s/it]
43%|████▎ | 18821/43939 [30:22:01<39:50:15, 5.71s/it]
43%|████▎ | 18822/43939 [30:22:07<39:44:02, 5.70s/it]
43%|████▎ | 18823/43939 [30:22:13<39:36:48, 5.68s/it]
43%|████▎ | 18824/43939 [30:22:18<39:47:07, 5.70s/it]
43%|████▎ | 18825/43939 [30:22:24<39:42:00, 5.69s/it]
43%|████▎ | 18826/43939 [30:22:30<39:41:02, 5.69s/it]
43%|████▎ | 18827/43939 [30:22:36<39:52:35, 5.72s/it]
43%|████▎ | 18828/43939 [30:22:41<39:51:03, 5.71s/it]
43%|████▎ | 18829/43939 [30:22:47<39:52:15, 5.72s/it]
43%|████▎ | 18830/43939 [30:22:53<39:49:56, 5.71s/it]
43%|████▎ | 18831/43939 [30:22:58<39:48:33, 5.71s/it]
43%|████▎ | 18832/43939 [30:23:04<39:48:44, 5.71s/it]
43%|████▎ | 18833/43939 [30:23:10<39:45:31, 5.70s/it]
43%|████▎ | 18834/43939 [30:23:16<39:49:42, 5.71s/it]
43%|████▎ | 18835/43939 [30:23:21<39:45:22, 5.70s/it]
43%|████▎ | 18836/43939 [30:23:27<39:44:42, 5.70s/it]
43%|████▎ | 18837/43939 [30:23:33<39:42:08, 5.69s/it]
43%|████▎ | 18838/43939 [30:23:38<40:08:20, 5.76s/it]
43%|████▎ | 18839/43939 [30:23:44<40:00:42, 5.74s/it]
43%|████▎ | 18840/43939 [30:23:50<39:58:24, 5.73s/it]
43%|████▎ | 18841/43939 [30:23:56<39:54:02, 5.72s/it]
43%|████▎ | 18842/43939 [30:24:01<39:50:32, 5.72s/it]
43%|████▎ | 18843/43939 [30:24:07<39:49:36, 5.71s/it]
43%|████▎ | 18844/43939 [30:24:13<39:55:09, 5.73s/it]
43%|████▎ | 18845/43939 [30:24:18<39:49:52, 5.71s/it]
43%|████▎ | 18846/43939 [30:24:24<39:48:09, 5.71s/it]
43%|████▎ | 18847/43939 [30:24:30<39:46:19, 5.71s/it]
43%|████▎ | 18848/43939 [30:24:36<39:52:31, 5.72s/it]
43%|████▎ | 18849/43939 [30:24:41<39:52:05, 5.72s/it]
43%|████▎ | 18850/43939 [30:24:47<39:49:23, 5.71s/it]
{'loss': 0.9167, 'grad_norm': 1.4621621370315552, 'learning_rate': 3.524854683495389e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.05, 'epoch': 1.72}
43%|████▎ | 18850/43939 [30:24:47<39:49:23, 5.71s/it]
43%|████▎ | 18851/43939 [30:24:53<39:50:10, 5.72s/it]
43%|████▎ | 18852/43939 [30:24:58<39:45:04, 5.70s/it]
43%|████▎ | 18853/43939 [30:25:04<39:43:15, 5.70s/it]
43%|████▎ | 18854/43939 [30:25:10<39:39:49, 5.69s/it]
43%|████▎ | 18855/43939 [30:25:16<39:46:10, 5.71s/it]
43%|████▎ | 18856/43939 [30:25:21<39:48:21, 5.71s/it]
43%|████▎ | 18857/43939 [30:25:27<39:48:50, 5.71s/it]
43%|████▎ | 18858/43939 [30:25:33<39:45:32, 5.71s/it]
43%|████▎ | 18859/43939 [30:25:39<40:14:54, 5.78s/it]
43%|████▎ | 18860/43939 [30:25:44<40:08:08, 5.76s/it]
43%|████▎ | 18861/43939 [30:25:50<39:55:23, 5.73s/it]
43%|████▎ | 18862/43939 [30:25:56<39:51:10, 5.72s/it]
43%|████▎ | 18863/43939 [30:26:01<39:51:04, 5.72s/it]
43%|████▎ | 18864/43939 [30:26:07<39:48:37, 5.72s/it]
43%|████▎ | 18865/43939 [30:26:13<39:46:50, 5.71s/it]
43%|████▎ | 18866/43939 [30:26:18<39:41:44, 5.70s/it]
43%|████▎ | 18867/43939 [30:26:24<39:42:37, 5.70s/it]
43%|████▎ | 18868/43939 [30:26:30<39:40:56, 5.70s/it]
43%|████▎ | 18869/43939 [30:26:36<39:40:03, 5.70s/it]
43%|████▎ | 18870/43939 [30:26:41<39:39:52, 5.70s/it]
43%|████▎ | 18871/43939 [30:26:47<39:42:48, 5.70s/it]
43%|████▎ | 18872/43939 [30:26:53<39:41:03, 5.70s/it]
43%|████▎ | 18873/43939 [30:26:58<39:46:39, 5.71s/it]
43%|████▎ | 18874/43939 [30:27:04<39:54:12, 5.73s/it]
43%|████▎ | 18875/43939 [30:27:10<39:57:48, 5.74s/it]
43%|████▎ | 18876/43939 [30:27:16<39:48:25, 5.72s/it]
43%|████▎ | 18877/43939 [30:27:21<39:48:14, 5.72s/it]
43%|████▎ | 18878/43939 [30:27:27<39:45:54, 5.71s/it]
43%|████▎ | 18879/43939 [30:27:33<39:42:29, 5.70s/it]
43%|████▎ | 18880/43939 [30:27:39<40:22:22, 5.80s/it]
43%|████▎ | 18881/43939 [30:27:44<40:05:08, 5.76s/it]
43%|████▎ | 18882/43939 [30:27:50<40:05:34, 5.76s/it]
43%|████▎ | 18883/43939 [30:27:56<40:02:10, 5.75s/it]
43%|████▎ | 18884/43939 [30:28:02<39:57:05, 5.74s/it]
43%|████▎ | 18885/43939 [30:28:07<39:50:47, 5.73s/it]
43%|████▎ | 18886/43939 [30:28:13<39:52:09, 5.73s/it]
43%|████▎ | 18887/43939 [30:28:19<39:56:29, 5.74s/it]
43%|████▎ | 18888/43939 [30:28:25<39:51:57, 5.73s/it]
43%|████▎ | 18889/43939 [30:28:30<39:50:21, 5.73s/it]
43%|████▎ | 18890/43939 [30:28:36<39:43:48, 5.71s/it]
43%|████▎ | 18891/43939 [30:28:42<39:44:25, 5.71s/it]
43%|████▎ | 18892/43939 [30:28:47<39:44:53, 5.71s/it]
43%|████▎ | 18893/43939 [30:28:53<39:41:58, 5.71s/it]
43%|████▎ | 18894/43939 [30:28:59<39:41:27, 5.71s/it]
43%|████▎ | 18895/43939 [30:29:04<39:38:21, 5.70s/it]
43%|████▎ | 18896/43939 [30:29:10<39:42:34, 5.71s/it]
43%|████▎ | 18897/43939 [30:29:16<39:42:04, 5.71s/it]
43%|████▎ | 18898/43939 [30:29:22<39:40:13, 5.70s/it]
43%|████▎ | 18899/43939 [30:29:27<39:36:10, 5.69s/it]
43%|████▎ | 18900/43939 [30:29:33<39:42:44, 5.71s/it]
{'loss': 0.9088, 'grad_norm': 1.3203855752944946, 'learning_rate': 3.515789185480763e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.55, 'epoch': 1.72}
43%|████▎ | 18900/43939 [30:29:33<39:42:44, 5.71s/it]
43%|████▎ | 18901/43939 [30:29:39<40:15:08, 5.79s/it]
43%|████▎ | 18902/43939 [30:29:45<40:13:06, 5.78s/it]
43%|████▎ | 18903/43939 [30:29:50<40:08:19, 5.77s/it]
43%|████▎ | 18904/43939 [30:29:56<39:58:39, 5.75s/it]
43%|████▎ | 18905/43939 [30:30:02<39:53:08, 5.74s/it]
43%|████▎ | 18906/43939 [30:30:08<39:48:25, 5.72s/it]
43%|████▎ | 18907/43939 [30:30:13<39:46:12, 5.72s/it]
43%|████▎ | 18908/43939 [30:30:19<39:40:03, 5.71s/it]
43%|████▎ | 18909/43939 [30:30:25<39:40:37, 5.71s/it]
43%|████▎ | 18910/43939 [30:30:30<39:43:41, 5.71s/it]
43%|████▎ | 18911/43939 [30:30:36<39:38:22, 5.70s/it]
43%|████▎ | 18912/43939 [30:30:42<39:41:19, 5.71s/it]
43%|████▎ | 18913/43939 [30:30:47<39:40:25, 5.71s/it]
43%|████▎ | 18914/43939 [30:30:53<39:41:01, 5.71s/it]
43%|████▎ | 18915/43939 [30:30:59<39:36:20, 5.70s/it]
43%|████▎ | 18916/43939 [30:31:05<39:42:08, 5.71s/it]
43%|████▎ | 18917/43939 [30:31:10<39:41:41, 5.71s/it]
43%|████▎ | 18918/43939 [30:31:16<39:35:40, 5.70s/it]
43%|████▎ | 18919/43939 [30:31:22<39:36:37, 5.70s/it]
43%|████▎ | 18920/43939 [30:31:27<39:38:05, 5.70s/it]
43%|████▎ | 18921/43939 [30:31:33<39:42:22, 5.71s/it]
43%|████▎ | 18922/43939 [30:31:39<39:42:47, 5.71s/it]
43%|████▎ | 18923/43939 [30:31:45<40:10:57, 5.78s/it]
43%|████▎ | 18924/43939 [30:31:50<39:59:06, 5.75s/it]
43%|████▎ | 18925/43939 [30:31:56<39:54:54, 5.74s/it]
43%|████▎ | 18926/43939 [30:32:02<39:50:01, 5.73s/it]
43%|████▎ | 18927/43939 [30:32:08<39:47:01, 5.73s/it]
43%|████▎ | 18928/43939 [30:32:13<39:44:30, 5.72s/it]
43%|████▎ | 18929/43939 [30:32:19<39:42:49, 5.72s/it]
43%|████▎ | 18930/43939 [30:32:25<39:40:24, 5.71s/it]
43%|████▎ | 18931/43939 [30:32:30<39:38:45, 5.71s/it]
43%|████▎ | 18932/43939 [30:32:36<39:39:52, 5.71s/it]
43%|████▎ | 18933/43939 [30:32:42<39:34:22, 5.70s/it]
43%|████▎ | 18934/43939 [30:32:48<39:35:52, 5.70s/it]
43%|████▎ | 18935/43939 [30:32:53<39:37:10, 5.70s/it]
43%|████▎ | 18936/43939 [30:32:59<39:39:33, 5.71s/it]
43%|████▎ | 18937/43939 [30:33:05<39:36:12, 5.70s/it]
43%|████▎ | 18938/43939 [30:33:10<39:35:41, 5.70s/it]
43%|████▎ | 18939/43939 [30:33:16<39:39:42, 5.71s/it]
43%|████▎ | 18940/43939 [30:33:22<39:39:28, 5.71s/it]
43%|████▎ | 18941/43939 [30:33:27<39:33:45, 5.70s/it]
43%|████▎ | 18942/43939 [30:33:33<39:30:55, 5.69s/it]
43%|████▎ | 18943/43939 [30:33:39<39:32:10, 5.69s/it]
43%|████▎ | 18944/43939 [30:33:45<39:57:10, 5.75s/it]
43%|████▎ | 18945/43939 [30:33:50<39:53:27, 5.75s/it]
43%|████▎ | 18946/43939 [30:33:56<39:49:14, 5.74s/it]
43%|████▎ | 18947/43939 [30:34:02<39:48:02, 5.73s/it]
43%|████▎ | 18948/43939 [30:34:08<39:43:55, 5.72s/it]
43%|████▎ | 18949/43939 [30:34:13<39:42:20, 5.72s/it]
43%|████▎ | 18950/43939 [30:34:19<39:38:58, 5.71s/it]
{'loss': 0.9222, 'grad_norm': 1.4525035619735718, 'learning_rate': 3.5067076610046637e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.94, 'epoch': 1.73}
43%|████▎ | 18950/43939 [30:34:19<39:38:58, 5.71s/it]
43%|████▎ | 18951/43939 [30:34:25<39:38:34, 5.71s/it]
43%|████▎ | 18952/43939 [30:34:30<39:42:50, 5.72s/it]
43%|████▎ | 18953/43939 [30:34:36<39:37:18, 5.71s/it]
43%|████▎ | 18954/43939 [30:34:42<39:32:13, 5.70s/it]
43%|████▎ | 18955/43939 [30:34:48<39:35:33, 5.70s/it]
43%|████▎ | 18956/43939 [30:34:53<39:39:21, 5.71s/it]
43%|████▎ | 18957/43939 [30:34:59<39:35:47, 5.71s/it]
43%|████▎ | 18958/43939 [30:35:05<39:39:26, 5.72s/it]
43%|████▎ | 18959/43939 [30:35:10<39:33:57, 5.70s/it]
43%|████▎ | 18960/43939 [30:35:16<39:32:19, 5.70s/it]
43%|████▎ | 18961/43939 [30:35:22<39:27:06, 5.69s/it]
43%|████▎ | 18962/43939 [30:35:27<39:33:37, 5.70s/it]
43%|████▎ | 18963/43939 [30:35:33<39:29:58, 5.69s/it]
43%|████▎ | 18964/43939 [30:35:39<39:26:00, 5.68s/it]
43%|████▎ | 18965/43939 [30:35:45<39:58:47, 5.76s/it]
43%|████▎ | 18966/43939 [30:35:50<39:51:08, 5.74s/it]
43%|████▎ | 18967/43939 [30:35:56<39:44:01, 5.73s/it]
43%|████▎ | 18968/43939 [30:36:02<39:40:52, 5.72s/it]
43%|████▎ | 18969/43939 [30:36:08<39:46:41, 5.73s/it]
43%|████▎ | 18970/43939 [30:36:13<39:43:31, 5.73s/it]
43%|████▎ | 18971/43939 [30:36:19<39:36:43, 5.71s/it]
43%|████▎ | 18972/43939 [30:36:25<39:31:15, 5.70s/it]
43%|████▎ | 18973/43939 [30:36:30<39:36:12, 5.71s/it]
43%|████▎ | 18974/43939 [30:36:36<39:33:56, 5.71s/it]
43%|████▎ | 18975/43939 [30:36:42<39:32:01, 5.70s/it]
43%|████▎ | 18976/43939 [30:36:47<39:35:18, 5.71s/it]
43%|████▎ | 18977/43939 [30:36:53<39:34:19, 5.71s/it]
43%|████▎ | 18978/43939 [30:36:59<39:39:37, 5.72s/it]
43%|████▎ | 18979/43939 [30:37:05<39:36:53, 5.71s/it]
43%|████▎ | 18980/43939 [30:37:10<39:38:01, 5.72s/it]
43%|████▎ | 18981/43939 [30:37:16<39:36:44, 5.71s/it]
43%|████▎ | 18982/43939 [30:37:22<39:30:59, 5.70s/it]
43%|████▎ | 18983/43939 [30:37:27<39:30:47, 5.70s/it]
43%|████▎ | 18984/43939 [30:37:33<39:34:48, 5.71s/it]
43%|████▎ | 18985/43939 [30:37:39<39:37:40, 5.72s/it]
43%|████▎ | 18986/43939 [30:37:45<40:09:56, 5.79s/it]
43%|████▎ | 18987/43939 [30:37:51<39:56:47, 5.76s/it]
43%|████▎ | 18988/43939 [30:37:56<39:46:47, 5.74s/it]
43%|████▎ | 18989/43939 [30:38:02<39:41:40, 5.73s/it]
43%|████▎ | 18990/43939 [30:38:08<39:45:13, 5.74s/it]
43%|████▎ | 18991/43939 [30:38:13<39:36:57, 5.72s/it]
43%|████▎ | 18992/43939 [30:38:19<39:37:36, 5.72s/it]
43%|████▎ | 18993/43939 [30:38:25<39:36:43, 5.72s/it]
43%|████▎ | 18994/43939 [30:38:31<39:39:53, 5.72s/it]
43%|████▎ | 18995/43939 [30:38:36<39:34:09, 5.71s/it]
43%|████▎ | 18996/43939 [30:38:42<39:33:32, 5.71s/it]
43%|████▎ | 18997/43939 [30:38:48<39:30:01, 5.70s/it]
43%|████▎ | 18998/43939 [30:38:53<39:31:59, 5.71s/it]
43%|████▎ | 18999/43939 [30:38:59<39:31:29, 5.71s/it]
43%|████▎ | 19000/43939 [30:39:05<39:31:06, 5.70s/it]
{'loss': 0.9337, 'grad_norm': 1.3682067394256592, 'learning_rate': 3.497610253349482e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.26, 'epoch': 1.73}
43%|████▎ | 19000/43939 [30:39:05<39:31:06, 5.70s/it]
43%|████▎ | 19001/43939 [30:39:10<39:30:17, 5.70s/it]
43%|████▎ | 19002/43939 [30:39:16<39:26:13, 5.69s/it]
43%|████▎ | 19003/43939 [30:39:22<39:32:13, 5.71s/it]
43%|████▎ | 19004/43939 [30:39:28<39:29:28, 5.70s/it]
43%|████▎ | 19005/43939 [30:39:33<39:24:27, 5.69s/it]
43%|████▎ | 19006/43939 [30:39:39<39:25:41, 5.69s/it]
43%|████▎ | 19007/43939 [30:39:45<39:26:02, 5.69s/it]
43%|████▎ | 19008/43939 [30:39:50<39:50:37, 5.75s/it]
43%|████▎ | 19009/43939 [30:39:56<39:44:42, 5.74s/it]
43%|████▎ | 19010/43939 [30:40:02<39:37:24, 5.72s/it]
43%|████▎ | 19011/43939 [30:40:08<39:32:50, 5.71s/it]
43%|████▎ | 19012/43939 [30:40:13<39:30:18, 5.71s/it]
43%|████▎ | 19013/43939 [30:40:19<39:31:21, 5.71s/it]
43%|████▎ | 19014/43939 [30:40:25<39:35:01, 5.72s/it]
43%|████▎ | 19015/43939 [30:40:30<39:29:40, 5.70s/it]
43%|████▎ | 19016/43939 [30:40:36<39:31:35, 5.71s/it]
43%|████▎ | 19017/43939 [30:40:42<39:30:49, 5.71s/it]
43%|████▎ | 19018/43939 [30:40:48<39:31:25, 5.71s/it]
43%|████▎ | 19019/43939 [30:40:53<39:37:18, 5.72s/it]
43%|████▎ | 19020/43939 [30:40:59<39:32:00, 5.71s/it]
43%|████▎ | 19021/43939 [30:41:05<39:30:12, 5.71s/it]
43%|████▎ | 19022/43939 [30:41:10<39:26:27, 5.70s/it]
43%|████▎ | 19023/43939 [30:41:16<39:26:11, 5.70s/it]
43%|████▎ | 19024/43939 [30:41:22<39:26:56, 5.70s/it]
43%|████▎ | 19025/43939 [30:41:27<39:27:13, 5.70s/it]
43%|████▎ | 19026/43939 [30:41:33<39:24:28, 5.69s/it]
43%|████▎ | 19027/43939 [30:41:39<39:24:39, 5.70s/it]
43%|████▎ | 19028/43939 [30:41:45<39:22:30, 5.69s/it]
43%|████▎ | 19029/43939 [30:41:50<39:51:20, 5.76s/it]
43%|████▎ | 19030/43939 [30:41:56<39:42:43, 5.74s/it]
43%|████▎ | 19031/43939 [30:42:02<39:37:59, 5.73s/it]
43%|████▎ | 19032/43939 [30:42:08<39:32:16, 5.71s/it]
43%|████▎ | 19033/43939 [30:42:13<39:29:55, 5.71s/it]
43%|████▎ | 19034/43939 [30:42:19<39:29:29, 5.71s/it]
43%|████▎ | 19035/43939 [30:42:25<39:30:36, 5.71s/it]
43%|████▎ | 19036/43939 [30:42:30<39:31:01, 5.71s/it]
43%|████▎ | 19037/43939 [30:42:36<39:28:35, 5.71s/it]
43%|████▎ | 19038/43939 [30:42:42<39:27:25, 5.70s/it]
43%|████▎ | 19039/43939 [30:42:47<39:25:14, 5.70s/it]
43%|████▎ | 19040/43939 [30:42:53<39:34:30, 5.72s/it]
43%|████▎ | 19041/43939 [30:42:59<39:32:57, 5.72s/it]
43%|████▎ | 19042/43939 [30:43:05<39:30:11, 5.71s/it]
43%|████▎ | 19043/43939 [30:43:10<39:24:36, 5.70s/it]
43%|████▎ | 19044/43939 [30:43:16<39:29:03, 5.71s/it]
43%|████▎ | 19045/43939 [30:43:22<39:25:14, 5.70s/it]
43%|████▎ | 19046/43939 [30:43:27<39:20:54, 5.69s/it]
43%|████▎ | 19047/43939 [30:43:33<39:19:06, 5.69s/it]
43%|████▎ | 19048/43939 [30:43:39<39:20:49, 5.69s/it]
43%|████▎ | 19049/43939 [30:43:44<39:18:37, 5.69s/it]
43%|████▎ | 19050/43939 [30:43:50<39:49:03, 5.76s/it]
{'loss': 0.9039, 'grad_norm': 1.1274014711380005, 'learning_rate': 3.488497106048202e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 441.7, 'epoch': 1.73}
43%|████▎ | 19050/43939 [30:43:50<39:49:03, 5.76s/it]
43%|████▎ | 19051/43939 [30:43:56<39:39:46, 5.74s/it]
43%|████▎ | 19052/43939 [30:44:02<39:32:07, 5.72s/it]
43%|████▎ | 19053/43939 [30:44:07<39:28:32, 5.71s/it]
43%|████▎ | 19054/43939 [30:44:13<39:29:42, 5.71s/it]
43%|████▎ | 19055/43939 [30:44:19<39:31:17, 5.72s/it]
43%|████▎ | 19056/43939 [30:44:25<39:31:12, 5.72s/it]
43%|████▎ | 19057/43939 [30:44:30<39:35:51, 5.73s/it]
43%|████▎ | 19058/43939 [30:44:36<39:37:32, 5.73s/it]
43%|████▎ | 19059/43939 [30:44:42<39:29:54, 5.72s/it]
43%|████▎ | 19060/43939 [30:44:47<39:28:37, 5.71s/it]
43%|████▎ | 19061/43939 [30:44:53<39:29:23, 5.71s/it]
43%|████▎ | 19062/43939 [30:44:59<39:24:33, 5.70s/it]
43%|████▎ | 19063/43939 [30:45:05<39:23:52, 5.70s/it]
43%|████▎ | 19064/43939 [30:45:10<39:21:24, 5.70s/it]
43%|████▎ | 19065/43939 [30:45:16<39:22:56, 5.70s/it]
43%|████▎ | 19066/43939 [30:45:22<39:24:39, 5.70s/it]
43%|████▎ | 19067/43939 [30:45:27<39:25:07, 5.71s/it]
43%|████▎ | 19068/43939 [30:45:33<39:26:28, 5.71s/it]
43%|████▎ | 19069/43939 [30:45:39<39:19:39, 5.69s/it]
43%|████▎ | 19070/43939 [30:45:44<39:30:10, 5.72s/it]
43%|████▎ | 19071/43939 [30:45:50<39:26:49, 5.71s/it]
43%|████▎ | 19072/43939 [30:45:56<39:52:11, 5.77s/it]
43%|████▎ | 19073/43939 [30:46:02<39:43:50, 5.75s/it]
43%|████▎ | 19074/43939 [30:46:08<39:39:42, 5.74s/it]
43%|████▎ | 19075/43939 [30:46:13<39:32:58, 5.73s/it]
43%|████▎ | 19076/43939 [30:46:19<39:27:22, 5.71s/it]
43%|████▎ | 19077/43939 [30:46:25<39:25:42, 5.71s/it]
43%|████▎ | 19078/43939 [30:46:30<39:26:09, 5.71s/it]
43%|████▎ | 19079/43939 [30:46:36<39:24:49, 5.71s/it]
43%|████▎ | 19080/43939 [30:46:42<39:23:27, 5.70s/it]
43%|████▎ | 19081/43939 [30:46:47<39:22:29, 5.70s/it]
43%|████▎ | 19082/43939 [30:46:53<39:20:34, 5.70s/it]
43%|████▎ | 19083/43939 [30:46:59<39:18:54, 5.69s/it]
43%|████▎ | 19084/43939 [30:47:05<39:23:09, 5.70s/it]
43%|████▎ | 19085/43939 [30:47:10<39:20:27, 5.70s/it]
43%|████▎ | 19086/43939 [30:47:16<39:25:13, 5.71s/it]
43%|████▎ | 19087/43939 [30:47:22<39:23:54, 5.71s/it]
43%|████▎ | 19088/43939 [30:47:27<39:24:35, 5.71s/it]
43%|████▎ | 19089/43939 [30:47:33<39:23:39, 5.71s/it]
43%|████▎ | 19090/43939 [30:47:39<39:27:42, 5.72s/it]
43%|████▎ | 19091/43939 [30:47:44<39:25:33, 5.71s/it]
43%|████▎ | 19092/43939 [30:47:50<39:21:54, 5.70s/it]
43%|████▎ | 19093/43939 [30:47:56<39:44:19, 5.76s/it]
43%|████▎ | 19094/43939 [30:48:02<39:38:29, 5.74s/it]
43%|████▎ | 19095/43939 [30:48:07<39:28:43, 5.72s/it]
43%|████▎ | 19096/43939 [30:48:13<39:22:59, 5.71s/it]
43%|████▎ | 19097/43939 [30:48:19<39:25:20, 5.71s/it]
43%|████▎ | 19098/43939 [30:48:25<39:21:00, 5.70s/it]
43%|████▎ | 19099/43939 [30:48:30<39:19:47, 5.70s/it]
43%|████▎ | 19100/43939 [30:48:36<39:17:00, 5.69s/it]
{'loss': 0.9051, 'grad_norm': 1.4040976762771606, 'learning_rate': 3.4793683628821375e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.65, 'epoch': 1.74}
43%|████▎ | 19100/43939 [30:48:36<39:17:00, 5.69s/it]
43%|████▎ | 19101/43939 [30:48:42<39:16:15, 5.69s/it]
43%|████▎ | 19102/43939 [30:48:47<39:15:06, 5.69s/it]
43%|████▎ | 19103/43939 [30:48:53<39:17:51, 5.70s/it]
43%|████▎ | 19104/43939 [30:48:59<39:20:14, 5.70s/it]
43%|████▎ | 19105/43939 [30:49:04<39:16:34, 5.69s/it]
43%|████▎ | 19106/43939 [30:49:10<39:12:03, 5.68s/it]
43%|████▎ | 19107/43939 [30:49:16<39:11:14, 5.68s/it]
43%|████▎ | 19108/43939 [30:49:21<39:14:07, 5.69s/it]
43%|████▎ | 19109/43939 [30:49:27<39:14:02, 5.69s/it]
43%|████▎ | 19110/43939 [30:49:33<39:16:08, 5.69s/it]
43%|████▎ | 19111/43939 [30:49:39<39:19:39, 5.70s/it]
43%|████▎ | 19112/43939 [30:49:44<39:17:58, 5.70s/it]
43%|████▎ | 19113/43939 [30:49:50<39:11:13, 5.68s/it]
44%|████▎ | 19114/43939 [30:49:56<39:42:11, 5.76s/it]
44%|████▎ | 19115/43939 [30:50:01<39:32:14, 5.73s/it]
44%|████▎ | 19116/43939 [30:50:07<39:25:20, 5.72s/it]
44%|████▎ | 19117/43939 [30:50:13<39:17:02, 5.70s/it]
44%|████▎ | 19118/43939 [30:50:18<39:17:50, 5.70s/it]
44%|████▎ | 19119/43939 [30:50:24<39:19:08, 5.70s/it]
44%|████▎ | 19120/43939 [30:50:30<39:18:40, 5.70s/it]
44%|████▎ | 19121/43939 [30:50:36<39:27:28, 5.72s/it]
44%|████▎ | 19122/43939 [30:50:41<39:22:49, 5.71s/it]
44%|████▎ | 19123/43939 [30:50:47<39:21:50, 5.71s/it]
44%|████▎ | 19124/43939 [30:50:53<39:18:09, 5.70s/it]
44%|████▎ | 19125/43939 [30:50:58<39:19:33, 5.71s/it]
44%|████▎ | 19126/43939 [30:51:04<39:21:20, 5.71s/it]
44%|████▎ | 19127/43939 [30:51:10<39:18:42, 5.70s/it]
44%|████▎ | 19128/43939 [30:51:16<39:14:47, 5.69s/it]
44%|████▎ | 19129/43939 [30:51:21<39:14:22, 5.69s/it]
44%|████▎ | 19130/43939 [30:51:27<39:11:48, 5.69s/it]
44%|████▎ | 19131/43939 [30:51:33<39:13:41, 5.69s/it]
44%|████▎ | 19132/43939 [30:51:38<39:16:49, 5.70s/it]
44%|████▎ | 19133/43939 [30:51:44<39:13:49, 5.69s/it]
44%|████▎ | 19134/43939 [30:51:50<39:14:01, 5.69s/it]
44%|████▎ | 19135/43939 [30:51:56<39:43:13, 5.76s/it]
44%|████▎ | 19136/43939 [30:52:01<39:35:11, 5.75s/it]
44%|████▎ | 19137/43939 [30:52:07<39:32:05, 5.74s/it]
44%|████▎ | 19138/43939 [30:52:13<39:26:33, 5.73s/it]
44%|████▎ | 19139/43939 [30:52:18<39:25:16, 5.72s/it]
44%|████▎ | 19140/43939 [30:52:24<39:24:39, 5.72s/it]
44%|████▎ | 19141/43939 [30:52:30<39:23:21, 5.72s/it]
44%|████▎ | 19142/43939 [30:52:36<39:22:31, 5.72s/it]
44%|████▎ | 19143/43939 [30:52:41<39:22:41, 5.72s/it]
44%|████▎ | 19144/43939 [30:52:47<39:20:03, 5.71s/it]
44%|████▎ | 19145/43939 [30:52:53<39:16:30, 5.70s/it]
44%|████▎ | 19146/43939 [30:52:58<39:15:27, 5.70s/it]
44%|████▎ | 19147/43939 [30:53:04<39:11:30, 5.69s/it]
44%|████▎ | 19148/43939 [30:53:10<39:08:33, 5.68s/it]
44%|████▎ | 19149/43939 [30:53:15<39:14:10, 5.70s/it]
44%|████▎ | 19150/43939 [30:53:21<39:14:26, 5.70s/it]
{'loss': 0.8916, 'grad_norm': 1.467485785484314, 'learning_rate': 3.470224167878665e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.92, 'epoch': 1.74}
44%|████▎ | 19150/43939 [30:53:21<39:14:26, 5.70s/it]
44%|████▎ | 19151/43939 [30:53:27<39:11:33, 5.69s/it]
44%|████▎ | 19152/43939 [30:53:33<39:12:45, 5.70s/it]
44%|████▎ | 19153/43939 [30:53:38<39:17:59, 5.71s/it]
44%|████▎ | 19154/43939 [30:53:44<39:14:30, 5.70s/it]
44%|████▎ | 19155/43939 [30:53:50<39:15:06, 5.70s/it]
44%|████▎ | 19156/43939 [30:53:55<39:12:09, 5.69s/it]
44%|████▎ | 19157/43939 [30:54:01<39:39:09, 5.76s/it]
44%|████▎ | 19158/43939 [30:54:07<39:35:36, 5.75s/it]
44%|████▎ | 19159/43939 [30:54:13<39:24:55, 5.73s/it]
44%|████▎ | 19160/43939 [30:54:18<39:21:12, 5.72s/it]
44%|████▎ | 19161/43939 [30:54:24<39:21:17, 5.72s/it]
44%|████▎ | 19162/43939 [30:54:30<39:17:54, 5.71s/it]
44%|████▎ | 19163/43939 [30:54:35<39:13:50, 5.70s/it]
44%|████▎ | 19164/43939 [30:54:41<39:11:51, 5.70s/it]
44%|████▎ | 19165/43939 [30:54:47<39:09:26, 5.69s/it]
44%|████▎ | 19166/43939 [30:54:53<39:14:20, 5.70s/it]
44%|████▎ | 19167/43939 [30:54:58<39:19:29, 5.71s/it]
44%|████▎ | 19168/43939 [30:55:04<39:27:01, 5.73s/it]
44%|████▎ | 19169/43939 [30:55:10<39:22:31, 5.72s/it]
44%|████▎ | 19170/43939 [30:55:15<39:23:19, 5.72s/it]
44%|████▎ | 19171/43939 [30:55:21<39:24:35, 5.73s/it]
44%|████▎ | 19172/43939 [30:55:27<39:18:50, 5.71s/it]
44%|████▎ | 19173/43939 [30:55:33<39:15:37, 5.71s/it]
44%|████▎ | 19174/43939 [30:55:38<39:19:12, 5.72s/it]
44%|████▎ | 19175/43939 [30:55:44<39:23:15, 5.73s/it]
44%|████▎ | 19176/43939 [30:55:50<39:20:02, 5.72s/it]
44%|████▎ | 19177/43939 [30:55:55<39:15:48, 5.71s/it]
44%|████▎ | 19178/43939 [30:56:01<39:42:32, 5.77s/it]
44%|████▎ | 19179/43939 [30:56:07<39:40:31, 5.77s/it]
44%|████▎ | 19180/43939 [30:56:13<39:28:33, 5.74s/it]
44%|████▎ | 19181/43939 [30:56:19<39:23:46, 5.73s/it]
44%|████▎ | 19182/43939 [30:56:24<39:18:46, 5.72s/it]
44%|████▎ | 19183/43939 [30:56:30<39:16:16, 5.71s/it]
44%|████▎ | 19184/43939 [30:56:36<39:18:02, 5.72s/it]
44%|████▎ | 19185/43939 [30:56:41<39:19:03, 5.72s/it]
44%|████▎ | 19186/43939 [30:56:47<39:15:20, 5.71s/it]
44%|████▎ | 19187/43939 [30:56:53<39:10:54, 5.70s/it]
44%|████▎ | 19188/43939 [30:56:59<39:23:26, 5.73s/it]
44%|████▎ | 19189/43939 [30:57:04<39:20:46, 5.72s/it]
44%|████▎ | 19190/43939 [30:57:10<39:17:45, 5.72s/it]
44%|████▎ | 19191/43939 [30:57:16<39:17:31, 5.72s/it]
44%|████▎ | 19192/43939 [30:57:21<39:15:51, 5.71s/it]
44%|████▎ | 19193/43939 [30:57:27<39:18:01, 5.72s/it]
44%|████▎ | 19194/43939 [30:57:33<39:13:45, 5.71s/it]
44%|████▎ | 19195/43939 [30:57:39<39:17:10, 5.72s/it]
44%|████▎ | 19196/43939 [30:57:44<39:11:06, 5.70s/it]
44%|████▎ | 19197/43939 [30:57:50<39:09:12, 5.70s/it]
44%|████▎ | 19198/43939 [30:57:56<39:17:10, 5.72s/it]
44%|████▎ | 19199/43939 [30:58:02<39:51:41, 5.80s/it]
44%|████▎ | 19200/43939 [30:58:07<39:38:08, 5.77s/it]
{'loss': 0.9076, 'grad_norm': 1.0342504978179932, 'learning_rate': 3.461064665308948e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.85, 'epoch': 1.75}
44%|████▎ | 19200/43939 [30:58:07<39:38:08, 5.77s/it]
44%|████▎ | 19201/43939 [30:58:13<39:28:17, 5.74s/it]
44%|████▎ | 19202/43939 [30:58:19<39:32:13, 5.75s/it]
44%|████▎ | 19203/43939 [30:58:24<39:21:13, 5.73s/it]
44%|████▎ | 19204/43939 [30:58:30<39:19:57, 5.72s/it]
44%|████▎ | 19205/43939 [30:58:36<39:11:07, 5.70s/it]
44%|████▎ | 19206/43939 [30:58:42<39:11:11, 5.70s/it]
44%|████▎ | 19207/43939 [30:58:47<39:11:28, 5.70s/it]
44%|████▎ | 19208/43939 [30:58:53<39:05:46, 5.69s/it]
44%|████▎ | 19209/43939 [30:58:59<39:12:31, 5.71s/it]
44%|████▎ | 19210/43939 [30:59:04<39:12:35, 5.71s/it]
44%|████▎ | 19211/43939 [30:59:10<39:10:18, 5.70s/it]
44%|████▎ | 19212/43939 [30:59:16<39:10:14, 5.70s/it]
44%|████▎ | 19213/43939 [30:59:21<39:10:29, 5.70s/it]
44%|████▎ | 19214/43939 [30:59:27<39:09:25, 5.70s/it]
44%|████▎ | 19215/43939 [30:59:33<39:09:59, 5.70s/it]
44%|████▎ | 19216/43939 [30:59:39<39:09:01, 5.70s/it]
44%|████▎ | 19217/43939 [30:59:44<39:07:56, 5.70s/it]
44%|████▎ | 19218/43939 [30:59:50<39:08:57, 5.70s/it]
44%|████▎ | 19219/43939 [30:59:56<39:09:35, 5.70s/it]
44%|████▎ | 19220/43939 [31:00:02<39:35:53, 5.77s/it]
44%|████▎ | 19221/43939 [31:00:07<39:27:03, 5.75s/it]
44%|████▎ | 19222/43939 [31:00:13<39:20:55, 5.73s/it]
44%|████▎ | 19223/43939 [31:00:19<39:15:44, 5.72s/it]
44%|████▍ | 19224/43939 [31:00:24<39:18:46, 5.73s/it]
44%|████▍ | 19225/43939 [31:00:30<39:18:40, 5.73s/it]
44%|████▍ | 19226/43939 [31:00:36<39:13:00, 5.71s/it]
44%|████▍ | 19227/43939 [31:00:41<39:08:29, 5.70s/it]
44%|████▍ | 19228/43939 [31:00:47<39:07:08, 5.70s/it]
44%|████▍ | 19229/43939 [31:00:53<39:05:35, 5.70s/it]
44%|████▍ | 19230/43939 [31:00:59<39:08:41, 5.70s/it]
44%|████▍ | 19231/43939 [31:01:04<39:02:47, 5.69s/it]
44%|████▍ | 19232/43939 [31:01:10<39:07:26, 5.70s/it]
44%|████▍ | 19233/43939 [31:01:16<39:06:01, 5.70s/it]
44%|████▍ | 19234/43939 [31:01:21<39:08:40, 5.70s/it]
44%|████▍ | 19235/43939 [31:01:27<39:06:17, 5.70s/it]
44%|████▍ | 19236/43939 [31:01:33<39:05:16, 5.70s/it]
44%|████▍ | 19237/43939 [31:01:38<39:06:40, 5.70s/it]
44%|████▍ | 19238/43939 [31:01:44<39:02:19, 5.69s/it]
44%|████▍ | 19239/43939 [31:01:50<39:04:18, 5.69s/it]
44%|████▍ | 19240/43939 [31:01:56<39:10:39, 5.71s/it]
44%|████▍ | 19241/43939 [31:02:01<39:09:15, 5.71s/it]
44%|████▍ | 19242/43939 [31:02:07<39:33:04, 5.77s/it]
44%|████▍ | 19243/43939 [31:02:13<39:23:55, 5.74s/it]
44%|████▍ | 19244/43939 [31:02:19<39:24:16, 5.74s/it]
44%|████▍ | 19245/43939 [31:02:24<39:19:07, 5.73s/it]
44%|████▍ | 19246/43939 [31:02:30<39:15:39, 5.72s/it]
44%|████▍ | 19247/43939 [31:02:36<39:10:05, 5.71s/it]
44%|████▍ | 19248/43939 [31:02:41<39:08:06, 5.71s/it]
44%|████▍ | 19249/43939 [31:02:47<39:11:38, 5.71s/it]
44%|████▍ | 19250/43939 [31:02:53<39:07:04, 5.70s/it]
{'loss': 0.9466, 'grad_norm': 1.26194167137146, 'learning_rate': 3.4518899996856646e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.34, 'epoch': 1.75}
44%|████▍ | 19250/43939 [31:02:53<39:07:04, 5.70s/it]
44%|████▍ | 19251/43939 [31:02:58<39:01:01, 5.69s/it]
44%|████▍ | 19252/43939 [31:03:04<38:56:46, 5.68s/it]
44%|████▍ | 19253/43939 [31:03:10<38:59:58, 5.69s/it]
44%|████▍ | 19254/43939 [31:03:16<39:09:03, 5.71s/it]
44%|████▍ | 19255/43939 [31:03:21<39:08:11, 5.71s/it]
44%|████▍ | 19256/43939 [31:03:27<39:05:56, 5.70s/it]
44%|████▍ | 19257/43939 [31:03:33<39:07:21, 5.71s/it]
44%|████▍ | 19258/43939 [31:03:38<39:02:12, 5.69s/it]
44%|████▍ | 19259/43939 [31:03:44<39:00:51, 5.69s/it]
44%|████▍ | 19260/43939 [31:03:50<39:09:35, 5.71s/it]
44%|████▍ | 19261/43939 [31:03:56<39:10:34, 5.72s/it]
44%|████▍ | 19262/43939 [31:04:01<39:05:54, 5.70s/it]
44%|████▍ | 19263/43939 [31:04:07<39:34:40, 5.77s/it]
44%|████▍ | 19264/43939 [31:04:13<39:26:45, 5.76s/it]
44%|████▍ | 19265/43939 [31:04:19<39:16:49, 5.73s/it]
44%|████▍ | 19266/43939 [31:04:24<39:13:44, 5.72s/it]
44%|████▍ | 19267/43939 [31:04:30<39:11:15, 5.72s/it]
44%|████▍ | 19268/43939 [31:04:36<39:04:46, 5.70s/it]
44%|████▍ | 19269/43939 [31:04:41<39:06:47, 5.71s/it]
44%|████▍ | 19270/43939 [31:04:47<39:09:42, 5.71s/it]
44%|████▍ | 19271/43939 [31:04:53<39:08:56, 5.71s/it]
44%|████▍ | 19272/43939 [31:04:58<39:09:04, 5.71s/it]
44%|████▍ | 19273/43939 [31:05:04<39:06:47, 5.71s/it]
44%|████▍ | 19274/43939 [31:05:10<39:12:44, 5.72s/it]
44%|████▍ | 19275/43939 [31:05:16<39:13:42, 5.73s/it]
44%|████▍ | 19276/43939 [31:05:21<39:08:15, 5.71s/it]
44%|████▍ | 19277/43939 [31:05:27<39:10:37, 5.72s/it]
44%|████▍ | 19278/43939 [31:05:33<39:08:36, 5.71s/it]
44%|████▍ | 19279/43939 [31:05:38<39:06:02, 5.71s/it]
44%|████▍ | 19280/43939 [31:05:44<39:07:52, 5.71s/it]
44%|████▍ | 19281/43939 [31:05:50<39:06:21, 5.71s/it]
44%|████▍ | 19282/43939 [31:05:56<39:03:35, 5.70s/it]
44%|████▍ | 19283/43939 [31:06:01<39:01:14, 5.70s/it]
44%|████▍ | 19284/43939 [31:06:07<39:27:44, 5.76s/it]
44%|████▍ | 19285/43939 [31:06:13<39:17:16, 5.74s/it]
44%|████▍ | 19286/43939 [31:06:19<39:15:49, 5.73s/it]
44%|████▍ | 19287/43939 [31:06:24<39:10:41, 5.72s/it]
44%|████▍ | 19288/43939 [31:06:30<39:09:28, 5.72s/it]
44%|████▍ | 19289/43939 [31:06:36<39:07:46, 5.71s/it]
44%|████▍ | 19290/43939 [31:06:41<39:09:44, 5.72s/it]
44%|████▍ | 19291/43939 [31:06:47<39:13:31, 5.73s/it]
44%|████▍ | 19292/43939 [31:06:53<39:14:09, 5.73s/it]
44%|████▍ | 19293/43939 [31:06:59<39:10:44, 5.72s/it]
44%|████▍ | 19294/43939 [31:07:04<39:05:33, 5.71s/it]
44%|████▍ | 19295/43939 [31:07:10<39:06:42, 5.71s/it]
44%|████▍ | 19296/43939 [31:07:16<39:04:53, 5.71s/it]
44%|████▍ | 19297/43939 [31:07:21<39:04:31, 5.71s/it]
44%|████▍ | 19298/43939 [31:07:27<39:09:54, 5.72s/it]
44%|████▍ | 19299/43939 [31:07:33<39:09:03, 5.72s/it]
44%|████▍ | 19300/43939 [31:07:39<39:11:44, 5.73s/it]
{'loss': 0.9154, 'grad_norm': 1.0763055086135864, 'learning_rate': 3.442700315760726e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.37, 'epoch': 1.76}
44%|████▍ | 19300/43939 [31:07:39<39:11:44, 5.73s/it]
44%|████▍ | 19301/43939 [31:07:44<39:07:19, 5.72s/it]
44%|████▍ | 19302/43939 [31:07:50<39:03:56, 5.71s/it]
44%|████▍ | 19303/43939 [31:07:56<39:03:51, 5.71s/it]
44%|████▍ | 19304/43939 [31:08:01<39:01:28, 5.70s/it]
44%|████▍ | 19305/43939 [31:08:07<39:01:33, 5.70s/it]
44%|████▍ | 19306/43939 [31:08:13<39:36:29, 5.79s/it]
44%|████▍ | 19307/43939 [31:08:19<39:25:31, 5.76s/it]
44%|████▍ | 19308/43939 [31:08:25<39:17:34, 5.74s/it]
44%|████▍ | 19309/43939 [31:08:30<39:14:18, 5.74s/it]
44%|████▍ | 19310/43939 [31:08:36<39:09:52, 5.72s/it]
44%|████▍ | 19311/43939 [31:08:42<39:02:32, 5.71s/it]
44%|████▍ | 19312/43939 [31:08:47<38:58:09, 5.70s/it]
44%|████▍ | 19313/43939 [31:08:53<39:06:11, 5.72s/it]
44%|████▍ | 19314/43939 [31:08:59<39:08:22, 5.72s/it]
44%|████▍ | 19315/43939 [31:09:04<39:05:26, 5.72s/it]
44%|████▍ | 19316/43939 [31:09:10<39:03:13, 5.71s/it]
44%|████▍ | 19317/43939 [31:09:16<39:02:51, 5.71s/it]
44%|████▍ | 19318/43939 [31:09:22<39:00:47, 5.70s/it]
44%|████▍ | 19319/43939 [31:09:27<38:59:21, 5.70s/it]
44%|████▍ | 19320/43939 [31:09:33<39:02:19, 5.71s/it]
44%|████▍ | 19321/43939 [31:09:39<39:00:08, 5.70s/it]
44%|████▍ | 19322/43939 [31:09:44<39:02:25, 5.71s/it]
44%|████▍ | 19323/43939 [31:09:50<39:10:19, 5.73s/it]
44%|████▍ | 19324/43939 [31:09:56<39:06:32, 5.72s/it]
44%|████▍ | 19325/43939 [31:10:02<38:59:52, 5.70s/it]
44%|████▍ | 19326/43939 [31:10:07<38:57:40, 5.70s/it]
44%|████▍ | 19327/43939 [31:10:13<39:24:45, 5.76s/it]
44%|████▍ | 19328/43939 [31:10:19<39:19:32, 5.75s/it]
44%|████▍ | 19329/43939 [31:10:25<39:12:40, 5.74s/it]
44%|████▍ | 19330/43939 [31:10:30<39:05:02, 5.72s/it]
44%|████▍ | 19331/43939 [31:10:36<39:03:36, 5.71s/it]
44%|████▍ | 19332/43939 [31:10:42<38:57:44, 5.70s/it]
44%|████▍ | 19333/43939 [31:10:47<38:57:40, 5.70s/it]
44%|████▍ | 19334/43939 [31:10:53<38:57:20, 5.70s/it]
44%|████▍ | 19335/43939 [31:10:59<38:54:29, 5.69s/it]
44%|████▍ | 19336/43939 [31:11:04<38:56:22, 5.70s/it]
44%|████▍ | 19337/43939 [31:11:10<38:55:53, 5.70s/it]
44%|████▍ | 19338/43939 [31:11:16<39:05:04, 5.72s/it]
44%|████▍ | 19339/43939 [31:11:22<39:04:06, 5.72s/it]
44%|████▍ | 19340/43939 [31:11:27<39:00:17, 5.71s/it]
44%|████▍ | 19341/43939 [31:11:33<38:59:18, 5.71s/it]
44%|████▍ | 19342/43939 [31:11:39<38:54:08, 5.69s/it]
44%|████▍ | 19343/43939 [31:11:44<38:56:37, 5.70s/it]
44%|████▍ | 19344/43939 [31:11:50<38:59:43, 5.71s/it]
44%|████▍ | 19345/43939 [31:11:56<38:56:18, 5.70s/it]
44%|████▍ | 19346/43939 [31:12:01<38:54:27, 5.70s/it]
44%|████▍ | 19347/43939 [31:12:07<38:57:01, 5.70s/it]
44%|████▍ | 19348/43939 [31:12:13<39:26:04, 5.77s/it]
44%|████▍ | 19349/43939 [31:12:19<39:21:33, 5.76s/it]
44%|████▍ | 19350/43939 [31:12:25<39:14:51, 5.75s/it]
{'loss': 0.9233, 'grad_norm': 1.1700631380081177, 'learning_rate': 3.433495758522991e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.16, 'epoch': 1.76}
44%|████▍ | 19350/43939 [31:12:25<39:14:51, 5.75s/it]
44%|████▍ | 19351/43939 [31:12:30<39:09:47, 5.73s/it]
44%|████▍ | 19352/43939 [31:12:36<39:02:03, 5.72s/it]
44%|████▍ | 19353/43939 [31:12:42<38:57:07, 5.70s/it]
44%|████▍ | 19354/43939 [31:12:47<38:59:02, 5.71s/it]
44%|████▍ | 19355/43939 [31:12:53<38:55:57, 5.70s/it]
44%|████▍ | 19356/43939 [31:12:59<38:52:23, 5.69s/it]
44%|████▍ | 19357/43939 [31:13:04<38:52:58, 5.69s/it]
44%|████▍ | 19358/43939 [31:13:10<38:55:24, 5.70s/it]
44%|████▍ | 19359/43939 [31:13:16<38:56:33, 5.70s/it]
44%|████▍ | 19360/43939 [31:13:22<38:59:31, 5.71s/it]
44%|████▍ | 19361/43939 [31:13:27<38:53:39, 5.70s/it]
44%|████▍ | 19362/43939 [31:13:33<38:52:52, 5.70s/it]
44%|████▍ | 19363/43939 [31:13:39<38:51:47, 5.69s/it]
44%|████▍ | 19364/43939 [31:13:44<39:04:20, 5.72s/it]
44%|████▍ | 19365/43939 [31:13:50<38:53:45, 5.70s/it]
44%|████▍ | 19366/43939 [31:13:56<38:54:50, 5.70s/it]
44%|████▍ | 19367/43939 [31:14:01<38:52:45, 5.70s/it]
44%|████▍ | 19368/43939 [31:14:07<38:50:00, 5.69s/it]
44%|████▍ | 19369/43939 [31:14:13<39:18:07, 5.76s/it]
44%|████▍ | 19370/43939 [31:14:19<39:11:43, 5.74s/it]
44%|████▍ | 19371/43939 [31:14:24<39:06:44, 5.73s/it]
44%|████▍ | 19372/43939 [31:14:30<39:01:34, 5.72s/it]
44%|████▍ | 19373/43939 [31:14:36<38:59:54, 5.71s/it]
44%|████▍ | 19374/43939 [31:14:41<38:55:18, 5.70s/it]
44%|████▍ | 19375/43939 [31:14:47<38:54:07, 5.70s/it]
44%|████▍ | 19376/43939 [31:14:53<38:52:53, 5.70s/it]
44%|████▍ | 19377/43939 [31:14:59<38:54:08, 5.70s/it]
44%|████▍ | 19378/43939 [31:15:04<39:01:06, 5.72s/it]
44%|████▍ | 19379/43939 [31:15:10<39:03:41, 5.73s/it]
44%|████▍ | 19380/43939 [31:15:16<38:59:42, 5.72s/it]
44%|████▍ | 19381/43939 [31:15:21<38:59:27, 5.72s/it]
44%|████▍ | 19382/43939 [31:15:27<38:59:49, 5.72s/it]
44%|████▍ | 19383/43939 [31:15:33<39:05:40, 5.73s/it]
44%|████▍ | 19384/43939 [31:15:39<39:05:10, 5.73s/it]
44%|████▍ | 19385/43939 [31:15:44<39:01:59, 5.72s/it]
44%|████▍ | 19386/43939 [31:15:50<38:56:43, 5.71s/it]
44%|████▍ | 19387/43939 [31:15:56<38:55:58, 5.71s/it]
44%|████▍ | 19388/43939 [31:16:01<38:53:21, 5.70s/it]
44%|████▍ | 19389/43939 [31:16:07<38:50:41, 5.70s/it]
44%|████▍ | 19390/43939 [31:16:13<39:19:24, 5.77s/it]
44%|████▍ | 19391/43939 [31:16:19<39:17:08, 5.76s/it]
44%|████▍ | 19392/43939 [31:16:25<39:10:11, 5.74s/it]
44%|████▍ | 19393/43939 [31:16:30<39:02:33, 5.73s/it]
44%|████▍ | 19394/43939 [31:16:36<38:57:52, 5.71s/it]
44%|████▍ | 19395/43939 [31:16:42<38:55:57, 5.71s/it]
44%|████▍ | 19396/43939 [31:16:47<38:56:49, 5.71s/it]
44%|████▍ | 19397/43939 [31:16:53<38:54:18, 5.71s/it]
44%|████▍ | 19398/43939 [31:16:59<38:54:26, 5.71s/it]
44%|████▍ | 19399/43939 [31:17:04<38:55:19, 5.71s/it]
44%|████▍ | 19400/43939 [31:17:10<38:56:55, 5.71s/it]
{'loss': 0.8992, 'grad_norm': 1.190200924873352, 'learning_rate': 3.4242764731959806e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.31, 'epoch': 1.77}
44%|████▍ | 19400/43939 [31:17:10<38:56:55, 5.71s/it]
44%|████▍ | 19401/43939 [31:17:16<39:03:02, 5.73s/it]
44%|████▍ | 19402/43939 [31:17:22<38:58:44, 5.72s/it]
44%|████▍ | 19403/43939 [31:17:27<38:54:37, 5.71s/it]
44%|████▍ | 19404/43939 [31:17:33<38:56:50, 5.71s/it]
44%|████▍ | 19405/43939 [31:17:39<38:56:04, 5.71s/it]
44%|████▍ | 19406/43939 [31:17:44<38:51:14, 5.70s/it]
44%|████▍ | 19407/43939 [31:17:50<38:53:14, 5.71s/it]
44%|████▍ | 19408/43939 [31:17:56<38:53:41, 5.71s/it]
44%|████▍ | 19409/43939 [31:18:02<38:52:07, 5.70s/it]
44%|████▍ | 19410/43939 [31:18:07<38:52:56, 5.71s/it]
44%|████▍ | 19411/43939 [31:18:13<39:21:05, 5.78s/it]
44%|████▍ | 19412/43939 [31:18:19<39:16:16, 5.76s/it]
44%|████▍ | 19413/43939 [31:18:25<39:11:52, 5.75s/it]
44%|████▍ | 19414/43939 [31:18:30<39:02:55, 5.73s/it]
44%|████▍ | 19415/43939 [31:18:36<39:01:24, 5.73s/it]
44%|████▍ | 19416/43939 [31:18:42<38:57:51, 5.72s/it]
44%|████▍ | 19417/43939 [31:18:47<38:55:38, 5.71s/it]
44%|████▍ | 19418/43939 [31:18:53<38:57:49, 5.72s/it]
44%|████▍ | 19419/43939 [31:18:59<38:59:23, 5.72s/it]
44%|████▍ | 19420/43939 [31:19:05<38:58:54, 5.72s/it]
44%|████▍ | 19421/43939 [31:19:10<38:58:04, 5.72s/it]
44%|████▍ | 19422/43939 [31:19:16<38:57:00, 5.72s/it]
44%|████▍ | 19423/43939 [31:19:22<38:56:31, 5.72s/it]
44%|████▍ | 19424/43939 [31:19:27<38:52:03, 5.71s/it]
44%|████▍ | 19425/43939 [31:19:33<38:51:48, 5.71s/it]
44%|████▍ | 19426/43939 [31:19:39<38:53:39, 5.71s/it]
44%|████▍ | 19427/43939 [31:19:45<38:51:44, 5.71s/it]
44%|████▍ | 19428/43939 [31:19:50<38:48:25, 5.70s/it]
44%|████▍ | 19429/43939 [31:19:56<38:42:03, 5.68s/it]
44%|████▍ | 19430/43939 [31:20:02<38:46:12, 5.69s/it]
44%|████▍ | 19431/43939 [31:20:07<38:42:52, 5.69s/it]
44%|████▍ | 19432/43939 [31:20:13<39:15:40, 5.77s/it]
44%|████▍ | 19433/43939 [31:20:19<39:14:57, 5.77s/it]
44%|████▍ | 19434/43939 [31:20:25<39:07:34, 5.75s/it]
44%|████▍ | 19435/43939 [31:20:30<39:00:42, 5.73s/it]
44%|████▍ | 19436/43939 [31:20:36<38:58:42, 5.73s/it]
44%|████▍ | 19437/43939 [31:20:42<38:50:19, 5.71s/it]
44%|████▍ | 19438/43939 [31:20:48<38:49:05, 5.70s/it]
44%|████▍ | 19439/43939 [31:20:53<38:52:39, 5.71s/it]
44%|████▍ | 19440/43939 [31:20:59<38:47:59, 5.70s/it]
44%|████▍ | 19441/43939 [31:21:05<38:53:32, 5.72s/it]
44%|████▍ | 19442/43939 [31:21:10<38:54:17, 5.72s/it]
44%|████▍ | 19443/43939 [31:21:16<38:48:18, 5.70s/it]
44%|████▍ | 19444/43939 [31:21:22<38:48:12, 5.70s/it]
44%|████▍ | 19445/43939 [31:21:27<38:47:41, 5.70s/it]
44%|████▍ | 19446/43939 [31:21:33<38:45:35, 5.70s/it]
44%|████▍ | 19447/43939 [31:21:39<38:44:22, 5.69s/it]
44%|████▍ | 19448/43939 [31:21:45<38:45:46, 5.70s/it]
44%|████▍ | 19449/43939 [31:21:50<38:48:49, 5.71s/it]
44%|████▍ | 19450/43939 [31:21:56<38:52:34, 5.71s/it]
{'loss': 0.9174, 'grad_norm': 1.285638451576233, 'learning_rate': 3.415042605235585e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.07, 'epoch': 1.77}
44%|████▍ | 19450/43939 [31:21:56<38:52:34, 5.71s/it]
44%|████▍ | 19451/43939 [31:22:02<38:51:11, 5.71s/it]
44%|████▍ | 19452/43939 [31:22:07<38:49:19, 5.71s/it]
44%|████▍ | 19453/43939 [31:22:13<38:48:27, 5.71s/it]
44%|████▍ | 19454/43939 [31:22:19<39:14:06, 5.77s/it]
44%|████▍ | 19455/43939 [31:22:25<39:08:06, 5.75s/it]
44%|████▍ | 19456/43939 [31:22:30<39:04:07, 5.74s/it]
44%|████▍ | 19457/43939 [31:22:36<38:56:55, 5.73s/it]
44%|████▍ | 19458/43939 [31:22:42<38:53:03, 5.72s/it]
44%|████▍ | 19459/43939 [31:22:48<38:51:30, 5.71s/it]
44%|████▍ | 19460/43939 [31:22:53<38:48:56, 5.71s/it]
44%|████▍ | 19461/43939 [31:22:59<38:41:56, 5.69s/it]
44%|████▍ | 19462/43939 [31:23:05<38:48:38, 5.71s/it]
44%|████▍ | 19463/43939 [31:23:10<38:51:04, 5.71s/it]
44%|████▍ | 19464/43939 [31:23:16<38:56:18, 5.73s/it]
44%|████▍ | 19465/43939 [31:23:22<38:47:07, 5.71s/it]
44%|████▍ | 19466/43939 [31:23:27<38:43:11, 5.70s/it]
44%|████▍ | 19467/43939 [31:23:33<38:43:49, 5.70s/it]
44%|████▍ | 19468/43939 [31:23:39<38:43:12, 5.70s/it]
44%|████▍ | 19469/43939 [31:23:45<38:44:27, 5.70s/it]
44%|████▍ | 19470/43939 [31:23:50<38:42:53, 5.70s/it]
44%|████▍ | 19471/43939 [31:23:56<38:44:33, 5.70s/it]
44%|████▍ | 19472/43939 [31:24:02<38:46:15, 5.70s/it]
44%|████▍ | 19473/43939 [31:24:07<38:45:39, 5.70s/it]
44%|████▍ | 19474/43939 [31:24:13<38:49:21, 5.71s/it]
44%|████▍ | 19475/43939 [31:24:19<39:18:26, 5.78s/it]
44%|████▍ | 19476/43939 [31:24:25<39:09:00, 5.76s/it]
44%|████▍ | 19477/43939 [31:24:30<39:01:10, 5.74s/it]
44%|████▍ | 19478/43939 [31:24:36<38:57:33, 5.73s/it]
44%|████▍ | 19479/43939 [31:24:42<38:56:47, 5.73s/it]
44%|████▍ | 19480/43939 [31:24:48<38:55:58, 5.73s/it]
44%|████▍ | 19481/43939 [31:24:53<38:55:37, 5.73s/it]
44%|████▍ | 19482/43939 [31:24:59<38:51:40, 5.72s/it]
44%|████▍ | 19483/43939 [31:25:05<38:49:06, 5.71s/it]
44%|████▍ | 19484/43939 [31:25:10<38:41:11, 5.70s/it]
44%|████▍ | 19485/43939 [31:25:16<38:42:14, 5.70s/it]
44%|████▍ | 19486/43939 [31:25:22<38:43:56, 5.70s/it]
44%|████▍ | 19487/43939 [31:25:28<38:39:46, 5.69s/it]
44%|████▍ | 19488/43939 [31:25:33<38:44:26, 5.70s/it]
44%|████▍ | 19489/43939 [31:25:39<38:42:59, 5.70s/it]
44%|████▍ | 19490/43939 [31:25:45<38:44:56, 5.71s/it]
44%|████▍ | 19491/43939 [31:25:50<38:47:33, 5.71s/it]
44%|████▍ | 19492/43939 [31:25:56<38:44:21, 5.70s/it]
44%|████▍ | 19493/43939 [31:26:02<38:47:25, 5.71s/it]
44%|████▍ | 19494/43939 [31:26:08<38:48:26, 5.72s/it]
44%|████▍ | 19495/43939 [31:26:13<38:46:28, 5.71s/it]
44%|████▍ | 19496/43939 [31:26:19<39:08:11, 5.76s/it]
44%|████▍ | 19497/43939 [31:26:25<39:05:08, 5.76s/it]
44%|████▍ | 19498/43939 [31:26:31<39:02:27, 5.75s/it]
44%|████▍ | 19499/43939 [31:26:36<38:53:28, 5.73s/it]
44%|████▍ | 19500/43939 [31:26:42<38:48:14, 5.72s/it]
{'loss': 0.9081, 'grad_norm': 1.2392321825027466, 'learning_rate': 3.4057943003277715e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.25, 'epoch': 1.78}
44%|████▍ | 19500/43939 [31:26:42<38:48:14, 5.72s/it]
44%|████▍ | 19501/43939 [31:26:48<38:45:36, 5.71s/it]
44%|████▍ | 19502/43939 [31:26:53<38:46:03, 5.71s/it]
44%|████▍ | 19503/43939 [31:26:59<38:41:41, 5.70s/it]
44%|████▍ | 19504/43939 [31:27:05<38:39:31, 5.70s/it]
44%|████▍ | 19505/43939 [31:27:10<38:44:55, 5.71s/it]
44%|████▍ | 19506/43939 [31:27:16<38:42:28, 5.70s/it]
44%|████▍ | 19507/43939 [31:27:22<38:46:13, 5.71s/it]
44%|████▍ | 19508/43939 [31:27:28<38:46:20, 5.71s/it]
44%|████▍ | 19509/43939 [31:27:33<38:44:36, 5.71s/it]
44%|████▍ | 19510/43939 [31:27:39<38:38:38, 5.69s/it]
44%|████▍ | 19511/43939 [31:27:45<38:35:23, 5.69s/it]
44%|████▍ | 19512/43939 [31:27:50<38:37:51, 5.69s/it]
44%|████▍ | 19513/43939 [31:27:56<38:38:32, 5.70s/it]
44%|████▍ | 19514/43939 [31:28:02<38:36:39, 5.69s/it]
44%|████▍ | 19515/43939 [31:28:07<38:44:08, 5.71s/it]
44%|████▍ | 19516/43939 [31:28:13<38:44:35, 5.71s/it]
44%|████▍ | 19517/43939 [31:28:19<39:17:52, 5.79s/it]
44%|████▍ | 19518/43939 [31:28:25<39:07:35, 5.77s/it]
44%|████▍ | 19519/43939 [31:28:31<38:58:10, 5.74s/it]
44%|████▍ | 19520/43939 [31:28:36<38:52:23, 5.73s/it]
44%|████▍ | 19521/43939 [31:28:42<38:44:44, 5.71s/it]
44%|████▍ | 19522/43939 [31:28:48<38:44:57, 5.71s/it]
44%|████▍ | 19523/43939 [31:28:53<38:36:38, 5.69s/it]
44%|████▍ | 19524/43939 [31:28:59<38:37:28, 5.70s/it]
44%|████▍ | 19525/43939 [31:29:05<38:37:19, 5.70s/it]
44%|████▍ | 19526/43939 [31:29:10<38:37:35, 5.70s/it]
44%|████▍ | 19527/43939 [31:29:16<38:35:28, 5.69s/it]
44%|████▍ | 19528/43939 [31:29:22<38:38:41, 5.70s/it]
44%|████▍ | 19529/43939 [31:29:27<38:33:32, 5.69s/it]
44%|████▍ | 19530/43939 [31:29:33<38:34:55, 5.69s/it]
44%|████▍ | 19531/43939 [31:29:39<38:35:53, 5.69s/it]
44%|████▍ | 19532/43939 [31:29:45<38:42:23, 5.71s/it]
44%|████▍ | 19533/43939 [31:29:50<38:44:40, 5.71s/it]
44%|████▍ | 19534/43939 [31:29:56<38:42:36, 5.71s/it]
44%|████▍ | 19535/43939 [31:30:02<38:41:13, 5.71s/it]
44%|████▍ | 19536/43939 [31:30:07<38:36:57, 5.70s/it]
44%|████▍ | 19537/43939 [31:30:13<38:39:45, 5.70s/it]
44%|████▍ | 19538/43939 [31:30:19<38:44:34, 5.72s/it]
44%|████▍ | 19539/43939 [31:30:25<39:20:00, 5.80s/it]
44%|████▍ | 19540/43939 [31:30:31<39:07:44, 5.77s/it]
44%|████▍ | 19541/43939 [31:30:36<39:01:14, 5.76s/it]
44%|████▍ | 19542/43939 [31:30:42<38:57:52, 5.75s/it]
44%|████▍ | 19543/43939 [31:30:48<38:53:33, 5.74s/it]
44%|████▍ | 19544/43939 [31:30:54<38:58:14, 5.75s/it]
44%|████▍ | 19545/43939 [31:30:59<38:51:12, 5.73s/it]
44%|████▍ | 19546/43939 [31:31:05<38:47:37, 5.73s/it]
44%|████▍ | 19547/43939 [31:31:11<38:41:31, 5.71s/it]
44%|████▍ | 19548/43939 [31:31:16<38:35:19, 5.70s/it]
44%|████▍ | 19549/43939 [31:31:22<38:40:16, 5.71s/it]
44%|████▍ | 19550/43939 [31:31:28<38:38:33, 5.70s/it]
{'loss': 0.9154, 'grad_norm': 1.3192247152328491, 'learning_rate': 3.396531704386282e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.57, 'epoch': 1.78}
44%|████▍ | 19550/43939 [31:31:28<38:38:33, 5.70s/it]
44%|████▍ | 19551/43939 [31:31:33<38:41:33, 5.71s/it]
44%|████▍ | 19552/43939 [31:31:39<38:37:14, 5.70s/it]
45%|████▍ | 19553/43939 [31:31:45<38:35:35, 5.70s/it]
45%|████▍ | 19554/43939 [31:31:50<38:33:44, 5.69s/it]
45%|████▍ | 19555/43939 [31:31:56<38:32:06, 5.69s/it]
45%|████▍ | 19556/43939 [31:32:02<38:36:56, 5.70s/it]
45%|████▍ | 19557/43939 [31:32:08<38:40:19, 5.71s/it]
45%|████▍ | 19558/43939 [31:32:13<38:52:08, 5.74s/it]
45%|████▍ | 19559/43939 [31:32:19<38:52:33, 5.74s/it]
45%|████▍ | 19560/43939 [31:32:25<39:18:26, 5.80s/it]
45%|████▍ | 19561/43939 [31:32:31<39:07:50, 5.78s/it]
45%|████▍ | 19562/43939 [31:32:37<39:01:35, 5.76s/it]
45%|████▍ | 19563/43939 [31:32:42<39:01:14, 5.76s/it]
45%|████▍ | 19564/43939 [31:32:48<38:50:59, 5.74s/it]
45%|████▍ | 19565/43939 [31:32:54<38:46:00, 5.73s/it]
45%|████▍ | 19566/43939 [31:32:59<38:47:11, 5.73s/it]
45%|████▍ | 19567/43939 [31:33:05<38:38:55, 5.71s/it]
45%|████▍ | 19568/43939 [31:33:11<38:38:03, 5.71s/it]
45%|████▍ | 19569/43939 [31:33:16<38:36:57, 5.70s/it]
45%|████▍ | 19570/43939 [31:33:22<38:36:59, 5.70s/it]
45%|████▍ | 19571/43939 [31:33:28<38:34:47, 5.70s/it]
45%|████▍ | 19572/43939 [31:33:34<38:32:12, 5.69s/it]
45%|████▍ | 19573/43939 [31:33:39<38:32:09, 5.69s/it]
45%|████▍ | 19574/43939 [31:33:45<38:35:04, 5.70s/it]
45%|████▍ | 19575/43939 [31:33:51<38:36:11, 5.70s/it]
45%|████▍ | 19576/43939 [31:33:56<38:39:15, 5.71s/it]
45%|████▍ | 19577/43939 [31:34:02<38:44:34, 5.73s/it]
45%|████▍ | 19578/43939 [31:34:08<38:38:15, 5.71s/it]
45%|████▍ | 19579/43939 [31:34:14<38:36:48, 5.71s/it]
45%|████▍ | 19580/43939 [31:34:19<38:39:28, 5.71s/it]
45%|████▍ | 19581/43939 [31:34:25<39:14:12, 5.80s/it]
45%|████▍ | 19582/43939 [31:34:31<38:59:49, 5.76s/it]
45%|████▍ | 19583/43939 [31:34:37<38:56:38, 5.76s/it]
45%|████▍ | 19584/43939 [31:34:42<38:48:45, 5.74s/it]
45%|████▍ | 19585/43939 [31:34:48<38:41:49, 5.72s/it]
45%|████▍ | 19586/43939 [31:34:54<38:40:41, 5.72s/it]
45%|████▍ | 19587/43939 [31:34:59<38:33:39, 5.70s/it]
45%|████▍ | 19588/43939 [31:35:05<38:37:24, 5.71s/it]
45%|████▍ | 19589/43939 [31:35:11<38:38:36, 5.71s/it]
45%|████▍ | 19590/43939 [31:35:17<38:36:11, 5.71s/it]
45%|████▍ | 19591/43939 [31:35:22<38:39:30, 5.72s/it]
45%|████▍ | 19592/43939 [31:35:28<38:33:43, 5.70s/it]
45%|████▍ | 19593/43939 [31:35:34<38:32:12, 5.70s/it]
45%|████▍ | 19594/43939 [31:35:39<38:41:23, 5.72s/it]
45%|████▍ | 19595/43939 [31:35:45<38:40:31, 5.72s/it]
45%|████▍ | 19596/43939 [31:35:51<38:36:59, 5.71s/it]
45%|████▍ | 19597/43939 [31:35:57<38:36:41, 5.71s/it]
45%|████▍ | 19598/43939 [31:36:02<38:37:12, 5.71s/it]
45%|████▍ | 19599/43939 [31:36:08<38:32:15, 5.70s/it]
45%|████▍ | 19600/43939 [31:36:14<38:29:55, 5.69s/it]
{'loss': 0.9097, 'grad_norm': 1.1055296659469604, 'learning_rate': 3.387254963550337e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.22, 'epoch': 1.78}
45%|████▍ | 19600/43939 [31:36:14<38:29:55, 5.69s/it]
45%|████▍ | 19601/43939 [31:36:19<38:30:56, 5.70s/it]
45%|████▍ | 19602/43939 [31:36:25<38:55:28, 5.76s/it]
45%|████▍ | 19603/43939 [31:36:31<38:48:39, 5.74s/it]
45%|████▍ | 19604/43939 [31:36:37<38:39:50, 5.72s/it]
45%|████▍ | 19605/43939 [31:36:42<38:39:45, 5.72s/it]
45%|████▍ | 19606/43939 [31:36:48<38:35:27, 5.71s/it]
45%|████▍ | 19607/43939 [31:36:54<38:33:28, 5.70s/it]
45%|████▍ | 19608/43939 [31:36:59<38:40:15, 5.72s/it]
45%|████▍ | 19609/43939 [31:37:05<38:37:48, 5.72s/it]
45%|████▍ | 19610/43939 [31:37:11<38:37:57, 5.72s/it]
45%|████▍ | 19611/43939 [31:37:17<38:44:17, 5.73s/it]
45%|████▍ | 19612/43939 [31:37:22<38:40:34, 5.72s/it]
45%|████▍ | 19613/43939 [31:37:28<38:43:53, 5.73s/it]
45%|████▍ | 19614/43939 [31:37:34<38:38:56, 5.72s/it]
45%|████▍ | 19615/43939 [31:37:39<38:35:46, 5.71s/it]
45%|████▍ | 19616/43939 [31:37:45<38:45:34, 5.74s/it]
45%|████▍ | 19617/43939 [31:37:51<38:39:19, 5.72s/it]
45%|████▍ | 19618/43939 [31:37:57<38:42:15, 5.73s/it]
45%|████▍ | 19619/43939 [31:38:02<38:37:31, 5.72s/it]
45%|████▍ | 19620/43939 [31:38:08<38:37:24, 5.72s/it]
45%|████▍ | 19621/43939 [31:38:14<38:39:51, 5.72s/it]
45%|████▍ | 19622/43939 [31:38:20<38:41:48, 5.73s/it]
45%|████▍ | 19623/43939 [31:38:25<39:00:57, 5.78s/it]
45%|████▍ | 19624/43939 [31:38:31<38:53:25, 5.76s/it]
45%|████▍ | 19625/43939 [31:38:37<38:47:13, 5.74s/it]
45%|████▍ | 19626/43939 [31:38:43<38:46:06, 5.74s/it]
45%|████▍ | 19627/43939 [31:38:48<38:45:50, 5.74s/it]
45%|████▍ | 19628/43939 [31:38:54<38:41:21, 5.73s/it]
45%|████▍ | 19629/43939 [31:39:00<38:38:38, 5.72s/it]
45%|████▍ | 19630/43939 [31:39:06<38:41:55, 5.73s/it]
45%|████▍ | 19631/43939 [31:39:11<38:37:39, 5.72s/it]
45%|████▍ | 19632/43939 [31:39:17<38:37:56, 5.72s/it]
45%|████▍ | 19633/43939 [31:39:23<38:35:20, 5.72s/it]
45%|████▍ | 19634/43939 [31:39:28<38:32:45, 5.71s/it]
45%|████▍ | 19635/43939 [31:39:34<38:33:21, 5.71s/it]
45%|████▍ | 19636/43939 [31:39:40<38:32:36, 5.71s/it]
45%|████▍ | 19637/43939 [31:39:45<38:29:01, 5.70s/it]
45%|████▍ | 19638/43939 [31:39:51<38:39:15, 5.73s/it]
45%|████▍ | 19639/43939 [31:39:57<38:37:29, 5.72s/it]
45%|████▍ | 19640/43939 [31:40:03<38:30:48, 5.71s/it]
45%|████▍ | 19641/43939 [31:40:08<38:28:24, 5.70s/it]
45%|████▍ | 19642/43939 [31:40:14<38:27:06, 5.70s/it]
45%|████▍ | 19643/43939 [31:40:20<38:34:10, 5.71s/it]
45%|████▍ | 19644/43939 [31:40:26<39:01:57, 5.78s/it]
45%|████▍ | 19645/43939 [31:40:31<38:53:30, 5.76s/it]
45%|████▍ | 19646/43939 [31:40:37<38:43:59, 5.74s/it]
45%|████▍ | 19647/43939 [31:40:43<38:41:04, 5.73s/it]
45%|████▍ | 19648/43939 [31:40:49<38:34:37, 5.72s/it]
45%|████▍ | 19649/43939 [31:40:54<38:40:09, 5.73s/it]
45%|████▍ | 19650/43939 [31:41:00<38:37:25, 5.72s/it]
{'loss': 0.9446, 'grad_norm': 1.2381501197814941, 'learning_rate': 3.37796422418232e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 540.49, 'epoch': 1.79}
45%|████▍ | 19650/43939 [31:41:00<38:37:25, 5.72s/it]
45%|████▍ | 19651/43939 [31:41:06<38:29:30, 5.71s/it]
45%|████▍ | 19652/43939 [31:41:11<38:28:12, 5.70s/it]
45%|████▍ | 19653/43939 [31:41:17<38:26:08, 5.70s/it]
45%|████▍ | 19654/43939 [31:41:23<38:25:10, 5.70s/it]
45%|████▍ | 19655/43939 [31:41:28<38:24:08, 5.69s/it]
45%|████▍ | 19656/43939 [31:41:34<38:21:14, 5.69s/it]
45%|████▍ | 19657/43939 [31:41:40<38:21:59, 5.69s/it]
45%|████▍ | 19658/43939 [31:41:45<38:24:35, 5.69s/it]
45%|████▍ | 19659/43939 [31:41:51<38:21:33, 5.69s/it]
45%|████▍ | 19660/43939 [31:41:57<38:24:58, 5.70s/it]
45%|████▍ | 19661/43939 [31:42:03<38:23:30, 5.69s/it]
45%|████▍ | 19662/43939 [31:42:08<38:24:20, 5.70s/it]
45%|████▍ | 19663/43939 [31:42:14<38:24:05, 5.69s/it]
45%|████▍ | 19664/43939 [31:42:20<38:22:35, 5.69s/it]
45%|████▍ | 19665/43939 [31:42:26<38:49:39, 5.76s/it]
45%|████▍ | 19666/43939 [31:42:31<38:45:16, 5.75s/it]
45%|████▍ | 19667/43939 [31:42:37<38:37:53, 5.73s/it]
45%|████▍ | 19668/43939 [31:42:43<38:36:27, 5.73s/it]
45%|████▍ | 19669/43939 [31:42:48<38:32:46, 5.72s/it]
45%|████▍ | 19670/43939 [31:42:54<38:27:45, 5.71s/it]
45%|████▍ | 19671/43939 [31:43:00<38:28:08, 5.71s/it]
45%|████▍ | 19672/43939 [31:43:05<38:25:36, 5.70s/it]
45%|████▍ | 19673/43939 [31:43:11<38:22:06, 5.69s/it]
45%|████▍ | 19674/43939 [31:43:17<38:34:27, 5.72s/it]
45%|████▍ | 19675/43939 [31:43:23<38:33:17, 5.72s/it]
45%|████▍ | 19676/43939 [31:43:28<38:42:06, 5.74s/it]
45%|████▍ | 19677/43939 [31:43:34<38:35:52, 5.73s/it]
45%|████▍ | 19678/43939 [31:43:40<38:28:00, 5.71s/it]
45%|████▍ | 19679/43939 [31:43:45<38:27:47, 5.71s/it]
45%|████▍ | 19680/43939 [31:43:51<38:21:41, 5.69s/it]
45%|████▍ | 19681/43939 [31:43:57<38:24:32, 5.70s/it]
45%|████▍ | 19682/43939 [31:44:03<38:23:31, 5.70s/it]
45%|████▍ | 19683/43939 [31:44:08<38:29:07, 5.71s/it]
45%|████▍ | 19684/43939 [31:44:14<38:21:15, 5.69s/it]
45%|████▍ | 19685/43939 [31:44:20<38:18:27, 5.69s/it]
45%|████▍ | 19686/43939 [31:44:26<38:51:46, 5.77s/it]
45%|████▍ | 19687/43939 [31:44:31<38:43:05, 5.75s/it]
45%|████▍ | 19688/43939 [31:44:37<38:44:25, 5.75s/it]
45%|████▍ | 19689/43939 [31:44:43<38:34:01, 5.73s/it]
45%|████▍ | 19690/43939 [31:44:48<38:24:32, 5.70s/it]
45%|████▍ | 19691/43939 [31:44:54<38:22:38, 5.70s/it]
45%|████▍ | 19692/43939 [31:45:00<38:26:08, 5.71s/it]
45%|████▍ | 19693/43939 [31:45:05<38:23:59, 5.70s/it]
45%|████▍ | 19694/43939 [31:45:11<38:29:53, 5.72s/it]
45%|████▍ | 19695/43939 [31:45:17<38:27:28, 5.71s/it]
45%|████▍ | 19696/43939 [31:45:23<38:24:39, 5.70s/it]
45%|████▍ | 19697/43939 [31:45:28<38:21:25, 5.70s/it]
45%|████▍ | 19698/43939 [31:45:34<38:16:45, 5.68s/it]
45%|████▍ | 19699/43939 [31:45:40<38:15:14, 5.68s/it]
45%|████▍ | 19700/43939 [31:45:45<38:19:47, 5.69s/it]
{'loss': 0.8913, 'grad_norm': 1.1021363735198975, 'learning_rate': 3.36865963286548e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.57, 'epoch': 1.79}
45%|████▍ | 19700/43939 [31:45:45<38:19:47, 5.69s/it]
45%|████▍ | 19701/43939 [31:45:51<38:22:41, 5.70s/it]
45%|████▍ | 19702/43939 [31:45:57<38:27:36, 5.71s/it]
45%|████▍ | 19703/43939 [31:46:02<38:21:41, 5.70s/it]
45%|████▍ | 19704/43939 [31:46:08<38:23:09, 5.70s/it]
45%|████▍ | 19705/43939 [31:46:14<38:23:15, 5.70s/it]
45%|████▍ | 19706/43939 [31:46:20<38:22:14, 5.70s/it]
45%|████▍ | 19707/43939 [31:46:25<38:51:25, 5.77s/it]
45%|████▍ | 19708/43939 [31:46:31<38:42:03, 5.75s/it]
45%|████▍ | 19709/43939 [31:46:37<38:33:59, 5.73s/it]
45%|████▍ | 19710/43939 [31:46:43<38:27:22, 5.71s/it]
45%|████▍ | 19711/43939 [31:46:48<38:28:00, 5.72s/it]
45%|████▍ | 19712/43939 [31:46:54<38:26:40, 5.71s/it]
45%|████▍ | 19713/43939 [31:47:00<38:24:18, 5.71s/it]
45%|████▍ | 19714/43939 [31:47:05<38:16:33, 5.69s/it]
45%|████▍ | 19715/43939 [31:47:11<38:20:00, 5.70s/it]
45%|████▍ | 19716/43939 [31:47:17<38:20:30, 5.70s/it]
45%|████▍ | 19717/43939 [31:47:22<38:25:59, 5.71s/it]
45%|████▍ | 19718/43939 [31:47:28<38:21:37, 5.70s/it]
45%|████▍ | 19719/43939 [31:47:34<38:25:59, 5.71s/it]
45%|████▍ | 19720/43939 [31:47:40<38:22:08, 5.70s/it]
45%|████▍ | 19721/43939 [31:47:45<38:17:36, 5.69s/it]
45%|████▍ | 19722/43939 [31:47:51<38:16:10, 5.69s/it]
45%|████▍ | 19723/43939 [31:47:57<38:20:17, 5.70s/it]
45%|████▍ | 19724/43939 [31:48:02<38:21:48, 5.70s/it]
45%|████▍ | 19725/43939 [31:48:08<38:21:40, 5.70s/it]
45%|████▍ | 19726/43939 [31:48:14<38:25:49, 5.71s/it]
45%|████▍ | 19727/43939 [31:48:19<38:20:35, 5.70s/it]
45%|████▍ | 19728/43939 [31:48:25<38:55:10, 5.79s/it]
45%|████▍ | 19729/43939 [31:48:31<38:42:43, 5.76s/it]
45%|████▍ | 19730/43939 [31:48:37<38:32:32, 5.73s/it]
45%|████▍ | 19731/43939 [31:48:43<38:29:05, 5.72s/it]
45%|████▍ | 19732/43939 [31:48:48<38:31:12, 5.73s/it]
45%|████▍ | 19733/43939 [31:48:54<38:24:38, 5.71s/it]
45%|████▍ | 19734/43939 [31:49:00<38:19:47, 5.70s/it]
45%|████▍ | 19735/43939 [31:49:05<38:19:38, 5.70s/it]
45%|████▍ | 19736/43939 [31:49:11<38:17:34, 5.70s/it]
45%|████▍ | 19737/43939 [31:49:17<38:14:27, 5.69s/it]
45%|████▍ | 19738/43939 [31:49:22<38:18:18, 5.70s/it]
45%|████▍ | 19739/43939 [31:49:28<38:18:52, 5.70s/it]
45%|████▍ | 19740/43939 [31:49:34<38:20:51, 5.70s/it]
45%|████▍ | 19741/43939 [31:49:40<38:25:17, 5.72s/it]
45%|████▍ | 19742/43939 [31:49:45<38:29:09, 5.73s/it]
45%|████▍ | 19743/43939 [31:49:51<38:27:16, 5.72s/it]
45%|████▍ | 19744/43939 [31:49:57<38:26:38, 5.72s/it]
45%|████▍ | 19745/43939 [31:50:03<38:34:10, 5.74s/it]
45%|████▍ | 19746/43939 [31:50:08<38:27:40, 5.72s/it]
45%|████▍ | 19747/43939 [31:50:14<38:21:49, 5.71s/it]
45%|████▍ | 19748/43939 [31:50:20<38:20:43, 5.71s/it]
45%|████▍ | 19749/43939 [31:50:25<38:19:54, 5.70s/it]
45%|████▍ | 19750/43939 [31:50:31<38:54:56, 5.79s/it]
{'loss': 0.9434, 'grad_norm': 1.132875680923462, 'learning_rate': 3.3593413364016094e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.41, 'epoch': 1.8}
45%|████▍ | 19750/43939 [31:50:31<38:54:56, 5.79s/it]
45%|████▍ | 19751/43939 [31:50:37<38:44:01, 5.76s/it]
45%|████▍ | 19752/43939 [31:50:43<38:35:38, 5.74s/it]
45%|████▍ | 19753/43939 [31:50:48<38:30:01, 5.73s/it]
45%|████▍ | 19754/43939 [31:50:54<38:33:07, 5.74s/it]
45%|████▍ | 19755/43939 [31:51:00<38:27:29, 5.72s/it]
45%|████▍ | 19756/43939 [31:51:06<38:24:38, 5.72s/it]
45%|████▍ | 19757/43939 [31:51:11<38:24:45, 5.72s/it]
45%|████▍ | 19758/43939 [31:51:17<38:23:13, 5.71s/it]
45%|████▍ | 19759/43939 [31:51:23<38:24:35, 5.72s/it]
45%|████▍ | 19760/43939 [31:51:28<38:22:33, 5.71s/it]
45%|████▍ | 19761/43939 [31:51:34<38:25:44, 5.72s/it]
45%|████▍ | 19762/43939 [31:51:40<38:20:08, 5.71s/it]
45%|████▍ | 19763/43939 [31:51:45<38:14:58, 5.70s/it]
45%|████▍ | 19764/43939 [31:51:51<38:12:10, 5.69s/it]
45%|████▍ | 19765/43939 [31:51:57<38:12:24, 5.69s/it]
45%|████▍ | 19766/43939 [31:52:02<38:10:36, 5.69s/it]
45%|████▍ | 19767/43939 [31:52:08<38:16:50, 5.70s/it]
45%|████▍ | 19768/43939 [31:52:14<38:15:41, 5.70s/it]
45%|████▍ | 19769/43939 [31:52:20<38:16:33, 5.70s/it]
45%|████▍ | 19770/43939 [31:52:25<38:13:08, 5.69s/it]
45%|████▍ | 19771/43939 [31:52:31<38:44:31, 5.77s/it]
45%|████▍ | 19772/43939 [31:52:37<38:41:59, 5.76s/it]
45%|████▌ | 19773/43939 [31:52:43<38:37:12, 5.75s/it][2025-11-30 01:56:56,550] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.94s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:48, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.69s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.72s/it][A
19%|█▉ | 14/72 [00:35<02:38, 2.73s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:41<02:32, 2.73s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:25, 2.74s/it][A
28%|██▊ | 20/72 [00:52<02:22, 2.74s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:17, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.74s/it][A
38%|███▊ | 27/72 [01:11<02:03, 2.74s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.74s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:55, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:33<01:41, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:44<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.74s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.74s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.74s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:52, 2.74s/it][A
75%|███████▌ | 54/72 [02:25<00:49, 2.74s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.74s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.74s/it][A
81%|████████ | 58/72 [02:36<00:38, 2.74s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.74s/it][A
96%|█████████▌| 69/72 [03:06<00:08, 2.74s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.74s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9705602526664734, 'eval_runtime': 196.6587, 'eval_samples_per_second': 5.08, 'eval_steps_per_second': 0.366, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 1.8}
45%|████▌ | 19773/43939 [31:55:59<38:37:12, 5.75s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
45%|████▌ | 19774/43939 [31:56:05<434:49:27, 64.78s/it]
45%|████▌ | 19775/43939 [31:56:11<315:45:51, 47.04s/it]
45%|████▌ | 19776/43939 [31:56:17<232:28:36, 34.64s/it]
45%|████▌ | 19777/43939 [31:56:22<174:15:34, 25.96s/it]
45%|████▌ | 19778/43939 [31:56:28<133:24:56, 19.88s/it]
45%|████▌ | 19779/43939 [31:56:34<104:48:23, 15.62s/it]
45%|████▌ | 19780/43939 [31:56:39<84:50:32, 12.64s/it]
45%|████▌ | 19781/43939 [31:56:45<70:58:07, 10.58s/it]
45%|████▌ | 19782/43939 [31:56:51<61:03:50, 9.10s/it]
45%|████▌ | 19783/43939 [31:56:56<54:08:26, 8.07s/it]
45%|████▌ | 19784/43939 [31:57:02<49:24:13, 7.36s/it]
45%|████▌ | 19785/43939 [31:57:08<46:01:06, 6.86s/it]
45%|████▌ | 19786/43939 [31:57:14<43:43:07, 6.52s/it]
45%|████▌ | 19787/43939 [31:57:19<41:57:45, 6.25s/it]
45%|████▌ | 19788/43939 [31:57:25<40:55:04, 6.10s/it]
45%|████▌ | 19789/43939 [31:57:31<40:09:10, 5.99s/it]
45%|████▌ | 19790/43939 [31:57:36<39:34:28, 5.90s/it]
45%|████▌ | 19791/43939 [31:57:42<39:13:19, 5.85s/it]
45%|████▌ | 19792/43939 [31:57:48<39:36:10, 5.90s/it]
45%|████▌ | 19793/43939 [31:57:54<39:11:01, 5.84s/it]
45%|████▌ | 19794/43939 [31:58:00<38:52:22, 5.80s/it]
45%|████▌ | 19795/43939 [31:58:05<38:41:13, 5.77s/it]
45%|████▌ | 19796/43939 [31:58:11<38:34:37, 5.75s/it]
45%|████▌ | 19797/43939 [31:58:17<38:26:13, 5.73s/it]
45%|████▌ | 19798/43939 [31:58:22<38:23:44, 5.73s/it]
45%|████▌ | 19799/43939 [31:58:28<38:21:41, 5.72s/it]
45%|████▌ | 19800/43939 [31:58:34<38:19:25, 5.72s/it]
{'loss': 0.9099, 'grad_norm': 1.2874317169189453, 'learning_rate': 3.3500094818087316e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 288.13, 'epoch': 1.8}
45%|████▌ | 19800/43939 [31:58:34<38:19:25, 5.72s/it]
45%|████▌ | 19801/43939 [31:58:39<38:17:06, 5.71s/it]
45%|████▌ | 19802/43939 [31:58:45<38:14:00, 5.70s/it]
45%|████▌ | 19803/43939 [31:58:51<38:12:23, 5.70s/it]
45%|████▌ | 19804/43939 [31:58:57<38:14:18, 5.70s/it]
45%|████▌ | 19805/43939 [31:59:02<38:19:26, 5.72s/it]
45%|████▌ | 19806/43939 [31:59:08<38:16:51, 5.71s/it]
45%|████▌ | 19807/43939 [31:59:14<38:16:43, 5.71s/it]
45%|████▌ | 19808/43939 [31:59:19<38:12:28, 5.70s/it]
45%|████▌ | 19809/43939 [31:59:25<38:09:53, 5.69s/it]
45%|████▌ | 19810/43939 [31:59:31<38:09:09, 5.69s/it]
45%|████▌ | 19811/43939 [31:59:36<38:11:32, 5.70s/it]
45%|████▌ | 19812/43939 [31:59:42<38:11:44, 5.70s/it]
45%|████▌ | 19813/43939 [31:59:48<38:37:45, 5.76s/it]
45%|████▌ | 19814/43939 [31:59:54<38:30:36, 5.75s/it]
45%|████▌ | 19815/43939 [31:59:59<38:23:09, 5.73s/it]
45%|████▌ | 19816/43939 [32:00:05<38:19:43, 5.72s/it]
45%|████▌ | 19817/43939 [32:00:11<38:21:00, 5.72s/it]
45%|████▌ | 19818/43939 [32:00:17<38:12:45, 5.70s/it]
45%|████▌ | 19819/43939 [32:00:22<38:12:02, 5.70s/it]
45%|████▌ | 19820/43939 [32:00:28<38:13:46, 5.71s/it]
45%|████▌ | 19821/43939 [32:00:34<38:16:50, 5.71s/it]
45%|████▌ | 19822/43939 [32:00:39<38:12:56, 5.70s/it]
45%|████▌ | 19823/43939 [32:00:45<38:18:36, 5.72s/it]
45%|████▌ | 19824/43939 [32:00:51<38:12:47, 5.70s/it]
45%|████▌ | 19825/43939 [32:00:57<38:15:30, 5.71s/it]
45%|████▌ | 19826/43939 [32:01:02<38:13:08, 5.71s/it]
45%|████▌ | 19827/43939 [32:01:08<38:12:09, 5.70s/it]
45%|████▌ | 19828/43939 [32:01:14<38:12:39, 5.71s/it]
45%|████▌ | 19829/43939 [32:01:19<38:13:37, 5.71s/it]
45%|████▌ | 19830/43939 [32:01:25<38:15:22, 5.71s/it]
45%|████▌ | 19831/43939 [32:01:31<38:10:24, 5.70s/it]
45%|████▌ | 19832/43939 [32:01:36<38:13:22, 5.71s/it]
45%|████▌ | 19833/43939 [32:01:42<38:10:24, 5.70s/it]
45%|████▌ | 19834/43939 [32:01:48<38:07:21, 5.69s/it]
45%|████▌ | 19835/43939 [32:01:54<38:33:03, 5.76s/it]
45%|████▌ | 19836/43939 [32:01:59<38:33:14, 5.76s/it]
45%|████▌ | 19837/43939 [32:02:05<38:28:47, 5.75s/it]
45%|████▌ | 19838/43939 [32:02:11<38:22:54, 5.73s/it]
45%|████▌ | 19839/43939 [32:02:17<38:15:56, 5.72s/it]
45%|████▌ | 19840/43939 [32:02:22<38:09:46, 5.70s/it]
45%|████▌ | 19841/43939 [32:02:28<38:08:22, 5.70s/it]
45%|████▌ | 19842/43939 [32:02:34<38:09:30, 5.70s/it]
45%|████▌ | 19843/43939 [32:02:39<38:08:02, 5.70s/it]
45%|████▌ | 19844/43939 [32:02:45<38:08:25, 5.70s/it]
45%|████▌ | 19845/43939 [32:02:51<38:11:31, 5.71s/it]
45%|████▌ | 19846/43939 [32:02:56<38:11:27, 5.71s/it]
45%|████▌ | 19847/43939 [32:03:02<38:10:20, 5.70s/it]
45%|████▌ | 19848/43939 [32:03:08<38:09:05, 5.70s/it]
45%|████▌ | 19849/43939 [32:03:14<38:11:33, 5.71s/it]
45%|████▌ | 19850/43939 [32:03:19<38:17:00, 5.72s/it]
{'loss': 0.9277, 'grad_norm': 1.1182738542556763, 'learning_rate': 3.340664216318784e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.5, 'epoch': 1.81}
45%|████▌ | 19850/43939 [32:03:19<38:17:00, 5.72s/it]
45%|████▌ | 19851/43939 [32:03:25<38:20:04, 5.73s/it]
45%|████▌ | 19852/43939 [32:03:31<38:12:36, 5.71s/it]
45%|████▌ | 19853/43939 [32:03:36<38:08:22, 5.70s/it]
45%|████▌ | 19854/43939 [32:03:42<38:17:57, 5.72s/it]
45%|████▌ | 19855/43939 [32:03:48<38:13:44, 5.71s/it]
45%|████▌ | 19856/43939 [32:03:54<38:41:04, 5.78s/it]
45%|████▌ | 19857/43939 [32:04:00<38:32:57, 5.76s/it]
45%|████▌ | 19858/43939 [32:04:05<38:30:28, 5.76s/it]
45%|████▌ | 19859/43939 [32:04:11<38:17:29, 5.72s/it]
45%|████▌ | 19860/43939 [32:04:17<38:12:21, 5.71s/it]
45%|████▌ | 19861/43939 [32:04:22<38:06:03, 5.70s/it]
45%|████▌ | 19862/43939 [32:04:28<38:06:34, 5.70s/it]
45%|████▌ | 19863/43939 [32:04:34<38:12:25, 5.71s/it]
45%|████▌ | 19864/43939 [32:04:39<38:13:59, 5.72s/it]
45%|████▌ | 19865/43939 [32:04:45<38:09:03, 5.71s/it]
45%|████▌ | 19866/43939 [32:04:51<38:11:12, 5.71s/it]
45%|████▌ | 19867/43939 [32:04:57<38:09:23, 5.71s/it]
45%|████▌ | 19868/43939 [32:05:02<38:09:11, 5.71s/it]
45%|████▌ | 19869/43939 [32:05:08<38:11:10, 5.71s/it]
45%|████▌ | 19870/43939 [32:05:14<38:07:11, 5.70s/it]
45%|████▌ | 19871/43939 [32:05:19<38:05:20, 5.70s/it]
45%|████▌ | 19872/43939 [32:05:25<38:05:36, 5.70s/it]
45%|████▌ | 19873/43939 [32:05:31<38:08:33, 5.71s/it]
45%|████▌ | 19874/43939 [32:05:36<38:11:16, 5.71s/it]
45%|████▌ | 19875/43939 [32:05:42<38:08:00, 5.70s/it]
45%|████▌ | 19876/43939 [32:05:48<38:10:17, 5.71s/it]
45%|████▌ | 19877/43939 [32:05:54<38:39:31, 5.78s/it]
45%|████▌ | 19878/43939 [32:06:00<38:32:00, 5.77s/it]
45%|████▌ | 19879/43939 [32:06:05<38:22:52, 5.74s/it]
45%|████▌ | 19880/43939 [32:06:11<38:17:13, 5.73s/it]
45%|████▌ | 19881/43939 [32:06:17<38:10:30, 5.71s/it]
45%|████▌ | 19882/43939 [32:06:22<38:11:02, 5.71s/it]
45%|████▌ | 19883/43939 [32:06:28<38:10:05, 5.71s/it]
45%|████▌ | 19884/43939 [32:06:34<38:05:24, 5.70s/it]
45%|████▌ | 19885/43939 [32:06:39<38:04:17, 5.70s/it]
45%|████▌ | 19886/43939 [32:06:45<38:01:28, 5.69s/it]
45%|████▌ | 19887/43939 [32:06:51<38:01:29, 5.69s/it]
45%|████▌ | 19888/43939 [32:06:57<38:02:52, 5.70s/it]
45%|████▌ | 19889/43939 [32:07:02<38:06:18, 5.70s/it]
45%|████▌ | 19890/43939 [32:07:08<38:09:37, 5.71s/it]
45%|████▌ | 19891/43939 [32:07:14<38:09:32, 5.71s/it]
45%|████▌ | 19892/43939 [32:07:19<38:07:49, 5.71s/it]
45%|████▌ | 19893/43939 [32:07:25<38:07:58, 5.71s/it]
45%|████▌ | 19894/43939 [32:07:31<38:14:03, 5.72s/it]
45%|████▌ | 19895/43939 [32:07:37<38:12:38, 5.72s/it]
45%|████▌ | 19896/43939 [32:07:42<38:10:39, 5.72s/it]
45%|████▌ | 19897/43939 [32:07:48<38:14:23, 5.73s/it]
45%|████▌ | 19898/43939 [32:07:54<38:40:27, 5.79s/it]
45%|████▌ | 19899/43939 [32:08:00<38:31:18, 5.77s/it]
45%|████▌ | 19900/43939 [32:08:05<38:22:51, 5.75s/it]
{'loss': 0.9219, 'grad_norm': 1.5167587995529175, 'learning_rate': 3.3313056873752904e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.06, 'epoch': 1.81}
45%|████▌ | 19900/43939 [32:08:05<38:22:51, 5.75s/it]
45%|████▌ | 19901/43939 [32:08:11<38:22:42, 5.75s/it]
45%|████▌ | 19902/43939 [32:08:17<38:16:49, 5.73s/it]
45%|████▌ | 19903/43939 [32:08:23<38:14:21, 5.73s/it]
45%|████▌ | 19904/43939 [32:08:28<38:14:50, 5.73s/it]
45%|████▌ | 19905/43939 [32:08:34<38:11:24, 5.72s/it]
45%|████▌ | 19906/43939 [32:08:40<38:11:34, 5.72s/it]
45%|████▌ | 19907/43939 [32:08:45<38:13:17, 5.73s/it]
45%|████▌ | 19908/43939 [32:08:51<38:09:30, 5.72s/it]
45%|████▌ | 19909/43939 [32:08:57<38:09:47, 5.72s/it]
45%|████▌ | 19910/43939 [32:09:03<38:08:43, 5.71s/it]
45%|████▌ | 19911/43939 [32:09:08<38:15:07, 5.73s/it]
45%|████▌ | 19912/43939 [32:09:14<38:07:28, 5.71s/it]
45%|████▌ | 19913/43939 [32:09:20<38:07:41, 5.71s/it]
45%|████▌ | 19914/43939 [32:09:25<38:07:09, 5.71s/it]
45%|████▌ | 19915/43939 [32:09:31<38:15:11, 5.73s/it]
45%|████▌ | 19916/43939 [32:09:37<38:07:52, 5.71s/it]
45%|████▌ | 19917/43939 [32:09:43<38:03:19, 5.70s/it]
45%|████▌ | 19918/43939 [32:09:48<38:01:51, 5.70s/it]
45%|████▌ | 19919/43939 [32:09:54<37:56:06, 5.69s/it]
45%|████▌ | 19920/43939 [32:10:00<38:24:26, 5.76s/it]
45%|████▌ | 19921/43939 [32:10:05<38:16:58, 5.74s/it]
45%|████▌ | 19922/43939 [32:10:11<38:13:59, 5.73s/it]
45%|████▌ | 19923/43939 [32:10:17<38:05:13, 5.71s/it]
45%|████▌ | 19924/43939 [32:10:23<38:08:38, 5.72s/it]
45%|████▌ | 19925/43939 [32:10:28<38:08:58, 5.72s/it]
45%|████▌ | 19926/43939 [32:10:34<38:04:21, 5.71s/it]
45%|████▌ | 19927/43939 [32:10:40<38:02:44, 5.70s/it]
45%|████▌ | 19928/43939 [32:10:45<38:03:27, 5.71s/it]
45%|████▌ | 19929/43939 [32:10:51<38:02:45, 5.70s/it]
45%|████▌ | 19930/43939 [32:10:57<37:58:04, 5.69s/it]
45%|████▌ | 19931/43939 [32:11:02<37:56:58, 5.69s/it]
45%|████▌ | 19932/43939 [32:11:08<37:58:53, 5.70s/it]
45%|████▌ | 19933/43939 [32:11:14<37:53:10, 5.68s/it]
45%|████▌ | 19934/43939 [32:11:20<38:01:53, 5.70s/it]
45%|████▌ | 19935/43939 [32:11:25<38:02:03, 5.70s/it]
45%|████▌ | 19936/43939 [32:11:31<37:59:18, 5.70s/it]
45%|████▌ | 19937/43939 [32:11:37<37:59:45, 5.70s/it]
45%|████▌ | 19938/43939 [32:11:42<37:59:53, 5.70s/it]
45%|████▌ | 19939/43939 [32:11:48<38:01:09, 5.70s/it]
45%|████▌ | 19940/43939 [32:11:54<38:07:09, 5.72s/it]
45%|████▌ | 19941/43939 [32:12:00<38:03:01, 5.71s/it]
45%|████▌ | 19942/43939 [32:12:05<38:29:00, 5.77s/it]
45%|████▌ | 19943/43939 [32:12:11<38:19:46, 5.75s/it]
45%|████▌ | 19944/43939 [32:12:17<38:13:51, 5.74s/it]
45%|████▌ | 19945/43939 [32:12:23<38:08:28, 5.72s/it]
45%|████▌ | 19946/43939 [32:12:28<38:01:40, 5.71s/it]
45%|████▌ | 19947/43939 [32:12:34<37:56:21, 5.69s/it]
45%|████▌ | 19948/43939 [32:12:40<37:55:26, 5.69s/it]
45%|████▌ | 19949/43939 [32:12:45<37:56:04, 5.69s/it]
45%|████▌ | 19950/43939 [32:12:51<37:56:38, 5.69s/it]
{'loss': 0.9023, 'grad_norm': 1.1592515707015991, 'learning_rate': 3.3219340426310366e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.44, 'epoch': 1.82}
45%|████▌ | 19950/43939 [32:12:51<37:56:38, 5.69s/it]
45%|████▌ | 19951/43939 [32:12:57<37:59:05, 5.70s/it]
45%|████▌ | 19952/43939 [32:13:02<37:57:33, 5.70s/it]
45%|████▌ | 19953/43939 [32:13:08<38:01:53, 5.71s/it]
45%|████▌ | 19954/43939 [32:13:14<38:03:11, 5.71s/it]
45%|████▌ | 19955/43939 [32:13:19<37:58:06, 5.70s/it]
45%|████▌ | 19956/43939 [32:13:25<38:04:40, 5.72s/it]
45%|████▌ | 19957/43939 [32:13:31<38:00:41, 5.71s/it]
45%|████▌ | 19958/43939 [32:13:37<37:59:46, 5.70s/it]
45%|████▌ | 19959/43939 [32:13:42<37:59:14, 5.70s/it]
45%|████▌ | 19960/43939 [32:13:48<37:55:40, 5.69s/it]
45%|████▌ | 19961/43939 [32:13:54<37:52:02, 5.69s/it]
45%|████▌ | 19962/43939 [32:14:00<38:21:28, 5.76s/it]
45%|████▌ | 19963/43939 [32:14:05<38:12:05, 5.74s/it]
45%|████▌ | 19964/43939 [32:14:11<38:06:06, 5.72s/it]
45%|████▌ | 19965/43939 [32:14:17<38:06:52, 5.72s/it]
45%|████▌ | 19966/43939 [32:14:22<38:03:48, 5.72s/it]
45%|████▌ | 19967/43939 [32:14:28<38:00:07, 5.71s/it]
45%|████▌ | 19968/43939 [32:14:34<38:02:48, 5.71s/it]
45%|████▌ | 19969/43939 [32:14:40<38:05:05, 5.72s/it]
45%|████▌ | 19970/43939 [32:14:45<38:01:19, 5.71s/it]
45%|████▌ | 19971/43939 [32:14:51<37:57:22, 5.70s/it]
45%|████▌ | 19972/43939 [32:14:57<37:56:30, 5.70s/it]
45%|████▌ | 19973/43939 [32:15:02<38:00:38, 5.71s/it]
45%|████▌ | 19974/43939 [32:15:08<38:00:54, 5.71s/it]
45%|████▌ | 19975/43939 [32:15:14<38:06:05, 5.72s/it]
45%|████▌ | 19976/43939 [32:15:19<38:02:56, 5.72s/it]
45%|████▌ | 19977/43939 [32:15:25<38:02:51, 5.72s/it]
45%|████▌ | 19978/43939 [32:15:31<37:58:05, 5.70s/it]
45%|████▌ | 19979/43939 [32:15:37<37:52:32, 5.69s/it]
45%|████▌ | 19980/43939 [32:15:42<37:53:27, 5.69s/it]
45%|████▌ | 19981/43939 [32:15:48<37:58:05, 5.71s/it]
45%|████▌ | 19982/43939 [32:15:54<37:55:47, 5.70s/it]
45%|████▌ | 19983/43939 [32:16:00<38:21:56, 5.77s/it]
45%|████▌ | 19984/43939 [32:16:05<38:11:56, 5.74s/it]
45%|████▌ | 19985/43939 [32:16:11<38:06:25, 5.73s/it]
45%|████▌ | 19986/43939 [32:16:17<38:01:44, 5.72s/it]
45%|████▌ | 19987/43939 [32:16:22<38:03:47, 5.72s/it]
45%|████▌ | 19988/43939 [32:16:28<38:03:32, 5.72s/it]
45%|████▌ | 19989/43939 [32:16:34<37:59:49, 5.71s/it]
45%|████▌ | 19990/43939 [32:16:39<37:58:57, 5.71s/it]
45%|████▌ | 19991/43939 [32:16:45<37:52:29, 5.69s/it]
45%|████▌ | 19992/43939 [32:16:51<37:55:48, 5.70s/it]
46%|████▌ | 19993/43939 [32:16:57<37:49:52, 5.69s/it]
46%|████▌ | 19994/43939 [32:17:02<37:51:59, 5.69s/it]
46%|████▌ | 19995/43939 [32:17:08<37:50:50, 5.69s/it]
46%|████▌ | 19996/43939 [32:17:14<37:52:55, 5.70s/it]
46%|████▌ | 19997/43939 [32:17:19<37:51:26, 5.69s/it]
46%|████▌ | 19998/43939 [32:17:25<37:51:40, 5.69s/it]
46%|████▌ | 19999/43939 [32:17:31<37:53:59, 5.70s/it]
46%|████▌ | 20000/43939 [32:17:36<37:52:55, 5.70s/it]
{'loss': 0.9122, 'grad_norm': 1.0731903314590454, 'learning_rate': 3.3125494299457416e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.3, 'epoch': 1.82}
46%|████▌ | 20000/43939 [32:17:36<37:52:55, 5.70s/it]
46%|████▌ | 20001/43939 [32:17:42<37:54:21, 5.70s/it]
46%|████▌ | 20002/43939 [32:17:48<37:56:48, 5.71s/it]
46%|████▌ | 20003/43939 [32:17:54<37:52:03, 5.70s/it]
46%|████▌ | 20004/43939 [32:17:59<38:18:36, 5.76s/it]
46%|████▌ | 20005/43939 [32:18:05<38:11:33, 5.74s/it]
46%|████▌ | 20006/43939 [32:18:11<38:05:25, 5.73s/it]
46%|████▌ | 20007/43939 [32:18:17<38:03:36, 5.73s/it]
46%|████▌ | 20008/43939 [32:18:22<37:58:38, 5.71s/it]
46%|████▌ | 20009/43939 [32:18:28<38:02:49, 5.72s/it]
46%|████▌ | 20010/43939 [32:18:34<38:02:08, 5.72s/it]
46%|████▌ | 20011/43939 [32:18:39<37:57:56, 5.71s/it]
46%|████▌ | 20012/43939 [32:18:45<37:54:40, 5.70s/it]
46%|████▌ | 20013/43939 [32:18:51<37:55:37, 5.71s/it]
46%|████▌ | 20014/43939 [32:18:56<37:53:52, 5.70s/it]
46%|████▌ | 20015/43939 [32:19:02<37:55:31, 5.71s/it]
46%|████▌ | 20016/43939 [32:19:08<37:53:17, 5.70s/it]
46%|████▌ | 20017/43939 [32:19:14<37:54:22, 5.70s/it]
46%|████▌ | 20018/43939 [32:19:19<37:49:21, 5.69s/it]
46%|████▌ | 20019/43939 [32:19:25<37:52:42, 5.70s/it]
46%|████▌ | 20020/43939 [32:19:31<37:49:21, 5.69s/it]
46%|████▌ | 20021/43939 [32:19:36<37:50:17, 5.70s/it]
46%|████▌ | 20022/43939 [32:19:42<37:50:03, 5.69s/it]
46%|████▌ | 20023/43939 [32:19:48<37:56:35, 5.71s/it]
46%|████▌ | 20024/43939 [32:19:54<38:33:00, 5.80s/it]
46%|████▌ | 20025/43939 [32:20:00<38:26:54, 5.79s/it]
46%|████▌ | 20026/43939 [32:20:05<38:15:02, 5.76s/it]
46%|████▌ | 20027/43939 [32:20:11<38:17:02, 5.76s/it]
46%|████▌ | 20028/43939 [32:20:17<38:06:02, 5.74s/it]
46%|████▌ | 20029/43939 [32:20:22<38:10:38, 5.75s/it]
46%|████▌ | 20030/43939 [32:20:28<38:07:19, 5.74s/it]
46%|████▌ | 20031/43939 [32:20:34<38:05:24, 5.74s/it]
46%|████▌ | 20032/43939 [32:20:40<38:07:01, 5.74s/it]
46%|████▌ | 20033/43939 [32:20:45<38:02:25, 5.73s/it]
46%|████▌ | 20034/43939 [32:20:51<37:56:35, 5.71s/it]
46%|████▌ | 20035/43939 [32:20:57<37:51:11, 5.70s/it]
46%|████▌ | 20036/43939 [32:21:02<37:54:40, 5.71s/it]
46%|████▌ | 20037/43939 [32:21:08<37:53:46, 5.71s/it]
46%|████▌ | 20038/43939 [32:21:14<37:54:10, 5.71s/it]
46%|████▌ | 20039/43939 [32:21:20<37:52:24, 5.70s/it]
46%|████▌ | 20040/43939 [32:21:25<37:58:20, 5.72s/it]
46%|████▌ | 20041/43939 [32:21:31<37:53:25, 5.71s/it]
46%|████▌ | 20042/43939 [32:21:37<37:54:11, 5.71s/it]
46%|████▌ | 20043/43939 [32:21:42<37:52:11, 5.71s/it]
46%|████▌ | 20044/43939 [32:21:48<37:50:57, 5.70s/it]
46%|████▌ | 20045/43939 [32:21:54<38:17:32, 5.77s/it]
46%|████▌ | 20046/43939 [32:22:00<38:11:53, 5.76s/it]
46%|████▌ | 20047/43939 [32:22:05<38:06:58, 5.74s/it]
46%|████▌ | 20048/43939 [32:22:11<38:01:51, 5.73s/it]
46%|████▌ | 20049/43939 [32:22:17<37:57:56, 5.72s/it]
46%|████▌ | 20050/43939 [32:22:23<37:59:53, 5.73s/it]
{'loss': 0.9147, 'grad_norm': 1.0888726711273193, 'learning_rate': 3.303151997383725e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.11, 'epoch': 1.83}
46%|████▌ | 20050/43939 [32:22:23<37:59:53, 5.73s/it]
46%|████▌ | 20051/43939 [32:22:28<38:01:04, 5.73s/it]
46%|████▌ | 20052/43939 [32:22:34<37:56:55, 5.72s/it]
46%|████▌ | 20053/43939 [32:22:40<37:49:47, 5.70s/it]
46%|████▌ | 20054/43939 [32:22:45<37:48:53, 5.70s/it]
46%|████▌ | 20055/43939 [32:22:51<37:47:33, 5.70s/it]
46%|████▌ | 20056/43939 [32:22:57<37:48:09, 5.70s/it]
46%|████▌ | 20057/43939 [32:23:02<37:44:08, 5.69s/it]
46%|████▌ | 20058/43939 [32:23:08<37:44:59, 5.69s/it]
46%|████▌ | 20059/43939 [32:23:14<37:52:28, 5.71s/it]
46%|████▌ | 20060/43939 [32:23:20<37:53:47, 5.71s/it]
46%|████▌ | 20061/43939 [32:23:25<37:51:15, 5.71s/it]
46%|████▌ | 20062/43939 [32:23:31<37:52:44, 5.71s/it]
46%|████▌ | 20063/43939 [32:23:37<37:48:57, 5.70s/it]
46%|████▌ | 20064/43939 [32:23:42<37:47:00, 5.70s/it]
46%|████▌ | 20065/43939 [32:23:48<37:44:26, 5.69s/it]
46%|████▌ | 20066/43939 [32:23:54<37:44:56, 5.69s/it]
46%|████▌ | 20067/43939 [32:24:00<38:17:01, 5.77s/it]
46%|████▌ | 20068/43939 [32:24:05<38:10:26, 5.76s/it]
46%|████▌ | 20069/43939 [32:24:11<38:02:17, 5.74s/it]
46%|████▌ | 20070/43939 [32:24:17<37:59:47, 5.73s/it]
46%|████▌ | 20071/43939 [32:24:23<37:52:20, 5.71s/it]
46%|████▌ | 20072/43939 [32:24:28<37:53:01, 5.71s/it]
46%|████▌ | 20073/43939 [32:24:34<37:49:13, 5.70s/it]
46%|████▌ | 20074/43939 [32:24:40<37:54:15, 5.72s/it]
46%|████▌ | 20075/43939 [32:24:45<37:50:23, 5.71s/it]
46%|████▌ | 20076/43939 [32:24:51<37:49:27, 5.71s/it]
46%|████▌ | 20077/43939 [32:24:57<37:47:26, 5.70s/it]
46%|████▌ | 20078/43939 [32:25:02<37:44:28, 5.69s/it]
46%|████▌ | 20079/43939 [32:25:08<37:50:17, 5.71s/it]
46%|████▌ | 20080/43939 [32:25:14<37:52:07, 5.71s/it]
46%|████▌ | 20081/43939 [32:25:20<37:45:15, 5.70s/it]
46%|████▌ | 20082/43939 [32:25:25<37:45:43, 5.70s/it]
46%|████▌ | 20083/43939 [32:25:31<37:43:59, 5.69s/it]
46%|████▌ | 20084/43939 [32:25:37<37:46:18, 5.70s/it]
46%|████▌ | 20085/43939 [32:25:42<37:47:03, 5.70s/it]
46%|████▌ | 20086/43939 [32:25:48<37:43:31, 5.69s/it]
46%|████▌ | 20087/43939 [32:25:54<37:52:02, 5.72s/it]
46%|████▌ | 20088/43939 [32:26:00<38:19:33, 5.78s/it]
46%|████▌ | 20089/43939 [32:26:05<38:04:44, 5.75s/it]
46%|████▌ | 20090/43939 [32:26:11<37:56:48, 5.73s/it]
46%|████▌ | 20091/43939 [32:26:17<37:51:12, 5.71s/it]
46%|████▌ | 20092/43939 [32:26:22<37:46:34, 5.70s/it]
46%|████▌ | 20093/43939 [32:26:28<37:49:33, 5.71s/it]
46%|████▌ | 20094/43939 [32:26:34<37:46:38, 5.70s/it]
46%|████▌ | 20095/43939 [32:26:40<37:50:13, 5.71s/it]
46%|████▌ | 20096/43939 [32:26:45<37:50:58, 5.71s/it]
46%|████▌ | 20097/43939 [32:26:51<37:50:02, 5.71s/it]
46%|████▌ | 20098/43939 [32:26:57<37:46:23, 5.70s/it]
46%|████▌ | 20099/43939 [32:27:02<37:41:17, 5.69s/it]
46%|████▌ | 20100/43939 [32:27:08<37:40:56, 5.69s/it]
{'loss': 0.9166, 'grad_norm': 1.3090068101882935, 'learning_rate': 3.293741893211567e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.9, 'epoch': 1.83}
46%|████▌ | 20100/43939 [32:27:08<37:40:56, 5.69s/it]
46%|████▌ | 20101/43939 [32:27:14<37:45:15, 5.70s/it]
46%|████▌ | 20102/43939 [32:27:19<37:44:48, 5.70s/it]
46%|████▌ | 20103/43939 [32:27:25<37:43:59, 5.70s/it]
46%|████▌ | 20104/43939 [32:27:31<37:45:28, 5.70s/it]
46%|████▌ | 20105/43939 [32:27:37<37:47:48, 5.71s/it]
46%|████▌ | 20106/43939 [32:27:42<37:42:56, 5.70s/it]
46%|████▌ | 20107/43939 [32:27:48<37:44:52, 5.70s/it]
46%|████▌ | 20108/43939 [32:27:54<37:44:31, 5.70s/it]
46%|████▌ | 20109/43939 [32:28:00<38:10:15, 5.77s/it]
46%|████▌ | 20110/43939 [32:28:05<38:04:42, 5.75s/it]
46%|████▌ | 20111/43939 [32:28:11<37:58:43, 5.74s/it]
46%|████▌ | 20112/43939 [32:28:17<37:52:37, 5.72s/it]
46%|████▌ | 20113/43939 [32:28:22<37:47:36, 5.71s/it]
46%|████▌ | 20114/43939 [32:28:28<37:43:32, 5.70s/it]
46%|████▌ | 20115/43939 [32:28:34<37:49:24, 5.72s/it]
46%|████▌ | 20116/43939 [32:28:40<37:43:53, 5.70s/it]
46%|████▌ | 20117/43939 [32:28:45<37:40:58, 5.69s/it]
46%|████▌ | 20118/43939 [32:28:51<37:39:55, 5.69s/it]
46%|████▌ | 20119/43939 [32:28:57<37:39:15, 5.69s/it]
46%|████▌ | 20120/43939 [32:29:02<37:40:24, 5.69s/it]
46%|████▌ | 20121/43939 [32:29:08<37:43:51, 5.70s/it]
46%|████▌ | 20122/43939 [32:29:14<37:44:49, 5.71s/it]
46%|████▌ | 20123/43939 [32:29:19<37:48:08, 5.71s/it]
46%|████▌ | 20124/43939 [32:29:25<37:47:39, 5.71s/it]
46%|████▌ | 20125/43939 [32:29:31<37:42:12, 5.70s/it]
46%|████▌ | 20126/43939 [32:29:37<37:44:13, 5.71s/it]
46%|████▌ | 20127/43939 [32:29:42<37:47:12, 5.71s/it]
46%|████▌ | 20128/43939 [32:29:48<37:44:47, 5.71s/it]
46%|████▌ | 20129/43939 [32:29:54<37:42:36, 5.70s/it]
46%|████▌ | 20130/43939 [32:30:00<38:11:23, 5.77s/it]
46%|████▌ | 20131/43939 [32:30:05<38:01:52, 5.75s/it]
46%|████▌ | 20132/43939 [32:30:11<37:54:53, 5.73s/it]
46%|████▌ | 20133/43939 [32:30:17<37:51:09, 5.72s/it]
46%|████▌ | 20134/43939 [32:30:22<37:53:06, 5.73s/it]
46%|████▌ | 20135/43939 [32:30:28<37:52:18, 5.73s/it]
46%|████▌ | 20136/43939 [32:30:34<37:53:42, 5.73s/it]
46%|████▌ | 20137/43939 [32:30:40<37:45:39, 5.71s/it]
46%|████▌ | 20138/43939 [32:30:45<37:51:54, 5.73s/it]
46%|████▌ | 20139/43939 [32:30:51<37:46:05, 5.71s/it]
46%|████▌ | 20140/43939 [32:30:57<37:42:38, 5.70s/it]
46%|████▌ | 20141/43939 [32:31:02<37:44:32, 5.71s/it]
46%|████▌ | 20142/43939 [32:31:08<37:41:25, 5.70s/it]
46%|████▌ | 20143/43939 [32:31:14<37:44:52, 5.71s/it]
46%|████▌ | 20144/43939 [32:31:20<37:52:31, 5.73s/it]
46%|████▌ | 20145/43939 [32:31:25<37:50:43, 5.73s/it]
46%|████▌ | 20146/43939 [32:31:31<37:47:28, 5.72s/it]
46%|████▌ | 20147/43939 [32:31:37<37:43:16, 5.71s/it]
46%|████▌ | 20148/43939 [32:31:42<37:43:21, 5.71s/it]
46%|████▌ | 20149/43939 [32:31:48<37:40:33, 5.70s/it]
46%|████▌ | 20150/43939 [32:31:54<37:40:23, 5.70s/it]
{'loss': 0.9221, 'grad_norm': 1.3456158638000488, 'learning_rate': 3.2843192658957774e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.15, 'epoch': 1.83}
46%|████▌ | 20150/43939 [32:31:54<37:40:23, 5.70s/it]
46%|████▌ | 20151/43939 [32:32:00<37:42:14, 5.71s/it]
46%|████▌ | 20152/43939 [32:32:05<38:08:44, 5.77s/it]
46%|████▌ | 20153/43939 [32:32:11<37:59:21, 5.75s/it]
46%|████▌ | 20154/43939 [32:32:17<37:50:34, 5.73s/it]
46%|████▌ | 20155/43939 [32:32:23<37:50:39, 5.73s/it]
46%|████▌ | 20156/43939 [32:32:28<37:43:09, 5.71s/it]
46%|████▌ | 20157/43939 [32:32:34<37:47:45, 5.72s/it]
46%|████▌ | 20158/43939 [32:32:40<37:42:51, 5.71s/it]
46%|████▌ | 20159/43939 [32:32:45<37:43:11, 5.71s/it]
46%|████▌ | 20160/43939 [32:32:51<37:38:52, 5.70s/it]
46%|████▌ | 20161/43939 [32:32:57<37:43:50, 5.71s/it]
46%|████▌ | 20162/43939 [32:33:02<37:41:57, 5.71s/it]
46%|████▌ | 20163/43939 [32:33:08<37:43:11, 5.71s/it]
46%|████▌ | 20164/43939 [32:33:14<37:42:16, 5.71s/it]
46%|████▌ | 20165/43939 [32:33:20<37:41:02, 5.71s/it]
46%|████▌ | 20166/43939 [32:33:25<37:43:47, 5.71s/it]
46%|████▌ | 20167/43939 [32:33:31<37:39:48, 5.70s/it]
46%|████▌ | 20168/43939 [32:33:37<37:37:32, 5.70s/it]
46%|████▌ | 20169/43939 [32:33:42<37:43:22, 5.71s/it]
46%|████▌ | 20170/43939 [32:33:48<37:45:59, 5.72s/it]
46%|████▌ | 20171/43939 [32:33:54<37:53:00, 5.74s/it]
46%|████▌ | 20172/43939 [32:34:00<38:13:27, 5.79s/it]
46%|████▌ | 20173/43939 [32:34:06<38:03:53, 5.77s/it]
46%|████▌ | 20174/43939 [32:34:11<37:56:46, 5.75s/it]
46%|████▌ | 20175/43939 [32:34:17<37:53:19, 5.74s/it]
46%|████▌ | 20176/43939 [32:34:23<37:46:52, 5.72s/it]
46%|████▌ | 20177/43939 [32:34:28<37:47:05, 5.72s/it]
46%|████▌ | 20178/43939 [32:34:34<37:50:03, 5.73s/it]
46%|████▌ | 20179/43939 [32:34:40<37:46:55, 5.72s/it]
46%|████▌ | 20180/43939 [32:34:46<37:48:29, 5.73s/it]
46%|████▌ | 20181/43939 [32:34:51<37:43:53, 5.72s/it]
46%|████▌ | 20182/43939 [32:34:57<37:42:17, 5.71s/it]
46%|████▌ | 20183/43939 [32:35:03<37:39:49, 5.71s/it]
46%|████▌ | 20184/43939 [32:35:08<37:45:19, 5.72s/it]
46%|████▌ | 20185/43939 [32:35:14<37:45:29, 5.72s/it]
46%|████▌ | 20186/43939 [32:35:20<37:53:51, 5.74s/it]
46%|████▌ | 20187/43939 [32:35:26<37:48:47, 5.73s/it]
46%|████▌ | 20188/43939 [32:35:31<37:47:14, 5.73s/it]
46%|████▌ | 20189/43939 [32:35:37<37:48:37, 5.73s/it]
46%|████▌ | 20190/43939 [32:35:43<37:40:31, 5.71s/it]
46%|████▌ | 20191/43939 [32:35:48<37:39:47, 5.71s/it]
46%|████▌ | 20192/43939 [32:35:54<37:35:59, 5.70s/it]
46%|████▌ | 20193/43939 [32:36:00<38:02:09, 5.77s/it]
46%|████▌ | 20194/43939 [32:36:06<37:55:49, 5.75s/it]
46%|████▌ | 20195/43939 [32:36:11<37:48:36, 5.73s/it]
46%|████▌ | 20196/43939 [32:36:17<37:45:15, 5.72s/it]
46%|████▌ | 20197/43939 [32:36:23<37:46:17, 5.73s/it]
46%|████▌ | 20198/43939 [32:36:29<37:47:22, 5.73s/it]
46%|████▌ | 20199/43939 [32:36:34<37:43:03, 5.72s/it]
46%|████▌ | 20200/43939 [32:36:40<37:44:36, 5.72s/it]
{'loss': 0.9416, 'grad_norm': 1.1804425716400146, 'learning_rate': 3.274884264100442e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.2, 'epoch': 1.84}
46%|████▌ | 20200/43939 [32:36:40<37:44:36, 5.72s/it]
46%|████▌ | 20201/43939 [32:36:46<37:43:59, 5.72s/it]
46%|████▌ | 20202/43939 [32:36:51<37:37:40, 5.71s/it]
46%|████▌ | 20203/43939 [32:36:57<37:42:40, 5.72s/it]
46%|████▌ | 20204/43939 [32:37:03<37:36:57, 5.71s/it]
46%|████▌ | 20205/43939 [32:37:09<37:38:22, 5.71s/it]
46%|████▌ | 20206/43939 [32:37:14<37:38:02, 5.71s/it]
46%|████▌ | 20207/43939 [32:37:20<37:35:35, 5.70s/it]
46%|████▌ | 20208/43939 [32:37:26<37:34:11, 5.70s/it]
46%|████▌ | 20209/43939 [32:37:31<37:28:35, 5.69s/it]
46%|████▌ | 20210/43939 [32:37:37<37:29:30, 5.69s/it]
46%|████▌ | 20211/43939 [32:37:43<37:26:47, 5.68s/it]
46%|████▌ | 20212/43939 [32:37:48<37:30:30, 5.69s/it]
46%|████▌ | 20213/43939 [32:37:54<37:33:22, 5.70s/it]
46%|████▌ | 20214/43939 [32:38:00<37:59:37, 5.77s/it]
46%|████▌ | 20215/43939 [32:38:06<37:47:14, 5.73s/it]
46%|████▌ | 20216/43939 [32:38:11<37:47:39, 5.74s/it]
46%|████▌ | 20217/43939 [32:38:17<37:44:22, 5.73s/it]
46%|████▌ | 20218/43939 [32:38:23<37:40:35, 5.72s/it]
46%|████▌ | 20219/43939 [32:38:29<37:35:30, 5.71s/it]
46%|████▌ | 20220/43939 [32:38:34<37:33:32, 5.70s/it]
46%|████▌ | 20221/43939 [32:38:40<37:36:48, 5.71s/it]
46%|████▌ | 20222/43939 [32:38:46<37:37:14, 5.71s/it]
46%|████▌ | 20223/43939 [32:38:51<37:34:48, 5.70s/it]
46%|████▌ | 20224/43939 [32:38:57<37:28:54, 5.69s/it]
46%|████▌ | 20225/43939 [32:39:03<37:28:23, 5.69s/it]
46%|████▌ | 20226/43939 [32:39:08<37:26:20, 5.68s/it]
46%|████▌ | 20227/43939 [32:39:14<37:36:26, 5.71s/it]
46%|████▌ | 20228/43939 [32:39:20<37:33:48, 5.70s/it]
46%|████▌ | 20229/43939 [32:39:26<37:38:51, 5.72s/it]
46%|████▌ | 20230/43939 [32:39:31<37:32:55, 5.70s/it]
46%|████▌ | 20231/43939 [32:39:37<37:32:56, 5.70s/it]
46%|████▌ | 20232/43939 [32:39:43<37:33:56, 5.70s/it]
46%|████▌ | 20233/43939 [32:39:48<37:34:48, 5.71s/it]
46%|████▌ | 20234/43939 [32:39:54<37:36:39, 5.71s/it]
46%|████▌ | 20235/43939 [32:40:00<38:05:47, 5.79s/it]
46%|████▌ | 20236/43939 [32:40:06<37:55:45, 5.76s/it]
46%|████▌ | 20237/43939 [32:40:11<37:45:02, 5.73s/it]
46%|████▌ | 20238/43939 [32:40:17<37:42:35, 5.73s/it]
46%|████▌ | 20239/43939 [32:40:23<37:48:43, 5.74s/it]
46%|████▌ | 20240/43939 [32:40:29<37:42:21, 5.73s/it]
46%|████▌ | 20241/43939 [32:40:34<37:36:26, 5.71s/it]
46%|████▌ | 20242/43939 [32:40:40<37:36:04, 5.71s/it]
46%|████▌ | 20243/43939 [32:40:46<37:40:16, 5.72s/it]
46%|████▌ | 20244/43939 [32:40:52<37:41:18, 5.73s/it]
46%|████▌ | 20245/43939 [32:40:57<37:42:51, 5.73s/it]
46%|████▌ | 20246/43939 [32:41:03<37:35:46, 5.71s/it]
46%|████▌ | 20247/43939 [32:41:09<37:30:26, 5.70s/it]
46%|████▌ | 20248/43939 [32:41:14<37:33:43, 5.71s/it]
46%|████▌ | 20249/43939 [32:41:20<37:28:51, 5.70s/it]
46%|████▌ | 20250/43939 [32:41:26<37:27:02, 5.69s/it]
{'loss': 0.9423, 'grad_norm': 1.1315155029296875, 'learning_rate': 3.265437036684889e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.39, 'epoch': 1.84}
46%|████▌ | 20250/43939 [32:41:26<37:27:02, 5.69s/it]
46%|████▌ | 20251/43939 [32:41:31<37:31:43, 5.70s/it]
46%|████▌ | 20252/43939 [32:41:37<37:31:06, 5.70s/it]
46%|████▌ | 20253/43939 [32:41:43<37:28:38, 5.70s/it]
46%|████▌ | 20254/43939 [32:41:49<37:32:54, 5.71s/it]
46%|████▌ | 20255/43939 [32:41:54<37:31:22, 5.70s/it]
46%|████▌ | 20256/43939 [32:42:00<38:01:11, 5.78s/it]
46%|████▌ | 20257/43939 [32:42:06<37:51:43, 5.76s/it]
46%|████▌ | 20258/43939 [32:42:12<37:41:39, 5.73s/it]
46%|████▌ | 20259/43939 [32:42:17<37:44:11, 5.74s/it]
46%|████▌ | 20260/43939 [32:42:23<37:42:28, 5.73s/it]
46%|████▌ | 20261/43939 [32:42:29<37:40:54, 5.73s/it]
46%|████▌ | 20262/43939 [32:42:34<37:34:19, 5.71s/it]
46%|████▌ | 20263/43939 [32:42:40<37:33:50, 5.71s/it]
46%|████▌ | 20264/43939 [32:42:46<37:28:57, 5.70s/it]
46%|████▌ | 20265/43939 [32:42:51<37:25:13, 5.69s/it]
46%|████▌ | 20266/43939 [32:42:57<37:26:51, 5.69s/it]
46%|████▌ | 20267/43939 [32:43:03<37:28:31, 5.70s/it]
46%|████▌ | 20268/43939 [32:43:09<37:31:07, 5.71s/it]
46%|████▌ | 20269/43939 [32:43:14<37:30:07, 5.70s/it]
46%|████▌ | 20270/43939 [32:43:20<37:42:44, 5.74s/it]
46%|████▌ | 20271/43939 [32:43:26<37:35:29, 5.72s/it]
46%|████▌ | 20272/43939 [32:43:31<37:31:18, 5.71s/it]
46%|████▌ | 20273/43939 [32:43:37<37:29:46, 5.70s/it]
46%|████▌ | 20274/43939 [32:43:43<37:31:38, 5.71s/it]
46%|████▌ | 20275/43939 [32:43:49<37:31:42, 5.71s/it]
46%|████▌ | 20276/43939 [32:43:54<37:29:22, 5.70s/it]
46%|████▌ | 20277/43939 [32:44:00<37:30:07, 5.71s/it]
46%|████▌ | 20278/43939 [32:44:06<38:03:31, 5.79s/it]
46%|████▌ | 20279/43939 [32:44:12<37:56:41, 5.77s/it]
46%|████▌ | 20280/43939 [32:44:17<37:47:06, 5.75s/it]
46%|████▌ | 20281/43939 [32:44:23<37:40:49, 5.73s/it]
46%|████▌ | 20282/43939 [32:44:29<37:34:37, 5.72s/it]
46%|████▌ | 20283/43939 [32:44:35<37:37:27, 5.73s/it]
46%|████▌ | 20284/43939 [32:44:40<37:30:55, 5.71s/it]
46%|████▌ | 20285/43939 [32:44:46<37:25:13, 5.70s/it]
46%|████▌ | 20286/43939 [32:44:52<37:27:22, 5.70s/it]
46%|████▌ | 20287/43939 [32:44:57<37:28:09, 5.70s/it]
46%|████▌ | 20288/43939 [32:45:03<37:30:37, 5.71s/it]
46%|████▌ | 20289/43939 [32:45:09<37:28:17, 5.70s/it]
46%|████▌ | 20290/43939 [32:45:14<37:27:51, 5.70s/it]
46%|████▌ | 20291/43939 [32:45:20<37:29:25, 5.71s/it]
46%|████▌ | 20292/43939 [32:45:26<37:28:15, 5.70s/it]
46%|████▌ | 20293/43939 [32:45:31<37:23:14, 5.69s/it]
46%|████▌ | 20294/43939 [32:45:37<37:24:38, 5.70s/it]
46%|████▌ | 20295/43939 [32:45:43<37:24:47, 5.70s/it]
46%|████▌ | 20296/43939 [32:45:49<37:26:47, 5.70s/it]
46%|████▌ | 20297/43939 [32:45:54<37:36:16, 5.73s/it]
46%|████▌ | 20298/43939 [32:46:00<37:34:27, 5.72s/it]
46%|████▌ | 20299/43939 [32:46:06<37:32:22, 5.72s/it]
46%|████▌ | 20300/43939 [32:46:12<37:59:38, 5.79s/it]
{'loss': 0.9118, 'grad_norm': 1.3613213300704956, 'learning_rate': 3.255977732701329e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 453.31, 'epoch': 1.85}
46%|████▌ | 20300/43939 [32:46:12<37:59:38, 5.79s/it]
46%|████▌ | 20301/43939 [32:46:17<37:50:38, 5.76s/it]
46%|████▌ | 20302/43939 [32:46:23<37:48:31, 5.76s/it]
46%|████▌ | 20303/43939 [32:46:29<37:46:10, 5.75s/it]
46%|████▌ | 20304/43939 [32:46:35<37:39:05, 5.73s/it]
46%|████▌ | 20305/43939 [32:46:40<37:36:43, 5.73s/it]
46%|████▌ | 20306/43939 [32:46:46<37:37:58, 5.73s/it]
46%|████▌ | 20307/43939 [32:46:52<37:33:51, 5.72s/it]
46%|████▌ | 20308/43939 [32:46:57<37:27:38, 5.71s/it]
46%|████▌ | 20309/43939 [32:47:03<37:23:33, 5.70s/it]
46%|████▌ | 20310/43939 [32:47:09<37:31:52, 5.72s/it]
46%|████▌ | 20311/43939 [32:47:15<37:28:27, 5.71s/it]
46%|████▌ | 20312/43939 [32:47:20<37:25:53, 5.70s/it]
46%|████▌ | 20313/43939 [32:47:26<37:29:35, 5.71s/it]
46%|████▌ | 20314/43939 [32:47:32<37:23:21, 5.70s/it]
46%|████▌ | 20315/43939 [32:47:37<37:22:54, 5.70s/it]
46%|████▌ | 20316/43939 [32:47:43<37:26:31, 5.71s/it]
46%|████▌ | 20317/43939 [32:47:49<37:25:59, 5.70s/it]
46%|████▌ | 20318/43939 [32:47:55<37:26:40, 5.71s/it]
46%|████▌ | 20319/43939 [32:48:00<37:21:40, 5.69s/it]
46%|████▌ | 20320/43939 [32:48:06<37:49:33, 5.77s/it]
46%|████▌ | 20321/43939 [32:48:12<37:41:02, 5.74s/it]
46%|████▋ | 20322/43939 [32:48:17<37:33:28, 5.73s/it]
46%|████▋ | 20323/43939 [32:48:23<37:29:51, 5.72s/it]
46%|████▋ | 20324/43939 [32:48:29<37:25:29, 5.71s/it]
46%|████▋ | 20325/43939 [32:48:35<37:25:16, 5.70s/it]
46%|████▋ | 20326/43939 [32:48:40<37:27:18, 5.71s/it]
46%|████▋ | 20327/43939 [32:48:46<37:28:01, 5.71s/it]
46%|████▋ | 20328/43939 [32:48:52<37:27:31, 5.71s/it]
46%|████▋ | 20329/43939 [32:48:57<37:32:01, 5.72s/it]
46%|████▋ | 20330/43939 [32:49:03<37:30:20, 5.72s/it]
46%|████▋ | 20331/43939 [32:49:09<37:30:20, 5.72s/it]
46%|████▋ | 20332/43939 [32:49:15<37:29:24, 5.72s/it]
46%|████▋ | 20333/43939 [32:49:20<37:21:08, 5.70s/it]
46%|████▋ | 20334/43939 [32:49:26<37:33:47, 5.73s/it]
46%|████▋ | 20335/43939 [32:49:32<37:27:19, 5.71s/it]
46%|████▋ | 20336/43939 [32:49:37<37:29:47, 5.72s/it]
46%|████▋ | 20337/43939 [32:49:43<37:28:55, 5.72s/it]
46%|████▋ | 20338/43939 [32:49:49<37:26:02, 5.71s/it]
46%|████▋ | 20339/43939 [32:49:55<37:24:21, 5.71s/it]
46%|████▋ | 20340/43939 [32:50:00<37:20:04, 5.70s/it]
46%|████▋ | 20341/43939 [32:50:06<37:44:57, 5.76s/it]
46%|████▋ | 20342/43939 [32:50:12<37:40:28, 5.75s/it]
46%|████▋ | 20343/43939 [32:50:18<37:30:29, 5.72s/it]
46%|████▋ | 20344/43939 [32:50:23<37:30:50, 5.72s/it]
46%|████▋ | 20345/43939 [32:50:29<37:33:15, 5.73s/it]
46%|████▋ | 20346/43939 [32:50:35<37:28:59, 5.72s/it]
46%|████▋ | 20347/43939 [32:50:40<37:21:15, 5.70s/it]
46%|████▋ | 20348/43939 [32:50:46<37:22:36, 5.70s/it]
46%|████▋ | 20349/43939 [32:50:52<37:21:59, 5.70s/it]
46%|████▋ | 20350/43939 [32:50:57<37:18:39, 5.69s/it]
{'loss': 0.9314, 'grad_norm': 1.228337287902832, 'learning_rate': 3.246506501392512e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.61, 'epoch': 1.85}
46%|████▋ | 20350/43939 [32:50:57<37:18:39, 5.69s/it]
46%|████▋ | 20351/43939 [32:51:03<37:26:51, 5.72s/it]
46%|████▋ | 20352/43939 [32:51:09<37:22:18, 5.70s/it]
46%|████▋ | 20353/43939 [32:51:15<37:22:56, 5.71s/it]
46%|████▋ | 20354/43939 [32:51:20<37:22:17, 5.70s/it]
46%|████▋ | 20355/43939 [32:51:26<37:20:51, 5.70s/it]
46%|████▋ | 20356/43939 [32:51:32<37:19:50, 5.70s/it]
46%|████▋ | 20357/43939 [32:51:37<37:19:26, 5.70s/it]
46%|████▋ | 20358/43939 [32:51:43<37:20:40, 5.70s/it]
46%|████▋ | 20359/43939 [32:51:49<37:20:39, 5.70s/it]
46%|████▋ | 20360/43939 [32:51:54<37:15:54, 5.69s/it]
46%|████▋ | 20361/43939 [32:52:00<37:16:46, 5.69s/it]
46%|████▋ | 20362/43939 [32:52:06<37:30:13, 5.73s/it]
46%|████▋ | 20363/43939 [32:52:12<37:54:49, 5.79s/it]
46%|████▋ | 20364/43939 [32:52:18<37:48:48, 5.77s/it]
46%|████▋ | 20365/43939 [32:52:23<37:42:18, 5.76s/it]
46%|████▋ | 20366/43939 [32:52:29<37:40:53, 5.75s/it]
46%|████▋ | 20367/43939 [32:52:35<37:30:52, 5.73s/it]
46%|████▋ | 20368/43939 [32:52:40<37:24:42, 5.71s/it]
46%|████▋ | 20369/43939 [32:52:46<37:22:44, 5.71s/it]
46%|████▋ | 20370/43939 [32:52:52<37:17:32, 5.70s/it]
46%|████▋ | 20371/43939 [32:52:57<37:16:57, 5.69s/it]
46%|████▋ | 20372/43939 [32:53:03<37:18:26, 5.70s/it]
46%|████▋ | 20373/43939 [32:53:09<37:18:17, 5.70s/it]
46%|████▋ | 20374/43939 [32:53:15<37:25:31, 5.72s/it]
46%|████▋ | 20375/43939 [32:53:20<37:24:45, 5.72s/it]
46%|████▋ | 20376/43939 [32:53:26<37:23:32, 5.71s/it]
46%|████▋ | 20377/43939 [32:53:32<37:21:17, 5.71s/it]
46%|████▋ | 20378/43939 [32:53:37<37:17:52, 5.70s/it]
46%|████▋ | 20379/43939 [32:53:43<37:18:54, 5.70s/it]
46%|████▋ | 20380/43939 [32:53:49<37:13:50, 5.69s/it]
46%|████▋ | 20381/43939 [32:53:54<37:12:32, 5.69s/it]
46%|████▋ | 20382/43939 [32:54:00<37:20:39, 5.71s/it]
46%|████▋ | 20383/43939 [32:54:06<37:16:07, 5.70s/it]
46%|████▋ | 20384/43939 [32:54:12<37:44:31, 5.77s/it]
46%|████▋ | 20385/43939 [32:54:18<37:31:19, 5.73s/it]
46%|████▋ | 20386/43939 [32:54:23<37:28:42, 5.73s/it]
46%|████▋ | 20387/43939 [32:54:29<37:22:47, 5.71s/it]
46%|████▋ | 20388/43939 [32:54:35<37:17:23, 5.70s/it]
46%|████▋ | 20389/43939 [32:54:40<37:20:02, 5.71s/it]
46%|████▋ | 20390/43939 [32:54:46<37:19:38, 5.71s/it]
46%|████▋ | 20391/43939 [32:54:52<37:14:30, 5.69s/it]
46%|████▋ | 20392/43939 [32:54:57<37:12:01, 5.69s/it]
46%|████▋ | 20393/43939 [32:55:03<37:15:20, 5.70s/it]
46%|████▋ | 20394/43939 [32:55:09<37:18:09, 5.70s/it]
46%|████▋ | 20395/43939 [32:55:15<37:27:00, 5.73s/it]
46%|████▋ | 20396/43939 [32:55:20<37:22:30, 5.72s/it]
46%|████▋ | 20397/43939 [32:55:26<37:21:44, 5.71s/it]
46%|████▋ | 20398/43939 [32:55:32<37:21:42, 5.71s/it]
46%|████▋ | 20399/43939 [32:55:37<37:19:13, 5.71s/it]
46%|████▋ | 20400/43939 [32:55:43<37:22:36, 5.72s/it]
{'loss': 0.8986, 'grad_norm': 1.2855277061462402, 'learning_rate': 3.237023492189369e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.9, 'epoch': 1.86}
46%|████▋ | 20400/43939 [32:55:43<37:22:36, 5.72s/it]
46%|████▋ | 20401/43939 [32:55:49<37:27:11, 5.73s/it]
46%|████▋ | 20402/43939 [32:55:55<37:24:18, 5.72s/it]
46%|████▋ | 20403/43939 [32:56:00<37:18:49, 5.71s/it]
46%|████▋ | 20404/43939 [32:56:06<37:19:48, 5.71s/it]
46%|████▋ | 20405/43939 [32:56:12<37:47:33, 5.78s/it]
46%|████▋ | 20406/43939 [32:56:18<37:40:04, 5.76s/it]
46%|████▋ | 20407/43939 [32:56:23<37:32:36, 5.74s/it]
46%|████▋ | 20408/43939 [32:56:29<37:25:17, 5.73s/it]
46%|████▋ | 20409/43939 [32:56:35<37:24:58, 5.72s/it]
46%|████▋ | 20410/43939 [32:56:40<37:21:49, 5.72s/it]
46%|████▋ | 20411/43939 [32:56:46<37:21:57, 5.72s/it]
46%|████▋ | 20412/43939 [32:56:52<37:21:58, 5.72s/it]
46%|████▋ | 20413/43939 [32:56:58<37:19:53, 5.71s/it]
46%|████▋ | 20414/43939 [32:57:03<37:15:13, 5.70s/it]
46%|████▋ | 20415/43939 [32:57:09<37:12:38, 5.69s/it]
46%|████▋ | 20416/43939 [32:57:15<37:17:22, 5.71s/it]
46%|████▋ | 20417/43939 [32:57:20<37:18:41, 5.71s/it]
46%|████▋ | 20418/43939 [32:57:26<37:17:48, 5.71s/it]
46%|████▋ | 20419/43939 [32:57:32<37:15:05, 5.70s/it]
46%|████▋ | 20420/43939 [32:57:37<37:17:59, 5.71s/it]
46%|████▋ | 20421/43939 [32:57:43<37:17:11, 5.71s/it]
46%|████▋ | 20422/43939 [32:57:49<37:16:49, 5.71s/it]
46%|████▋ | 20423/43939 [32:57:55<37:14:00, 5.70s/it]
46%|████▋ | 20424/43939 [32:58:00<37:18:23, 5.71s/it]
46%|████▋ | 20425/43939 [32:58:06<37:14:24, 5.70s/it]
46%|████▋ | 20426/43939 [32:58:12<37:14:54, 5.70s/it]
46%|████▋ | 20427/43939 [32:58:18<37:45:16, 5.78s/it]
46%|████▋ | 20428/43939 [32:58:23<37:38:38, 5.76s/it]
46%|████▋ | 20429/43939 [32:58:29<37:30:37, 5.74s/it]
46%|████▋ | 20430/43939 [32:58:35<37:24:55, 5.73s/it]
46%|████▋ | 20431/43939 [32:58:40<37:22:48, 5.72s/it]
47%|████▋ | 20432/43939 [32:58:46<37:17:16, 5.71s/it]
47%|████▋ | 20433/43939 [32:58:52<37:17:41, 5.71s/it]
47%|████▋ | 20434/43939 [32:58:58<37:09:44, 5.69s/it]
47%|████▋ | 20435/43939 [32:59:03<37:10:00, 5.69s/it]
47%|████▋ | 20436/43939 [32:59:09<37:12:42, 5.70s/it]
47%|████▋ | 20437/43939 [32:59:15<37:10:26, 5.69s/it]
47%|████▋ | 20438/43939 [32:59:20<37:08:48, 5.69s/it]
47%|████▋ | 20439/43939 [32:59:26<37:16:50, 5.71s/it]
47%|████▋ | 20440/43939 [32:59:32<37:11:38, 5.70s/it]
47%|████▋ | 20441/43939 [32:59:37<37:10:05, 5.69s/it]
47%|████▋ | 20442/43939 [32:59:43<37:13:43, 5.70s/it]
47%|████▋ | 20443/43939 [32:59:49<37:15:52, 5.71s/it]
47%|████▋ | 20444/43939 [32:59:55<37:15:03, 5.71s/it]
47%|████▋ | 20445/43939 [33:00:00<37:08:30, 5.69s/it]
47%|████▋ | 20446/43939 [33:00:06<37:14:20, 5.71s/it]
47%|████▋ | 20447/43939 [33:00:12<37:17:13, 5.71s/it]
47%|████▋ | 20448/43939 [33:00:18<37:43:55, 5.78s/it]
47%|████▋ | 20449/43939 [33:00:23<37:37:16, 5.77s/it]
47%|████▋ | 20450/43939 [33:00:29<37:26:04, 5.74s/it]
{'loss': 0.9312, 'grad_norm': 1.2580296993255615, 'learning_rate': 3.227528854708656e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.12, 'epoch': 1.86}
47%|████▋ | 20450/43939 [33:00:29<37:26:04, 5.74s/it]
47%|████▋ | 20451/43939 [33:00:35<37:21:11, 5.73s/it]
47%|████▋ | 20452/43939 [33:00:40<37:14:06, 5.71s/it]
47%|████▋ | 20453/43939 [33:00:46<37:15:46, 5.71s/it]
47%|████▋ | 20454/43939 [33:00:52<37:09:17, 5.70s/it]
47%|████▋ | 20455/43939 [33:00:57<37:11:59, 5.70s/it]
47%|████▋ | 20456/43939 [33:01:03<37:12:35, 5.70s/it]
47%|████▋ | 20457/43939 [33:01:09<37:12:40, 5.70s/it]
47%|████▋ | 20458/43939 [33:01:15<37:12:58, 5.71s/it]
47%|████▋ | 20459/43939 [33:01:20<37:09:23, 5.70s/it]
47%|████▋ | 20460/43939 [33:01:26<37:05:17, 5.69s/it]
47%|████▋ | 20461/43939 [33:01:32<37:10:07, 5.70s/it]
47%|████▋ | 20462/43939 [33:01:37<37:06:12, 5.69s/it]
47%|████▋ | 20463/43939 [33:01:43<37:14:53, 5.71s/it]
47%|████▋ | 20464/43939 [33:01:49<37:10:06, 5.70s/it]
47%|████▋ | 20465/43939 [33:01:54<37:09:27, 5.70s/it]
47%|████▋ | 20466/43939 [33:02:00<37:08:24, 5.70s/it]
47%|████▋ | 20467/43939 [33:02:06<37:10:51, 5.70s/it]
47%|████▋ | 20468/43939 [33:02:12<37:15:17, 5.71s/it]
47%|████▋ | 20469/43939 [33:02:18<37:37:46, 5.77s/it]
47%|████▋ | 20470/43939 [33:02:23<37:28:07, 5.75s/it]
47%|████▋ | 20471/43939 [33:02:29<37:19:13, 5.72s/it]
47%|████▋ | 20472/43939 [33:02:35<37:18:03, 5.72s/it]
47%|████▋ | 20473/43939 [33:02:40<37:13:29, 5.71s/it]
47%|████▋ | 20474/43939 [33:02:46<37:09:11, 5.70s/it]
47%|████▋ | 20475/43939 [33:02:52<37:11:39, 5.71s/it]
47%|████▋ | 20476/43939 [33:02:57<37:06:08, 5.69s/it]
47%|████▋ | 20477/43939 [33:03:03<37:05:45, 5.69s/it]
47%|████▋ | 20478/43939 [33:03:09<37:09:56, 5.70s/it]
47%|████▋ | 20479/43939 [33:03:15<37:12:29, 5.71s/it]
47%|████▋ | 20480/43939 [33:03:20<37:06:57, 5.70s/it]
47%|████▋ | 20481/43939 [33:03:26<37:10:26, 5.70s/it]
47%|████▋ | 20482/43939 [33:03:32<37:10:31, 5.71s/it]
47%|████▋ | 20483/43939 [33:03:37<37:10:26, 5.71s/it]
47%|████▋ | 20484/43939 [33:03:43<37:07:52, 5.70s/it]
47%|████▋ | 20485/43939 [33:03:49<37:08:25, 5.70s/it]
47%|████▋ | 20486/43939 [33:03:54<37:02:23, 5.69s/it]
47%|████▋ | 20487/43939 [33:04:00<37:09:54, 5.71s/it]
47%|████▋ | 20488/43939 [33:04:06<37:08:48, 5.70s/it]
47%|████▋ | 20489/43939 [33:04:11<37:07:42, 5.70s/it]
47%|████▋ | 20490/43939 [33:04:17<37:38:02, 5.78s/it]
47%|████▋ | 20491/43939 [33:04:23<37:32:31, 5.76s/it]
47%|████▋ | 20492/43939 [33:04:29<37:24:06, 5.74s/it]
47%|████▋ | 20493/43939 [33:04:35<37:18:46, 5.73s/it]
47%|████▋ | 20494/43939 [33:04:40<37:17:56, 5.73s/it]
47%|████▋ | 20495/43939 [33:04:46<37:11:27, 5.71s/it]
47%|████▋ | 20496/43939 [33:04:52<37:09:57, 5.71s/it]
47%|████▋ | 20497/43939 [33:04:57<37:06:21, 5.70s/it]
47%|████▋ | 20498/43939 [33:05:03<37:07:50, 5.70s/it]
47%|████▋ | 20499/43939 [33:05:09<37:04:45, 5.69s/it]
47%|████▋ | 20500/43939 [33:05:14<36:57:30, 5.68s/it]
{'loss': 0.9247, 'grad_norm': 1.3489645719528198, 'learning_rate': 3.21802273875059e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.42, 'epoch': 1.87}
47%|████▋ | 20500/43939 [33:05:14<36:57:30, 5.68s/it]
47%|████▋ | 20501/43939 [33:05:20<37:07:37, 5.70s/it]
47%|████▋ | 20502/43939 [33:05:26<37:16:11, 5.72s/it]
47%|████▋ | 20503/43939 [33:05:32<37:19:06, 5.73s/it]
47%|████▋ | 20504/43939 [33:05:37<37:13:50, 5.72s/it]
47%|████▋ | 20505/43939 [33:05:43<37:11:38, 5.71s/it]
47%|████▋ | 20506/43939 [33:05:49<37:09:45, 5.71s/it]
47%|████▋ | 20507/43939 [33:05:54<37:07:24, 5.70s/it]
47%|████▋ | 20508/43939 [33:06:00<37:05:14, 5.70s/it]
47%|████▋ | 20509/43939 [33:06:06<37:05:39, 5.70s/it]
47%|████▋ | 20510/43939 [33:06:12<37:05:47, 5.70s/it]
47%|████▋ | 20511/43939 [33:06:17<37:00:15, 5.69s/it]
47%|████▋ | 20512/43939 [33:06:23<37:30:25, 5.76s/it]
47%|████▋ | 20513/43939 [33:06:29<37:25:59, 5.75s/it]
47%|████▋ | 20514/43939 [33:06:35<37:24:11, 5.75s/it]
47%|████▋ | 20515/43939 [33:06:40<37:13:00, 5.72s/it]
47%|████▋ | 20516/43939 [33:06:46<37:12:00, 5.72s/it]
47%|████▋ | 20517/43939 [33:06:52<37:07:18, 5.71s/it]
47%|████▋ | 20518/43939 [33:06:57<37:08:02, 5.71s/it]
47%|████▋ | 20519/43939 [33:07:03<37:08:28, 5.71s/it]
47%|████▋ | 20520/43939 [33:07:09<37:10:12, 5.71s/it]
47%|████▋ | 20521/43939 [33:07:14<37:09:49, 5.71s/it]
47%|████▋ | 20522/43939 [33:07:20<37:08:06, 5.71s/it]
47%|████▋ | 20523/43939 [33:07:26<37:08:28, 5.71s/it]
47%|████▋ | 20524/43939 [33:07:32<37:08:20, 5.71s/it]
47%|████▋ | 20525/43939 [33:07:37<37:05:15, 5.70s/it]
47%|████▋ | 20526/43939 [33:07:43<37:04:48, 5.70s/it]
47%|████▋ | 20527/43939 [33:07:49<37:02:29, 5.70s/it]
47%|████▋ | 20528/43939 [33:07:55<37:18:30, 5.74s/it]
47%|████▋ | 20529/43939 [33:08:00<37:18:42, 5.74s/it]
47%|████▋ | 20530/43939 [33:08:06<37:19:30, 5.74s/it]
47%|████▋ | 20531/43939 [33:08:12<37:14:21, 5.73s/it]
47%|████▋ | 20532/43939 [33:08:17<37:08:51, 5.71s/it]
47%|████▋ | 20533/43939 [33:08:23<37:33:27, 5.78s/it]
47%|████▋ | 20534/43939 [33:08:29<37:20:47, 5.74s/it]
47%|████▋ | 20535/43939 [33:08:35<37:12:29, 5.72s/it]
47%|████▋ | 20536/43939 [33:08:40<37:07:12, 5.71s/it]
47%|████▋ | 20537/43939 [33:08:46<37:03:33, 5.70s/it]
47%|████▋ | 20538/43939 [33:08:52<37:01:30, 5.70s/it]
47%|████▋ | 20539/43939 [33:08:57<37:01:49, 5.70s/it]
47%|████▋ | 20540/43939 [33:09:03<37:04:18, 5.70s/it]
47%|████▋ | 20541/43939 [33:09:09<37:05:03, 5.71s/it]
47%|████▋ | 20542/43939 [33:09:15<37:11:06, 5.72s/it]
47%|████▋ | 20543/43939 [33:09:20<37:04:31, 5.70s/it]
47%|████▋ | 20544/43939 [33:09:26<37:06:47, 5.71s/it]
47%|████▋ | 20545/43939 [33:09:32<37:05:02, 5.71s/it]
47%|████▋ | 20546/43939 [33:09:37<37:01:13, 5.70s/it]
47%|████▋ | 20547/43939 [33:09:43<37:06:32, 5.71s/it]
47%|████▋ | 20548/43939 [33:09:49<37:02:50, 5.70s/it]
47%|████▋ | 20549/43939 [33:09:54<37:03:32, 5.70s/it]
47%|████▋ | 20550/43939 [33:10:00<36:59:28, 5.69s/it]
{'loss': 0.9119, 'grad_norm': 1.118877649307251, 'learning_rate': 3.208505294296491e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.96, 'epoch': 1.87}
47%|████▋ | 20550/43939 [33:10:00<36:59:28, 5.69s/it]
47%|████▋ | 20551/43939 [33:10:06<37:01:19, 5.70s/it]
47%|████▋ | 20552/43939 [33:10:12<37:03:15, 5.70s/it]
47%|████▋ | 20553/43939 [33:10:17<36:56:38, 5.69s/it]
47%|████▋ | 20554/43939 [33:10:23<37:21:02, 5.75s/it]
47%|████▋ | 20555/43939 [33:10:29<37:14:38, 5.73s/it]
47%|████▋ | 20556/43939 [33:10:35<37:13:43, 5.73s/it]
47%|████▋ | 20557/43939 [33:10:40<37:08:47, 5.72s/it]
47%|████▋ | 20558/43939 [33:10:46<37:07:15, 5.72s/it]
47%|████▋ | 20559/43939 [33:10:52<37:06:46, 5.71s/it]
47%|████▋ | 20560/43939 [33:10:57<37:09:12, 5.72s/it]
47%|████▋ | 20561/43939 [33:11:03<37:01:42, 5.70s/it]
47%|████▋ | 20562/43939 [33:11:09<36:59:39, 5.70s/it]
47%|████▋ | 20563/43939 [33:11:14<37:03:03, 5.71s/it]
47%|████▋ | 20564/43939 [33:11:20<36:59:20, 5.70s/it]
47%|████▋ | 20565/43939 [33:11:26<36:59:06, 5.70s/it]
47%|████▋ | 20566/43939 [33:11:32<37:00:30, 5.70s/it]
47%|████▋ | 20567/43939 [33:11:37<36:57:26, 5.69s/it]
47%|████▋ | 20568/43939 [33:11:43<36:55:21, 5.69s/it]
47%|████▋ | 20569/43939 [33:11:49<36:56:24, 5.69s/it]
47%|████▋ | 20570/43939 [33:11:54<36:56:30, 5.69s/it]
47%|████▋ | 20571/43939 [33:12:00<36:58:49, 5.70s/it]
47%|████▋ | 20572/43939 [33:12:06<36:59:03, 5.70s/it]
47%|████▋ | 20573/43939 [33:12:11<37:02:16, 5.71s/it]
47%|████▋ | 20574/43939 [33:12:17<37:04:32, 5.71s/it]
47%|████▋ | 20575/43939 [33:12:23<37:36:49, 5.80s/it]
47%|████▋ | 20576/43939 [33:12:29<37:28:25, 5.77s/it]
47%|████▋ | 20577/43939 [33:12:35<37:20:14, 5.75s/it]
47%|████▋ | 20578/43939 [33:12:40<37:16:17, 5.74s/it]
47%|████▋ | 20579/43939 [33:12:46<37:10:10, 5.73s/it]
47%|████▋ | 20580/43939 [33:12:52<37:10:28, 5.73s/it]
47%|████▋ | 20581/43939 [33:12:57<37:03:00, 5.71s/it]
47%|████▋ | 20582/43939 [33:13:03<37:09:47, 5.73s/it]
47%|████▋ | 20583/43939 [33:13:09<37:09:20, 5.73s/it]
47%|████▋ | 20584/43939 [33:13:15<37:08:48, 5.73s/it]
47%|████▋ | 20585/43939 [33:13:20<37:03:44, 5.71s/it]
47%|████▋ | 20586/43939 [33:13:26<36:57:31, 5.70s/it]
47%|████▋ | 20587/43939 [33:13:32<36:56:57, 5.70s/it]
47%|████▋ | 20588/43939 [33:13:37<36:55:09, 5.69s/it]
47%|████▋ | 20589/43939 [33:13:43<36:56:49, 5.70s/it]
47%|████▋ | 20590/43939 [33:13:49<37:04:04, 5.72s/it]
47%|████▋ | 20591/43939 [33:13:54<37:03:55, 5.72s/it]
47%|████▋ | 20592/43939 [33:14:00<37:03:22, 5.71s/it]
47%|████▋ | 20593/43939 [33:14:06<36:58:00, 5.70s/it]
47%|████▋ | 20594/43939 [33:14:12<36:59:12, 5.70s/it]
47%|████▋ | 20595/43939 [33:14:17<37:02:59, 5.71s/it]
47%|████▋ | 20596/43939 [33:14:23<37:00:55, 5.71s/it]
47%|████▋ | 20597/43939 [33:14:29<37:31:12, 5.79s/it]
47%|████▋ | 20598/43939 [33:14:35<37:19:09, 5.76s/it]
47%|████▋ | 20599/43939 [33:14:40<37:10:00, 5.73s/it]
47%|████▋ | 20600/43939 [33:14:46<37:02:22, 5.71s/it]
{'loss': 0.9069, 'grad_norm': 1.4261683225631714, 'learning_rate': 3.198976671506409e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.09, 'epoch': 1.88}
47%|████▋ | 20600/43939 [33:14:46<37:02:22, 5.71s/it]
47%|████▋ | 20601/43939 [33:14:52<36:59:23, 5.71s/it]
47%|████▋ | 20602/43939 [33:14:57<37:00:12, 5.71s/it]
47%|████▋ | 20603/43939 [33:15:03<37:05:46, 5.72s/it]
47%|████▋ | 20604/43939 [33:15:09<37:05:30, 5.72s/it]
47%|████▋ | 20605/43939 [33:15:15<37:00:44, 5.71s/it]
47%|████▋ | 20606/43939 [33:15:20<36:57:34, 5.70s/it]
47%|████▋ | 20607/43939 [33:15:26<36:57:24, 5.70s/it]
47%|████▋ | 20608/43939 [33:15:32<36:58:29, 5.71s/it]
47%|████▋ | 20609/43939 [33:15:37<36:57:44, 5.70s/it]
47%|████▋ | 20610/43939 [33:15:43<36:50:53, 5.69s/it]
47%|████▋ | 20611/43939 [33:15:49<36:49:08, 5.68s/it]
47%|████▋ | 20612/43939 [33:15:54<36:49:51, 5.68s/it]
47%|████▋ | 20613/43939 [33:16:00<36:50:55, 5.69s/it]
47%|████▋ | 20614/43939 [33:16:06<36:53:40, 5.69s/it]
47%|████▋ | 20615/43939 [33:16:12<37:00:09, 5.71s/it]
47%|████▋ | 20616/43939 [33:16:17<36:59:30, 5.71s/it]
47%|████▋ | 20617/43939 [33:16:23<37:00:30, 5.71s/it]
47%|████▋ | 20618/43939 [33:16:29<37:36:39, 5.81s/it]
47%|████▋ | 20619/43939 [33:16:35<37:26:01, 5.78s/it]
47%|████▋ | 20620/43939 [33:16:40<37:13:19, 5.75s/it]
47%|████▋ | 20621/43939 [33:16:46<37:10:55, 5.74s/it]
47%|████▋ | 20622/43939 [33:16:52<37:04:57, 5.73s/it]
47%|████▋ | 20623/43939 [33:16:57<37:02:30, 5.72s/it]
47%|████▋ | 20624/43939 [33:17:03<37:03:53, 5.72s/it]
47%|████▋ | 20625/43939 [33:17:09<36:55:18, 5.70s/it]
47%|████▋ | 20626/43939 [33:17:15<36:57:29, 5.71s/it]
47%|████▋ | 20627/43939 [33:17:20<36:55:02, 5.70s/it]
47%|████▋ | 20628/43939 [33:17:26<36:53:58, 5.70s/it]
47%|████▋ | 20629/43939 [33:17:32<37:00:08, 5.71s/it]
47%|████▋ | 20630/43939 [33:17:37<36:57:30, 5.71s/it]
47%|████▋ | 20631/43939 [33:17:43<36:56:39, 5.71s/it]
47%|████▋ | 20632/43939 [33:17:49<36:58:54, 5.71s/it]
47%|████▋ | 20633/43939 [33:17:55<36:53:22, 5.70s/it]
47%|████▋ | 20634/43939 [33:18:00<36:49:56, 5.69s/it]
47%|████▋ | 20635/43939 [33:18:06<36:47:40, 5.68s/it]
47%|████▋ | 20636/43939 [33:18:12<36:48:43, 5.69s/it]
47%|████▋ | 20637/43939 [33:18:17<36:53:00, 5.70s/it]
47%|████▋ | 20638/43939 [33:18:23<36:57:48, 5.71s/it]
47%|████▋ | 20639/43939 [33:18:29<37:28:57, 5.79s/it]
47%|████▋ | 20640/43939 [33:18:35<37:15:35, 5.76s/it]
47%|████▋ | 20641/43939 [33:18:40<37:10:50, 5.75s/it]
47%|████▋ | 20642/43939 [33:18:46<37:04:30, 5.73s/it]
47%|████▋ | 20643/43939 [33:18:52<37:01:23, 5.72s/it]
47%|████▋ | 20644/43939 [33:18:57<36:54:29, 5.70s/it]
47%|████▋ | 20645/43939 [33:19:03<36:49:34, 5.69s/it]
47%|████▋ | 20646/43939 [33:19:09<36:48:00, 5.69s/it]
47%|████▋ | 20647/43939 [33:19:15<36:55:27, 5.71s/it]
47%|████▋ | 20648/43939 [33:19:20<36:50:36, 5.69s/it]
47%|████▋ | 20649/43939 [33:19:26<36:52:31, 5.70s/it]
47%|████▋ | 20650/43939 [33:19:32<36:49:18, 5.69s/it]
{'loss': 0.9471, 'grad_norm': 1.2352209091186523, 'learning_rate': 3.1894370207167625e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.79, 'epoch': 1.88}
47%|████▋ | 20650/43939 [33:19:32<36:49:18, 5.69s/it]
47%|████▋ | 20651/43939 [33:19:37<36:52:20, 5.70s/it]
47%|████▋ | 20652/43939 [33:19:43<36:52:10, 5.70s/it]
47%|████▋ | 20653/43939 [33:19:49<36:49:00, 5.69s/it]
47%|████▋ | 20654/43939 [33:19:54<36:45:25, 5.68s/it]
47%|████▋ | 20655/43939 [33:20:00<36:49:12, 5.69s/it]
47%|████▋ | 20656/43939 [33:20:06<36:53:05, 5.70s/it]
47%|████▋ | 20657/43939 [33:20:11<36:53:36, 5.70s/it]
47%|████▋ | 20658/43939 [33:20:17<36:52:25, 5.70s/it]
47%|████▋ | 20659/43939 [33:20:23<36:47:04, 5.69s/it]
47%|████▋ | 20660/43939 [33:20:29<37:13:57, 5.76s/it]
47%|████▋ | 20661/43939 [33:20:35<37:15:25, 5.76s/it]
47%|████▋ | 20662/43939 [33:20:40<37:03:29, 5.73s/it]
47%|████▋ | 20663/43939 [33:20:46<36:58:20, 5.72s/it]
47%|████▋ | 20664/43939 [33:20:52<37:05:00, 5.74s/it]
47%|████▋ | 20665/43939 [33:20:57<36:59:50, 5.72s/it]
47%|████▋ | 20666/43939 [33:21:03<37:02:09, 5.73s/it]
47%|████▋ | 20667/43939 [33:21:09<37:00:07, 5.72s/it]
47%|████▋ | 20668/43939 [33:21:14<36:55:32, 5.71s/it]
47%|████▋ | 20669/43939 [33:21:20<36:56:36, 5.72s/it]
47%|████▋ | 20670/43939 [33:21:26<36:53:21, 5.71s/it]
47%|████▋ | 20671/43939 [33:21:32<36:54:37, 5.71s/it]
47%|████▋ | 20672/43939 [33:21:37<36:51:51, 5.70s/it]
47%|████▋ | 20673/43939 [33:21:43<37:00:36, 5.73s/it]
47%|████▋ | 20674/43939 [33:21:49<36:56:23, 5.72s/it]
47%|████▋ | 20675/43939 [33:21:54<36:52:08, 5.71s/it]
47%|████▋ | 20676/43939 [33:22:00<36:51:08, 5.70s/it]
47%|████▋ | 20677/43939 [33:22:06<36:53:48, 5.71s/it]
47%|████▋ | 20678/43939 [33:22:12<36:54:45, 5.71s/it]
47%|████▋ | 20679/43939 [33:22:17<36:50:22, 5.70s/it]
47%|████▋ | 20680/43939 [33:22:23<36:45:00, 5.69s/it]
47%|████▋ | 20681/43939 [33:22:29<36:49:56, 5.70s/it]
47%|████▋ | 20682/43939 [33:22:35<37:16:39, 5.77s/it]
47%|████▋ | 20683/43939 [33:22:40<37:08:36, 5.75s/it]
47%|████▋ | 20684/43939 [33:22:46<37:04:33, 5.74s/it]
47%|████▋ | 20685/43939 [33:22:52<36:58:02, 5.72s/it]
47%|████▋ | 20686/43939 [33:22:57<36:58:02, 5.72s/it]
47%|████▋ | 20687/43939 [33:23:03<36:54:23, 5.71s/it]
47%|████▋ | 20688/43939 [33:23:09<36:51:52, 5.71s/it]
47%|████▋ | 20689/43939 [33:23:14<36:48:36, 5.70s/it]
47%|████▋ | 20690/43939 [33:23:20<36:47:09, 5.70s/it]
47%|████▋ | 20691/43939 [33:23:26<36:45:16, 5.69s/it]
47%|████▋ | 20692/43939 [33:23:32<36:46:36, 5.70s/it]
47%|████▋ | 20693/43939 [33:23:37<36:44:55, 5.69s/it]
47%|████▋ | 20694/43939 [33:23:43<36:46:26, 5.70s/it]
47%|████▋ | 20695/43939 [33:23:49<36:48:58, 5.70s/it]
47%|████▋ | 20696/43939 [33:23:54<36:50:37, 5.71s/it]
47%|████▋ | 20697/43939 [33:24:00<36:48:42, 5.70s/it]
47%|████▋ | 20698/43939 [33:24:06<36:51:06, 5.71s/it]
47%|████▋ | 20699/43939 [33:24:11<36:49:56, 5.71s/it]
47%|████▋ | 20700/43939 [33:24:17<36:45:31, 5.69s/it]
{'loss': 0.9172, 'grad_norm': 1.480117678642273, 'learning_rate': 3.179886492437958e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.22, 'epoch': 1.88}
47%|████▋ | 20700/43939 [33:24:17<36:45:31, 5.69s/it]
47%|████▋ | 20701/43939 [33:24:23<36:47:24, 5.70s/it]
47%|████▋ | 20702/43939 [33:24:29<36:50:57, 5.71s/it]
47%|████▋ | 20703/43939 [33:24:34<36:50:08, 5.71s/it]
47%|████▋ | 20704/43939 [33:24:40<37:15:46, 5.77s/it]
47%|████▋ | 20705/43939 [33:24:46<37:09:12, 5.76s/it]
47%|████▋ | 20706/43939 [33:24:52<37:00:54, 5.74s/it]
47%|████▋ | 20707/43939 [33:24:57<36:58:09, 5.73s/it]
47%|████▋ | 20708/43939 [33:25:03<36:50:54, 5.71s/it]
47%|████▋ | 20709/43939 [33:25:09<36:46:28, 5.70s/it]
47%|████▋ | 20710/43939 [33:25:14<36:46:08, 5.70s/it]
47%|████▋ | 20711/43939 [33:25:20<36:48:35, 5.70s/it]
47%|████▋ | 20712/43939 [33:25:26<36:55:18, 5.72s/it]
47%|████▋ | 20713/43939 [33:25:32<36:51:13, 5.71s/it]
47%|████▋ | 20714/43939 [33:25:37<36:50:22, 5.71s/it]
47%|████▋ | 20715/43939 [33:25:43<36:47:43, 5.70s/it]
47%|████▋ | 20716/43939 [33:25:49<36:51:05, 5.71s/it]
47%|████▋ | 20717/43939 [33:25:54<36:51:13, 5.71s/it]
47%|████▋ | 20718/43939 [33:26:00<36:47:40, 5.70s/it]
47%|████▋ | 20719/43939 [33:26:06<36:49:04, 5.71s/it]
47%|████▋ | 20720/43939 [33:26:12<36:48:07, 5.71s/it]
47%|████▋ | 20721/43939 [33:26:17<36:45:20, 5.70s/it]
47%|████▋ | 20722/43939 [33:26:23<36:42:16, 5.69s/it]
47%|████▋ | 20723/43939 [33:26:29<36:45:08, 5.70s/it]
47%|████▋ | 20724/43939 [33:26:35<37:11:46, 5.77s/it]
47%|████▋ | 20725/43939 [33:26:40<37:05:20, 5.75s/it]
47%|████▋ | 20726/43939 [33:26:46<37:05:33, 5.75s/it]
47%|████▋ | 20727/43939 [33:26:52<36:58:13, 5.73s/it]
47%|████▋ | 20728/43939 [33:26:57<36:50:43, 5.71s/it]
47%|████▋ | 20729/43939 [33:27:03<36:45:58, 5.70s/it]
47%|████▋ | 20730/43939 [33:27:09<36:44:43, 5.70s/it]
47%|████▋ | 20731/43939 [33:27:14<36:49:12, 5.71s/it]
47%|████▋ | 20732/43939 [33:27:20<36:50:55, 5.72s/it]
47%|████▋ | 20733/43939 [33:27:26<36:50:23, 5.72s/it]
47%|████▋ | 20734/43939 [33:27:32<36:49:29, 5.71s/it]
47%|████▋ | 20735/43939 [33:27:37<36:45:21, 5.70s/it]
47%|████▋ | 20736/43939 [33:27:43<36:44:39, 5.70s/it]
47%|████▋ | 20737/43939 [33:27:49<36:43:44, 5.70s/it]
47%|████▋ | 20738/43939 [33:27:54<36:41:14, 5.69s/it]
47%|████▋ | 20739/43939 [33:28:00<36:46:59, 5.71s/it]
47%|████▋ | 20740/43939 [33:28:06<36:50:33, 5.72s/it]
47%|████▋ | 20741/43939 [33:28:12<36:53:28, 5.72s/it]
47%|████▋ | 20742/43939 [33:28:17<36:56:09, 5.73s/it]
47%|████▋ | 20743/43939 [33:28:23<36:51:53, 5.72s/it]
47%|████▋ | 20744/43939 [33:28:29<36:50:19, 5.72s/it]
47%|████▋ | 20745/43939 [33:28:35<37:19:24, 5.79s/it]
47%|████▋ | 20746/43939 [33:28:40<37:06:32, 5.76s/it]
47%|████▋ | 20747/43939 [33:28:46<37:02:06, 5.75s/it]
47%|████▋ | 20748/43939 [33:28:52<36:59:13, 5.74s/it]
47%|████▋ | 20749/43939 [33:28:58<36:56:55, 5.74s/it]
47%|████▋ | 20750/43939 [33:29:03<36:53:47, 5.73s/it]
{'loss': 0.9114, 'grad_norm': 1.1455492973327637, 'learning_rate': 3.170325237352024e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.15, 'epoch': 1.89}
47%|████▋ | 20750/43939 [33:29:03<36:53:47, 5.73s/it]
47%|████▋ | 20751/43939 [33:29:09<36:50:31, 5.72s/it]
47%|████▋ | 20752/43939 [33:29:15<36:55:10, 5.73s/it]
47%|████▋ | 20753/43939 [33:29:20<36:51:08, 5.72s/it]
47%|████▋ | 20754/43939 [33:29:26<36:48:39, 5.72s/it]
47%|████▋ | 20755/43939 [33:29:32<36:48:45, 5.72s/it]
47%|████▋ | 20756/43939 [33:29:38<36:46:03, 5.71s/it]
47%|████▋ | 20757/43939 [33:29:43<36:42:43, 5.70s/it]
47%|████▋ | 20758/43939 [33:29:49<36:43:59, 5.70s/it]
47%|████▋ | 20759/43939 [33:29:55<36:39:27, 5.69s/it]
47%|████▋ | 20760/43939 [33:30:00<36:36:56, 5.69s/it]
47%|████▋ | 20761/43939 [33:30:06<36:40:22, 5.70s/it]
47%|████▋ | 20762/43939 [33:30:12<36:45:00, 5.71s/it]
47%|████▋ | 20763/43939 [33:30:17<36:40:01, 5.70s/it]
47%|████▋ | 20764/43939 [33:30:23<36:42:09, 5.70s/it]
47%|████▋ | 20765/43939 [33:30:29<36:41:51, 5.70s/it]
47%|████▋ | 20766/43939 [33:30:35<37:08:24, 5.77s/it]
47%|████▋ | 20767/43939 [33:30:40<37:04:21, 5.76s/it]
47%|████▋ | 20768/43939 [33:30:46<37:01:29, 5.75s/it]
47%|████▋ | 20769/43939 [33:30:52<36:55:03, 5.74s/it]
47%|████▋ | 20770/43939 [33:30:58<36:51:42, 5.73s/it]
47%|████▋ | 20771/43939 [33:31:03<36:53:04, 5.73s/it]
47%|████▋ | 20772/43939 [33:31:09<36:48:22, 5.72s/it]
47%|████▋ | 20773/43939 [33:31:15<36:44:33, 5.71s/it]
47%|████▋ | 20774/43939 [33:31:20<36:43:25, 5.71s/it]
47%|████▋ | 20775/43939 [33:31:26<36:40:03, 5.70s/it]
47%|████▋ | 20776/43939 [33:31:32<36:40:34, 5.70s/it]
47%|████▋ | 20777/43939 [33:31:37<36:39:06, 5.70s/it]
47%|████▋ | 20778/43939 [33:31:43<36:38:41, 5.70s/it]
47%|████▋ | 20779/43939 [33:31:49<36:39:43, 5.70s/it]
47%|████▋ | 20780/43939 [33:31:55<36:41:43, 5.70s/it]
47%|████▋ | 20781/43939 [33:32:00<36:47:29, 5.72s/it]
47%|████▋ | 20782/43939 [33:32:06<36:47:17, 5.72s/it]
47%|████▋ | 20783/43939 [33:32:12<36:47:10, 5.72s/it]
47%|████▋ | 20784/43939 [33:32:17<36:43:09, 5.71s/it]
47%|████▋ | 20785/43939 [33:32:23<36:43:45, 5.71s/it]
47%|████▋ | 20786/43939 [33:32:29<36:42:44, 5.71s/it]
47%|████▋ | 20787/43939 [33:32:35<37:12:46, 5.79s/it]
47%|████▋ | 20788/43939 [33:32:41<37:02:10, 5.76s/it]
47%|████▋ | 20789/43939 [33:32:46<36:57:41, 5.75s/it]
47%|████▋ | 20790/43939 [33:32:52<36:52:51, 5.74s/it]
47%|████▋ | 20791/43939 [33:32:58<36:49:31, 5.73s/it]
47%|████▋ | 20792/43939 [33:33:03<36:50:55, 5.73s/it]
47%|████▋ | 20793/43939 [33:33:09<36:47:00, 5.72s/it]
47%|████▋ | 20794/43939 [33:33:15<36:48:31, 5.73s/it]
47%|████▋ | 20795/43939 [33:33:21<36:43:00, 5.71s/it]
47%|████▋ | 20796/43939 [33:33:26<36:42:52, 5.71s/it]
47%|████▋ | 20797/43939 [33:33:32<36:49:44, 5.73s/it]
47%|████▋ | 20798/43939 [33:33:38<36:42:33, 5.71s/it]
47%|████▋ | 20799/43939 [33:33:43<36:43:27, 5.71s/it]
47%|████▋ | 20800/43939 [33:33:49<36:40:33, 5.71s/it]
{'loss': 0.9251, 'grad_norm': 1.2906856536865234, 'learning_rate': 3.160753406310229e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.01, 'epoch': 1.89}
47%|████▋ | 20800/43939 [33:33:49<36:40:33, 5.71s/it]
47%|████▋ | 20801/43939 [33:33:55<36:40:40, 5.71s/it]
47%|████▋ | 20802/43939 [33:34:00<36:35:44, 5.69s/it]
47%|████▋ | 20803/43939 [33:34:06<36:37:39, 5.70s/it]
47%|████▋ | 20804/43939 [33:34:12<36:39:52, 5.71s/it]
47%|████▋ | 20805/43939 [33:34:18<36:35:26, 5.69s/it]
47%|████▋ | 20806/43939 [33:34:23<36:44:57, 5.72s/it]
47%|████▋ | 20807/43939 [33:34:29<36:48:16, 5.73s/it]
47%|████▋ | 20808/43939 [33:34:35<37:07:04, 5.78s/it]
47%|████▋ | 20809/43939 [33:34:41<36:58:01, 5.75s/it]
47%|████▋ | 20810/43939 [33:34:46<36:52:40, 5.74s/it]
47%|████▋ | 20811/43939 [33:34:52<36:47:17, 5.73s/it]
47%|████▋ | 20812/43939 [33:34:58<36:39:16, 5.71s/it]
47%|████▋ | 20813/43939 [33:35:03<36:38:48, 5.70s/it]
47%|████▋ | 20814/43939 [33:35:09<36:39:41, 5.71s/it]
47%|████▋ | 20815/43939 [33:35:15<36:39:09, 5.71s/it]
47%|████▋ | 20816/43939 [33:35:21<36:40:14, 5.71s/it]
47%|████▋ | 20817/43939 [33:35:26<36:43:36, 5.72s/it]
47%|████▋ | 20818/43939 [33:35:32<36:34:56, 5.70s/it]
47%|████▋ | 20819/43939 [33:35:38<36:37:19, 5.70s/it]
47%|████▋ | 20820/43939 [33:35:43<36:35:09, 5.70s/it]
47%|████▋ | 20821/43939 [33:35:49<36:36:07, 5.70s/it]
47%|████▋ | 20822/43939 [33:35:55<36:41:16, 5.71s/it]
47%|████▋ | 20823/43939 [33:36:01<36:43:49, 5.72s/it]
47%|████▋ | 20824/43939 [33:36:06<36:38:49, 5.71s/it]
47%|████▋ | 20825/43939 [33:36:12<36:43:12, 5.72s/it]
47%|████▋ | 20826/43939 [33:36:18<36:48:14, 5.73s/it]
47%|████▋ | 20827/43939 [33:36:23<36:44:57, 5.72s/it]
47%|████▋ | 20828/43939 [33:36:29<36:42:43, 5.72s/it]
47%|████▋ | 20829/43939 [33:36:35<36:39:44, 5.71s/it]
47%|████▋ | 20830/43939 [33:36:41<37:01:35, 5.77s/it]
47%|████▋ | 20831/43939 [33:36:46<36:52:30, 5.74s/it]
47%|████▋ | 20832/43939 [33:36:52<36:47:45, 5.73s/it]
47%|████▋ | 20833/43939 [33:36:58<36:41:56, 5.72s/it]
47%|████▋ | 20834/43939 [33:37:04<36:40:24, 5.71s/it]
47%|████▋ | 20835/43939 [33:37:09<36:39:00, 5.71s/it]
47%|████▋ | 20836/43939 [33:37:15<36:39:23, 5.71s/it]
47%|████▋ | 20837/43939 [33:37:21<36:37:42, 5.71s/it]
47%|████▋ | 20838/43939 [33:37:26<36:41:59, 5.72s/it]
47%|████▋ | 20839/43939 [33:37:32<36:44:32, 5.73s/it]
47%|████▋ | 20840/43939 [33:37:38<36:49:51, 5.74s/it]
47%|████▋ | 20841/43939 [33:37:44<36:44:30, 5.73s/it]
47%|████▋ | 20842/43939 [33:37:49<36:43:44, 5.72s/it]
47%|████▋ | 20843/43939 [33:37:55<36:40:42, 5.72s/it]
47%|████▋ | 20844/43939 [33:38:01<36:39:08, 5.71s/it]
47%|████▋ | 20845/43939 [33:38:06<36:35:09, 5.70s/it]
47%|████▋ | 20846/43939 [33:38:12<36:36:35, 5.71s/it]
47%|████▋ | 20847/43939 [33:38:18<36:36:29, 5.71s/it]
47%|████▋ | 20848/43939 [33:38:24<36:38:41, 5.71s/it]
47%|████▋ | 20849/43939 [33:38:29<36:39:10, 5.71s/it]
47%|████▋ | 20850/43939 [33:38:35<36:41:16, 5.72s/it]
{'loss': 0.9145, 'grad_norm': 1.0654765367507935, 'learning_rate': 3.1511711503307e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.46, 'epoch': 1.9}
47%|████▋ | 20850/43939 [33:38:35<36:41:16, 5.72s/it]
47%|████▋ | 20851/43939 [33:38:41<37:03:46, 5.78s/it]
47%|████▋ | 20852/43939 [33:38:47<36:50:42, 5.75s/it]
47%|████▋ | 20853/43939 [33:38:52<36:45:14, 5.73s/it]
47%|████▋ | 20854/43939 [33:38:58<36:38:37, 5.71s/it]
47%|████▋ | 20855/43939 [33:39:04<36:45:43, 5.73s/it]
47%|████▋ | 20856/43939 [33:39:09<36:38:34, 5.71s/it]
47%|████▋ | 20857/43939 [33:39:15<36:35:09, 5.71s/it]
47%|████▋ | 20858/43939 [33:39:21<36:31:58, 5.70s/it]
47%|████▋ | 20859/43939 [33:39:26<36:30:01, 5.69s/it]
47%|████▋ | 20860/43939 [33:39:32<36:29:06, 5.69s/it]
47%|████▋ | 20861/43939 [33:39:38<36:29:33, 5.69s/it]
47%|████▋ | 20862/43939 [33:39:44<36:28:30, 5.69s/it]
47%|████▋ | 20863/43939 [33:39:49<36:30:08, 5.69s/it]
47%|████▋ | 20864/43939 [33:39:55<36:26:56, 5.69s/it]
47%|████▋ | 20865/43939 [33:40:01<36:25:10, 5.68s/it]
47%|████▋ | 20866/43939 [33:40:06<36:31:40, 5.70s/it]
47%|████▋ | 20867/43939 [33:40:12<36:29:02, 5.69s/it]
47%|████▋ | 20868/43939 [33:40:18<36:29:51, 5.70s/it]
47%|████▋ | 20869/43939 [33:40:23<36:29:48, 5.70s/it]
47%|████▋ | 20870/43939 [33:40:29<36:33:58, 5.71s/it]
47%|████▋ | 20871/43939 [33:40:35<36:29:58, 5.70s/it]
48%|████▊ | 20872/43939 [33:40:41<37:02:47, 5.78s/it]
48%|████▊ | 20873/43939 [33:40:46<36:54:48, 5.76s/it]
48%|████▊ | 20874/43939 [33:40:52<36:51:16, 5.75s/it]
48%|████▊ | 20875/43939 [33:40:58<36:43:31, 5.73s/it]
48%|████▊ | 20876/43939 [33:41:04<36:37:29, 5.72s/it]
48%|████▊ | 20877/43939 [33:41:09<36:30:10, 5.70s/it]
48%|████▊ | 20878/43939 [33:41:15<36:30:08, 5.70s/it]
48%|████▊ | 20879/43939 [33:41:21<36:30:32, 5.70s/it]
48%|████▊ | 20880/43939 [33:41:26<36:29:56, 5.70s/it]
48%|████▊ | 20881/43939 [33:41:32<36:32:16, 5.70s/it]
48%|████▊ | 20882/43939 [33:41:38<36:31:07, 5.70s/it]
48%|████▊ | 20883/43939 [33:41:43<36:30:51, 5.70s/it]
48%|████▊ | 20884/43939 [33:41:49<36:37:06, 5.72s/it]
48%|████▊ | 20885/43939 [33:41:55<36:33:54, 5.71s/it]
48%|████▊ | 20886/43939 [33:42:01<36:36:01, 5.72s/it]
48%|████▊ | 20887/43939 [33:42:06<36:33:57, 5.71s/it]
48%|████▊ | 20888/43939 [33:42:12<36:35:41, 5.72s/it]
48%|████▊ | 20889/43939 [33:42:18<36:37:06, 5.72s/it]
48%|████▊ | 20890/43939 [33:42:23<36:34:21, 5.71s/it]
48%|████▊ | 20891/43939 [33:42:29<36:31:15, 5.70s/it]
48%|████▊ | 20892/43939 [33:42:35<36:28:48, 5.70s/it]
48%|████▊ | 20893/43939 [33:42:41<36:50:16, 5.75s/it]
48%|████▊ | 20894/43939 [33:42:46<36:44:49, 5.74s/it]
48%|████▊ | 20895/43939 [33:42:52<36:42:55, 5.74s/it]
48%|████▊ | 20896/43939 [33:42:58<36:36:21, 5.72s/it]
48%|████▊ | 20897/43939 [33:43:04<36:36:25, 5.72s/it]
48%|████▊ | 20898/43939 [33:43:09<36:33:51, 5.71s/it]
48%|████▊ | 20899/43939 [33:43:15<36:31:23, 5.71s/it]
48%|████▊ | 20900/43939 [33:43:21<36:32:47, 5.71s/it]
{'loss': 0.8849, 'grad_norm': 1.24265456199646, 'learning_rate': 3.141578620596043e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 457.22, 'epoch': 1.9}
48%|████▊ | 20900/43939 [33:43:21<36:32:47, 5.71s/it]
48%|████▊ | 20901/43939 [33:43:26<36:35:56, 5.72s/it]
48%|████▊ | 20902/43939 [33:43:32<36:37:25, 5.72s/it]
48%|████▊ | 20903/43939 [33:43:38<36:37:37, 5.72s/it]
48%|████▊ | 20904/43939 [33:43:44<36:32:31, 5.71s/it]
48%|████▊ | 20905/43939 [33:43:49<36:30:45, 5.71s/it]
48%|████▊ | 20906/43939 [33:43:55<36:32:57, 5.71s/it]
48%|████▊ | 20907/43939 [33:44:01<36:31:08, 5.71s/it]
48%|████▊ | 20908/43939 [33:44:06<36:29:40, 5.70s/it]
48%|████▊ | 20909/43939 [33:44:12<36:25:00, 5.69s/it]
48%|████▊ | 20910/43939 [33:44:18<36:30:15, 5.71s/it]
48%|████▊ | 20911/43939 [33:44:24<36:33:06, 5.71s/it]
48%|████▊ | 20912/43939 [33:44:29<36:30:47, 5.71s/it]
48%|████▊ | 20913/43939 [33:44:35<36:28:12, 5.70s/it]
48%|████▊ | 20914/43939 [33:44:41<36:25:31, 5.70s/it]
48%|████▊ | 20915/43939 [33:44:47<36:56:38, 5.78s/it]
48%|████▊ | 20916/43939 [33:44:52<36:47:41, 5.75s/it]
48%|████▊ | 20917/43939 [33:44:58<36:37:44, 5.73s/it]
48%|████▊ | 20918/43939 [33:45:04<36:30:06, 5.71s/it]
48%|████▊ | 20919/43939 [33:45:09<36:31:18, 5.71s/it]
48%|████▊ | 20920/43939 [33:45:15<36:32:03, 5.71s/it]
48%|████▊ | 20921/43939 [33:45:21<36:30:54, 5.71s/it]
48%|████▊ | 20922/43939 [33:45:26<36:35:20, 5.72s/it]
48%|████▊ | 20923/43939 [33:45:32<36:35:55, 5.72s/it]
48%|████▊ | 20924/43939 [33:45:38<36:39:14, 5.73s/it]
48%|████▊ | 20925/43939 [33:45:44<36:37:37, 5.73s/it]
48%|████▊ | 20926/43939 [33:45:49<36:39:27, 5.73s/it]
48%|████▊ | 20927/43939 [33:45:55<36:41:15, 5.74s/it]
48%|████▊ | 20928/43939 [33:46:01<36:46:24, 5.75s/it]
48%|████▊ | 20929/43939 [33:46:07<36:37:35, 5.73s/it]
48%|████▊ | 20930/43939 [33:46:12<36:33:32, 5.72s/it]
48%|████▊ | 20931/43939 [33:46:18<36:33:54, 5.72s/it]
48%|████▊ | 20932/43939 [33:46:24<36:31:33, 5.72s/it]
48%|████▊ | 20933/43939 [33:46:29<36:28:35, 5.71s/it]
48%|████▊ | 20934/43939 [33:46:35<36:25:05, 5.70s/it]
48%|████▊ | 20935/43939 [33:46:41<36:26:27, 5.70s/it]
48%|████▊ | 20936/43939 [33:46:47<36:24:59, 5.70s/it]
48%|████▊ | 20937/43939 [33:46:52<36:51:56, 5.77s/it]
48%|████▊ | 20938/43939 [33:46:58<36:43:43, 5.75s/it]
48%|████▊ | 20939/43939 [33:47:04<36:40:04, 5.74s/it]
48%|████▊ | 20940/43939 [33:47:10<36:36:23, 5.73s/it]
48%|████▊ | 20941/43939 [33:47:15<36:32:13, 5.72s/it]
48%|████▊ | 20942/43939 [33:47:21<36:42:33, 5.75s/it]
48%|████▊ | 20943/43939 [33:47:27<36:32:59, 5.72s/it]
48%|████▊ | 20944/43939 [33:47:32<36:30:20, 5.72s/it]
48%|████▊ | 20945/43939 [33:47:38<36:30:46, 5.72s/it]
48%|████▊ | 20946/43939 [33:47:44<36:28:55, 5.71s/it]
48%|████▊ | 20947/43939 [33:47:50<36:36:01, 5.73s/it]
48%|████▊ | 20948/43939 [33:47:55<36:29:06, 5.71s/it]
48%|████▊ | 20949/43939 [33:48:01<36:24:04, 5.70s/it]
48%|████▊ | 20950/43939 [33:48:07<36:27:48, 5.71s/it]
{'loss': 0.9141, 'grad_norm': 1.462532639503479, 'learning_rate': 3.1319759684509584e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.24, 'epoch': 1.91}
48%|████▊ | 20950/43939 [33:48:07<36:27:48, 5.71s/it]
48%|████▊ | 20951/43939 [33:48:12<36:20:45, 5.69s/it]
48%|████▊ | 20952/43939 [33:48:18<36:25:50, 5.71s/it]
48%|████▊ | 20953/43939 [33:48:24<36:29:36, 5.72s/it]
48%|████▊ | 20954/43939 [33:48:30<36:30:30, 5.72s/it]
48%|████▊ | 20955/43939 [33:48:35<36:27:26, 5.71s/it]
48%|████▊ | 20956/43939 [33:48:41<36:27:26, 5.71s/it]
48%|████▊ | 20957/43939 [33:48:47<36:50:14, 5.77s/it]
48%|████▊ | 20958/43939 [33:48:53<36:41:09, 5.75s/it]
48%|████▊ | 20959/43939 [33:48:58<36:38:38, 5.74s/it]
48%|████▊ | 20960/43939 [33:49:04<36:28:38, 5.71s/it]
48%|████▊ | 20961/43939 [33:49:10<36:28:09, 5.71s/it]
48%|████▊ | 20962/43939 [33:49:15<36:26:12, 5.71s/it]
48%|████▊ | 20963/43939 [33:49:21<36:22:41, 5.70s/it]
48%|████▊ | 20964/43939 [33:49:27<36:23:13, 5.70s/it]
48%|████▊ | 20965/43939 [33:49:32<36:28:03, 5.71s/it]
48%|████▊ | 20966/43939 [33:49:38<36:26:12, 5.71s/it]
48%|████▊ | 20967/43939 [33:49:44<36:31:26, 5.72s/it]
48%|████▊ | 20968/43939 [33:49:50<36:27:14, 5.71s/it]
48%|████▊ | 20969/43939 [33:49:55<36:23:57, 5.70s/it]
48%|████▊ | 20970/43939 [33:50:01<36:23:33, 5.70s/it]
48%|████▊ | 20971/43939 [33:50:07<36:23:42, 5.70s/it]
48%|████▊ | 20972/43939 [33:50:12<36:19:53, 5.69s/it]
48%|████▊ | 20973/43939 [33:50:18<36:16:19, 5.69s/it]
48%|████▊ | 20974/43939 [33:50:24<36:22:05, 5.70s/it]
48%|████▊ | 20975/43939 [33:50:30<36:21:56, 5.70s/it]
48%|████▊ | 20976/43939 [33:50:35<36:23:45, 5.71s/it]
48%|████▊ | 20977/43939 [33:50:41<36:27:04, 5.71s/it]
48%|████▊ | 20978/43939 [33:50:47<36:52:18, 5.78s/it]
48%|████▊ | 20979/43939 [33:50:53<36:39:50, 5.75s/it]
48%|████▊ | 20980/43939 [33:50:58<36:32:18, 5.73s/it]
48%|████▊ | 20981/43939 [33:51:04<36:28:47, 5.72s/it]
48%|████▊ | 20982/43939 [33:51:10<36:26:32, 5.71s/it]
48%|████▊ | 20983/43939 [33:51:15<36:21:15, 5.70s/it]
48%|████▊ | 20984/43939 [33:51:21<36:20:53, 5.70s/it]
48%|████▊ | 20985/43939 [33:51:27<36:21:07, 5.70s/it]
48%|████▊ | 20986/43939 [33:51:32<36:22:03, 5.70s/it]
48%|████▊ | 20987/43939 [33:51:38<36:28:08, 5.72s/it]
48%|████▊ | 20988/43939 [33:51:44<36:22:52, 5.71s/it]
48%|████▊ | 20989/43939 [33:51:50<36:27:27, 5.72s/it]
48%|████▊ | 20990/43939 [33:51:55<36:25:56, 5.72s/it]
48%|████▊ | 20991/43939 [33:52:01<36:25:41, 5.71s/it]
48%|████▊ | 20992/43939 [33:52:07<36:18:19, 5.70s/it]
48%|████▊ | 20993/43939 [33:52:12<36:13:25, 5.68s/it]
48%|████▊ | 20994/43939 [33:52:18<36:15:50, 5.69s/it]
48%|████▊ | 20995/43939 [33:52:24<36:20:38, 5.70s/it]
48%|████▊ | 20996/43939 [33:52:29<36:23:09, 5.71s/it]
48%|████▊ | 20997/43939 [33:52:35<36:20:31, 5.70s/it]
48%|████▊ | 20998/43939 [33:52:41<36:19:17, 5.70s/it]
48%|████▊ | 20999/43939 [33:52:47<36:44:28, 5.77s/it]
48%|████▊ | 21000/43939 [33:52:52<36:36:36, 5.75s/it]
{'loss': 0.913, 'grad_norm': 1.3084138631820679, 'learning_rate': 3.1223633453998466e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.48, 'epoch': 1.91}
48%|████▊ | 21000/43939 [33:52:53<36:36:36, 5.75s/it]
48%|████▊ | 21001/43939 [33:52:58<36:31:55, 5.73s/it]
48%|████▊ | 21002/43939 [33:53:04<36:23:28, 5.71s/it]
48%|████▊ | 21003/43939 [33:53:10<36:27:03, 5.72s/it]
48%|████▊ | 21004/43939 [33:53:15<36:23:32, 5.71s/it]
48%|████▊ | 21005/43939 [33:53:21<36:21:06, 5.71s/it]
48%|████▊ | 21006/43939 [33:53:27<36:23:29, 5.71s/it]
48%|████▊ | 21007/43939 [33:53:32<36:19:07, 5.70s/it]
48%|████▊ | 21008/43939 [33:53:38<36:16:41, 5.70s/it]
48%|████▊ | 21009/43939 [33:53:44<36:11:46, 5.68s/it]
48%|████▊ | 21010/43939 [33:53:49<36:08:29, 5.67s/it]
48%|████▊ | 21011/43939 [33:53:55<36:09:32, 5.68s/it]
48%|████▊ | 21012/43939 [33:54:01<36:12:26, 5.69s/it]
48%|████▊ | 21013/43939 [33:54:06<36:15:20, 5.69s/it]
48%|████▊ | 21014/43939 [33:54:12<36:17:31, 5.70s/it]
48%|████▊ | 21015/43939 [33:54:18<36:18:43, 5.70s/it]
48%|████▊ | 21016/43939 [33:54:24<36:21:49, 5.71s/it]
48%|████▊ | 21017/43939 [33:54:29<36:24:13, 5.72s/it]
48%|████▊ | 21018/43939 [33:54:35<36:19:46, 5.71s/it]
48%|████▊ | 21019/43939 [33:54:41<36:21:30, 5.71s/it]
48%|████▊ | 21020/43939 [33:54:47<36:49:19, 5.78s/it]
48%|████▊ | 21021/43939 [33:54:52<36:36:42, 5.75s/it]
48%|████▊ | 21022/43939 [33:54:58<36:36:53, 5.75s/it]
48%|████▊ | 21023/43939 [33:55:04<36:28:52, 5.73s/it]
48%|████▊ | 21024/43939 [33:55:10<36:23:41, 5.72s/it]
48%|████▊ | 21025/43939 [33:55:15<36:27:07, 5.73s/it]
48%|████▊ | 21026/43939 [33:55:21<36:28:36, 5.73s/it]
48%|████▊ | 21027/43939 [33:55:27<36:21:13, 5.71s/it]
48%|████▊ | 21028/43939 [33:55:32<36:22:20, 5.72s/it]
48%|████▊ | 21029/43939 [33:55:38<36:17:27, 5.70s/it]
48%|████▊ | 21030/43939 [33:55:44<36:17:19, 5.70s/it]
48%|████▊ | 21031/43939 [33:55:49<36:12:26, 5.69s/it]
48%|████▊ | 21032/43939 [33:55:55<36:14:27, 5.70s/it]
48%|████▊ | 21033/43939 [33:56:01<36:12:37, 5.69s/it]
48%|████▊ | 21034/43939 [33:56:06<36:09:09, 5.68s/it]
48%|████▊ | 21035/43939 [33:56:12<36:10:55, 5.69s/it]
48%|████▊ | 21036/43939 [33:56:18<36:09:43, 5.68s/it]
48%|████▊ | 21037/43939 [33:56:24<36:12:11, 5.69s/it]
48%|████▊ | 21038/43939 [33:56:29<36:13:00, 5.69s/it]
48%|████▊ | 21039/43939 [33:56:35<36:11:19, 5.69s/it]
48%|████▊ | 21040/43939 [33:56:41<36:09:12, 5.68s/it]
48%|████▊ | 21041/43939 [33:56:47<36:41:22, 5.77s/it]
48%|████▊ | 21042/43939 [33:56:52<36:43:35, 5.77s/it]
48%|████▊ | 21043/43939 [33:56:58<36:37:50, 5.76s/it]
48%|████▊ | 21044/43939 [33:57:04<36:29:26, 5.74s/it]
48%|████▊ | 21045/43939 [33:57:09<36:22:53, 5.72s/it]
48%|████▊ | 21046/43939 [33:57:15<36:16:55, 5.71s/it]
48%|████▊ | 21047/43939 [33:57:21<36:12:56, 5.70s/it]
48%|████▊ | 21048/43939 [33:57:27<36:19:13, 5.71s/it]
48%|████▊ | 21049/43939 [33:57:32<36:18:22, 5.71s/it]
48%|████▊ | 21050/43939 [33:57:38<36:16:11, 5.70s/it]
{'loss': 0.9466, 'grad_norm': 1.0528351068496704, 'learning_rate': 3.112740903104428e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.61, 'epoch': 1.92}
48%|████▊ | 21050/43939 [33:57:38<36:16:11, 5.70s/it]
48%|████▊ | 21051/43939 [33:57:44<36:23:10, 5.72s/it]
48%|████▊ | 21052/43939 [33:57:49<36:23:12, 5.72s/it]
48%|████▊ | 21053/43939 [33:57:55<36:25:27, 5.73s/it]
48%|████▊ | 21054/43939 [33:58:01<36:24:10, 5.73s/it]
48%|████▊ | 21055/43939 [33:58:07<36:22:55, 5.72s/it]
48%|████▊ | 21056/43939 [33:58:12<36:17:59, 5.71s/it]
48%|████▊ | 21057/43939 [33:58:18<36:14:14, 5.70s/it]
48%|████▊ | 21058/43939 [33:58:24<36:13:56, 5.70s/it]
48%|████▊ | 21059/43939 [33:58:29<36:13:49, 5.70s/it]
48%|████▊ | 21060/43939 [33:58:35<36:14:25, 5.70s/it]
48%|████▊ | 21061/43939 [33:58:41<36:10:42, 5.69s/it]
48%|████▊ | 21062/43939 [33:58:46<36:12:51, 5.70s/it]
48%|████▊ | 21063/43939 [33:58:52<36:38:19, 5.77s/it]
48%|████▊ | 21064/43939 [33:58:58<36:32:29, 5.75s/it]
48%|████▊ | 21065/43939 [33:59:04<36:25:05, 5.73s/it]
48%|████▊ | 21066/43939 [33:59:09<36:19:16, 5.72s/it]
48%|████▊ | 21067/43939 [33:59:15<36:12:36, 5.70s/it]
48%|████▊ | 21068/43939 [33:59:21<36:11:08, 5.70s/it]
48%|████▊ | 21069/43939 [33:59:26<36:07:26, 5.69s/it]
48%|████▊ | 21070/43939 [33:59:32<36:08:35, 5.69s/it]
48%|████▊ | 21071/43939 [33:59:38<36:07:18, 5.69s/it]
48%|████▊ | 21072/43939 [33:59:44<36:11:20, 5.70s/it]
48%|████▊ | 21073/43939 [33:59:49<36:13:40, 5.70s/it]
48%|████▊ | 21074/43939 [33:59:55<36:16:05, 5.71s/it]
48%|████▊ | 21075/43939 [34:00:01<36:20:27, 5.72s/it]
48%|████▊ | 21076/43939 [34:00:07<36:20:08, 5.72s/it]
48%|████▊ | 21077/43939 [34:00:12<36:16:42, 5.71s/it]
48%|████▊ | 21078/43939 [34:00:18<36:15:19, 5.71s/it]
48%|████▊ | 21079/43939 [34:00:24<36:08:01, 5.69s/it]
48%|████▊ | 21080/43939 [34:00:29<36:05:40, 5.68s/it]
48%|████▊ | 21081/43939 [34:00:35<36:06:14, 5.69s/it]
48%|████▊ | 21082/43939 [34:00:41<36:12:46, 5.70s/it]
48%|████▊ | 21083/43939 [34:00:46<36:14:53, 5.71s/it]
48%|████▊ | 21084/43939 [34:00:52<36:42:37, 5.78s/it]
48%|████▊ | 21085/43939 [34:00:58<36:29:31, 5.75s/it]
48%|████▊ | 21086/43939 [34:01:04<36:23:50, 5.73s/it]
48%|████▊ | 21087/43939 [34:01:09<36:20:17, 5.72s/it]
48%|████▊ | 21088/43939 [34:01:15<36:14:03, 5.71s/it]
48%|████▊ | 21089/43939 [34:01:21<36:23:23, 5.73s/it]
48%|████▊ | 21090/43939 [34:01:27<36:20:45, 5.73s/it]
48%|████▊ | 21091/43939 [34:01:32<36:21:16, 5.73s/it]
48%|████▊ | 21092/43939 [34:01:38<36:17:55, 5.72s/it]
48%|████▊ | 21093/43939 [34:01:44<36:22:15, 5.73s/it]
48%|████▊ | 21094/43939 [34:01:49<36:18:44, 5.72s/it]
48%|████▊ | 21095/43939 [34:01:55<36:15:44, 5.71s/it]
48%|████▊ | 21096/43939 [34:02:01<36:13:54, 5.71s/it]
48%|████▊ | 21097/43939 [34:02:07<36:25:42, 5.74s/it]
48%|████▊ | 21098/43939 [34:02:12<36:19:54, 5.73s/it]
48%|████▊ | 21099/43939 [34:02:18<36:17:20, 5.72s/it]
48%|████▊ | 21100/43939 [34:02:24<36:22:01, 5.73s/it]
{'loss': 0.8976, 'grad_norm': 1.1360912322998047, 'learning_rate': 3.103108793381341e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.0, 'epoch': 1.92}
48%|████▊ | 21100/43939 [34:02:24<36:22:01, 5.73s/it]
48%|████▊ | 21101/43939 [34:02:30<36:16:42, 5.72s/it]
48%|████▊ | 21102/43939 [34:02:35<36:16:53, 5.72s/it]
48%|████▊ | 21103/43939 [34:02:41<36:14:43, 5.71s/it]
48%|████▊ | 21104/43939 [34:02:47<36:11:22, 5.71s/it]
48%|████▊ | 21105/43939 [34:02:53<36:36:04, 5.77s/it]
48%|████▊ | 21106/43939 [34:02:58<36:32:10, 5.76s/it]
48%|████▊ | 21107/43939 [34:03:04<36:25:20, 5.74s/it]
48%|████▊ | 21108/43939 [34:03:10<36:20:59, 5.73s/it]
48%|████▊ | 21109/43939 [34:03:15<36:17:18, 5.72s/it]
48%|████▊ | 21110/43939 [34:03:21<36:10:03, 5.70s/it]
48%|████▊ | 21111/43939 [34:03:27<36:13:06, 5.71s/it]
48%|████▊ | 21112/43939 [34:03:32<36:09:54, 5.70s/it]
48%|████▊ | 21113/43939 [34:03:38<36:12:45, 5.71s/it]
48%|████▊ | 21114/43939 [34:03:44<36:08:04, 5.70s/it]
48%|████▊ | 21115/43939 [34:03:50<36:11:42, 5.71s/it]
48%|████▊ | 21116/43939 [34:03:55<36:14:32, 5.72s/it]
48%|████▊ | 21117/43939 [34:04:01<36:12:34, 5.71s/it]
48%|████▊ | 21118/43939 [34:04:07<36:09:51, 5.70s/it]
48%|████▊ | 21119/43939 [34:04:12<36:06:51, 5.70s/it]
48%|████▊ | 21120/43939 [34:04:18<36:04:53, 5.69s/it]
48%|████▊ | 21121/43939 [34:04:24<36:04:57, 5.69s/it]
48%|████▊ | 21122/43939 [34:04:29<36:08:00, 5.70s/it]
48%|████▊ | 21123/43939 [34:04:35<36:08:51, 5.70s/it]
48%|████▊ | 21124/43939 [34:04:41<36:04:54, 5.69s/it]
48%|████▊ | 21125/43939 [34:04:47<36:04:57, 5.69s/it]
48%|████▊ | 21126/43939 [34:04:52<36:04:12, 5.69s/it]
48%|████▊ | 21127/43939 [34:04:58<36:32:27, 5.77s/it]
48%|████▊ | 21128/43939 [34:05:04<36:31:56, 5.77s/it]
48%|████▊ | 21129/43939 [34:05:10<36:24:48, 5.75s/it]
48%|████▊ | 21130/43939 [34:05:15<36:23:35, 5.74s/it]
48%|████▊ | 21131/43939 [34:05:21<36:16:14, 5.72s/it]
48%|████▊ | 21132/43939 [34:05:27<36:14:25, 5.72s/it]
48%|████▊ | 21133/43939 [34:05:33<36:14:33, 5.72s/it]
48%|████▊ | 21134/43939 [34:05:38<36:16:37, 5.73s/it]
48%|████▊ | 21135/43939 [34:05:44<36:17:46, 5.73s/it]
48%|████▊ | 21136/43939 [34:05:50<36:12:26, 5.72s/it]
48%|████▊ | 21137/43939 [34:05:55<36:09:57, 5.71s/it]
48%|████▊ | 21138/43939 [34:06:01<36:12:44, 5.72s/it]
48%|████▊ | 21139/43939 [34:06:07<36:09:38, 5.71s/it]
48%|████▊ | 21140/43939 [34:06:13<36:09:39, 5.71s/it]
48%|████▊ | 21141/43939 [34:06:18<36:10:49, 5.71s/it]
48%|████▊ | 21142/43939 [34:06:24<36:09:27, 5.71s/it]
48%|████▊ | 21143/43939 [34:06:30<36:07:48, 5.71s/it]
48%|████▊ | 21144/43939 [34:06:35<36:07:20, 5.70s/it]
48%|████▊ | 21145/43939 [34:06:41<36:10:29, 5.71s/it]
48%|████▊ | 21146/43939 [34:06:47<36:05:08, 5.70s/it]
48%|████▊ | 21147/43939 [34:06:52<36:08:05, 5.71s/it]
48%|████▊ | 21148/43939 [34:06:58<36:35:39, 5.78s/it]
48%|████▊ | 21149/43939 [34:07:04<36:27:07, 5.76s/it]
48%|████▊ | 21150/43939 [34:07:10<36:24:05, 5.75s/it]
{'loss': 0.924, 'grad_norm': 1.2134554386138916, 'learning_rate': 3.093467168199754e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.54, 'epoch': 1.93}
48%|████▊ | 21150/43939 [34:07:10<36:24:05, 5.75s/it]
48%|████▊ | 21151/43939 [34:07:16<36:23:55, 5.75s/it]
48%|████▊ | 21152/43939 [34:07:21<36:23:27, 5.75s/it]
48%|████▊ | 21153/43939 [34:07:27<36:18:02, 5.74s/it]
48%|████▊ | 21154/43939 [34:07:33<36:21:12, 5.74s/it]
48%|████▊ | 21155/43939 [34:07:39<36:16:16, 5.73s/it]
48%|████▊ | 21156/43939 [34:07:44<36:12:04, 5.72s/it]
48%|████▊ | 21157/43939 [34:07:50<36:09:03, 5.71s/it]
48%|████▊ | 21158/43939 [34:07:56<36:08:17, 5.71s/it]
48%|████▊ | 21159/43939 [34:08:01<36:05:11, 5.70s/it]
48%|████▊ | 21160/43939 [34:08:07<36:04:16, 5.70s/it]
48%|████▊ | 21161/43939 [34:08:13<36:10:31, 5.72s/it]
48%|████▊ | 21162/43939 [34:08:18<36:09:29, 5.71s/it]
48%|████▊ | 21163/43939 [34:08:24<36:06:24, 5.71s/it]
48%|████▊ | 21164/43939 [34:08:30<36:06:31, 5.71s/it]
48%|████▊ | 21165/43939 [34:08:36<36:15:38, 5.73s/it]
48%|████▊ | 21166/43939 [34:08:41<36:12:47, 5.72s/it]
48%|████▊ | 21167/43939 [34:08:47<36:04:51, 5.70s/it]
48%|████▊ | 21168/43939 [34:08:53<36:04:00, 5.70s/it]
48%|████▊ | 21169/43939 [34:08:59<36:27:26, 5.76s/it]
48%|████▊ | 21170/43939 [34:09:04<36:22:21, 5.75s/it]
48%|████▊ | 21171/43939 [34:09:10<36:19:43, 5.74s/it]
48%|████▊ | 21172/43939 [34:09:16<36:18:30, 5.74s/it]
48%|████▊ | 21173/43939 [34:09:22<36:16:09, 5.74s/it]
48%|████▊ | 21174/43939 [34:09:27<36:16:20, 5.74s/it]
48%|████▊ | 21175/43939 [34:09:33<36:08:22, 5.72s/it]
48%|████▊ | 21176/43939 [34:09:39<36:12:06, 5.73s/it]
48%|████▊ | 21177/43939 [34:09:44<36:06:59, 5.71s/it]
48%|████▊ | 21178/43939 [34:09:50<35:59:17, 5.69s/it]
48%|████▊ | 21179/43939 [34:09:56<36:04:00, 5.70s/it]
48%|████▊ | 21180/43939 [34:10:01<36:00:44, 5.70s/it]
48%|████▊ | 21181/43939 [34:10:07<35:58:10, 5.69s/it]
48%|████▊ | 21182/43939 [34:10:13<35:55:58, 5.68s/it]
48%|████▊ | 21183/43939 [34:10:18<35:57:45, 5.69s/it]
48%|████▊ | 21184/43939 [34:10:24<36:00:38, 5.70s/it]
48%|████▊ | 21185/43939 [34:10:30<36:12:08, 5.73s/it]
48%|████▊ | 21186/43939 [34:10:36<36:07:34, 5.72s/it]
48%|████▊ | 21187/43939 [34:10:41<36:03:07, 5.70s/it]
48%|████▊ | 21188/43939 [34:10:47<36:03:21, 5.71s/it]
48%|████▊ | 21189/43939 [34:10:53<36:01:02, 5.70s/it]
48%|████▊ | 21190/43939 [34:10:59<36:29:59, 5.78s/it]
48%|████▊ | 21191/43939 [34:11:04<36:26:32, 5.77s/it]
48%|████▊ | 21192/43939 [34:11:10<36:13:26, 5.73s/it]
48%|████▊ | 21193/43939 [34:11:16<36:04:30, 5.71s/it]
48%|████▊ | 21194/43939 [34:11:21<36:04:21, 5.71s/it]
48%|████▊ | 21195/43939 [34:11:27<36:09:26, 5.72s/it]
48%|████▊ | 21196/43939 [34:11:33<36:07:59, 5.72s/it]
48%|████▊ | 21197/43939 [34:11:39<36:08:51, 5.72s/it]
48%|████▊ | 21198/43939 [34:11:44<36:07:46, 5.72s/it]
48%|████▊ | 21199/43939 [34:11:50<36:05:03, 5.71s/it]
48%|████▊ | 21200/43939 [34:11:56<36:02:46, 5.71s/it]
{'loss': 0.9113, 'grad_norm': 1.0968073606491089, 'learning_rate': 3.083816179678958e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.57, 'epoch': 1.93}
48%|████▊ | 21200/43939 [34:11:56<36:02:46, 5.71s/it]
48%|████▊ | 21201/43939 [34:12:01<36:07:27, 5.72s/it]
48%|████▊ | 21202/43939 [34:12:07<36:08:54, 5.72s/it]
48%|████▊ | 21203/43939 [34:12:13<36:07:17, 5.72s/it]
48%|████▊ | 21204/43939 [34:12:19<35:59:17, 5.70s/it]
48%|████▊ | 21205/43939 [34:12:24<35:59:49, 5.70s/it]
48%|████▊ | 21206/43939 [34:12:30<36:02:16, 5.71s/it]
48%|████▊ | 21207/43939 [34:12:36<35:56:54, 5.69s/it]
48%|████▊ | 21208/43939 [34:12:41<35:58:09, 5.70s/it]
48%|████▊ | 21209/43939 [34:12:47<35:57:56, 5.70s/it]
48%|████▊ | 21210/43939 [34:12:53<35:57:51, 5.70s/it]
48%|████▊ | 21211/43939 [34:12:58<36:00:19, 5.70s/it]
48%|████▊ | 21212/43939 [34:13:04<36:26:49, 5.77s/it]
48%|████▊ | 21213/43939 [34:13:10<36:17:31, 5.75s/it]
48%|████▊ | 21214/43939 [34:13:16<36:12:24, 5.74s/it]
48%|████▊ | 21215/43939 [34:13:22<36:09:48, 5.73s/it]
48%|████▊ | 21216/43939 [34:13:27<36:06:45, 5.72s/it]
48%|████▊ | 21217/43939 [34:13:33<36:02:39, 5.71s/it]
48%|████▊ | 21218/43939 [34:13:39<35:59:27, 5.70s/it]
48%|████▊ | 21219/43939 [34:13:44<35:58:24, 5.70s/it]
48%|████▊ | 21220/43939 [34:13:50<36:02:57, 5.71s/it]
48%|████▊ | 21221/43939 [34:13:56<36:09:11, 5.73s/it]
48%|████▊ | 21222/43939 [34:14:01<36:03:49, 5.72s/it]
48%|████▊ | 21223/43939 [34:14:07<35:59:16, 5.70s/it]
48%|████▊ | 21224/43939 [34:14:13<36:03:34, 5.71s/it]
48%|████▊ | 21225/43939 [34:14:19<36:06:50, 5.72s/it]
48%|████▊ | 21226/43939 [34:14:24<36:03:52, 5.72s/it]
48%|████▊ | 21227/43939 [34:14:30<36:02:13, 5.71s/it]
48%|████▊ | 21228/43939 [34:14:36<36:00:40, 5.71s/it]
48%|████▊ | 21229/43939 [34:14:42<36:05:33, 5.72s/it]
48%|████▊ | 21230/43939 [34:14:47<36:01:26, 5.71s/it]
48%|████▊ | 21231/43939 [34:14:53<36:03:34, 5.72s/it]
48%|████▊ | 21232/43939 [34:14:59<36:00:19, 5.71s/it]
48%|████▊ | 21233/43939 [34:15:04<35:57:46, 5.70s/it]
48%|████▊ | 21234/43939 [34:15:10<36:23:34, 5.77s/it]
48%|████▊ | 21235/43939 [34:15:16<36:13:26, 5.74s/it]
48%|████▊ | 21236/43939 [34:15:22<36:07:57, 5.73s/it]
48%|████▊ | 21237/43939 [34:15:27<36:02:26, 5.72s/it]
48%|████▊ | 21238/43939 [34:15:33<35:58:26, 5.70s/it]
48%|████▊ | 21239/43939 [34:15:39<35:59:07, 5.71s/it]
48%|████▊ | 21240/43939 [34:15:44<35:57:14, 5.70s/it]
48%|████▊ | 21241/43939 [34:15:50<36:00:15, 5.71s/it]
48%|████▊ | 21242/43939 [34:15:56<36:04:27, 5.72s/it]
48%|████▊ | 21243/43939 [34:16:02<36:01:15, 5.71s/it]
48%|████▊ | 21244/43939 [34:16:07<36:00:43, 5.71s/it]
48%|████▊ | 21245/43939 [34:16:13<35:57:56, 5.71s/it]
48%|████▊ | 21246/43939 [34:16:19<36:00:29, 5.71s/it]
48%|████▊ | 21247/43939 [34:16:24<36:01:50, 5.72s/it]
48%|████▊ | 21248/43939 [34:16:30<36:02:13, 5.72s/it]
48%|████▊ | 21249/43939 [34:16:36<36:01:39, 5.72s/it]
48%|████▊ | 21250/43939 [34:16:42<36:00:08, 5.71s/it]
{'loss': 0.9019, 'grad_norm': 1.332603096961975, 'learning_rate': 3.074155980085979e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.06, 'epoch': 1.93}
48%|████▊ | 21250/43939 [34:16:42<36:00:08, 5.71s/it]
48%|████▊ | 21251/43939 [34:16:47<35:58:16, 5.71s/it]
48%|████▊ | 21252/43939 [34:16:53<35:52:35, 5.69s/it]
48%|████▊ | 21253/43939 [34:16:59<35:55:24, 5.70s/it]
48%|████▊ | 21254/43939 [34:17:05<36:25:58, 5.78s/it]
48%|████▊ | 21255/43939 [34:17:10<36:15:31, 5.75s/it]
48%|████▊ | 21256/43939 [34:17:16<36:06:15, 5.73s/it]
48%|████▊ | 21257/43939 [34:17:22<35:59:12, 5.71s/it]
48%|████▊ | 21258/43939 [34:17:27<36:02:30, 5.72s/it]
48%|████▊ | 21259/43939 [34:17:33<36:00:11, 5.71s/it]
48%|████▊ | 21260/43939 [34:17:39<35:55:49, 5.70s/it]
48%|████▊ | 21261/43939 [34:17:44<35:56:16, 5.70s/it]
48%|████▊ | 21262/43939 [34:17:50<35:55:08, 5.70s/it]
48%|████▊ | 21263/43939 [34:17:56<35:55:25, 5.70s/it]
48%|████▊ | 21264/43939 [34:18:02<35:58:27, 5.71s/it]
48%|████▊ | 21265/43939 [34:18:07<35:56:44, 5.71s/it]
48%|████▊ | 21266/43939 [34:18:13<35:57:18, 5.71s/it]
48%|████▊ | 21267/43939 [34:18:19<35:53:37, 5.70s/it]
48%|████▊ | 21268/43939 [34:18:24<35:58:06, 5.71s/it]
48%|████▊ | 21269/43939 [34:18:30<35:56:15, 5.71s/it]
48%|████▊ | 21270/43939 [34:18:36<35:54:21, 5.70s/it]
48%|████▊ | 21271/43939 [34:18:41<35:52:07, 5.70s/it]
48%|████▊ | 21272/43939 [34:18:47<35:52:17, 5.70s/it]
48%|████▊ | 21273/43939 [34:18:53<35:58:51, 5.71s/it]
48%|████▊ | 21274/43939 [34:18:59<35:59:04, 5.72s/it]
48%|████▊ | 21275/43939 [34:19:05<36:26:10, 5.79s/it]
48%|████▊ | 21276/43939 [34:19:10<36:14:18, 5.76s/it]
48%|████▊ | 21277/43939 [34:19:16<36:07:59, 5.74s/it]
48%|████▊ | 21278/43939 [34:19:22<36:02:12, 5.72s/it]
48%|████▊ | 21279/43939 [34:19:27<35:58:33, 5.72s/it]
48%|████▊ | 21280/43939 [34:19:33<35:54:55, 5.71s/it]
48%|████▊ | 21281/43939 [34:19:39<35:54:40, 5.71s/it]
48%|████▊ | 21282/43939 [34:19:44<35:53:05, 5.70s/it]
48%|████▊ | 21283/43939 [34:19:50<35:53:15, 5.70s/it]
48%|████▊ | 21284/43939 [34:19:56<35:52:31, 5.70s/it]
48%|████▊ | 21285/43939 [34:20:01<35:45:56, 5.68s/it]
48%|████▊ | 21286/43939 [34:20:07<35:50:03, 5.69s/it]
48%|████▊ | 21287/43939 [34:20:13<35:50:52, 5.70s/it]
48%|████▊ | 21288/43939 [34:20:19<35:51:01, 5.70s/it]
48%|████▊ | 21289/43939 [34:20:24<35:54:12, 5.71s/it]
48%|████▊ | 21290/43939 [34:20:30<35:56:06, 5.71s/it]
48%|████▊ | 21291/43939 [34:20:36<35:55:12, 5.71s/it]
48%|████▊ | 21292/43939 [34:20:41<35:51:00, 5.70s/it]
48%|████▊ | 21293/43939 [34:20:47<35:48:36, 5.69s/it]
48%|████▊ | 21294/43939 [34:20:53<35:52:42, 5.70s/it]
48%|████▊ | 21295/43939 [34:20:59<35:53:27, 5.71s/it]
48%|████▊ | 21296/43939 [34:21:04<35:57:57, 5.72s/it]
48%|████▊ | 21297/43939 [34:21:10<36:26:48, 5.79s/it]
48%|████▊ | 21298/43939 [34:21:16<36:17:30, 5.77s/it]
48%|████▊ | 21299/43939 [34:21:22<36:08:13, 5.75s/it]
48%|████▊ | 21300/43939 [34:21:27<36:02:49, 5.73s/it]
{'loss': 0.9203, 'grad_norm': 1.3806955814361572, 'learning_rate': 3.0644867218331665e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.01, 'epoch': 1.94}
48%|████▊ | 21300/43939 [34:21:27<36:02:49, 5.73s/it]
48%|████▊ | 21301/43939 [34:21:33<36:02:12, 5.73s/it]
48%|████▊ | 21302/43939 [34:21:39<35:55:57, 5.71s/it]
48%|████▊ | 21303/43939 [34:21:44<35:54:05, 5.71s/it]
48%|████▊ | 21304/43939 [34:21:50<35:54:02, 5.71s/it]
48%|████▊ | 21305/43939 [34:21:56<35:55:48, 5.71s/it]
48%|████▊ | 21306/43939 [34:22:02<35:50:47, 5.70s/it]
48%|████▊ | 21307/43939 [34:22:07<35:49:09, 5.70s/it]
48%|████▊ | 21308/43939 [34:22:13<35:53:50, 5.71s/it]
48%|████▊ | 21309/43939 [34:22:19<35:51:40, 5.70s/it]
48%|████▊ | 21310/43939 [34:22:24<35:45:26, 5.69s/it]
49%|████▊ | 21311/43939 [34:22:30<35:44:37, 5.69s/it]
49%|████▊ | 21312/43939 [34:22:36<35:44:23, 5.69s/it]
49%|████▊ | 21313/43939 [34:22:41<35:40:44, 5.68s/it]
49%|████▊ | 21314/43939 [34:22:47<35:45:13, 5.69s/it]
49%|████▊ | 21315/43939 [34:22:53<35:48:34, 5.70s/it]
49%|████▊ | 21316/43939 [34:22:59<35:46:02, 5.69s/it]
49%|████▊ | 21317/43939 [34:23:04<36:08:04, 5.75s/it]
49%|████▊ | 21318/43939 [34:23:10<36:11:20, 5.76s/it]
49%|████▊ | 21319/43939 [34:23:16<36:05:28, 5.74s/it]
49%|████▊ | 21320/43939 [34:23:22<35:59:53, 5.73s/it]
49%|████▊ | 21321/43939 [34:23:27<35:55:55, 5.72s/it]
49%|████▊ | 21322/43939 [34:23:33<35:57:45, 5.72s/it]
49%|████▊ | 21323/43939 [34:23:39<35:51:19, 5.71s/it]
49%|████▊ | 21324/43939 [34:23:44<35:51:32, 5.71s/it]
49%|████▊ | 21325/43939 [34:23:50<35:53:56, 5.71s/it]
49%|████▊ | 21326/43939 [34:23:56<35:52:38, 5.71s/it]
49%|████▊ | 21327/43939 [34:24:02<35:52:58, 5.71s/it]
49%|████▊ | 21328/43939 [34:24:07<35:51:45, 5.71s/it]
49%|████▊ | 21329/43939 [34:24:13<35:53:51, 5.72s/it]
49%|████▊ | 21330/43939 [34:24:19<35:53:09, 5.71s/it]
49%|████▊ | 21331/43939 [34:24:24<35:51:14, 5.71s/it]
49%|████▊ | 21332/43939 [34:24:30<35:58:50, 5.73s/it]
49%|████▊ | 21333/43939 [34:24:36<35:55:09, 5.72s/it]
49%|████▊ | 21334/43939 [34:24:42<35:54:36, 5.72s/it]
49%|████▊ | 21335/43939 [34:24:47<35:55:28, 5.72s/it]
49%|████▊ | 21336/43939 [34:24:53<35:53:24, 5.72s/it]
49%|████▊ | 21337/43939 [34:24:59<35:53:32, 5.72s/it]
49%|████▊ | 21338/43939 [34:25:05<36:16:50, 5.78s/it]
49%|████▊ | 21339/43939 [34:25:10<36:07:03, 5.75s/it]
49%|████▊ | 21340/43939 [34:25:16<35:59:01, 5.73s/it]
49%|████▊ | 21341/43939 [34:25:22<35:53:01, 5.72s/it]
49%|████▊ | 21342/43939 [34:25:27<35:50:14, 5.71s/it]
49%|████▊ | 21343/43939 [34:25:33<36:00:31, 5.74s/it]
49%|████▊ | 21344/43939 [34:25:39<35:56:20, 5.73s/it]
49%|████▊ | 21345/43939 [34:25:45<35:52:15, 5.72s/it]
49%|████▊ | 21346/43939 [34:25:50<35:46:05, 5.70s/it]
49%|████▊ | 21347/43939 [34:25:56<35:47:56, 5.70s/it]
49%|████▊ | 21348/43939 [34:26:02<35:48:06, 5.71s/it]
49%|████▊ | 21349/43939 [34:26:07<35:50:53, 5.71s/it]
49%|████▊ | 21350/43939 [34:26:13<35:59:56, 5.74s/it]
{'loss': 0.9105, 'grad_norm': 0.9688365459442139, 'learning_rate': 3.0548085574757905e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.31, 'epoch': 1.94}
49%|████▊ | 21350/43939 [34:26:13<35:59:56, 5.74s/it]
49%|████▊ | 21351/43939 [34:26:19<35:58:07, 5.73s/it]
49%|████▊ | 21352/43939 [34:26:25<35:54:07, 5.72s/it]
49%|████▊ | 21353/43939 [34:26:30<35:55:15, 5.73s/it]
49%|████▊ | 21354/43939 [34:26:36<35:53:04, 5.72s/it]
49%|████▊ | 21355/43939 [34:26:42<35:49:08, 5.71s/it]
49%|████▊ | 21356/43939 [34:26:47<35:48:48, 5.71s/it]
49%|████▊ | 21357/43939 [34:26:53<35:44:12, 5.70s/it]
49%|████▊ | 21358/43939 [34:26:59<35:41:20, 5.69s/it]
49%|████▊ | 21359/43939 [34:27:04<35:43:23, 5.70s/it]
49%|████▊ | 21360/43939 [34:27:10<36:03:39, 5.75s/it]
49%|████▊ | 21361/43939 [34:27:16<35:58:49, 5.74s/it]
49%|████▊ | 21362/43939 [34:27:22<35:57:06, 5.73s/it]
49%|████▊ | 21363/43939 [34:27:28<35:56:23, 5.73s/it]
49%|████▊ | 21364/43939 [34:27:33<35:52:29, 5.72s/it]
49%|████▊ | 21365/43939 [34:27:39<35:52:13, 5.72s/it]
49%|████▊ | 21366/43939 [34:27:45<35:50:41, 5.72s/it]
49%|████▊ | 21367/43939 [34:27:50<35:50:07, 5.72s/it]
49%|████▊ | 21368/43939 [34:27:56<35:49:02, 5.71s/it]
49%|████▊ | 21369/43939 [34:28:02<35:48:37, 5.71s/it]
49%|████▊ | 21370/43939 [34:28:08<35:49:21, 5.71s/it]
49%|████▊ | 21371/43939 [34:28:13<35:47:50, 5.71s/it]
49%|████▊ | 21372/43939 [34:28:19<35:47:57, 5.71s/it]
49%|████▊ | 21373/43939 [34:28:25<35:50:27, 5.72s/it]
49%|████▊ | 21374/43939 [34:28:30<35:51:36, 5.72s/it]
49%|████▊ | 21375/43939 [34:28:36<35:45:00, 5.70s/it]
49%|████▊ | 21376/43939 [34:28:42<35:46:00, 5.71s/it]
49%|████▊ | 21377/43939 [34:28:47<35:41:09, 5.69s/it]
49%|████▊ | 21378/43939 [34:28:53<35:42:02, 5.70s/it]
49%|████▊ | 21379/43939 [34:28:59<35:43:48, 5.70s/it]
49%|████▊ | 21380/43939 [34:29:05<35:42:08, 5.70s/it]
49%|████▊ | 21381/43939 [34:29:10<36:05:57, 5.76s/it]
49%|████▊ | 21382/43939 [34:29:16<35:58:06, 5.74s/it]
49%|████▊ | 21383/43939 [34:29:22<35:54:17, 5.73s/it]
49%|████▊ | 21384/43939 [34:29:28<35:56:10, 5.74s/it]
49%|████▊ | 21385/43939 [34:29:33<35:51:50, 5.72s/it]
49%|████▊ | 21386/43939 [34:29:39<35:48:50, 5.72s/it]
49%|████▊ | 21387/43939 [34:29:45<35:41:30, 5.70s/it]
49%|████▊ | 21388/43939 [34:29:50<35:40:19, 5.69s/it]
49%|████▊ | 21389/43939 [34:29:56<35:44:01, 5.70s/it]
49%|████▊ | 21390/43939 [34:30:02<35:44:03, 5.71s/it]
49%|████▊ | 21391/43939 [34:30:07<35:44:08, 5.71s/it]
49%|████▊ | 21392/43939 [34:30:13<35:45:11, 5.71s/it]
49%|████▊ | 21393/43939 [34:30:19<35:41:52, 5.70s/it]
49%|████▊ | 21394/43939 [34:30:25<35:41:00, 5.70s/it]
49%|████▊ | 21395/43939 [34:30:30<35:42:36, 5.70s/it]
49%|████▊ | 21396/43939 [34:30:36<35:45:43, 5.71s/it]
49%|████▊ | 21397/43939 [34:30:42<35:41:21, 5.70s/it]
49%|████▊ | 21398/43939 [34:30:47<35:43:00, 5.70s/it]
49%|████▊ | 21399/43939 [34:30:53<35:40:51, 5.70s/it]
49%|████▊ | 21400/43939 [34:30:59<35:40:55, 5.70s/it]
{'loss': 0.918, 'grad_norm': 1.2917383909225464, 'learning_rate': 3.0451216397096392e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.58, 'epoch': 1.95}
49%|████▊ | 21400/43939 [34:30:59<35:40:55, 5.70s/it]
49%|████▊ | 21401/43939 [34:31:04<35:43:17, 5.71s/it]
49%|████▊ | 21402/43939 [34:31:10<36:11:26, 5.78s/it]
49%|████▊ | 21403/43939 [34:31:16<36:09:49, 5.78s/it]
49%|████▊ | 21404/43939 [34:31:22<36:01:26, 5.75s/it]
49%|████▊ | 21405/43939 [34:31:28<35:51:49, 5.73s/it]
49%|████▊ | 21406/43939 [34:31:33<35:48:22, 5.72s/it]
49%|████▊ | 21407/43939 [34:31:39<35:46:39, 5.72s/it]
49%|████▊ | 21408/43939 [34:31:45<35:44:27, 5.71s/it]
49%|████▊ | 21409/43939 [34:31:50<35:40:26, 5.70s/it]
49%|████▊ | 21410/43939 [34:31:56<35:41:26, 5.70s/it]
49%|████▊ | 21411/43939 [34:32:02<35:40:47, 5.70s/it]
49%|████▊ | 21412/43939 [34:32:07<35:42:08, 5.71s/it]
49%|████▊ | 21413/43939 [34:32:13<35:37:44, 5.69s/it]
49%|████▊ | 21414/43939 [34:32:19<35:35:23, 5.69s/it]
49%|████▊ | 21415/43939 [34:32:25<35:36:34, 5.69s/it]
49%|████▊ | 21416/43939 [34:32:30<35:39:00, 5.70s/it]
49%|████▊ | 21417/43939 [34:32:36<35:36:43, 5.69s/it]
49%|████▊ | 21418/43939 [34:32:42<35:32:31, 5.68s/it]
49%|████▊ | 21419/43939 [34:32:47<35:34:14, 5.69s/it]
49%|████▊ | 21420/43939 [34:32:53<35:35:39, 5.69s/it]
49%|████▉ | 21421/43939 [34:32:59<35:35:55, 5.69s/it]
49%|████▉ | 21422/43939 [34:33:04<35:39:43, 5.70s/it]
49%|████▉ | 21423/43939 [34:33:10<35:38:12, 5.70s/it]
49%|████▉ | 21424/43939 [34:33:16<36:00:53, 5.76s/it]
49%|████▉ | 21425/43939 [34:33:22<36:01:53, 5.76s/it]
49%|████▉ | 21426/43939 [34:33:27<35:54:18, 5.74s/it]
49%|████▉ | 21427/43939 [34:33:33<35:45:18, 5.72s/it]
49%|████▉ | 21428/43939 [34:33:39<35:42:25, 5.71s/it]
49%|████▉ | 21429/43939 [34:33:44<35:38:44, 5.70s/it]
49%|████▉ | 21430/43939 [34:33:50<35:39:22, 5.70s/it]
49%|████▉ | 21431/43939 [34:33:56<35:43:49, 5.71s/it]
49%|████▉ | 21432/43939 [34:34:02<35:47:25, 5.72s/it]
49%|████▉ | 21433/43939 [34:34:07<35:50:26, 5.73s/it]
49%|████▉ | 21434/43939 [34:34:13<35:51:52, 5.74s/it]
49%|████▉ | 21435/43939 [34:34:19<35:52:38, 5.74s/it]
49%|████▉ | 21436/43939 [34:34:25<35:48:21, 5.73s/it]
49%|████▉ | 21437/43939 [34:34:30<35:43:48, 5.72s/it]
49%|████▉ | 21438/43939 [34:34:36<35:40:47, 5.71s/it]
49%|████▉ | 21439/43939 [34:34:42<35:41:11, 5.71s/it]
49%|████▉ | 21440/43939 [34:34:47<35:44:08, 5.72s/it]
49%|████▉ | 21441/43939 [34:34:53<35:41:45, 5.71s/it]
49%|████▉ | 21442/43939 [34:34:59<35:41:06, 5.71s/it]
49%|████▉ | 21443/43939 [34:35:05<35:39:46, 5.71s/it]
49%|████▉ | 21444/43939 [34:35:10<35:40:31, 5.71s/it]
49%|████▉ | 21445/43939 [34:35:16<36:06:56, 5.78s/it]
49%|████▉ | 21446/43939 [34:35:22<35:55:29, 5.75s/it]
49%|████▉ | 21447/43939 [34:35:28<35:51:31, 5.74s/it]
49%|████▉ | 21448/43939 [34:35:33<35:44:28, 5.72s/it]
49%|████▉ | 21449/43939 [34:35:39<35:46:47, 5.73s/it]
49%|████▉ | 21450/43939 [34:35:45<35:39:57, 5.71s/it]
{'loss': 0.9047, 'grad_norm': 1.3599040508270264, 'learning_rate': 3.035426121368603e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.0, 'epoch': 1.95}
49%|████▉ | 21450/43939 [34:35:45<35:39:57, 5.71s/it]
49%|████▉ | 21451/43939 [34:35:50<35:39:18, 5.71s/it]
49%|████▉ | 21452/43939 [34:35:56<35:44:37, 5.72s/it]
49%|████▉ | 21453/43939 [34:36:02<35:38:24, 5.71s/it]
49%|████▉ | 21454/43939 [34:36:08<35:39:52, 5.71s/it]
49%|████▉ | 21455/43939 [34:36:13<35:38:14, 5.71s/it]
49%|████▉ | 21456/43939 [34:36:19<35:35:00, 5.70s/it]
49%|████▉ | 21457/43939 [34:36:25<35:31:12, 5.69s/it]
49%|████▉ | 21458/43939 [34:36:30<35:32:09, 5.69s/it]
49%|████▉ | 21459/43939 [34:36:36<35:33:18, 5.69s/it]
49%|████▉ | 21460/43939 [34:36:42<35:37:07, 5.70s/it]
49%|████▉ | 21461/43939 [34:36:47<35:34:13, 5.70s/it]
49%|████▉ | 21462/43939 [34:36:53<35:38:05, 5.71s/it]
49%|████▉ | 21463/43939 [34:36:59<35:41:23, 5.72s/it]
49%|████▉ | 21464/43939 [34:37:05<35:41:56, 5.72s/it]
49%|████▉ | 21465/43939 [34:37:10<35:44:10, 5.72s/it]
49%|████▉ | 21466/43939 [34:37:16<36:09:12, 5.79s/it]
49%|████▉ | 21467/43939 [34:37:22<36:07:35, 5.79s/it]
49%|████▉ | 21468/43939 [34:37:28<36:00:12, 5.77s/it]
49%|████▉ | 21469/43939 [34:37:34<35:58:46, 5.76s/it]
49%|████▉ | 21470/43939 [34:37:39<35:49:37, 5.74s/it]
49%|████▉ | 21471/43939 [34:37:45<35:44:12, 5.73s/it]
49%|████▉ | 21472/43939 [34:37:51<35:44:56, 5.73s/it]
49%|████▉ | 21473/43939 [34:37:56<35:38:27, 5.71s/it]
49%|████▉ | 21474/43939 [34:38:02<35:38:53, 5.71s/it]
49%|████▉ | 21475/43939 [34:38:08<35:45:53, 5.73s/it]
49%|████▉ | 21476/43939 [34:38:13<35:40:33, 5.72s/it]
49%|████▉ | 21477/43939 [34:38:19<35:34:00, 5.70s/it]
49%|████▉ | 21478/43939 [34:38:25<35:31:54, 5.69s/it]
49%|████▉ | 21479/43939 [34:38:31<35:31:53, 5.70s/it]
49%|████▉ | 21480/43939 [34:38:36<35:33:15, 5.70s/it]
49%|████▉ | 21481/43939 [34:38:42<35:34:07, 5.70s/it]
49%|████▉ | 21482/43939 [34:38:48<35:32:59, 5.70s/it]
49%|████▉ | 21483/43939 [34:38:53<35:35:19, 5.71s/it]
49%|████▉ | 21484/43939 [34:38:59<35:32:21, 5.70s/it]
49%|████▉ | 21485/43939 [34:39:05<35:33:03, 5.70s/it]
49%|████▉ | 21486/43939 [34:39:10<35:28:55, 5.69s/it]
49%|████▉ | 21487/43939 [34:39:16<35:53:56, 5.76s/it]
49%|████▉ | 21488/43939 [34:39:22<35:46:19, 5.74s/it]
49%|████▉ | 21489/43939 [34:39:28<35:41:34, 5.72s/it]
49%|████▉ | 21490/43939 [34:39:33<35:43:45, 5.73s/it]
49%|████▉ | 21491/43939 [34:39:39<35:36:49, 5.71s/it]
49%|████▉ | 21492/43939 [34:39:45<35:42:08, 5.73s/it]
49%|████▉ | 21493/43939 [34:39:51<35:46:19, 5.74s/it]
49%|████▉ | 21494/43939 [34:39:56<35:42:27, 5.73s/it]
49%|████▉ | 21495/43939 [34:40:02<35:41:53, 5.73s/it]
49%|████▉ | 21496/43939 [34:40:08<35:45:17, 5.74s/it]
49%|████▉ | 21497/43939 [34:40:14<35:47:59, 5.74s/it]
49%|████▉ | 21498/43939 [34:40:19<35:43:56, 5.73s/it]
49%|████▉ | 21499/43939 [34:40:25<35:38:34, 5.72s/it]
49%|████▉ | 21500/43939 [34:40:31<35:35:08, 5.71s/it]
{'loss': 0.9173, 'grad_norm': 1.3930363655090332, 'learning_rate': 3.025722155422268e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.52, 'epoch': 1.96}
49%|████▉ | 21500/43939 [34:40:31<35:35:08, 5.71s/it]
49%|████▉ | 21501/43939 [34:40:36<35:34:25, 5.71s/it]
49%|████▉ | 21502/43939 [34:40:42<35:39:01, 5.72s/it]
49%|████▉ | 21503/43939 [34:40:48<35:40:03, 5.72s/it]
49%|████▉ | 21504/43939 [34:40:54<35:38:04, 5.72s/it]
49%|████▉ | 21505/43939 [34:40:59<35:37:02, 5.72s/it]
49%|████▉ | 21506/43939 [34:41:05<35:35:30, 5.71s/it]
49%|████▉ | 21507/43939 [34:41:11<35:37:45, 5.72s/it]
49%|████▉ | 21508/43939 [34:41:16<35:32:31, 5.70s/it]
49%|████▉ | 21509/43939 [34:41:22<35:58:25, 5.77s/it]
49%|████▉ | 21510/43939 [34:41:28<35:54:30, 5.76s/it]
49%|████▉ | 21511/43939 [34:41:34<35:49:17, 5.75s/it]
49%|████▉ | 21512/43939 [34:41:39<35:44:57, 5.74s/it]
49%|████▉ | 21513/43939 [34:41:45<35:38:05, 5.72s/it]
49%|████▉ | 21514/43939 [34:41:51<35:36:46, 5.72s/it]
49%|████▉ | 21515/43939 [34:41:57<35:34:17, 5.71s/it]
49%|████▉ | 21516/43939 [34:42:02<35:31:05, 5.70s/it]
49%|████▉ | 21517/43939 [34:42:08<35:30:03, 5.70s/it]
49%|████▉ | 21518/43939 [34:42:14<35:26:32, 5.69s/it]
49%|████▉ | 21519/43939 [34:42:19<35:28:13, 5.70s/it]
49%|████▉ | 21520/43939 [34:42:25<35:29:05, 5.70s/it]
49%|████▉ | 21521/43939 [34:42:31<35:29:24, 5.70s/it]
49%|████▉ | 21522/43939 [34:42:36<35:30:56, 5.70s/it]
49%|████▉ | 21523/43939 [34:42:42<35:29:29, 5.70s/it]
49%|████▉ | 21524/43939 [34:42:48<35:32:09, 5.71s/it]
49%|████▉ | 21525/43939 [34:42:54<35:32:21, 5.71s/it]
49%|████▉ | 21526/43939 [34:42:59<35:31:48, 5.71s/it]
49%|████▉ | 21527/43939 [34:43:05<35:29:57, 5.70s/it]
49%|████▉ | 21528/43939 [34:43:11<35:30:02, 5.70s/it]
49%|████▉ | 21529/43939 [34:43:16<35:32:17, 5.71s/it]
49%|████▉ | 21530/43939 [34:43:22<35:32:56, 5.71s/it]
49%|████▉ | 21531/43939 [34:43:28<35:52:44, 5.76s/it]
49%|████▉ | 21532/43939 [34:43:34<35:42:25, 5.74s/it]
49%|████▉ | 21533/43939 [34:43:39<35:35:41, 5.72s/it]
49%|████▉ | 21534/43939 [34:43:45<35:28:45, 5.70s/it]
49%|████▉ | 21535/43939 [34:43:51<35:32:20, 5.71s/it]
49%|████▉ | 21536/43939 [34:43:56<35:29:57, 5.70s/it]
49%|████▉ | 21537/43939 [34:44:02<35:29:14, 5.70s/it]
49%|████▉ | 21538/43939 [34:44:08<35:26:14, 5.70s/it]
49%|████▉ | 21539/43939 [34:44:13<35:24:34, 5.69s/it]
49%|████▉ | 21540/43939 [34:44:19<35:26:03, 5.70s/it]
49%|████▉ | 21541/43939 [34:44:25<35:30:27, 5.71s/it]
49%|████▉ | 21542/43939 [34:44:31<35:29:21, 5.70s/it]
49%|████▉ | 21543/43939 [34:44:36<35:30:50, 5.71s/it]
49%|████▉ | 21544/43939 [34:44:42<35:31:19, 5.71s/it]
49%|████▉ | 21545/43939 [34:44:48<35:34:35, 5.72s/it]
49%|████▉ | 21546/43939 [34:44:53<35:30:01, 5.71s/it]
49%|████▉ | 21547/43939 [34:44:59<35:28:44, 5.70s/it]
49%|████▉ | 21548/43939 [34:45:05<35:25:26, 5.70s/it]
49%|████▉ | 21549/43939 [34:45:11<35:29:25, 5.71s/it]
49%|████▉ | 21550/43939 [34:45:16<35:28:44, 5.70s/it]
{'loss': 0.914, 'grad_norm': 0.9727678894996643, 'learning_rate': 3.0160098949735012e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.03, 'epoch': 1.96}
49%|████▉ | 21550/43939 [34:45:16<35:28:44, 5.70s/it]
49%|████▉ | 21551/43939 [34:45:22<35:54:25, 5.77s/it]
49%|████▉ | 21552/43939 [34:45:28<35:43:21, 5.74s/it]
49%|████▉ | 21553/43939 [34:45:34<35:37:25, 5.73s/it]
49%|████▉ | 21554/43939 [34:45:39<35:31:59, 5.71s/it]
49%|████▉ | 21555/43939 [34:45:45<35:26:45, 5.70s/it]
49%|████▉ | 21556/43939 [34:45:51<35:28:17, 5.71s/it]
49%|████▉ | 21557/43939 [34:45:56<35:23:08, 5.69s/it]
49%|████▉ | 21558/43939 [34:46:02<35:23:07, 5.69s/it]
49%|████▉ | 21559/43939 [34:46:08<35:18:54, 5.68s/it]
49%|████▉ | 21560/43939 [34:46:13<35:19:47, 5.68s/it]
49%|████▉ | 21561/43939 [34:46:19<35:20:06, 5.68s/it]
49%|████▉ | 21562/43939 [34:46:25<35:20:28, 5.69s/it]
49%|████▉ | 21563/43939 [34:46:30<35:21:43, 5.69s/it]
49%|████▉ | 21564/43939 [34:46:36<35:21:40, 5.69s/it]
49%|████▉ | 21565/43939 [34:46:42<35:18:26, 5.68s/it]
49%|████▉ | 21566/43939 [34:46:47<35:18:49, 5.68s/it]
49%|████▉ | 21567/43939 [34:46:53<35:21:41, 5.69s/it]
49%|████▉ | 21568/43939 [34:46:59<35:23:53, 5.70s/it]
49%|████▉ | 21569/43939 [34:47:05<35:25:18, 5.70s/it]
49%|████▉ | 21570/43939 [34:47:10<35:30:05, 5.71s/it]
49%|████▉ | 21571/43939 [34:47:16<35:27:13, 5.71s/it]
49%|████▉ | 21572/43939 [34:47:22<35:52:34, 5.77s/it]
49%|████▉ | 21573/43939 [34:47:28<35:47:31, 5.76s/it]
49%|████▉ | 21574/43939 [34:47:33<35:43:27, 5.75s/it]
49%|████▉ | 21575/43939 [34:47:39<35:37:18, 5.73s/it]
49%|████▉ | 21576/43939 [34:47:45<35:35:14, 5.73s/it]
49%|████▉ | 21577/43939 [34:47:51<35:32:23, 5.72s/it]
49%|████▉ | 21578/43939 [34:47:56<35:34:33, 5.73s/it]
49%|████▉ | 21579/43939 [34:48:02<35:34:11, 5.73s/it]
49%|████▉ | 21580/43939 [34:48:08<35:31:05, 5.72s/it]
49%|████▉ | 21581/43939 [34:48:13<35:32:59, 5.72s/it]
49%|████▉ | 21582/43939 [34:48:19<35:30:43, 5.72s/it]
49%|████▉ | 21583/43939 [34:48:25<35:29:32, 5.72s/it]
49%|████▉ | 21584/43939 [34:48:31<35:25:49, 5.71s/it]
49%|████▉ | 21585/43939 [34:48:36<35:21:52, 5.70s/it]
49%|████▉ | 21586/43939 [34:48:42<35:21:52, 5.70s/it]
49%|████▉ | 21587/43939 [34:48:48<35:19:44, 5.69s/it]
49%|████▉ | 21588/43939 [34:48:53<35:27:50, 5.71s/it]
49%|████▉ | 21589/43939 [34:48:59<35:23:57, 5.70s/it]
49%|████▉ | 21590/43939 [34:49:05<35:22:42, 5.70s/it]
49%|████▉ | 21591/43939 [34:49:10<35:27:06, 5.71s/it]
49%|████▉ | 21592/43939 [34:49:16<35:27:01, 5.71s/it]
49%|████▉ | 21593/43939 [34:49:22<35:52:04, 5.78s/it]
49%|████▉ | 21594/43939 [34:49:28<35:45:20, 5.76s/it]
49%|████▉ | 21595/43939 [34:49:33<35:37:27, 5.74s/it]
49%|████▉ | 21596/43939 [34:49:39<35:32:35, 5.73s/it]
49%|████▉ | 21597/43939 [34:49:45<35:28:18, 5.72s/it]
49%|████▉ | 21598/43939 [34:49:51<35:32:45, 5.73s/it]
49%|████▉ | 21599/43939 [34:49:56<35:27:01, 5.71s/it]
49%|████▉ | 21600/43939 [34:50:02<35:27:36, 5.71s/it]
{'loss': 0.9023, 'grad_norm': 1.0985159873962402, 'learning_rate': 3.0062894932560337e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.01, 'epoch': 1.97}
49%|████▉ | 21600/43939 [34:50:02<35:27:36, 5.71s/it]
49%|████▉ | 21601/43939 [34:50:08<35:25:56, 5.71s/it]
49%|████▉ | 21602/43939 [34:50:13<35:27:32, 5.71s/it]
49%|████▉ | 21603/43939 [34:50:19<35:27:07, 5.71s/it]
49%|████▉ | 21604/43939 [34:50:25<35:25:56, 5.71s/it]
49%|████▉ | 21605/43939 [34:50:31<35:25:00, 5.71s/it]
49%|████▉ | 21606/43939 [34:50:36<35:24:19, 5.71s/it]
49%|████▉ | 21607/43939 [34:50:42<35:26:21, 5.71s/it]
49%|████▉ | 21608/43939 [34:50:48<35:29:38, 5.72s/it]
49%|████▉ | 21609/43939 [34:50:53<35:32:47, 5.73s/it]
49%|████▉ | 21610/43939 [34:50:59<35:32:55, 5.73s/it]
49%|████▉ | 21611/43939 [34:51:05<35:36:04, 5.74s/it]
49%|████▉ | 21612/43939 [34:51:11<35:28:45, 5.72s/it]
49%|████▉ | 21613/43939 [34:51:16<35:25:02, 5.71s/it]
49%|████▉ | 21614/43939 [34:51:22<35:52:31, 5.79s/it]
49%|████▉ | 21615/43939 [34:51:28<35:46:45, 5.77s/it]
49%|████▉ | 21616/43939 [34:51:34<35:38:10, 5.75s/it]
49%|████▉ | 21617/43939 [34:51:39<35:36:17, 5.74s/it]
49%|████▉ | 21618/43939 [34:51:45<35:36:33, 5.74s/it]
49%|████▉ | 21619/43939 [34:51:51<35:33:15, 5.73s/it]
49%|████▉ | 21620/43939 [34:51:57<35:29:06, 5.72s/it]
49%|████▉ | 21621/43939 [34:52:02<35:28:04, 5.72s/it]
49%|████▉ | 21622/43939 [34:52:08<35:25:31, 5.71s/it]
49%|████▉ | 21623/43939 [34:52:14<35:25:13, 5.71s/it]
49%|████▉ | 21624/43939 [34:52:19<35:21:48, 5.71s/it]
49%|████▉ | 21625/43939 [34:52:25<35:22:11, 5.71s/it]
49%|████▉ | 21626/43939 [34:52:31<35:26:02, 5.72s/it]
49%|████▉ | 21627/43939 [34:52:37<35:23:48, 5.71s/it]
49%|████▉ | 21628/43939 [34:52:42<35:23:56, 5.71s/it]
49%|████▉ | 21629/43939 [34:52:48<35:25:12, 5.72s/it]
49%|████▉ | 21630/43939 [34:52:54<35:28:14, 5.72s/it]
49%|████▉ | 21631/43939 [34:52:59<35:25:36, 5.72s/it]
49%|████▉ | 21632/43939 [34:53:05<35:25:22, 5.72s/it]
49%|████▉ | 21633/43939 [34:53:11<35:21:50, 5.71s/it]
49%|████▉ | 21634/43939 [34:53:17<35:22:30, 5.71s/it]
49%|████▉ | 21635/43939 [34:53:23<35:54:26, 5.80s/it]
49%|████▉ | 21636/43939 [34:53:28<35:44:17, 5.77s/it]
49%|████▉ | 21637/43939 [34:53:34<35:36:25, 5.75s/it]
49%|████▉ | 21638/43939 [34:53:40<35:38:29, 5.75s/it]
49%|████▉ | 21639/43939 [34:53:45<35:37:06, 5.75s/it]
49%|████▉ | 21640/43939 [34:53:51<35:31:38, 5.74s/it]
49%|████▉ | 21641/43939 [34:53:57<35:22:50, 5.71s/it]
49%|████▉ | 21642/43939 [34:54:03<35:18:00, 5.70s/it]
49%|████▉ | 21643/43939 [34:54:08<35:17:41, 5.70s/it]
49%|████▉ | 21644/43939 [34:54:14<35:16:37, 5.70s/it]
49%|████▉ | 21645/43939 [34:54:20<35:16:23, 5.70s/it]
49%|████▉ | 21646/43939 [34:54:25<35:14:40, 5.69s/it]
49%|████▉ | 21647/43939 [34:54:31<35:17:15, 5.70s/it]
49%|████▉ | 21648/43939 [34:54:37<35:16:51, 5.70s/it]
49%|████▉ | 21649/43939 [34:54:42<35:14:43, 5.69s/it]
49%|████▉ | 21650/43939 [34:54:48<35:18:12, 5.70s/it]
{'loss': 0.9078, 'grad_norm': 1.1219627857208252, 'learning_rate': 2.9965611036320475e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.04, 'epoch': 1.97}
49%|████▉ | 21650/43939 [34:54:48<35:18:12, 5.70s/it]
49%|████▉ | 21651/43939 [34:54:54<35:18:21, 5.70s/it]
49%|████▉ | 21652/43939 [34:55:00<35:18:05, 5.70s/it]
49%|████▉ | 21653/43939 [34:55:05<35:16:35, 5.70s/it]
49%|████▉ | 21654/43939 [34:55:11<35:18:17, 5.70s/it]
49%|████▉ | 21655/43939 [34:55:17<35:18:44, 5.70s/it]
49%|████▉ | 21656/43939 [34:55:23<35:41:34, 5.77s/it]
49%|████▉ | 21657/43939 [34:55:28<35:32:54, 5.74s/it]
49%|████▉ | 21658/43939 [34:55:34<35:26:18, 5.73s/it]
49%|████▉ | 21659/43939 [34:55:40<35:23:04, 5.72s/it]
49%|████▉ | 21660/43939 [34:55:45<35:21:15, 5.71s/it]
49%|████▉ | 21661/43939 [34:55:51<35:16:08, 5.70s/it]
49%|████▉ | 21662/43939 [34:55:57<35:14:28, 5.70s/it]
49%|████▉ | 21663/43939 [34:56:02<35:17:42, 5.70s/it]
49%|████▉ | 21664/43939 [34:56:08<35:11:47, 5.69s/it]
49%|████▉ | 21665/43939 [34:56:14<35:10:27, 5.68s/it]
49%|████▉ | 21666/43939 [34:56:19<35:08:49, 5.68s/it]
49%|████▉ | 21667/43939 [34:56:25<35:10:40, 5.69s/it]
49%|████▉ | 21668/43939 [34:56:31<35:08:54, 5.68s/it]
49%|████▉ | 21669/43939 [34:56:36<35:10:52, 5.69s/it]
49%|████▉ | 21670/43939 [34:56:42<35:07:29, 5.68s/it]
49%|████▉ | 21671/43939 [34:56:48<35:07:16, 5.68s/it]
49%|████▉ | 21672/43939 [34:56:53<35:10:32, 5.69s/it]
49%|████▉ | 21673/43939 [34:56:59<35:07:39, 5.68s/it]
49%|████▉ | 21674/43939 [34:57:05<35:10:16, 5.69s/it]
49%|████▉ | 21675/43939 [34:57:11<35:09:20, 5.68s/it]
49%|████▉ | 21676/43939 [34:57:16<35:10:20, 5.69s/it]
49%|████▉ | 21677/43939 [34:57:22<35:09:00, 5.68s/it]
49%|████▉ | 21678/43939 [34:57:28<35:41:15, 5.77s/it]
49%|████▉ | 21679/43939 [34:57:34<35:32:53, 5.75s/it]
49%|████▉ | 21680/43939 [34:57:39<35:27:06, 5.73s/it]
49%|████▉ | 21681/43939 [34:57:45<35:18:55, 5.71s/it]
49%|████▉ | 21682/43939 [34:57:51<35:14:44, 5.70s/it]
49%|████▉ | 21683/43939 [34:57:56<35:15:33, 5.70s/it]
49%|████▉ | 21684/43939 [34:58:02<35:16:30, 5.71s/it]
49%|████▉ | 21685/43939 [34:58:08<35:15:03, 5.70s/it]
49%|████▉ | 21686/43939 [34:58:13<35:15:13, 5.70s/it]
49%|████▉ | 21687/43939 [34:58:19<35:14:23, 5.70s/it]
49%|████▉ | 21688/43939 [34:58:25<35:15:02, 5.70s/it]
49%|████▉ | 21689/43939 [34:58:31<35:17:26, 5.71s/it]
49%|████▉ | 21690/43939 [34:58:36<35:16:04, 5.71s/it]
49%|████▉ | 21691/43939 [34:58:42<35:15:51, 5.71s/it]
49%|████▉ | 21692/43939 [34:58:48<35:11:26, 5.69s/it]
49%|████▉ | 21693/43939 [34:58:53<35:13:56, 5.70s/it]
49%|████▉ | 21694/43939 [34:58:59<35:15:40, 5.71s/it]
49%|████▉ | 21695/43939 [34:59:05<35:16:55, 5.71s/it]
49%|████▉ | 21696/43939 [34:59:10<35:16:33, 5.71s/it]
49%|████▉ | 21697/43939 [34:59:16<35:18:30, 5.71s/it]
49%|████▉ | 21698/43939 [34:59:22<35:18:58, 5.72s/it]
49%|████▉ | 21699/43939 [34:59:28<35:43:04, 5.78s/it]
49%|████▉ | 21700/43939 [34:59:34<35:30:34, 5.75s/it]
{'loss': 0.9023, 'grad_norm': 1.338707447052002, 'learning_rate': 2.986824879589748e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.85, 'epoch': 1.98}
49%|████▉ | 21700/43939 [34:59:34<35:30:34, 5.75s/it]
49%|████▉ | 21701/43939 [34:59:39<35:22:25, 5.73s/it]
49%|████▉ | 21702/43939 [34:59:45<35:18:31, 5.72s/it]
49%|████▉ | 21703/43939 [34:59:51<35:13:01, 5.70s/it]
49%|████▉ | 21704/43939 [34:59:56<35:15:53, 5.71s/it]
49%|████▉ | 21705/43939 [35:00:02<35:11:41, 5.70s/it]
49%|████▉ | 21706/43939 [35:00:08<35:09:28, 5.69s/it]
49%|████▉ | 21707/43939 [35:00:13<35:07:43, 5.69s/it]
49%|████▉ | 21708/43939 [35:00:19<35:15:06, 5.71s/it]
49%|████▉ | 21709/43939 [35:00:25<35:14:38, 5.71s/it]
49%|████▉ | 21710/43939 [35:00:31<35:18:38, 5.72s/it]
49%|████▉ | 21711/43939 [35:00:36<35:17:32, 5.72s/it]
49%|████▉ | 21712/43939 [35:00:42<35:19:34, 5.72s/it]
49%|████▉ | 21713/43939 [35:00:48<35:15:05, 5.71s/it]
49%|████▉ | 21714/43939 [35:00:53<35:12:36, 5.70s/it]
49%|████▉ | 21715/43939 [35:00:59<35:12:55, 5.70s/it]
49%|████▉ | 21716/43939 [35:01:05<35:11:06, 5.70s/it]
49%|████▉ | 21717/43939 [35:01:10<35:08:40, 5.69s/it]
49%|████▉ | 21718/43939 [35:01:16<35:12:59, 5.71s/it]
49%|████▉ | 21719/43939 [35:01:22<35:14:19, 5.71s/it]
49%|████▉ | 21720/43939 [35:01:28<35:47:26, 5.80s/it]
49%|████▉ | 21721/43939 [35:01:34<35:34:06, 5.76s/it]
49%|████▉ | 21722/43939 [35:01:39<35:28:05, 5.75s/it]
49%|████▉ | 21723/43939 [35:01:45<35:17:17, 5.72s/it]
49%|████▉ | 21724/43939 [35:01:51<35:13:40, 5.71s/it]
49%|████▉ | 21725/43939 [35:01:56<35:09:54, 5.70s/it]
49%|████▉ | 21726/43939 [35:02:02<35:16:08, 5.72s/it]
49%|████▉ | 21727/43939 [35:02:08<35:12:02, 5.71s/it]
49%|████▉ | 21728/43939 [35:02:13<35:12:20, 5.71s/it]
49%|████▉ | 21729/43939 [35:02:19<35:16:05, 5.72s/it]
49%|████▉ | 21730/43939 [35:02:25<35:14:14, 5.71s/it]
49%|████▉ | 21731/43939 [35:02:31<35:11:20, 5.70s/it]
49%|████▉ | 21732/43939 [35:02:36<35:09:38, 5.70s/it]
49%|████▉ | 21733/43939 [35:02:42<35:08:06, 5.70s/it]
49%|████▉ | 21734/43939 [35:02:48<35:09:31, 5.70s/it]
49%|████▉ | 21735/43939 [35:02:53<35:05:00, 5.69s/it]
49%|████▉ | 21736/43939 [35:02:59<35:03:11, 5.68s/it]
49%|████▉ | 21737/43939 [35:03:05<35:14:46, 5.72s/it]
49%|████▉ | 21738/43939 [35:03:10<35:14:54, 5.72s/it]
49%|████▉ | 21739/43939 [35:03:16<35:16:48, 5.72s/it]
49%|████▉ | 21740/43939 [35:03:22<35:11:06, 5.71s/it]
49%|████▉ | 21741/43939 [35:03:28<35:10:00, 5.70s/it]
49%|████▉ | 21742/43939 [35:03:34<35:35:59, 5.77s/it]
49%|████▉ | 21743/43939 [35:03:39<35:25:14, 5.74s/it]
49%|████▉ | 21744/43939 [35:03:45<35:22:54, 5.74s/it]
49%|████▉ | 21745/43939 [35:03:51<35:14:11, 5.72s/it]
49%|████▉ | 21746/43939 [35:03:56<35:13:32, 5.71s/it]
49%|████▉ | 21747/43939 [35:04:02<35:11:57, 5.71s/it]
49%|████▉ | 21748/43939 [35:04:08<35:08:07, 5.70s/it]
49%|████▉ | 21749/43939 [35:04:13<35:04:33, 5.69s/it]
50%|████▉ | 21750/43939 [35:04:19<35:07:08, 5.70s/it]
{'loss': 0.9288, 'grad_norm': 1.3754671812057495, 'learning_rate': 2.9770809747409507e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.93, 'epoch': 1.98}
50%|████▉ | 21750/43939 [35:04:19<35:07:08, 5.70s/it]
50%|████▉ | 21751/43939 [35:04:25<35:09:47, 5.71s/it]
50%|████▉ | 21752/43939 [35:04:30<35:09:29, 5.70s/it]
50%|████▉ | 21753/43939 [35:04:36<35:09:36, 5.71s/it]
50%|████▉ | 21754/43939 [35:04:42<35:06:54, 5.70s/it]
50%|████▉ | 21755/43939 [35:04:48<35:06:31, 5.70s/it]
50%|████▉ | 21756/43939 [35:04:53<35:04:58, 5.69s/it]
50%|████▉ | 21757/43939 [35:04:59<35:04:41, 5.69s/it]
50%|████▉ | 21758/43939 [35:05:05<35:07:48, 5.70s/it]
50%|████▉ | 21759/43939 [35:05:10<35:10:49, 5.71s/it]
50%|████▉ | 21760/43939 [35:05:16<35:13:51, 5.72s/it]
50%|████▉ | 21761/43939 [35:05:22<35:11:23, 5.71s/it]
50%|████▉ | 21762/43939 [35:05:28<35:07:46, 5.70s/it]
50%|████▉ | 21763/43939 [35:05:33<35:07:48, 5.70s/it]
50%|████▉ | 21764/43939 [35:05:39<35:32:48, 5.77s/it]
50%|████▉ | 21765/43939 [35:05:45<35:23:12, 5.75s/it]
50%|████▉ | 21766/43939 [35:05:51<35:22:09, 5.74s/it]
50%|████▉ | 21767/43939 [35:05:56<35:17:55, 5.73s/it]
50%|████▉ | 21768/43939 [35:06:02<35:13:16, 5.72s/it]
50%|████▉ | 21769/43939 [35:06:08<35:09:22, 5.71s/it]
50%|████▉ | 21770/43939 [35:06:13<35:13:10, 5.72s/it]
50%|████▉ | 21771/43939 [35:06:19<35:14:40, 5.72s/it]
50%|████▉ | 21772/43939 [35:06:25<35:12:05, 5.72s/it]
50%|████▉ | 21773/43939 [35:06:31<35:10:52, 5.71s/it]
50%|████▉ | 21774/43939 [35:06:36<35:07:10, 5.70s/it]
50%|████▉ | 21775/43939 [35:06:42<35:04:57, 5.70s/it]
50%|████▉ | 21776/43939 [35:06:48<35:06:07, 5.70s/it]
50%|████▉ | 21777/43939 [35:06:53<35:04:53, 5.70s/it]
50%|████▉ | 21778/43939 [35:06:59<35:07:51, 5.71s/it]
50%|████▉ | 21779/43939 [35:07:05<35:03:46, 5.70s/it]
50%|████▉ | 21780/43939 [35:07:10<35:10:56, 5.72s/it]
50%|████▉ | 21781/43939 [35:07:16<35:05:48, 5.70s/it]
50%|████▉ | 21782/43939 [35:07:22<35:07:12, 5.71s/it]
50%|████▉ | 21783/43939 [35:07:28<35:31:38, 5.77s/it]
50%|████▉ | 21784/43939 [35:07:34<35:30:14, 5.77s/it]
50%|████▉ | 21785/43939 [35:07:39<35:19:10, 5.74s/it]
50%|████▉ | 21786/43939 [35:07:45<35:15:14, 5.73s/it]
50%|████▉ | 21787/43939 [35:07:51<35:13:18, 5.72s/it]
50%|████▉ | 21788/43939 [35:07:56<35:09:52, 5.71s/it]
50%|████▉ | 21789/43939 [35:08:02<35:05:01, 5.70s/it]
50%|████▉ | 21790/43939 [35:08:08<35:06:14, 5.71s/it]
50%|████▉ | 21791/43939 [35:08:13<35:02:47, 5.70s/it]
50%|████▉ | 21792/43939 [35:08:19<35:02:52, 5.70s/it]
50%|████▉ | 21793/43939 [35:08:25<35:03:09, 5.70s/it]
50%|████▉ | 21794/43939 [35:08:30<34:59:25, 5.69s/it]
50%|████▉ | 21795/43939 [35:08:36<34:58:31, 5.69s/it]
50%|████▉ | 21796/43939 [35:08:42<35:03:34, 5.70s/it]
50%|████▉ | 21797/43939 [35:08:48<34:57:59, 5.69s/it]
50%|████▉ | 21798/43939 [35:08:53<35:02:49, 5.70s/it]
50%|████▉ | 21799/43939 [35:08:59<35:02:06, 5.70s/it]
50%|████▉ | 21800/43939 [35:09:05<34:58:31, 5.69s/it]
{'loss': 0.9, 'grad_norm': 1.4110934734344482, 'learning_rate': 2.9673295428186514e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.3, 'epoch': 1.98}
50%|████▉ | 21800/43939 [35:09:05<34:58:31, 5.69s/it]
50%|████▉ | 21801/43939 [35:09:10<35:01:47, 5.70s/it]
50%|████▉ | 21802/43939 [35:09:16<35:04:33, 5.70s/it]
50%|████▉ | 21803/43939 [35:09:22<35:10:24, 5.72s/it]
50%|████▉ | 21804/43939 [35:09:28<35:08:41, 5.72s/it]
50%|████▉ | 21805/43939 [35:09:33<35:32:42, 5.78s/it]
50%|████▉ | 21806/43939 [35:09:39<35:20:49, 5.75s/it]
50%|████▉ | 21807/43939 [35:09:45<35:11:58, 5.73s/it]
50%|████▉ | 21808/43939 [35:09:50<35:10:07, 5.72s/it]
50%|████▉ | 21809/43939 [35:09:56<35:13:55, 5.73s/it]
50%|████▉ | 21810/43939 [35:10:02<35:11:43, 5.73s/it]
50%|████▉ | 21811/43939 [35:10:08<35:10:37, 5.72s/it]
50%|████▉ | 21812/43939 [35:10:13<35:04:09, 5.71s/it]
50%|████▉ | 21813/43939 [35:10:19<35:00:38, 5.70s/it]
50%|████▉ | 21814/43939 [35:10:25<34:55:33, 5.68s/it]
50%|████▉ | 21815/43939 [35:10:30<35:01:36, 5.70s/it]
50%|████▉ | 21816/43939 [35:10:36<35:01:56, 5.70s/it]
50%|████▉ | 21817/43939 [35:10:42<35:10:53, 5.73s/it]
50%|████▉ | 21818/43939 [35:10:48<35:08:26, 5.72s/it]
50%|████▉ | 21819/43939 [35:10:53<35:06:18, 5.71s/it]
50%|████▉ | 21820/43939 [35:10:59<35:03:44, 5.71s/it]
50%|████▉ | 21821/43939 [35:11:05<35:05:18, 5.71s/it]
50%|████▉ | 21822/43939 [35:11:10<35:00:34, 5.70s/it]
50%|████▉ | 21823/43939 [35:11:16<35:01:41, 5.70s/it]
50%|████▉ | 21824/43939 [35:11:22<35:02:42, 5.70s/it]
50%|████▉ | 21825/43939 [35:11:27<34:59:43, 5.70s/it]
50%|████▉ | 21826/43939 [35:11:33<35:25:28, 5.77s/it]
50%|████▉ | 21827/43939 [35:11:39<35:19:10, 5.75s/it]
50%|████▉ | 21828/43939 [35:11:45<35:13:03, 5.73s/it]
50%|████▉ | 21829/43939 [35:11:51<35:11:09, 5.73s/it]
50%|████▉ | 21830/43939 [35:11:56<35:06:06, 5.72s/it]
50%|████▉ | 21831/43939 [35:12:02<35:05:48, 5.72s/it]
50%|████▉ | 21832/43939 [35:12:08<35:10:15, 5.73s/it]
50%|████▉ | 21833/43939 [35:12:13<35:08:53, 5.72s/it]
50%|████▉ | 21834/43939 [35:12:19<35:06:38, 5.72s/it]
50%|████▉ | 21835/43939 [35:12:25<35:00:56, 5.70s/it]
50%|████▉ | 21836/43939 [35:12:31<35:05:35, 5.72s/it]
50%|████▉ | 21837/43939 [35:12:36<35:01:25, 5.70s/it]
50%|████▉ | 21838/43939 [35:12:42<35:00:24, 5.70s/it]
50%|████▉ | 21839/43939 [35:12:48<35:01:46, 5.71s/it]
50%|████▉ | 21840/43939 [35:12:53<34:58:10, 5.70s/it]
50%|████▉ | 21841/43939 [35:12:59<34:58:07, 5.70s/it]
50%|████▉ | 21842/43939 [35:13:05<35:02:19, 5.71s/it]
50%|████▉ | 21843/43939 [35:13:10<35:02:41, 5.71s/it]
50%|████▉ | 21844/43939 [35:13:16<35:12:44, 5.74s/it]
50%|████▉ | 21845/43939 [35:13:22<35:05:30, 5.72s/it]
50%|████▉ | 21846/43939 [35:13:28<35:02:28, 5.71s/it]
50%|████▉ | 21847/43939 [35:13:34<35:26:21, 5.77s/it]
50%|████▉ | 21848/43939 [35:13:39<35:20:15, 5.76s/it]
50%|████▉ | 21849/43939 [35:13:45<35:12:33, 5.74s/it]
50%|████▉ | 21850/43939 [35:13:51<35:04:22, 5.72s/it]
{'loss': 0.9236, 'grad_norm': 1.1868780851364136, 'learning_rate': 2.9575707376746053e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.74, 'epoch': 1.99}
50%|████▉ | 21850/43939 [35:13:51<35:04:22, 5.72s/it]
50%|████▉ | 21851/43939 [35:13:56<35:04:58, 5.72s/it]
50%|████▉ | 21852/43939 [35:14:02<35:01:14, 5.71s/it]
50%|████▉ | 21853/43939 [35:14:08<35:03:20, 5.71s/it]
50%|████▉ | 21854/43939 [35:14:13<35:00:34, 5.71s/it]
50%|████▉ | 21855/43939 [35:14:19<34:56:18, 5.70s/it]
50%|████▉ | 21856/43939 [35:14:25<34:53:58, 5.69s/it]
50%|████▉ | 21857/43939 [35:14:30<34:50:28, 5.68s/it]
50%|████▉ | 21858/43939 [35:14:36<34:53:50, 5.69s/it]
50%|████▉ | 21859/43939 [35:14:42<34:53:53, 5.69s/it]
50%|████▉ | 21860/43939 [35:14:48<35:01:40, 5.71s/it]
50%|████▉ | 21861/43939 [35:14:53<35:03:45, 5.72s/it]
50%|████▉ | 21862/43939 [35:14:59<35:00:18, 5.71s/it]
50%|████▉ | 21863/43939 [35:15:05<35:04:16, 5.72s/it]
50%|████▉ | 21864/43939 [35:15:10<35:00:10, 5.71s/it]
50%|████▉ | 21865/43939 [35:15:16<34:57:50, 5.70s/it]
50%|████▉ | 21866/43939 [35:15:22<34:55:34, 5.70s/it]
50%|████▉ | 21867/43939 [35:15:28<34:57:51, 5.70s/it]
50%|████▉ | 21868/43939 [35:15:33<35:20:56, 5.77s/it]
50%|████▉ | 21869/43939 [35:15:39<35:12:42, 5.74s/it]
50%|████▉ | 21870/43939 [35:15:45<35:04:10, 5.72s/it]
50%|████▉ | 21871/43939 [35:15:50<35:00:17, 5.71s/it]
50%|████▉ | 21872/43939 [35:15:56<35:00:25, 5.71s/it]
50%|████▉ | 21873/43939 [35:16:02<34:58:29, 5.71s/it]
50%|████▉ | 21874/43939 [35:16:08<34:53:27, 5.69s/it]
50%|████▉ | 21875/43939 [35:16:13<34:57:29, 5.70s/it]
50%|████▉ | 21876/43939 [35:16:19<34:57:53, 5.71s/it]
50%|████▉ | 21877/43939 [35:16:25<34:53:54, 5.69s/it]
50%|████▉ | 21878/43939 [35:16:30<34:51:00, 5.69s/it]
50%|████▉ | 21879/43939 [35:16:36<34:52:17, 5.69s/it]
50%|████▉ | 21880/43939 [35:16:42<34:55:52, 5.70s/it]
50%|████▉ | 21881/43939 [35:16:47<34:56:57, 5.70s/it]
50%|████▉ | 21882/43939 [35:16:53<34:56:05, 5.70s/it]
50%|████▉ | 21883/43939 [35:16:59<34:53:49, 5.70s/it]
50%|████▉ | 21884/43939 [35:17:05<34:53:29, 5.70s/it]
50%|████▉ | 21885/43939 [35:17:10<34:51:28, 5.69s/it]
50%|████▉ | 21886/43939 [35:17:16<34:54:47, 5.70s/it]
50%|████▉ | 21887/43939 [35:17:22<34:53:20, 5.70s/it]
50%|████▉ | 21888/43939 [35:17:27<34:53:05, 5.70s/it]
50%|████▉ | 21889/43939 [35:17:33<34:54:18, 5.70s/it]
50%|████▉ | 21890/43939 [35:17:39<35:14:35, 5.75s/it]
50%|████▉ | 21891/43939 [35:17:45<35:05:45, 5.73s/it]
50%|████▉ | 21892/43939 [35:17:50<35:00:08, 5.72s/it]
50%|████▉ | 21893/43939 [35:17:56<34:57:35, 5.71s/it]
50%|████▉ | 21894/43939 [35:18:02<34:55:57, 5.70s/it]
50%|████▉ | 21895/43939 [35:18:07<34:55:18, 5.70s/it]
50%|████▉ | 21896/43939 [35:18:13<34:55:43, 5.70s/it]
50%|████▉ | 21897/43939 [35:18:19<34:55:23, 5.70s/it]
50%|████▉ | 21898/43939 [35:18:24<34:56:10, 5.71s/it]
50%|████▉ | 21899/43939 [35:18:30<34:59:57, 5.72s/it]
50%|████▉ | 21900/43939 [35:18:36<34:57:01, 5.71s/it]
{'loss': 0.9106, 'grad_norm': 1.1212882995605469, 'learning_rate': 2.9478047132768955e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.25, 'epoch': 1.99}
50%|████▉ | 21900/43939 [35:18:36<34:57:01, 5.71s/it]
50%|████▉ | 21901/43939 [35:18:42<34:55:04, 5.70s/it]
50%|████▉ | 21902/43939 [35:18:47<34:55:49, 5.71s/it]
50%|████▉ | 21903/43939 [35:18:53<34:55:04, 5.70s/it]
50%|████▉ | 21904/43939 [35:18:59<34:54:08, 5.70s/it]
50%|████▉ | 21905/43939 [35:19:04<34:49:45, 5.69s/it]
50%|████▉ | 21906/43939 [35:19:10<34:50:55, 5.69s/it]
50%|████▉ | 21907/43939 [35:19:16<34:54:37, 5.70s/it]
50%|████▉ | 21908/43939 [35:19:22<34:56:07, 5.71s/it]
50%|████▉ | 21909/43939 [35:19:27<34:53:36, 5.70s/it]
50%|████▉ | 21910/43939 [35:19:33<34:49:22, 5.69s/it]
50%|████▉ | 21911/43939 [35:19:39<35:19:36, 5.77s/it]
50%|████▉ | 21912/43939 [35:19:45<35:09:12, 5.75s/it]
50%|████▉ | 21913/43939 [35:19:50<35:08:20, 5.74s/it]
50%|████▉ | 21914/43939 [35:19:56<35:05:27, 5.74s/it]
50%|████▉ | 21915/43939 [35:20:02<35:06:01, 5.74s/it]
50%|████▉ | 21916/43939 [35:20:07<35:02:53, 5.73s/it]
50%|████▉ | 21917/43939 [35:20:13<34:56:09, 5.71s/it]
50%|████▉ | 21918/43939 [35:20:19<34:59:37, 5.72s/it]
50%|████▉ | 21919/43939 [35:20:25<34:58:34, 5.72s/it]
50%|████▉ | 21920/43939 [35:20:30<34:55:16, 5.71s/it]
50%|████▉ | 21921/43939 [35:20:36<34:56:04, 5.71s/it]
50%|████▉ | 21922/43939 [35:20:42<34:53:29, 5.71s/it]
50%|████▉ | 21923/43939 [35:20:47<34:52:00, 5.70s/it]
50%|████▉ | 21924/43939 [35:20:53<34:53:27, 5.71s/it]
50%|████▉ | 21925/43939 [35:20:59<34:54:17, 5.71s/it]
50%|████▉ | 21926/43939 [35:21:05<34:58:50, 5.72s/it]
50%|████▉ | 21927/43939 [35:21:10<34:56:49, 5.72s/it]
50%|████▉ | 21928/43939 [35:21:16<34:51:33, 5.70s/it]
50%|████▉ | 21929/43939 [35:21:22<34:52:45, 5.70s/it]
50%|████▉ | 21930/43939 [35:21:27<34:56:49, 5.72s/it]
50%|████▉ | 21931/43939 [35:21:33<34:52:19, 5.70s/it]
50%|████▉ | 21932/43939 [35:21:39<35:16:42, 5.77s/it]
50%|████▉ | 21933/43939 [35:21:45<35:04:58, 5.74s/it]
50%|████▉ | 21934/43939 [35:21:50<35:02:20, 5.73s/it]
50%|████▉ | 21935/43939 [35:21:56<35:00:38, 5.73s/it]
50%|████▉ | 21936/43939 [35:22:02<35:01:15, 5.73s/it]
50%|████▉ | 21937/43939 [35:22:07<34:54:38, 5.71s/it]
50%|████▉ | 21938/43939 [35:22:13<34:54:21, 5.71s/it]
50%|████▉ | 21939/43939 [35:22:19<34:53:36, 5.71s/it]
50%|████▉ | 21940/43939 [35:22:25<34:50:01, 5.70s/it]
50%|████▉ | 21941/43939 [35:22:30<34:49:58, 5.70s/it]
50%|████▉ | 21942/43939 [35:22:36<34:48:13, 5.70s/it]
50%|████▉ | 21943/43939 [35:22:42<34:50:11, 5.70s/it]
50%|████▉ | 21944/43939 [35:22:47<34:51:46, 5.71s/it]
50%|████▉ | 21945/43939 [35:22:53<34:53:20, 5.71s/it]
50%|████▉ | 21946/43939 [35:22:59<34:49:18, 5.70s/it]
50%|████▉ | 21947/43939 [35:23:04<34:51:48, 5.71s/it]
50%|████▉ | 21948/43939 [35:23:10<34:53:05, 5.71s/it]
50%|████▉ | 21949/43939 [35:23:16<34:55:47, 5.72s/it]
50%|████▉ | 21950/43939 [35:23:22<34:52:48, 5.71s/it]
{'loss': 0.88, 'grad_norm': 1.2349896430969238, 'learning_rate': 2.9380316237075073e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.6, 'epoch': 2.0}
50%|████▉ | 21950/43939 [35:23:22<34:52:48, 5.71s/it]
50%|████▉ | 21951/43939 [35:23:27<35:00:51, 5.73s/it]
50%|████▉ | 21952/43939 [35:23:33<34:58:19, 5.73s/it]
50%|████▉ | 21953/43939 [35:23:39<34:54:39, 5.72s/it]
50%|████▉ | 21954/43939 [35:23:45<35:16:55, 5.78s/it]
50%|████▉ | 21955/43939 [35:23:50<35:09:12, 5.76s/it]
50%|████▉ | 21956/43939 [35:23:56<35:04:37, 5.74s/it]
50%|████▉ | 21957/43939 [35:24:02<34:59:16, 5.73s/it]
50%|████▉ | 21958/43939 [35:24:08<35:01:27, 5.74s/it]
50%|████▉ | 21959/43939 [35:24:13<34:57:25, 5.73s/it]
50%|████▉ | 21960/43939 [35:24:19<34:58:54, 5.73s/it]
50%|████▉ | 21961/43939 [35:24:25<34:50:47, 5.71s/it]
50%|████▉ | 21962/43939 [35:24:30<34:47:51, 5.70s/it]
50%|████▉ | 21963/43939 [35:24:36<34:55:33, 5.72s/it]
50%|████▉ | 21964/43939 [35:24:42<34:54:29, 5.72s/it]
50%|████▉ | 21965/43939 [35:24:48<34:52:18, 5.71s/it]
50%|████▉ | 21966/43939 [35:24:53<34:55:55, 5.72s/it]
50%|████▉ | 21967/43939 [35:24:59<34:52:25, 5.71s/it]
50%|████▉ | 21968/43939 [35:25:05<34:51:00, 5.71s/it]
50%|████▉ | 21969/43939 [35:25:10<34:50:02, 5.71s/it]
50%|█████ | 21970/43939 [35:25:14<31:33:43, 5.17s/it][2025-11-30 05:29:28,160] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.36s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:48, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.72s/it][A
29%|██▉ | 21/72 [00:54<02:18, 2.72s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.72s/it][A
40%|████ | 29/72 [01:16<01:57, 2.72s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.72s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.72s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.72s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.72s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.26s/it][A
[A{'eval_loss': 0.962039589881897, 'eval_runtime': 200.4223, 'eval_samples_per_second': 4.984, 'eval_steps_per_second': 0.359, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 2.0}
50%|█████ | 21970/43939 [35:28:35<31:33:43, 5.17s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
50%|█████ | 21971/43939 [35:28:43<403:09:50, 66.07s/it]
50%|█████ | 21972/43939 [35:28:48<292:39:09, 47.96s/it][2025-11-30 05:33:02,030] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-21972
50%|█████ | 21973/43939 [35:28:56<219:19:44, 35.95s/it]
50%|█████ | 21974/43939 [35:29:02<164:29:46, 26.96s/it]
50%|█████ | 21975/43939 [35:29:08<125:35:26, 20.58s/it]
50%|█████ | 21976/43939 [35:29:14<98:20:51, 16.12s/it]
50%|█████ | 21977/43939 [35:29:19<79:17:54, 13.00s/it]
50%|█████ | 21978/43939 [35:29:25<65:57:20, 10.81s/it]
50%|█████ | 21979/43939 [35:29:31<56:39:02, 9.29s/it]
50%|█████ | 21980/43939 [35:29:36<50:01:26, 8.20s/it]
50%|█████ | 21981/43939 [35:29:42<45:25:00, 7.45s/it]
50%|█████ | 21982/43939 [35:29:48<42:15:05, 6.93s/it]
50%|█████ | 21983/43939 [35:29:53<39:56:44, 6.55s/it]
50%|█████ | 21984/43939 [35:29:59<38:22:37, 6.29s/it]
50%|█████ | 21985/43939 [35:30:05<37:20:08, 6.12s/it]
50%|█████ | 21986/43939 [35:30:11<36:32:39, 5.99s/it]
50%|█████ | 21987/43939 [35:30:16<35:58:38, 5.90s/it]
50%|█████ | 21988/43939 [35:30:22<35:38:02, 5.84s/it]
50%|█████ | 21989/43939 [35:30:28<35:20:37, 5.80s/it]
50%|█████ | 21990/43939 [35:30:33<35:08:29, 5.76s/it]
50%|█████ | 21991/43939 [35:30:39<34:58:32, 5.74s/it]
50%|█████ | 21992/43939 [35:30:45<34:56:09, 5.73s/it]
50%|█████ | 21993/43939 [35:30:50<34:50:24, 5.72s/it]
50%|█████ | 21994/43939 [35:30:56<34:49:55, 5.71s/it]
50%|█████ | 21995/43939 [35:31:02<35:15:30, 5.78s/it]
50%|█████ | 21996/43939 [35:31:08<35:05:39, 5.76s/it]
50%|█████ | 21997/43939 [35:31:13<34:59:07, 5.74s/it]
50%|█████ | 21998/43939 [35:31:19<35:00:17, 5.74s/it]
50%|█████ | 21999/43939 [35:31:25<35:00:21, 5.74s/it]
50%|█████ | 22000/43939 [35:31:31<35:03:32, 5.75s/it]
{'loss': 0.8407, 'grad_norm': 1.1923445463180542, 'learning_rate': 2.928251623159895e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 289.45, 'epoch': 2.0}
50%|█████ | 22000/43939 [35:31:31<35:03:32, 5.75s/it]
50%|█████ | 22001/43939 [35:31:36<34:57:45, 5.74s/it]
50%|█████ | 22002/43939 [35:31:42<34:57:03, 5.74s/it]
50%|█████ | 22003/43939 [35:31:48<34:56:11, 5.73s/it]
50%|█████ | 22004/43939 [35:31:54<35:02:00, 5.75s/it]
50%|█████ | 22005/43939 [35:31:59<35:01:53, 5.75s/it]
50%|█████ | 22006/43939 [35:32:05<34:56:03, 5.73s/it]
50%|█████ | 22007/43939 [35:32:11<34:54:55, 5.73s/it]
50%|█████ | 22008/43939 [35:32:17<34:56:28, 5.74s/it]
50%|█████ | 22009/43939 [35:32:22<34:52:44, 5.73s/it]
50%|█████ | 22010/43939 [35:32:28<34:54:25, 5.73s/it]
50%|█████ | 22011/43939 [35:32:34<34:51:40, 5.72s/it]
50%|█████ | 22012/43939 [35:32:39<34:50:21, 5.72s/it]
50%|█████ | 22013/43939 [35:32:45<34:54:27, 5.73s/it]
50%|█████ | 22014/43939 [35:32:51<34:54:05, 5.73s/it]
50%|█████ | 22015/43939 [35:32:57<34:55:42, 5.74s/it]
50%|█████ | 22016/43939 [35:33:03<35:19:29, 5.80s/it]
50%|█████ | 22017/43939 [35:33:08<35:10:48, 5.78s/it]
50%|█████ | 22018/43939 [35:33:14<34:58:56, 5.75s/it]
50%|█████ | 22019/43939 [35:33:20<34:58:13, 5.74s/it]
50%|█████ | 22020/43939 [35:33:25<34:59:31, 5.75s/it]
50%|█████ | 22021/43939 [35:33:31<34:53:25, 5.73s/it]
50%|█████ | 22022/43939 [35:33:37<34:54:34, 5.73s/it]
50%|█████ | 22023/43939 [35:33:43<34:49:55, 5.72s/it]
50%|█████ | 22024/43939 [35:33:48<34:54:27, 5.73s/it]
50%|█████ | 22025/43939 [35:33:54<34:49:53, 5.72s/it]
50%|█████ | 22026/43939 [35:34:00<34:49:13, 5.72s/it]
50%|█████ | 22027/43939 [35:34:05<34:45:39, 5.71s/it]
50%|█████ | 22028/43939 [35:34:11<34:47:08, 5.72s/it]
50%|█████ | 22029/43939 [35:34:17<34:45:15, 5.71s/it]
50%|█████ | 22030/43939 [35:34:23<34:48:21, 5.72s/it]
50%|█████ | 22031/43939 [35:34:28<34:45:13, 5.71s/it]
50%|█████ | 22032/43939 [35:34:34<34:45:48, 5.71s/it]
50%|█████ | 22033/43939 [35:34:40<34:43:03, 5.71s/it]
50%|█████ | 22034/43939 [35:34:46<34:49:28, 5.72s/it]
50%|█████ | 22035/43939 [35:34:51<34:49:43, 5.72s/it]
50%|█████ | 22036/43939 [35:34:57<34:45:29, 5.71s/it]
50%|█████ | 22037/43939 [35:35:03<34:42:59, 5.71s/it]
50%|█████ | 22038/43939 [35:35:09<35:09:59, 5.78s/it]
50%|█████ | 22039/43939 [35:35:14<34:58:51, 5.75s/it]
50%|█████ | 22040/43939 [35:35:20<34:53:58, 5.74s/it]
50%|█████ | 22041/43939 [35:35:26<34:50:19, 5.73s/it]
50%|█████ | 22042/43939 [35:35:31<34:42:52, 5.71s/it]
50%|█████ | 22043/43939 [35:35:37<34:43:41, 5.71s/it]
50%|█████ | 22044/43939 [35:35:43<34:41:55, 5.71s/it]
50%|█████ | 22045/43939 [35:35:48<34:40:55, 5.70s/it]
50%|█████ | 22046/43939 [35:35:54<34:41:03, 5.70s/it]
50%|█████ | 22047/43939 [35:36:00<34:38:17, 5.70s/it]
50%|█████ | 22048/43939 [35:36:06<34:38:34, 5.70s/it]
50%|█████ | 22049/43939 [35:36:11<34:39:36, 5.70s/it]
50%|█████ | 22050/43939 [35:36:17<34:47:25, 5.72s/it]
{'loss': 0.7675, 'grad_norm': 1.174663782119751, 'learning_rate': 2.91846486593655e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.7, 'epoch': 2.01}
50%|█████ | 22050/43939 [35:36:17<34:47:25, 5.72s/it]
50%|█████ | 22051/43939 [35:36:23<34:41:58, 5.71s/it]
50%|█████ | 22052/43939 [35:36:28<34:40:14, 5.70s/it]
50%|█████ | 22053/43939 [35:36:34<34:38:19, 5.70s/it]
50%|█████ | 22054/43939 [35:36:40<34:40:12, 5.70s/it]
50%|█████ | 22055/43939 [35:36:45<34:38:18, 5.70s/it]
50%|█████ | 22056/43939 [35:36:51<34:35:54, 5.69s/it]
50%|█████ | 22057/43939 [35:36:57<34:38:18, 5.70s/it]
50%|█████ | 22058/43939 [35:37:03<34:35:02, 5.69s/it]
50%|█████ | 22059/43939 [35:37:08<35:06:32, 5.78s/it]
50%|█████ | 22060/43939 [35:37:14<34:54:47, 5.74s/it]
50%|█████ | 22061/43939 [35:37:20<34:49:24, 5.73s/it]
50%|█████ | 22062/43939 [35:37:26<34:46:46, 5.72s/it]
50%|█████ | 22063/43939 [35:37:31<34:44:20, 5.72s/it]
50%|█████ | 22064/43939 [35:37:37<34:44:16, 5.72s/it]
50%|█████ | 22065/43939 [35:37:43<34:44:04, 5.72s/it]
50%|█████ | 22066/43939 [35:37:48<34:44:24, 5.72s/it]
50%|█████ | 22067/43939 [35:37:54<34:43:37, 5.72s/it]
50%|█████ | 22068/43939 [35:38:00<34:42:15, 5.71s/it]
50%|█████ | 22069/43939 [35:38:06<34:48:03, 5.73s/it]
50%|█████ | 22070/43939 [35:38:11<34:43:01, 5.71s/it]
50%|█████ | 22071/43939 [35:38:17<34:42:08, 5.71s/it]
50%|█████ | 22072/43939 [35:38:23<34:39:25, 5.71s/it]
50%|█████ | 22073/43939 [35:38:28<34:38:05, 5.70s/it]
50%|█████ | 22074/43939 [35:38:34<34:42:17, 5.71s/it]
50%|█████ | 22075/43939 [35:38:40<34:44:54, 5.72s/it]
50%|█████ | 22076/43939 [35:38:46<34:43:59, 5.72s/it]
50%|█████ | 22077/43939 [35:38:51<34:40:38, 5.71s/it]
50%|█████ | 22078/43939 [35:38:57<34:36:35, 5.70s/it]
50%|█████ | 22079/43939 [35:39:03<34:37:55, 5.70s/it]
50%|█████ | 22080/43939 [35:39:09<35:06:27, 5.78s/it]
50%|█████ | 22081/43939 [35:39:14<34:58:07, 5.76s/it]
50%|█████ | 22082/43939 [35:39:20<34:50:28, 5.74s/it]
50%|█████ | 22083/43939 [35:39:26<34:45:32, 5.73s/it]
50%|█████ | 22084/43939 [35:39:31<34:45:33, 5.73s/it]
50%|█████ | 22085/43939 [35:39:37<34:39:19, 5.71s/it]
50%|█████ | 22086/43939 [35:39:43<34:49:44, 5.74s/it]
50%|█████ | 22087/43939 [35:39:49<34:43:40, 5.72s/it]
50%|█████ | 22088/43939 [35:39:54<34:41:01, 5.71s/it]
50%|█████ | 22089/43939 [35:40:00<34:35:39, 5.70s/it]
50%|█████ | 22090/43939 [35:40:06<34:45:09, 5.73s/it]
50%|█████ | 22091/43939 [35:40:11<34:40:20, 5.71s/it]
50%|█████ | 22092/43939 [35:40:17<34:43:09, 5.72s/it]
50%|█████ | 22093/43939 [35:40:23<34:40:26, 5.71s/it]
50%|█████ | 22094/43939 [35:40:29<34:37:47, 5.71s/it]
50%|█████ | 22095/43939 [35:40:34<34:43:26, 5.72s/it]
50%|█████ | 22096/43939 [35:40:40<34:46:16, 5.73s/it]
50%|█████ | 22097/43939 [35:40:46<34:41:47, 5.72s/it]
50%|█████ | 22098/43939 [35:40:51<34:37:02, 5.71s/it]
50%|█████ | 22099/43939 [35:40:57<34:35:27, 5.70s/it]
50%|█████ | 22100/43939 [35:41:03<34:38:13, 5.71s/it]
{'loss': 0.7855, 'grad_norm': 1.2221143245697021, 'learning_rate': 2.908671506446566e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.95, 'epoch': 2.01}
50%|█████ | 22100/43939 [35:41:03<34:38:13, 5.71s/it]
50%|█████ | 22101/43939 [35:41:09<35:07:32, 5.79s/it]
50%|█████ | 22102/43939 [35:41:14<34:55:56, 5.76s/it]
50%|█████ | 22103/43939 [35:41:20<34:50:34, 5.74s/it]
50%|█████ | 22104/43939 [35:41:26<34:49:13, 5.74s/it]
50%|█████ | 22105/43939 [35:41:32<34:48:56, 5.74s/it]
50%|█████ | 22106/43939 [35:41:37<34:41:25, 5.72s/it]
50%|█████ | 22107/43939 [35:41:43<34:45:30, 5.73s/it]
50%|█████ | 22108/43939 [35:41:49<34:42:55, 5.72s/it]
50%|█████ | 22109/43939 [35:41:55<34:48:13, 5.74s/it]
50%|█████ | 22110/43939 [35:42:00<34:39:30, 5.72s/it]
50%|█████ | 22111/43939 [35:42:06<34:37:12, 5.71s/it]
50%|█████ | 22112/43939 [35:42:12<34:34:00, 5.70s/it]
50%|█████ | 22113/43939 [35:42:17<34:29:37, 5.69s/it]
50%|█████ | 22114/43939 [35:42:23<34:31:21, 5.69s/it]
50%|█████ | 22115/43939 [35:42:29<34:29:28, 5.69s/it]
50%|█████ | 22116/43939 [35:42:34<34:32:50, 5.70s/it]
50%|█████ | 22117/43939 [35:42:40<34:28:27, 5.69s/it]
50%|█████ | 22118/43939 [35:42:46<34:30:30, 5.69s/it]
50%|█████ | 22119/43939 [35:42:51<34:32:02, 5.70s/it]
50%|█████ | 22120/43939 [35:42:57<34:38:49, 5.72s/it]
50%|█████ | 22121/43939 [35:43:03<34:35:24, 5.71s/it]
50%|█████ | 22122/43939 [35:43:09<34:30:51, 5.70s/it]
50%|█████ | 22123/43939 [35:43:14<34:50:56, 5.75s/it]
50%|█████ | 22124/43939 [35:43:20<34:53:45, 5.76s/it]
50%|█████ | 22125/43939 [35:43:26<34:45:16, 5.74s/it]
50%|█████ | 22126/43939 [35:43:32<34:39:27, 5.72s/it]
50%|█████ | 22127/43939 [35:43:37<34:36:30, 5.71s/it]
50%|█████ | 22128/43939 [35:43:43<34:35:36, 5.71s/it]
50%|█████ | 22129/43939 [35:43:49<34:36:37, 5.71s/it]
50%|█████ | 22130/43939 [35:43:54<34:36:06, 5.71s/it]
50%|█████ | 22131/43939 [35:44:00<34:30:12, 5.70s/it]
50%|█████ | 22132/43939 [35:44:06<34:27:38, 5.69s/it]
50%|█████ | 22133/43939 [35:44:11<34:30:53, 5.70s/it]
50%|█████ | 22134/43939 [35:44:17<34:28:14, 5.69s/it]
50%|█████ | 22135/43939 [35:44:23<34:35:36, 5.71s/it]
50%|█████ | 22136/43939 [35:44:29<34:33:47, 5.71s/it]
50%|█████ | 22137/43939 [35:44:34<34:32:50, 5.70s/it]
50%|█████ | 22138/43939 [35:44:40<34:31:39, 5.70s/it]
50%|█████ | 22139/43939 [35:44:46<34:29:39, 5.70s/it]
50%|█████ | 22140/43939 [35:44:51<34:32:42, 5.70s/it]
50%|█████ | 22141/43939 [35:44:57<34:33:31, 5.71s/it]
50%|█████ | 22142/43939 [35:45:03<34:32:00, 5.70s/it]
50%|█████ | 22143/43939 [35:45:09<34:35:32, 5.71s/it]
50%|█████ | 22144/43939 [35:45:14<34:54:15, 5.77s/it]
50%|█████ | 22145/43939 [35:45:20<34:47:27, 5.75s/it]
50%|█████ | 22146/43939 [35:45:26<34:41:43, 5.73s/it]
50%|█████ | 22147/43939 [35:45:32<34:33:17, 5.71s/it]
50%|█████ | 22148/43939 [35:45:37<34:34:18, 5.71s/it]
50%|█████ | 22149/43939 [35:45:43<34:33:14, 5.71s/it]
50%|█████ | 22150/43939 [35:45:49<34:32:00, 5.71s/it]
{'loss': 0.8037, 'grad_norm': 1.3807263374328613, 'learning_rate': 2.8988716992032028e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.08, 'epoch': 2.02}
50%|█████ | 22150/43939 [35:45:49<34:32:00, 5.71s/it]
50%|█████ | 22151/43939 [35:45:54<34:34:20, 5.71s/it]
50%|█████ | 22152/43939 [35:46:00<34:32:25, 5.71s/it]
50%|█████ | 22153/43939 [35:46:06<34:31:17, 5.70s/it]
50%|█████ | 22154/43939 [35:46:11<34:33:46, 5.71s/it]
50%|█████ | 22155/43939 [35:46:17<34:37:16, 5.72s/it]
50%|█████ | 22156/43939 [35:46:23<34:37:15, 5.72s/it]
50%|█████ | 22157/43939 [35:46:29<34:35:37, 5.72s/it]
50%|█████ | 22158/43939 [35:46:34<34:39:08, 5.73s/it]
50%|█████ | 22159/43939 [35:46:40<34:33:34, 5.71s/it]
50%|█████ | 22160/43939 [35:46:46<34:28:29, 5.70s/it]
50%|█████ | 22161/43939 [35:46:51<34:28:09, 5.70s/it]
50%|█████ | 22162/43939 [35:46:57<34:26:04, 5.69s/it]
50%|█████ | 22163/43939 [35:47:03<34:26:16, 5.69s/it]
50%|█████ | 22164/43939 [35:47:09<34:27:10, 5.70s/it]
50%|█████ | 22165/43939 [35:47:14<34:57:11, 5.78s/it]
50%|█████ | 22166/43939 [35:47:20<34:52:54, 5.77s/it]
50%|█████ | 22167/43939 [35:47:26<34:42:39, 5.74s/it]
50%|█████ | 22168/43939 [35:47:32<34:36:09, 5.72s/it]
50%|█████ | 22169/43939 [35:47:37<34:35:23, 5.72s/it]
50%|█████ | 22170/43939 [35:47:43<34:40:31, 5.73s/it]
50%|█████ | 22171/43939 [35:47:49<34:36:51, 5.72s/it]
50%|█████ | 22172/43939 [35:47:54<34:33:45, 5.72s/it]
50%|█████ | 22173/43939 [35:48:00<34:32:07, 5.71s/it]
50%|█████ | 22174/43939 [35:48:06<34:34:35, 5.72s/it]
50%|█████ | 22175/43939 [35:48:12<34:36:08, 5.72s/it]
50%|█████ | 22176/43939 [35:48:17<34:33:08, 5.72s/it]
50%|█████ | 22177/43939 [35:48:23<34:34:03, 5.72s/it]
50%|█████ | 22178/43939 [35:48:29<34:29:28, 5.71s/it]
50%|█████ | 22179/43939 [35:48:34<34:27:56, 5.70s/it]
50%|█████ | 22180/43939 [35:48:40<34:27:05, 5.70s/it]
50%|█████ | 22181/43939 [35:48:46<34:32:12, 5.71s/it]
50%|█████ | 22182/43939 [35:48:52<34:32:15, 5.71s/it]
50%|█████ | 22183/43939 [35:48:57<34:31:58, 5.71s/it]
50%|█████ | 22184/43939 [35:49:03<34:34:36, 5.72s/it]
50%|█████ | 22185/43939 [35:49:09<34:29:14, 5.71s/it]
50%|█████ | 22186/43939 [35:49:14<34:31:34, 5.71s/it]
50%|█████ | 22187/43939 [35:49:20<34:51:58, 5.77s/it]
50%|█████ | 22188/43939 [35:49:26<34:41:58, 5.74s/it]
50%|█████ | 22189/43939 [35:49:32<34:40:31, 5.74s/it]
51%|█████ | 22190/43939 [35:49:37<34:34:50, 5.72s/it]
51%|█████ | 22191/43939 [35:49:43<34:27:19, 5.70s/it]
51%|█████ | 22192/43939 [35:49:49<34:25:50, 5.70s/it]
51%|█████ | 22193/43939 [35:49:54<34:26:10, 5.70s/it]
51%|█████ | 22194/43939 [35:50:00<34:27:03, 5.70s/it]
51%|█████ | 22195/43939 [35:50:06<34:26:59, 5.70s/it]
51%|█████ | 22196/43939 [35:50:12<34:23:31, 5.69s/it]
51%|█████ | 22197/43939 [35:50:17<34:26:30, 5.70s/it]
51%|█████ | 22198/43939 [35:50:23<34:20:48, 5.69s/it]
51%|█████ | 22199/43939 [35:50:29<34:23:15, 5.69s/it]
51%|█████ | 22200/43939 [35:50:34<34:26:40, 5.70s/it]
{'loss': 0.7763, 'grad_norm': 1.2672842741012573, 'learning_rate': 2.8890655988214483e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.79, 'epoch': 2.02}
51%|█████ | 22200/43939 [35:50:34<34:26:40, 5.70s/it]
51%|█████ | 22201/43939 [35:50:40<34:29:27, 5.71s/it]
51%|█████ | 22202/43939 [35:50:46<34:28:24, 5.71s/it]
51%|█████ | 22203/43939 [35:50:52<34:25:44, 5.70s/it]
51%|█████ | 22204/43939 [35:50:57<34:26:17, 5.70s/it]
51%|█████ | 22205/43939 [35:51:03<34:28:35, 5.71s/it]
51%|█████ | 22206/43939 [35:51:09<34:29:10, 5.71s/it]
51%|█████ | 22207/43939 [35:51:14<34:27:05, 5.71s/it]
51%|█████ | 22208/43939 [35:51:20<34:58:29, 5.79s/it]
51%|█████ | 22209/43939 [35:51:26<34:48:36, 5.77s/it]
51%|█████ | 22210/43939 [35:51:32<34:39:28, 5.74s/it]
51%|█████ | 22211/43939 [35:51:37<34:40:07, 5.74s/it]
51%|█████ | 22212/43939 [35:51:43<34:37:00, 5.74s/it]
51%|█████ | 22213/43939 [35:51:49<34:31:08, 5.72s/it]
51%|█████ | 22214/43939 [35:51:55<34:30:10, 5.72s/it]
51%|█████ | 22215/43939 [35:52:00<34:27:31, 5.71s/it]
51%|█████ | 22216/43939 [35:52:06<34:21:28, 5.69s/it]
51%|█████ | 22217/43939 [35:52:12<34:24:28, 5.70s/it]
51%|█████ | 22218/43939 [35:52:17<34:24:31, 5.70s/it]
51%|█████ | 22219/43939 [35:52:23<34:25:37, 5.71s/it]
51%|█████ | 22220/43939 [35:52:29<34:22:33, 5.70s/it]
51%|█████ | 22221/43939 [35:52:34<34:23:26, 5.70s/it]
51%|█████ | 22222/43939 [35:52:40<34:24:10, 5.70s/it]
51%|█████ | 22223/43939 [35:52:46<34:24:56, 5.71s/it]
51%|█████ | 22224/43939 [35:52:52<34:27:09, 5.71s/it]
51%|█████ | 22225/43939 [35:52:57<34:26:05, 5.71s/it]
51%|█████ | 22226/43939 [35:53:03<34:25:37, 5.71s/it]
51%|█████ | 22227/43939 [35:53:09<34:26:24, 5.71s/it]
51%|█████ | 22228/43939 [35:53:14<34:20:14, 5.69s/it]
51%|█████ | 22229/43939 [35:53:20<34:42:49, 5.76s/it]
51%|█████ | 22230/43939 [35:53:26<34:35:18, 5.74s/it]
51%|█████ | 22231/43939 [35:53:32<34:32:33, 5.73s/it]
51%|█████ | 22232/43939 [35:53:37<34:34:02, 5.73s/it]
51%|█████ | 22233/43939 [35:53:43<34:30:09, 5.72s/it]
51%|█████ | 22234/43939 [35:53:49<34:26:55, 5.71s/it]
51%|█████ | 22235/43939 [35:53:55<34:27:33, 5.72s/it]
51%|█████ | 22236/43939 [35:54:00<34:28:11, 5.72s/it]
51%|█████ | 22237/43939 [35:54:06<34:27:21, 5.72s/it]
51%|█████ | 22238/43939 [35:54:12<34:22:49, 5.70s/it]
51%|█████ | 22239/43939 [35:54:17<34:23:30, 5.71s/it]
51%|█████ | 22240/43939 [35:54:23<34:20:57, 5.70s/it]
51%|█████ | 22241/43939 [35:54:29<34:22:24, 5.70s/it]
51%|█████ | 22242/43939 [35:54:34<34:23:10, 5.71s/it]
51%|█████ | 22243/43939 [35:54:40<34:30:48, 5.73s/it]
51%|█████ | 22244/43939 [35:54:46<34:29:18, 5.72s/it]
51%|█████ | 22245/43939 [35:54:52<34:29:42, 5.72s/it]
51%|█████ | 22246/43939 [35:54:57<34:31:31, 5.73s/it]
51%|█████ | 22247/43939 [35:55:03<34:27:25, 5.72s/it]
51%|█████ | 22248/43939 [35:55:09<34:24:30, 5.71s/it]
51%|█████ | 22249/43939 [35:55:15<34:22:01, 5.70s/it]
51%|█████ | 22250/43939 [35:55:20<34:42:05, 5.76s/it]
{'loss': 0.781, 'grad_norm': 1.3480112552642822, 'learning_rate': 2.8792533600155813e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.58, 'epoch': 2.03}
51%|█████ | 22250/43939 [35:55:20<34:42:05, 5.76s/it]
51%|█████ | 22251/43939 [35:55:26<34:35:55, 5.74s/it]
51%|█████ | 22252/43939 [35:55:32<34:32:04, 5.73s/it]
51%|█████ | 22253/43939 [35:55:38<34:26:12, 5.72s/it]
51%|█████ | 22254/43939 [35:55:43<34:21:48, 5.70s/it]
51%|█████ | 22255/43939 [35:55:49<34:25:11, 5.71s/it]
51%|█████ | 22256/43939 [35:55:55<34:21:49, 5.71s/it]
51%|█████ | 22257/43939 [35:56:00<34:17:05, 5.69s/it]
51%|█████ | 22258/43939 [35:56:06<34:16:42, 5.69s/it]
51%|█████ | 22259/43939 [35:56:12<34:17:45, 5.69s/it]
51%|█████ | 22260/43939 [35:56:17<34:20:16, 5.70s/it]
51%|█████ | 22261/43939 [35:56:23<34:21:48, 5.71s/it]
51%|█████ | 22262/43939 [35:56:29<34:21:55, 5.71s/it]
51%|█████ | 22263/43939 [35:56:34<34:16:52, 5.69s/it]
51%|█████ | 22264/43939 [35:56:40<34:18:55, 5.70s/it]
51%|█████ | 22265/43939 [35:56:46<34:19:40, 5.70s/it]
51%|█████ | 22266/43939 [35:56:52<34:19:40, 5.70s/it]
51%|█████ | 22267/43939 [35:56:57<34:18:30, 5.70s/it]
51%|█████ | 22268/43939 [35:57:03<34:24:43, 5.72s/it]
51%|█████ | 22269/43939 [35:57:09<34:22:56, 5.71s/it]
51%|█████ | 22270/43939 [35:57:14<34:22:17, 5.71s/it]
51%|█████ | 22271/43939 [35:57:20<34:21:04, 5.71s/it]
51%|█████ | 22272/43939 [35:57:26<34:48:34, 5.78s/it]
51%|█████ | 22273/43939 [35:57:32<34:41:12, 5.76s/it]
51%|█████ | 22274/43939 [35:57:38<34:36:07, 5.75s/it]
51%|█████ | 22275/43939 [35:57:43<34:32:49, 5.74s/it]
51%|█████ | 22276/43939 [35:57:49<34:31:09, 5.74s/it]
51%|█████ | 22277/43939 [35:57:55<34:26:15, 5.72s/it]
51%|█████ | 22278/43939 [35:58:00<34:29:45, 5.73s/it]
51%|█████ | 22279/43939 [35:58:06<34:28:33, 5.73s/it]
51%|█████ | 22280/43939 [35:58:12<34:21:49, 5.71s/it]
51%|█████ | 22281/43939 [35:58:17<34:17:36, 5.70s/it]
51%|█████ | 22282/43939 [35:58:23<34:16:37, 5.70s/it]
51%|█████ | 22283/43939 [35:58:29<34:19:39, 5.71s/it]
51%|█████ | 22284/43939 [35:58:35<34:22:41, 5.72s/it]
51%|█████ | 22285/43939 [35:58:40<34:18:46, 5.70s/it]
51%|█████ | 22286/43939 [35:58:46<34:12:48, 5.69s/it]
51%|█████ | 22287/43939 [35:58:52<34:15:17, 5.70s/it]
51%|█████ | 22288/43939 [35:58:57<34:16:28, 5.70s/it]
51%|█████ | 22289/43939 [35:59:03<34:22:44, 5.72s/it]
51%|█████ | 22290/43939 [35:59:09<34:19:24, 5.71s/it]
51%|█████ | 22291/43939 [35:59:15<34:20:21, 5.71s/it]
51%|█████ | 22292/43939 [35:59:20<34:22:47, 5.72s/it]
51%|█████ | 22293/43939 [35:59:26<34:42:30, 5.77s/it]
51%|█████ | 22294/43939 [35:59:32<34:36:25, 5.76s/it]
51%|█████ | 22295/43939 [35:59:38<34:29:53, 5.74s/it]
51%|█████ | 22296/43939 [35:59:43<34:30:28, 5.74s/it]
51%|█████ | 22297/43939 [35:59:49<34:29:44, 5.74s/it]
51%|█████ | 22298/43939 [35:59:55<34:24:39, 5.72s/it]
51%|█████ | 22299/43939 [36:00:00<34:21:12, 5.72s/it]
51%|█████ | 22300/43939 [36:00:06<34:23:37, 5.72s/it]
{'loss': 0.7686, 'grad_norm': 1.296072244644165, 'learning_rate': 2.869435137596727e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.56, 'epoch': 2.03}
51%|█████ | 22300/43939 [36:00:06<34:23:37, 5.72s/it]
51%|█████ | 22301/43939 [36:00:12<34:19:57, 5.71s/it]
51%|█████ | 22302/43939 [36:00:18<34:25:04, 5.73s/it]
51%|█████ | 22303/43939 [36:00:23<34:31:03, 5.74s/it]
51%|█████ | 22304/43939 [36:00:29<34:29:23, 5.74s/it]
51%|█████ | 22305/43939 [36:00:35<34:20:12, 5.71s/it]
51%|█████ | 22306/43939 [36:00:41<34:17:47, 5.71s/it]
51%|█████ | 22307/43939 [36:00:46<34:13:45, 5.70s/it]
51%|█████ | 22308/43939 [36:00:52<34:12:37, 5.69s/it]
51%|█████ | 22309/43939 [36:00:58<34:14:35, 5.70s/it]
51%|█████ | 22310/43939 [36:01:03<34:10:24, 5.69s/it]
51%|█████ | 22311/43939 [36:01:09<34:08:47, 5.68s/it]
51%|█████ | 22312/43939 [36:01:15<34:08:31, 5.68s/it]
51%|█████ | 22313/43939 [36:01:20<34:09:22, 5.69s/it]
51%|█████ | 22314/43939 [36:01:26<34:37:11, 5.76s/it]
51%|█████ | 22315/43939 [36:01:32<34:27:57, 5.74s/it]
51%|█████ | 22316/43939 [36:01:38<34:26:48, 5.74s/it]
51%|█████ | 22317/43939 [36:01:43<34:20:48, 5.72s/it]
51%|█████ | 22318/43939 [36:01:49<34:23:25, 5.73s/it]
51%|█████ | 22319/43939 [36:01:55<34:24:13, 5.73s/it]
51%|█████ | 22320/43939 [36:02:01<34:24:30, 5.73s/it]
51%|█████ | 22321/43939 [36:02:06<34:19:03, 5.71s/it]
51%|█████ | 22322/43939 [36:02:12<34:18:43, 5.71s/it]
51%|█████ | 22323/43939 [36:02:18<34:20:06, 5.72s/it]
51%|█████ | 22324/43939 [36:02:23<34:15:18, 5.71s/it]
51%|█████ | 22325/43939 [36:02:29<34:11:49, 5.70s/it]
51%|█████ | 22326/43939 [36:02:35<34:14:47, 5.70s/it]
51%|█████ | 22327/43939 [36:02:40<34:13:03, 5.70s/it]
51%|█████ | 22328/43939 [36:02:46<34:15:11, 5.71s/it]
51%|█████ | 22329/43939 [36:02:52<34:17:59, 5.71s/it]
51%|█████ | 22330/43939 [36:02:58<34:16:54, 5.71s/it]
51%|█████ | 22331/43939 [36:03:03<34:16:13, 5.71s/it]
51%|█████ | 22332/43939 [36:03:09<34:12:58, 5.70s/it]
51%|█████ | 22333/43939 [36:03:15<34:07:05, 5.68s/it]
51%|█████ | 22334/43939 [36:03:20<34:10:11, 5.69s/it]
51%|█████ | 22335/43939 [36:03:26<34:08:17, 5.69s/it]
51%|█████ | 22336/43939 [36:03:32<34:32:40, 5.76s/it]
51%|█████ | 22337/43939 [36:03:38<34:26:43, 5.74s/it]
51%|█████ | 22338/43939 [36:03:43<34:24:35, 5.73s/it]
51%|█████ | 22339/43939 [36:03:49<34:21:44, 5.73s/it]
51%|█████ | 22340/43939 [36:03:55<34:27:24, 5.74s/it]
51%|█████ | 22341/43939 [36:04:01<34:20:52, 5.73s/it]
51%|█████ | 22342/43939 [36:04:06<34:17:31, 5.72s/it]
51%|█████ | 22343/43939 [36:04:12<34:10:50, 5.70s/it]
51%|█████ | 22344/43939 [36:04:18<34:11:42, 5.70s/it]
51%|█████ | 22345/43939 [36:04:23<34:12:49, 5.70s/it]
51%|█████ | 22346/43939 [36:04:29<34:12:27, 5.70s/it]
51%|█████ | 22347/43939 [36:04:35<34:08:36, 5.69s/it]
51%|█████ | 22348/43939 [36:04:40<34:07:08, 5.69s/it]
51%|█████ | 22349/43939 [36:04:46<34:08:12, 5.69s/it]
51%|█████ | 22350/43939 [36:04:52<34:07:27, 5.69s/it]
{'loss': 0.7609, 'grad_norm': 1.4430197477340698, 'learning_rate': 2.8596110864704168e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.36, 'epoch': 2.03}
51%|█████ | 22350/43939 [36:04:52<34:07:27, 5.69s/it]
51%|█████ | 22351/43939 [36:04:57<34:05:52, 5.69s/it]
51%|█████ | 22352/43939 [36:05:03<34:05:57, 5.69s/it]
51%|█████ | 22353/43939 [36:05:09<34:20:51, 5.73s/it]
51%|█████ | 22354/43939 [36:05:15<34:17:18, 5.72s/it]
51%|█████ | 22355/43939 [36:05:20<34:12:58, 5.71s/it]
51%|█████ | 22356/43939 [36:05:26<34:11:17, 5.70s/it]
51%|█████ | 22357/43939 [36:05:32<34:38:17, 5.78s/it]
51%|█████ | 22358/43939 [36:05:38<34:26:29, 5.75s/it]
51%|█████ | 22359/43939 [36:05:43<34:21:25, 5.73s/it]
51%|█████ | 22360/43939 [36:05:49<34:17:51, 5.72s/it]
51%|█████ | 22361/43939 [36:05:55<34:21:44, 5.73s/it]
51%|█████ | 22362/43939 [36:06:00<34:17:29, 5.72s/it]
51%|█████ | 22363/43939 [36:06:06<34:13:41, 5.71s/it]
51%|█████ | 22364/43939 [36:06:12<34:14:52, 5.71s/it]
51%|█████ | 22365/43939 [36:06:18<34:10:24, 5.70s/it]
51%|█████ | 22366/43939 [36:06:23<34:09:31, 5.70s/it]
51%|█████ | 22367/43939 [36:06:29<34:16:46, 5.72s/it]
51%|█████ | 22368/43939 [36:06:35<34:14:53, 5.72s/it]
51%|█████ | 22369/43939 [36:06:40<34:12:34, 5.71s/it]
51%|█████ | 22370/43939 [36:06:46<34:19:23, 5.73s/it]
51%|█████ | 22371/43939 [36:06:52<34:16:21, 5.72s/it]
51%|█████ | 22372/43939 [36:06:58<34:24:39, 5.74s/it]
51%|█████ | 22373/43939 [36:07:03<34:20:40, 5.73s/it]
51%|█████ | 22374/43939 [36:07:09<34:20:59, 5.73s/it]
51%|█████ | 22375/43939 [36:07:15<34:19:46, 5.73s/it]
51%|█████ | 22376/43939 [36:07:21<34:19:31, 5.73s/it]
51%|█████ | 22377/43939 [36:07:26<34:16:22, 5.72s/it]
51%|█████ | 22378/43939 [36:07:32<34:41:18, 5.79s/it]
51%|█████ | 22379/43939 [36:07:38<34:31:40, 5.77s/it]
51%|█████ | 22380/43939 [36:07:44<34:25:07, 5.75s/it]
51%|█████ | 22381/43939 [36:07:49<34:18:08, 5.73s/it]
51%|█████ | 22382/43939 [36:07:55<34:12:14, 5.71s/it]
51%|█████ | 22383/43939 [36:08:01<34:10:39, 5.71s/it]
51%|█████ | 22384/43939 [36:08:06<34:13:58, 5.72s/it]
51%|█████ | 22385/43939 [36:08:12<34:10:06, 5.71s/it]
51%|█████ | 22386/43939 [36:08:18<34:15:38, 5.72s/it]
51%|█████ | 22387/43939 [36:08:24<34:15:03, 5.72s/it]
51%|█████ | 22388/43939 [36:08:29<34:13:32, 5.72s/it]
51%|█████ | 22389/43939 [36:08:35<34:13:59, 5.72s/it]
51%|█████ | 22390/43939 [36:08:41<34:11:52, 5.71s/it]
51%|█████ | 22391/43939 [36:08:46<34:09:31, 5.71s/it]
51%|█████ | 22392/43939 [36:08:52<34:12:58, 5.72s/it]
51%|█████ | 22393/43939 [36:08:58<34:16:15, 5.73s/it]
51%|█████ | 22394/43939 [36:09:04<34:11:16, 5.71s/it]
51%|█████ | 22395/43939 [36:09:09<34:10:23, 5.71s/it]
51%|█████ | 22396/43939 [36:09:15<34:08:59, 5.71s/it]
51%|█████ | 22397/43939 [36:09:21<34:10:56, 5.71s/it]
51%|█████ | 22398/43939 [36:09:26<34:07:53, 5.70s/it]
51%|█████ | 22399/43939 [36:09:32<34:28:11, 5.76s/it]
51%|█████ | 22400/43939 [36:09:38<34:19:38, 5.74s/it]
{'loss': 0.7875, 'grad_norm': 1.331939458847046, 'learning_rate': 2.849781361634143e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 527.72, 'epoch': 2.04}
51%|█████ | 22400/43939 [36:09:38<34:19:38, 5.74s/it]
51%|█████ | 22401/43939 [36:09:44<34:17:08, 5.73s/it]
51%|█████ | 22402/43939 [36:09:49<34:17:31, 5.73s/it]
51%|█████ | 22403/43939 [36:09:55<34:12:54, 5.72s/it]
51%|█████ | 22404/43939 [36:10:01<34:06:27, 5.70s/it]
51%|█████ | 22405/43939 [36:10:06<34:03:40, 5.69s/it]
51%|█████ | 22406/43939 [36:10:12<34:03:02, 5.69s/it]
51%|█████ | 22407/43939 [36:10:18<34:06:46, 5.70s/it]
51%|█████ | 22408/43939 [36:10:24<34:07:13, 5.70s/it]
51%|█████ | 22409/43939 [36:10:29<34:13:52, 5.72s/it]
51%|█████ | 22410/43939 [36:10:35<34:12:34, 5.72s/it]
51%|█████ | 22411/43939 [36:10:41<34:15:22, 5.73s/it]
51%|█████ | 22412/43939 [36:10:47<34:12:19, 5.72s/it]
51%|█████ | 22413/43939 [36:10:52<34:07:59, 5.71s/it]
51%|█████ | 22414/43939 [36:10:58<34:13:13, 5.72s/it]
51%|█████ | 22415/43939 [36:11:04<34:15:22, 5.73s/it]
51%|█████ | 22416/43939 [36:11:09<34:16:32, 5.73s/it]
51%|█████ | 22417/43939 [36:11:15<34:09:44, 5.71s/it]
51%|█████ | 22418/43939 [36:11:21<34:12:11, 5.72s/it]
51%|█████ | 22419/43939 [36:11:27<34:08:41, 5.71s/it]
51%|█████ | 22420/43939 [36:11:32<34:08:07, 5.71s/it]
51%|█████ | 22421/43939 [36:11:38<34:29:14, 5.77s/it]
51%|█████ | 22422/43939 [36:11:44<34:18:37, 5.74s/it]
51%|█████ | 22423/43939 [36:11:49<34:08:37, 5.71s/it]
51%|█████ | 22424/43939 [36:11:55<34:05:34, 5.70s/it]
51%|█████ | 22425/43939 [36:12:01<34:06:49, 5.71s/it]
51%|█████ | 22426/43939 [36:12:07<34:05:12, 5.70s/it]
51%|█████ | 22427/43939 [36:12:12<34:03:49, 5.70s/it]
51%|█████ | 22428/43939 [36:12:18<34:02:18, 5.70s/it]
51%|█████ | 22429/43939 [36:12:24<33:58:29, 5.69s/it]
51%|█████ | 22430/43939 [36:12:29<33:58:28, 5.69s/it]
51%|█████ | 22431/43939 [36:12:35<33:59:02, 5.69s/it]
51%|█████ | 22432/43939 [36:12:41<33:58:36, 5.69s/it]
51%|█████ | 22433/43939 [36:12:46<34:00:29, 5.69s/it]
51%|█████ | 22434/43939 [36:12:52<33:59:01, 5.69s/it]
51%|█████ | 22435/43939 [36:12:58<33:59:39, 5.69s/it]
51%|█████ | 22436/43939 [36:13:03<34:01:53, 5.70s/it]
51%|█████ | 22437/43939 [36:13:09<33:58:23, 5.69s/it]
51%|█████ | 22438/43939 [36:13:15<33:56:53, 5.68s/it]
51%|█████ | 22439/43939 [36:13:21<34:00:29, 5.69s/it]
51%|█████ | 22440/43939 [36:13:26<34:03:48, 5.70s/it]
51%|█████ | 22441/43939 [36:13:32<33:59:26, 5.69s/it]
51%|█████ | 22442/43939 [36:13:38<34:02:28, 5.70s/it]
51%|█████ | 22443/43939 [36:13:44<34:27:32, 5.77s/it]
51%|█████ | 22444/43939 [36:13:49<34:19:28, 5.75s/it]
51%|█████ | 22445/43939 [36:13:55<34:13:53, 5.73s/it]
51%|█████ | 22446/43939 [36:14:01<34:08:50, 5.72s/it]
51%|█████ | 22447/43939 [36:14:06<34:03:45, 5.71s/it]
51%|█████ | 22448/43939 [36:14:12<34:06:53, 5.71s/it]
51%|█████ | 22449/43939 [36:14:18<34:07:39, 5.72s/it]
51%|█████ | 22450/43939 [36:14:23<34:06:42, 5.71s/it]
{'loss': 0.7658, 'grad_norm': 1.3159470558166504, 'learning_rate': 2.839946118174914e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.05, 'epoch': 2.04}
51%|█████ | 22450/43939 [36:14:23<34:06:42, 5.71s/it]
51%|█████ | 22451/43939 [36:14:29<34:05:55, 5.71s/it]
51%|█████ | 22452/43939 [36:14:35<34:07:15, 5.72s/it]
51%|█████ | 22453/43939 [36:14:41<34:01:37, 5.70s/it]
51%|█████ | 22454/43939 [36:14:46<34:05:42, 5.71s/it]
51%|█████ | 22455/43939 [36:14:52<34:05:07, 5.71s/it]
51%|█████ | 22456/43939 [36:14:58<34:06:24, 5.72s/it]
51%|█████ | 22457/43939 [36:15:03<34:05:55, 5.71s/it]
51%|█████ | 22458/43939 [36:15:09<34:11:25, 5.73s/it]
51%|█████ | 22459/43939 [36:15:15<34:05:48, 5.71s/it]
51%|█████ | 22460/43939 [36:15:21<34:02:28, 5.71s/it]
51%|█████ | 22461/43939 [36:15:26<34:03:20, 5.71s/it]
51%|█████ | 22462/43939 [36:15:32<34:06:23, 5.72s/it]
51%|█████ | 22463/43939 [36:15:38<34:28:23, 5.78s/it]
51%|█████ | 22464/43939 [36:15:44<34:20:49, 5.76s/it]
51%|█████ | 22465/43939 [36:15:49<34:14:08, 5.74s/it]
51%|█████ | 22466/43939 [36:15:55<34:07:14, 5.72s/it]
51%|█████ | 22467/43939 [36:16:01<34:02:26, 5.71s/it]
51%|█████ | 22468/43939 [36:16:06<33:58:47, 5.70s/it]
51%|█████ | 22469/43939 [36:16:12<33:59:46, 5.70s/it]
51%|█████ | 22470/43939 [36:16:18<34:03:19, 5.71s/it]
51%|█████ | 22471/43939 [36:16:24<33:59:07, 5.70s/it]
51%|█████ | 22472/43939 [36:16:29<34:05:55, 5.72s/it]
51%|█████ | 22473/43939 [36:16:35<34:04:01, 5.71s/it]
51%|█████ | 22474/43939 [36:16:41<34:12:10, 5.74s/it]
51%|█████ | 22475/43939 [36:16:46<34:06:06, 5.72s/it]
51%|█████ | 22476/43939 [36:16:52<34:00:33, 5.70s/it]
51%|█████ | 22477/43939 [36:16:58<33:59:30, 5.70s/it]
51%|█████ | 22478/43939 [36:17:04<33:57:46, 5.70s/it]
51%|█████ | 22479/43939 [36:17:09<34:01:07, 5.71s/it]
51%|█████ | 22480/43939 [36:17:15<33:58:47, 5.70s/it]
51%|█████ | 22481/43939 [36:17:21<34:03:38, 5.71s/it]
51%|█████ | 22482/43939 [36:17:26<34:01:48, 5.71s/it]
51%|█████ | 22483/43939 [36:17:32<34:05:00, 5.72s/it]
51%|█████ | 22484/43939 [36:17:38<34:28:47, 5.79s/it]
51%|█████ | 22485/43939 [36:17:44<34:22:35, 5.77s/it]
51%|█████ | 22486/43939 [36:17:49<34:13:19, 5.74s/it]
51%|█████ | 22487/43939 [36:17:55<34:11:22, 5.74s/it]
51%|█████ | 22488/43939 [36:18:01<34:09:07, 5.73s/it]
51%|█████ | 22489/43939 [36:18:07<34:03:48, 5.72s/it]
51%|█████ | 22490/43939 [36:18:12<34:01:29, 5.71s/it]
51%|█████ | 22491/43939 [36:18:18<34:00:33, 5.71s/it]
51%|█████ | 22492/43939 [36:18:24<34:03:30, 5.72s/it]
51%|█████ | 22493/43939 [36:18:29<34:05:21, 5.72s/it]
51%|█████ | 22494/43939 [36:18:35<34:07:08, 5.73s/it]
51%|█████ | 22495/43939 [36:18:41<34:04:57, 5.72s/it]
51%|█████ | 22496/43939 [36:18:47<34:05:57, 5.72s/it]
51%|█████ | 22497/43939 [36:18:52<34:01:45, 5.71s/it]
51%|█████ | 22498/43939 [36:18:58<33:58:13, 5.70s/it]
51%|█████ | 22499/43939 [36:19:04<33:58:37, 5.71s/it]
51%|█████ | 22500/43939 [36:19:09<34:01:36, 5.71s/it]
{'loss': 0.7939, 'grad_norm': 1.3375674486160278, 'learning_rate': 2.8301055112668083e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 521.71, 'epoch': 2.05}
51%|█████ | 22500/43939 [36:19:09<34:01:36, 5.71s/it]
51%|█████ | 22501/43939 [36:19:15<34:03:46, 5.72s/it]
51%|█████ | 22502/43939 [36:19:21<34:00:32, 5.71s/it]
51%|█████ | 22503/43939 [36:19:27<34:01:09, 5.71s/it]
51%|█████ | 22504/43939 [36:19:32<34:03:41, 5.72s/it]
51%|█████ | 22505/43939 [36:19:38<34:21:20, 5.77s/it]
51%|█████ | 22506/43939 [36:19:44<34:16:28, 5.76s/it]
51%|█████ | 22507/43939 [36:19:50<34:07:05, 5.73s/it]
51%|█████ | 22508/43939 [36:19:55<34:05:23, 5.73s/it]
51%|█████ | 22509/43939 [36:20:01<34:04:31, 5.72s/it]
51%|█████ | 22510/43939 [36:20:07<33:58:15, 5.71s/it]
51%|█████ | 22511/43939 [36:20:12<33:55:06, 5.70s/it]
51%|█████ | 22512/43939 [36:20:18<33:54:54, 5.70s/it]
51%|█████ | 22513/43939 [36:20:24<34:02:24, 5.72s/it]
51%|█████ | 22514/43939 [36:20:30<34:03:35, 5.72s/it]
51%|█████ | 22515/43939 [36:20:35<33:57:37, 5.71s/it]
51%|█████ | 22516/43939 [36:20:41<33:59:39, 5.71s/it]
51%|█████ | 22517/43939 [36:20:47<33:58:12, 5.71s/it]
51%|█████ | 22518/43939 [36:20:52<34:02:13, 5.72s/it]
51%|█████▏ | 22519/43939 [36:20:58<34:01:25, 5.72s/it]
51%|█████▏ | 22520/43939 [36:21:04<33:57:30, 5.71s/it]
51%|█████▏ | 22521/43939 [36:21:10<33:57:09, 5.71s/it]
51%|█████▏ | 22522/43939 [36:21:15<33:58:41, 5.71s/it]
51%|█████▏ | 22523/43939 [36:21:21<33:55:45, 5.70s/it]
51%|█████▏ | 22524/43939 [36:21:27<33:58:52, 5.71s/it]
51%|█████▏ | 22525/43939 [36:21:32<33:53:55, 5.70s/it]
51%|█████▏ | 22526/43939 [36:21:38<34:19:38, 5.77s/it]
51%|█████▏ | 22527/43939 [36:21:44<34:11:37, 5.75s/it]
51%|█████▏ | 22528/43939 [36:21:50<34:04:32, 5.73s/it]
51%|█████▏ | 22529/43939 [36:21:55<34:06:00, 5.73s/it]
51%|█████▏ | 22530/43939 [36:22:01<34:03:34, 5.73s/it]
51%|█████▏ | 22531/43939 [36:22:07<34:06:51, 5.74s/it]
51%|█████▏ | 22532/43939 [36:22:13<34:05:04, 5.73s/it]
51%|█████▏ | 22533/43939 [36:22:18<34:00:29, 5.72s/it]
51%|█████▏ | 22534/43939 [36:22:24<34:01:34, 5.72s/it]
51%|█████▏ | 22535/43939 [36:22:30<34:02:44, 5.73s/it]
51%|█████▏ | 22536/43939 [36:22:36<34:04:59, 5.73s/it]
51%|█████▏ | 22537/43939 [36:22:41<34:02:18, 5.73s/it]
51%|█████▏ | 22538/43939 [36:22:47<33:59:40, 5.72s/it]
51%|█████▏ | 22539/43939 [36:22:53<33:56:49, 5.71s/it]
51%|█████▏ | 22540/43939 [36:22:58<34:01:50, 5.73s/it]
51%|█████▏ | 22541/43939 [36:23:04<33:56:03, 5.71s/it]
51%|█████▏ | 22542/43939 [36:23:10<33:55:26, 5.71s/it]
51%|█████▏ | 22543/43939 [36:23:16<34:01:46, 5.73s/it]
51%|█████▏ | 22544/43939 [36:23:21<33:57:13, 5.71s/it]
51%|█████▏ | 22545/43939 [36:23:27<33:55:26, 5.71s/it]
51%|█████▏ | 22546/43939 [36:23:33<33:54:11, 5.71s/it]
51%|█████▏ | 22547/43939 [36:23:39<34:16:46, 5.77s/it]
51%|█████▏ | 22548/43939 [36:23:44<34:09:17, 5.75s/it]
51%|█████▏ | 22549/43939 [36:23:50<34:04:15, 5.73s/it]
51%|█████▏ | 22550/43939 [36:23:56<33:56:24, 5.71s/it]
{'loss': 0.7653, 'grad_norm': 1.4380762577056885, 'learning_rate': 2.8202596961685256e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.08, 'epoch': 2.05}
51%|█████▏ | 22550/43939 [36:23:56<33:56:24, 5.71s/it]
51%|█████▏ | 22551/43939 [36:24:01<33:52:21, 5.70s/it]
51%|█████▏ | 22552/43939 [36:24:07<33:50:54, 5.70s/it]
51%|█████▏ | 22553/43939 [36:24:13<33:53:28, 5.71s/it]
51%|█████▏ | 22554/43939 [36:24:18<33:55:15, 5.71s/it]
51%|█████▏ | 22555/43939 [36:24:24<33:52:53, 5.70s/it]
51%|█████▏ | 22556/43939 [36:24:30<33:55:11, 5.71s/it]
51%|█████▏ | 22557/43939 [36:24:36<33:58:46, 5.72s/it]
51%|█████▏ | 22558/43939 [36:24:41<33:58:02, 5.72s/it]
51%|█████▏ | 22559/43939 [36:24:47<33:58:23, 5.72s/it]
51%|█████▏ | 22560/43939 [36:24:53<33:54:51, 5.71s/it]
51%|█████▏ | 22561/43939 [36:24:58<33:48:48, 5.69s/it]
51%|█████▏ | 22562/43939 [36:25:04<33:50:06, 5.70s/it]
51%|█████▏ | 22563/43939 [36:25:10<33:47:31, 5.69s/it]
51%|█████▏ | 22564/43939 [36:25:15<33:48:31, 5.69s/it]
51%|█████▏ | 22565/43939 [36:25:21<33:48:10, 5.69s/it]
51%|█████▏ | 22566/43939 [36:25:27<33:50:04, 5.70s/it]
51%|█████▏ | 22567/43939 [36:25:32<33:46:06, 5.69s/it]
51%|█████▏ | 22568/43939 [36:25:38<33:50:39, 5.70s/it]
51%|█████▏ | 22569/43939 [36:25:44<34:15:09, 5.77s/it]
51%|█████▏ | 22570/43939 [36:25:50<34:03:01, 5.74s/it]
51%|█████▏ | 22571/43939 [36:25:55<33:57:15, 5.72s/it]
51%|█████▏ | 22572/43939 [36:26:01<33:53:44, 5.71s/it]
51%|█████▏ | 22573/43939 [36:26:07<33:50:54, 5.70s/it]
51%|█████▏ | 22574/43939 [36:26:13<33:52:58, 5.71s/it]
51%|█████▏ | 22575/43939 [36:26:18<33:55:32, 5.72s/it]
51%|█████▏ | 22576/43939 [36:26:24<33:54:25, 5.71s/it]
51%|█████▏ | 22577/43939 [36:26:30<33:53:56, 5.71s/it]
51%|█████▏ | 22578/43939 [36:26:35<33:56:55, 5.72s/it]
51%|█████▏ | 22579/43939 [36:26:41<33:51:15, 5.71s/it]
51%|█████▏ | 22580/43939 [36:26:47<33:52:10, 5.71s/it]
51%|█████▏ | 22581/43939 [36:26:53<33:58:46, 5.73s/it]
51%|█████▏ | 22582/43939 [36:26:58<33:55:51, 5.72s/it]
51%|█████▏ | 22583/43939 [36:27:04<33:52:49, 5.71s/it]
51%|█████▏ | 22584/43939 [36:27:10<33:50:29, 5.70s/it]
51%|█████▏ | 22585/43939 [36:27:15<33:48:52, 5.70s/it]
51%|█████▏ | 22586/43939 [36:27:21<33:54:11, 5.72s/it]
51%|█████▏ | 22587/43939 [36:27:27<33:53:38, 5.71s/it]
51%|█████▏ | 22588/43939 [36:27:33<33:54:20, 5.72s/it]
51%|█████▏ | 22589/43939 [36:27:38<33:53:00, 5.71s/it]
51%|█████▏ | 22590/43939 [36:27:44<33:51:49, 5.71s/it]
51%|█████▏ | 22591/43939 [36:27:50<34:17:51, 5.78s/it]
51%|█████▏ | 22592/43939 [36:27:56<34:12:56, 5.77s/it]
51%|█████▏ | 22593/43939 [36:28:01<34:02:43, 5.74s/it]
51%|█████▏ | 22594/43939 [36:28:07<33:59:34, 5.73s/it]
51%|█████▏ | 22595/43939 [36:28:13<33:57:56, 5.73s/it]
51%|█████▏ | 22596/43939 [36:28:19<34:02:19, 5.74s/it]
51%|█████▏ | 22597/43939 [36:28:24<34:04:04, 5.75s/it]
51%|█████▏ | 22598/43939 [36:28:30<33:58:54, 5.73s/it]
51%|█████▏ | 22599/43939 [36:28:36<33:54:35, 5.72s/it]
51%|█████▏ | 22600/43939 [36:28:41<33:51:32, 5.71s/it]
{'loss': 0.7688, 'grad_norm': 1.2875185012817383, 'learning_rate': 2.8104088282209357e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.39, 'epoch': 2.06}
51%|█████▏ | 22600/43939 [36:28:41<33:51:32, 5.71s/it]
51%|█████▏ | 22601/43939 [36:28:47<33:56:51, 5.73s/it]
51%|█████▏ | 22602/43939 [36:28:53<33:52:03, 5.71s/it]
51%|█████▏ | 22603/43939 [36:28:59<33:50:31, 5.71s/it]
51%|█████▏ | 22604/43939 [36:29:04<33:52:02, 5.71s/it]
51%|█████▏ | 22605/43939 [36:29:10<33:48:23, 5.70s/it]
51%|█████▏ | 22606/43939 [36:29:16<33:45:52, 5.70s/it]
51%|█████▏ | 22607/43939 [36:29:21<33:44:03, 5.69s/it]
51%|█████▏ | 22608/43939 [36:29:27<33:41:12, 5.69s/it]
51%|█████▏ | 22609/43939 [36:29:33<33:42:46, 5.69s/it]
51%|█████▏ | 22610/43939 [36:29:38<33:44:17, 5.69s/it]
51%|█████▏ | 22611/43939 [36:29:44<34:06:47, 5.76s/it]
51%|█████▏ | 22612/43939 [36:29:50<34:00:23, 5.74s/it]
51%|█████▏ | 22613/43939 [36:29:56<33:59:45, 5.74s/it]
51%|█████▏ | 22614/43939 [36:30:01<34:00:34, 5.74s/it]
51%|█████▏ | 22615/43939 [36:30:07<33:57:30, 5.73s/it]
51%|█████▏ | 22616/43939 [36:30:13<33:52:07, 5.72s/it]
51%|█████▏ | 22617/43939 [36:30:19<33:56:53, 5.73s/it]
51%|█████▏ | 22618/43939 [36:30:24<33:52:08, 5.72s/it]
51%|█████▏ | 22619/43939 [36:30:30<33:47:26, 5.71s/it]
51%|█████▏ | 22620/43939 [36:30:36<33:41:42, 5.69s/it]
51%|█████▏ | 22621/43939 [36:30:41<33:37:42, 5.68s/it]
51%|█████▏ | 22622/43939 [36:30:47<33:41:46, 5.69s/it]
51%|█████▏ | 22623/43939 [36:30:53<33:44:49, 5.70s/it]
51%|█████▏ | 22624/43939 [36:30:58<33:42:08, 5.69s/it]
51%|█████▏ | 22625/43939 [36:31:04<33:41:49, 5.69s/it]
51%|█████▏ | 22626/43939 [36:31:10<33:38:56, 5.68s/it]
51%|█████▏ | 22627/43939 [36:31:15<33:42:08, 5.69s/it]
51%|█████▏ | 22628/43939 [36:31:21<33:44:19, 5.70s/it]
52%|█████▏ | 22629/43939 [36:31:27<33:42:47, 5.70s/it]
52%|█████▏ | 22630/43939 [36:31:33<33:40:44, 5.69s/it]
52%|█████▏ | 22631/43939 [36:31:38<33:43:18, 5.70s/it]
52%|█████▏ | 22632/43939 [36:31:44<34:12:40, 5.78s/it]
52%|█████▏ | 22633/43939 [36:31:50<34:12:15, 5.78s/it]
52%|█████▏ | 22634/43939 [36:31:56<34:04:07, 5.76s/it]
52%|█████▏ | 22635/43939 [36:32:01<33:59:51, 5.75s/it]
52%|█████▏ | 22636/43939 [36:32:07<33:54:11, 5.73s/it]
52%|█████▏ | 22637/43939 [36:32:13<33:53:13, 5.73s/it]
52%|█████▏ | 22638/43939 [36:32:19<33:51:17, 5.72s/it]
52%|█████▏ | 22639/43939 [36:32:24<33:52:42, 5.73s/it]
52%|█████▏ | 22640/43939 [36:32:30<33:48:39, 5.71s/it]
52%|█████▏ | 22641/43939 [36:32:36<33:44:53, 5.70s/it]
52%|█████▏ | 22642/43939 [36:32:41<33:47:45, 5.71s/it]
52%|█████▏ | 22643/43939 [36:32:47<33:43:26, 5.70s/it]
52%|█████▏ | 22644/43939 [36:32:53<33:40:34, 5.69s/it]
52%|█████▏ | 22645/43939 [36:32:58<33:41:19, 5.70s/it]
52%|█████▏ | 22646/43939 [36:33:04<33:37:56, 5.69s/it]
52%|█████▏ | 22647/43939 [36:33:10<33:41:25, 5.70s/it]
52%|█████▏ | 22648/43939 [36:33:16<33:45:26, 5.71s/it]
52%|█████▏ | 22649/43939 [36:33:21<33:41:00, 5.70s/it]
52%|█████▏ | 22650/43939 [36:33:27<33:50:21, 5.72s/it]
{'loss': 0.7784, 'grad_norm': 1.0484875440597534, 'learning_rate': 2.8005530628446297e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.61, 'epoch': 2.06}
52%|█████▏ | 22650/43939 [36:33:27<33:50:21, 5.72s/it]
52%|█████▏ | 22651/43939 [36:33:33<33:50:03, 5.72s/it]
52%|█████▏ | 22652/43939 [36:33:38<33:43:47, 5.70s/it]
52%|█████▏ | 22653/43939 [36:33:44<33:44:34, 5.71s/it]
52%|█████▏ | 22654/43939 [36:33:50<34:06:51, 5.77s/it]
52%|█████▏ | 22655/43939 [36:33:56<33:58:43, 5.75s/it]
52%|█████▏ | 22656/43939 [36:34:01<33:52:59, 5.73s/it]
52%|█████▏ | 22657/43939 [36:34:07<33:54:11, 5.73s/it]
52%|█████▏ | 22658/43939 [36:34:13<33:47:46, 5.72s/it]
52%|█████▏ | 22659/43939 [36:34:19<33:49:27, 5.72s/it]
52%|█████▏ | 22660/43939 [36:34:24<33:52:06, 5.73s/it]
52%|█████▏ | 22661/43939 [36:34:30<33:48:19, 5.72s/it]
52%|█████▏ | 22662/43939 [36:34:36<33:45:05, 5.71s/it]
52%|█████▏ | 22663/43939 [36:34:41<33:47:34, 5.72s/it]
52%|█████▏ | 22664/43939 [36:34:47<33:44:31, 5.71s/it]
52%|█████▏ | 22665/43939 [36:34:53<33:40:41, 5.70s/it]
52%|█████▏ | 22666/43939 [36:34:58<33:35:52, 5.69s/it]
52%|█████▏ | 22667/43939 [36:35:04<33:42:00, 5.70s/it]
52%|█████▏ | 22668/43939 [36:35:10<33:41:58, 5.70s/it]
52%|█████▏ | 22669/43939 [36:35:16<33:38:50, 5.69s/it]
52%|█████▏ | 22670/43939 [36:35:21<33:39:56, 5.70s/it]
52%|█████▏ | 22671/43939 [36:35:27<33:40:29, 5.70s/it]
52%|█████▏ | 22672/43939 [36:35:33<33:46:55, 5.72s/it]
52%|█████▏ | 22673/43939 [36:35:38<33:46:29, 5.72s/it]
52%|█████▏ | 22674/43939 [36:35:45<34:17:06, 5.80s/it]
52%|█████▏ | 22675/43939 [36:35:50<34:09:43, 5.78s/it]
52%|█████▏ | 22676/43939 [36:35:56<34:03:59, 5.77s/it]
52%|█████▏ | 22677/43939 [36:36:02<34:01:10, 5.76s/it]
52%|█████▏ | 22678/43939 [36:36:07<33:55:03, 5.74s/it]
52%|█████▏ | 22679/43939 [36:36:13<33:48:02, 5.72s/it]
52%|█████▏ | 22680/43939 [36:36:19<33:47:07, 5.72s/it]
52%|█████▏ | 22681/43939 [36:36:24<33:42:22, 5.71s/it]
52%|█████▏ | 22682/43939 [36:36:30<33:41:44, 5.71s/it]
52%|█████▏ | 22683/43939 [36:36:36<33:41:30, 5.71s/it]
52%|█████▏ | 22684/43939 [36:36:42<33:43:08, 5.71s/it]
52%|█████▏ | 22685/43939 [36:36:47<33:40:16, 5.70s/it]
52%|█████▏ | 22686/43939 [36:36:53<33:42:56, 5.71s/it]
52%|█████▏ | 22687/43939 [36:36:59<33:38:38, 5.70s/it]
52%|█████▏ | 22688/43939 [36:37:04<33:37:48, 5.70s/it]
52%|█████▏ | 22689/43939 [36:37:10<33:35:04, 5.69s/it]
52%|█████▏ | 22690/43939 [36:37:16<33:36:00, 5.69s/it]
52%|█████▏ | 22691/43939 [36:37:22<33:43:04, 5.71s/it]
52%|█████▏ | 22692/43939 [36:37:27<33:39:45, 5.70s/it]
52%|█████▏ | 22693/43939 [36:37:33<33:46:57, 5.72s/it]
52%|█████▏ | 22694/43939 [36:37:39<33:43:32, 5.71s/it]
52%|█████▏ | 22695/43939 [36:37:45<34:14:02, 5.80s/it]
52%|█████▏ | 22696/43939 [36:37:50<34:05:46, 5.78s/it]
52%|█████▏ | 22697/43939 [36:37:56<33:57:16, 5.75s/it]
52%|█████▏ | 22698/43939 [36:38:02<33:54:40, 5.75s/it]
52%|█████▏ | 22699/43939 [36:38:07<33:45:29, 5.72s/it]
52%|█████▏ | 22700/43939 [36:38:13<33:45:37, 5.72s/it]
{'loss': 0.7757, 'grad_norm': 1.2280105352401733, 'learning_rate': 2.790692555537468e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.42, 'epoch': 2.07}
52%|█████▏ | 22700/43939 [36:38:13<33:45:37, 5.72s/it]
52%|█████▏ | 22701/43939 [36:38:19<33:45:20, 5.72s/it]
52%|█████▏ | 22702/43939 [36:38:25<33:42:21, 5.71s/it]
52%|█████▏ | 22703/43939 [36:38:30<33:40:31, 5.71s/it]
52%|█████▏ | 22704/43939 [36:38:36<33:34:57, 5.69s/it]
52%|█████▏ | 22705/43939 [36:38:42<33:40:10, 5.71s/it]
52%|█████▏ | 22706/43939 [36:38:47<33:35:30, 5.70s/it]
52%|█████▏ | 22707/43939 [36:38:53<33:36:55, 5.70s/it]
52%|█████▏ | 22708/43939 [36:38:59<33:34:57, 5.69s/it]
52%|█████▏ | 22709/43939 [36:39:05<33:40:03, 5.71s/it]
52%|█████▏ | 22710/43939 [36:39:10<33:40:19, 5.71s/it]
52%|█████▏ | 22711/43939 [36:39:16<33:39:17, 5.71s/it]
52%|█████▏ | 22712/43939 [36:39:22<33:40:19, 5.71s/it]
52%|█████▏ | 22713/43939 [36:39:27<33:35:37, 5.70s/it]
52%|█████▏ | 22714/43939 [36:39:33<33:36:26, 5.70s/it]
52%|█████▏ | 22715/43939 [36:39:39<33:36:56, 5.70s/it]
52%|█████▏ | 22716/43939 [36:39:45<34:03:05, 5.78s/it]
52%|█████▏ | 22717/43939 [36:39:50<33:54:03, 5.75s/it]
52%|█████▏ | 22718/43939 [36:39:56<33:53:31, 5.75s/it]
52%|█████▏ | 22719/43939 [36:40:02<33:52:24, 5.75s/it]
52%|█████▏ | 22720/43939 [36:40:08<33:46:27, 5.73s/it]
52%|█████▏ | 22721/43939 [36:40:13<33:42:36, 5.72s/it]
52%|█████▏ | 22722/43939 [36:40:19<33:37:22, 5.70s/it]
52%|█████▏ | 22723/43939 [36:40:25<33:36:23, 5.70s/it]
52%|█████▏ | 22724/43939 [36:40:30<33:38:47, 5.71s/it]
52%|█████▏ | 22725/43939 [36:40:36<33:37:47, 5.71s/it]
52%|█████▏ | 22726/43939 [36:40:42<33:36:57, 5.70s/it]
52%|█████▏ | 22727/43939 [36:40:47<33:36:22, 5.70s/it]
52%|█████▏ | 22728/43939 [36:40:53<33:35:54, 5.70s/it]
52%|█████▏ | 22729/43939 [36:40:59<33:36:41, 5.70s/it]
52%|█████▏ | 22730/43939 [36:41:05<33:37:16, 5.71s/it]
52%|█████▏ | 22731/43939 [36:41:10<33:38:23, 5.71s/it]
52%|█████▏ | 22732/43939 [36:41:16<33:37:52, 5.71s/it]
52%|█████▏ | 22733/43939 [36:41:22<33:33:55, 5.70s/it]
52%|█████▏ | 22734/43939 [36:41:27<33:32:30, 5.69s/it]
52%|█████▏ | 22735/43939 [36:41:33<33:33:31, 5.70s/it]
52%|█████▏ | 22736/43939 [36:41:39<33:35:41, 5.70s/it]
52%|█████▏ | 22737/43939 [36:41:45<34:03:09, 5.78s/it]
52%|█████▏ | 22738/43939 [36:41:50<33:56:00, 5.76s/it]
52%|█████▏ | 22739/43939 [36:41:56<33:52:18, 5.75s/it]
52%|█████▏ | 22740/43939 [36:42:02<33:47:12, 5.74s/it]
52%|█████▏ | 22741/43939 [36:42:08<33:42:50, 5.73s/it]
52%|█████▏ | 22742/43939 [36:42:13<33:44:31, 5.73s/it]
52%|█████▏ | 22743/43939 [36:42:19<33:44:01, 5.73s/it]
52%|█████▏ | 22744/43939 [36:42:25<33:40:50, 5.72s/it]
52%|█████▏ | 22745/43939 [36:42:30<33:35:26, 5.71s/it]
52%|█████▏ | 22746/43939 [36:42:36<33:37:17, 5.71s/it]
52%|█████▏ | 22747/43939 [36:42:42<33:37:08, 5.71s/it]
52%|█████▏ | 22748/43939 [36:42:48<33:36:06, 5.71s/it]
52%|█████▏ | 22749/43939 [36:42:53<33:38:02, 5.71s/it]
52%|█████▏ | 22750/43939 [36:42:59<33:33:37, 5.70s/it]
{'loss': 0.7623, 'grad_norm': 1.350639820098877, 'learning_rate': 2.7808274618721242e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.39, 'epoch': 2.07}
52%|█████▏ | 22750/43939 [36:42:59<33:33:37, 5.70s/it]
52%|█████▏ | 22751/43939 [36:43:05<33:38:00, 5.71s/it]
52%|█████▏ | 22752/43939 [36:43:10<33:40:19, 5.72s/it]
52%|█████▏ | 22753/43939 [36:43:16<33:36:16, 5.71s/it]
52%|█████▏ | 22754/43939 [36:43:22<33:37:54, 5.72s/it]
52%|█████▏ | 22755/43939 [36:43:28<33:36:01, 5.71s/it]
52%|█████▏ | 22756/43939 [36:43:33<33:31:37, 5.70s/it]
52%|█████▏ | 22757/43939 [36:43:39<33:34:35, 5.71s/it]
52%|█████▏ | 22758/43939 [36:43:45<34:01:19, 5.78s/it]
52%|█████▏ | 22759/43939 [36:43:51<33:52:45, 5.76s/it]
52%|█████▏ | 22760/43939 [36:43:56<33:49:31, 5.75s/it]
52%|█████▏ | 22761/43939 [36:44:02<33:45:35, 5.74s/it]
52%|█████▏ | 22762/43939 [36:44:08<33:46:13, 5.74s/it]
52%|█████▏ | 22763/43939 [36:44:14<33:49:49, 5.75s/it]
52%|█████▏ | 22764/43939 [36:44:19<33:44:07, 5.74s/it]
52%|█████▏ | 22765/43939 [36:44:25<33:40:51, 5.73s/it]
52%|█████▏ | 22766/43939 [36:44:31<33:36:35, 5.71s/it]
52%|█████▏ | 22767/43939 [36:44:36<33:34:58, 5.71s/it]
52%|█████▏ | 22768/43939 [36:44:42<33:33:47, 5.71s/it]
52%|█████▏ | 22769/43939 [36:44:48<33:35:17, 5.71s/it]
52%|█████▏ | 22770/43939 [36:44:54<33:35:52, 5.71s/it]
52%|█████▏ | 22771/43939 [36:44:59<33:37:45, 5.72s/it]
52%|█████▏ | 22772/43939 [36:45:05<33:43:17, 5.74s/it]
52%|█████▏ | 22773/43939 [36:45:11<33:39:10, 5.72s/it]
52%|█████▏ | 22774/43939 [36:45:16<33:37:22, 5.72s/it]
52%|█████▏ | 22775/43939 [36:45:22<33:32:41, 5.71s/it]
52%|█████▏ | 22776/43939 [36:45:28<33:31:23, 5.70s/it]
52%|█████▏ | 22777/43939 [36:45:33<33:26:30, 5.69s/it]
52%|█████▏ | 22778/43939 [36:45:39<33:30:16, 5.70s/it]
52%|█████▏ | 22779/43939 [36:45:45<33:32:08, 5.71s/it]
52%|█████▏ | 22780/43939 [36:45:51<34:00:03, 5.78s/it]
52%|█████▏ | 22781/43939 [36:45:57<33:49:02, 5.75s/it]
52%|█████▏ | 22782/43939 [36:46:02<33:42:21, 5.74s/it]
52%|█████▏ | 22783/43939 [36:46:08<33:40:14, 5.73s/it]
52%|█████▏ | 22784/43939 [36:46:14<33:34:58, 5.71s/it]
52%|█████▏ | 22785/43939 [36:46:19<33:38:48, 5.73s/it]
52%|█████▏ | 22786/43939 [36:46:25<33:34:03, 5.71s/it]
52%|█████▏ | 22787/43939 [36:46:31<33:27:52, 5.70s/it]
52%|█████▏ | 22788/43939 [36:46:36<33:27:49, 5.70s/it]
52%|█████▏ | 22789/43939 [36:46:42<33:28:14, 5.70s/it]
52%|█████▏ | 22790/43939 [36:46:48<33:30:05, 5.70s/it]
52%|█████▏ | 22791/43939 [36:46:54<33:32:17, 5.71s/it]
52%|█████▏ | 22792/43939 [36:46:59<33:36:18, 5.72s/it]
52%|█████▏ | 22793/43939 [36:47:05<33:32:01, 5.71s/it]
52%|█████▏ | 22794/43939 [36:47:11<33:31:56, 5.71s/it]
52%|█████▏ | 22795/43939 [36:47:16<33:31:01, 5.71s/it]
52%|█████▏ | 22796/43939 [36:47:22<33:33:45, 5.71s/it]
52%|█████▏ | 22797/43939 [36:47:28<33:40:49, 5.74s/it]
52%|█████▏ | 22798/43939 [36:47:34<33:37:19, 5.73s/it]
52%|█████▏ | 22799/43939 [36:47:39<33:36:17, 5.72s/it]
52%|█████▏ | 22800/43939 [36:47:45<33:30:44, 5.71s/it]
{'loss': 0.7781, 'grad_norm': 1.449352502822876, 'learning_rate': 2.7709579374936328e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.84, 'epoch': 2.08}
52%|█████▏ | 22800/43939 [36:47:45<33:30:44, 5.71s/it]
52%|█████▏ | 22801/43939 [36:47:51<33:59:12, 5.79s/it]
52%|█████▏ | 22802/43939 [36:47:57<33:47:17, 5.75s/it]
52%|█████▏ | 22803/43939 [36:48:02<33:39:58, 5.73s/it]
52%|█████▏ | 22804/43939 [36:48:08<33:39:19, 5.73s/it]
52%|█████▏ | 22805/43939 [36:48:14<33:32:53, 5.71s/it]
52%|█████▏ | 22806/43939 [36:48:19<33:34:57, 5.72s/it]
52%|█████▏ | 22807/43939 [36:48:25<33:31:39, 5.71s/it]
52%|█████▏ | 22808/43939 [36:48:31<33:32:01, 5.71s/it]
52%|█████▏ | 22809/43939 [36:48:37<33:29:27, 5.71s/it]
52%|█████▏ | 22810/43939 [36:48:42<33:31:28, 5.71s/it]
52%|█████▏ | 22811/43939 [36:48:48<33:24:40, 5.69s/it]
52%|█████▏ | 22812/43939 [36:48:54<33:35:00, 5.72s/it]
52%|█████▏ | 22813/43939 [36:48:59<33:33:57, 5.72s/it]
52%|█████▏ | 22814/43939 [36:49:05<33:41:47, 5.74s/it]
52%|█████▏ | 22815/43939 [36:49:11<33:37:20, 5.73s/it]
52%|█████▏ | 22816/43939 [36:49:17<33:34:55, 5.72s/it]
52%|█████▏ | 22817/43939 [36:49:22<33:31:32, 5.71s/it]
52%|█████▏ | 22818/43939 [36:49:28<33:24:23, 5.69s/it]
52%|█████▏ | 22819/43939 [36:49:34<33:32:45, 5.72s/it]
52%|█████▏ | 22820/43939 [36:49:39<33:33:24, 5.72s/it]
52%|█████▏ | 22821/43939 [36:49:45<33:31:49, 5.72s/it]
52%|█████▏ | 22822/43939 [36:49:51<33:54:37, 5.78s/it]
52%|█████▏ | 22823/43939 [36:49:57<33:43:50, 5.75s/it]
52%|█████▏ | 22824/43939 [36:50:03<33:36:44, 5.73s/it]
52%|█████▏ | 22825/43939 [36:50:08<33:29:15, 5.71s/it]
52%|█████▏ | 22826/43939 [36:50:14<33:26:54, 5.70s/it]
52%|█████▏ | 22827/43939 [36:50:20<33:22:28, 5.69s/it]
52%|█████▏ | 22828/43939 [36:50:25<33:20:49, 5.69s/it]
52%|█████▏ | 22829/43939 [36:50:31<33:22:35, 5.69s/it]
52%|█████▏ | 22830/43939 [36:50:37<33:28:53, 5.71s/it]
52%|█████▏ | 22831/43939 [36:50:42<33:32:11, 5.72s/it]
52%|█████▏ | 22832/43939 [36:50:48<33:31:46, 5.72s/it]
52%|█████▏ | 22833/43939 [36:50:54<33:30:52, 5.72s/it]
52%|█████▏ | 22834/43939 [36:51:00<33:30:56, 5.72s/it]
52%|█████▏ | 22835/43939 [36:51:05<33:35:07, 5.73s/it]
52%|█████▏ | 22836/43939 [36:51:11<33:31:20, 5.72s/it]
52%|█████▏ | 22837/43939 [36:51:17<33:26:28, 5.71s/it]
52%|█████▏ | 22838/43939 [36:51:22<33:23:38, 5.70s/it]
52%|█████▏ | 22839/43939 [36:51:28<33:30:30, 5.72s/it]
52%|█████▏ | 22840/43939 [36:51:34<33:32:38, 5.72s/it]
52%|█████▏ | 22841/43939 [36:51:40<33:33:33, 5.73s/it]
52%|█████▏ | 22842/43939 [36:51:45<33:33:23, 5.73s/it]
52%|█████▏ | 22843/43939 [36:51:51<33:53:16, 5.78s/it]
52%|█████▏ | 22844/43939 [36:51:57<33:41:35, 5.75s/it]
52%|█████▏ | 22845/43939 [36:52:03<33:33:13, 5.73s/it]
52%|█████▏ | 22846/43939 [36:52:08<33:28:47, 5.71s/it]
52%|█████▏ | 22847/43939 [36:52:14<33:24:39, 5.70s/it]
52%|█████▏ | 22848/43939 [36:52:20<33:29:58, 5.72s/it]
52%|█████▏ | 22849/43939 [36:52:25<33:27:01, 5.71s/it]
52%|█████▏ | 22850/43939 [36:52:31<33:23:45, 5.70s/it]
{'loss': 0.7661, 'grad_norm': 1.2269128561019897, 'learning_rate': 2.7610841381169362e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.16, 'epoch': 2.08}
52%|█████▏ | 22850/43939 [36:52:31<33:23:45, 5.70s/it]
52%|█████▏ | 22851/43939 [36:52:37<33:26:06, 5.71s/it]
52%|█████▏ | 22852/43939 [36:52:42<33:24:19, 5.70s/it]
52%|█████▏ | 22853/43939 [36:52:48<33:21:03, 5.69s/it]
52%|█████▏ | 22854/43939 [36:52:54<33:22:36, 5.70s/it]
52%|█████▏ | 22855/43939 [36:53:00<33:24:31, 5.70s/it]
52%|█████▏ | 22856/43939 [36:53:05<33:27:13, 5.71s/it]
52%|█████▏ | 22857/43939 [36:53:11<33:29:50, 5.72s/it]
52%|█████▏ | 22858/43939 [36:53:17<33:31:46, 5.73s/it]
52%|█████▏ | 22859/43939 [36:53:22<33:25:17, 5.71s/it]
52%|█████▏ | 22860/43939 [36:53:28<33:25:05, 5.71s/it]
52%|█████▏ | 22861/43939 [36:53:34<33:23:14, 5.70s/it]
52%|█████▏ | 22862/43939 [36:53:40<33:22:46, 5.70s/it]
52%|█████▏ | 22863/43939 [36:53:45<33:22:04, 5.70s/it]
52%|█████▏ | 22864/43939 [36:53:51<33:22:28, 5.70s/it]
52%|█████▏ | 22865/43939 [36:53:57<33:42:23, 5.76s/it]
52%|█████▏ | 22866/43939 [36:54:03<33:38:15, 5.75s/it]
52%|█████▏ | 22867/43939 [36:54:08<33:30:19, 5.72s/it]
52%|█████▏ | 22868/43939 [36:54:14<33:22:47, 5.70s/it]
52%|█████▏ | 22869/43939 [36:54:20<33:25:17, 5.71s/it]
52%|█████▏ | 22870/43939 [36:54:25<33:27:40, 5.72s/it]
52%|█████▏ | 22871/43939 [36:54:31<33:27:13, 5.72s/it]
52%|█████▏ | 22872/43939 [36:54:37<33:22:23, 5.70s/it]
52%|█████▏ | 22873/43939 [36:54:42<33:21:22, 5.70s/it]
52%|█████▏ | 22874/43939 [36:54:48<33:22:52, 5.70s/it]
52%|█████▏ | 22875/43939 [36:54:54<33:17:10, 5.69s/it]
52%|█████▏ | 22876/43939 [36:54:59<33:19:38, 5.70s/it]
52%|█████▏ | 22877/43939 [36:55:05<33:19:43, 5.70s/it]
52%|█████▏ | 22878/43939 [36:55:11<33:17:55, 5.69s/it]
52%|█████▏ | 22879/43939 [36:55:17<33:19:20, 5.70s/it]
52%|█████▏ | 22880/43939 [36:55:22<33:18:12, 5.69s/it]
52%|█████▏ | 22881/43939 [36:55:28<33:18:08, 5.69s/it]
52%|█████▏ | 22882/43939 [36:55:34<33:24:11, 5.71s/it]
52%|█████▏ | 22883/43939 [36:55:39<33:25:15, 5.71s/it]
52%|█████▏ | 22884/43939 [36:55:45<33:24:43, 5.71s/it]
52%|█████▏ | 22885/43939 [36:55:51<33:26:31, 5.72s/it]
52%|█████▏ | 22886/43939 [36:55:57<33:47:33, 5.78s/it]
52%|█████▏ | 22887/43939 [36:56:02<33:40:41, 5.76s/it]
52%|█████▏ | 22888/43939 [36:56:08<33:39:48, 5.76s/it]
52%|█████▏ | 22889/43939 [36:56:14<33:33:08, 5.74s/it]
52%|█████▏ | 22890/43939 [36:56:20<33:32:18, 5.74s/it]
52%|█████▏ | 22891/43939 [36:56:25<33:24:23, 5.71s/it]
52%|█████▏ | 22892/43939 [36:56:31<33:27:34, 5.72s/it]
52%|█████▏ | 22893/43939 [36:56:37<33:23:03, 5.71s/it]
52%|█████▏ | 22894/43939 [36:56:42<33:19:34, 5.70s/it]
52%|█████▏ | 22895/43939 [36:56:48<33:25:31, 5.72s/it]
52%|█████▏ | 22896/43939 [36:56:54<33:24:39, 5.72s/it]
52%|█████▏ | 22897/43939 [36:57:00<33:21:29, 5.71s/it]
52%|█████▏ | 22898/43939 [36:57:05<33:19:42, 5.70s/it]
52%|█████▏ | 22899/43939 [36:57:11<33:21:54, 5.71s/it]
52%|█████▏ | 22900/43939 [36:57:17<33:25:16, 5.72s/it]
{'loss': 0.7809, 'grad_norm': 1.1848582029342651, 'learning_rate': 2.7512062195244215e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.9, 'epoch': 2.08}
52%|█████▏ | 22900/43939 [36:57:17<33:25:16, 5.72s/it]
52%|█████▏ | 22901/43939 [36:57:22<33:25:23, 5.72s/it]
52%|█████▏ | 22902/43939 [36:57:28<33:31:51, 5.74s/it]
52%|█████▏ | 22903/43939 [36:57:34<33:26:38, 5.72s/it]
52%|█████▏ | 22904/43939 [36:57:40<33:22:56, 5.71s/it]
52%|█████▏ | 22905/43939 [36:57:45<33:19:34, 5.70s/it]
52%|█████▏ | 22906/43939 [36:57:51<33:19:14, 5.70s/it]
52%|█████▏ | 22907/43939 [36:57:57<33:42:04, 5.77s/it]
52%|█████▏ | 22908/43939 [36:58:03<33:34:33, 5.75s/it]
52%|█████▏ | 22909/43939 [36:58:08<33:33:53, 5.75s/it]
52%|█████▏ | 22910/43939 [36:58:14<33:24:46, 5.72s/it]
52%|█████▏ | 22911/43939 [36:58:20<33:23:48, 5.72s/it]
52%|█████▏ | 22912/43939 [36:58:25<33:21:23, 5.71s/it]
52%|█████▏ | 22913/43939 [36:58:31<33:19:26, 5.71s/it]
52%|█████▏ | 22914/43939 [36:58:37<33:16:41, 5.70s/it]
52%|█████▏ | 22915/43939 [36:58:43<33:24:24, 5.72s/it]
52%|█████▏ | 22916/43939 [36:58:48<33:20:45, 5.71s/it]
52%|█████▏ | 22917/43939 [36:58:54<33:24:46, 5.72s/it]
52%|█████▏ | 22918/43939 [36:59:00<33:22:42, 5.72s/it]
52%|█████▏ | 22919/43939 [36:59:05<33:27:36, 5.73s/it]
52%|█████▏ | 22920/43939 [36:59:11<33:25:29, 5.72s/it]
52%|█████▏ | 22921/43939 [36:59:17<33:24:46, 5.72s/it]
52%|█████▏ | 22922/43939 [36:59:23<33:23:29, 5.72s/it]
52%|█████▏ | 22923/43939 [36:59:28<33:22:54, 5.72s/it]
52%|█████▏ | 22924/43939 [36:59:34<33:21:40, 5.71s/it]
52%|█████▏ | 22925/43939 [36:59:40<33:21:49, 5.72s/it]
52%|█████▏ | 22926/43939 [36:59:45<33:21:41, 5.72s/it]
52%|█████▏ | 22927/43939 [36:59:51<33:14:49, 5.70s/it]
52%|█████▏ | 22928/43939 [36:59:57<33:14:41, 5.70s/it]
52%|█████▏ | 22929/43939 [37:00:03<33:37:49, 5.76s/it]
52%|█████▏ | 22930/43939 [37:00:08<33:34:45, 5.75s/it]
52%|█████▏ | 22931/43939 [37:00:14<33:25:44, 5.73s/it]
52%|█████▏ | 22932/43939 [37:00:20<33:21:01, 5.72s/it]
52%|█████▏ | 22933/43939 [37:00:26<33:22:21, 5.72s/it]
52%|█████▏ | 22934/43939 [37:00:31<33:18:46, 5.71s/it]
52%|█████▏ | 22935/43939 [37:00:37<33:19:49, 5.71s/it]
52%|█████▏ | 22936/43939 [37:00:43<33:22:42, 5.72s/it]
52%|█████▏ | 22937/43939 [37:00:48<33:19:57, 5.71s/it]
52%|█████▏ | 22938/43939 [37:00:54<33:15:16, 5.70s/it]
52%|█████▏ | 22939/43939 [37:01:00<33:15:39, 5.70s/it]
52%|█████▏ | 22940/43939 [37:01:05<33:15:42, 5.70s/it]
52%|█████▏ | 22941/43939 [37:01:11<33:18:02, 5.71s/it]
52%|█████▏ | 22942/43939 [37:01:17<33:16:35, 5.71s/it]
52%|█████▏ | 22943/43939 [37:01:23<33:20:20, 5.72s/it]
52%|█████▏ | 22944/43939 [37:01:28<33:19:17, 5.71s/it]
52%|█████▏ | 22945/43939 [37:01:34<33:15:05, 5.70s/it]
52%|█████▏ | 22946/43939 [37:01:40<33:13:19, 5.70s/it]
52%|█████▏ | 22947/43939 [37:01:45<33:13:36, 5.70s/it]
52%|█████▏ | 22948/43939 [37:01:51<33:12:50, 5.70s/it]
52%|█████▏ | 22949/43939 [37:01:57<33:13:18, 5.70s/it]
52%|█████▏ | 22950/43939 [37:02:03<33:11:45, 5.69s/it]
{'loss': 0.7771, 'grad_norm': 1.2756482362747192, 'learning_rate': 2.741324337563467e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.14, 'epoch': 2.09}
52%|█████▏ | 22950/43939 [37:02:03<33:11:45, 5.69s/it]
52%|█████▏ | 22951/43939 [37:02:08<33:31:26, 5.75s/it]
52%|█████▏ | 22952/43939 [37:02:14<33:29:54, 5.75s/it]
52%|█████▏ | 22953/43939 [37:02:20<33:29:11, 5.74s/it]
52%|█████▏ | 22954/43939 [37:02:26<33:29:55, 5.75s/it]
52%|█████▏ | 22955/43939 [37:02:31<33:28:30, 5.74s/it]
52%|█████▏ | 22956/43939 [37:02:37<33:23:21, 5.73s/it]
52%|█████▏ | 22957/43939 [37:02:43<33:17:03, 5.71s/it]
52%|█████▏ | 22958/43939 [37:02:48<33:17:33, 5.71s/it]
52%|█████▏ | 22959/43939 [37:02:54<33:22:40, 5.73s/it]
52%|█████▏ | 22960/43939 [37:03:00<33:18:46, 5.72s/it]
52%|█████▏ | 22961/43939 [37:03:06<33:15:41, 5.71s/it]
52%|█████▏ | 22962/43939 [37:03:11<33:21:16, 5.72s/it]
52%|█████▏ | 22963/43939 [37:03:17<33:20:11, 5.72s/it]
52%|█████▏ | 22964/43939 [37:03:23<33:15:56, 5.71s/it]
52%|█████▏ | 22965/43939 [37:03:28<33:19:32, 5.72s/it]
52%|█████▏ | 22966/43939 [37:03:34<33:20:28, 5.72s/it]
52%|█████▏ | 22967/43939 [37:03:40<33:18:45, 5.72s/it]
52%|█████▏ | 22968/43939 [37:03:46<33:14:42, 5.71s/it]
52%|█████▏ | 22969/43939 [37:03:51<33:11:21, 5.70s/it]
52%|█████▏ | 22970/43939 [37:03:57<33:09:06, 5.69s/it]
52%|█████▏ | 22971/43939 [37:04:03<33:32:09, 5.76s/it]
52%|█████▏ | 22972/43939 [37:04:09<33:21:42, 5.73s/it]
52%|█████▏ | 22973/43939 [37:04:14<33:22:20, 5.73s/it]
52%|█████▏ | 22974/43939 [37:04:20<33:16:53, 5.71s/it]
52%|█████▏ | 22975/43939 [37:04:26<33:13:02, 5.70s/it]
52%|█████▏ | 22976/43939 [37:04:31<33:11:12, 5.70s/it]
52%|█████▏ | 22977/43939 [37:04:37<33:13:18, 5.71s/it]
52%|█████▏ | 22978/43939 [37:04:43<33:15:59, 5.71s/it]
52%|█████▏ | 22979/43939 [37:04:48<33:14:02, 5.71s/it]
52%|█████▏ | 22980/43939 [37:04:54<33:18:39, 5.72s/it]
52%|█████▏ | 22981/43939 [37:05:00<33:12:54, 5.71s/it]
52%|█████▏ | 22982/43939 [37:05:06<33:11:38, 5.70s/it]
52%|█████▏ | 22983/43939 [37:05:11<33:10:47, 5.70s/it]
52%|█████▏ | 22984/43939 [37:05:17<33:06:28, 5.69s/it]
52%|█████▏ | 22985/43939 [37:05:23<33:09:21, 5.70s/it]
52%|█████▏ | 22986/43939 [37:05:28<33:10:29, 5.70s/it]
52%|█████▏ | 22987/43939 [37:05:34<33:13:59, 5.71s/it]
52%|█████▏ | 22988/43939 [37:05:40<33:11:42, 5.70s/it]
52%|█████▏ | 22989/43939 [37:05:46<33:15:18, 5.71s/it]
52%|█████▏ | 22990/43939 [37:05:51<33:11:44, 5.70s/it]
52%|█████▏ | 22991/43939 [37:05:57<33:12:23, 5.71s/it]
52%|█████▏ | 22992/43939 [37:06:03<33:35:38, 5.77s/it]
52%|█████▏ | 22993/43939 [37:06:08<33:23:40, 5.74s/it]
52%|█████▏ | 22994/43939 [37:06:14<33:18:01, 5.72s/it]
52%|█████▏ | 22995/43939 [37:06:20<33:15:39, 5.72s/it]
52%|█████▏ | 22996/43939 [37:06:26<33:12:49, 5.71s/it]
52%|█████▏ | 22997/43939 [37:06:31<33:15:32, 5.72s/it]
52%|█████▏ | 22998/43939 [37:06:37<33:13:56, 5.71s/it]
52%|█████▏ | 22999/43939 [37:06:43<33:13:55, 5.71s/it]
52%|█████▏ | 23000/43939 [37:06:48<33:14:06, 5.71s/it]
{'loss': 0.767, 'grad_norm': 1.1643390655517578, 'learning_rate': 2.7314386481439835e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.41, 'epoch': 2.09}
52%|█████▏ | 23000/43939 [37:06:48<33:14:06, 5.71s/it]
52%|█████▏ | 23001/43939 [37:06:54<33:12:28, 5.71s/it]
52%|█████▏ | 23002/43939 [37:07:00<33:12:10, 5.71s/it]
52%|█████▏ | 23003/43939 [37:07:06<33:11:42, 5.71s/it]
52%|█████▏ | 23004/43939 [37:07:11<33:13:49, 5.71s/it]
52%|█████▏ | 23005/43939 [37:07:17<33:16:53, 5.72s/it]
52%|█████▏ | 23006/43939 [37:07:23<33:15:01, 5.72s/it]
52%|█████▏ | 23007/43939 [37:07:29<33:21:07, 5.74s/it]
52%|█████▏ | 23008/43939 [37:07:34<33:13:55, 5.72s/it]
52%|█████▏ | 23009/43939 [37:07:40<33:08:50, 5.70s/it]
52%|█████▏ | 23010/43939 [37:07:46<33:09:03, 5.70s/it]
52%|█████▏ | 23011/43939 [37:07:51<33:09:40, 5.70s/it]
52%|█████▏ | 23012/43939 [37:07:57<33:12:15, 5.71s/it]
52%|█████▏ | 23013/43939 [37:08:03<33:33:29, 5.77s/it]
52%|█████▏ | 23014/43939 [37:08:09<33:26:38, 5.75s/it]
52%|█████▏ | 23015/43939 [37:08:14<33:19:41, 5.73s/it]
52%|█████▏ | 23016/43939 [37:08:20<33:15:48, 5.72s/it]
52%|█████▏ | 23017/43939 [37:08:26<33:15:34, 5.72s/it]
52%|█████▏ | 23018/43939 [37:08:31<33:10:32, 5.71s/it]
52%|█████▏ | 23019/43939 [37:08:37<33:10:39, 5.71s/it]
52%|█████▏ | 23020/43939 [37:08:43<33:13:56, 5.72s/it]
52%|█████▏ | 23021/43939 [37:08:49<33:18:06, 5.73s/it]
52%|█████▏ | 23022/43939 [37:08:54<33:15:17, 5.72s/it]
52%|█████▏ | 23023/43939 [37:09:00<33:14:41, 5.72s/it]
52%|█████▏ | 23024/43939 [37:09:06<33:16:22, 5.73s/it]
52%|█████▏ | 23025/43939 [37:09:12<33:17:48, 5.73s/it]
52%|█████▏ | 23026/43939 [37:09:17<33:10:56, 5.71s/it]
52%|█████▏ | 23027/43939 [37:09:23<33:05:32, 5.70s/it]
52%|█████▏ | 23028/43939 [37:09:29<33:03:27, 5.69s/it]
52%|█████▏ | 23029/43939 [37:09:34<33:03:26, 5.69s/it]
52%|█████▏ | 23030/43939 [37:09:40<33:01:30, 5.69s/it]
52%|█████▏ | 23031/43939 [37:09:46<33:01:19, 5.69s/it]
52%|█████▏ | 23032/43939 [37:09:51<33:03:00, 5.69s/it]
52%|█████▏ | 23033/43939 [37:09:57<33:06:59, 5.70s/it]
52%|█████▏ | 23034/43939 [37:10:03<33:35:22, 5.78s/it]
52%|█████▏ | 23035/43939 [37:10:09<33:29:38, 5.77s/it]
52%|█████▏ | 23036/43939 [37:10:14<33:18:15, 5.74s/it]
52%|█████▏ | 23037/43939 [37:10:20<33:10:52, 5.71s/it]
52%|█████▏ | 23038/43939 [37:10:26<33:08:49, 5.71s/it]
52%|█████▏ | 23039/43939 [37:10:31<33:09:34, 5.71s/it]
52%|█████▏ | 23040/43939 [37:10:37<33:10:44, 5.72s/it]
52%|█████▏ | 23041/43939 [37:10:43<33:10:01, 5.71s/it]
52%|█████▏ | 23042/43939 [37:10:49<33:09:44, 5.71s/it]
52%|█████▏ | 23043/43939 [37:10:54<33:06:57, 5.71s/it]
52%|█████▏ | 23044/43939 [37:11:00<33:05:09, 5.70s/it]
52%|█████▏ | 23045/43939 [37:11:06<33:01:51, 5.69s/it]
52%|█████▏ | 23046/43939 [37:11:11<33:03:33, 5.70s/it]
52%|█████▏ | 23047/43939 [37:11:17<33:03:42, 5.70s/it]
52%|█████▏ | 23048/43939 [37:11:23<33:06:54, 5.71s/it]
52%|█████▏ | 23049/43939 [37:11:28<33:05:16, 5.70s/it]
52%|█████▏ | 23050/43939 [37:11:34<33:03:31, 5.70s/it]
{'loss': 0.7718, 'grad_norm': 1.5686687231063843, 'learning_rate': 2.7215493072359516e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.55, 'epoch': 2.1}
52%|█████▏ | 23050/43939 [37:11:34<33:03:31, 5.70s/it]
52%|█████▏ | 23051/43939 [37:11:40<33:02:50, 5.70s/it]
52%|█████▏ | 23052/43939 [37:11:46<33:05:57, 5.70s/it]
52%|█████▏ | 23053/43939 [37:11:51<33:03:48, 5.70s/it]
52%|█████▏ | 23054/43939 [37:11:57<33:06:14, 5.71s/it]
52%|█████▏ | 23055/43939 [37:12:03<33:37:06, 5.80s/it]
52%|█████▏ | 23056/43939 [37:12:09<33:28:21, 5.77s/it]
52%|█████▏ | 23057/43939 [37:12:14<33:21:23, 5.75s/it]
52%|█████▏ | 23058/43939 [37:12:20<33:15:16, 5.73s/it]
52%|█████▏ | 23059/43939 [37:12:26<33:08:03, 5.71s/it]
52%|█████▏ | 23060/43939 [37:12:31<33:03:35, 5.70s/it]
52%|█████▏ | 23061/43939 [37:12:37<33:02:07, 5.70s/it]
52%|█████▏ | 23062/43939 [37:12:43<33:00:03, 5.69s/it]
52%|█████▏ | 23063/43939 [37:12:48<32:57:40, 5.68s/it]
52%|█████▏ | 23064/43939 [37:12:54<33:01:12, 5.69s/it]
52%|█████▏ | 23065/43939 [37:13:00<32:58:40, 5.69s/it]
52%|█████▏ | 23066/43939 [37:13:06<32:55:55, 5.68s/it]
52%|█████▏ | 23067/43939 [37:13:11<32:58:13, 5.69s/it]
53%|█████▎ | 23068/43939 [37:13:17<33:04:34, 5.71s/it]
53%|█████▎ | 23069/43939 [37:13:23<33:06:44, 5.71s/it]
53%|█████▎ | 23070/43939 [37:13:28<33:06:34, 5.71s/it]
53%|█████▎ | 23071/43939 [37:13:34<33:02:41, 5.70s/it]
53%|█████▎ | 23072/43939 [37:13:40<33:08:29, 5.72s/it]
53%|█████▎ | 23073/43939 [37:13:46<33:06:07, 5.71s/it]
53%|█████▎ | 23074/43939 [37:13:51<33:01:12, 5.70s/it]
53%|█████▎ | 23075/43939 [37:13:57<33:03:17, 5.70s/it]
53%|█████▎ | 23076/43939 [37:14:03<33:06:47, 5.71s/it]
53%|█████▎ | 23077/43939 [37:14:09<33:29:34, 5.78s/it]
53%|█████▎ | 23078/43939 [37:14:14<33:30:17, 5.78s/it]
53%|█████▎ | 23079/43939 [37:14:20<33:23:41, 5.76s/it]
53%|█████▎ | 23080/43939 [37:14:26<33:19:39, 5.75s/it]
53%|█████▎ | 23081/43939 [37:14:32<33:14:20, 5.74s/it]
53%|█████▎ | 23082/43939 [37:14:37<33:11:04, 5.73s/it]
53%|█████▎ | 23083/43939 [37:14:43<33:05:22, 5.71s/it]
53%|█████▎ | 23084/43939 [37:14:49<33:02:54, 5.70s/it]
53%|█████▎ | 23085/43939 [37:14:54<32:58:25, 5.69s/it]
53%|█████▎ | 23086/43939 [37:15:00<32:58:33, 5.69s/it]
53%|█████▎ | 23087/43939 [37:15:06<32:57:56, 5.69s/it]
53%|█████▎ | 23088/43939 [37:15:11<32:58:00, 5.69s/it]
53%|█████▎ | 23089/43939 [37:15:17<32:55:26, 5.68s/it]
53%|█████▎ | 23090/43939 [37:15:23<33:01:38, 5.70s/it]
53%|█████▎ | 23091/43939 [37:15:28<33:03:02, 5.71s/it]
53%|█████▎ | 23092/43939 [37:15:34<33:07:25, 5.72s/it]
53%|█████▎ | 23093/43939 [37:15:40<33:05:13, 5.71s/it]
53%|█████▎ | 23094/43939 [37:15:46<33:01:03, 5.70s/it]
53%|█████▎ | 23095/43939 [37:15:51<33:01:51, 5.70s/it]
53%|█████▎ | 23096/43939 [37:15:57<33:08:25, 5.72s/it]
53%|█████▎ | 23097/43939 [37:16:03<33:08:34, 5.72s/it]
53%|█████▎ | 23098/43939 [37:16:09<33:09:51, 5.73s/it]
53%|█████▎ | 23099/43939 [37:16:15<33:41:34, 5.82s/it]
53%|█████▎ | 23100/43939 [37:16:20<33:25:44, 5.77s/it]
{'loss': 0.7766, 'grad_norm': 1.5676182508468628, 'learning_rate': 2.711656470866964e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 455.68, 'epoch': 2.1}
53%|█████▎ | 23100/43939 [37:16:20<33:25:44, 5.77s/it]
53%|█████▎ | 23101/43939 [37:16:26<33:14:57, 5.74s/it]
53%|█████▎ | 23102/43939 [37:16:32<33:16:15, 5.75s/it]
53%|█████▎ | 23103/43939 [37:16:37<33:11:28, 5.73s/it]
53%|█████▎ | 23104/43939 [37:16:43<33:06:36, 5.72s/it]
53%|█████▎ | 23105/43939 [37:16:49<33:02:29, 5.71s/it]
53%|█████▎ | 23106/43939 [37:16:54<32:58:32, 5.70s/it]
53%|█████▎ | 23107/43939 [37:17:00<32:56:35, 5.69s/it]
53%|█████▎ | 23108/43939 [37:17:06<32:56:09, 5.69s/it]
53%|█████▎ | 23109/43939 [37:17:11<32:50:12, 5.68s/it]
53%|█████▎ | 23110/43939 [37:17:17<32:52:22, 5.68s/it]
53%|█████▎ | 23111/43939 [37:17:23<32:54:48, 5.69s/it]
53%|█████▎ | 23112/43939 [37:17:28<32:53:42, 5.69s/it]
53%|█████▎ | 23113/43939 [37:17:34<32:55:04, 5.69s/it]
53%|█████▎ | 23114/43939 [37:17:40<32:55:31, 5.69s/it]
53%|█████▎ | 23115/43939 [37:17:46<32:54:09, 5.69s/it]
53%|█████▎ | 23116/43939 [37:17:51<32:57:02, 5.70s/it]
53%|█████▎ | 23117/43939 [37:17:57<32:59:03, 5.70s/it]
53%|█████▎ | 23118/43939 [37:18:03<32:56:11, 5.69s/it]
53%|█████▎ | 23119/43939 [37:18:09<33:18:05, 5.76s/it]
53%|█████▎ | 23120/43939 [37:18:14<33:12:46, 5.74s/it]
53%|█████▎ | 23121/43939 [37:18:20<33:06:23, 5.73s/it]
53%|█████▎ | 23122/43939 [37:18:26<33:04:14, 5.72s/it]
53%|█████▎ | 23123/43939 [37:18:31<32:59:23, 5.71s/it]
53%|█████▎ | 23124/43939 [37:18:37<33:03:44, 5.72s/it]
53%|█████▎ | 23125/43939 [37:18:43<33:01:06, 5.71s/it]
53%|█████▎ | 23126/43939 [37:18:49<33:02:38, 5.72s/it]
53%|█████▎ | 23127/43939 [37:18:54<33:02:46, 5.72s/it]
53%|█████▎ | 23128/43939 [37:19:00<33:01:11, 5.71s/it]
53%|█████▎ | 23129/43939 [37:19:06<33:00:08, 5.71s/it]
53%|█████▎ | 23130/43939 [37:19:11<33:05:12, 5.72s/it]
53%|█████▎ | 23131/43939 [37:19:17<33:02:20, 5.72s/it]
53%|█████▎ | 23132/43939 [37:19:23<32:56:35, 5.70s/it]
53%|█████▎ | 23133/43939 [37:19:28<32:51:25, 5.69s/it]
53%|█████▎ | 23134/43939 [37:19:34<32:53:26, 5.69s/it]
53%|█████▎ | 23135/43939 [37:19:40<32:49:38, 5.68s/it]
53%|█████▎ | 23136/43939 [37:19:45<32:50:45, 5.68s/it]
53%|█████▎ | 23137/43939 [37:19:51<32:52:35, 5.69s/it]
53%|█████▎ | 23138/43939 [37:19:57<32:53:34, 5.69s/it]
53%|█████▎ | 23139/43939 [37:20:03<32:54:51, 5.70s/it]
53%|█████▎ | 23140/43939 [37:20:08<33:15:43, 5.76s/it]
53%|█████▎ | 23141/43939 [37:20:14<33:10:25, 5.74s/it]
53%|█████▎ | 23142/43939 [37:20:20<33:10:12, 5.74s/it]
53%|█████▎ | 23143/43939 [37:20:26<33:05:41, 5.73s/it]
53%|█████▎ | 23144/43939 [37:20:31<33:02:52, 5.72s/it]
53%|█████▎ | 23145/43939 [37:20:37<33:00:22, 5.71s/it]
53%|█████▎ | 23146/43939 [37:20:43<33:05:37, 5.73s/it]
53%|█████▎ | 23147/43939 [37:20:49<33:04:03, 5.73s/it]
53%|█████▎ | 23148/43939 [37:20:54<32:59:31, 5.71s/it]
53%|█████▎ | 23149/43939 [37:21:00<33:02:40, 5.72s/it]
53%|█████▎ | 23150/43939 [37:21:06<32:59:31, 5.71s/it]
{'loss': 0.7783, 'grad_norm': 1.47994863986969, 'learning_rate': 2.7017602951197636e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.33, 'epoch': 2.11}
53%|█████▎ | 23150/43939 [37:21:06<32:59:31, 5.71s/it]
53%|█████▎ | 23151/43939 [37:21:11<33:02:56, 5.72s/it]
53%|█████▎ | 23152/43939 [37:21:17<33:01:04, 5.72s/it]
53%|█████▎ | 23153/43939 [37:21:23<32:54:52, 5.70s/it]
53%|█████▎ | 23154/43939 [37:21:28<32:57:01, 5.71s/it]
53%|█████▎ | 23155/43939 [37:21:34<32:56:50, 5.71s/it]
53%|█████▎ | 23156/43939 [37:21:40<32:55:11, 5.70s/it]
53%|█████▎ | 23157/43939 [37:21:46<32:56:15, 5.71s/it]
53%|█████▎ | 23158/43939 [37:21:51<33:00:45, 5.72s/it]
53%|█████▎ | 23159/43939 [37:21:57<33:00:28, 5.72s/it]
53%|█████▎ | 23160/43939 [37:22:03<32:56:22, 5.71s/it]
53%|█████▎ | 23161/43939 [37:22:08<32:51:13, 5.69s/it]
53%|█████▎ | 23162/43939 [37:22:14<33:25:52, 5.79s/it]
53%|█████▎ | 23163/43939 [37:22:20<33:15:34, 5.76s/it]
53%|█████▎ | 23164/43939 [37:22:26<33:12:26, 5.75s/it]
53%|█████▎ | 23165/43939 [37:22:32<33:07:17, 5.74s/it]
53%|█████▎ | 23166/43939 [37:22:37<33:02:54, 5.73s/it]
53%|█████▎ | 23167/43939 [37:22:43<33:01:42, 5.72s/it]
53%|█████▎ | 23168/43939 [37:22:49<33:07:25, 5.74s/it]
53%|█████▎ | 23169/43939 [37:22:54<33:03:57, 5.73s/it]
53%|█████▎ | 23170/43939 [37:23:00<32:59:08, 5.72s/it]
53%|█████▎ | 23171/43939 [37:23:06<32:58:00, 5.71s/it]
53%|█████▎ | 23172/43939 [37:23:12<32:55:53, 5.71s/it]
53%|█████▎ | 23173/43939 [37:23:17<32:53:01, 5.70s/it]
53%|█████▎ | 23174/43939 [37:23:23<32:59:40, 5.72s/it]
53%|█████▎ | 23175/43939 [37:23:29<32:55:30, 5.71s/it]
53%|█████▎ | 23176/43939 [37:23:34<33:00:09, 5.72s/it]
53%|█████▎ | 23177/43939 [37:23:40<32:57:16, 5.71s/it]
53%|█████▎ | 23178/43939 [37:23:46<32:57:36, 5.72s/it]
53%|█████▎ | 23179/43939 [37:23:52<32:57:27, 5.72s/it]
53%|█████▎ | 23180/43939 [37:23:57<32:54:15, 5.71s/it]
53%|█████▎ | 23181/43939 [37:24:03<32:51:30, 5.70s/it]
53%|█████▎ | 23182/43939 [37:24:09<32:53:47, 5.71s/it]
53%|█████▎ | 23183/43939 [37:24:15<33:17:51, 5.78s/it]
53%|█████▎ | 23184/43939 [37:24:20<33:09:00, 5.75s/it]
53%|█████▎ | 23185/43939 [37:24:26<33:04:08, 5.74s/it]
53%|█████▎ | 23186/43939 [37:24:32<33:03:46, 5.74s/it]
53%|█████▎ | 23187/43939 [37:24:37<33:00:27, 5.73s/it]
53%|█████▎ | 23188/43939 [37:24:43<32:56:29, 5.71s/it]
53%|█████▎ | 23189/43939 [37:24:49<32:57:32, 5.72s/it]
53%|█████▎ | 23190/43939 [37:24:54<32:52:01, 5.70s/it]
53%|█████▎ | 23191/43939 [37:25:00<32:54:26, 5.71s/it]
53%|█████▎ | 23192/43939 [37:25:06<32:57:24, 5.72s/it]
53%|█████▎ | 23193/43939 [37:25:12<32:57:22, 5.72s/it]
53%|█████▎ | 23194/43939 [37:25:17<32:55:46, 5.71s/it]
53%|█████▎ | 23195/43939 [37:25:23<32:56:03, 5.72s/it]
53%|█████▎ | 23196/43939 [37:25:29<32:53:47, 5.71s/it]
53%|█████▎ | 23197/43939 [37:25:35<32:56:38, 5.72s/it]
53%|█████▎ | 23198/43939 [37:25:40<32:55:44, 5.72s/it]
53%|█████▎ | 23199/43939 [37:25:46<32:53:33, 5.71s/it]
53%|█████▎ | 23200/43939 [37:25:52<32:51:58, 5.71s/it]
{'loss': 0.7853, 'grad_norm': 1.0682467222213745, 'learning_rate': 2.6918609361297768e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.59, 'epoch': 2.11}
53%|█████▎ | 23200/43939 [37:25:52<32:51:58, 5.71s/it]
53%|█████▎ | 23201/43939 [37:25:57<33:01:00, 5.73s/it]
53%|█████▎ | 23202/43939 [37:26:03<32:58:45, 5.73s/it]
53%|█████▎ | 23203/43939 [37:26:09<32:58:19, 5.72s/it]
53%|█████▎ | 23204/43939 [37:26:15<33:18:06, 5.78s/it]
53%|█████▎ | 23205/43939 [37:26:20<33:09:07, 5.76s/it]
53%|█████▎ | 23206/43939 [37:26:26<33:02:50, 5.74s/it]
53%|█████▎ | 23207/43939 [37:26:32<32:58:21, 5.73s/it]
53%|█████▎ | 23208/43939 [37:26:38<32:53:06, 5.71s/it]
53%|█████▎ | 23209/43939 [37:26:43<32:51:50, 5.71s/it]
53%|█████▎ | 23210/43939 [37:26:49<32:50:07, 5.70s/it]
53%|█████▎ | 23211/43939 [37:26:55<32:56:59, 5.72s/it]
53%|█████▎ | 23212/43939 [37:27:00<32:49:56, 5.70s/it]
53%|█████▎ | 23213/43939 [37:27:06<32:50:44, 5.71s/it]
53%|█████▎ | 23214/43939 [37:27:12<32:48:24, 5.70s/it]
53%|█████▎ | 23215/43939 [37:27:17<32:45:12, 5.69s/it]
53%|█████▎ | 23216/43939 [37:27:23<32:53:14, 5.71s/it]
53%|█████▎ | 23217/43939 [37:27:29<32:53:11, 5.71s/it]
53%|█████▎ | 23218/43939 [37:27:35<32:52:35, 5.71s/it]
53%|█████▎ | 23219/43939 [37:27:40<32:46:33, 5.69s/it]
53%|█████▎ | 23220/43939 [37:27:46<32:45:42, 5.69s/it]
53%|█████▎ | 23221/43939 [37:27:52<32:51:16, 5.71s/it]
53%|█████▎ | 23222/43939 [37:27:57<32:50:11, 5.71s/it]
53%|█████▎ | 23223/43939 [37:28:03<32:47:38, 5.70s/it]
53%|█████▎ | 23224/43939 [37:28:09<32:48:06, 5.70s/it]
53%|█████▎ | 23225/43939 [37:28:14<32:47:01, 5.70s/it]
53%|█████▎ | 23226/43939 [37:28:20<33:09:54, 5.76s/it]
53%|█████▎ | 23227/43939 [37:28:26<33:02:21, 5.74s/it]
53%|█████▎ | 23228/43939 [37:28:32<32:56:00, 5.72s/it]
53%|█████▎ | 23229/43939 [37:28:37<32:54:02, 5.72s/it]
53%|█████▎ | 23230/43939 [37:28:43<32:50:46, 5.71s/it]
53%|█████▎ | 23231/43939 [37:28:49<32:50:20, 5.71s/it]
53%|█████▎ | 23232/43939 [37:28:55<32:51:45, 5.71s/it]
53%|█████▎ | 23233/43939 [37:29:00<32:56:49, 5.73s/it]
53%|█████▎ | 23234/43939 [37:29:06<32:51:02, 5.71s/it]
53%|█████▎ | 23235/43939 [37:29:12<32:49:03, 5.71s/it]
53%|█████▎ | 23236/43939 [37:29:17<32:46:54, 5.70s/it]
53%|█████▎ | 23237/43939 [37:29:23<32:46:04, 5.70s/it]
53%|█████▎ | 23238/43939 [37:29:29<32:46:32, 5.70s/it]
53%|█████▎ | 23239/43939 [37:29:34<32:46:38, 5.70s/it]
53%|█████▎ | 23240/43939 [37:29:40<32:46:09, 5.70s/it]
53%|█████▎ | 23241/43939 [37:29:46<32:45:29, 5.70s/it]
53%|█████▎ | 23242/43939 [37:29:52<32:45:37, 5.70s/it]
53%|█████▎ | 23243/43939 [37:29:57<32:46:37, 5.70s/it]
53%|█████▎ | 23244/43939 [37:30:03<32:48:43, 5.71s/it]
53%|█████▎ | 23245/43939 [37:30:09<32:51:46, 5.72s/it]
53%|█████▎ | 23246/43939 [37:30:14<32:46:21, 5.70s/it]
53%|█████▎ | 23247/43939 [37:30:20<33:11:13, 5.77s/it]
53%|█████▎ | 23248/43939 [37:30:26<33:02:04, 5.75s/it]
53%|█████▎ | 23249/43939 [37:30:32<32:56:18, 5.73s/it]
53%|█████▎ | 23250/43939 [37:30:37<32:52:25, 5.72s/it]
{'loss': 0.7667, 'grad_norm': 1.425839900970459, 'learning_rate': 2.6819585500826565e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.54, 'epoch': 2.12}
53%|█████▎ | 23250/43939 [37:30:37<32:52:25, 5.72s/it]
53%|█████▎ | 23251/43939 [37:30:43<32:51:49, 5.72s/it]
53%|█████▎ | 23252/43939 [37:30:49<32:54:42, 5.73s/it]
53%|█████▎ | 23253/43939 [37:30:55<32:49:32, 5.71s/it]
53%|█████▎ | 23254/43939 [37:31:00<32:48:55, 5.71s/it]
53%|█████▎ | 23255/43939 [37:31:06<32:49:20, 5.71s/it]
53%|█████▎ | 23256/43939 [37:31:12<32:47:21, 5.71s/it]
53%|█████▎ | 23257/43939 [37:31:17<32:47:49, 5.71s/it]
53%|█████▎ | 23258/43939 [37:31:23<32:54:33, 5.73s/it]
53%|█████▎ | 23259/43939 [37:31:29<32:52:40, 5.72s/it]
53%|█████▎ | 23260/43939 [37:31:35<32:57:48, 5.74s/it]
53%|█████▎ | 23261/43939 [37:31:40<32:56:39, 5.74s/it]
53%|█████▎ | 23262/43939 [37:31:46<32:54:39, 5.73s/it]
53%|█████▎ | 23263/43939 [37:31:52<32:59:16, 5.74s/it]
53%|█████▎ | 23264/43939 [37:31:58<32:55:56, 5.73s/it]
53%|█████▎ | 23265/43939 [37:32:03<32:50:16, 5.72s/it]
53%|█████▎ | 23266/43939 [37:32:09<32:45:55, 5.71s/it]
53%|█████▎ | 23267/43939 [37:32:15<32:46:35, 5.71s/it]
53%|█████▎ | 23268/43939 [37:32:21<33:10:35, 5.78s/it]
53%|█████▎ | 23269/43939 [37:32:26<33:04:11, 5.76s/it]
53%|█████▎ | 23270/43939 [37:32:32<32:56:50, 5.74s/it]
53%|█████▎ | 23271/43939 [37:32:38<32:57:18, 5.74s/it]
53%|█████▎ | 23272/43939 [37:32:43<32:50:18, 5.72s/it]
53%|█████▎ | 23273/43939 [37:32:49<32:45:25, 5.71s/it]
53%|█████▎ | 23274/43939 [37:32:55<32:48:40, 5.72s/it]
53%|█████▎ | 23275/43939 [37:33:01<32:48:39, 5.72s/it]
53%|█████▎ | 23276/43939 [37:33:06<32:45:24, 5.71s/it]
53%|█████▎ | 23277/43939 [37:33:12<32:43:24, 5.70s/it]
53%|█████▎ | 23278/43939 [37:33:18<32:43:40, 5.70s/it]
53%|█████▎ | 23279/43939 [37:33:23<32:44:01, 5.70s/it]
53%|█████▎ | 23280/43939 [37:33:29<32:44:51, 5.71s/it]
53%|█████▎ | 23281/43939 [37:33:35<32:45:02, 5.71s/it]
53%|█████▎ | 23282/43939 [37:33:40<32:41:42, 5.70s/it]
53%|█████▎ | 23283/43939 [37:33:46<32:35:51, 5.68s/it]
53%|█████▎ | 23284/43939 [37:33:52<32:38:04, 5.69s/it]
53%|█████▎ | 23285/43939 [37:33:57<32:36:49, 5.68s/it]
53%|█████▎ | 23286/43939 [37:34:03<32:36:13, 5.68s/it]
53%|█████▎ | 23287/43939 [37:34:09<32:38:45, 5.69s/it]
53%|█████▎ | 23288/43939 [37:34:15<32:37:27, 5.69s/it]
53%|█████▎ | 23289/43939 [37:34:20<32:38:54, 5.69s/it]
53%|█████▎ | 23290/43939 [37:34:26<33:06:50, 5.77s/it]
53%|█████▎ | 23291/43939 [37:34:32<33:00:11, 5.75s/it]
53%|█████▎ | 23292/43939 [37:34:38<32:55:33, 5.74s/it]
53%|█████▎ | 23293/43939 [37:34:43<32:53:24, 5.73s/it]
53%|█████▎ | 23294/43939 [37:34:49<32:51:03, 5.73s/it]
53%|█████▎ | 23295/43939 [37:34:55<32:44:18, 5.71s/it]
53%|█████▎ | 23296/43939 [37:35:00<32:40:35, 5.70s/it]
53%|█████▎ | 23297/43939 [37:35:06<32:40:45, 5.70s/it]
53%|█████▎ | 23298/43939 [37:35:12<32:39:34, 5.70s/it]
53%|█████▎ | 23299/43939 [37:35:17<32:38:38, 5.69s/it]
53%|█████▎ | 23300/43939 [37:35:23<32:35:18, 5.68s/it]
{'loss': 0.7465, 'grad_norm': 1.410832166671753, 'learning_rate': 2.672053293211811e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.33, 'epoch': 2.12}
53%|█████▎ | 23300/43939 [37:35:23<32:35:18, 5.68s/it]
53%|█████▎ | 23301/43939 [37:35:29<32:35:01, 5.68s/it]
53%|█████▎ | 23302/43939 [37:35:35<32:35:00, 5.68s/it]
53%|█████▎ | 23303/43939 [37:35:40<32:41:04, 5.70s/it]
53%|█████▎ | 23304/43939 [37:35:46<32:38:29, 5.69s/it]
53%|█████▎ | 23305/43939 [37:35:52<32:40:07, 5.70s/it]
53%|█████▎ | 23306/43939 [37:35:57<32:41:02, 5.70s/it]
53%|█████▎ | 23307/43939 [37:36:03<32:41:00, 5.70s/it]
53%|█████▎ | 23308/43939 [37:36:09<32:41:05, 5.70s/it]
53%|█████▎ | 23309/43939 [37:36:14<32:42:26, 5.71s/it]
53%|█████▎ | 23310/43939 [37:36:20<32:42:40, 5.71s/it]
53%|█████▎ | 23311/43939 [37:36:26<33:09:40, 5.79s/it]
53%|█████▎ | 23312/43939 [37:36:32<32:58:57, 5.76s/it]
53%|█████▎ | 23313/43939 [37:36:38<33:01:11, 5.76s/it]
53%|█████▎ | 23314/43939 [37:36:43<32:56:08, 5.75s/it]
53%|█████▎ | 23315/43939 [37:36:49<32:53:48, 5.74s/it]
53%|█████▎ | 23316/43939 [37:36:55<32:51:46, 5.74s/it]
53%|█████▎ | 23317/43939 [37:37:00<32:45:18, 5.72s/it]
53%|█████▎ | 23318/43939 [37:37:06<32:49:05, 5.73s/it]
53%|█████▎ | 23319/43939 [37:37:12<32:49:19, 5.73s/it]
53%|█████▎ | 23320/43939 [37:37:18<32:52:04, 5.74s/it]
53%|█████▎ | 23321/43939 [37:37:23<32:50:05, 5.73s/it]
53%|█████▎ | 23322/43939 [37:37:29<32:47:12, 5.73s/it]
53%|█████▎ | 23323/43939 [37:37:35<32:44:12, 5.72s/it]
53%|█████▎ | 23324/43939 [37:37:41<32:40:31, 5.71s/it]
53%|█████▎ | 23325/43939 [37:37:46<32:36:28, 5.69s/it]
53%|█████▎ | 23326/43939 [37:37:52<32:35:06, 5.69s/it]
53%|█████▎ | 23327/43939 [37:37:58<32:36:07, 5.69s/it]
53%|█████▎ | 23328/43939 [37:38:03<32:38:52, 5.70s/it]
53%|█████▎ | 23329/43939 [37:38:09<32:38:12, 5.70s/it]
53%|█████▎ | 23330/43939 [37:38:15<32:38:08, 5.70s/it]
53%|█████▎ | 23331/43939 [37:38:20<32:40:51, 5.71s/it]
53%|█████▎ | 23332/43939 [37:38:26<33:04:14, 5.78s/it]
53%|█████▎ | 23333/43939 [37:38:32<32:56:47, 5.76s/it]
53%|█████▎ | 23334/43939 [37:38:38<32:50:16, 5.74s/it]
53%|█████▎ | 23335/43939 [37:38:43<32:44:02, 5.72s/it]
53%|█████▎ | 23336/43939 [37:38:49<32:45:31, 5.72s/it]
53%|█████▎ | 23337/43939 [37:38:55<32:42:02, 5.71s/it]
53%|█████▎ | 23338/43939 [37:39:01<32:40:20, 5.71s/it]
53%|█████▎ | 23339/43939 [37:39:06<32:40:25, 5.71s/it]
53%|█████▎ | 23340/43939 [37:39:12<32:40:14, 5.71s/it]
53%|█████▎ | 23341/43939 [37:39:18<32:40:36, 5.71s/it]
53%|█████▎ | 23342/43939 [37:39:23<32:42:49, 5.72s/it]
53%|█████▎ | 23343/43939 [37:39:29<32:38:43, 5.71s/it]
53%|█████▎ | 23344/43939 [37:39:35<32:40:47, 5.71s/it]
53%|█████▎ | 23345/43939 [37:39:41<32:42:46, 5.72s/it]
53%|█████▎ | 23346/43939 [37:39:46<32:47:29, 5.73s/it]
53%|█████▎ | 23347/43939 [37:39:52<32:48:04, 5.73s/it]
53%|█████▎ | 23348/43939 [37:39:58<32:44:48, 5.73s/it]
53%|█████▎ | 23349/43939 [37:40:03<32:44:54, 5.73s/it]
53%|█████▎ | 23350/43939 [37:40:09<32:44:26, 5.72s/it]
{'loss': 0.7666, 'grad_norm': 1.1838409900665283, 'learning_rate': 2.6621453217959454e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.97, 'epoch': 2.13}
53%|█████▎ | 23350/43939 [37:40:09<32:44:26, 5.72s/it]
53%|█████▎ | 23351/43939 [37:40:15<32:42:46, 5.72s/it]
53%|█████▎ | 23352/43939 [37:40:21<32:44:00, 5.72s/it]
53%|█████▎ | 23353/43939 [37:40:27<33:09:55, 5.80s/it]
53%|█████▎ | 23354/43939 [37:40:32<32:59:40, 5.77s/it]
53%|█████▎ | 23355/43939 [37:40:38<32:50:19, 5.74s/it]
53%|█████▎ | 23356/43939 [37:40:44<32:46:14, 5.73s/it]
53%|█████▎ | 23357/43939 [37:40:49<32:46:43, 5.73s/it]
53%|█████▎ | 23358/43939 [37:40:55<32:43:08, 5.72s/it]
53%|█████▎ | 23359/43939 [37:41:01<32:41:47, 5.72s/it]
53%|█████▎ | 23360/43939 [37:41:07<32:40:01, 5.71s/it]
53%|█████▎ | 23361/43939 [37:41:12<32:42:07, 5.72s/it]
53%|█████▎ | 23362/43939 [37:41:18<32:40:25, 5.72s/it]
53%|█████▎ | 23363/43939 [37:41:24<32:37:22, 5.71s/it]
53%|█████▎ | 23364/43939 [37:41:29<32:41:14, 5.72s/it]
53%|█████▎ | 23365/43939 [37:41:35<32:38:37, 5.71s/it]
53%|█████▎ | 23366/43939 [37:41:41<32:41:51, 5.72s/it]
53%|█████▎ | 23367/43939 [37:41:47<32:38:38, 5.71s/it]
53%|█████▎ | 23368/43939 [37:41:52<32:37:43, 5.71s/it]
53%|█████▎ | 23369/43939 [37:41:58<32:36:13, 5.71s/it]
53%|█████▎ | 23370/43939 [37:42:04<32:32:23, 5.70s/it]
53%|█████▎ | 23371/43939 [37:42:09<32:31:55, 5.69s/it]
53%|█████▎ | 23372/43939 [37:42:15<32:32:54, 5.70s/it]
53%|█████▎ | 23373/43939 [37:42:21<32:33:23, 5.70s/it]
53%|█████▎ | 23374/43939 [37:42:26<32:39:00, 5.72s/it]
53%|█████▎ | 23375/43939 [37:42:32<33:01:50, 5.78s/it]
53%|█████▎ | 23376/43939 [37:42:38<32:52:19, 5.75s/it]
53%|█████▎ | 23377/43939 [37:42:44<32:48:18, 5.74s/it]
53%|█████▎ | 23378/43939 [37:42:50<32:43:15, 5.73s/it]
53%|█████▎ | 23379/43939 [37:42:55<32:46:15, 5.74s/it]
53%|█████▎ | 23380/43939 [37:43:01<32:45:56, 5.74s/it]
53%|█████▎ | 23381/43939 [37:43:07<32:46:51, 5.74s/it]
53%|█████▎ | 23382/43939 [37:43:12<32:44:07, 5.73s/it]
53%|█████▎ | 23383/43939 [37:43:18<32:44:53, 5.74s/it]
53%|█████▎ | 23384/43939 [37:43:24<32:47:29, 5.74s/it]
53%|█████▎ | 23385/43939 [37:43:30<32:43:19, 5.73s/it]
53%|█████▎ | 23386/43939 [37:43:35<32:42:12, 5.73s/it]
53%|█████▎ | 23387/43939 [37:43:41<32:44:21, 5.73s/it]
53%|█████▎ | 23388/43939 [37:43:47<32:41:42, 5.73s/it]
53%|█████▎ | 23389/43939 [37:43:53<32:34:54, 5.71s/it]
53%|█████▎ | 23390/43939 [37:43:58<32:31:35, 5.70s/it]
53%|█████▎ | 23391/43939 [37:44:04<32:32:51, 5.70s/it]
53%|█████▎ | 23392/43939 [37:44:10<32:34:24, 5.71s/it]
53%|█████▎ | 23393/43939 [37:44:15<32:32:07, 5.70s/it]
53%|█████▎ | 23394/43939 [37:44:21<32:33:17, 5.70s/it]
53%|█████▎ | 23395/43939 [37:44:27<32:32:01, 5.70s/it]
53%|█████▎ | 23396/43939 [37:44:32<32:29:33, 5.69s/it]
53%|█████▎ | 23397/43939 [37:44:38<32:51:39, 5.76s/it]
53%|█████▎ | 23398/43939 [37:44:44<32:43:46, 5.74s/it]
53%|█████▎ | 23399/43939 [37:44:50<32:36:35, 5.72s/it]
53%|█████▎ | 23400/43939 [37:44:55<32:40:33, 5.73s/it]
{'loss': 0.7888, 'grad_norm': 1.1752251386642456, 'learning_rate': 2.652234792156592e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.93, 'epoch': 2.13}
53%|█████▎ | 23400/43939 [37:44:55<32:40:33, 5.73s/it]
53%|█████▎ | 23401/43939 [37:45:01<32:40:13, 5.73s/it]
53%|█████▎ | 23402/43939 [37:45:07<32:40:20, 5.73s/it]
53%|█████▎ | 23403/43939 [37:45:13<32:36:28, 5.72s/it]
53%|█████▎ | 23404/43939 [37:45:18<32:32:38, 5.71s/it]
53%|█████▎ | 23405/43939 [37:45:24<32:42:35, 5.73s/it]
53%|█████▎ | 23406/43939 [37:45:30<32:38:29, 5.72s/it]
53%|█████▎ | 23407/43939 [37:45:35<32:37:00, 5.72s/it]
53%|█████▎ | 23408/43939 [37:45:41<32:37:38, 5.72s/it]
53%|█████▎ | 23409/43939 [37:45:47<32:37:26, 5.72s/it]
53%|█████▎ | 23410/43939 [37:45:53<32:35:39, 5.72s/it]
53%|█████▎ | 23411/43939 [37:45:58<32:37:37, 5.72s/it]
53%|█████▎ | 23412/43939 [37:46:04<32:38:05, 5.72s/it]
53%|█████▎ | 23413/43939 [37:46:10<32:34:54, 5.71s/it]
53%|█████▎ | 23414/43939 [37:46:15<32:30:22, 5.70s/it]
53%|█████▎ | 23415/43939 [37:46:21<32:30:27, 5.70s/it]
53%|█████▎ | 23416/43939 [37:46:27<32:25:43, 5.69s/it]
53%|█████▎ | 23417/43939 [37:46:33<32:49:24, 5.76s/it]
53%|█████▎ | 23418/43939 [37:46:38<32:46:44, 5.75s/it]
53%|█████▎ | 23419/43939 [37:46:44<32:45:33, 5.75s/it]
53%|█████▎ | 23420/43939 [37:46:50<32:41:23, 5.74s/it]
53%|█████▎ | 23421/43939 [37:46:56<32:37:12, 5.72s/it]
53%|█████▎ | 23422/43939 [37:47:01<32:33:50, 5.71s/it]
53%|█████▎ | 23423/43939 [37:47:07<32:33:25, 5.71s/it]
53%|█████▎ | 23424/43939 [37:47:13<32:29:21, 5.70s/it]
53%|█████▎ | 23425/43939 [37:47:18<32:27:34, 5.70s/it]
53%|█████▎ | 23426/43939 [37:47:24<32:27:42, 5.70s/it]
53%|█████▎ | 23427/43939 [37:47:30<32:28:31, 5.70s/it]
53%|█████▎ | 23428/43939 [37:47:36<32:37:15, 5.73s/it]
53%|█████▎ | 23429/43939 [37:47:41<32:34:32, 5.72s/it]
53%|█████▎ | 23430/43939 [37:47:47<32:34:41, 5.72s/it]
53%|█████▎ | 23431/43939 [37:47:53<32:32:19, 5.71s/it]
53%|█████▎ | 23432/43939 [37:47:58<32:30:59, 5.71s/it]
53%|█████▎ | 23433/43939 [37:48:04<32:25:48, 5.69s/it]
53%|█████▎ | 23434/43939 [37:48:10<32:26:16, 5.70s/it]
53%|█████▎ | 23435/43939 [37:48:15<32:26:40, 5.70s/it]
53%|█████▎ | 23436/43939 [37:48:21<32:27:27, 5.70s/it]
53%|█████▎ | 23437/43939 [37:48:27<32:26:24, 5.70s/it]
53%|█████▎ | 23438/43939 [37:48:33<32:46:58, 5.76s/it]
53%|█████▎ | 23439/43939 [37:48:39<32:49:20, 5.76s/it]
53%|█████▎ | 23440/43939 [37:48:44<32:43:29, 5.75s/it]
53%|█████▎ | 23441/43939 [37:48:50<32:39:13, 5.73s/it]
53%|█████▎ | 23442/43939 [37:48:56<32:36:09, 5.73s/it]
53%|█████▎ | 23443/43939 [37:49:01<32:41:08, 5.74s/it]
53%|█████▎ | 23444/43939 [37:49:07<32:37:07, 5.73s/it]
53%|█████▎ | 23445/43939 [37:49:13<32:36:48, 5.73s/it]
53%|█████▎ | 23446/43939 [37:49:18<32:30:35, 5.71s/it]
53%|█████▎ | 23447/43939 [37:49:24<32:32:52, 5.72s/it]
53%|█████▎ | 23448/43939 [37:49:30<32:33:41, 5.72s/it]
53%|█████▎ | 23449/43939 [37:49:36<32:32:08, 5.72s/it]
53%|█████▎ | 23450/43939 [37:49:41<32:28:11, 5.71s/it]
{'loss': 0.7784, 'grad_norm': 1.3073374032974243, 'learning_rate': 2.642321860655646e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.12, 'epoch': 2.13}
53%|█████▎ | 23450/43939 [37:49:41<32:28:11, 5.71s/it]
53%|█████▎ | 23451/43939 [37:49:47<32:30:11, 5.71s/it]
53%|█████▎ | 23452/43939 [37:49:53<32:30:48, 5.71s/it]
53%|█████▎ | 23453/43939 [37:49:58<32:28:51, 5.71s/it]
53%|█████▎ | 23454/43939 [37:50:04<32:28:03, 5.71s/it]
53%|█████▎ | 23455/43939 [37:50:10<32:25:54, 5.70s/it]
53%|█████▎ | 23456/43939 [37:50:16<32:25:38, 5.70s/it]
53%|█████▎ | 23457/43939 [37:50:21<32:26:05, 5.70s/it]
53%|█████▎ | 23458/43939 [37:50:27<32:30:50, 5.72s/it]
53%|█████▎ | 23459/43939 [37:50:33<32:28:56, 5.71s/it]
53%|█████▎ | 23460/43939 [37:50:39<32:46:29, 5.76s/it]
53%|█████▎ | 23461/43939 [37:50:44<32:38:36, 5.74s/it]
53%|█████▎ | 23462/43939 [37:50:50<32:35:14, 5.73s/it]
53%|█████▎ | 23463/43939 [37:50:56<32:31:08, 5.72s/it]
53%|█████▎ | 23464/43939 [37:51:01<32:28:47, 5.71s/it]
53%|█████▎ | 23465/43939 [37:51:07<32:25:19, 5.70s/it]
53%|█████▎ | 23466/43939 [37:51:13<32:22:26, 5.69s/it]
53%|█████▎ | 23467/43939 [37:51:18<32:25:39, 5.70s/it]
53%|█████▎ | 23468/43939 [37:51:24<32:30:41, 5.72s/it]
53%|█████▎ | 23469/43939 [37:51:30<32:33:10, 5.72s/it]
53%|█████▎ | 23470/43939 [37:51:36<32:31:55, 5.72s/it]
53%|█████▎ | 23471/43939 [37:51:41<32:31:16, 5.72s/it]
53%|█████▎ | 23472/43939 [37:51:47<32:27:07, 5.71s/it]
53%|█████▎ | 23473/43939 [37:51:53<32:32:06, 5.72s/it]
53%|█████▎ | 23474/43939 [37:51:59<32:36:21, 5.74s/it]
53%|█████▎ | 23475/43939 [37:52:04<32:28:56, 5.71s/it]
53%|█████▎ | 23476/43939 [37:52:10<32:26:36, 5.71s/it]
53%|█████▎ | 23477/43939 [37:52:16<32:23:16, 5.70s/it]
53%|█████▎ | 23478/43939 [37:52:21<32:24:53, 5.70s/it]
53%|█████▎ | 23479/43939 [37:52:27<32:24:45, 5.70s/it]
53%|█████▎ | 23480/43939 [37:52:33<32:25:41, 5.71s/it]
53%|█████▎ | 23481/43939 [37:52:39<32:52:41, 5.79s/it]
53%|█████▎ | 23482/43939 [37:52:44<32:45:42, 5.77s/it]
53%|█████▎ | 23483/43939 [37:52:50<32:39:36, 5.75s/it]
53%|█████▎ | 23484/43939 [37:52:56<32:31:54, 5.73s/it]
53%|█████▎ | 23485/43939 [37:53:01<32:26:16, 5.71s/it]
53%|█████▎ | 23486/43939 [37:53:07<32:26:45, 5.71s/it]
53%|█████▎ | 23487/43939 [37:53:13<32:25:13, 5.71s/it]
53%|█████▎ | 23488/43939 [37:53:19<32:27:34, 5.71s/it]
53%|█████▎ | 23489/43939 [37:53:24<32:22:56, 5.70s/it]
53%|█████▎ | 23490/43939 [37:53:30<32:25:38, 5.71s/it]
53%|█████▎ | 23491/43939 [37:53:36<32:25:26, 5.71s/it]
53%|█████▎ | 23492/43939 [37:53:41<32:27:59, 5.72s/it]
53%|█████▎ | 23493/43939 [37:53:47<32:27:40, 5.72s/it]
53%|█████▎ | 23494/43939 [37:53:53<32:27:01, 5.71s/it]
53%|█████▎ | 23495/43939 [37:53:59<32:27:46, 5.72s/it]
53%|█████▎ | 23496/43939 [37:54:04<32:25:56, 5.71s/it]
53%|█████▎ | 23497/43939 [37:54:10<32:25:46, 5.71s/it]
53%|█████▎ | 23498/43939 [37:54:16<32:23:24, 5.70s/it]
53%|█████▎ | 23499/43939 [37:54:21<32:26:57, 5.72s/it]
53%|█████▎ | 23500/43939 [37:54:27<32:25:01, 5.71s/it]
{'loss': 0.7972, 'grad_norm': 1.4693443775177002, 'learning_rate': 2.6324066836928964e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.86, 'epoch': 2.14}
53%|█████▎ | 23500/43939 [37:54:27<32:25:01, 5.71s/it]
53%|█████▎ | 23501/43939 [37:54:33<32:26:21, 5.71s/it]
53%|█████▎ | 23502/43939 [37:54:39<32:50:30, 5.79s/it]
53%|█████▎ | 23503/43939 [37:54:45<32:45:55, 5.77s/it]
53%|█████▎ | 23504/43939 [37:54:50<32:35:39, 5.74s/it]
53%|█████▎ | 23505/43939 [37:54:56<32:29:57, 5.73s/it]
53%|█████▎ | 23506/43939 [37:55:02<32:24:45, 5.71s/it]
53%|█████▎ | 23507/43939 [37:55:07<32:23:11, 5.71s/it]
54%|█████▎ | 23508/43939 [37:55:13<32:26:51, 5.72s/it]
54%|█████▎ | 23509/43939 [37:55:19<32:22:56, 5.71s/it]
54%|█████▎ | 23510/43939 [37:55:24<32:23:12, 5.71s/it]
54%|█████▎ | 23511/43939 [37:55:30<32:21:16, 5.70s/it]
54%|█████▎ | 23512/43939 [37:55:36<32:19:00, 5.70s/it]
54%|█████▎ | 23513/43939 [37:55:41<32:17:55, 5.69s/it]
54%|█████▎ | 23514/43939 [37:55:47<32:19:13, 5.70s/it]
54%|█████▎ | 23515/43939 [37:55:53<32:20:43, 5.70s/it]
54%|█████▎ | 23516/43939 [37:55:59<32:20:35, 5.70s/it]
54%|█████▎ | 23517/43939 [37:56:04<32:19:57, 5.70s/it]
54%|█████▎ | 23518/43939 [37:56:10<32:20:05, 5.70s/it]
54%|█████▎ | 23519/43939 [37:56:16<32:24:57, 5.71s/it]
54%|█████▎ | 23520/43939 [37:56:21<32:24:07, 5.71s/it]
54%|█████▎ | 23521/43939 [37:56:27<32:22:24, 5.71s/it]
54%|█████▎ | 23522/43939 [37:56:33<32:22:34, 5.71s/it]
54%|█████▎ | 23523/43939 [37:56:39<32:20:34, 5.70s/it]
54%|█████▎ | 23524/43939 [37:56:45<32:49:22, 5.79s/it]
54%|█████▎ | 23525/43939 [37:56:50<32:41:47, 5.77s/it]
54%|█████▎ | 23526/43939 [37:56:56<32:35:43, 5.75s/it]
54%|█████▎ | 23527/43939 [37:57:02<32:33:35, 5.74s/it]
54%|█████▎ | 23528/43939 [37:57:07<32:23:48, 5.71s/it]
54%|█████▎ | 23529/43939 [37:57:13<32:25:09, 5.72s/it]
54%|█████▎ | 23530/43939 [37:57:19<32:21:41, 5.71s/it]
54%|█████▎ | 23531/43939 [37:57:24<32:21:57, 5.71s/it]
54%|█████▎ | 23532/43939 [37:57:30<32:20:04, 5.70s/it]
54%|█████▎ | 23533/43939 [37:57:36<32:20:02, 5.70s/it]
54%|█████▎ | 23534/43939 [37:57:42<32:18:54, 5.70s/it]
54%|█████▎ | 23535/43939 [37:57:47<32:16:08, 5.69s/it]
54%|█████▎ | 23536/43939 [37:57:53<32:17:50, 5.70s/it]
54%|█████▎ | 23537/43939 [37:57:59<32:18:41, 5.70s/it]
54%|█████▎ | 23538/43939 [37:58:04<32:19:47, 5.70s/it]
54%|█████▎ | 23539/43939 [37:58:10<32:24:16, 5.72s/it]
54%|█████▎ | 23540/43939 [37:58:16<32:24:44, 5.72s/it]
54%|█████▎ | 23541/43939 [37:58:22<32:20:33, 5.71s/it]
54%|█████▎ | 23542/43939 [37:58:27<32:20:53, 5.71s/it]
54%|█████▎ | 23543/43939 [37:58:33<32:21:28, 5.71s/it]
54%|█████▎ | 23544/43939 [37:58:39<32:15:35, 5.69s/it]
54%|█████▎ | 23545/43939 [37:58:45<32:37:38, 5.76s/it]
54%|█████▎ | 23546/43939 [37:58:50<32:31:54, 5.74s/it]
54%|█████▎ | 23547/43939 [37:58:56<32:26:15, 5.73s/it]
54%|█████▎ | 23548/43939 [37:59:02<32:21:34, 5.71s/it]
54%|█████▎ | 23549/43939 [37:59:07<32:22:27, 5.72s/it]
54%|█████▎ | 23550/43939 [37:59:13<32:21:06, 5.71s/it]
{'loss': 0.768, 'grad_norm': 2.2674715518951416, 'learning_rate': 2.6224894177035602e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.52, 'epoch': 2.14}
54%|█████▎ | 23550/43939 [37:59:13<32:21:06, 5.71s/it]
54%|█████▎ | 23551/43939 [37:59:19<32:19:50, 5.71s/it]
54%|█████▎ | 23552/43939 [37:59:24<32:14:15, 5.69s/it]
54%|█████▎ | 23553/43939 [37:59:30<32:15:15, 5.70s/it]
54%|█████▎ | 23554/43939 [37:59:36<32:17:50, 5.70s/it]
54%|█████▎ | 23555/43939 [37:59:41<32:11:52, 5.69s/it]
54%|█████▎ | 23556/43939 [37:59:47<32:14:18, 5.69s/it]
54%|█████▎ | 23557/43939 [37:59:53<32:18:55, 5.71s/it]
54%|█████▎ | 23558/43939 [37:59:59<32:18:15, 5.71s/it]
54%|█████▎ | 23559/43939 [38:00:04<32:19:48, 5.71s/it]
54%|█████▎ | 23560/43939 [38:00:10<32:21:11, 5.72s/it]
54%|█████▎ | 23561/43939 [38:00:16<32:16:29, 5.70s/it]
54%|█████▎ | 23562/43939 [38:00:21<32:13:27, 5.69s/it]
54%|█████▎ | 23563/43939 [38:00:27<32:12:09, 5.69s/it]
54%|█████▎ | 23564/43939 [38:00:33<32:14:18, 5.70s/it]
54%|█████▎ | 23565/43939 [38:00:38<32:12:43, 5.69s/it]
54%|█████▎ | 23566/43939 [38:00:44<32:36:55, 5.76s/it]
54%|█████▎ | 23567/43939 [38:00:50<32:29:32, 5.74s/it]
54%|█████▎ | 23568/43939 [38:00:56<32:22:36, 5.72s/it]
54%|█████▎ | 23569/43939 [38:01:02<32:25:40, 5.73s/it]
54%|█████▎ | 23570/43939 [38:01:07<32:24:51, 5.73s/it]
54%|█████▎ | 23571/43939 [38:01:13<32:26:14, 5.73s/it]
54%|█████▎ | 23572/43939 [38:01:19<32:21:29, 5.72s/it]
54%|█████▎ | 23573/43939 [38:01:24<32:26:24, 5.73s/it]
54%|█████▎ | 23574/43939 [38:01:30<32:21:10, 5.72s/it]
54%|█████▎ | 23575/43939 [38:01:36<32:20:16, 5.72s/it]
54%|█████▎ | 23576/43939 [38:01:42<32:20:10, 5.72s/it]
54%|█████▎ | 23577/43939 [38:01:47<32:21:08, 5.72s/it]
54%|█████▎ | 23578/43939 [38:01:53<32:22:14, 5.72s/it]
54%|█████▎ | 23579/43939 [38:01:59<32:18:08, 5.71s/it]
54%|█████▎ | 23580/43939 [38:02:04<32:15:17, 5.70s/it]
54%|█████▎ | 23581/43939 [38:02:10<32:16:36, 5.71s/it]
54%|█████▎ | 23582/43939 [38:02:16<32:13:17, 5.70s/it]
54%|█████▎ | 23583/43939 [38:02:22<32:18:12, 5.71s/it]
54%|█████▎ | 23584/43939 [38:02:27<32:14:41, 5.70s/it]
54%|█████▎ | 23585/43939 [38:02:33<32:15:08, 5.70s/it]
54%|█████▎ | 23586/43939 [38:02:39<32:20:43, 5.72s/it]
54%|█████▎ | 23587/43939 [38:02:45<32:40:39, 5.78s/it]
54%|█████▎ | 23588/43939 [38:02:50<32:29:23, 5.75s/it]
54%|█████▎ | 23589/43939 [38:02:56<32:22:14, 5.73s/it]
54%|█████▎ | 23590/43939 [38:03:02<32:21:01, 5.72s/it]
54%|█████▎ | 23591/43939 [38:03:07<32:18:18, 5.72s/it]
54%|█████▎ | 23592/43939 [38:03:13<32:18:04, 5.72s/it]
54%|█████▎ | 23593/43939 [38:03:19<32:12:59, 5.70s/it]
54%|█████▎ | 23594/43939 [38:03:24<32:13:58, 5.70s/it]
54%|█████▎ | 23595/43939 [38:03:30<32:10:17, 5.69s/it]
54%|█████▎ | 23596/43939 [38:03:36<32:12:47, 5.70s/it]
54%|█████▎ | 23597/43939 [38:03:42<32:12:24, 5.70s/it]
54%|█████▎ | 23598/43939 [38:03:47<32:14:49, 5.71s/it]
54%|█████▎ | 23599/43939 [38:03:53<32:14:16, 5.71s/it]
54%|█████▎ | 23600/43939 [38:03:59<32:15:35, 5.71s/it]
{'loss': 0.788, 'grad_norm': 1.593753457069397, 'learning_rate': 2.6125702191558132e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.07, 'epoch': 2.15}
54%|█████▎ | 23600/43939 [38:03:59<32:15:35, 5.71s/it]
54%|█████▎ | 23601/43939 [38:04:04<32:19:20, 5.72s/it]
54%|█████▎ | 23602/43939 [38:04:10<32:21:25, 5.73s/it]
54%|█████▎ | 23603/43939 [38:04:16<32:20:52, 5.73s/it]
54%|█████▎ | 23604/43939 [38:04:22<32:25:14, 5.74s/it]
54%|█████▎ | 23605/43939 [38:04:27<32:19:50, 5.72s/it]
54%|█████▎ | 23606/43939 [38:04:33<32:16:13, 5.71s/it]
54%|█████▎ | 23607/43939 [38:04:39<32:13:51, 5.71s/it]
54%|█████▎ | 23608/43939 [38:04:45<32:36:02, 5.77s/it]
54%|█████▎ | 23609/43939 [38:04:50<32:29:30, 5.75s/it]
54%|█████▎ | 23610/43939 [38:04:56<32:32:08, 5.76s/it]
54%|█████▎ | 23611/43939 [38:05:02<32:29:29, 5.75s/it]
54%|█████▎ | 23612/43939 [38:05:08<32:28:15, 5.75s/it]
54%|█████▎ | 23613/43939 [38:05:13<32:21:36, 5.73s/it]
54%|█████▎ | 23614/43939 [38:05:19<32:18:12, 5.72s/it]
54%|█████▎ | 23615/43939 [38:05:25<32:15:43, 5.71s/it]
54%|█████▎ | 23616/43939 [38:05:30<32:15:42, 5.71s/it]
54%|█████▎ | 23617/43939 [38:05:36<32:12:45, 5.71s/it]
54%|█████▍ | 23618/43939 [38:05:42<32:10:19, 5.70s/it]
54%|█████▍ | 23619/43939 [38:05:47<32:11:03, 5.70s/it]
54%|█████▍ | 23620/43939 [38:05:53<32:11:06, 5.70s/it]
54%|█████▍ | 23621/43939 [38:05:59<32:07:13, 5.69s/it]
54%|█████▍ | 23622/43939 [38:06:05<32:14:38, 5.71s/it]
54%|█████▍ | 23623/43939 [38:06:10<32:17:09, 5.72s/it]
54%|█████▍ | 23624/43939 [38:06:16<32:17:39, 5.72s/it]
54%|█████▍ | 23625/43939 [38:06:22<32:15:05, 5.72s/it]
54%|█████▍ | 23626/43939 [38:06:27<32:13:18, 5.71s/it]
54%|█████▍ | 23627/43939 [38:06:33<32:18:54, 5.73s/it]
54%|█████▍ | 23628/43939 [38:06:39<32:19:13, 5.73s/it]
54%|█████▍ | 23629/43939 [38:06:45<32:16:52, 5.72s/it]
54%|█████▍ | 23630/43939 [38:06:51<32:35:22, 5.78s/it]
54%|█████▍ | 23631/43939 [38:06:56<32:29:51, 5.76s/it]
54%|█████▍ | 23632/43939 [38:07:02<32:27:20, 5.75s/it]
54%|█████▍ | 23633/43939 [38:07:08<32:21:54, 5.74s/it]
54%|█████▍ | 23634/43939 [38:07:13<32:20:21, 5.73s/it]
54%|█████▍ | 23635/43939 [38:07:19<32:16:32, 5.72s/it]
54%|█████▍ | 23636/43939 [38:07:25<32:13:26, 5.71s/it]
54%|█████▍ | 23637/43939 [38:07:31<32:12:43, 5.71s/it]
54%|█████▍ | 23638/43939 [38:07:36<32:11:15, 5.71s/it]
54%|█████▍ | 23639/43939 [38:07:42<32:11:16, 5.71s/it]
54%|█████▍ | 23640/43939 [38:07:48<32:12:54, 5.71s/it]
54%|█████▍ | 23641/43939 [38:07:53<32:10:03, 5.71s/it]
54%|█████▍ | 23642/43939 [38:07:59<32:08:32, 5.70s/it]
54%|█████▍ | 23643/43939 [38:08:05<32:07:22, 5.70s/it]
54%|█████▍ | 23644/43939 [38:08:10<32:08:15, 5.70s/it]
54%|█████▍ | 23645/43939 [38:08:16<32:06:27, 5.70s/it]
54%|█████▍ | 23646/43939 [38:08:22<32:10:04, 5.71s/it]
54%|█████▍ | 23647/43939 [38:08:28<32:08:08, 5.70s/it]
54%|█████▍ | 23648/43939 [38:08:33<32:07:19, 5.70s/it]
54%|█████▍ | 23649/43939 [38:08:39<32:05:07, 5.69s/it]
54%|█████▍ | 23650/43939 [38:08:45<32:02:45, 5.69s/it]
{'loss': 0.7793, 'grad_norm': 1.3500062227249146, 'learning_rate': 2.602649244548322e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.81, 'epoch': 2.15}
54%|█████▍ | 23650/43939 [38:08:45<32:02:45, 5.69s/it]
54%|█████▍ | 23651/43939 [38:08:51<32:24:12, 5.75s/it]
54%|█████▍ | 23652/43939 [38:08:56<32:18:21, 5.73s/it]
54%|█████▍ | 23653/43939 [38:09:02<32:11:14, 5.71s/it]
54%|█████▍ | 23654/43939 [38:09:08<32:11:45, 5.71s/it]
54%|█████▍ | 23655/43939 [38:09:13<32:10:06, 5.71s/it]
54%|█████▍ | 23656/43939 [38:09:19<32:13:01, 5.72s/it]
54%|█████▍ | 23657/43939 [38:09:25<32:11:42, 5.71s/it]
54%|█████▍ | 23658/43939 [38:09:30<32:06:42, 5.70s/it]
54%|█████▍ | 23659/43939 [38:09:36<32:02:07, 5.69s/it]
54%|█████▍ | 23660/43939 [38:09:42<32:06:18, 5.70s/it]
54%|█████▍ | 23661/43939 [38:09:48<32:07:25, 5.70s/it]
54%|█████▍ | 23662/43939 [38:09:53<32:06:58, 5.70s/it]
54%|█████▍ | 23663/43939 [38:09:59<32:04:54, 5.70s/it]
54%|█████▍ | 23664/43939 [38:10:05<32:08:34, 5.71s/it]
54%|█████▍ | 23665/43939 [38:10:10<32:08:09, 5.71s/it]
54%|█████▍ | 23666/43939 [38:10:16<32:06:57, 5.70s/it]
54%|█████▍ | 23667/43939 [38:10:22<32:06:16, 5.70s/it]
54%|█████▍ | 23668/43939 [38:10:27<32:07:29, 5.71s/it]
54%|█████▍ | 23669/43939 [38:10:33<32:07:24, 5.71s/it]
54%|█████▍ | 23670/43939 [38:10:39<32:04:21, 5.70s/it]
54%|█████▍ | 23671/43939 [38:10:44<32:02:42, 5.69s/it]
54%|█████▍ | 23672/43939 [38:10:50<32:24:10, 5.76s/it]
54%|█████▍ | 23673/43939 [38:10:56<32:24:13, 5.76s/it]
54%|█████▍ | 23674/43939 [38:11:02<32:20:54, 5.75s/it]
54%|█████▍ | 23675/43939 [38:11:08<32:16:11, 5.73s/it]
54%|█████▍ | 23676/43939 [38:11:13<32:12:27, 5.72s/it]
54%|█████▍ | 23677/43939 [38:11:19<32:08:48, 5.71s/it]
54%|█████▍ | 23678/43939 [38:11:25<32:10:19, 5.72s/it]
54%|█████▍ | 23679/43939 [38:11:30<32:03:40, 5.70s/it]
54%|█████▍ | 23680/43939 [38:11:36<32:01:35, 5.69s/it]
54%|█████▍ | 23681/43939 [38:11:42<32:01:33, 5.69s/it]
54%|█████▍ | 23682/43939 [38:11:47<32:03:34, 5.70s/it]
54%|█████▍ | 23683/43939 [38:11:53<32:01:07, 5.69s/it]
54%|█████▍ | 23684/43939 [38:11:59<32:01:58, 5.69s/it]
54%|█████▍ | 23685/43939 [38:12:05<32:06:18, 5.71s/it]
54%|█████▍ | 23686/43939 [38:12:10<32:04:56, 5.70s/it]
54%|█████▍ | 23687/43939 [38:12:16<32:07:03, 5.71s/it]
54%|█████▍ | 23688/43939 [38:12:22<32:05:15, 5.70s/it]
54%|█████▍ | 23689/43939 [38:12:27<32:05:08, 5.70s/it]
54%|█████▍ | 23690/43939 [38:12:33<32:00:44, 5.69s/it]
54%|█████▍ | 23691/43939 [38:12:39<32:07:17, 5.71s/it]
54%|█████▍ | 23692/43939 [38:12:44<32:07:12, 5.71s/it]
54%|█████▍ | 23693/43939 [38:12:50<32:09:01, 5.72s/it]
54%|█████▍ | 23694/43939 [38:12:56<32:29:05, 5.78s/it]
54%|█████▍ | 23695/43939 [38:13:02<32:17:14, 5.74s/it]
54%|█████▍ | 23696/43939 [38:13:08<32:14:26, 5.73s/it]
54%|█████▍ | 23697/43939 [38:13:13<32:12:05, 5.73s/it]
54%|█████▍ | 23698/43939 [38:13:19<32:07:09, 5.71s/it]
54%|█████▍ | 23699/43939 [38:13:25<32:03:19, 5.70s/it]
54%|█████▍ | 23700/43939 [38:13:30<32:03:34, 5.70s/it]
{'loss': 0.7698, 'grad_norm': 1.6638089418411255, 'learning_rate': 2.592726650407777e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.88, 'epoch': 2.16}
54%|█████▍ | 23700/43939 [38:13:30<32:03:34, 5.70s/it]
54%|█████▍ | 23701/43939 [38:13:36<32:06:40, 5.71s/it]
54%|█████▍ | 23702/43939 [38:13:42<32:06:01, 5.71s/it]
54%|█████▍ | 23703/43939 [38:13:47<32:06:53, 5.71s/it]
54%|█████▍ | 23704/43939 [38:13:53<32:05:16, 5.71s/it]
54%|█████▍ | 23705/43939 [38:13:59<32:03:26, 5.70s/it]
54%|█████▍ | 23706/43939 [38:14:05<32:04:34, 5.71s/it]
54%|█████▍ | 23707/43939 [38:14:10<32:07:06, 5.72s/it]
54%|█████▍ | 23708/43939 [38:14:16<32:02:00, 5.70s/it]
54%|█████▍ | 23709/43939 [38:14:22<32:00:37, 5.70s/it]
54%|█████▍ | 23710/43939 [38:14:27<32:01:35, 5.70s/it]
54%|█████▍ | 23711/43939 [38:14:33<32:07:38, 5.72s/it]
54%|█████▍ | 23712/43939 [38:14:39<32:06:51, 5.72s/it]
54%|█████▍ | 23713/43939 [38:14:44<32:02:20, 5.70s/it]
54%|█████▍ | 23714/43939 [38:14:50<32:02:23, 5.70s/it]
54%|█████▍ | 23715/43939 [38:14:56<32:19:01, 5.75s/it]
54%|█████▍ | 23716/43939 [38:15:02<32:21:16, 5.76s/it]
54%|█████▍ | 23717/43939 [38:15:08<32:19:01, 5.75s/it]
54%|█████▍ | 23718/43939 [38:15:13<32:14:04, 5.74s/it]
54%|█████▍ | 23719/43939 [38:15:19<32:06:30, 5.72s/it]
54%|█████▍ | 23720/43939 [38:15:25<32:06:13, 5.72s/it]
54%|█████▍ | 23721/43939 [38:15:30<32:02:31, 5.71s/it]
54%|█████▍ | 23722/43939 [38:15:36<31:59:10, 5.70s/it]
54%|█████▍ | 23723/43939 [38:15:42<31:56:04, 5.69s/it]
54%|█████▍ | 23724/43939 [38:15:47<32:00:11, 5.70s/it]
54%|█████▍ | 23725/43939 [38:15:53<31:59:55, 5.70s/it]
54%|█████▍ | 23726/43939 [38:15:59<31:58:15, 5.69s/it]
54%|█████▍ | 23727/43939 [38:16:05<32:01:10, 5.70s/it]
54%|█████▍ | 23728/43939 [38:16:10<32:02:12, 5.71s/it]
54%|█████▍ | 23729/43939 [38:16:16<32:02:41, 5.71s/it]
54%|█████▍ | 23730/43939 [38:16:22<32:04:47, 5.71s/it]
54%|█████▍ | 23731/43939 [38:16:27<32:06:40, 5.72s/it]
54%|█████▍ | 23732/43939 [38:16:33<32:02:10, 5.71s/it]
54%|█████▍ | 23733/43939 [38:16:39<32:04:54, 5.72s/it]
54%|█████▍ | 23734/43939 [38:16:44<32:01:02, 5.70s/it]
54%|█████▍ | 23735/43939 [38:16:50<31:59:36, 5.70s/it]
54%|█████▍ | 23736/43939 [38:16:56<32:19:47, 5.76s/it]
54%|█████▍ | 23737/43939 [38:17:02<32:18:36, 5.76s/it]
54%|█████▍ | 23738/43939 [38:17:08<32:16:53, 5.75s/it]
54%|█████▍ | 23739/43939 [38:17:13<32:11:47, 5.74s/it]
54%|█████▍ | 23740/43939 [38:17:19<32:14:12, 5.75s/it]
54%|█████▍ | 23741/43939 [38:17:25<32:14:21, 5.75s/it]
54%|█████▍ | 23742/43939 [38:17:30<32:08:04, 5.73s/it]
54%|█████▍ | 23743/43939 [38:17:36<32:06:03, 5.72s/it]
54%|█████▍ | 23744/43939 [38:17:42<32:00:22, 5.71s/it]
54%|█████▍ | 23745/43939 [38:17:48<31:59:36, 5.70s/it]
54%|█████▍ | 23746/43939 [38:17:53<31:55:38, 5.69s/it]
54%|█████▍ | 23747/43939 [38:17:59<31:59:40, 5.70s/it]
54%|█████▍ | 23748/43939 [38:18:05<31:57:22, 5.70s/it]
54%|█████▍ | 23749/43939 [38:18:10<32:03:23, 5.72s/it]
54%|█████▍ | 23750/43939 [38:18:16<32:05:19, 5.72s/it]
{'loss': 0.7889, 'grad_norm': 1.292625904083252, 'learning_rate': 2.5828025932864164e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.5, 'epoch': 2.16}
54%|█████▍ | 23750/43939 [38:18:16<32:05:19, 5.72s/it]
54%|█████▍ | 23751/43939 [38:18:22<32:05:23, 5.72s/it]
54%|█████▍ | 23752/43939 [38:18:27<31:58:42, 5.70s/it]
54%|█████▍ | 23753/43939 [38:18:33<31:56:58, 5.70s/it]
54%|█████▍ | 23754/43939 [38:18:39<31:54:50, 5.69s/it]
54%|█████▍ | 23755/43939 [38:18:45<31:54:12, 5.69s/it]
54%|█████▍ | 23756/43939 [38:18:50<31:53:47, 5.69s/it]
54%|█████▍ | 23757/43939 [38:18:56<32:16:10, 5.76s/it]
54%|█████▍ | 23758/43939 [38:19:02<32:16:01, 5.76s/it]
54%|█████▍ | 23759/43939 [38:19:08<32:09:38, 5.74s/it]
54%|█████▍ | 23760/43939 [38:19:13<32:01:06, 5.71s/it]
54%|█████▍ | 23761/43939 [38:19:19<31:58:26, 5.70s/it]
54%|█████▍ | 23762/43939 [38:19:25<31:58:36, 5.71s/it]
54%|█████▍ | 23763/43939 [38:19:30<31:59:58, 5.71s/it]
54%|█████▍ | 23764/43939 [38:19:36<31:57:06, 5.70s/it]
54%|█████▍ | 23765/43939 [38:19:42<31:54:14, 5.69s/it]
54%|█████▍ | 23766/43939 [38:19:47<31:58:37, 5.71s/it]
54%|█████▍ | 23767/43939 [38:19:53<31:52:11, 5.69s/it]
54%|█████▍ | 23768/43939 [38:19:59<31:52:15, 5.69s/it]
54%|█████▍ | 23769/43939 [38:20:04<31:52:44, 5.69s/it]
54%|█████▍ | 23770/43939 [38:20:10<31:57:07, 5.70s/it]
54%|█████▍ | 23771/43939 [38:20:16<31:55:03, 5.70s/it]
54%|█████▍ | 23772/43939 [38:20:22<31:57:47, 5.71s/it]
54%|█████▍ | 23773/43939 [38:20:27<31:58:01, 5.71s/it]
54%|█████▍ | 23774/43939 [38:20:33<31:57:49, 5.71s/it]
54%|█████▍ | 23775/43939 [38:20:39<31:57:00, 5.70s/it]
54%|█████▍ | 23776/43939 [38:20:44<31:58:12, 5.71s/it]
54%|█████▍ | 23777/43939 [38:20:50<31:58:15, 5.71s/it]
54%|█████▍ | 23778/43939 [38:20:56<31:56:52, 5.70s/it]
54%|█████▍ | 23779/43939 [38:21:02<32:18:21, 5.77s/it]
54%|█████▍ | 23780/43939 [38:21:08<32:13:55, 5.76s/it]
54%|█████▍ | 23781/43939 [38:21:13<32:07:47, 5.74s/it]
54%|█████▍ | 23782/43939 [38:21:19<32:08:40, 5.74s/it]
54%|█████▍ | 23783/43939 [38:21:25<32:02:50, 5.72s/it]
54%|█████▍ | 23784/43939 [38:21:30<32:01:26, 5.72s/it]
54%|█████▍ | 23785/43939 [38:21:36<32:00:13, 5.72s/it]
54%|█████▍ | 23786/43939 [38:21:42<32:01:28, 5.72s/it]
54%|█████▍ | 23787/43939 [38:21:47<31:59:26, 5.71s/it]
54%|█████▍ | 23788/43939 [38:21:53<31:57:21, 5.71s/it]
54%|█████▍ | 23789/43939 [38:21:59<31:58:55, 5.71s/it]
54%|█████▍ | 23790/43939 [38:22:05<32:00:23, 5.72s/it]
54%|█████▍ | 23791/43939 [38:22:10<31:59:55, 5.72s/it]
54%|█████▍ | 23792/43939 [38:22:16<31:55:15, 5.70s/it]
54%|█████▍ | 23793/43939 [38:22:22<31:52:50, 5.70s/it]
54%|█████▍ | 23794/43939 [38:22:27<31:53:38, 5.70s/it]
54%|█████▍ | 23795/43939 [38:22:33<31:51:10, 5.69s/it]
54%|█████▍ | 23796/43939 [38:22:39<31:52:40, 5.70s/it]
54%|█████▍ | 23797/43939 [38:22:45<31:53:15, 5.70s/it]
54%|█████▍ | 23798/43939 [38:22:50<31:54:14, 5.70s/it]
54%|█████▍ | 23799/43939 [38:22:56<32:13:42, 5.76s/it]
54%|█████▍ | 23800/43939 [38:23:02<32:08:39, 5.75s/it]
{'loss': 0.765, 'grad_norm': 1.1630120277404785, 'learning_rate': 2.572877229759563e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.94, 'epoch': 2.17}
54%|█████▍ | 23800/43939 [38:23:02<32:08:39, 5.75s/it]
54%|█████▍ | 23801/43939 [38:23:08<32:04:09, 5.73s/it]
54%|█████▍ | 23802/43939 [38:23:13<32:01:32, 5.73s/it]
54%|█████▍ | 23803/43939 [38:23:19<31:58:48, 5.72s/it]
54%|█████▍ | 23804/43939 [38:23:25<31:55:36, 5.71s/it]
54%|█████▍ | 23805/43939 [38:23:30<31:54:11, 5.70s/it]
54%|█████▍ | 23806/43939 [38:23:36<31:56:33, 5.71s/it]
54%|█████▍ | 23807/43939 [38:23:42<31:50:33, 5.69s/it]
54%|█████▍ | 23808/43939 [38:23:47<31:53:47, 5.70s/it]
54%|█████▍ | 23809/43939 [38:23:53<32:01:54, 5.73s/it]
54%|█████▍ | 23810/43939 [38:23:59<31:58:24, 5.72s/it]
54%|█████▍ | 23811/43939 [38:24:05<31:59:00, 5.72s/it]
54%|█████▍ | 23812/43939 [38:24:10<31:54:24, 5.71s/it]
54%|█████▍ | 23813/43939 [38:24:16<31:57:26, 5.72s/it]
54%|█████▍ | 23814/43939 [38:24:22<31:56:55, 5.72s/it]
54%|█████▍ | 23815/43939 [38:24:27<31:51:46, 5.70s/it]
54%|█████▍ | 23816/43939 [38:24:33<31:49:59, 5.69s/it]
54%|█████▍ | 23817/43939 [38:24:39<31:51:55, 5.70s/it]
54%|█████▍ | 23818/43939 [38:24:45<31:52:29, 5.70s/it]
54%|█████▍ | 23819/43939 [38:24:50<31:50:44, 5.70s/it]
54%|█████▍ | 23820/43939 [38:24:56<32:10:23, 5.76s/it]
54%|█████▍ | 23821/43939 [38:25:02<32:10:01, 5.76s/it]
54%|█████▍ | 23822/43939 [38:25:08<32:04:39, 5.74s/it]
54%|█████▍ | 23823/43939 [38:25:13<31:58:47, 5.72s/it]
54%|█████▍ | 23824/43939 [38:25:19<31:57:06, 5.72s/it]
54%|█████▍ | 23825/43939 [38:25:25<31:56:42, 5.72s/it]
54%|█████▍ | 23826/43939 [38:25:30<32:01:37, 5.73s/it]
54%|█████▍ | 23827/43939 [38:25:36<31:58:54, 5.72s/it]
54%|█████▍ | 23828/43939 [38:25:42<31:58:55, 5.73s/it]
54%|█████▍ | 23829/43939 [38:25:48<32:00:18, 5.73s/it]
54%|█████▍ | 23830/43939 [38:25:53<31:57:20, 5.72s/it]
54%|█████▍ | 23831/43939 [38:25:59<31:52:29, 5.71s/it]
54%|█████▍ | 23832/43939 [38:26:05<31:52:31, 5.71s/it]
54%|█████▍ | 23833/43939 [38:26:10<32:02:06, 5.74s/it]
54%|█████▍ | 23834/43939 [38:26:16<31:59:11, 5.73s/it]
54%|█████▍ | 23835/43939 [38:26:22<31:59:26, 5.73s/it]
54%|█████▍ | 23836/43939 [38:26:28<32:04:11, 5.74s/it]
54%|█████▍ | 23837/43939 [38:26:33<31:57:55, 5.72s/it]
54%|█████▍ | 23838/43939 [38:26:39<31:55:06, 5.72s/it]
54%|█████▍ | 23839/43939 [38:26:45<31:50:52, 5.70s/it]
54%|█████▍ | 23840/43939 [38:26:50<31:50:12, 5.70s/it]
54%|█████▍ | 23841/43939 [38:26:56<31:50:20, 5.70s/it]
54%|█████▍ | 23842/43939 [38:27:02<32:08:48, 5.76s/it]
54%|█████▍ | 23843/43939 [38:27:08<32:04:22, 5.75s/it]
54%|█████▍ | 23844/43939 [38:27:13<32:01:02, 5.74s/it]
54%|█████▍ | 23845/43939 [38:27:19<31:57:06, 5.72s/it]
54%|█████▍ | 23846/43939 [38:27:25<31:54:38, 5.72s/it]
54%|█████▍ | 23847/43939 [38:27:31<31:54:22, 5.72s/it]
54%|█████▍ | 23848/43939 [38:27:36<31:51:03, 5.71s/it]
54%|█████▍ | 23849/43939 [38:27:42<31:50:25, 5.71s/it]
54%|█████▍ | 23850/43939 [38:27:48<31:52:55, 5.71s/it]
{'loss': 0.7917, 'grad_norm': 1.2248200178146362, 'learning_rate': 2.562950716423152e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 505.26, 'epoch': 2.17}
54%|█████▍ | 23850/43939 [38:27:48<31:52:55, 5.71s/it]
54%|█████▍ | 23851/43939 [38:27:53<31:59:02, 5.73s/it]
54%|█████▍ | 23852/43939 [38:27:59<31:55:20, 5.72s/it]
54%|█████▍ | 23853/43939 [38:28:05<31:52:49, 5.71s/it]
54%|█████▍ | 23854/43939 [38:28:11<31:52:15, 5.71s/it]
54%|█████▍ | 23855/43939 [38:28:16<31:50:33, 5.71s/it]
54%|█████▍ | 23856/43939 [38:28:22<31:48:20, 5.70s/it]
54%|█████▍ | 23857/43939 [38:28:28<31:49:30, 5.71s/it]
54%|█████▍ | 23858/43939 [38:28:33<31:52:32, 5.71s/it]
54%|█████▍ | 23859/43939 [38:28:39<31:51:34, 5.71s/it]
54%|█████▍ | 23860/43939 [38:28:45<31:52:39, 5.72s/it]
54%|█████▍ | 23861/43939 [38:28:51<31:49:56, 5.71s/it]
54%|█████▍ | 23862/43939 [38:28:56<31:49:19, 5.71s/it]
54%|█████▍ | 23863/43939 [38:29:02<32:17:04, 5.79s/it]
54%|█████▍ | 23864/43939 [38:29:08<32:09:07, 5.77s/it]
54%|█████▍ | 23865/43939 [38:29:14<32:01:59, 5.74s/it]
54%|█████▍ | 23866/43939 [38:29:19<31:57:02, 5.73s/it]
54%|█████▍ | 23867/43939 [38:29:25<31:56:34, 5.73s/it]
54%|█████▍ | 23868/43939 [38:29:31<31:49:37, 5.71s/it]
54%|█████▍ | 23869/43939 [38:29:36<31:50:42, 5.71s/it]
54%|█████▍ | 23870/43939 [38:29:42<31:49:05, 5.71s/it]
54%|█████▍ | 23871/43939 [38:29:48<31:47:23, 5.70s/it]
54%|█████▍ | 23872/43939 [38:29:54<31:49:38, 5.71s/it]
54%|█████▍ | 23873/43939 [38:29:59<31:58:09, 5.74s/it]
54%|█████▍ | 23874/43939 [38:30:05<31:57:35, 5.73s/it]
54%|█████▍ | 23875/43939 [38:30:11<31:53:57, 5.72s/it]
54%|█████▍ | 23876/43939 [38:30:16<31:51:47, 5.72s/it]
54%|█████▍ | 23877/43939 [38:30:22<31:56:42, 5.73s/it]
54%|█████▍ | 23878/43939 [38:30:28<31:52:20, 5.72s/it]
54%|█████▍ | 23879/43939 [38:30:34<31:52:42, 5.72s/it]
54%|█████▍ | 23880/43939 [38:30:39<31:51:11, 5.72s/it]
54%|█████▍ | 23881/43939 [38:30:45<31:52:41, 5.72s/it]
54%|█████▍ | 23882/43939 [38:30:51<31:49:47, 5.71s/it]
54%|█████▍ | 23883/43939 [38:30:57<31:48:56, 5.71s/it]
54%|█████▍ | 23884/43939 [38:31:02<32:12:29, 5.78s/it]
54%|█████▍ | 23885/43939 [38:31:08<32:09:11, 5.77s/it]
54%|█████▍ | 23886/43939 [38:31:14<32:02:54, 5.75s/it]
54%|█████▍ | 23887/43939 [38:31:20<32:01:51, 5.75s/it]
54%|█████▍ | 23888/43939 [38:31:25<31:55:33, 5.73s/it]
54%|█████▍ | 23889/43939 [38:31:31<31:51:34, 5.72s/it]
54%|█████▍ | 23890/43939 [38:31:37<31:48:08, 5.71s/it]
54%|█████▍ | 23891/43939 [38:31:42<31:51:10, 5.72s/it]
54%|█████▍ | 23892/43939 [38:31:48<31:48:10, 5.71s/it]
54%|█████▍ | 23893/43939 [38:31:54<31:46:23, 5.71s/it]
54%|█████▍ | 23894/43939 [38:32:00<31:45:00, 5.70s/it]
54%|█████▍ | 23895/43939 [38:32:05<31:46:12, 5.71s/it]
54%|█████▍ | 23896/43939 [38:32:11<31:48:07, 5.71s/it]
54%|█████▍ | 23897/43939 [38:32:17<31:49:04, 5.72s/it]
54%|█████▍ | 23898/43939 [38:32:22<31:47:44, 5.71s/it]
54%|█████▍ | 23899/43939 [38:32:28<31:45:05, 5.70s/it]
54%|█████▍ | 23900/43939 [38:32:34<31:41:05, 5.69s/it]
{'loss': 0.7987, 'grad_norm': 1.4750016927719116, 'learning_rate': 2.5530232098912566e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.5, 'epoch': 2.18}
54%|█████▍ | 23900/43939 [38:32:34<31:41:05, 5.69s/it]
54%|█████▍ | 23901/43939 [38:32:39<31:42:12, 5.70s/it]
54%|█████▍ | 23902/43939 [38:32:45<31:39:48, 5.69s/it]
54%|█████▍ | 23903/43939 [38:32:51<31:39:15, 5.69s/it]
54%|█████▍ | 23904/43939 [38:32:57<31:42:06, 5.70s/it]
54%|█████▍ | 23905/43939 [38:33:02<31:43:38, 5.70s/it]
54%|█████▍ | 23906/43939 [38:33:08<32:06:35, 5.77s/it]
54%|█████▍ | 23907/43939 [38:33:14<31:59:43, 5.75s/it]
54%|█████▍ | 23908/43939 [38:33:20<32:02:53, 5.76s/it]
54%|█████▍ | 23909/43939 [38:33:25<31:56:57, 5.74s/it]
54%|█████▍ | 23910/43939 [38:33:31<31:54:44, 5.74s/it]
54%|█████▍ | 23911/43939 [38:33:37<31:56:41, 5.74s/it]
54%|█████▍ | 23912/43939 [38:33:43<31:54:15, 5.74s/it]
54%|█████▍ | 23913/43939 [38:33:48<31:55:23, 5.74s/it]
54%|█████▍ | 23914/43939 [38:33:54<31:52:10, 5.73s/it]
54%|█████▍ | 23915/43939 [38:34:00<31:45:46, 5.71s/it]
54%|█████▍ | 23916/43939 [38:34:05<31:43:35, 5.70s/it]
54%|█████▍ | 23917/43939 [38:34:11<31:48:11, 5.72s/it]
54%|█████▍ | 23918/43939 [38:34:17<31:46:34, 5.71s/it]
54%|█████▍ | 23919/43939 [38:34:23<31:49:48, 5.72s/it]
54%|█████▍ | 23920/43939 [38:34:28<31:48:55, 5.72s/it]
54%|█████▍ | 23921/43939 [38:34:34<31:52:16, 5.73s/it]
54%|█████▍ | 23922/43939 [38:34:40<31:46:28, 5.71s/it]
54%|█████▍ | 23923/43939 [38:34:45<31:39:59, 5.70s/it]
54%|█████▍ | 23924/43939 [38:34:51<31:36:00, 5.68s/it]
54%|█████▍ | 23925/43939 [38:34:57<31:36:59, 5.69s/it]
54%|█████▍ | 23926/43939 [38:35:02<31:38:43, 5.69s/it]
54%|█████▍ | 23927/43939 [38:35:08<32:05:24, 5.77s/it]
54%|█████▍ | 23928/43939 [38:35:14<31:58:41, 5.75s/it]
54%|█████▍ | 23929/43939 [38:35:20<31:53:18, 5.74s/it]
54%|█████▍ | 23930/43939 [38:35:25<31:49:21, 5.73s/it]
54%|█████▍ | 23931/43939 [38:35:31<31:52:01, 5.73s/it]
54%|█████▍ | 23932/43939 [38:35:37<31:46:37, 5.72s/it]
54%|█████▍ | 23933/43939 [38:35:43<31:44:58, 5.71s/it]
54%|█████▍ | 23934/43939 [38:35:48<31:47:35, 5.72s/it]
54%|█████▍ | 23935/43939 [38:35:54<31:46:04, 5.72s/it]
54%|█████▍ | 23936/43939 [38:36:00<31:45:30, 5.72s/it]
54%|█████▍ | 23937/43939 [38:36:05<31:42:06, 5.71s/it]
54%|█████▍ | 23938/43939 [38:36:11<31:38:40, 5.70s/it]
54%|█████▍ | 23939/43939 [38:36:17<31:38:08, 5.69s/it]
54%|█████▍ | 23940/43939 [38:36:23<31:37:45, 5.69s/it]
54%|█████▍ | 23941/43939 [38:36:28<31:36:44, 5.69s/it]
54%|█████▍ | 23942/43939 [38:36:34<31:37:26, 5.69s/it]
54%|█████▍ | 23943/43939 [38:36:40<31:37:51, 5.69s/it]
54%|█████▍ | 23944/43939 [38:36:45<31:40:17, 5.70s/it]
54%|█████▍ | 23945/43939 [38:36:51<31:36:29, 5.69s/it]
54%|█████▍ | 23946/43939 [38:36:57<31:34:28, 5.69s/it]
55%|█████▍ | 23947/43939 [38:37:02<31:34:16, 5.69s/it]
55%|█████▍ | 23948/43939 [38:37:08<31:59:24, 5.76s/it]
55%|█████▍ | 23949/43939 [38:37:14<31:56:29, 5.75s/it]
55%|█████▍ | 23950/43939 [38:37:20<31:56:49, 5.75s/it]
{'loss': 0.7918, 'grad_norm': 1.1996573209762573, 'learning_rate': 2.5430948667936215e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.39, 'epoch': 2.18}
55%|█████▍ | 23950/43939 [38:37:20<31:56:49, 5.75s/it]
55%|█████▍ | 23951/43939 [38:37:26<31:55:37, 5.75s/it]
55%|█████▍ | 23952/43939 [38:37:31<31:53:03, 5.74s/it]
55%|█████▍ | 23953/43939 [38:37:37<31:56:13, 5.75s/it]
55%|█████▍ | 23954/43939 [38:37:43<31:48:10, 5.73s/it]
55%|█████▍ | 23955/43939 [38:37:48<31:47:57, 5.73s/it]
55%|█████▍ | 23956/43939 [38:37:54<31:44:27, 5.72s/it]
55%|█████▍ | 23957/43939 [38:38:00<31:41:34, 5.71s/it]
55%|█████▍ | 23958/43939 [38:38:06<31:43:07, 5.71s/it]
55%|█████▍ | 23959/43939 [38:38:11<31:40:21, 5.71s/it]
55%|█████▍ | 23960/43939 [38:38:17<31:39:27, 5.70s/it]
55%|█████▍ | 23961/43939 [38:38:23<31:37:10, 5.70s/it]
55%|█████▍ | 23962/43939 [38:38:28<31:32:46, 5.68s/it]
55%|█████▍ | 23963/43939 [38:38:34<31:35:37, 5.69s/it]
55%|█████▍ | 23964/43939 [38:38:40<31:35:58, 5.70s/it]
55%|█████▍ | 23965/43939 [38:38:45<31:42:31, 5.71s/it]
55%|█████▍ | 23966/43939 [38:38:51<31:40:03, 5.71s/it]
55%|█████▍ | 23967/43939 [38:38:57<31:39:15, 5.71s/it]
55%|█████▍ | 23968/43939 [38:39:03<31:36:29, 5.70s/it]
55%|█████▍ | 23969/43939 [38:39:08<32:04:06, 5.78s/it]
55%|█████▍ | 23970/43939 [38:39:14<31:58:21, 5.76s/it]
55%|█████▍ | 23971/43939 [38:39:20<31:53:54, 5.75s/it]
55%|█████▍ | 23972/43939 [38:39:26<31:52:44, 5.75s/it]
55%|█████▍ | 23973/43939 [38:39:31<31:52:15, 5.75s/it]
55%|█████▍ | 23974/43939 [38:39:37<31:46:03, 5.73s/it]
55%|█████▍ | 23975/43939 [38:39:43<31:40:21, 5.71s/it]
55%|█████▍ | 23976/43939 [38:39:48<31:40:23, 5.71s/it]
55%|█████▍ | 23977/43939 [38:39:54<31:42:23, 5.72s/it]
55%|█████▍ | 23978/43939 [38:40:00<31:40:28, 5.71s/it]
55%|█████▍ | 23979/43939 [38:40:06<31:38:29, 5.71s/it]
55%|█████▍ | 23980/43939 [38:40:11<31:35:09, 5.70s/it]
55%|█████▍ | 23981/43939 [38:40:17<31:35:56, 5.70s/it]
55%|█████▍ | 23982/43939 [38:40:23<31:40:03, 5.71s/it]
55%|█████▍ | 23983/43939 [38:40:29<31:46:11, 5.73s/it]
55%|█████▍ | 23984/43939 [38:40:34<31:39:22, 5.71s/it]
55%|█████▍ | 23985/43939 [38:40:40<31:47:46, 5.74s/it]
55%|█████▍ | 23986/43939 [38:40:46<31:45:06, 5.73s/it]
55%|█████▍ | 23987/43939 [38:40:51<31:40:39, 5.72s/it]
55%|█████▍ | 23988/43939 [38:40:57<31:43:50, 5.73s/it]
55%|█████▍ | 23989/43939 [38:41:03<31:41:50, 5.72s/it]
55%|█████▍ | 23990/43939 [38:41:09<31:39:43, 5.71s/it]
55%|█████▍ | 23991/43939 [38:41:14<32:03:44, 5.79s/it]
55%|█████▍ | 23992/43939 [38:41:20<31:58:30, 5.77s/it]
55%|█████▍ | 23993/43939 [38:41:26<31:57:27, 5.77s/it]
55%|█████▍ | 23994/43939 [38:41:32<31:49:28, 5.74s/it]
55%|█████▍ | 23995/43939 [38:41:37<31:46:09, 5.73s/it]
55%|█████▍ | 23996/43939 [38:41:43<31:43:49, 5.73s/it]
55%|█████▍ | 23997/43939 [38:41:49<31:43:16, 5.73s/it]
55%|█████▍ | 23998/43939 [38:41:54<31:37:17, 5.71s/it]
55%|█████▍ | 23999/43939 [38:42:00<31:34:05, 5.70s/it]
55%|█████▍ | 24000/43939 [38:42:06<31:36:16, 5.71s/it]
{'loss': 0.807, 'grad_norm': 1.2561222314834595, 'learning_rate': 2.5331658437731925e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 534.12, 'epoch': 2.18}
55%|█████▍ | 24000/43939 [38:42:06<31:36:16, 5.71s/it]
55%|█████▍ | 24001/43939 [38:42:12<31:40:22, 5.72s/it]
55%|█████▍ | 24002/43939 [38:42:17<31:43:52, 5.73s/it]
55%|█████▍ | 24003/43939 [38:42:23<31:43:19, 5.73s/it]
55%|█████▍ | 24004/43939 [38:42:29<31:39:27, 5.72s/it]
55%|█████▍ | 24005/43939 [38:42:34<31:37:46, 5.71s/it]
55%|█████▍ | 24006/43939 [38:42:40<31:36:01, 5.71s/it]
55%|█████▍ | 24007/43939 [38:42:46<31:31:46, 5.69s/it]
55%|█████▍ | 24008/43939 [38:42:52<31:35:57, 5.71s/it]
55%|█████▍ | 24009/43939 [38:42:57<31:36:43, 5.71s/it]
55%|█████▍ | 24010/43939 [38:43:03<31:38:44, 5.72s/it]
55%|█████▍ | 24011/43939 [38:43:09<31:37:59, 5.71s/it]
55%|█████▍ | 24012/43939 [38:43:15<32:07:22, 5.80s/it]
55%|█████▍ | 24013/43939 [38:43:20<31:54:57, 5.77s/it]
55%|█████▍ | 24014/43939 [38:43:26<31:48:23, 5.75s/it]
55%|█████▍ | 24015/43939 [38:43:32<31:43:31, 5.73s/it]
55%|█████▍ | 24016/43939 [38:43:38<31:40:39, 5.72s/it]
55%|█████▍ | 24017/43939 [38:43:43<31:36:48, 5.71s/it]
55%|█████▍ | 24018/43939 [38:43:49<31:38:13, 5.72s/it]
55%|█████▍ | 24019/43939 [38:43:55<31:34:42, 5.71s/it]
55%|█████▍ | 24020/43939 [38:44:00<31:36:43, 5.71s/it]
55%|█████▍ | 24021/43939 [38:44:06<31:35:37, 5.71s/it]
55%|█████▍ | 24022/43939 [38:44:12<31:30:20, 5.69s/it]
55%|█████▍ | 24023/43939 [38:44:17<31:31:10, 5.70s/it]
55%|█████▍ | 24024/43939 [38:44:23<31:33:15, 5.70s/it]
55%|█████▍ | 24025/43939 [38:44:29<31:33:46, 5.71s/it]
55%|█████▍ | 24026/43939 [38:44:35<31:37:42, 5.72s/it]
55%|█████▍ | 24027/43939 [38:44:40<31:31:49, 5.70s/it]
55%|█████▍ | 24028/43939 [38:44:46<31:34:41, 5.71s/it]
55%|█████▍ | 24029/43939 [38:44:52<31:32:43, 5.70s/it]
55%|█████▍ | 24030/43939 [38:44:57<31:34:52, 5.71s/it]
55%|█████▍ | 24031/43939 [38:45:03<31:35:32, 5.71s/it]
55%|█████▍ | 24032/43939 [38:45:09<31:35:54, 5.71s/it]
55%|█████▍ | 24033/43939 [38:45:15<31:58:09, 5.78s/it]
55%|█████▍ | 24034/43939 [38:45:20<31:46:42, 5.75s/it]
55%|█████▍ | 24035/43939 [38:45:26<31:41:49, 5.73s/it]
55%|█████▍ | 24036/43939 [38:45:32<31:41:49, 5.73s/it]
55%|█████▍ | 24037/43939 [38:45:38<31:38:21, 5.72s/it]
55%|█████▍ | 24038/43939 [38:45:43<31:39:41, 5.73s/it]
55%|█████▍ | 24039/43939 [38:45:49<31:32:11, 5.71s/it]
55%|█████▍ | 24040/43939 [38:45:55<31:33:17, 5.71s/it]
55%|█████▍ | 24041/43939 [38:46:00<31:32:28, 5.71s/it]
55%|█████▍ | 24042/43939 [38:46:06<31:34:15, 5.71s/it]
55%|█████▍ | 24043/43939 [38:46:12<31:38:07, 5.72s/it]
55%|█████▍ | 24044/43939 [38:46:18<31:34:56, 5.71s/it]
55%|█████▍ | 24045/43939 [38:46:23<31:37:54, 5.72s/it]
55%|█████▍ | 24046/43939 [38:46:29<31:37:31, 5.72s/it]
55%|█████▍ | 24047/43939 [38:46:35<31:34:48, 5.72s/it]
55%|█████▍ | 24048/43939 [38:46:40<31:31:30, 5.71s/it]
55%|█████▍ | 24049/43939 [38:46:46<31:29:09, 5.70s/it]
55%|█████▍ | 24050/43939 [38:46:52<31:28:48, 5.70s/it]
{'loss': 0.7788, 'grad_norm': 1.34530508518219, 'learning_rate': 2.5232362974836394e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.18, 'epoch': 2.19}
55%|█████▍ | 24050/43939 [38:46:52<31:28:48, 5.70s/it]
55%|█████▍ | 24051/43939 [38:46:58<31:30:34, 5.70s/it]
55%|█████▍ | 24052/43939 [38:47:03<31:27:39, 5.70s/it]
55%|█████▍ | 24053/43939 [38:47:09<31:30:46, 5.70s/it]
55%|█████▍ | 24054/43939 [38:47:15<31:54:58, 5.78s/it]
55%|█████▍ | 24055/43939 [38:47:21<31:49:52, 5.76s/it]
55%|█████▍ | 24056/43939 [38:47:26<31:49:08, 5.76s/it]
55%|█████▍ | 24057/43939 [38:47:32<31:43:29, 5.74s/it]
55%|█████▍ | 24058/43939 [38:47:38<31:36:18, 5.72s/it]
55%|█████▍ | 24059/43939 [38:47:43<31:35:56, 5.72s/it]
55%|█████▍ | 24060/43939 [38:47:49<31:31:16, 5.71s/it]
55%|█████▍ | 24061/43939 [38:47:55<31:30:29, 5.71s/it]
55%|█████▍ | 24062/43939 [38:48:01<31:32:05, 5.71s/it]
55%|█████▍ | 24063/43939 [38:48:06<31:30:01, 5.71s/it]
55%|█████▍ | 24064/43939 [38:48:12<31:30:29, 5.71s/it]
55%|█████▍ | 24065/43939 [38:48:18<31:30:55, 5.71s/it]
55%|█████▍ | 24066/43939 [38:48:23<31:29:08, 5.70s/it]
55%|█████▍ | 24067/43939 [38:48:29<31:32:41, 5.71s/it]
55%|█████▍ | 24068/43939 [38:48:35<31:28:54, 5.70s/it]
55%|█████▍ | 24069/43939 [38:48:40<31:27:59, 5.70s/it]
55%|█████▍ | 24070/43939 [38:48:46<31:28:40, 5.70s/it]
55%|█████▍ | 24071/43939 [38:48:52<31:31:06, 5.71s/it]
55%|█████▍ | 24072/43939 [38:48:58<31:32:18, 5.71s/it]
55%|█████▍ | 24073/43939 [38:49:03<31:34:57, 5.72s/it]
55%|█████▍ | 24074/43939 [38:49:09<31:34:26, 5.72s/it]
55%|█████▍ | 24075/43939 [38:49:15<31:29:11, 5.71s/it]
55%|█████▍ | 24076/43939 [38:49:21<31:48:40, 5.77s/it]
55%|█████▍ | 24077/43939 [38:49:26<31:45:01, 5.75s/it]
55%|█████▍ | 24078/43939 [38:49:32<31:41:40, 5.74s/it]
55%|█████▍ | 24079/43939 [38:49:38<31:37:51, 5.73s/it]
55%|█████▍ | 24080/43939 [38:49:44<31:36:23, 5.73s/it]
55%|█████▍ | 24081/43939 [38:49:49<31:32:46, 5.72s/it]
55%|█████▍ | 24082/43939 [38:49:55<31:31:34, 5.72s/it]
55%|█████▍ | 24083/43939 [38:50:01<31:30:44, 5.71s/it]
55%|█████▍ | 24084/43939 [38:50:06<31:29:37, 5.71s/it]
55%|█████▍ | 24085/43939 [38:50:12<31:25:57, 5.70s/it]
55%|█████▍ | 24086/43939 [38:50:18<31:26:41, 5.70s/it]
55%|█████▍ | 24087/43939 [38:50:23<31:25:07, 5.70s/it]
55%|█████▍ | 24088/43939 [38:50:29<31:28:23, 5.71s/it]
55%|█████▍ | 24089/43939 [38:50:35<31:26:24, 5.70s/it]
55%|█████▍ | 24090/43939 [38:50:41<31:27:14, 5.70s/it]
55%|█████▍ | 24091/43939 [38:50:46<31:28:41, 5.71s/it]
55%|█████▍ | 24092/43939 [38:50:52<31:26:01, 5.70s/it]
55%|█████▍ | 24093/43939 [38:50:58<31:22:56, 5.69s/it]
55%|█████▍ | 24094/43939 [38:51:03<31:21:16, 5.69s/it]
55%|█████▍ | 24095/43939 [38:51:09<31:34:10, 5.73s/it]
55%|█████▍ | 24096/43939 [38:51:15<31:32:18, 5.72s/it]
55%|█████▍ | 24097/43939 [38:51:21<31:29:05, 5.71s/it]
55%|█████▍ | 24098/43939 [38:51:26<31:49:53, 5.78s/it]
55%|█████▍ | 24099/43939 [38:51:32<31:44:23, 5.76s/it]
55%|█████▍ | 24100/43939 [38:51:38<31:36:37, 5.74s/it]
{'loss': 0.7896, 'grad_norm': 1.4375536441802979, 'learning_rate': 2.513306384586888e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.1, 'epoch': 2.19}
55%|█████▍ | 24100/43939 [38:51:38<31:36:37, 5.74s/it]
55%|█████▍ | 24101/43939 [38:51:44<31:31:18, 5.72s/it]
55%|█████▍ | 24102/43939 [38:51:49<31:30:33, 5.72s/it]
55%|█████▍ | 24103/43939 [38:51:55<31:28:40, 5.71s/it]
55%|█████▍ | 24104/43939 [38:52:01<31:30:09, 5.72s/it]
55%|█████▍ | 24105/43939 [38:52:06<31:30:49, 5.72s/it]
55%|█████▍ | 24106/43939 [38:52:12<31:29:58, 5.72s/it]
55%|█████▍ | 24107/43939 [38:52:18<31:26:31, 5.71s/it]
55%|█████▍ | 24108/43939 [38:52:23<31:21:06, 5.69s/it]
55%|█████▍ | 24109/43939 [38:52:29<31:19:21, 5.69s/it]
55%|█████▍ | 24110/43939 [38:52:35<31:17:25, 5.68s/it]
55%|█████▍ | 24111/43939 [38:52:40<31:16:44, 5.68s/it]
55%|█████▍ | 24112/43939 [38:52:46<31:20:01, 5.69s/it]
55%|█████▍ | 24113/43939 [38:52:52<31:19:22, 5.69s/it]
55%|█████▍ | 24114/43939 [38:52:58<31:18:41, 5.69s/it]
55%|█████▍ | 24115/43939 [38:53:03<31:18:23, 5.69s/it]
55%|█████▍ | 24116/43939 [38:53:09<31:20:43, 5.69s/it]
55%|█████▍ | 24117/43939 [38:53:15<31:24:13, 5.70s/it]
55%|█████▍ | 24118/43939 [38:53:21<31:47:53, 5.78s/it]
55%|█████▍ | 24119/43939 [38:53:26<31:39:50, 5.75s/it]
55%|█████▍ | 24120/43939 [38:53:32<31:31:53, 5.73s/it]
55%|█████▍ | 24121/43939 [38:53:38<31:30:03, 5.72s/it]
55%|█████▍ | 24122/43939 [38:53:43<31:29:54, 5.72s/it]
55%|█████▍ | 24123/43939 [38:53:49<31:30:00, 5.72s/it]
55%|█████▍ | 24124/43939 [38:53:55<31:30:47, 5.73s/it]
55%|█████▍ | 24125/43939 [38:54:01<31:26:13, 5.71s/it]
55%|█████▍ | 24126/43939 [38:54:06<31:28:05, 5.72s/it]
55%|█████▍ | 24127/43939 [38:54:12<31:28:22, 5.72s/it]
55%|█████▍ | 24128/43939 [38:54:18<31:26:02, 5.71s/it]
55%|█████▍ | 24129/43939 [38:54:23<31:22:11, 5.70s/it]
55%|█████▍ | 24130/43939 [38:54:29<31:22:18, 5.70s/it]
55%|█████▍ | 24131/43939 [38:54:35<31:19:40, 5.69s/it]
55%|█████▍ | 24132/43939 [38:54:40<31:21:32, 5.70s/it]
55%|█████▍ | 24133/43939 [38:54:46<31:23:56, 5.71s/it]
55%|█████▍ | 24134/43939 [38:54:52<31:26:33, 5.72s/it]
55%|█████▍ | 24135/43939 [38:54:58<31:27:04, 5.72s/it]
55%|█████▍ | 24136/43939 [38:55:03<31:28:07, 5.72s/it]
55%|█████▍ | 24137/43939 [38:55:09<31:26:50, 5.72s/it]
55%|█████▍ | 24138/43939 [38:55:15<31:24:57, 5.71s/it]
55%|█████▍ | 24139/43939 [38:55:21<31:50:59, 5.79s/it]
55%|█████▍ | 24140/43939 [38:55:26<31:38:59, 5.75s/it]
55%|█████▍ | 24141/43939 [38:55:32<31:33:07, 5.74s/it]
55%|█████▍ | 24142/43939 [38:55:38<31:28:16, 5.72s/it]
55%|█████▍ | 24143/43939 [38:55:43<31:23:33, 5.71s/it]
55%|█████▍ | 24144/43939 [38:55:49<31:24:29, 5.71s/it]
55%|█████▍ | 24145/43939 [38:55:55<31:20:13, 5.70s/it]
55%|█████▍ | 24146/43939 [38:56:01<31:22:32, 5.71s/it]
55%|█████▍ | 24147/43939 [38:56:06<31:21:00, 5.70s/it]
55%|█████▍ | 24148/43939 [38:56:12<31:19:12, 5.70s/it]
55%|█████▍ | 24149/43939 [38:56:18<31:14:37, 5.68s/it]
55%|█████▍ | 24150/43939 [38:56:23<31:21:54, 5.71s/it]
{'loss': 0.8008, 'grad_norm': 1.1456172466278076, 'learning_rate': 2.5033762617506507e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.79, 'epoch': 2.2}
55%|█████▍ | 24150/43939 [38:56:23<31:21:54, 5.71s/it]
55%|█████▍ | 24151/43939 [38:56:29<31:18:28, 5.70s/it]
55%|█████▍ | 24152/43939 [38:56:35<31:19:02, 5.70s/it]
55%|█████▍ | 24153/43939 [38:56:41<31:23:48, 5.71s/it]
55%|█████▍ | 24154/43939 [38:56:46<31:22:00, 5.71s/it]
55%|█████▍ | 24155/43939 [38:56:52<31:23:02, 5.71s/it]
55%|█████▍ | 24156/43939 [38:56:58<31:19:43, 5.70s/it]
55%|█████▍ | 24157/43939 [38:57:03<31:22:17, 5.71s/it]
55%|█████▍ | 24158/43939 [38:57:09<31:15:57, 5.69s/it]
55%|█████▍ | 24159/43939 [38:57:15<31:20:02, 5.70s/it]
55%|█████▍ | 24160/43939 [38:57:20<31:22:50, 5.71s/it]
55%|█████▍ | 24161/43939 [38:57:26<31:38:50, 5.76s/it]
55%|█████▍ | 24162/43939 [38:57:32<31:31:23, 5.74s/it]
55%|█████▍ | 24163/43939 [38:57:38<31:27:01, 5.73s/it]
55%|█████▍ | 24164/43939 [38:57:43<31:27:05, 5.73s/it]
55%|█████▍ | 24165/43939 [38:57:49<31:25:36, 5.72s/it]
55%|█████▍ | 24166/43939 [38:57:55<31:23:26, 5.72s/it]
55%|█████▌ | 24167/43939 [38:58:00<31:17:36, 5.70s/it][2025-11-30 09:02:14,312] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:41, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<01:00, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:51, 2.74s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9876511096954346, 'eval_runtime': 195.9364, 'eval_samples_per_second': 5.099, 'eval_steps_per_second': 0.367, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 2.2}
55%|█████▌ | 24167/43939 [39:01:16<31:17:36, 5.70s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
55%|█████▌ | 24168/43939 [39:01:22<354:23:10, 64.53s/it]
55%|█████▌ | 24169/43939 [39:01:28<257:26:37, 46.88s/it]
55%|█████▌ | 24170/43939 [39:01:34<189:32:36, 34.52s/it]
55%|█████▌ | 24171/43939 [39:01:39<142:00:26, 25.86s/it]
55%|█████▌ | 24172/43939 [39:01:45<108:50:41, 19.82s/it]
55%|█████▌ | 24173/43939 [39:01:51<85:41:45, 15.61s/it]
55%|█████▌ | 24174/43939 [39:01:57<69:23:22, 12.64s/it]
55%|█████▌ | 24175/43939 [39:02:02<57:56:09, 10.55s/it]
55%|█████▌ | 24176/43939 [39:02:08<49:55:00, 9.09s/it]
55%|█████▌ | 24177/43939 [39:02:14<44:17:57, 8.07s/it]
55%|█████▌ | 24178/43939 [39:02:19<40:23:34, 7.36s/it]
55%|█████▌ | 24179/43939 [39:02:25<37:36:48, 6.85s/it]
55%|█████▌ | 24180/43939 [39:02:31<35:42:11, 6.50s/it]
55%|█████▌ | 24181/43939 [39:02:37<34:47:41, 6.34s/it]
55%|█████▌ | 24182/43939 [39:02:42<33:44:14, 6.15s/it]
55%|█████▌ | 24183/43939 [39:02:48<33:00:26, 6.01s/it]
55%|█████▌ | 24184/43939 [39:02:54<32:32:01, 5.93s/it]
55%|█████▌ | 24185/43939 [39:02:59<32:11:34, 5.87s/it]
55%|█████▌ | 24186/43939 [39:03:05<31:58:21, 5.83s/it]
55%|█████▌ | 24187/43939 [39:03:11<31:42:57, 5.78s/it]
55%|█████▌ | 24188/43939 [39:03:17<31:37:14, 5.76s/it]
55%|█████▌ | 24189/43939 [39:03:22<31:33:24, 5.75s/it]
55%|█████▌ | 24190/43939 [39:03:28<31:30:52, 5.74s/it]
55%|█████▌ | 24191/43939 [39:03:34<31:26:36, 5.73s/it]
55%|█████▌ | 24192/43939 [39:03:39<31:21:59, 5.72s/it]
55%|█████▌ | 24193/43939 [39:03:45<31:18:57, 5.71s/it]
55%|█████▌ | 24194/43939 [39:03:51<31:22:00, 5.72s/it]
55%|█████▌ | 24195/43939 [39:03:57<31:21:55, 5.72s/it]
55%|█████▌ | 24196/43939 [39:04:02<31:17:01, 5.70s/it]
55%|█████▌ | 24197/43939 [39:04:08<31:25:05, 5.73s/it]
55%|█████▌ | 24198/43939 [39:04:14<31:26:39, 5.73s/it]
55%|█████▌ | 24199/43939 [39:04:20<31:24:07, 5.73s/it]
55%|█████▌ | 24200/43939 [39:04:25<31:23:04, 5.72s/it]
{'loss': 0.7625, 'grad_norm': 1.0658150911331177, 'learning_rate': 2.4934460856459503e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 312.8, 'epoch': 2.2}
55%|█████▌ | 24200/43939 [39:04:25<31:23:04, 5.72s/it]
55%|█████▌ | 24201/43939 [39:04:31<31:17:29, 5.71s/it]
55%|█████▌ | 24202/43939 [39:04:37<31:16:57, 5.71s/it]
55%|█████▌ | 24203/43939 [39:04:43<31:45:40, 5.79s/it]
55%|█████▌ | 24204/43939 [39:04:48<31:36:30, 5.77s/it]
55%|█████▌ | 24205/43939 [39:04:54<31:27:30, 5.74s/it]
55%|█████▌ | 24206/43939 [39:05:00<31:30:25, 5.75s/it]
55%|█████▌ | 24207/43939 [39:05:05<31:25:47, 5.73s/it]
55%|█████▌ | 24208/43939 [39:05:11<31:26:18, 5.74s/it]
55%|█████▌ | 24209/43939 [39:05:17<31:24:23, 5.73s/it]
55%|█████▌ | 24210/43939 [39:05:23<31:17:15, 5.71s/it]
55%|█████▌ | 24211/43939 [39:05:28<31:15:22, 5.70s/it]
55%|█████▌ | 24212/43939 [39:05:34<31:18:04, 5.71s/it]
55%|█████▌ | 24213/43939 [39:05:40<31:15:30, 5.70s/it]
55%|█████▌ | 24214/43939 [39:05:45<31:14:59, 5.70s/it]
55%|█████▌ | 24215/43939 [39:05:51<31:19:59, 5.72s/it]
55%|█████▌ | 24216/43939 [39:05:57<31:22:33, 5.73s/it]
55%|█████▌ | 24217/43939 [39:06:03<31:20:52, 5.72s/it]
55%|█████▌ | 24218/43939 [39:06:08<31:19:09, 5.72s/it]
55%|█████▌ | 24219/43939 [39:06:14<31:18:31, 5.72s/it]
55%|█████▌ | 24220/43939 [39:06:20<31:20:40, 5.72s/it]
55%|█████▌ | 24221/43939 [39:06:25<31:16:14, 5.71s/it]
55%|█████▌ | 24222/43939 [39:06:31<31:15:13, 5.71s/it]
55%|█████▌ | 24223/43939 [39:06:37<31:17:38, 5.71s/it]
55%|█████▌ | 24224/43939 [39:06:43<31:38:15, 5.78s/it]
55%|█████▌ | 24225/43939 [39:06:48<31:30:45, 5.75s/it]
55%|█████▌ | 24226/43939 [39:06:54<31:28:44, 5.75s/it]
55%|█████▌ | 24227/43939 [39:07:00<31:28:56, 5.75s/it]
55%|█████▌ | 24228/43939 [39:07:06<31:22:03, 5.73s/it]
55%|█████▌ | 24229/43939 [39:07:11<31:19:41, 5.72s/it]
55%|█████▌ | 24230/43939 [39:07:17<31:15:25, 5.71s/it]
55%|█████▌ | 24231/43939 [39:07:23<31:08:26, 5.69s/it]
55%|█████▌ | 24232/43939 [39:07:28<31:08:33, 5.69s/it]
55%|█████▌ | 24233/43939 [39:07:34<31:09:55, 5.69s/it]
55%|█████▌ | 24234/43939 [39:07:40<31:11:57, 5.70s/it]
55%|█████▌ | 24235/43939 [39:07:45<31:13:07, 5.70s/it]
55%|█████▌ | 24236/43939 [39:07:51<31:11:46, 5.70s/it]
55%|█████▌ | 24237/43939 [39:07:57<31:16:08, 5.71s/it]
55%|█████▌ | 24238/43939 [39:08:03<31:11:30, 5.70s/it]
55%|█████▌ | 24239/43939 [39:08:08<31:11:03, 5.70s/it]
55%|█████▌ | 24240/43939 [39:08:14<31:12:42, 5.70s/it]
55%|█████▌ | 24241/43939 [39:08:20<31:10:27, 5.70s/it]
55%|█████▌ | 24242/43939 [39:08:25<31:07:53, 5.69s/it]
55%|█████▌ | 24243/43939 [39:08:31<31:09:14, 5.69s/it]
55%|█████▌ | 24244/43939 [39:08:37<31:11:25, 5.70s/it]
55%|█████▌ | 24245/43939 [39:08:42<31:11:18, 5.70s/it]
55%|█████▌ | 24246/43939 [39:08:48<31:34:41, 5.77s/it]
55%|█████▌ | 24247/43939 [39:08:54<31:24:07, 5.74s/it]
55%|█████▌ | 24248/43939 [39:09:00<31:22:46, 5.74s/it]
55%|█████▌ | 24249/43939 [39:09:06<31:19:01, 5.73s/it]
55%|█████▌ | 24250/43939 [39:09:11<31:14:16, 5.71s/it]
{'loss': 0.7795, 'grad_norm': 1.3847614526748657, 'learning_rate': 2.48351601294465e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.79, 'epoch': 2.21}
55%|█████▌ | 24250/43939 [39:09:11<31:14:16, 5.71s/it]
55%|█████▌ | 24251/43939 [39:09:17<31:17:11, 5.72s/it]
55%|█████▌ | 24252/43939 [39:09:23<31:18:36, 5.73s/it]
55%|█████▌ | 24253/43939 [39:09:28<31:13:58, 5.71s/it]
55%|█████▌ | 24254/43939 [39:09:34<31:16:35, 5.72s/it]
55%|█████▌ | 24255/43939 [39:09:40<31:16:08, 5.72s/it]
55%|█████▌ | 24256/43939 [39:09:45<31:12:15, 5.71s/it]
55%|█████▌ | 24257/43939 [39:09:51<31:10:44, 5.70s/it]
55%|█████▌ | 24258/43939 [39:09:57<31:11:49, 5.71s/it]
55%|█████▌ | 24259/43939 [39:10:03<31:08:58, 5.70s/it]
55%|█████▌ | 24260/43939 [39:10:08<31:09:50, 5.70s/it]
55%|█████▌ | 24261/43939 [39:10:14<31:12:55, 5.71s/it]
55%|█████▌ | 24262/43939 [39:10:20<31:13:51, 5.71s/it]
55%|█████▌ | 24263/43939 [39:10:25<31:10:00, 5.70s/it]
55%|█████▌ | 24264/43939 [39:10:31<31:08:31, 5.70s/it]
55%|█████▌ | 24265/43939 [39:10:37<31:09:40, 5.70s/it]
55%|█████▌ | 24266/43939 [39:10:43<31:09:46, 5.70s/it]
55%|█████▌ | 24267/43939 [39:10:48<31:31:40, 5.77s/it]
55%|█████▌ | 24268/43939 [39:10:54<31:28:53, 5.76s/it]
55%|█████▌ | 24269/43939 [39:11:00<31:20:52, 5.74s/it]
55%|█████▌ | 24270/43939 [39:11:06<31:20:27, 5.74s/it]
55%|█████▌ | 24271/43939 [39:11:11<31:17:55, 5.73s/it]
55%|█████▌ | 24272/43939 [39:11:17<31:13:03, 5.71s/it]
55%|█████▌ | 24273/43939 [39:11:23<31:16:42, 5.73s/it]
55%|█████▌ | 24274/43939 [39:11:28<31:17:04, 5.73s/it]
55%|█████▌ | 24275/43939 [39:11:34<31:12:11, 5.71s/it]
55%|█████▌ | 24276/43939 [39:11:40<31:12:18, 5.71s/it]
55%|█████▌ | 24277/43939 [39:11:46<31:10:23, 5.71s/it]
55%|█████▌ | 24278/43939 [39:11:51<31:10:53, 5.71s/it]
55%|█████▌ | 24279/43939 [39:11:57<31:09:26, 5.71s/it]
55%|█████▌ | 24280/43939 [39:12:03<31:11:14, 5.71s/it]
55%|█████▌ | 24281/43939 [39:12:08<31:11:32, 5.71s/it]
55%|█████▌ | 24282/43939 [39:12:14<31:10:19, 5.71s/it]
55%|█████▌ | 24283/43939 [39:12:20<31:08:21, 5.70s/it]
55%|█████▌ | 24284/43939 [39:12:26<31:09:22, 5.71s/it]
55%|█████▌ | 24285/43939 [39:12:31<31:16:31, 5.73s/it]
55%|█████▌ | 24286/43939 [39:12:37<31:13:38, 5.72s/it]
55%|█████▌ | 24287/43939 [39:12:43<31:11:56, 5.72s/it]
55%|█████▌ | 24288/43939 [39:12:48<31:10:41, 5.71s/it]
55%|█████▌ | 24289/43939 [39:12:54<31:32:58, 5.78s/it]
55%|█████▌ | 24290/43939 [39:13:00<31:24:15, 5.75s/it]
55%|█████▌ | 24291/43939 [39:13:06<31:22:24, 5.75s/it]
55%|█████▌ | 24292/43939 [39:13:11<31:15:59, 5.73s/it]
55%|█████▌ | 24293/43939 [39:13:17<31:09:46, 5.71s/it]
55%|█████▌ | 24294/43939 [39:13:23<31:05:59, 5.70s/it]
55%|█████▌ | 24295/43939 [39:13:28<31:04:35, 5.70s/it]
55%|█████▌ | 24296/43939 [39:13:34<31:08:13, 5.71s/it]
55%|█████▌ | 24297/43939 [39:13:40<31:06:43, 5.70s/it]
55%|█████▌ | 24298/43939 [39:13:46<31:10:04, 5.71s/it]
55%|█████▌ | 24299/43939 [39:13:51<31:09:25, 5.71s/it]
55%|█████▌ | 24300/43939 [39:13:57<31:10:08, 5.71s/it]
{'loss': 0.7945, 'grad_norm': 1.2221823930740356, 'learning_rate': 2.4735862003169822e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 517.91, 'epoch': 2.21}
55%|█████▌ | 24300/43939 [39:13:57<31:10:08, 5.71s/it]
55%|█████▌ | 24301/43939 [39:14:03<31:05:35, 5.70s/it]
55%|█████▌ | 24302/43939 [39:14:08<31:09:59, 5.71s/it]
55%|█████▌ | 24303/43939 [39:14:14<31:15:16, 5.73s/it]
55%|█████▌ | 24304/43939 [39:14:20<31:12:24, 5.72s/it]
55%|█████▌ | 24305/43939 [39:14:26<31:08:29, 5.71s/it]
55%|█████▌ | 24306/43939 [39:14:31<31:04:33, 5.70s/it]
55%|█████▌ | 24307/43939 [39:14:37<31:05:32, 5.70s/it]
55%|█████▌ | 24308/43939 [39:14:43<31:03:33, 5.70s/it]
55%|█████▌ | 24309/43939 [39:14:48<31:04:26, 5.70s/it]
55%|█████▌ | 24310/43939 [39:14:54<31:29:23, 5.78s/it]
55%|█████▌ | 24311/43939 [39:15:00<31:24:13, 5.76s/it]
55%|█████▌ | 24312/43939 [39:15:06<31:18:45, 5.74s/it]
55%|█████▌ | 24313/43939 [39:15:12<31:15:59, 5.74s/it]
55%|█████▌ | 24314/43939 [39:15:17<31:13:05, 5.73s/it]
55%|█████▌ | 24315/43939 [39:15:23<31:09:50, 5.72s/it]
55%|█████▌ | 24316/43939 [39:15:29<31:09:10, 5.72s/it]
55%|█████▌ | 24317/43939 [39:15:34<31:05:53, 5.71s/it]
55%|█████▌ | 24318/43939 [39:15:40<31:04:59, 5.70s/it]
55%|█████▌ | 24319/43939 [39:15:46<31:02:52, 5.70s/it]
55%|█████▌ | 24320/43939 [39:15:51<31:08:04, 5.71s/it]
55%|█████▌ | 24321/43939 [39:15:57<31:04:44, 5.70s/it]
55%|█████▌ | 24322/43939 [39:16:03<31:05:17, 5.71s/it]
55%|█████▌ | 24323/43939 [39:16:09<31:04:23, 5.70s/it]
55%|█████▌ | 24324/43939 [39:16:14<31:03:37, 5.70s/it]
55%|█████▌ | 24325/43939 [39:16:20<31:01:26, 5.69s/it]
55%|█████▌ | 24326/43939 [39:16:26<30:59:05, 5.69s/it]
55%|█████▌ | 24327/43939 [39:16:31<31:01:22, 5.69s/it]
55%|█████▌ | 24328/43939 [39:16:37<31:00:43, 5.69s/it]
55%|█████▌ | 24329/43939 [39:16:43<31:04:39, 5.71s/it]
55%|█████▌ | 24330/43939 [39:16:48<31:04:46, 5.71s/it]
55%|█████▌ | 24331/43939 [39:16:54<31:32:44, 5.79s/it]
55%|█████▌ | 24332/43939 [39:17:00<31:19:06, 5.75s/it]
55%|█████▌ | 24333/43939 [39:17:06<31:13:22, 5.73s/it]
55%|█████▌ | 24334/43939 [39:17:11<31:08:38, 5.72s/it]
55%|█████▌ | 24335/43939 [39:17:17<31:12:29, 5.73s/it]
55%|█████▌ | 24336/43939 [39:17:23<31:06:53, 5.71s/it]
55%|█████▌ | 24337/43939 [39:17:29<31:10:20, 5.72s/it]
55%|█████▌ | 24338/43939 [39:17:34<31:08:00, 5.72s/it]
55%|█████▌ | 24339/43939 [39:17:40<31:06:22, 5.71s/it]
55%|█████▌ | 24340/43939 [39:17:46<31:06:51, 5.72s/it]
55%|█████▌ | 24341/43939 [39:17:51<31:05:40, 5.71s/it]
55%|█████▌ | 24342/43939 [39:17:57<31:05:16, 5.71s/it]
55%|█████▌ | 24343/43939 [39:18:03<31:06:17, 5.71s/it]
55%|█████▌ | 24344/43939 [39:18:09<31:07:53, 5.72s/it]
55%|█████▌ | 24345/43939 [39:18:14<31:04:17, 5.71s/it]
55%|█████▌ | 24346/43939 [39:18:20<31:04:22, 5.71s/it]
55%|█████▌ | 24347/43939 [39:18:26<31:05:19, 5.71s/it]
55%|█████▌ | 24348/43939 [39:18:31<31:00:35, 5.70s/it]
55%|█████▌ | 24349/43939 [39:18:37<31:02:26, 5.70s/it]
55%|█████▌ | 24350/43939 [39:18:43<31:03:53, 5.71s/it]
{'loss': 0.7816, 'grad_norm': 1.1758830547332764, 'learning_rate': 2.4636568044290763e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.86, 'epoch': 2.22}
55%|█████▌ | 24350/43939 [39:18:43<31:03:53, 5.71s/it]
55%|█████▌ | 24351/43939 [39:18:49<31:05:56, 5.72s/it]
55%|█████▌ | 24352/43939 [39:18:55<31:29:15, 5.79s/it]
55%|█████▌ | 24353/43939 [39:19:00<31:20:41, 5.76s/it]
55%|█████▌ | 24354/43939 [39:19:06<31:17:48, 5.75s/it]
55%|█████▌ | 24355/43939 [39:19:12<31:11:32, 5.73s/it]
55%|█████▌ | 24356/43939 [39:19:17<31:08:56, 5.73s/it]
55%|█████▌ | 24357/43939 [39:19:23<31:08:52, 5.73s/it]
55%|█████▌ | 24358/43939 [39:19:29<31:11:38, 5.74s/it]
55%|█████▌ | 24359/43939 [39:19:35<31:08:34, 5.73s/it]
55%|█████▌ | 24360/43939 [39:19:40<31:05:47, 5.72s/it]
55%|█████▌ | 24361/43939 [39:19:46<31:00:06, 5.70s/it]
55%|█████▌ | 24362/43939 [39:19:52<31:00:00, 5.70s/it]
55%|█████▌ | 24363/43939 [39:19:57<30:59:43, 5.70s/it]
55%|█████▌ | 24364/43939 [39:20:03<30:56:19, 5.69s/it]
55%|█████▌ | 24365/43939 [39:20:09<31:00:36, 5.70s/it]
55%|█████▌ | 24366/43939 [39:20:14<31:00:29, 5.70s/it]
55%|█████▌ | 24367/43939 [39:20:20<31:02:20, 5.71s/it]
55%|█████▌ | 24368/43939 [39:20:26<31:05:07, 5.72s/it]
55%|█████▌ | 24369/43939 [39:20:32<31:01:32, 5.71s/it]
55%|█████▌ | 24370/43939 [39:20:37<30:59:30, 5.70s/it]
55%|█████▌ | 24371/43939 [39:20:43<30:59:40, 5.70s/it]
55%|█████▌ | 24372/43939 [39:20:49<30:58:44, 5.70s/it]
55%|█████▌ | 24373/43939 [39:20:54<30:55:39, 5.69s/it]
55%|█████▌ | 24374/43939 [39:21:00<31:18:10, 5.76s/it]
55%|█████▌ | 24375/43939 [39:21:06<31:10:41, 5.74s/it]
55%|█████▌ | 24376/43939 [39:21:12<31:08:42, 5.73s/it]
55%|█████▌ | 24377/43939 [39:21:17<31:05:18, 5.72s/it]
55%|█████▌ | 24378/43939 [39:21:23<31:08:26, 5.73s/it]
55%|█████▌ | 24379/43939 [39:21:29<31:06:55, 5.73s/it]
55%|█████▌ | 24380/43939 [39:21:34<31:05:39, 5.72s/it]
55%|█████▌ | 24381/43939 [39:21:40<31:02:02, 5.71s/it]
55%|█████▌ | 24382/43939 [39:21:46<31:00:26, 5.71s/it]
55%|█████▌ | 24383/43939 [39:21:52<31:02:53, 5.72s/it]
55%|█████▌ | 24384/43939 [39:21:57<31:02:41, 5.72s/it]
55%|█████▌ | 24385/43939 [39:22:03<31:01:46, 5.71s/it]
55%|█████▌ | 24386/43939 [39:22:09<31:00:13, 5.71s/it]
56%|█████▌ | 24387/43939 [39:22:14<31:02:17, 5.71s/it]
56%|█████▌ | 24388/43939 [39:22:20<31:00:32, 5.71s/it]
56%|█████▌ | 24389/43939 [39:22:26<30:59:28, 5.71s/it]
56%|█████▌ | 24390/43939 [39:22:32<31:02:52, 5.72s/it]
56%|█████▌ | 24391/43939 [39:22:37<31:00:19, 5.71s/it]
56%|█████▌ | 24392/43939 [39:22:43<31:01:07, 5.71s/it]
56%|█████▌ | 24393/43939 [39:22:49<31:03:16, 5.72s/it]
56%|█████▌ | 24394/43939 [39:22:54<31:04:28, 5.72s/it]
56%|█████▌ | 24395/43939 [39:23:00<31:31:13, 5.81s/it]
56%|█████▌ | 24396/43939 [39:23:06<31:25:16, 5.79s/it]
56%|█████▌ | 24397/43939 [39:23:12<31:16:23, 5.76s/it]
56%|█████▌ | 24398/43939 [39:23:18<31:06:34, 5.73s/it]
56%|█████▌ | 24399/43939 [39:23:23<31:02:04, 5.72s/it]
56%|█████▌ | 24400/43939 [39:23:29<30:58:50, 5.71s/it]
{'loss': 0.7831, 'grad_norm': 1.3471750020980835, 'learning_rate': 2.453727981940486e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.74, 'epoch': 2.22}
56%|█████▌ | 24400/43939 [39:23:29<30:58:50, 5.71s/it]
56%|█████▌ | 24401/43939 [39:23:35<30:58:35, 5.71s/it]
56%|█████▌ | 24402/43939 [39:23:40<30:55:56, 5.70s/it]
56%|█████▌ | 24403/43939 [39:23:46<30:57:06, 5.70s/it]
56%|█████▌ | 24404/43939 [39:23:52<31:05:15, 5.73s/it]
56%|█████▌ | 24405/43939 [39:23:58<31:01:40, 5.72s/it]
56%|█████▌ | 24406/43939 [39:24:03<31:02:40, 5.72s/it]
56%|█████▌ | 24407/43939 [39:24:09<31:01:20, 5.72s/it]
56%|█████▌ | 24408/43939 [39:24:15<31:00:51, 5.72s/it]
56%|█████▌ | 24409/43939 [39:24:20<30:57:43, 5.71s/it]
56%|█████▌ | 24410/43939 [39:24:26<30:55:58, 5.70s/it]
56%|█████▌ | 24411/43939 [39:24:32<30:55:21, 5.70s/it]
56%|█████▌ | 24412/43939 [39:24:37<30:55:56, 5.70s/it]
56%|█████▌ | 24413/43939 [39:24:43<30:53:03, 5.69s/it]
56%|█████▌ | 24414/43939 [39:24:49<30:53:34, 5.70s/it]
56%|█████▌ | 24415/43939 [39:24:55<30:55:19, 5.70s/it]
56%|█████▌ | 24416/43939 [39:25:00<31:16:36, 5.77s/it]
56%|█████▌ | 24417/43939 [39:25:06<31:10:56, 5.75s/it]
56%|█████▌ | 24418/43939 [39:25:12<31:04:13, 5.73s/it]
56%|█████▌ | 24419/43939 [39:25:18<30:59:55, 5.72s/it]
56%|█████▌ | 24420/43939 [39:25:23<31:00:41, 5.72s/it]
56%|█████▌ | 24421/43939 [39:25:29<30:58:10, 5.71s/it]
56%|█████▌ | 24422/43939 [39:25:35<31:01:47, 5.72s/it]
56%|█████▌ | 24423/43939 [39:25:40<30:58:46, 5.71s/it]
56%|█████▌ | 24424/43939 [39:25:46<30:56:23, 5.71s/it]
56%|█████▌ | 24425/43939 [39:25:52<31:05:35, 5.74s/it]
56%|█████▌ | 24426/43939 [39:25:58<31:08:20, 5.74s/it]
56%|█████▌ | 24427/43939 [39:26:03<31:03:37, 5.73s/it]
56%|█████▌ | 24428/43939 [39:26:09<31:00:12, 5.72s/it]
56%|█████▌ | 24429/43939 [39:26:15<30:55:51, 5.71s/it]
56%|█████▌ | 24430/43939 [39:26:20<30:55:54, 5.71s/it]
56%|█████▌ | 24431/43939 [39:26:26<30:56:22, 5.71s/it]
56%|█████▌ | 24432/43939 [39:26:32<30:55:22, 5.71s/it]
56%|█████▌ | 24433/43939 [39:26:38<30:57:16, 5.71s/it]
56%|█████▌ | 24434/43939 [39:26:43<30:58:04, 5.72s/it]
56%|█████▌ | 24435/43939 [39:26:49<30:54:24, 5.70s/it]
56%|█████▌ | 24436/43939 [39:26:55<30:52:44, 5.70s/it]
56%|█████▌ | 24437/43939 [39:27:00<30:48:14, 5.69s/it]
56%|█████▌ | 24438/43939 [39:27:06<31:10:53, 5.76s/it]
56%|█████▌ | 24439/43939 [39:27:12<31:06:01, 5.74s/it]
56%|█████▌ | 24440/43939 [39:27:18<31:06:25, 5.74s/it]
56%|█████▌ | 24441/43939 [39:27:23<31:04:03, 5.74s/it]
56%|█████▌ | 24442/43939 [39:27:29<31:00:25, 5.73s/it]
56%|█████▌ | 24443/43939 [39:27:35<30:56:28, 5.71s/it]
56%|█████▌ | 24444/43939 [39:27:41<30:53:33, 5.70s/it]
56%|█████▌ | 24445/43939 [39:27:46<30:54:08, 5.71s/it]
56%|█████▌ | 24446/43939 [39:27:52<30:53:23, 5.70s/it]
56%|█████▌ | 24447/43939 [39:27:58<30:52:57, 5.70s/it]
56%|█████▌ | 24448/43939 [39:28:03<30:53:13, 5.70s/it]
56%|█████▌ | 24449/43939 [39:28:09<30:53:06, 5.70s/it]
56%|█████▌ | 24450/43939 [39:28:15<30:53:20, 5.71s/it]
{'loss': 0.7814, 'grad_norm': 1.4884663820266724, 'learning_rate': 2.443799889501718e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.79, 'epoch': 2.23}
56%|█████▌ | 24450/43939 [39:28:15<30:53:20, 5.71s/it]
56%|█████▌ | 24451/43939 [39:28:20<30:54:39, 5.71s/it]
56%|█████▌ | 24452/43939 [39:28:26<30:57:07, 5.72s/it]
56%|█████▌ | 24453/43939 [39:28:32<30:52:11, 5.70s/it]
56%|█████▌ | 24454/43939 [39:28:38<30:50:24, 5.70s/it]
56%|█████▌ | 24455/43939 [39:28:43<30:52:08, 5.70s/it]
56%|█████▌ | 24456/43939 [39:28:49<30:48:13, 5.69s/it]
56%|█████▌ | 24457/43939 [39:28:55<30:49:51, 5.70s/it]
56%|█████▌ | 24458/43939 [39:29:00<30:50:12, 5.70s/it]
56%|█████▌ | 24459/43939 [39:29:06<31:11:22, 5.76s/it]
56%|█████▌ | 24460/43939 [39:29:12<31:04:57, 5.74s/it]
56%|█████▌ | 24461/43939 [39:29:18<31:04:57, 5.74s/it]
56%|█████▌ | 24462/43939 [39:29:23<31:04:54, 5.74s/it]
56%|█████▌ | 24463/43939 [39:29:29<31:00:45, 5.73s/it]
56%|█████▌ | 24464/43939 [39:29:35<30:56:35, 5.72s/it]
56%|█████▌ | 24465/43939 [39:29:41<30:58:14, 5.73s/it]
56%|█████▌ | 24466/43939 [39:29:46<30:55:35, 5.72s/it]
56%|█████▌ | 24467/43939 [39:29:52<30:50:45, 5.70s/it]
56%|█████▌ | 24468/43939 [39:29:58<30:52:43, 5.71s/it]
56%|█████▌ | 24469/43939 [39:30:03<30:53:38, 5.71s/it]
56%|█████▌ | 24470/43939 [39:30:09<30:54:44, 5.72s/it]
56%|█████▌ | 24471/43939 [39:30:15<30:53:18, 5.71s/it]
56%|█████▌ | 24472/43939 [39:30:21<30:51:14, 5.71s/it]
56%|█████▌ | 24473/43939 [39:30:26<30:55:56, 5.72s/it]
56%|█████▌ | 24474/43939 [39:30:32<30:55:25, 5.72s/it]
56%|█████▌ | 24475/43939 [39:30:38<30:55:28, 5.72s/it]
56%|█████▌ | 24476/43939 [39:30:43<30:54:00, 5.72s/it]
56%|█████▌ | 24477/43939 [39:30:49<30:52:34, 5.71s/it]
56%|█████▌ | 24478/43939 [39:30:55<30:53:02, 5.71s/it]
56%|█████▌ | 24479/43939 [39:31:01<30:53:22, 5.71s/it]
56%|█████▌ | 24480/43939 [39:31:06<31:15:32, 5.78s/it]
56%|█████▌ | 24481/43939 [39:31:12<31:12:16, 5.77s/it]
56%|█████▌ | 24482/43939 [39:31:18<31:04:43, 5.75s/it]
56%|█████▌ | 24483/43939 [39:31:24<31:01:06, 5.74s/it]
56%|█████▌ | 24484/43939 [39:31:29<30:58:16, 5.73s/it]
56%|█████▌ | 24485/43939 [39:31:35<30:54:30, 5.72s/it]
56%|█████▌ | 24486/43939 [39:31:41<30:54:14, 5.72s/it]
56%|█████▌ | 24487/43939 [39:31:46<30:53:03, 5.72s/it]
56%|█████▌ | 24488/43939 [39:31:52<30:52:43, 5.72s/it]
56%|█████▌ | 24489/43939 [39:31:58<30:54:29, 5.72s/it]
56%|█████▌ | 24490/43939 [39:32:04<30:56:01, 5.73s/it]
56%|█████▌ | 24491/43939 [39:32:09<30:58:34, 5.73s/it]
56%|█████▌ | 24492/43939 [39:32:15<30:57:32, 5.73s/it]
56%|█████▌ | 24493/43939 [39:32:21<30:54:39, 5.72s/it]
56%|█████▌ | 24494/43939 [39:32:27<30:51:09, 5.71s/it]
56%|█████▌ | 24495/43939 [39:32:32<30:48:58, 5.71s/it]
56%|█████▌ | 24496/43939 [39:32:38<30:47:21, 5.70s/it]
56%|█████▌ | 24497/43939 [39:32:44<30:48:38, 5.71s/it]
56%|█████▌ | 24498/43939 [39:32:49<30:47:46, 5.70s/it]
56%|█████▌ | 24499/43939 [39:32:55<30:54:18, 5.72s/it]
56%|█████▌ | 24500/43939 [39:33:01<30:51:05, 5.71s/it]
{'loss': 0.7873, 'grad_norm': 1.2731692790985107, 'learning_rate': 2.433872683751762e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.67, 'epoch': 2.23}
56%|█████▌ | 24500/43939 [39:33:01<30:51:05, 5.71s/it]
56%|█████▌ | 24501/43939 [39:33:07<31:19:24, 5.80s/it]
56%|█████▌ | 24502/43939 [39:33:12<31:09:20, 5.77s/it]
56%|█████▌ | 24503/43939 [39:33:18<31:00:06, 5.74s/it]
56%|█████▌ | 24504/43939 [39:33:24<30:55:01, 5.73s/it]
56%|█████▌ | 24505/43939 [39:33:30<30:52:21, 5.72s/it]
56%|█████▌ | 24506/43939 [39:33:35<30:50:31, 5.71s/it]
56%|█████▌ | 24507/43939 [39:33:41<30:54:32, 5.73s/it]
56%|█████▌ | 24508/43939 [39:33:47<30:54:14, 5.73s/it]
56%|█████▌ | 24509/43939 [39:33:52<30:50:17, 5.71s/it]
56%|█████▌ | 24510/43939 [39:33:58<30:50:08, 5.71s/it]
56%|█████▌ | 24511/43939 [39:34:04<30:52:58, 5.72s/it]
56%|█████▌ | 24512/43939 [39:34:10<30:46:38, 5.70s/it]
56%|█████▌ | 24513/43939 [39:34:15<30:48:48, 5.71s/it]
56%|█████▌ | 24514/43939 [39:34:21<30:46:29, 5.70s/it]
56%|█████▌ | 24515/43939 [39:34:27<30:49:57, 5.71s/it]
56%|█████▌ | 24516/43939 [39:34:32<30:48:06, 5.71s/it]
56%|█████▌ | 24517/43939 [39:34:38<30:45:20, 5.70s/it]
56%|█████▌ | 24518/43939 [39:34:44<30:44:57, 5.70s/it]
56%|█████▌ | 24519/43939 [39:34:49<30:45:09, 5.70s/it]
56%|█████▌ | 24520/43939 [39:34:55<30:44:58, 5.70s/it]
56%|█████▌ | 24521/43939 [39:35:01<30:42:24, 5.69s/it]
56%|█████▌ | 24522/43939 [39:35:07<30:41:21, 5.69s/it]
56%|█████▌ | 24523/43939 [39:35:12<31:01:04, 5.75s/it]
56%|█████▌ | 24524/43939 [39:35:18<30:57:57, 5.74s/it]
56%|█████▌ | 24525/43939 [39:35:24<30:51:41, 5.72s/it]
56%|█████▌ | 24526/43939 [39:35:30<30:49:26, 5.72s/it]
56%|█████▌ | 24527/43939 [39:35:35<30:47:56, 5.71s/it]
56%|█████▌ | 24528/43939 [39:35:41<30:50:07, 5.72s/it]
56%|█████▌ | 24529/43939 [39:35:47<30:50:07, 5.72s/it]
56%|█████▌ | 24530/43939 [39:35:52<30:45:26, 5.70s/it]
56%|█████▌ | 24531/43939 [39:35:58<30:44:54, 5.70s/it]
56%|█████▌ | 24532/43939 [39:36:04<30:45:09, 5.70s/it]
56%|█████▌ | 24533/43939 [39:36:09<30:45:21, 5.71s/it]
56%|█████▌ | 24534/43939 [39:36:15<30:44:21, 5.70s/it]
56%|█████▌ | 24535/43939 [39:36:21<30:45:12, 5.71s/it]
56%|█████▌ | 24536/43939 [39:36:27<30:43:10, 5.70s/it]
56%|█████▌ | 24537/43939 [39:36:32<30:42:58, 5.70s/it]
56%|█████▌ | 24538/43939 [39:36:38<30:43:05, 5.70s/it]
56%|█████▌ | 24539/43939 [39:36:44<30:46:53, 5.71s/it]
56%|█████▌ | 24540/43939 [39:36:49<30:43:45, 5.70s/it]
56%|█████▌ | 24541/43939 [39:36:55<30:45:43, 5.71s/it]
56%|█████▌ | 24542/43939 [39:37:01<30:44:13, 5.70s/it]
56%|█████▌ | 24543/43939 [39:37:07<30:49:04, 5.72s/it]
56%|█████▌ | 24544/43939 [39:37:13<31:10:00, 5.79s/it]
56%|█████▌ | 24545/43939 [39:37:18<31:02:45, 5.76s/it]
56%|█████▌ | 24546/43939 [39:37:24<30:55:51, 5.74s/it]
56%|█████▌ | 24547/43939 [39:37:30<30:54:25, 5.74s/it]
56%|█████▌ | 24548/43939 [39:37:35<30:50:26, 5.73s/it]
56%|█████▌ | 24549/43939 [39:37:41<30:50:31, 5.73s/it]
56%|█████▌ | 24550/43939 [39:37:47<30:51:08, 5.73s/it]
{'loss': 0.7994, 'grad_norm': 1.2478971481323242, 'learning_rate': 2.4239465213156158e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.44, 'epoch': 2.23}
56%|█████▌ | 24550/43939 [39:37:47<30:51:08, 5.73s/it]
56%|█████▌ | 24551/43939 [39:37:52<30:47:47, 5.72s/it]
56%|█████▌ | 24552/43939 [39:37:58<30:47:28, 5.72s/it]
56%|█████▌ | 24553/43939 [39:38:04<30:47:47, 5.72s/it]
56%|█████▌ | 24554/43939 [39:38:10<30:53:43, 5.74s/it]
56%|█████▌ | 24555/43939 [39:38:15<30:50:57, 5.73s/it]
56%|█████▌ | 24556/43939 [39:38:21<30:46:41, 5.72s/it]
56%|█████▌ | 24557/43939 [39:38:27<30:47:46, 5.72s/it]
56%|█████▌ | 24558/43939 [39:38:33<30:45:47, 5.71s/it]
56%|█████▌ | 24559/43939 [39:38:38<30:48:24, 5.72s/it]
56%|█████▌ | 24560/43939 [39:38:44<30:48:24, 5.72s/it]
56%|█████▌ | 24561/43939 [39:38:50<30:43:16, 5.71s/it]
56%|█████▌ | 24562/43939 [39:38:55<30:39:35, 5.70s/it]
56%|█████▌ | 24563/43939 [39:39:01<30:39:04, 5.69s/it]
56%|█████▌ | 24564/43939 [39:39:07<30:35:52, 5.69s/it]
56%|█████▌ | 24565/43939 [39:39:13<30:59:01, 5.76s/it]
56%|█████▌ | 24566/43939 [39:39:18<30:50:58, 5.73s/it]
56%|█████▌ | 24567/43939 [39:39:24<30:48:07, 5.72s/it]
56%|█████▌ | 24568/43939 [39:39:30<30:52:31, 5.74s/it]
56%|█████▌ | 24569/43939 [39:39:35<30:45:40, 5.72s/it]
56%|█████▌ | 24570/43939 [39:39:41<30:46:19, 5.72s/it]
56%|█████▌ | 24571/43939 [39:39:47<30:46:45, 5.72s/it]
56%|█████▌ | 24572/43939 [39:39:53<30:46:00, 5.72s/it]
56%|█████▌ | 24573/43939 [39:39:58<30:44:18, 5.71s/it]
56%|█████▌ | 24574/43939 [39:40:04<30:47:08, 5.72s/it]
56%|█████▌ | 24575/43939 [39:40:10<30:48:13, 5.73s/it]
56%|█████▌ | 24576/43939 [39:40:15<30:46:04, 5.72s/it]
56%|█████▌ | 24577/43939 [39:40:21<30:44:07, 5.71s/it]
56%|█████▌ | 24578/43939 [39:40:27<30:45:07, 5.72s/it]
56%|█████▌ | 24579/43939 [39:40:33<30:40:31, 5.70s/it]
56%|█████▌ | 24580/43939 [39:40:38<30:37:12, 5.69s/it]
56%|█████▌ | 24581/43939 [39:40:44<30:38:54, 5.70s/it]
56%|█████▌ | 24582/43939 [39:40:50<30:37:21, 5.70s/it]
56%|█████▌ | 24583/43939 [39:40:55<30:35:39, 5.69s/it]
56%|█████▌ | 24584/43939 [39:41:01<30:35:48, 5.69s/it]
56%|█████▌ | 24585/43939 [39:41:07<30:38:38, 5.70s/it]
56%|█████▌ | 24586/43939 [39:41:12<30:41:15, 5.71s/it]
56%|█████▌ | 24587/43939 [39:41:18<31:00:09, 5.77s/it]
56%|█████▌ | 24588/43939 [39:41:24<30:56:05, 5.76s/it]
56%|█████▌ | 24589/43939 [39:41:30<30:50:47, 5.74s/it]
56%|█████▌ | 24590/43939 [39:41:35<30:45:50, 5.72s/it]
56%|█████▌ | 24591/43939 [39:41:41<30:42:10, 5.71s/it]
56%|█████▌ | 24592/43939 [39:41:47<30:38:02, 5.70s/it]
56%|█████▌ | 24593/43939 [39:41:53<30:37:47, 5.70s/it]
56%|█████▌ | 24594/43939 [39:41:58<30:37:21, 5.70s/it]
56%|█████▌ | 24595/43939 [39:42:04<30:37:48, 5.70s/it]
56%|█████▌ | 24596/43939 [39:42:10<30:39:28, 5.71s/it]
56%|█████▌ | 24597/43939 [39:42:15<30:37:07, 5.70s/it]
56%|█████▌ | 24598/43939 [39:42:21<30:42:02, 5.71s/it]
56%|█████▌ | 24599/43939 [39:42:27<30:43:50, 5.72s/it]
56%|█████▌ | 24600/43939 [39:42:33<30:42:59, 5.72s/it]
{'loss': 0.7835, 'grad_norm': 1.6925835609436035, 'learning_rate': 2.4140215588018204e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.4, 'epoch': 2.24}
56%|█████▌ | 24600/43939 [39:42:33<30:42:59, 5.72s/it]
56%|█████▌ | 24601/43939 [39:42:38<30:40:10, 5.71s/it]
56%|█████▌ | 24602/43939 [39:42:44<30:45:14, 5.73s/it]
56%|█████▌ | 24603/43939 [39:42:50<30:41:45, 5.72s/it]
56%|█████▌ | 24604/43939 [39:42:55<30:44:55, 5.73s/it]
56%|█████▌ | 24605/43939 [39:43:01<30:39:10, 5.71s/it]
56%|█████▌ | 24606/43939 [39:43:07<30:42:31, 5.72s/it]
56%|█████▌ | 24607/43939 [39:43:13<30:48:59, 5.74s/it]
56%|█████▌ | 24608/43939 [39:43:19<31:14:08, 5.82s/it]
56%|█████▌ | 24609/43939 [39:43:24<31:02:11, 5.78s/it]
56%|█████▌ | 24610/43939 [39:43:30<31:01:23, 5.78s/it]
56%|█████▌ | 24611/43939 [39:43:36<30:54:11, 5.76s/it]
56%|█████▌ | 24612/43939 [39:43:42<30:49:03, 5.74s/it]
56%|█████▌ | 24613/43939 [39:43:47<30:46:35, 5.73s/it]
56%|█████▌ | 24614/43939 [39:43:53<30:42:36, 5.72s/it]
56%|█████▌ | 24615/43939 [39:43:59<30:37:45, 5.71s/it]
56%|█████▌ | 24616/43939 [39:44:04<30:33:45, 5.69s/it]
56%|█████▌ | 24617/43939 [39:44:10<30:31:31, 5.69s/it]
56%|█████▌ | 24618/43939 [39:44:16<30:30:24, 5.68s/it]
56%|█████▌ | 24619/43939 [39:44:21<30:28:27, 5.68s/it]
56%|█████▌ | 24620/43939 [39:44:27<30:29:42, 5.68s/it]
56%|█████▌ | 24621/43939 [39:44:33<30:32:50, 5.69s/it]
56%|█████▌ | 24622/43939 [39:44:38<30:32:35, 5.69s/it]
56%|█████▌ | 24623/43939 [39:44:44<30:36:17, 5.70s/it]
56%|█████▌ | 24624/43939 [39:44:50<30:35:42, 5.70s/it]
56%|█████▌ | 24625/43939 [39:44:56<30:36:39, 5.71s/it]
56%|█████▌ | 24626/43939 [39:45:01<30:36:36, 5.71s/it]
56%|█████▌ | 24627/43939 [39:45:07<30:33:43, 5.70s/it]
56%|█████▌ | 24628/43939 [39:45:13<30:30:56, 5.69s/it]
56%|█████▌ | 24629/43939 [39:45:19<30:54:44, 5.76s/it]
56%|█████▌ | 24630/43939 [39:45:24<30:46:38, 5.74s/it]
56%|█████▌ | 24631/43939 [39:45:30<30:45:32, 5.74s/it]
56%|█████▌ | 24632/43939 [39:45:36<30:38:35, 5.71s/it]
56%|█████▌ | 24633/43939 [39:45:41<30:39:50, 5.72s/it]
56%|█████▌ | 24634/43939 [39:45:47<30:37:41, 5.71s/it]
56%|█████▌ | 24635/43939 [39:45:53<30:38:18, 5.71s/it]
56%|█████▌ | 24636/43939 [39:45:58<30:35:20, 5.70s/it]
56%|█████▌ | 24637/43939 [39:46:04<30:31:56, 5.69s/it]
56%|█████▌ | 24638/43939 [39:46:10<30:30:05, 5.69s/it]
56%|█████▌ | 24639/43939 [39:46:15<30:31:31, 5.69s/it]
56%|█████▌ | 24640/43939 [39:46:21<30:30:32, 5.69s/it]
56%|█████▌ | 24641/43939 [39:46:27<30:29:34, 5.69s/it]
56%|█████▌ | 24642/43939 [39:46:33<30:30:24, 5.69s/it]
56%|█████▌ | 24643/43939 [39:46:38<30:31:24, 5.69s/it]
56%|█████▌ | 24644/43939 [39:46:44<30:32:06, 5.70s/it]
56%|█████▌ | 24645/43939 [39:46:50<30:31:08, 5.69s/it]
56%|█████▌ | 24646/43939 [39:46:55<30:28:53, 5.69s/it]
56%|█████▌ | 24647/43939 [39:47:01<30:31:18, 5.70s/it]
56%|█████▌ | 24648/43939 [39:47:07<30:32:36, 5.70s/it]
56%|█████▌ | 24649/43939 [39:47:12<30:35:16, 5.71s/it]
56%|█████▌ | 24650/43939 [39:47:18<30:56:04, 5.77s/it]
{'loss': 0.7846, 'grad_norm': 1.3365802764892578, 'learning_rate': 2.404097952799981e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.07, 'epoch': 2.24}
56%|█████▌ | 24650/43939 [39:47:18<30:56:04, 5.77s/it]
56%|█████▌ | 24651/43939 [39:47:24<30:51:32, 5.76s/it]
56%|█████▌ | 24652/43939 [39:47:30<30:45:11, 5.74s/it]
56%|█████▌ | 24653/43939 [39:47:35<30:39:02, 5.72s/it]
56%|█████▌ | 24654/43939 [39:47:41<30:38:00, 5.72s/it]
56%|█████▌ | 24655/43939 [39:47:47<30:37:33, 5.72s/it]
56%|█████▌ | 24656/43939 [39:47:53<30:37:38, 5.72s/it]
56%|█████▌ | 24657/43939 [39:47:58<30:36:54, 5.72s/it]
56%|█████▌ | 24658/43939 [39:48:04<30:38:59, 5.72s/it]
56%|█████▌ | 24659/43939 [39:48:10<30:37:33, 5.72s/it]
56%|█████▌ | 24660/43939 [39:48:15<30:35:46, 5.71s/it]
56%|█████▌ | 24661/43939 [39:48:21<30:33:19, 5.71s/it]
56%|█████▌ | 24662/43939 [39:48:27<30:31:20, 5.70s/it]
56%|█████▌ | 24663/43939 [39:48:33<30:30:09, 5.70s/it]
56%|█████▌ | 24664/43939 [39:48:38<30:33:37, 5.71s/it]
56%|█████▌ | 24665/43939 [39:48:44<30:34:49, 5.71s/it]
56%|█████▌ | 24666/43939 [39:48:50<30:31:05, 5.70s/it]
56%|█████▌ | 24667/43939 [39:48:55<30:32:19, 5.70s/it]
56%|█████▌ | 24668/43939 [39:49:01<30:32:21, 5.70s/it]
56%|█████▌ | 24669/43939 [39:49:07<30:37:10, 5.72s/it]
56%|█████▌ | 24670/43939 [39:49:13<30:36:18, 5.72s/it]
56%|█████▌ | 24671/43939 [39:49:18<30:37:42, 5.72s/it]
56%|█████▌ | 24672/43939 [39:49:24<30:53:16, 5.77s/it]
56%|█████▌ | 24673/43939 [39:49:30<30:44:51, 5.75s/it]
56%|█████▌ | 24674/43939 [39:49:36<30:40:16, 5.73s/it]
56%|█████▌ | 24675/43939 [39:49:41<30:33:32, 5.71s/it]
56%|█████▌ | 24676/43939 [39:49:47<30:33:46, 5.71s/it]
56%|█████▌ | 24677/43939 [39:49:53<30:31:48, 5.71s/it]
56%|█████▌ | 24678/43939 [39:49:58<30:31:38, 5.71s/it]
56%|█████▌ | 24679/43939 [39:50:04<30:34:05, 5.71s/it]
56%|█████▌ | 24680/43939 [39:50:10<30:37:03, 5.72s/it]
56%|█████▌ | 24681/43939 [39:50:15<30:32:42, 5.71s/it]
56%|█████▌ | 24682/43939 [39:50:21<30:33:38, 5.71s/it]
56%|█████▌ | 24683/43939 [39:50:27<30:35:58, 5.72s/it]
56%|█████▌ | 24684/43939 [39:50:33<30:31:50, 5.71s/it]
56%|█████▌ | 24685/43939 [39:50:38<30:27:23, 5.69s/it]
56%|█████▌ | 24686/43939 [39:50:44<30:28:56, 5.70s/it]
56%|█████▌ | 24687/43939 [39:50:50<30:27:28, 5.70s/it]
56%|█████▌ | 24688/43939 [39:50:55<30:24:35, 5.69s/it]
56%|█████▌ | 24689/43939 [39:51:01<30:25:36, 5.69s/it]
56%|█████▌ | 24690/43939 [39:51:07<30:26:57, 5.69s/it]
56%|█████▌ | 24691/43939 [39:51:12<30:24:42, 5.69s/it]
56%|█████▌ | 24692/43939 [39:51:18<30:23:47, 5.69s/it]
56%|█████▌ | 24693/43939 [39:51:24<30:51:23, 5.77s/it]
56%|█████▌ | 24694/43939 [39:51:30<30:49:03, 5.76s/it]
56%|█████▌ | 24695/43939 [39:51:36<30:42:15, 5.74s/it]
56%|█████▌ | 24696/43939 [39:51:41<30:40:22, 5.74s/it]
56%|█████▌ | 24697/43939 [39:51:47<30:34:34, 5.72s/it]
56%|█████▌ | 24698/43939 [39:51:53<30:31:04, 5.71s/it]
56%|█████▌ | 24699/43939 [39:51:58<30:27:27, 5.70s/it]
56%|█████▌ | 24700/43939 [39:52:04<30:28:12, 5.70s/it]
{'loss': 0.7863, 'grad_norm': 1.4315035343170166, 'learning_rate': 2.3941758598783027e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.25, 'epoch': 2.25}
56%|█████▌ | 24700/43939 [39:52:04<30:28:12, 5.70s/it]
56%|█████▌ | 24701/43939 [39:52:10<30:28:20, 5.70s/it]
56%|█████▌ | 24702/43939 [39:52:15<30:25:43, 5.69s/it]
56%|█████▌ | 24703/43939 [39:52:21<30:24:35, 5.69s/it]
56%|█████▌ | 24704/43939 [39:52:27<30:27:52, 5.70s/it]
56%|█████▌ | 24705/43939 [39:52:33<30:30:26, 5.71s/it]
56%|█████▌ | 24706/43939 [39:52:38<30:28:38, 5.70s/it]
56%|█████▌ | 24707/43939 [39:52:44<30:34:52, 5.72s/it]
56%|█████▌ | 24708/43939 [39:52:50<30:35:18, 5.73s/it]
56%|█████▌ | 24709/43939 [39:52:55<30:35:18, 5.73s/it]
56%|█████▌ | 24710/43939 [39:53:01<30:30:03, 5.71s/it]
56%|█████▌ | 24711/43939 [39:53:07<30:23:27, 5.69s/it]
56%|█████▌ | 24712/43939 [39:53:12<30:23:57, 5.69s/it]
56%|█████▌ | 24713/43939 [39:53:18<30:21:08, 5.68s/it]
56%|█████▌ | 24714/43939 [39:53:24<30:48:58, 5.77s/it]
56%|█████▌ | 24715/43939 [39:53:30<30:44:11, 5.76s/it]
56%|█████▋ | 24716/43939 [39:53:36<30:39:48, 5.74s/it]
56%|█████▋ | 24717/43939 [39:53:41<30:32:50, 5.72s/it]
56%|█████▋ | 24718/43939 [39:53:47<30:26:50, 5.70s/it]
56%|█████▋ | 24719/43939 [39:53:53<30:25:42, 5.70s/it]
56%|█████▋ | 24720/43939 [39:53:58<30:21:39, 5.69s/it]
56%|█████▋ | 24721/43939 [39:54:04<30:24:11, 5.70s/it]
56%|█████▋ | 24722/43939 [39:54:10<30:21:59, 5.69s/it]
56%|█████▋ | 24723/43939 [39:54:15<30:27:54, 5.71s/it]
56%|█████▋ | 24724/43939 [39:54:21<30:26:57, 5.70s/it]
56%|█████▋ | 24725/43939 [39:54:27<30:29:21, 5.71s/it]
56%|█████▋ | 24726/43939 [39:54:32<30:25:13, 5.70s/it]
56%|█████▋ | 24727/43939 [39:54:38<30:22:49, 5.69s/it]
56%|█████▋ | 24728/43939 [39:54:44<30:23:04, 5.69s/it]
56%|█████▋ | 24729/43939 [39:54:50<30:28:16, 5.71s/it]
56%|█████▋ | 24730/43939 [39:54:55<30:30:00, 5.72s/it]
56%|█████▋ | 24731/43939 [39:55:01<30:27:48, 5.71s/it]
56%|█████▋ | 24732/43939 [39:55:07<30:26:06, 5.70s/it]
56%|█████▋ | 24733/43939 [39:55:12<30:23:37, 5.70s/it]
56%|█████▋ | 24734/43939 [39:55:18<30:25:55, 5.70s/it]
56%|█████▋ | 24735/43939 [39:55:24<30:26:08, 5.71s/it]
56%|█████▋ | 24736/43939 [39:55:30<30:47:40, 5.77s/it]
56%|█████▋ | 24737/43939 [39:55:35<30:46:29, 5.77s/it]
56%|█████▋ | 24738/43939 [39:55:41<30:41:29, 5.75s/it]
56%|█████▋ | 24739/43939 [39:55:47<30:39:02, 5.75s/it]
56%|█████▋ | 24740/43939 [39:55:53<30:35:00, 5.73s/it]
56%|█████▋ | 24741/43939 [39:55:58<30:29:35, 5.72s/it]
56%|█████▋ | 24742/43939 [39:56:04<30:27:22, 5.71s/it]
56%|█████▋ | 24743/43939 [39:56:10<30:24:27, 5.70s/it]
56%|█████▋ | 24744/43939 [39:56:15<30:25:27, 5.71s/it]
56%|█████▋ | 24745/43939 [39:56:21<30:28:58, 5.72s/it]
56%|█████▋ | 24746/43939 [39:56:27<30:26:04, 5.71s/it]
56%|█████▋ | 24747/43939 [39:56:32<30:21:38, 5.70s/it]
56%|█████▋ | 24748/43939 [39:56:38<30:18:48, 5.69s/it]
56%|█████▋ | 24749/43939 [39:56:44<30:17:42, 5.68s/it]
56%|█████▋ | 24750/43939 [39:56:49<30:15:30, 5.68s/it]
{'loss': 0.7891, 'grad_norm': 1.3341989517211914, 'learning_rate': 2.384255436581118e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.09, 'epoch': 2.25}
56%|█████▋ | 24750/43939 [39:56:50<30:15:30, 5.68s/it]
56%|█████▋ | 24751/43939 [39:56:55<30:22:02, 5.70s/it]
56%|█████▋ | 24752/43939 [39:57:01<30:26:07, 5.71s/it]
56%|█████▋ | 24753/43939 [39:57:07<30:21:15, 5.70s/it]
56%|█████▋ | 24754/43939 [39:57:12<30:23:44, 5.70s/it]
56%|█████▋ | 24755/43939 [39:57:18<30:24:42, 5.71s/it]
56%|█████▋ | 24756/43939 [39:57:24<30:23:51, 5.70s/it]
56%|█████▋ | 24757/43939 [39:57:30<30:41:02, 5.76s/it]
56%|█████▋ | 24758/43939 [39:57:35<30:40:32, 5.76s/it]
56%|█████▋ | 24759/43939 [39:57:41<30:31:36, 5.73s/it]
56%|█████▋ | 24760/43939 [39:57:47<30:30:30, 5.73s/it]
56%|█████▋ | 24761/43939 [39:57:52<30:26:26, 5.71s/it]
56%|█████▋ | 24762/43939 [39:57:58<30:26:31, 5.71s/it]
56%|█████▋ | 24763/43939 [39:58:04<30:23:39, 5.71s/it]
56%|█████▋ | 24764/43939 [39:58:10<30:19:50, 5.69s/it]
56%|█████▋ | 24765/43939 [39:58:15<30:26:34, 5.72s/it]
56%|█████▋ | 24766/43939 [39:58:21<30:25:40, 5.71s/it]
56%|█████▋ | 24767/43939 [39:58:27<30:24:27, 5.71s/it]
56%|█████▋ | 24768/43939 [39:58:32<30:23:10, 5.71s/it]
56%|█████▋ | 24769/43939 [39:58:38<30:21:37, 5.70s/it]
56%|█████▋ | 24770/43939 [39:58:44<30:20:17, 5.70s/it]
56%|█████▋ | 24771/43939 [39:58:50<30:19:23, 5.70s/it]
56%|█████▋ | 24772/43939 [39:58:55<30:20:59, 5.70s/it]
56%|█████▋ | 24773/43939 [39:59:01<30:20:23, 5.70s/it]
56%|█████▋ | 24774/43939 [39:59:07<30:24:28, 5.71s/it]
56%|█████▋ | 24775/43939 [39:59:12<30:25:11, 5.71s/it]
56%|█████▋ | 24776/43939 [39:59:18<30:24:23, 5.71s/it]
56%|█████▋ | 24777/43939 [39:59:24<30:28:49, 5.73s/it]
56%|█████▋ | 24778/43939 [39:59:30<30:49:42, 5.79s/it]
56%|█████▋ | 24779/43939 [39:59:36<30:44:56, 5.78s/it]
56%|█████▋ | 24780/43939 [39:59:41<30:41:26, 5.77s/it]
56%|█████▋ | 24781/43939 [39:59:47<30:32:29, 5.74s/it]
56%|█████▋ | 24782/43939 [39:59:53<30:29:26, 5.73s/it]
56%|█████▋ | 24783/43939 [39:59:58<30:24:39, 5.72s/it]
56%|█████▋ | 24784/43939 [40:00:04<30:25:11, 5.72s/it]
56%|█████▋ | 24785/43939 [40:00:10<30:25:10, 5.72s/it]
56%|█████▋ | 24786/43939 [40:00:16<30:28:00, 5.73s/it]
56%|█████▋ | 24787/43939 [40:00:21<30:24:08, 5.71s/it]
56%|█████▋ | 24788/43939 [40:00:27<30:22:45, 5.71s/it]
56%|█████▋ | 24789/43939 [40:00:33<30:17:55, 5.70s/it]
56%|█████▋ | 24790/43939 [40:00:38<30:18:47, 5.70s/it]
56%|█████▋ | 24791/43939 [40:00:44<30:16:01, 5.69s/it]
56%|█████▋ | 24792/43939 [40:00:50<30:18:19, 5.70s/it]
56%|█████▋ | 24793/43939 [40:00:55<30:26:52, 5.73s/it]
56%|█████▋ | 24794/43939 [40:01:01<30:23:32, 5.71s/it]
56%|█████▋ | 24795/43939 [40:01:07<30:26:48, 5.73s/it]
56%|█████▋ | 24796/43939 [40:01:13<30:23:57, 5.72s/it]
56%|█████▋ | 24797/43939 [40:01:18<30:21:36, 5.71s/it]
56%|█████▋ | 24798/43939 [40:01:24<30:19:16, 5.70s/it]
56%|█████▋ | 24799/43939 [40:01:30<30:43:31, 5.78s/it]
56%|█████▋ | 24800/43939 [40:01:36<30:31:20, 5.74s/it]
{'loss': 0.7807, 'grad_norm': 1.4974217414855957, 'learning_rate': 2.374336839426418e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.86, 'epoch': 2.26}
56%|█████▋ | 24800/43939 [40:01:36<30:31:20, 5.74s/it]
56%|█████▋ | 24801/43939 [40:01:41<30:27:29, 5.73s/it]
56%|█████▋ | 24802/43939 [40:01:47<30:27:35, 5.73s/it]
56%|█████▋ | 24803/43939 [40:01:53<30:32:32, 5.75s/it]
56%|█████▋ | 24804/43939 [40:01:59<30:35:56, 5.76s/it]
56%|█████▋ | 24805/43939 [40:02:04<30:29:33, 5.74s/it]
56%|█████▋ | 24806/43939 [40:02:10<30:25:15, 5.72s/it]
56%|█████▋ | 24807/43939 [40:02:16<30:21:18, 5.71s/it]
56%|█████▋ | 24808/43939 [40:02:21<30:18:54, 5.70s/it]
56%|█████▋ | 24809/43939 [40:02:27<30:18:49, 5.70s/it]
56%|█████▋ | 24810/43939 [40:02:33<30:18:49, 5.70s/it]
56%|█████▋ | 24811/43939 [40:02:38<30:19:02, 5.71s/it]
56%|█████▋ | 24812/43939 [40:02:44<30:20:21, 5.71s/it]
56%|█████▋ | 24813/43939 [40:02:50<30:18:59, 5.71s/it]
56%|█████▋ | 24814/43939 [40:02:56<30:17:44, 5.70s/it]
56%|█████▋ | 24815/43939 [40:03:01<30:19:12, 5.71s/it]
56%|█████▋ | 24816/43939 [40:03:07<30:19:09, 5.71s/it]
56%|█████▋ | 24817/43939 [40:03:13<30:20:10, 5.71s/it]
56%|█████▋ | 24818/43939 [40:03:18<30:17:35, 5.70s/it]
56%|█████▋ | 24819/43939 [40:03:24<30:17:20, 5.70s/it]
56%|█████▋ | 24820/43939 [40:03:30<30:12:26, 5.69s/it]
56%|█████▋ | 24821/43939 [40:03:36<30:33:36, 5.75s/it]
56%|█████▋ | 24822/43939 [40:03:41<30:28:34, 5.74s/it]
56%|█████▋ | 24823/43939 [40:03:47<30:24:33, 5.73s/it]
56%|█████▋ | 24824/43939 [40:03:53<30:17:59, 5.71s/it]
56%|█████▋ | 24825/43939 [40:03:58<30:16:11, 5.70s/it]
57%|█████▋ | 24826/43939 [40:04:04<30:17:01, 5.70s/it]
57%|█████▋ | 24827/43939 [40:04:10<30:17:34, 5.71s/it]
57%|█████▋ | 24828/43939 [40:04:16<30:20:08, 5.71s/it]
57%|█████▋ | 24829/43939 [40:04:21<30:14:38, 5.70s/it]
57%|█████▋ | 24830/43939 [40:04:27<30:14:19, 5.70s/it]
57%|█████▋ | 24831/43939 [40:04:33<30:16:08, 5.70s/it]
57%|█████▋ | 24832/43939 [40:04:38<30:20:12, 5.72s/it]
57%|█████▋ | 24833/43939 [40:04:44<30:18:57, 5.71s/it]
57%|█████▋ | 24834/43939 [40:04:50<30:21:39, 5.72s/it]
57%|█████▋ | 24835/43939 [40:04:55<30:16:03, 5.70s/it]
57%|█████▋ | 24836/43939 [40:05:01<30:14:55, 5.70s/it]
57%|█████▋ | 24837/43939 [40:05:07<30:11:37, 5.69s/it]
57%|█████▋ | 24838/43939 [40:05:13<30:15:54, 5.70s/it]
57%|█████▋ | 24839/43939 [40:05:18<30:12:40, 5.69s/it]
57%|█████▋ | 24840/43939 [40:05:24<30:18:27, 5.71s/it]
57%|█████▋ | 24841/43939 [40:05:30<30:19:42, 5.72s/it]
57%|█████▋ | 24842/43939 [40:05:35<30:19:48, 5.72s/it]
57%|█████▋ | 24843/43939 [40:05:41<30:40:48, 5.78s/it]
57%|█████▋ | 24844/43939 [40:05:47<30:34:42, 5.76s/it]
57%|█████▋ | 24845/43939 [40:05:53<30:29:04, 5.75s/it]
57%|█████▋ | 24846/43939 [40:05:59<30:23:05, 5.73s/it]
57%|█████▋ | 24847/43939 [40:06:04<30:22:04, 5.73s/it]
57%|█████▋ | 24848/43939 [40:06:10<30:19:33, 5.72s/it]
57%|█████▋ | 24849/43939 [40:06:16<30:19:55, 5.72s/it]
57%|█████▋ | 24850/43939 [40:06:21<30:23:52, 5.73s/it]
{'loss': 0.766, 'grad_norm': 1.382554531097412, 'learning_rate': 2.3644202249033794e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.26, 'epoch': 2.26}
57%|█████▋ | 24850/43939 [40:06:21<30:23:52, 5.73s/it]
57%|█████▋ | 24851/43939 [40:06:27<30:23:59, 5.73s/it]
57%|█████▋ | 24852/43939 [40:06:33<30:21:17, 5.73s/it]
57%|█████▋ | 24853/43939 [40:06:39<30:20:43, 5.72s/it]
57%|█████▋ | 24854/43939 [40:06:44<30:18:16, 5.72s/it]
57%|█████▋ | 24855/43939 [40:06:50<30:16:33, 5.71s/it]
57%|█████▋ | 24856/43939 [40:06:56<30:16:36, 5.71s/it]
57%|█████▋ | 24857/43939 [40:07:01<30:18:16, 5.72s/it]
57%|█████▋ | 24858/43939 [40:07:07<30:14:44, 5.71s/it]
57%|█████▋ | 24859/43939 [40:07:13<30:15:09, 5.71s/it]
57%|█████▋ | 24860/43939 [40:07:19<30:14:32, 5.71s/it]
57%|█████▋ | 24861/43939 [40:07:24<30:13:20, 5.70s/it]
57%|█████▋ | 24862/43939 [40:07:30<30:13:04, 5.70s/it]
57%|█████▋ | 24863/43939 [40:07:36<30:31:46, 5.76s/it]
57%|█████▋ | 24864/43939 [40:07:42<30:25:04, 5.74s/it]
57%|█████▋ | 24865/43939 [40:07:47<30:21:14, 5.73s/it]
57%|█████▋ | 24866/43939 [40:07:53<30:18:06, 5.72s/it]
57%|█████▋ | 24867/43939 [40:07:59<30:15:12, 5.71s/it]
57%|█████▋ | 24868/43939 [40:08:04<30:14:35, 5.71s/it]
57%|█████▋ | 24869/43939 [40:08:10<30:13:54, 5.71s/it]
57%|█████▋ | 24870/43939 [40:08:16<30:17:05, 5.72s/it]
57%|█████▋ | 24871/43939 [40:08:21<30:12:20, 5.70s/it]
57%|█████▋ | 24872/43939 [40:08:27<30:12:46, 5.70s/it]
57%|█████▋ | 24873/43939 [40:08:33<30:08:03, 5.69s/it]
57%|█████▋ | 24874/43939 [40:08:38<30:10:42, 5.70s/it]
57%|█████▋ | 24875/43939 [40:08:44<30:12:42, 5.71s/it]
57%|█████▋ | 24876/43939 [40:08:50<30:11:12, 5.70s/it]
57%|█████▋ | 24877/43939 [40:08:56<30:10:02, 5.70s/it]
57%|█████▋ | 24878/43939 [40:09:01<30:16:32, 5.72s/it]
57%|█████▋ | 24879/43939 [40:09:07<30:20:34, 5.73s/it]
57%|█████▋ | 24880/43939 [40:09:13<30:21:07, 5.73s/it]
57%|█████▋ | 24881/43939 [40:09:19<30:19:52, 5.73s/it]
57%|█████▋ | 24882/43939 [40:09:24<30:15:00, 5.71s/it]
57%|█████▋ | 24883/43939 [40:09:30<30:09:48, 5.70s/it]
57%|█████▋ | 24884/43939 [40:09:36<30:27:17, 5.75s/it]
57%|█████▋ | 24885/43939 [40:09:42<30:23:09, 5.74s/it]
57%|█████▋ | 24886/43939 [40:09:47<30:19:29, 5.73s/it]
57%|█████▋ | 24887/43939 [40:09:53<30:21:08, 5.74s/it]
57%|█████▋ | 24888/43939 [40:09:59<30:19:24, 5.73s/it]
57%|█████▋ | 24889/43939 [40:10:04<30:15:12, 5.72s/it]
57%|█████▋ | 24890/43939 [40:10:10<30:11:26, 5.71s/it]
57%|█████▋ | 24891/43939 [40:10:16<30:08:30, 5.70s/it]
57%|█████▋ | 24892/43939 [40:10:21<30:07:48, 5.69s/it]
57%|█████▋ | 24893/43939 [40:10:27<30:10:07, 5.70s/it]
57%|█████▋ | 24894/43939 [40:10:33<30:08:02, 5.70s/it]
57%|█████▋ | 24895/43939 [40:10:39<30:09:52, 5.70s/it]
57%|█████▋ | 24896/43939 [40:10:44<30:07:57, 5.70s/it]
57%|█████▋ | 24897/43939 [40:10:50<30:06:10, 5.69s/it]
57%|█████▋ | 24898/43939 [40:10:56<30:05:40, 5.69s/it]
57%|█████▋ | 24899/43939 [40:11:01<30:06:34, 5.69s/it]
57%|█████▋ | 24900/43939 [40:11:07<30:06:18, 5.69s/it]
{'loss': 0.7652, 'grad_norm': 1.6081000566482544, 'learning_rate': 2.3545057494699013e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.15, 'epoch': 2.27}
57%|█████▋ | 24900/43939 [40:11:07<30:06:18, 5.69s/it]
57%|█████▋ | 24901/43939 [40:11:13<30:10:13, 5.71s/it]
57%|█████▋ | 24902/43939 [40:11:18<30:12:40, 5.71s/it]
57%|█████▋ | 24903/43939 [40:11:24<30:12:25, 5.71s/it]
57%|█████▋ | 24904/43939 [40:11:30<30:14:27, 5.72s/it]
57%|█████▋ | 24905/43939 [40:11:36<30:32:51, 5.78s/it]
57%|█████▋ | 24906/43939 [40:11:41<30:22:58, 5.75s/it]
57%|█████▋ | 24907/43939 [40:11:47<30:18:12, 5.73s/it]
57%|█████▋ | 24908/43939 [40:11:53<30:15:44, 5.72s/it]
57%|█████▋ | 24909/43939 [40:11:59<30:19:30, 5.74s/it]
57%|█████▋ | 24910/43939 [40:12:04<30:19:00, 5.74s/it]
57%|█████▋ | 24911/43939 [40:12:10<30:17:58, 5.73s/it]
57%|█████▋ | 24912/43939 [40:12:16<30:12:53, 5.72s/it]
57%|█████▋ | 24913/43939 [40:12:22<30:14:00, 5.72s/it]
57%|█████▋ | 24914/43939 [40:12:27<30:11:28, 5.71s/it]
57%|█████▋ | 24915/43939 [40:12:33<30:09:55, 5.71s/it]
57%|█████▋ | 24916/43939 [40:12:39<30:13:39, 5.72s/it]
57%|█████▋ | 24917/43939 [40:12:44<30:14:53, 5.72s/it]
57%|█████▋ | 24918/43939 [40:12:50<30:09:48, 5.71s/it]
57%|█████▋ | 24919/43939 [40:12:56<30:13:48, 5.72s/it]
57%|█████▋ | 24920/43939 [40:13:02<30:12:40, 5.72s/it]
57%|█████▋ | 24921/43939 [40:13:07<30:11:06, 5.71s/it]
57%|█████▋ | 24922/43939 [40:13:13<30:08:18, 5.71s/it]
57%|█████▋ | 24923/43939 [40:13:19<30:07:10, 5.70s/it]
57%|█████▋ | 24924/43939 [40:13:24<30:07:13, 5.70s/it]
57%|█████▋ | 24925/43939 [40:13:30<30:07:24, 5.70s/it]
57%|█████▋ | 24926/43939 [40:13:36<30:29:23, 5.77s/it]
57%|█████▋ | 24927/43939 [40:13:42<30:20:33, 5.75s/it]
57%|█████▋ | 24928/43939 [40:13:47<30:14:12, 5.73s/it]
57%|█████▋ | 24929/43939 [40:13:53<30:07:57, 5.71s/it]
57%|█████▋ | 24930/43939 [40:13:59<30:04:36, 5.70s/it]
57%|█████▋ | 24931/43939 [40:14:04<30:06:53, 5.70s/it]
57%|█████▋ | 24932/43939 [40:14:10<30:07:18, 5.71s/it]
57%|█████▋ | 24933/43939 [40:14:16<30:08:06, 5.71s/it]
57%|█████▋ | 24934/43939 [40:14:22<30:09:32, 5.71s/it]
57%|█████▋ | 24935/43939 [40:14:27<30:09:59, 5.71s/it]
57%|█████▋ | 24936/43939 [40:14:33<30:07:17, 5.71s/it]
57%|█████▋ | 24937/43939 [40:14:39<30:06:32, 5.70s/it]
57%|█████▋ | 24938/43939 [40:14:44<30:05:27, 5.70s/it]
57%|█████▋ | 24939/43939 [40:14:50<30:02:56, 5.69s/it]
57%|█████▋ | 24940/43939 [40:14:56<30:06:40, 5.71s/it]
57%|█████▋ | 24941/43939 [40:15:01<30:07:13, 5.71s/it]
57%|█████▋ | 24942/43939 [40:15:07<30:02:50, 5.69s/it]
57%|█████▋ | 24943/43939 [40:15:13<30:03:32, 5.70s/it]
57%|█████▋ | 24944/43939 [40:15:18<30:03:36, 5.70s/it]
57%|█████▋ | 24945/43939 [40:15:24<30:05:12, 5.70s/it]
57%|█████▋ | 24946/43939 [40:15:30<30:03:09, 5.70s/it]
57%|█████▋ | 24947/43939 [40:15:36<30:28:05, 5.78s/it]
57%|█████▋ | 24948/43939 [40:15:42<30:19:39, 5.75s/it]
57%|█████▋ | 24949/43939 [40:15:47<30:21:37, 5.76s/it]
57%|█████▋ | 24950/43939 [40:15:53<30:14:59, 5.73s/it]
{'loss': 0.7904, 'grad_norm': 1.4605951309204102, 'learning_rate': 2.3445935695501326e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.98, 'epoch': 2.27}
57%|█████▋ | 24950/43939 [40:15:53<30:14:59, 5.73s/it]
57%|█████▋ | 24951/43939 [40:15:59<30:15:10, 5.74s/it]
57%|█████▋ | 24952/43939 [40:16:04<30:09:23, 5.72s/it]
57%|█████▋ | 24953/43939 [40:16:10<30:05:04, 5.70s/it]
57%|█████▋ | 24954/43939 [40:16:16<30:07:16, 5.71s/it]
57%|█████▋ | 24955/43939 [40:16:22<30:15:58, 5.74s/it]
57%|█████▋ | 24956/43939 [40:16:27<30:11:08, 5.72s/it]
57%|█████▋ | 24957/43939 [40:16:33<30:08:15, 5.72s/it]
57%|█████▋ | 24958/43939 [40:16:39<30:05:34, 5.71s/it]
57%|█████▋ | 24959/43939 [40:16:44<30:03:42, 5.70s/it]
57%|█████▋ | 24960/43939 [40:16:50<30:02:41, 5.70s/it]
57%|█████▋ | 24961/43939 [40:16:56<30:03:14, 5.70s/it]
57%|█████▋ | 24962/43939 [40:17:01<30:04:08, 5.70s/it]
57%|█████▋ | 24963/43939 [40:17:07<30:05:49, 5.71s/it]
57%|█████▋ | 24964/43939 [40:17:13<30:09:08, 5.72s/it]
57%|█████▋ | 24965/43939 [40:17:19<30:04:14, 5.71s/it]
57%|█████▋ | 24966/43939 [40:17:24<30:05:15, 5.71s/it]
57%|█████▋ | 24967/43939 [40:17:30<30:05:41, 5.71s/it]
57%|█████▋ | 24968/43939 [40:17:36<30:05:41, 5.71s/it]
57%|█████▋ | 24969/43939 [40:17:42<30:27:21, 5.78s/it]
57%|█████▋ | 24970/43939 [40:17:47<30:19:47, 5.76s/it]
57%|█████▋ | 24971/43939 [40:17:53<30:14:26, 5.74s/it]
57%|█████▋ | 24972/43939 [40:17:59<30:08:51, 5.72s/it]
57%|█████▋ | 24973/43939 [40:18:05<30:11:43, 5.73s/it]
57%|█████▋ | 24974/43939 [40:18:10<30:09:43, 5.73s/it]
57%|█████▋ | 24975/43939 [40:18:16<30:07:42, 5.72s/it]
57%|█████▋ | 24976/43939 [40:18:22<30:04:28, 5.71s/it]
57%|█████▋ | 24977/43939 [40:18:27<30:03:31, 5.71s/it]
57%|█████▋ | 24978/43939 [40:18:33<30:00:49, 5.70s/it]
57%|█████▋ | 24979/43939 [40:18:39<29:59:29, 5.69s/it]
57%|█████▋ | 24980/43939 [40:18:44<29:59:17, 5.69s/it]
57%|█████▋ | 24981/43939 [40:18:50<30:02:04, 5.70s/it]
57%|█████▋ | 24982/43939 [40:18:56<30:01:26, 5.70s/it]
57%|█████▋ | 24983/43939 [40:19:02<30:04:05, 5.71s/it]
57%|█████▋ | 24984/43939 [40:19:07<30:07:33, 5.72s/it]
57%|█████▋ | 24985/43939 [40:19:13<30:06:59, 5.72s/it]
57%|█████▋ | 24986/43939 [40:19:19<30:08:34, 5.73s/it]
57%|█████▋ | 24987/43939 [40:19:24<30:04:51, 5.71s/it]
57%|█████▋ | 24988/43939 [40:19:30<30:00:33, 5.70s/it]
57%|█████▋ | 24989/43939 [40:19:36<30:02:06, 5.71s/it]
57%|█████▋ | 24990/43939 [40:19:42<30:00:51, 5.70s/it]
57%|█████▋ | 24991/43939 [40:19:47<30:22:15, 5.77s/it]
57%|█████▋ | 24992/43939 [40:19:53<30:17:41, 5.76s/it]
57%|█████▋ | 24993/43939 [40:19:59<30:10:54, 5.73s/it]
57%|█████▋ | 24994/43939 [40:20:05<30:13:31, 5.74s/it]
57%|█████▋ | 24995/43939 [40:20:10<30:10:52, 5.74s/it]
57%|█████▋ | 24996/43939 [40:20:16<30:04:36, 5.72s/it]
57%|█████▋ | 24997/43939 [40:20:22<29:59:18, 5.70s/it]
57%|█████▋ | 24998/43939 [40:20:27<29:59:38, 5.70s/it]
57%|█████▋ | 24999/43939 [40:20:33<30:01:14, 5.71s/it]
57%|█████▋ | 25000/43939 [40:20:39<29:56:33, 5.69s/it]
{'loss': 0.7897, 'grad_norm': 1.3324612379074097, 'learning_rate': 2.334683841532004e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.96, 'epoch': 2.28}
57%|█████▋ | 25000/43939 [40:20:39<29:56:33, 5.69s/it]
57%|█████▋ | 25001/43939 [40:20:45<30:06:17, 5.72s/it]
57%|█████▋ | 25002/43939 [40:20:50<30:03:14, 5.71s/it]
57%|█████▋ | 25003/43939 [40:20:56<29:59:32, 5.70s/it]
57%|█████▋ | 25004/43939 [40:21:02<29:57:09, 5.69s/it]
57%|█████▋ | 25005/43939 [40:21:07<30:05:16, 5.72s/it]
57%|█████▋ | 25006/43939 [40:21:13<30:00:58, 5.71s/it]
57%|█████▋ | 25007/43939 [40:21:19<30:01:28, 5.71s/it]
57%|█████▋ | 25008/43939 [40:21:24<30:00:41, 5.71s/it]
57%|█████▋ | 25009/43939 [40:21:30<29:59:34, 5.70s/it]
57%|█████▋ | 25010/43939 [40:21:36<30:01:38, 5.71s/it]
57%|█████▋ | 25011/43939 [40:21:42<30:21:34, 5.77s/it]
57%|█████▋ | 25012/43939 [40:21:47<30:11:38, 5.74s/it]
57%|█████▋ | 25013/43939 [40:21:53<30:09:05, 5.74s/it]
57%|█████▋ | 25014/43939 [40:21:59<30:07:32, 5.73s/it]
57%|█████▋ | 25015/43939 [40:22:05<30:03:38, 5.72s/it]
57%|█████▋ | 25016/43939 [40:22:10<30:03:43, 5.72s/it]
57%|█████▋ | 25017/43939 [40:22:16<30:06:32, 5.73s/it]
57%|█████▋ | 25018/43939 [40:22:22<30:02:45, 5.72s/it]
57%|█████▋ | 25019/43939 [40:22:27<30:00:58, 5.71s/it]
57%|█████▋ | 25020/43939 [40:22:33<30:04:04, 5.72s/it]
57%|█████▋ | 25021/43939 [40:22:39<30:00:48, 5.71s/it]
57%|█████▋ | 25022/43939 [40:22:45<29:59:45, 5.71s/it]
57%|█████▋ | 25023/43939 [40:22:50<30:05:35, 5.73s/it]
57%|█████▋ | 25024/43939 [40:22:56<30:08:55, 5.74s/it]
57%|█████▋ | 25025/43939 [40:23:02<30:05:26, 5.73s/it]
57%|█████▋ | 25026/43939 [40:23:08<30:06:10, 5.73s/it]
57%|█████▋ | 25027/43939 [40:23:13<30:06:48, 5.73s/it]
57%|█████▋ | 25028/43939 [40:23:19<30:03:21, 5.72s/it]
57%|█████▋ | 25029/43939 [40:23:25<30:03:06, 5.72s/it]
57%|█████▋ | 25030/43939 [40:23:30<30:03:22, 5.72s/it]
57%|█████▋ | 25031/43939 [40:23:36<29:59:30, 5.71s/it]
57%|█████▋ | 25032/43939 [40:23:42<30:24:40, 5.79s/it]
57%|█████▋ | 25033/43939 [40:23:48<30:20:17, 5.78s/it]
57%|█████▋ | 25034/43939 [40:23:54<30:16:59, 5.77s/it]
57%|█████▋ | 25035/43939 [40:23:59<30:09:13, 5.74s/it]
57%|█████▋ | 25036/43939 [40:24:05<30:06:18, 5.73s/it]
57%|█████▋ | 25037/43939 [40:24:11<30:05:28, 5.73s/it]
57%|█████▋ | 25038/43939 [40:24:16<29:58:22, 5.71s/it]
57%|█████▋ | 25039/43939 [40:24:22<29:57:31, 5.71s/it]
57%|█████▋ | 25040/43939 [40:24:28<29:57:53, 5.71s/it]
57%|█████▋ | 25041/43939 [40:24:33<29:55:25, 5.70s/it]
57%|█████▋ | 25042/43939 [40:24:39<29:59:59, 5.72s/it]
57%|█████▋ | 25043/43939 [40:24:45<29:59:17, 5.71s/it]
57%|█████▋ | 25044/43939 [40:24:51<29:53:29, 5.70s/it]
57%|█████▋ | 25045/43939 [40:24:56<29:54:43, 5.70s/it]
57%|█████▋ | 25046/43939 [40:25:02<29:58:24, 5.71s/it]
57%|█████▋ | 25047/43939 [40:25:08<29:53:53, 5.70s/it]
57%|█████▋ | 25048/43939 [40:25:13<29:54:46, 5.70s/it]
57%|█████▋ | 25049/43939 [40:25:19<29:51:48, 5.69s/it]
57%|█████▋ | 25050/43939 [40:25:25<29:59:29, 5.72s/it]
{'loss': 0.7878, 'grad_norm': 1.165163278579712, 'learning_rate': 2.324776721764763e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.98, 'epoch': 2.28}
57%|█████▋ | 25050/43939 [40:25:25<29:59:29, 5.72s/it]
57%|█████▋ | 25051/43939 [40:25:31<29:55:11, 5.70s/it]
57%|█████▋ | 25052/43939 [40:25:36<29:53:54, 5.70s/it]
57%|█████▋ | 25053/43939 [40:25:42<30:22:45, 5.79s/it]
57%|█████▋ | 25054/43939 [40:25:48<30:19:34, 5.78s/it]
57%|█████▋ | 25055/43939 [40:25:54<30:10:39, 5.75s/it]
57%|█████▋ | 25056/43939 [40:25:59<30:04:13, 5.73s/it]
57%|█████▋ | 25057/43939 [40:26:05<30:02:15, 5.73s/it]
57%|█████▋ | 25058/43939 [40:26:11<30:00:14, 5.72s/it]
57%|█████▋ | 25059/43939 [40:26:17<29:59:55, 5.72s/it]
57%|█████▋ | 25060/43939 [40:26:22<29:56:08, 5.71s/it]
57%|█████▋ | 25061/43939 [40:26:28<30:00:00, 5.72s/it]
57%|█████▋ | 25062/43939 [40:26:34<29:56:46, 5.71s/it]
57%|█████▋ | 25063/43939 [40:26:39<29:55:11, 5.71s/it]
57%|█████▋ | 25064/43939 [40:26:45<29:54:22, 5.70s/it]
57%|█████▋ | 25065/43939 [40:26:51<29:50:30, 5.69s/it]
57%|█████▋ | 25066/43939 [40:26:56<29:50:07, 5.69s/it]
57%|█████▋ | 25067/43939 [40:27:02<29:54:04, 5.70s/it]
57%|█████▋ | 25068/43939 [40:27:08<29:59:05, 5.72s/it]
57%|█████▋ | 25069/43939 [40:27:14<29:55:58, 5.71s/it]
57%|█████▋ | 25070/43939 [40:27:19<29:53:45, 5.70s/it]
57%|█████▋ | 25071/43939 [40:27:25<29:55:49, 5.71s/it]
57%|█████▋ | 25072/43939 [40:27:31<30:01:44, 5.73s/it]
57%|█████▋ | 25073/43939 [40:27:36<29:57:04, 5.72s/it]
57%|█████▋ | 25074/43939 [40:27:42<30:17:03, 5.78s/it]
57%|█████▋ | 25075/43939 [40:27:48<30:07:51, 5.75s/it]
57%|█████▋ | 25076/43939 [40:27:54<30:00:39, 5.73s/it]
57%|█████▋ | 25077/43939 [40:27:59<29:59:42, 5.72s/it]
57%|█████▋ | 25078/43939 [40:28:05<29:54:30, 5.71s/it]
57%|█████▋ | 25079/43939 [40:28:11<29:56:12, 5.71s/it]
57%|█████▋ | 25080/43939 [40:28:17<29:55:16, 5.71s/it]
57%|█████▋ | 25081/43939 [40:28:22<29:53:24, 5.71s/it]
57%|█████▋ | 25082/43939 [40:28:28<29:55:40, 5.71s/it]
57%|█████▋ | 25083/43939 [40:28:34<29:56:10, 5.72s/it]
57%|█████▋ | 25084/43939 [40:28:39<29:55:55, 5.71s/it]
57%|█████▋ | 25085/43939 [40:28:45<29:56:14, 5.72s/it]
57%|█████▋ | 25086/43939 [40:28:51<29:53:51, 5.71s/it]
57%|█████▋ | 25087/43939 [40:28:56<29:50:42, 5.70s/it]
57%|█████▋ | 25088/43939 [40:29:02<29:53:00, 5.71s/it]
57%|█████▋ | 25089/43939 [40:29:08<29:52:51, 5.71s/it]
57%|█████▋ | 25090/43939 [40:29:14<29:55:45, 5.72s/it]
57%|█████▋ | 25091/43939 [40:29:19<29:53:33, 5.71s/it]
57%|█████▋ | 25092/43939 [40:29:25<29:51:02, 5.70s/it]
57%|█████▋ | 25093/43939 [40:29:31<29:48:46, 5.69s/it]
57%|█████▋ | 25094/43939 [40:29:36<29:51:11, 5.70s/it]
57%|█████▋ | 25095/43939 [40:29:42<30:09:40, 5.76s/it]
57%|█████▋ | 25096/43939 [40:29:48<30:03:32, 5.74s/it]
57%|█████▋ | 25097/43939 [40:29:54<29:59:15, 5.73s/it]
57%|█████▋ | 25098/43939 [40:29:59<29:55:55, 5.72s/it]
57%|█████▋ | 25099/43939 [40:30:05<29:56:59, 5.72s/it]
57%|█████▋ | 25100/43939 [40:30:11<30:00:05, 5.73s/it]
{'loss': 0.7987, 'grad_norm': 1.2003819942474365, 'learning_rate': 2.3148723665565053e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 521.37, 'epoch': 2.28}
57%|█████▋ | 25100/43939 [40:30:11<30:00:05, 5.73s/it]
57%|█████▋ | 25101/43939 [40:30:17<30:02:02, 5.74s/it]
57%|█████▋ | 25102/43939 [40:30:22<29:57:52, 5.73s/it]
57%|█████▋ | 25103/43939 [40:30:28<29:53:17, 5.71s/it]
57%|█████▋ | 25104/43939 [40:30:34<29:53:33, 5.71s/it]
57%|█████▋ | 25105/43939 [40:30:39<29:51:17, 5.71s/it]
57%|█████▋ | 25106/43939 [40:30:45<29:52:36, 5.71s/it]
57%|█████▋ | 25107/43939 [40:30:51<29:54:08, 5.72s/it]
57%|█████▋ | 25108/43939 [40:30:57<29:48:48, 5.70s/it]
57%|█████▋ | 25109/43939 [40:31:02<29:45:13, 5.69s/it]
57%|█████▋ | 25110/43939 [40:31:08<29:48:56, 5.70s/it]
57%|█████▋ | 25111/43939 [40:31:14<29:56:16, 5.72s/it]
57%|█████▋ | 25112/43939 [40:31:19<29:58:15, 5.73s/it]
57%|█████▋ | 25113/43939 [40:31:25<29:57:27, 5.73s/it]
57%|█████▋ | 25114/43939 [40:31:31<29:51:05, 5.71s/it]
57%|█████▋ | 25115/43939 [40:31:37<29:50:39, 5.71s/it]
57%|█████▋ | 25116/43939 [40:31:42<30:08:07, 5.76s/it]
57%|█████▋ | 25117/43939 [40:31:48<30:00:20, 5.74s/it]
57%|█████▋ | 25118/43939 [40:31:54<29:56:41, 5.73s/it]
57%|█████▋ | 25119/43939 [40:32:00<29:52:25, 5.71s/it]
57%|█████▋ | 25120/43939 [40:32:05<29:54:20, 5.72s/it]
57%|█████▋ | 25121/43939 [40:32:11<29:51:41, 5.71s/it]
57%|█████▋ | 25122/43939 [40:32:17<29:53:27, 5.72s/it]
57%|█████▋ | 25123/43939 [40:32:22<29:52:12, 5.71s/it]
57%|█████▋ | 25124/43939 [40:32:28<29:49:56, 5.71s/it]
57%|█████▋ | 25125/43939 [40:32:34<29:44:53, 5.69s/it]
57%|█████▋ | 25126/43939 [40:32:39<29:44:45, 5.69s/it]
57%|█████▋ | 25127/43939 [40:32:45<29:44:00, 5.69s/it]
57%|█████▋ | 25128/43939 [40:32:51<29:43:39, 5.69s/it]
57%|█████▋ | 25129/43939 [40:32:57<29:46:04, 5.70s/it]
57%|█████▋ | 25130/43939 [40:33:02<29:44:24, 5.69s/it]
57%|█████▋ | 25131/43939 [40:33:08<29:49:32, 5.71s/it]
57%|█████▋ | 25132/43939 [40:33:14<29:56:04, 5.73s/it]
57%|█████▋ | 25133/43939 [40:33:19<29:57:52, 5.74s/it]
57%|█████▋ | 25134/43939 [40:33:25<29:54:35, 5.73s/it]
57%|█████▋ | 25135/43939 [40:33:31<29:52:13, 5.72s/it]
57%|█████▋ | 25136/43939 [40:33:37<29:47:10, 5.70s/it]
57%|█████▋ | 25137/43939 [40:33:42<29:47:22, 5.70s/it]
57%|█████▋ | 25138/43939 [40:33:48<30:07:30, 5.77s/it]
57%|█████▋ | 25139/43939 [40:33:54<30:00:19, 5.75s/it]
57%|█████▋ | 25140/43939 [40:34:00<29:56:35, 5.73s/it]
57%|█████▋ | 25141/43939 [40:34:05<29:54:02, 5.73s/it]
57%|█████▋ | 25142/43939 [40:34:11<29:51:01, 5.72s/it]
57%|█████▋ | 25143/43939 [40:34:17<29:50:07, 5.71s/it]
57%|█████▋ | 25144/43939 [40:34:22<29:47:43, 5.71s/it]
57%|█████▋ | 25145/43939 [40:34:28<29:42:17, 5.69s/it]
57%|█████▋ | 25146/43939 [40:34:34<29:44:46, 5.70s/it]
57%|█████▋ | 25147/43939 [40:34:39<29:44:43, 5.70s/it]
57%|█████▋ | 25148/43939 [40:34:45<29:49:06, 5.71s/it]
57%|█████▋ | 25149/43939 [40:34:51<29:48:27, 5.71s/it]
57%|█████▋ | 25150/43939 [40:34:57<29:47:42, 5.71s/it]
{'loss': 0.7858, 'grad_norm': 1.0604358911514282, 'learning_rate': 2.3049709321717085e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 512.62, 'epoch': 2.29}
57%|█████▋ | 25150/43939 [40:34:57<29:47:42, 5.71s/it]
57%|█████▋ | 25151/43939 [40:35:02<29:49:38, 5.72s/it]
57%|█████▋ | 25152/43939 [40:35:08<29:48:34, 5.71s/it]
57%|█████▋ | 25153/43939 [40:35:14<29:47:22, 5.71s/it]
57%|█████▋ | 25154/43939 [40:35:20<29:56:14, 5.74s/it]
57%|█████▋ | 25155/43939 [40:35:25<29:52:22, 5.73s/it]
57%|█████▋ | 25156/43939 [40:35:31<29:50:33, 5.72s/it]
57%|█████▋ | 25157/43939 [40:35:37<29:49:49, 5.72s/it]
57%|█████▋ | 25158/43939 [40:35:42<29:44:49, 5.70s/it]
57%|█████▋ | 25159/43939 [40:35:48<30:14:55, 5.80s/it]
57%|█████▋ | 25160/43939 [40:35:54<30:05:52, 5.77s/it]
57%|█████▋ | 25161/43939 [40:36:00<30:00:19, 5.75s/it]
57%|█████▋ | 25162/43939 [40:36:05<29:51:41, 5.73s/it]
57%|█████▋ | 25163/43939 [40:36:11<29:49:03, 5.72s/it]
57%|█████▋ | 25164/43939 [40:36:17<29:50:24, 5.72s/it]
57%|█████▋ | 25165/43939 [40:36:23<29:45:09, 5.71s/it]
57%|█████▋ | 25166/43939 [40:36:28<29:42:58, 5.70s/it]
57%|█████▋ | 25167/43939 [40:36:34<29:41:07, 5.69s/it]
57%|█████▋ | 25168/43939 [40:36:40<29:39:15, 5.69s/it]
57%|█████▋ | 25169/43939 [40:36:45<29:42:52, 5.70s/it]
57%|█████▋ | 25170/43939 [40:36:51<29:43:02, 5.70s/it]
57%|█████▋ | 25171/43939 [40:36:57<29:44:34, 5.71s/it]
57%|█████▋ | 25172/43939 [40:37:02<29:46:00, 5.71s/it]
57%|█████▋ | 25173/43939 [40:37:08<29:47:06, 5.71s/it]
57%|█████▋ | 25174/43939 [40:37:14<29:42:19, 5.70s/it]
57%|█████▋ | 25175/43939 [40:37:20<29:44:10, 5.71s/it]
57%|█████▋ | 25176/43939 [40:37:25<29:44:26, 5.71s/it]
57%|█████▋ | 25177/43939 [40:37:31<29:43:34, 5.70s/it]
57%|█████▋ | 25178/43939 [40:37:37<29:40:37, 5.69s/it]
57%|█████▋ | 25179/43939 [40:37:42<29:36:05, 5.68s/it]
57%|█████▋ | 25180/43939 [40:37:48<30:02:23, 5.76s/it]
57%|█████▋ | 25181/43939 [40:37:54<29:53:58, 5.74s/it]
57%|█████▋ | 25182/43939 [40:38:00<29:49:42, 5.72s/it]
57%|█████▋ | 25183/43939 [40:38:05<29:48:29, 5.72s/it]
57%|█████▋ | 25184/43939 [40:38:11<29:49:55, 5.73s/it]
57%|█████▋ | 25185/43939 [40:38:17<29:52:12, 5.73s/it]
57%|█████▋ | 25186/43939 [40:38:22<29:49:26, 5.73s/it]
57%|█████▋ | 25187/43939 [40:38:28<29:49:37, 5.73s/it]
57%|█████▋ | 25188/43939 [40:38:34<29:44:38, 5.71s/it]
57%|█████▋ | 25189/43939 [40:38:40<29:46:37, 5.72s/it]
57%|█████▋ | 25190/43939 [40:38:45<29:43:51, 5.71s/it]
57%|█████▋ | 25191/43939 [40:38:51<29:43:29, 5.71s/it]
57%|█████▋ | 25192/43939 [40:38:57<29:43:42, 5.71s/it]
57%|█████▋ | 25193/43939 [40:39:02<29:39:10, 5.69s/it]
57%|█████▋ | 25194/43939 [40:39:08<29:39:18, 5.70s/it]
57%|█████▋ | 25195/43939 [40:39:14<29:43:58, 5.71s/it]
57%|█████▋ | 25196/43939 [40:39:20<29:46:41, 5.72s/it]
57%|█████▋ | 25197/43939 [40:39:25<29:43:02, 5.71s/it]
57%|█████▋ | 25198/43939 [40:39:31<29:43:22, 5.71s/it]
57%|█████▋ | 25199/43939 [40:39:37<29:40:57, 5.70s/it]
57%|█████▋ | 25200/43939 [40:39:42<29:45:29, 5.72s/it]
{'loss': 0.7859, 'grad_norm': 1.1268080472946167, 'learning_rate': 2.2950725748287695e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 509.62, 'epoch': 2.29}
57%|█████▋ | 25200/43939 [40:39:42<29:45:29, 5.72s/it]
57%|█████▋ | 25201/43939 [40:39:48<30:07:26, 5.79s/it]
57%|█████▋ | 25202/43939 [40:39:54<30:02:47, 5.77s/it]
57%|█████▋ | 25203/43939 [40:40:00<29:57:17, 5.76s/it]
57%|█████▋ | 25204/43939 [40:40:06<29:55:50, 5.75s/it]
57%|█████▋ | 25205/43939 [40:40:11<29:51:11, 5.74s/it]
57%|█████▋ | 25206/43939 [40:40:17<29:49:56, 5.73s/it]
57%|█████▋ | 25207/43939 [40:40:23<29:48:07, 5.73s/it]
57%|█████▋ | 25208/43939 [40:40:28<29:44:03, 5.71s/it]
57%|█████▋ | 25209/43939 [40:40:34<29:42:08, 5.71s/it]
57%|█████▋ | 25210/43939 [40:40:40<29:40:52, 5.71s/it]
57%|█████▋ | 25211/43939 [40:40:45<29:39:39, 5.70s/it]
57%|█████▋ | 25212/43939 [40:40:51<29:39:05, 5.70s/it]
57%|█████▋ | 25213/43939 [40:40:57<29:38:31, 5.70s/it]
57%|█████▋ | 25214/43939 [40:41:03<29:38:13, 5.70s/it]
57%|█████▋ | 25215/43939 [40:41:08<29:39:22, 5.70s/it]
57%|█████▋ | 25216/43939 [40:41:14<29:36:22, 5.69s/it]
57%|█████▋ | 25217/43939 [40:41:20<29:33:05, 5.68s/it]
57%|█████▋ | 25218/43939 [40:41:25<29:32:47, 5.68s/it]
57%|█████▋ | 25219/43939 [40:41:31<29:34:05, 5.69s/it]
57%|█████▋ | 25220/43939 [40:41:37<29:37:46, 5.70s/it]
57%|█████▋ | 25221/43939 [40:41:42<29:42:39, 5.71s/it]
57%|█████▋ | 25222/43939 [40:41:48<30:06:55, 5.79s/it]
57%|█████▋ | 25223/43939 [40:41:54<29:58:23, 5.77s/it]
57%|█████▋ | 25224/43939 [40:42:00<29:53:44, 5.75s/it]
57%|█████▋ | 25225/43939 [40:42:06<29:46:26, 5.73s/it]
57%|█████▋ | 25226/43939 [40:42:11<29:45:31, 5.72s/it]
57%|█████▋ | 25227/43939 [40:42:17<29:41:24, 5.71s/it]
57%|█████▋ | 25228/43939 [40:42:23<29:38:09, 5.70s/it]
57%|█████▋ | 25229/43939 [40:42:28<29:33:38, 5.69s/it]
57%|█████▋ | 25230/43939 [40:42:34<29:36:12, 5.70s/it]
57%|█████▋ | 25231/43939 [40:42:40<29:38:49, 5.71s/it]
57%|█████▋ | 25232/43939 [40:42:45<29:41:38, 5.71s/it]
57%|█████▋ | 25233/43939 [40:42:51<29:45:44, 5.73s/it]
57%|█████▋ | 25234/43939 [40:42:57<29:42:46, 5.72s/it]
57%|█████▋ | 25235/43939 [40:43:03<29:41:06, 5.71s/it]
57%|█████▋ | 25236/43939 [40:43:08<29:41:35, 5.72s/it]
57%|█████▋ | 25237/43939 [40:43:14<29:44:01, 5.72s/it]
57%|█████▋ | 25238/43939 [40:43:20<29:41:36, 5.72s/it]
57%|█████▋ | 25239/43939 [40:43:25<29:36:50, 5.70s/it]
57%|█████▋ | 25240/43939 [40:43:31<29:37:07, 5.70s/it]
57%|█████▋ | 25241/43939 [40:43:37<29:36:10, 5.70s/it]
57%|█████▋ | 25242/43939 [40:43:43<29:39:31, 5.71s/it]
57%|█████▋ | 25243/43939 [40:43:48<30:02:17, 5.78s/it]
57%|█████▋ | 25244/43939 [40:43:54<29:54:59, 5.76s/it]
57%|█████▋ | 25245/43939 [40:44:00<29:55:46, 5.76s/it]
57%|█████▋ | 25246/43939 [40:44:06<29:52:08, 5.75s/it]
57%|█████▋ | 25247/43939 [40:44:11<29:47:46, 5.74s/it]
57%|█████▋ | 25248/43939 [40:44:17<29:41:15, 5.72s/it]
57%|█████▋ | 25249/43939 [40:44:23<29:42:45, 5.72s/it]
57%|█████▋ | 25250/43939 [40:44:29<29:43:45, 5.73s/it]
{'loss': 0.7769, 'grad_norm': 1.4554740190505981, 'learning_rate': 2.285177450697535e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.16, 'epoch': 2.3}
57%|█████▋ | 25250/43939 [40:44:29<29:43:45, 5.73s/it]
57%|█████▋ | 25251/43939 [40:44:34<29:39:26, 5.71s/it]
57%|█████▋ | 25252/43939 [40:44:40<29:41:09, 5.72s/it]
57%|█████▋ | 25253/43939 [40:44:46<29:46:53, 5.74s/it]
57%|█████▋ | 25254/43939 [40:44:51<29:45:50, 5.73s/it]
57%|█████▋ | 25255/43939 [40:44:57<29:42:11, 5.72s/it]
57%|█████▋ | 25256/43939 [40:45:03<29:43:00, 5.73s/it]
57%|█████▋ | 25257/43939 [40:45:09<29:43:44, 5.73s/it]
57%|█████▋ | 25258/43939 [40:45:14<29:41:29, 5.72s/it]
57%|█████▋ | 25259/43939 [40:45:20<29:41:19, 5.72s/it]
57%|█████▋ | 25260/43939 [40:45:26<29:42:30, 5.73s/it]
57%|█████▋ | 25261/43939 [40:45:32<29:41:48, 5.72s/it]
57%|█████▋ | 25262/43939 [40:45:37<29:39:56, 5.72s/it]
57%|█████▋ | 25263/43939 [40:45:43<29:35:04, 5.70s/it]
57%|█████▋ | 25264/43939 [40:45:49<29:40:50, 5.72s/it]
58%|█████▊ | 25265/43939 [40:45:55<30:04:36, 5.80s/it]
58%|█████▊ | 25266/43939 [40:46:00<29:55:14, 5.77s/it]
58%|█████▊ | 25267/43939 [40:46:06<29:49:42, 5.75s/it]
58%|█████▊ | 25268/43939 [40:46:12<29:39:54, 5.72s/it]
58%|█████▊ | 25269/43939 [40:46:17<29:37:19, 5.71s/it]
58%|█████▊ | 25270/43939 [40:46:23<29:36:43, 5.71s/it]
58%|█████▊ | 25271/43939 [40:46:29<29:34:36, 5.70s/it]
58%|█████▊ | 25272/43939 [40:46:34<29:32:51, 5.70s/it]
58%|█████▊ | 25273/43939 [40:46:40<29:32:48, 5.70s/it]
58%|█████▊ | 25274/43939 [40:46:46<29:30:39, 5.69s/it]
58%|█████▊ | 25275/43939 [40:46:52<29:31:13, 5.69s/it]
58%|█████▊ | 25276/43939 [40:46:57<29:29:42, 5.69s/it]
58%|█████▊ | 25277/43939 [40:47:03<29:30:21, 5.69s/it]
58%|█████▊ | 25278/43939 [40:47:09<29:28:13, 5.69s/it]
58%|█████▊ | 25279/43939 [40:47:14<29:29:20, 5.69s/it]
58%|█████▊ | 25280/43939 [40:47:20<29:29:30, 5.69s/it]
58%|█████▊ | 25281/43939 [40:47:26<29:36:10, 5.71s/it]
58%|█████▊ | 25282/43939 [40:47:31<29:31:23, 5.70s/it]
58%|█████▊ | 25283/43939 [40:47:37<29:30:13, 5.69s/it]
58%|█████▊ | 25284/43939 [40:47:43<29:30:45, 5.70s/it]
58%|█████▊ | 25285/43939 [40:47:49<29:32:59, 5.70s/it]
58%|█████▊ | 25286/43939 [40:47:54<29:34:10, 5.71s/it]
58%|█████▊ | 25287/43939 [40:48:00<29:55:33, 5.78s/it]
58%|█████▊ | 25288/43939 [40:48:06<29:48:18, 5.75s/it]
58%|█████▊ | 25289/43939 [40:48:12<29:38:54, 5.72s/it]
58%|█████▊ | 25290/43939 [40:48:17<29:33:58, 5.71s/it]
58%|█████▊ | 25291/43939 [40:48:23<29:36:25, 5.72s/it]
58%|█████▊ | 25292/43939 [40:48:29<29:37:28, 5.72s/it]
58%|█████▊ | 25293/43939 [40:48:34<29:33:03, 5.71s/it]
58%|█████▊ | 25294/43939 [40:48:40<29:30:18, 5.70s/it]
58%|█████▊ | 25295/43939 [40:48:46<29:32:24, 5.70s/it]
58%|█████▊ | 25296/43939 [40:48:51<29:33:56, 5.71s/it]
58%|█████▊ | 25297/43939 [40:48:57<29:30:58, 5.70s/it]
58%|█████▊ | 25298/43939 [40:49:03<29:27:15, 5.69s/it]
58%|█████▊ | 25299/43939 [40:49:08<29:26:07, 5.68s/it]
58%|█████▊ | 25300/43939 [40:49:14<29:25:02, 5.68s/it]
{'loss': 0.7842, 'grad_norm': 1.524244785308838, 'learning_rate': 2.2752857158968422e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.96, 'epoch': 2.3}
58%|█████▊ | 25300/43939 [40:49:14<29:25:02, 5.68s/it]
58%|█████▊ | 25301/43939 [40:49:20<29:23:43, 5.68s/it]
58%|█████▊ | 25302/43939 [40:49:25<29:25:49, 5.68s/it]
58%|█████▊ | 25303/43939 [40:49:31<29:25:54, 5.69s/it]
58%|█████▊ | 25304/43939 [40:49:37<29:28:41, 5.69s/it]
58%|█████▊ | 25305/43939 [40:49:43<29:29:28, 5.70s/it]
58%|█████▊ | 25306/43939 [40:49:48<29:31:19, 5.70s/it]
58%|█████▊ | 25307/43939 [40:49:54<29:31:37, 5.71s/it]
58%|█████▊ | 25308/43939 [40:50:00<29:56:51, 5.79s/it]
58%|█████▊ | 25309/43939 [40:50:06<29:52:03, 5.77s/it]
58%|█████▊ | 25310/43939 [40:50:11<29:43:15, 5.74s/it]
58%|█████▊ | 25311/43939 [40:50:17<29:40:07, 5.73s/it]
58%|█████▊ | 25312/43939 [40:50:23<29:36:23, 5.72s/it]
58%|█████▊ | 25313/43939 [40:50:29<29:34:01, 5.71s/it]
58%|█████▊ | 25314/43939 [40:50:34<29:30:47, 5.70s/it]
58%|█████▊ | 25315/43939 [40:50:40<29:30:17, 5.70s/it]
58%|█████▊ | 25316/43939 [40:50:46<29:30:57, 5.71s/it]
58%|█████▊ | 25317/43939 [40:50:51<29:33:53, 5.72s/it]
58%|█████▊ | 25318/43939 [40:50:57<29:37:05, 5.73s/it]
58%|█████▊ | 25319/43939 [40:51:03<29:36:29, 5.72s/it]
58%|█████▊ | 25320/43939 [40:51:09<29:36:33, 5.72s/it]
58%|█████▊ | 25321/43939 [40:51:14<29:37:59, 5.73s/it]
58%|█████▊ | 25322/43939 [40:51:20<29:37:46, 5.73s/it]
58%|█████▊ | 25323/43939 [40:51:26<29:36:19, 5.73s/it]
58%|█████▊ | 25324/43939 [40:51:31<29:30:55, 5.71s/it]
58%|█████▊ | 25325/43939 [40:51:37<29:26:28, 5.69s/it]
58%|█████▊ | 25326/43939 [40:51:43<29:26:44, 5.70s/it]
58%|█████▊ | 25327/43939 [40:51:48<29:23:32, 5.69s/it]
58%|█████▊ | 25328/43939 [40:51:54<29:24:35, 5.69s/it]
58%|█████▊ | 25329/43939 [40:52:00<29:42:41, 5.75s/it]
58%|█████▊ | 25330/43939 [40:52:06<29:39:46, 5.74s/it]
58%|█████▊ | 25331/43939 [40:52:11<29:30:58, 5.71s/it]
58%|█████▊ | 25332/43939 [40:52:17<29:26:40, 5.70s/it]
58%|█████▊ | 25333/43939 [40:52:23<29:22:03, 5.68s/it]
58%|█████▊ | 25334/43939 [40:52:28<29:25:39, 5.69s/it]
58%|█████▊ | 25335/43939 [40:52:34<29:25:23, 5.69s/it]
58%|█████▊ | 25336/43939 [40:52:40<29:30:12, 5.71s/it]
58%|█████▊ | 25337/43939 [40:52:46<29:31:54, 5.72s/it]
58%|█████▊ | 25338/43939 [40:52:51<29:35:22, 5.73s/it]
58%|█████▊ | 25339/43939 [40:52:57<29:30:38, 5.71s/it]
58%|█████▊ | 25340/43939 [40:53:03<29:30:17, 5.71s/it]
58%|█████▊ | 25341/43939 [40:53:08<29:29:04, 5.71s/it]
58%|█████▊ | 25342/43939 [40:53:14<29:26:33, 5.70s/it]
58%|█████▊ | 25343/43939 [40:53:20<29:27:37, 5.70s/it]
58%|█████▊ | 25344/43939 [40:53:26<29:29:02, 5.71s/it]
58%|█████▊ | 25345/43939 [40:53:31<29:32:16, 5.72s/it]
58%|█████▊ | 25346/43939 [40:53:37<29:34:53, 5.73s/it]
58%|█████▊ | 25347/43939 [40:53:43<29:33:47, 5.72s/it]
58%|█████▊ | 25348/43939 [40:53:48<29:30:01, 5.71s/it]
58%|█████▊ | 25349/43939 [40:53:54<29:25:00, 5.70s/it]
58%|█████▊ | 25350/43939 [40:54:00<29:22:26, 5.69s/it]
{'loss': 0.7811, 'grad_norm': 1.4134726524353027, 'learning_rate': 2.265397526492052e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.78, 'epoch': 2.31}
58%|█████▊ | 25350/43939 [40:54:00<29:22:26, 5.69s/it]
58%|█████▊ | 25351/43939 [40:54:06<29:46:18, 5.77s/it]
58%|█████▊ | 25352/43939 [40:54:11<29:39:42, 5.74s/it]
58%|█████▊ | 25353/43939 [40:54:17<29:34:41, 5.73s/it]
58%|█████▊ | 25354/43939 [40:54:23<29:32:14, 5.72s/it]
58%|█████▊ | 25355/43939 [40:54:28<29:29:52, 5.71s/it]
58%|█████▊ | 25356/43939 [40:54:34<29:28:42, 5.71s/it]
58%|█████▊ | 25357/43939 [40:54:40<29:30:04, 5.72s/it]
58%|█████▊ | 25358/43939 [40:54:46<29:27:51, 5.71s/it]
58%|█████▊ | 25359/43939 [40:54:51<29:29:23, 5.71s/it]
58%|█████▊ | 25360/43939 [40:54:57<29:25:51, 5.70s/it]
58%|█████▊ | 25361/43939 [40:55:03<29:29:03, 5.71s/it]
58%|█████▊ | 25362/43939 [40:55:08<29:26:11, 5.70s/it]
58%|█████▊ | 25363/43939 [40:55:14<29:24:48, 5.70s/it]
58%|█████▊ | 25364/43939 [40:55:20<29:23:59, 5.70s/it]
58%|█████▊ | 25365/43939 [40:55:25<29:22:07, 5.69s/it]
58%|█████▊ | 25366/43939 [40:55:31<29:27:50, 5.71s/it]
58%|█████▊ | 25367/43939 [40:55:37<29:28:48, 5.71s/it]
58%|█████▊ | 25368/43939 [40:55:43<29:26:06, 5.71s/it]
58%|█████▊ | 25369/43939 [40:55:48<29:25:10, 5.70s/it]
58%|█████▊ | 25370/43939 [40:55:54<29:24:05, 5.70s/it]
58%|█████▊ | 25371/43939 [40:56:00<29:19:39, 5.69s/it]
58%|█████▊ | 25372/43939 [40:56:05<29:19:37, 5.69s/it]
58%|█████▊ | 25373/43939 [40:56:11<29:40:39, 5.75s/it]
58%|█████▊ | 25374/43939 [40:56:17<29:38:06, 5.75s/it]
58%|█████▊ | 25375/43939 [40:56:23<29:32:50, 5.73s/it]
58%|█████▊ | 25376/43939 [40:56:28<29:27:57, 5.71s/it]
58%|█████▊ | 25377/43939 [40:56:34<29:31:10, 5.73s/it]
58%|█████▊ | 25378/43939 [40:56:40<29:28:40, 5.72s/it]
58%|█████▊ | 25379/43939 [40:56:46<29:26:36, 5.71s/it]
58%|█████▊ | 25380/43939 [40:56:51<29:21:19, 5.69s/it]
58%|█████▊ | 25381/43939 [40:56:57<29:21:30, 5.70s/it]
58%|█████▊ | 25382/43939 [40:57:03<29:22:13, 5.70s/it]
58%|█████▊ | 25383/43939 [40:57:08<29:21:57, 5.70s/it]
58%|█████▊ | 25384/43939 [40:57:14<29:20:45, 5.69s/it]
58%|█████▊ | 25385/43939 [40:57:20<29:22:56, 5.70s/it]
58%|█████▊ | 25386/43939 [40:57:25<29:22:12, 5.70s/it]
58%|█████▊ | 25387/43939 [40:57:31<29:21:00, 5.70s/it]
58%|█████▊ | 25388/43939 [40:57:37<29:24:37, 5.71s/it]
58%|█████▊ | 25389/43939 [40:57:43<29:25:00, 5.71s/it]
58%|█████▊ | 25390/43939 [40:57:48<29:27:25, 5.72s/it]
58%|█████▊ | 25391/43939 [40:57:54<29:25:55, 5.71s/it]
58%|█████▊ | 25392/43939 [40:58:00<29:28:35, 5.72s/it]
58%|█████▊ | 25393/43939 [40:58:05<29:24:11, 5.71s/it]
58%|█████▊ | 25394/43939 [40:58:11<29:43:08, 5.77s/it]
58%|█████▊ | 25395/43939 [40:58:17<29:41:59, 5.77s/it]
58%|█████▊ | 25396/43939 [40:58:23<29:34:10, 5.74s/it]
58%|█████▊ | 25397/43939 [40:58:28<29:30:59, 5.73s/it]
58%|█████▊ | 25398/43939 [40:58:34<29:29:18, 5.73s/it]
58%|█████▊ | 25399/43939 [40:58:40<29:27:56, 5.72s/it]
58%|█████▊ | 25400/43939 [40:58:46<29:26:55, 5.72s/it]
{'loss': 0.7762, 'grad_norm': 1.3532620668411255, 'learning_rate': 2.2555130384925902e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.13, 'epoch': 2.31}
58%|█████▊ | 25400/43939 [40:58:46<29:26:55, 5.72s/it]
58%|█████▊ | 25401/43939 [40:58:51<29:27:52, 5.72s/it]
58%|█████▊ | 25402/43939 [40:58:57<29:26:07, 5.72s/it]
58%|█████▊ | 25403/43939 [40:59:03<29:26:58, 5.72s/it]
58%|█████▊ | 25404/43939 [40:59:08<29:25:44, 5.72s/it]
58%|█████▊ | 25405/43939 [40:59:14<29:22:36, 5.71s/it]
58%|█████▊ | 25406/43939 [40:59:20<29:20:20, 5.70s/it]
58%|█████▊ | 25407/43939 [40:59:26<29:21:12, 5.70s/it]
58%|█████▊ | 25408/43939 [40:59:31<29:18:37, 5.69s/it]
58%|█████▊ | 25409/43939 [40:59:37<29:21:06, 5.70s/it]
58%|█████▊ | 25410/43939 [40:59:43<29:20:13, 5.70s/it]
58%|█████▊ | 25411/43939 [40:59:48<29:18:29, 5.69s/it]
58%|█████▊ | 25412/43939 [40:59:54<29:17:27, 5.69s/it]
58%|█████▊ | 25413/43939 [41:00:00<29:18:24, 5.69s/it]
58%|█████▊ | 25414/43939 [41:00:05<29:23:48, 5.71s/it]
58%|█████▊ | 25415/43939 [41:00:11<29:41:11, 5.77s/it]
58%|█████▊ | 25416/43939 [41:00:17<29:40:26, 5.77s/it]
58%|█████▊ | 25417/43939 [41:00:23<29:34:25, 5.75s/it]
58%|█████▊ | 25418/43939 [41:00:28<29:28:30, 5.73s/it]
58%|█████▊ | 25419/43939 [41:00:34<29:26:34, 5.72s/it]
58%|█████▊ | 25420/43939 [41:00:40<29:27:17, 5.73s/it]
58%|█████▊ | 25421/43939 [41:00:46<29:29:12, 5.73s/it]
58%|█████▊ | 25422/43939 [41:00:51<29:24:02, 5.72s/it]
58%|█████▊ | 25423/43939 [41:00:57<29:23:37, 5.71s/it]
58%|█████▊ | 25424/43939 [41:01:03<29:23:42, 5.72s/it]
58%|█████▊ | 25425/43939 [41:01:09<29:25:36, 5.72s/it]
58%|█████▊ | 25426/43939 [41:01:14<29:24:06, 5.72s/it]
58%|█████▊ | 25427/43939 [41:01:20<29:25:37, 5.72s/it]
58%|█████▊ | 25428/43939 [41:01:26<29:18:13, 5.70s/it]
58%|█████▊ | 25429/43939 [41:01:31<29:18:01, 5.70s/it]
58%|█████▊ | 25430/43939 [41:01:37<29:19:57, 5.71s/it]
58%|█████▊ | 25431/43939 [41:01:43<29:16:23, 5.69s/it]
58%|█████▊ | 25432/43939 [41:01:48<29:18:14, 5.70s/it]
58%|█████▊ | 25433/43939 [41:01:54<29:21:48, 5.71s/it]
58%|█████▊ | 25434/43939 [41:02:00<29:18:30, 5.70s/it]
58%|█████▊ | 25435/43939 [41:02:06<29:17:13, 5.70s/it]
58%|█████▊ | 25436/43939 [41:02:11<29:39:03, 5.77s/it]
58%|█████▊ | 25437/43939 [41:02:17<29:32:36, 5.75s/it]
58%|█████▊ | 25438/43939 [41:02:23<29:28:59, 5.74s/it]
58%|█████▊ | 25439/43939 [41:02:29<29:25:51, 5.73s/it]
58%|█████▊ | 25440/43939 [41:02:34<29:27:39, 5.73s/it]
58%|█████▊ | 25441/43939 [41:02:40<29:26:29, 5.73s/it]
58%|█████▊ | 25442/43939 [41:02:46<29:32:34, 5.75s/it]
58%|█████▊ | 25443/43939 [41:02:52<29:32:25, 5.75s/it]
58%|█████▊ | 25444/43939 [41:02:57<29:26:56, 5.73s/it]
58%|█████▊ | 25445/43939 [41:03:03<29:22:19, 5.72s/it]
58%|█████▊ | 25446/43939 [41:03:09<29:19:53, 5.71s/it]
58%|█████▊ | 25447/43939 [41:03:14<29:17:33, 5.70s/it]
58%|█████▊ | 25448/43939 [41:03:20<29:16:41, 5.70s/it]
58%|█████▊ | 25449/43939 [41:03:26<29:15:04, 5.70s/it]
58%|█████▊ | 25450/43939 [41:03:31<29:19:49, 5.71s/it]
{'loss': 0.7758, 'grad_norm': 1.1549259424209595, 'learning_rate': 2.2456324078494817e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.57, 'epoch': 2.32}
58%|█████▊ | 25450/43939 [41:03:31<29:19:49, 5.71s/it]
58%|█████▊ | 25451/43939 [41:03:37<29:19:50, 5.71s/it]
58%|█████▊ | 25452/43939 [41:03:43<29:20:23, 5.71s/it]
58%|█████▊ | 25453/43939 [41:03:49<29:19:48, 5.71s/it]
58%|█████▊ | 25454/43939 [41:03:54<29:21:26, 5.72s/it]
58%|█████▊ | 25455/43939 [41:04:00<29:19:09, 5.71s/it]
58%|█████▊ | 25456/43939 [41:04:06<29:16:32, 5.70s/it]
58%|█████▊ | 25457/43939 [41:04:11<29:13:02, 5.69s/it]
58%|█████▊ | 25458/43939 [41:04:17<29:35:48, 5.77s/it]
58%|█████▊ | 25459/43939 [41:04:23<29:34:01, 5.76s/it]
58%|█████▊ | 25460/43939 [41:04:29<29:27:55, 5.74s/it]
58%|█████▊ | 25461/43939 [41:04:34<29:24:14, 5.73s/it]
58%|█████▊ | 25462/43939 [41:04:40<29:20:33, 5.72s/it]
58%|█████▊ | 25463/43939 [41:04:46<29:22:13, 5.72s/it]
58%|█████▊ | 25464/43939 [41:04:52<29:22:09, 5.72s/it]
58%|█████▊ | 25465/43939 [41:04:57<29:22:48, 5.73s/it]
58%|█████▊ | 25466/43939 [41:05:03<29:20:55, 5.72s/it]
58%|█████▊ | 25467/43939 [41:05:09<29:17:30, 5.71s/it]
58%|█████▊ | 25468/43939 [41:05:14<29:14:11, 5.70s/it]
58%|█████▊ | 25469/43939 [41:05:20<29:15:43, 5.70s/it]
58%|█████▊ | 25470/43939 [41:05:26<29:15:46, 5.70s/it]
58%|█████▊ | 25471/43939 [41:05:32<29:20:36, 5.72s/it]
58%|█████▊ | 25472/43939 [41:05:37<29:23:40, 5.73s/it]
58%|█████▊ | 25473/43939 [41:05:43<29:21:30, 5.72s/it]
58%|█████▊ | 25474/43939 [41:05:49<29:20:44, 5.72s/it]
58%|█████▊ | 25475/43939 [41:05:54<29:21:11, 5.72s/it]
58%|█████▊ | 25476/43939 [41:06:00<29:21:53, 5.73s/it]
58%|█████▊ | 25477/43939 [41:06:06<29:14:38, 5.70s/it]
58%|█████▊ | 25478/43939 [41:06:12<29:18:28, 5.72s/it]
58%|█████▊ | 25479/43939 [41:06:18<29:38:51, 5.78s/it]
58%|█████▊ | 25480/43939 [41:06:23<29:32:07, 5.76s/it]
58%|█████▊ | 25481/43939 [41:06:29<29:27:13, 5.74s/it]
58%|█████▊ | 25482/43939 [41:06:35<29:24:38, 5.74s/it]
58%|█████▊ | 25483/43939 [41:06:40<29:21:27, 5.73s/it]
58%|█████▊ | 25484/43939 [41:06:46<29:18:35, 5.72s/it]
58%|█████▊ | 25485/43939 [41:06:52<29:19:04, 5.72s/it]
58%|█████▊ | 25486/43939 [41:06:57<29:16:15, 5.71s/it]
58%|█████▊ | 25487/43939 [41:07:03<29:15:52, 5.71s/it]
58%|█████▊ | 25488/43939 [41:07:09<29:15:34, 5.71s/it]
58%|█████▊ | 25489/43939 [41:07:15<29:12:37, 5.70s/it]
58%|█████▊ | 25490/43939 [41:07:20<29:13:20, 5.70s/it]
58%|█████▊ | 25491/43939 [41:07:26<29:12:18, 5.70s/it]
58%|█████▊ | 25492/43939 [41:07:32<29:10:05, 5.69s/it]
58%|█████▊ | 25493/43939 [41:07:37<29:06:50, 5.68s/it]
58%|█████▊ | 25494/43939 [41:07:43<29:08:24, 5.69s/it]
58%|█████▊ | 25495/43939 [41:07:49<29:13:14, 5.70s/it]
58%|█████▊ | 25496/43939 [41:07:55<29:19:35, 5.72s/it]
58%|█████▊ | 25497/43939 [41:08:00<29:18:42, 5.72s/it]
58%|█████▊ | 25498/43939 [41:08:06<29:18:25, 5.72s/it]
58%|█████▊ | 25499/43939 [41:08:12<29:17:44, 5.72s/it]
58%|█████▊ | 25500/43939 [41:08:17<29:16:05, 5.71s/it]
{'loss': 0.7813, 'grad_norm': 1.4226809740066528, 'learning_rate': 2.2357557904528954e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 520.0, 'epoch': 2.32}
58%|█████▊ | 25500/43939 [41:08:17<29:16:05, 5.71s/it]
58%|█████▊ | 25501/43939 [41:08:23<29:38:22, 5.79s/it]
58%|█████▊ | 25502/43939 [41:08:29<29:28:57, 5.76s/it]
58%|█████▊ | 25503/43939 [41:08:35<29:22:10, 5.74s/it]
58%|█████▊ | 25504/43939 [41:08:40<29:19:28, 5.73s/it]
58%|█████▊ | 25505/43939 [41:08:46<29:21:07, 5.73s/it]
58%|█████▊ | 25506/43939 [41:08:52<29:20:51, 5.73s/it]
58%|█████▊ | 25507/43939 [41:08:58<29:18:22, 5.72s/it]
58%|█████▊ | 25508/43939 [41:09:03<29:19:18, 5.73s/it]
58%|█████▊ | 25509/43939 [41:09:09<29:17:08, 5.72s/it]
58%|█████▊ | 25510/43939 [41:09:15<29:15:03, 5.71s/it]
58%|█████▊ | 25511/43939 [41:09:20<29:14:13, 5.71s/it]
58%|█████▊ | 25512/43939 [41:09:26<29:12:22, 5.71s/it]
58%|█████▊ | 25513/43939 [41:09:32<29:11:10, 5.70s/it]
58%|█████▊ | 25514/43939 [41:09:38<29:12:22, 5.71s/it]
58%|█████▊ | 25515/43939 [41:09:43<29:12:53, 5.71s/it]
58%|█████▊ | 25516/43939 [41:09:49<29:08:10, 5.69s/it]
58%|█████▊ | 25517/43939 [41:09:55<29:08:13, 5.69s/it]
58%|█████▊ | 25518/43939 [41:10:00<29:07:47, 5.69s/it]
58%|█████▊ | 25519/43939 [41:10:06<29:05:49, 5.69s/it]
58%|█████▊ | 25520/43939 [41:10:12<29:07:03, 5.69s/it]
58%|█████▊ | 25521/43939 [41:10:17<29:09:15, 5.70s/it]
58%|█████▊ | 25522/43939 [41:10:23<29:37:30, 5.79s/it]
58%|█████▊ | 25523/43939 [41:10:29<29:27:02, 5.76s/it]
58%|█████▊ | 25524/43939 [41:10:35<29:21:37, 5.74s/it]
58%|█████▊ | 25525/43939 [41:10:40<29:18:38, 5.73s/it]
58%|█████▊ | 25526/43939 [41:10:46<29:15:46, 5.72s/it]
58%|█████▊ | 25527/43939 [41:10:52<29:14:05, 5.72s/it]
58%|█████▊ | 25528/43939 [41:10:58<29:14:25, 5.72s/it]
58%|█████▊ | 25529/43939 [41:11:03<29:13:19, 5.71s/it]
58%|█████▊ | 25530/43939 [41:11:09<29:13:56, 5.72s/it]
58%|█████▊ | 25531/43939 [41:11:15<29:11:25, 5.71s/it]
58%|█████▊ | 25532/43939 [41:11:20<29:07:35, 5.70s/it]
58%|█████▊ | 25533/43939 [41:11:26<29:07:54, 5.70s/it]
58%|█████▊ | 25534/43939 [41:11:32<29:08:12, 5.70s/it]
58%|█████▊ | 25535/43939 [41:11:38<29:08:36, 5.70s/it]
58%|█████▊ | 25536/43939 [41:11:43<29:07:22, 5.70s/it]
58%|█████▊ | 25537/43939 [41:11:49<29:07:53, 5.70s/it]
58%|█████▊ | 25538/43939 [41:11:55<29:07:22, 5.70s/it]
58%|█████▊ | 25539/43939 [41:12:00<29:08:05, 5.70s/it]
58%|█████▊ | 25540/43939 [41:12:06<29:09:53, 5.71s/it]
58%|█████▊ | 25541/43939 [41:12:12<29:10:19, 5.71s/it]
58%|█████▊ | 25542/43939 [41:12:17<29:06:02, 5.69s/it]
58%|█████▊ | 25543/43939 [41:12:23<29:28:43, 5.77s/it]
58%|█████▊ | 25544/43939 [41:12:29<29:21:50, 5.75s/it]
58%|█████▊ | 25545/43939 [41:12:35<29:22:31, 5.75s/it]
58%|█████▊ | 25546/43939 [41:12:41<29:19:25, 5.74s/it]
58%|█████▊ | 25547/43939 [41:12:46<29:15:42, 5.73s/it]
58%|█████▊ | 25548/43939 [41:12:52<29:16:19, 5.73s/it]
58%|█████▊ | 25549/43939 [41:12:58<29:11:28, 5.71s/it]
58%|█████▊ | 25550/43939 [41:13:03<29:12:11, 5.72s/it]
{'loss': 0.783, 'grad_norm': 1.2407307624816895, 'learning_rate': 2.2258833421296805e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.26, 'epoch': 2.33}
58%|█████▊ | 25550/43939 [41:13:03<29:12:11, 5.72s/it]
58%|█████▊ | 25551/43939 [41:13:09<29:11:00, 5.71s/it]
58%|█████▊ | 25552/43939 [41:13:15<29:09:40, 5.71s/it]
58%|█████▊ | 25553/43939 [41:13:20<29:08:49, 5.71s/it]
58%|█████▊ | 25554/43939 [41:13:26<29:07:54, 5.70s/it]
58%|█████▊ | 25555/43939 [41:13:32<29:05:59, 5.70s/it]
58%|█████▊ | 25556/43939 [41:13:38<29:05:37, 5.70s/it]
58%|█████▊ | 25557/43939 [41:13:43<29:06:56, 5.70s/it]
58%|█████▊ | 25558/43939 [41:13:49<29:05:01, 5.70s/it]
58%|█████▊ | 25559/43939 [41:13:55<29:08:19, 5.71s/it]
58%|█████▊ | 25560/43939 [41:14:00<29:06:09, 5.70s/it]
58%|█████▊ | 25561/43939 [41:14:06<29:05:59, 5.70s/it]
58%|█████▊ | 25562/43939 [41:14:12<29:03:12, 5.69s/it]
58%|█████▊ | 25563/43939 [41:14:17<29:02:41, 5.69s/it]
58%|█████▊ | 25564/43939 [41:14:23<29:22:42, 5.76s/it]
58%|█████▊ | 25565/43939 [41:14:29<29:20:19, 5.75s/it]
58%|█████▊ | 25566/43939 [41:14:35<29:17:27, 5.74s/it]
58%|█████▊ | 25567/43939 [41:14:40<29:10:09, 5.72s/it]
58%|█████▊ | 25568/43939 [41:14:46<29:08:30, 5.71s/it]
58%|█████▊ | 25569/43939 [41:14:52<29:08:40, 5.71s/it]
58%|█████▊ | 25570/43939 [41:14:58<29:10:46, 5.72s/it]
58%|█████▊ | 25571/43939 [41:15:03<29:10:18, 5.72s/it]
58%|█████▊ | 25572/43939 [41:15:09<29:11:01, 5.72s/it]
58%|█████▊ | 25573/43939 [41:15:15<29:08:05, 5.71s/it]
58%|█████▊ | 25574/43939 [41:15:20<29:08:38, 5.71s/it]
58%|█████▊ | 25575/43939 [41:15:26<29:09:29, 5.72s/it]
58%|█████▊ | 25576/43939 [41:15:32<29:08:44, 5.71s/it]
58%|█████▊ | 25577/43939 [41:15:38<29:08:53, 5.71s/it]
58%|█████▊ | 25578/43939 [41:15:43<29:06:05, 5.71s/it]
58%|█████▊ | 25579/43939 [41:15:49<29:03:33, 5.70s/it]
58%|█████▊ | 25580/43939 [41:15:55<29:08:21, 5.71s/it]
58%|█████▊ | 25581/43939 [41:16:00<29:08:09, 5.71s/it]
58%|█████▊ | 25582/43939 [41:16:06<29:09:20, 5.72s/it]
58%|█████▊ | 25583/43939 [41:16:12<29:09:07, 5.72s/it]
58%|█████▊ | 25584/43939 [41:16:18<29:05:29, 5.71s/it]
58%|█████▊ | 25585/43939 [41:16:23<29:05:30, 5.71s/it]
58%|█████▊ | 25586/43939 [41:16:29<29:27:25, 5.78s/it]
58%|█████▊ | 25587/43939 [41:16:35<29:21:03, 5.76s/it]
58%|█████▊ | 25588/43939 [41:16:41<29:13:54, 5.73s/it]
58%|█████▊ | 25589/43939 [41:16:46<29:13:03, 5.73s/it]
58%|█████▊ | 25590/43939 [41:16:52<29:08:32, 5.72s/it]
58%|█████▊ | 25591/43939 [41:16:58<29:04:30, 5.70s/it]
58%|█████▊ | 25592/43939 [41:17:03<29:03:47, 5.70s/it]
58%|█████▊ | 25593/43939 [41:17:09<29:03:04, 5.70s/it]
58%|█████▊ | 25594/43939 [41:17:15<29:01:56, 5.70s/it]
58%|█████▊ | 25595/43939 [41:17:20<29:04:36, 5.71s/it]
58%|█████▊ | 25596/43939 [41:17:26<29:05:13, 5.71s/it]
58%|█████▊ | 25597/43939 [41:17:32<29:01:21, 5.70s/it]
58%|█████▊ | 25598/43939 [41:17:38<29:00:33, 5.69s/it]
58%|█████▊ | 25599/43939 [41:17:43<29:00:20, 5.69s/it]
58%|█████▊ | 25600/43939 [41:17:49<28:59:11, 5.69s/it]
{'loss': 0.7754, 'grad_norm': 1.4377738237380981, 'learning_rate': 2.2160152186409086e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.07, 'epoch': 2.33}
58%|█████▊ | 25600/43939 [41:17:49<28:59:11, 5.69s/it]
58%|█████▊ | 25601/43939 [41:17:55<29:02:58, 5.70s/it]
58%|█████▊ | 25602/43939 [41:18:00<29:02:47, 5.70s/it]
58%|█████▊ | 25603/43939 [41:18:06<29:02:22, 5.70s/it]
58%|█████▊ | 25604/43939 [41:18:12<29:02:38, 5.70s/it]
58%|█████▊ | 25605/43939 [41:18:17<29:03:28, 5.71s/it]
58%|█████▊ | 25606/43939 [41:18:23<29:02:55, 5.70s/it]
58%|█████▊ | 25607/43939 [41:18:29<29:24:40, 5.78s/it]
58%|█████▊ | 25608/43939 [41:18:35<29:17:55, 5.75s/it]
58%|█████▊ | 25609/43939 [41:18:41<29:14:09, 5.74s/it]
58%|█████▊ | 25610/43939 [41:18:46<29:09:35, 5.73s/it]
58%|█████▊ | 25611/43939 [41:18:52<29:07:19, 5.72s/it]
58%|█████▊ | 25612/43939 [41:18:58<29:04:55, 5.71s/it]
58%|█████▊ | 25613/43939 [41:19:03<29:05:46, 5.72s/it]
58%|█████▊ | 25614/43939 [41:19:09<29:06:46, 5.72s/it]
58%|█████▊ | 25615/43939 [41:19:15<29:04:37, 5.71s/it]
58%|█████▊ | 25616/43939 [41:19:21<29:11:18, 5.73s/it]
58%|█████▊ | 25617/43939 [41:19:26<29:07:34, 5.72s/it]
58%|█████▊ | 25618/43939 [41:19:32<29:08:57, 5.73s/it]
58%|█████▊ | 25619/43939 [41:19:38<29:05:03, 5.72s/it]
58%|█████▊ | 25620/43939 [41:19:43<28:59:36, 5.70s/it]
58%|█████▊ | 25621/43939 [41:19:49<29:00:36, 5.70s/it]
58%|█████▊ | 25622/43939 [41:19:55<28:58:22, 5.69s/it]
58%|█████▊ | 25623/43939 [41:20:00<28:57:46, 5.69s/it]
58%|█████▊ | 25624/43939 [41:20:06<28:59:53, 5.70s/it]
58%|█████▊ | 25625/43939 [41:20:12<29:00:46, 5.70s/it]
58%|█████▊ | 25626/43939 [41:20:18<29:01:19, 5.71s/it]
58%|█████▊ | 25627/43939 [41:20:23<29:01:19, 5.71s/it]
58%|█████▊ | 25628/43939 [41:20:29<29:24:01, 5.78s/it]
58%|█████▊ | 25629/43939 [41:20:35<29:18:33, 5.76s/it]
58%|█████▊ | 25630/43939 [41:20:41<29:16:00, 5.75s/it]
58%|█████▊ | 25631/43939 [41:20:46<29:10:13, 5.74s/it]
58%|█████▊ | 25632/43939 [41:20:52<29:06:05, 5.72s/it]
58%|█████▊ | 25633/43939 [41:20:58<29:04:46, 5.72s/it]
58%|█████▊ | 25634/43939 [41:21:03<29:03:29, 5.71s/it]
58%|█████▊ | 25635/43939 [41:21:09<29:01:39, 5.71s/it]
58%|█████▊ | 25636/43939 [41:21:15<29:00:11, 5.70s/it]
58%|█████▊ | 25637/43939 [41:21:21<28:58:43, 5.70s/it]
58%|█████▊ | 25638/43939 [41:21:26<28:56:47, 5.69s/it]
58%|█████▊ | 25639/43939 [41:21:32<28:57:23, 5.70s/it]
58%|█████▊ | 25640/43939 [41:21:38<28:54:03, 5.69s/it]
58%|█████▊ | 25641/43939 [41:21:43<28:52:48, 5.68s/it]
58%|█████▊ | 25642/43939 [41:21:49<28:55:03, 5.69s/it]
58%|█████▊ | 25643/43939 [41:21:55<28:56:52, 5.70s/it]
58%|█████▊ | 25644/43939 [41:22:00<29:02:16, 5.71s/it]
58%|█████▊ | 25645/43939 [41:22:06<28:59:59, 5.71s/it]
58%|█████▊ | 25646/43939 [41:22:12<28:58:09, 5.70s/it]
58%|█████▊ | 25647/43939 [41:22:18<29:00:37, 5.71s/it]
58%|█████▊ | 25648/43939 [41:22:23<29:00:20, 5.71s/it]
58%|█████▊ | 25649/43939 [41:22:29<29:00:55, 5.71s/it]
58%|█████▊ | 25650/43939 [41:22:35<29:20:01, 5.77s/it]
{'loss': 0.7503, 'grad_norm': 1.3697234392166138, 'learning_rate': 2.206151575679418e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 450.66, 'epoch': 2.34}
58%|█████▊ | 25650/43939 [41:22:35<29:20:01, 5.77s/it]
58%|█████▊ | 25651/43939 [41:22:41<29:11:38, 5.75s/it]
58%|█████▊ | 25652/43939 [41:22:46<29:06:59, 5.73s/it]
58%|█████▊ | 25653/43939 [41:22:52<29:05:14, 5.73s/it]
58%|█████▊ | 25654/43939 [41:22:58<29:03:51, 5.72s/it]
58%|█████▊ | 25655/43939 [41:23:03<29:05:03, 5.73s/it]
58%|█████▊ | 25656/43939 [41:23:09<29:03:31, 5.72s/it]
58%|█████▊ | 25657/43939 [41:23:15<29:00:03, 5.71s/it]
58%|█████▊ | 25658/43939 [41:23:21<28:59:50, 5.71s/it]
58%|█████▊ | 25659/43939 [41:23:26<28:57:20, 5.70s/it]
58%|█████▊ | 25660/43939 [41:23:32<29:00:36, 5.71s/it]
58%|█████▊ | 25661/43939 [41:23:38<28:56:47, 5.70s/it]
58%|█████▊ | 25662/43939 [41:23:43<28:52:41, 5.69s/it]
58%|█████▊ | 25663/43939 [41:23:49<28:53:33, 5.69s/it]
58%|█████▊ | 25664/43939 [41:23:55<28:59:50, 5.71s/it]
58%|█████▊ | 25665/43939 [41:24:00<29:01:21, 5.72s/it]
58%|█████▊ | 25666/43939 [41:24:06<28:57:33, 5.71s/it]
58%|█████▊ | 25667/43939 [41:24:12<28:57:46, 5.71s/it]
58%|█████▊ | 25668/43939 [41:24:18<28:56:39, 5.70s/it]
58%|█████▊ | 25669/43939 [41:24:23<28:54:56, 5.70s/it]
58%|█████▊ | 25670/43939 [41:24:29<28:52:27, 5.69s/it]
58%|█████▊ | 25671/43939 [41:24:35<29:19:03, 5.78s/it]
58%|█████▊ | 25672/43939 [41:24:41<29:08:57, 5.74s/it]
58%|█████▊ | 25673/43939 [41:24:46<29:04:58, 5.73s/it]
58%|█████▊ | 25674/43939 [41:24:52<29:01:18, 5.72s/it]
58%|█████▊ | 25675/43939 [41:24:58<28:58:58, 5.71s/it]
58%|█████▊ | 25676/43939 [41:25:03<28:58:32, 5.71s/it]
58%|█████▊ | 25677/43939 [41:25:09<28:57:24, 5.71s/it]
58%|█████▊ | 25678/43939 [41:25:15<28:59:20, 5.71s/it]
58%|█████▊ | 25679/43939 [41:25:21<29:02:03, 5.72s/it]
58%|█████▊ | 25680/43939 [41:25:26<28:58:31, 5.71s/it]
58%|█████▊ | 25681/43939 [41:25:32<29:02:11, 5.73s/it]
58%|█████▊ | 25682/43939 [41:25:38<29:05:12, 5.74s/it]
58%|█████▊ | 25683/43939 [41:25:43<28:58:34, 5.71s/it]
58%|█████▊ | 25684/43939 [41:25:49<28:57:25, 5.71s/it]
58%|█████▊ | 25685/43939 [41:25:55<28:57:52, 5.71s/it]
58%|█████▊ | 25686/43939 [41:26:00<28:54:32, 5.70s/it]
58%|█████▊ | 25687/43939 [41:26:06<29:00:36, 5.72s/it]
58%|█████▊ | 25688/43939 [41:26:12<28:56:56, 5.71s/it]
58%|█████▊ | 25689/43939 [41:26:18<28:56:27, 5.71s/it]
58%|█████▊ | 25690/43939 [41:26:23<28:52:08, 5.70s/it]
58%|█████▊ | 25691/43939 [41:26:29<28:53:03, 5.70s/it]
58%|█████▊ | 25692/43939 [41:26:35<29:15:08, 5.77s/it]
58%|█████▊ | 25693/43939 [41:26:41<29:07:26, 5.75s/it]
58%|█████▊ | 25694/43939 [41:26:46<29:03:59, 5.74s/it]
58%|█████▊ | 25695/43939 [41:26:52<29:02:23, 5.73s/it]
58%|█████▊ | 25696/43939 [41:26:58<29:00:07, 5.72s/it]
58%|█████▊ | 25697/43939 [41:27:03<28:54:57, 5.71s/it]
58%|█████▊ | 25698/43939 [41:27:09<28:53:20, 5.70s/it]
58%|█████▊ | 25699/43939 [41:27:15<28:52:45, 5.70s/it]
58%|█████▊ | 25700/43939 [41:27:21<28:54:20, 5.71s/it]
{'loss': 0.777, 'grad_norm': 1.3014477491378784, 'learning_rate': 2.196292568867355e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.95, 'epoch': 2.34}
58%|█████▊ | 25700/43939 [41:27:21<28:54:20, 5.71s/it]
58%|█████▊ | 25701/43939 [41:27:26<28:53:35, 5.70s/it]
58%|█████▊ | 25702/43939 [41:27:32<28:50:21, 5.69s/it]
58%|█████▊ | 25703/43939 [41:27:38<28:53:17, 5.70s/it]
58%|█████▊ | 25704/43939 [41:27:43<28:57:58, 5.72s/it]
59%|█████▊ | 25705/43939 [41:27:49<29:00:50, 5.73s/it]
59%|█████▊ | 25706/43939 [41:27:55<28:55:38, 5.71s/it]
59%|█████▊ | 25707/43939 [41:28:01<28:58:00, 5.72s/it]
59%|█████▊ | 25708/43939 [41:28:06<28:56:09, 5.71s/it]
59%|█████▊ | 25709/43939 [41:28:12<28:51:38, 5.70s/it]
59%|█████▊ | 25710/43939 [41:28:18<28:51:45, 5.70s/it]
59%|█████▊ | 25711/43939 [41:28:23<28:52:47, 5.70s/it]
59%|█████▊ | 25712/43939 [41:28:29<28:55:10, 5.71s/it]
59%|█████▊ | 25713/43939 [41:28:35<29:16:49, 5.78s/it]
59%|█████▊ | 25714/43939 [41:28:41<29:08:18, 5.76s/it]
59%|█████▊ | 25715/43939 [41:28:46<29:06:41, 5.75s/it]
59%|█████▊ | 25716/43939 [41:28:52<29:01:48, 5.73s/it]
59%|█████▊ | 25717/43939 [41:28:58<28:58:17, 5.72s/it]
59%|█████▊ | 25718/43939 [41:29:04<28:54:58, 5.71s/it]
59%|█████▊ | 25719/43939 [41:29:09<28:57:29, 5.72s/it]
59%|█████▊ | 25720/43939 [41:29:15<28:58:58, 5.73s/it]
59%|█████▊ | 25721/43939 [41:29:21<28:53:43, 5.71s/it]
59%|█████▊ | 25722/43939 [41:29:26<28:57:38, 5.72s/it]
59%|█████▊ | 25723/43939 [41:29:32<28:59:19, 5.73s/it]
59%|█████▊ | 25724/43939 [41:29:38<28:59:02, 5.73s/it]
59%|█████▊ | 25725/43939 [41:29:44<28:59:01, 5.73s/it]
59%|█████▊ | 25726/43939 [41:29:49<28:58:40, 5.73s/it]
59%|█████▊ | 25727/43939 [41:29:55<28:57:42, 5.72s/it]
59%|█████▊ | 25728/43939 [41:30:01<28:56:20, 5.72s/it]
59%|█████▊ | 25729/43939 [41:30:06<28:53:03, 5.71s/it]
59%|█████▊ | 25730/43939 [41:30:12<28:47:53, 5.69s/it]
59%|█████▊ | 25731/43939 [41:30:18<28:50:27, 5.70s/it]
59%|█████▊ | 25732/43939 [41:30:24<28:48:49, 5.70s/it]
59%|█████▊ | 25733/43939 [41:30:29<28:50:03, 5.70s/it]
59%|█████▊ | 25734/43939 [41:30:35<28:47:05, 5.69s/it]
59%|█████▊ | 25735/43939 [41:30:41<29:09:50, 5.77s/it]
59%|█████▊ | 25736/43939 [41:30:47<29:02:25, 5.74s/it]
59%|█████▊ | 25737/43939 [41:30:52<28:55:04, 5.72s/it]
59%|█████▊ | 25738/43939 [41:30:58<28:53:35, 5.71s/it]
59%|█████▊ | 25739/43939 [41:31:04<28:57:09, 5.73s/it]
59%|█████▊ | 25740/43939 [41:31:09<28:56:47, 5.73s/it]
59%|█████▊ | 25741/43939 [41:31:15<28:54:32, 5.72s/it]
59%|█████▊ | 25742/43939 [41:31:21<28:51:23, 5.71s/it]
59%|█████▊ | 25743/43939 [41:31:27<28:52:49, 5.71s/it]
59%|█████▊ | 25744/43939 [41:31:32<28:51:25, 5.71s/it]
59%|█████▊ | 25745/43939 [41:31:38<28:49:02, 5.70s/it]
59%|█████▊ | 25746/43939 [41:31:44<28:51:31, 5.71s/it]
59%|█████▊ | 25747/43939 [41:31:49<28:53:06, 5.72s/it]
59%|█████▊ | 25748/43939 [41:31:55<28:53:19, 5.72s/it]
59%|█████▊ | 25749/43939 [41:32:01<28:49:35, 5.71s/it]
59%|█████▊ | 25750/43939 [41:32:06<28:47:20, 5.70s/it]
{'loss': 0.7826, 'grad_norm': 1.505482792854309, 'learning_rate': 2.1864383537537202e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.49, 'epoch': 2.34}
59%|█████▊ | 25750/43939 [41:32:06<28:47:20, 5.70s/it]
59%|█████▊ | 25751/43939 [41:32:12<28:46:40, 5.70s/it]
59%|█████▊ | 25752/43939 [41:32:18<28:50:24, 5.71s/it]
59%|█████▊ | 25753/43939 [41:32:24<28:48:03, 5.70s/it]
59%|█████▊ | 25754/43939 [41:32:29<28:45:46, 5.69s/it]
59%|█████▊ | 25755/43939 [41:32:35<28:46:08, 5.70s/it]
59%|█████▊ | 25756/43939 [41:32:41<29:08:39, 5.77s/it]
59%|█████▊ | 25757/43939 [41:32:47<29:02:37, 5.75s/it]
59%|█████▊ | 25758/43939 [41:32:52<28:59:53, 5.74s/it]
59%|█████▊ | 25759/43939 [41:32:58<28:58:49, 5.74s/it]
59%|█████▊ | 25760/43939 [41:33:04<28:55:12, 5.73s/it]
59%|█████▊ | 25761/43939 [41:33:09<28:57:13, 5.73s/it]
59%|█████▊ | 25762/43939 [41:33:15<28:55:39, 5.73s/it]
59%|█████▊ | 25763/43939 [41:33:21<28:51:18, 5.72s/it]
59%|█████▊ | 25764/43939 [41:33:27<28:48:10, 5.71s/it]
59%|█████▊ | 25765/43939 [41:33:32<28:47:19, 5.70s/it]
59%|█████▊ | 25766/43939 [41:33:38<28:46:58, 5.70s/it]
59%|█████▊ | 25767/43939 [41:33:44<28:46:37, 5.70s/it]
59%|█████▊ | 25768/43939 [41:33:49<28:45:49, 5.70s/it]
59%|█████▊ | 25769/43939 [41:33:55<28:46:41, 5.70s/it]
59%|█████▊ | 25770/43939 [41:34:01<28:47:00, 5.70s/it]
59%|█████▊ | 25771/43939 [41:34:06<28:48:32, 5.71s/it]
59%|█████▊ | 25772/43939 [41:34:12<28:46:13, 5.70s/it]
59%|█████▊ | 25773/43939 [41:34:18<28:43:12, 5.69s/it]
59%|█████▊ | 25774/43939 [41:34:23<28:41:23, 5.69s/it]
59%|█████▊ | 25775/43939 [41:34:29<28:44:29, 5.70s/it]
59%|█████▊ | 25776/43939 [41:34:35<28:45:52, 5.70s/it]
59%|█████▊ | 25777/43939 [41:34:41<29:09:19, 5.78s/it]
59%|█████▊ | 25778/43939 [41:34:47<29:03:32, 5.76s/it]
59%|█████▊ | 25779/43939 [41:34:52<28:58:38, 5.74s/it]
59%|█████▊ | 25780/43939 [41:34:58<28:56:00, 5.74s/it]
59%|█████▊ | 25781/43939 [41:35:04<28:52:44, 5.73s/it]
59%|█████▊ | 25782/43939 [41:35:09<28:52:13, 5.72s/it]
59%|█████▊ | 25783/43939 [41:35:15<28:51:36, 5.72s/it]
59%|█████▊ | 25784/43939 [41:35:21<28:51:09, 5.72s/it]
59%|█████▊ | 25785/43939 [41:35:27<28:49:36, 5.72s/it]
59%|█████▊ | 25786/43939 [41:35:32<28:48:55, 5.71s/it]
59%|█████▊ | 25787/43939 [41:35:38<28:53:11, 5.73s/it]
59%|█████▊ | 25788/43939 [41:35:44<28:54:41, 5.73s/it]
59%|█████▊ | 25789/43939 [41:35:50<28:51:57, 5.73s/it]
59%|█████▊ | 25790/43939 [41:35:55<28:51:54, 5.73s/it]
59%|█████▊ | 25791/43939 [41:36:01<28:50:29, 5.72s/it]
59%|█████▊ | 25792/43939 [41:36:07<28:50:23, 5.72s/it]
59%|█████▊ | 25793/43939 [41:36:12<28:48:11, 5.71s/it]
59%|█████▊ | 25794/43939 [41:36:18<28:48:41, 5.72s/it]
59%|█████▊ | 25795/43939 [41:36:24<28:52:39, 5.73s/it]
59%|█████▊ | 25796/43939 [41:36:30<28:47:10, 5.71s/it]
59%|█████▊ | 25797/43939 [41:36:35<28:43:59, 5.70s/it]
59%|█████▊ | 25798/43939 [41:36:41<29:05:22, 5.77s/it]
59%|█████▊ | 25799/43939 [41:36:47<29:02:04, 5.76s/it]
59%|█████▊ | 25800/43939 [41:36:53<28:57:41, 5.75s/it]
{'loss': 0.7862, 'grad_norm': 1.2439770698547363, 'learning_rate': 2.1765890858119155e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 523.79, 'epoch': 2.35}
59%|█████▊ | 25800/43939 [41:36:53<28:57:41, 5.75s/it]
59%|█████▊ | 25801/43939 [41:36:58<28:54:20, 5.74s/it]
59%|█████▊ | 25802/43939 [41:37:04<28:55:00, 5.74s/it]
59%|█████▊ | 25803/43939 [41:37:10<28:51:30, 5.73s/it]
59%|█████▊ | 25804/43939 [41:37:16<28:55:11, 5.74s/it]
59%|█████▊ | 25805/43939 [41:37:21<28:48:46, 5.72s/it]
59%|█████▊ | 25806/43939 [41:37:27<28:47:23, 5.72s/it]
59%|█████▊ | 25807/43939 [41:37:33<28:45:35, 5.71s/it]
59%|█████▊ | 25808/43939 [41:37:38<28:44:42, 5.71s/it]
59%|█████▊ | 25809/43939 [41:37:44<28:43:24, 5.70s/it]
59%|█████▊ | 25810/43939 [41:37:50<28:42:00, 5.70s/it]
59%|█████▊ | 25811/43939 [41:37:55<28:48:17, 5.72s/it]
59%|█████▊ | 25812/43939 [41:38:01<28:47:28, 5.72s/it]
59%|█████▊ | 25813/43939 [41:38:07<28:47:48, 5.72s/it]
59%|█████▊ | 25814/43939 [41:38:13<28:45:35, 5.71s/it]
59%|█████▉ | 25815/43939 [41:38:18<28:45:21, 5.71s/it]
59%|█████▉ | 25816/43939 [41:38:24<28:42:16, 5.70s/it]
59%|█████▉ | 25817/43939 [41:38:30<28:43:57, 5.71s/it]
59%|█████▉ | 25818/43939 [41:38:35<28:46:27, 5.72s/it]
59%|█████▉ | 25819/43939 [41:38:41<28:46:56, 5.72s/it]
59%|█████▉ | 25820/43939 [41:38:47<29:06:17, 5.78s/it]
59%|█████▉ | 25821/43939 [41:38:53<28:58:34, 5.76s/it]
59%|█████▉ | 25822/43939 [41:38:59<28:55:47, 5.75s/it]
59%|█████▉ | 25823/43939 [41:39:04<28:49:11, 5.73s/it]
59%|█████▉ | 25824/43939 [41:39:10<28:47:30, 5.72s/it]
59%|█████▉ | 25825/43939 [41:39:16<28:47:52, 5.72s/it]
59%|█████▉ | 25826/43939 [41:39:21<28:47:42, 5.72s/it]
59%|█████▉ | 25827/43939 [41:39:27<28:44:16, 5.71s/it]
59%|█████▉ | 25828/43939 [41:39:33<28:46:10, 5.72s/it]
59%|█████▉ | 25829/43939 [41:39:38<28:42:24, 5.71s/it]
59%|█████▉ | 25830/43939 [41:39:44<28:42:28, 5.71s/it]
59%|█████▉ | 25831/43939 [41:39:50<28:41:41, 5.70s/it]
59%|█████▉ | 25832/43939 [41:39:56<28:39:16, 5.70s/it]
59%|█████▉ | 25833/43939 [41:40:01<28:38:41, 5.70s/it]
59%|█████▉ | 25834/43939 [41:40:07<28:41:48, 5.71s/it]
59%|█████▉ | 25835/43939 [41:40:13<28:40:39, 5.70s/it]
59%|█████▉ | 25836/43939 [41:40:18<28:41:11, 5.70s/it]
59%|█████▉ | 25837/43939 [41:40:24<28:39:05, 5.70s/it]
59%|█████▉ | 25838/43939 [41:40:30<28:34:53, 5.68s/it]
59%|█████▉ | 25839/43939 [41:40:35<28:37:58, 5.69s/it]
59%|█████▉ | 25840/43939 [41:40:41<28:38:07, 5.70s/it]
59%|█████▉ | 25841/43939 [41:40:47<28:38:55, 5.70s/it]
59%|█████▉ | 25842/43939 [41:40:53<29:21:02, 5.84s/it]
59%|█████▉ | 25843/43939 [41:40:59<29:08:43, 5.80s/it]
59%|█████▉ | 25844/43939 [41:41:04<29:02:47, 5.78s/it]
59%|█████▉ | 25845/43939 [41:41:10<28:53:15, 5.75s/it]
59%|█████▉ | 25846/43939 [41:41:16<28:55:14, 5.75s/it]
59%|█████▉ | 25847/43939 [41:41:22<28:48:15, 5.73s/it]
59%|█████▉ | 25848/43939 [41:41:27<28:45:55, 5.72s/it]
59%|█████▉ | 25849/43939 [41:41:33<28:38:40, 5.70s/it]
59%|█████▉ | 25850/43939 [41:41:39<28:42:00, 5.71s/it]
{'loss': 0.7912, 'grad_norm': 1.4096859693527222, 'learning_rate': 2.166744920437287e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.92, 'epoch': 2.35}
59%|█████▉ | 25850/43939 [41:41:39<28:42:00, 5.71s/it]
59%|█████▉ | 25851/43939 [41:41:44<28:39:25, 5.70s/it]
59%|█████▉ | 25852/43939 [41:41:50<28:39:19, 5.70s/it]
59%|█████▉ | 25853/43939 [41:41:56<28:45:24, 5.72s/it]
59%|█████▉ | 25854/43939 [41:42:01<28:43:06, 5.72s/it]
59%|█████▉ | 25855/43939 [41:42:07<28:42:20, 5.71s/it]
59%|█████▉ | 25856/43939 [41:42:13<28:45:45, 5.73s/it]
59%|█████▉ | 25857/43939 [41:42:19<28:43:26, 5.72s/it]
59%|█████▉ | 25858/43939 [41:42:24<28:43:00, 5.72s/it]
59%|█████▉ | 25859/43939 [41:42:30<28:36:47, 5.70s/it]
59%|█████▉ | 25860/43939 [41:42:36<28:37:43, 5.70s/it]
59%|█████▉ | 25861/43939 [41:42:41<28:34:53, 5.69s/it]
59%|█████▉ | 25862/43939 [41:42:47<28:57:54, 5.77s/it]
59%|█████▉ | 25863/43939 [41:42:53<28:50:47, 5.75s/it]
59%|█████▉ | 25864/43939 [41:42:59<28:46:42, 5.73s/it]
59%|█████▉ | 25865/43939 [41:43:04<28:42:14, 5.72s/it]
59%|█████▉ | 25866/43939 [41:43:10<28:41:49, 5.72s/it]
59%|█████▉ | 25867/43939 [41:43:16<28:46:52, 5.73s/it]
59%|█████▉ | 25868/43939 [41:43:22<28:44:05, 5.72s/it]
59%|█████▉ | 25869/43939 [41:43:27<28:45:02, 5.73s/it]
59%|█████▉ | 25870/43939 [41:43:33<28:40:56, 5.71s/it]
59%|█████▉ | 25871/43939 [41:43:39<28:41:04, 5.72s/it]
59%|█████▉ | 25872/43939 [41:43:44<28:35:21, 5.70s/it]
59%|█████▉ | 25873/43939 [41:43:50<28:36:37, 5.70s/it]
59%|█████▉ | 25874/43939 [41:43:56<28:36:15, 5.70s/it]
59%|█████▉ | 25875/43939 [41:44:02<28:34:47, 5.70s/it]
59%|█████▉ | 25876/43939 [41:44:07<28:35:20, 5.70s/it]
59%|█████▉ | 25877/43939 [41:44:13<28:39:14, 5.71s/it]
59%|█████▉ | 25878/43939 [41:44:19<28:36:35, 5.70s/it]
59%|█████▉ | 25879/43939 [41:44:24<28:36:24, 5.70s/it]
59%|█████▉ | 25880/43939 [41:44:30<28:38:54, 5.71s/it]
59%|█████▉ | 25881/43939 [41:44:36<28:41:57, 5.72s/it]
59%|█████▉ | 25882/43939 [41:44:42<28:41:59, 5.72s/it]
59%|█████▉ | 25883/43939 [41:44:48<29:07:07, 5.81s/it]
59%|█████▉ | 25884/43939 [41:44:53<29:01:27, 5.79s/it]
59%|█████▉ | 25885/43939 [41:44:59<28:52:49, 5.76s/it]
59%|█████▉ | 25886/43939 [41:45:05<28:44:35, 5.73s/it]
59%|█████▉ | 25887/43939 [41:45:10<28:44:52, 5.73s/it]
59%|█████▉ | 25888/43939 [41:45:16<28:45:06, 5.73s/it]
59%|█████▉ | 25889/43939 [41:45:22<28:43:17, 5.73s/it]
59%|█████▉ | 25890/43939 [41:45:28<28:44:23, 5.73s/it]
59%|█████▉ | 25891/43939 [41:45:33<28:40:46, 5.72s/it]
59%|█████▉ | 25892/43939 [41:45:39<28:40:31, 5.72s/it]
59%|█████▉ | 25893/43939 [41:45:45<28:37:54, 5.71s/it]
59%|█████▉ | 25894/43939 [41:45:50<28:35:58, 5.71s/it]
59%|█████▉ | 25895/43939 [41:45:56<28:38:40, 5.71s/it]
59%|█████▉ | 25896/43939 [41:46:02<28:37:10, 5.71s/it]
59%|█████▉ | 25897/43939 [41:46:08<28:38:44, 5.72s/it]
59%|█████▉ | 25898/43939 [41:46:13<28:36:28, 5.71s/it]
59%|█████▉ | 25899/43939 [41:46:19<28:37:13, 5.71s/it]
59%|█████▉ | 25900/43939 [41:46:25<28:37:57, 5.71s/it]
{'loss': 0.8071, 'grad_norm': 1.6719779968261719, 'learning_rate': 2.1569060129446777e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.04, 'epoch': 2.36}
59%|█████▉ | 25900/43939 [41:46:25<28:37:57, 5.71s/it]
59%|█████▉ | 25901/43939 [41:46:30<28:34:00, 5.70s/it]
59%|█████▉ | 25902/43939 [41:46:36<28:30:31, 5.69s/it]
59%|█████▉ | 25903/43939 [41:46:42<28:34:17, 5.70s/it]
59%|█████▉ | 25904/43939 [41:46:47<28:31:32, 5.69s/it]
59%|█████▉ | 25905/43939 [41:46:53<28:52:40, 5.76s/it]
59%|█████▉ | 25906/43939 [41:46:59<28:46:26, 5.74s/it]
59%|█████▉ | 25907/43939 [41:47:05<28:44:02, 5.74s/it]
59%|█████▉ | 25908/43939 [41:47:10<28:43:16, 5.73s/it]
59%|█████▉ | 25909/43939 [41:47:16<28:41:48, 5.73s/it]
59%|█████▉ | 25910/43939 [41:47:22<28:38:56, 5.72s/it]
59%|█████▉ | 25911/43939 [41:47:28<28:37:30, 5.72s/it]
59%|█████▉ | 25912/43939 [41:47:33<28:39:14, 5.72s/it]
59%|█████▉ | 25913/43939 [41:47:39<28:36:33, 5.71s/it]
59%|█████▉ | 25914/43939 [41:47:45<28:34:54, 5.71s/it]
59%|█████▉ | 25915/43939 [41:47:50<28:36:58, 5.72s/it]
59%|█████▉ | 25916/43939 [41:47:56<28:35:41, 5.71s/it]
59%|█████▉ | 25917/43939 [41:48:02<28:35:48, 5.71s/it]
59%|█████▉ | 25918/43939 [41:48:08<28:36:53, 5.72s/it]
59%|█████▉ | 25919/43939 [41:48:13<28:36:48, 5.72s/it]
59%|█████▉ | 25920/43939 [41:48:19<28:34:16, 5.71s/it]
59%|█████▉ | 25921/43939 [41:48:25<28:40:24, 5.73s/it]
59%|█████▉ | 25922/43939 [41:48:31<28:39:19, 5.73s/it]
59%|█████▉ | 25923/43939 [41:48:36<28:41:42, 5.73s/it]
59%|█████▉ | 25924/43939 [41:48:42<28:34:40, 5.71s/it]
59%|█████▉ | 25925/43939 [41:48:48<28:34:04, 5.71s/it]
59%|█████▉ | 25926/43939 [41:48:54<28:53:17, 5.77s/it]
59%|█████▉ | 25927/43939 [41:48:59<28:50:44, 5.77s/it]
59%|█████▉ | 25928/43939 [41:49:05<28:46:45, 5.75s/it]
59%|█████▉ | 25929/43939 [41:49:11<28:45:09, 5.75s/it]
59%|█████▉ | 25930/43939 [41:49:16<28:43:57, 5.74s/it]
59%|█████▉ | 25931/43939 [41:49:22<28:38:42, 5.73s/it]
59%|█████▉ | 25932/43939 [41:49:28<28:32:56, 5.71s/it]
59%|█████▉ | 25933/43939 [41:49:34<28:32:31, 5.71s/it]
59%|█████▉ | 25934/43939 [41:49:39<28:33:56, 5.71s/it]
59%|█████▉ | 25935/43939 [41:49:45<28:33:47, 5.71s/it]
59%|█████▉ | 25936/43939 [41:49:51<28:36:52, 5.72s/it]
59%|█████▉ | 25937/43939 [41:49:56<28:32:21, 5.71s/it]
59%|█████▉ | 25938/43939 [41:50:02<28:29:21, 5.70s/it]
59%|█████▉ | 25939/43939 [41:50:08<28:28:10, 5.69s/it]
59%|█████▉ | 25940/43939 [41:50:13<28:32:54, 5.71s/it]
59%|█████▉ | 25941/43939 [41:50:19<28:35:16, 5.72s/it]
59%|█████▉ | 25942/43939 [41:50:25<28:35:02, 5.72s/it]
59%|█████▉ | 25943/43939 [41:50:31<28:30:27, 5.70s/it]
59%|█████▉ | 25944/43939 [41:50:36<28:30:45, 5.70s/it]
59%|█████▉ | 25945/43939 [41:50:42<28:27:35, 5.69s/it]
59%|█████▉ | 25946/43939 [41:50:48<28:33:02, 5.71s/it]
59%|█████▉ | 25947/43939 [41:50:54<28:52:01, 5.78s/it]
59%|█████▉ | 25948/43939 [41:50:59<28:48:45, 5.77s/it]
59%|█████▉ | 25949/43939 [41:51:05<28:46:32, 5.76s/it]
59%|█████▉ | 25950/43939 [41:51:11<28:43:21, 5.75s/it]
{'loss': 0.7829, 'grad_norm': 1.2982172966003418, 'learning_rate': 2.147072518565974e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 525.2, 'epoch': 2.36}
59%|█████▉ | 25950/43939 [41:51:11<28:43:21, 5.75s/it]
59%|█████▉ | 25951/43939 [41:51:17<28:39:39, 5.74s/it]
59%|█████▉ | 25952/43939 [41:51:22<28:37:29, 5.73s/it]
59%|█████▉ | 25953/43939 [41:51:28<28:34:32, 5.72s/it]
59%|█████▉ | 25954/43939 [41:51:34<28:32:55, 5.71s/it]
59%|█████▉ | 25955/43939 [41:51:39<28:30:23, 5.71s/it]
59%|█████▉ | 25956/43939 [41:51:45<28:32:01, 5.71s/it]
59%|█████▉ | 25957/43939 [41:51:51<28:32:07, 5.71s/it]
59%|█████▉ | 25958/43939 [41:51:57<28:30:09, 5.71s/it]
59%|█████▉ | 25959/43939 [41:52:02<28:28:49, 5.70s/it]
59%|█████▉ | 25960/43939 [41:52:08<28:26:50, 5.70s/it]
59%|█████▉ | 25961/43939 [41:52:14<28:24:50, 5.69s/it]
59%|█████▉ | 25962/43939 [41:52:19<28:26:03, 5.69s/it]
59%|█████▉ | 25963/43939 [41:52:25<28:22:20, 5.68s/it]
59%|█████▉ | 25964/43939 [41:52:31<28:27:16, 5.70s/it]
59%|█████▉ | 25965/43939 [41:52:36<28:27:58, 5.70s/it]
59%|█████▉ | 25966/43939 [41:52:42<28:25:59, 5.70s/it]
59%|█████▉ | 25967/43939 [41:52:48<28:30:33, 5.71s/it]
59%|█████▉ | 25968/43939 [41:52:54<28:50:19, 5.78s/it]
59%|█████▉ | 25969/43939 [41:52:59<28:47:14, 5.77s/it]
59%|█████▉ | 25970/43939 [41:53:05<28:42:52, 5.75s/it]
59%|█████▉ | 25971/43939 [41:53:11<28:41:33, 5.75s/it]
59%|█████▉ | 25972/43939 [41:53:17<28:33:27, 5.72s/it]
59%|█████▉ | 25973/43939 [41:53:22<28:30:11, 5.71s/it]
59%|█████▉ | 25974/43939 [41:53:28<28:30:19, 5.71s/it]
59%|█████▉ | 25975/43939 [41:53:34<28:32:26, 5.72s/it]
59%|█████▉ | 25976/43939 [41:53:39<28:30:46, 5.71s/it]
59%|█████▉ | 25977/43939 [41:53:45<28:31:08, 5.72s/it]
59%|█████▉ | 25978/43939 [41:53:51<28:34:46, 5.73s/it]
59%|█████▉ | 25979/43939 [41:53:57<28:34:16, 5.73s/it]
59%|█████▉ | 25980/43939 [41:54:02<28:30:33, 5.71s/it]
59%|█████▉ | 25981/43939 [41:54:08<28:28:57, 5.71s/it]
59%|█████▉ | 25982/43939 [41:54:14<28:24:16, 5.69s/it]
59%|█████▉ | 25983/43939 [41:54:19<28:23:56, 5.69s/it]
59%|█████▉ | 25984/43939 [41:54:25<28:27:06, 5.70s/it]
59%|█████▉ | 25985/43939 [41:54:31<28:27:15, 5.71s/it]
59%|█████▉ | 25986/43939 [41:54:37<28:30:51, 5.72s/it]
59%|█████▉ | 25987/43939 [41:54:42<28:25:31, 5.70s/it]
59%|█████▉ | 25988/43939 [41:54:48<28:27:20, 5.71s/it]
59%|█████▉ | 25989/43939 [41:54:54<28:46:49, 5.77s/it]
59%|█████▉ | 25990/43939 [41:55:00<28:45:35, 5.77s/it]
59%|█████▉ | 25991/43939 [41:55:05<28:39:44, 5.75s/it]
59%|█████▉ | 25992/43939 [41:55:11<28:35:51, 5.74s/it]
59%|█████▉ | 25993/43939 [41:55:17<28:31:54, 5.72s/it]
59%|█████▉ | 25994/43939 [41:55:22<28:29:37, 5.72s/it]
59%|█████▉ | 25995/43939 [41:55:28<28:28:35, 5.71s/it]
59%|█████▉ | 25996/43939 [41:55:34<28:28:40, 5.71s/it]
59%|█████▉ | 25997/43939 [41:55:40<28:31:00, 5.72s/it]
59%|█████▉ | 25998/43939 [41:55:45<28:28:53, 5.72s/it]
59%|█████▉ | 25999/43939 [41:55:51<28:30:51, 5.72s/it]
59%|█████▉ | 26000/43939 [41:55:57<28:29:49, 5.72s/it]
{'loss': 0.7864, 'grad_norm': 1.4265486001968384, 'learning_rate': 2.137244592447658e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.69, 'epoch': 2.37}
59%|█████▉ | 26000/43939 [41:55:57<28:29:49, 5.72s/it]
59%|█████▉ | 26001/43939 [41:56:02<28:25:07, 5.70s/it]
59%|█████▉ | 26002/43939 [41:56:08<28:21:47, 5.69s/it]
59%|█████▉ | 26003/43939 [41:56:14<28:21:08, 5.69s/it]
59%|█████▉ | 26004/43939 [41:56:19<28:23:55, 5.70s/it]
59%|█████▉ | 26005/43939 [41:56:25<28:26:39, 5.71s/it]
59%|█████▉ | 26006/43939 [41:56:31<28:29:30, 5.72s/it]
59%|█████▉ | 26007/43939 [41:56:37<28:25:54, 5.71s/it]
59%|█████▉ | 26008/43939 [41:56:42<28:27:56, 5.72s/it]
59%|█████▉ | 26009/43939 [41:56:48<28:25:58, 5.71s/it]
59%|█████▉ | 26010/43939 [41:56:54<28:46:57, 5.78s/it]
59%|█████▉ | 26011/43939 [41:57:00<28:40:30, 5.76s/it]
59%|█████▉ | 26012/43939 [41:57:05<28:33:16, 5.73s/it]
59%|█████▉ | 26013/43939 [41:57:11<28:29:41, 5.72s/it]
59%|█████▉ | 26014/43939 [41:57:17<28:29:10, 5.72s/it]
59%|█████▉ | 26015/43939 [41:57:23<28:29:03, 5.72s/it]
59%|█████▉ | 26016/43939 [41:57:28<28:29:25, 5.72s/it]
59%|█████▉ | 26017/43939 [41:57:34<28:26:53, 5.71s/it]
59%|█████▉ | 26018/43939 [41:57:40<28:26:25, 5.71s/it]
59%|█████▉ | 26019/43939 [41:57:45<28:21:47, 5.70s/it]
59%|█████▉ | 26020/43939 [41:57:51<28:23:43, 5.70s/it]
59%|█████▉ | 26021/43939 [41:57:57<28:23:08, 5.70s/it]
59%|█████▉ | 26022/43939 [41:58:02<28:20:58, 5.70s/it]
59%|█████▉ | 26023/43939 [41:58:08<28:20:33, 5.70s/it]
59%|█████▉ | 26024/43939 [41:58:14<28:20:30, 5.70s/it]
59%|█████▉ | 26025/43939 [41:58:20<28:23:34, 5.71s/it]
59%|█████▉ | 26026/43939 [41:58:25<28:24:43, 5.71s/it]
59%|█████▉ | 26027/43939 [41:58:31<28:20:43, 5.70s/it]
59%|█████▉ | 26028/43939 [41:58:37<28:21:32, 5.70s/it]
59%|█████▉ | 26029/43939 [41:58:42<28:18:05, 5.69s/it]
59%|█████▉ | 26030/43939 [41:58:48<28:15:43, 5.68s/it]
59%|█████▉ | 26031/43939 [41:58:54<28:16:56, 5.69s/it]
59%|█████▉ | 26032/43939 [41:59:00<28:43:28, 5.77s/it]
59%|█████▉ | 26033/43939 [41:59:05<28:36:49, 5.75s/it]
59%|█████▉ | 26034/43939 [41:59:11<28:30:37, 5.73s/it]
59%|█████▉ | 26035/43939 [41:59:17<28:25:35, 5.72s/it]
59%|█████▉ | 26036/43939 [41:59:22<28:26:51, 5.72s/it]
59%|█████▉ | 26037/43939 [41:59:28<28:24:39, 5.71s/it]
59%|█████▉ | 26038/43939 [41:59:34<28:24:51, 5.71s/it]
59%|█████▉ | 26039/43939 [41:59:40<28:22:34, 5.71s/it]
59%|█████▉ | 26040/43939 [41:59:45<28:21:43, 5.70s/it]
59%|█████▉ | 26041/43939 [41:59:51<28:20:21, 5.70s/it]
59%|█████▉ | 26042/43939 [41:59:57<28:16:54, 5.69s/it]
59%|█████▉ | 26043/43939 [42:00:02<28:13:58, 5.68s/it]
59%|█████▉ | 26044/43939 [42:00:08<28:16:00, 5.69s/it]
59%|█████▉ | 26045/43939 [42:00:14<28:16:29, 5.69s/it]
59%|█████▉ | 26046/43939 [42:00:19<28:18:17, 5.69s/it]
59%|█████▉ | 26047/43939 [42:00:25<28:21:13, 5.70s/it]
59%|█████▉ | 26048/43939 [42:00:31<28:19:08, 5.70s/it]
59%|█████▉ | 26049/43939 [42:00:36<28:17:16, 5.69s/it]
59%|█████▉ | 26050/43939 [42:00:42<28:17:37, 5.69s/it]
{'loss': 0.7814, 'grad_norm': 1.3955106735229492, 'learning_rate': 2.1274223896483588e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.84, 'epoch': 2.37}
59%|█████▉ | 26050/43939 [42:00:42<28:17:37, 5.69s/it]
59%|█████▉ | 26051/43939 [42:00:48<28:18:40, 5.70s/it]
59%|█████▉ | 26052/43939 [42:00:54<28:17:25, 5.69s/it]
59%|█████▉ | 26053/43939 [42:00:59<28:41:07, 5.77s/it]
59%|█████▉ | 26054/43939 [42:01:05<28:37:59, 5.76s/it]
59%|█████▉ | 26055/43939 [42:01:11<28:32:07, 5.74s/it]
59%|█████▉ | 26056/43939 [42:01:17<28:25:10, 5.72s/it]
59%|█████▉ | 26057/43939 [42:01:22<28:27:15, 5.73s/it]
59%|█████▉ | 26058/43939 [42:01:28<28:26:51, 5.73s/it]
59%|█████▉ | 26059/43939 [42:01:34<28:22:03, 5.71s/it]
59%|█████▉ | 26060/43939 [42:01:39<28:19:27, 5.70s/it]
59%|█████▉ | 26061/43939 [42:01:45<28:19:33, 5.70s/it]
59%|█████▉ | 26062/43939 [42:01:51<28:22:07, 5.71s/it]
59%|█████▉ | 26063/43939 [42:01:57<28:19:25, 5.70s/it]
59%|█████▉ | 26064/43939 [42:02:02<28:22:18, 5.71s/it]
59%|█████▉ | 26065/43939 [42:02:08<28:18:37, 5.70s/it]
59%|█████▉ | 26066/43939 [42:02:14<28:16:47, 5.70s/it]
59%|█████▉ | 26067/43939 [42:02:19<28:14:47, 5.69s/it]
59%|█████▉ | 26068/43939 [42:02:25<28:17:31, 5.70s/it]
59%|█████▉ | 26069/43939 [42:02:31<28:13:50, 5.69s/it]
59%|█████▉ | 26070/43939 [42:02:36<28:17:27, 5.70s/it]
59%|█████▉ | 26071/43939 [42:02:42<28:17:43, 5.70s/it]
59%|█████▉ | 26072/43939 [42:02:48<28:20:07, 5.71s/it]
59%|█████▉ | 26073/43939 [42:02:54<28:18:23, 5.70s/it]
59%|█████▉ | 26074/43939 [42:02:59<28:38:21, 5.77s/it]
59%|█████▉ | 26075/43939 [42:03:05<28:32:42, 5.75s/it]
59%|█████▉ | 26076/43939 [42:03:11<28:28:58, 5.74s/it]
59%|█████▉ | 26077/43939 [42:03:17<28:28:18, 5.74s/it]
59%|█████▉ | 26078/43939 [42:03:22<28:23:29, 5.72s/it]
59%|█████▉ | 26079/43939 [42:03:28<28:23:05, 5.72s/it]
59%|█████▉ | 26080/43939 [42:03:34<28:23:08, 5.72s/it]
59%|█████▉ | 26081/43939 [42:03:39<28:19:22, 5.71s/it]
59%|█████▉ | 26082/43939 [42:03:45<28:20:27, 5.71s/it]
59%|█████▉ | 26083/43939 [42:03:51<28:20:04, 5.71s/it]
59%|█████▉ | 26084/43939 [42:03:57<28:17:26, 5.70s/it]
59%|█████▉ | 26085/43939 [42:04:02<28:16:55, 5.70s/it]
59%|█████▉ | 26086/43939 [42:04:08<28:12:49, 5.69s/it]
59%|█████▉ | 26087/43939 [42:04:14<28:15:44, 5.70s/it]
59%|█████▉ | 26088/43939 [42:04:19<28:18:57, 5.71s/it]
59%|█████▉ | 26089/43939 [42:04:25<28:18:25, 5.71s/it]
59%|█████▉ | 26090/43939 [42:04:31<28:17:43, 5.71s/it]
59%|█████▉ | 26091/43939 [42:04:37<28:22:18, 5.72s/it]
59%|█████▉ | 26092/43939 [42:04:42<28:25:17, 5.73s/it]
59%|█████▉ | 26093/43939 [42:04:48<28:26:54, 5.74s/it]
59%|█████▉ | 26094/43939 [42:04:54<28:19:46, 5.72s/it]
59%|█████▉ | 26095/43939 [42:05:00<28:37:56, 5.78s/it]
59%|█████▉ | 26096/43939 [42:05:05<28:30:19, 5.75s/it]
59%|█████▉ | 26097/43939 [42:05:11<28:32:35, 5.76s/it]
59%|█████▉ | 26098/43939 [42:05:17<28:26:33, 5.74s/it]
59%|█████▉ | 26099/43939 [42:05:23<28:24:25, 5.73s/it]
59%|█████▉ | 26100/43939 [42:05:28<28:21:05, 5.72s/it]
{'loss': 0.7529, 'grad_norm': 1.1367692947387695, 'learning_rate': 2.117606065136409e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.36, 'epoch': 2.38}
59%|█████▉ | 26100/43939 [42:05:28<28:21:05, 5.72s/it]
59%|█████▉ | 26101/43939 [42:05:34<28:25:40, 5.74s/it]
59%|█████▉ | 26102/43939 [42:05:40<28:23:20, 5.73s/it]
59%|█████▉ | 26103/43939 [42:05:45<28:19:26, 5.72s/it]
59%|█████▉ | 26104/43939 [42:05:51<28:15:55, 5.71s/it]
59%|█████▉ | 26105/43939 [42:05:57<28:14:47, 5.70s/it]
59%|█████▉ | 26106/43939 [42:06:02<28:15:35, 5.70s/it]
59%|█████▉ | 26107/43939 [42:06:08<28:15:08, 5.70s/it]
59%|█████▉ | 26108/43939 [42:06:14<28:14:00, 5.70s/it]
59%|█████▉ | 26109/43939 [42:06:20<28:11:39, 5.69s/it]
59%|█████▉ | 26110/43939 [42:06:25<28:13:50, 5.70s/it]
59%|█████▉ | 26111/43939 [42:06:31<28:19:35, 5.72s/it]
59%|█████▉ | 26112/43939 [42:06:37<28:20:39, 5.72s/it]
59%|█████▉ | 26113/43939 [42:06:42<28:16:51, 5.71s/it]
59%|█████▉ | 26114/43939 [42:06:48<28:18:18, 5.72s/it]
59%|█████▉ | 26115/43939 [42:06:54<28:14:34, 5.70s/it]
59%|█████▉ | 26116/43939 [42:07:00<28:17:24, 5.71s/it]
59%|█████▉ | 26117/43939 [42:07:06<28:37:59, 5.78s/it]
59%|█████▉ | 26118/43939 [42:07:11<28:31:26, 5.76s/it]
59%|█████▉ | 26119/43939 [42:07:17<28:26:27, 5.75s/it]
59%|█████▉ | 26120/43939 [42:07:23<28:22:55, 5.73s/it]
59%|█████▉ | 26121/43939 [42:07:28<28:20:56, 5.73s/it]
59%|█████▉ | 26122/43939 [42:07:34<28:17:11, 5.72s/it]
59%|█████▉ | 26123/43939 [42:07:40<28:19:58, 5.73s/it]
59%|█████▉ | 26124/43939 [42:07:46<28:22:17, 5.73s/it]
59%|█████▉ | 26125/43939 [42:07:51<28:19:16, 5.72s/it]
59%|█████▉ | 26126/43939 [42:07:57<28:16:57, 5.72s/it]
59%|█████▉ | 26127/43939 [42:08:03<28:13:27, 5.70s/it]
59%|█████▉ | 26128/43939 [42:08:08<28:10:17, 5.69s/it]
59%|█████▉ | 26129/43939 [42:08:14<28:11:57, 5.70s/it]
59%|█████▉ | 26130/43939 [42:08:20<28:14:14, 5.71s/it]
59%|█████▉ | 26131/43939 [42:08:25<28:11:01, 5.70s/it]
59%|█████▉ | 26132/43939 [42:08:31<28:11:45, 5.70s/it]
59%|█████▉ | 26133/43939 [42:08:37<28:12:52, 5.70s/it]
59%|█████▉ | 26134/43939 [42:08:42<28:09:31, 5.69s/it]
59%|█████▉ | 26135/43939 [42:08:48<28:12:36, 5.70s/it]
59%|█████▉ | 26136/43939 [42:08:54<28:12:09, 5.70s/it]
59%|█████▉ | 26137/43939 [42:09:00<28:10:10, 5.70s/it]
59%|█████▉ | 26138/43939 [42:09:06<28:32:00, 5.77s/it]
59%|█████▉ | 26139/43939 [42:09:11<28:29:14, 5.76s/it]
59%|█████▉ | 26140/43939 [42:09:17<28:24:38, 5.75s/it]
59%|█████▉ | 26141/43939 [42:09:23<28:18:13, 5.73s/it]
59%|█████▉ | 26142/43939 [42:09:28<28:15:48, 5.72s/it]
59%|█████▉ | 26143/43939 [42:09:34<28:14:45, 5.71s/it]
60%|█████▉ | 26144/43939 [42:09:40<28:11:44, 5.70s/it]
60%|█████▉ | 26145/43939 [42:09:45<28:08:49, 5.69s/it]
60%|█████▉ | 26146/43939 [42:09:51<28:11:01, 5.70s/it]
60%|█████▉ | 26147/43939 [42:09:57<28:12:33, 5.71s/it]
60%|█████▉ | 26148/43939 [42:10:03<28:17:20, 5.72s/it]
60%|█████▉ | 26149/43939 [42:10:08<28:18:29, 5.73s/it]
60%|█████▉ | 26150/43939 [42:10:14<28:17:39, 5.73s/it]
{'loss': 0.8125, 'grad_norm': 1.3815332651138306, 'learning_rate': 2.107795773787395e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.08, 'epoch': 2.38}
60%|█████▉ | 26150/43939 [42:10:14<28:17:39, 5.73s/it]
60%|█████▉ | 26151/43939 [42:10:20<28:16:13, 5.72s/it]
60%|█████▉ | 26152/43939 [42:10:26<28:15:30, 5.72s/it]
60%|█████▉ | 26153/43939 [42:10:31<28:12:43, 5.71s/it]
60%|█████▉ | 26154/43939 [42:10:37<28:11:58, 5.71s/it]
60%|█████▉ | 26155/43939 [42:10:43<28:10:25, 5.70s/it]
60%|█████▉ | 26156/43939 [42:10:48<28:11:32, 5.71s/it]
60%|█████▉ | 26157/43939 [42:10:54<28:08:59, 5.70s/it]
60%|█████▉ | 26158/43939 [42:11:00<28:09:06, 5.70s/it]
60%|█████▉ | 26159/43939 [42:11:06<28:29:26, 5.77s/it]
60%|█████▉ | 26160/43939 [42:11:11<28:19:44, 5.74s/it]
60%|█████▉ | 26161/43939 [42:11:17<28:17:41, 5.73s/it]
60%|█████▉ | 26162/43939 [42:11:23<28:22:41, 5.75s/it]
60%|█████▉ | 26163/43939 [42:11:28<28:18:22, 5.73s/it]
60%|█████▉ | 26164/43939 [42:11:34<28:20:30, 5.74s/it]
60%|█████▉ | 26165/43939 [42:11:40<28:19:32, 5.74s/it]
60%|█████▉ | 26166/43939 [42:11:46<28:20:12, 5.74s/it]
60%|█████▉ | 26167/43939 [42:11:51<28:16:32, 5.73s/it]
60%|█████▉ | 26168/43939 [42:11:57<28:24:38, 5.76s/it]
60%|█████▉ | 26169/43939 [42:12:03<28:21:05, 5.74s/it]
60%|█████▉ | 26170/43939 [42:12:09<28:16:49, 5.73s/it]
60%|█████▉ | 26171/43939 [42:12:14<28:11:45, 5.71s/it]
60%|█████▉ | 26172/43939 [42:12:20<28:11:02, 5.71s/it]
60%|█████▉ | 26173/43939 [42:12:26<28:11:43, 5.71s/it]
60%|█████▉ | 26174/43939 [42:12:31<28:05:35, 5.69s/it]
60%|█████▉ | 26175/43939 [42:12:37<28:04:51, 5.69s/it]
60%|█████▉ | 26176/43939 [42:12:43<28:08:43, 5.70s/it]
60%|█████▉ | 26177/43939 [42:12:49<28:10:20, 5.71s/it]
60%|█████▉ | 26178/43939 [42:12:54<28:06:38, 5.70s/it]
60%|█████▉ | 26179/43939 [42:13:00<28:07:32, 5.70s/it]
60%|█████▉ | 26180/43939 [42:13:06<28:07:22, 5.70s/it]
60%|█████▉ | 26181/43939 [42:13:12<28:30:21, 5.78s/it]
60%|█████▉ | 26182/43939 [42:13:17<28:23:32, 5.76s/it]
60%|█████▉ | 26183/43939 [42:13:23<28:19:57, 5.74s/it]
60%|█████▉ | 26184/43939 [42:13:29<28:13:56, 5.72s/it]
60%|█████▉ | 26185/43939 [42:13:34<28:10:20, 5.71s/it]
60%|█████▉ | 26186/43939 [42:13:40<28:05:56, 5.70s/it]
60%|█████▉ | 26187/43939 [42:13:46<28:03:15, 5.69s/it]
60%|█████▉ | 26188/43939 [42:13:51<28:03:59, 5.69s/it]
60%|█████▉ | 26189/43939 [42:13:57<28:03:49, 5.69s/it]
60%|█████▉ | 26190/43939 [42:14:03<28:00:21, 5.68s/it]
60%|█████▉ | 26191/43939 [42:14:08<27:58:18, 5.67s/it]
60%|█████▉ | 26192/43939 [42:14:14<28:02:03, 5.69s/it]
60%|█████▉ | 26193/43939 [42:14:20<28:04:45, 5.70s/it]
60%|█████▉ | 26194/43939 [42:14:26<28:04:30, 5.70s/it]
60%|█████▉ | 26195/43939 [42:14:31<28:07:15, 5.71s/it]
60%|█████▉ | 26196/43939 [42:14:37<28:04:45, 5.70s/it]
60%|█████▉ | 26197/43939 [42:14:43<28:07:50, 5.71s/it]
60%|█████▉ | 26198/43939 [42:14:48<28:07:28, 5.71s/it]
60%|█████▉ | 26199/43939 [42:14:54<28:05:41, 5.70s/it]
60%|█████▉ | 26200/43939 [42:15:00<28:02:47, 5.69s/it]
{'loss': 0.7574, 'grad_norm': 1.6012133359909058, 'learning_rate': 2.097991670381717e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.46, 'epoch': 2.39}
60%|█████▉ | 26200/43939 [42:15:00<28:02:47, 5.69s/it]
60%|█████▉ | 26201/43939 [42:15:05<28:07:37, 5.71s/it]
60%|█████▉ | 26202/43939 [42:15:11<28:29:02, 5.78s/it]
60%|█████▉ | 26203/43939 [42:15:17<28:21:15, 5.76s/it]
60%|█████▉ | 26204/43939 [42:15:23<28:16:57, 5.74s/it]
60%|█████▉ | 26205/43939 [42:15:29<28:10:50, 5.72s/it]
60%|█████▉ | 26206/43939 [42:15:34<28:11:03, 5.72s/it]
60%|█████▉ | 26207/43939 [42:15:40<28:10:52, 5.72s/it]
60%|█████▉ | 26208/43939 [42:15:46<28:08:14, 5.71s/it]
60%|█████▉ | 26209/43939 [42:15:51<28:06:47, 5.71s/it]
60%|█████▉ | 26210/43939 [42:15:57<28:08:17, 5.71s/it]
60%|█████▉ | 26211/43939 [42:16:03<28:06:35, 5.71s/it]
60%|█████▉ | 26212/43939 [42:16:09<28:09:23, 5.72s/it]
60%|█████▉ | 26213/43939 [42:16:14<28:07:20, 5.71s/it]
60%|█████▉ | 26214/43939 [42:16:20<28:04:45, 5.70s/it]
60%|█████▉ | 26215/43939 [42:16:26<28:04:41, 5.70s/it]
60%|█████▉ | 26216/43939 [42:16:31<28:02:04, 5.69s/it]
60%|█████▉ | 26217/43939 [42:16:37<28:01:58, 5.69s/it]
60%|█████▉ | 26218/43939 [42:16:43<27:58:28, 5.68s/it]
60%|█████▉ | 26219/43939 [42:16:48<28:01:52, 5.69s/it]
60%|█████▉ | 26220/43939 [42:16:54<28:00:52, 5.69s/it]
60%|█████▉ | 26221/43939 [42:17:00<28:04:25, 5.70s/it]
60%|█████▉ | 26222/43939 [42:17:06<28:22:33, 5.77s/it]
60%|█████▉ | 26223/43939 [42:17:11<28:13:52, 5.74s/it]
60%|█████▉ | 26224/43939 [42:17:17<28:09:42, 5.72s/it]
60%|█████▉ | 26225/43939 [42:17:23<28:06:39, 5.71s/it]
60%|█████▉ | 26226/43939 [42:17:28<28:04:26, 5.71s/it]
60%|█████▉ | 26227/43939 [42:17:34<28:01:40, 5.70s/it]
60%|█████▉ | 26228/43939 [42:17:40<28:01:59, 5.70s/it]
60%|█████▉ | 26229/43939 [42:17:45<28:01:23, 5.70s/it]
60%|█████▉ | 26230/43939 [42:17:51<28:00:55, 5.70s/it]
60%|█████▉ | 26231/43939 [42:17:57<27:58:07, 5.69s/it]
60%|█████▉ | 26232/43939 [42:18:03<27:58:54, 5.69s/it]
60%|█████▉ | 26233/43939 [42:18:08<28:03:08, 5.70s/it]
60%|█████▉ | 26234/43939 [42:18:14<28:02:25, 5.70s/it]
60%|█████▉ | 26235/43939 [42:18:20<27:59:01, 5.69s/it]
60%|█████▉ | 26236/43939 [42:18:25<28:00:39, 5.70s/it]
60%|█████▉ | 26237/43939 [42:18:31<28:02:08, 5.70s/it]
60%|█████▉ | 26238/43939 [42:18:37<28:02:59, 5.70s/it]
60%|█████▉ | 26239/43939 [42:18:42<28:01:47, 5.70s/it]
60%|█████▉ | 26240/43939 [42:18:48<28:02:34, 5.70s/it]
60%|█████▉ | 26241/43939 [42:18:54<27:59:44, 5.69s/it]
60%|█████▉ | 26242/43939 [42:19:00<27:57:27, 5.69s/it]
60%|█████▉ | 26243/43939 [42:19:05<28:19:24, 5.76s/it]
60%|█████▉ | 26244/43939 [42:19:11<28:10:55, 5.73s/it]
60%|█████▉ | 26245/43939 [42:19:17<28:09:01, 5.73s/it]
60%|█████▉ | 26246/43939 [42:19:23<28:07:26, 5.72s/it]
60%|█████▉ | 26247/43939 [42:19:28<28:01:52, 5.70s/it]
60%|█████▉ | 26248/43939 [42:19:34<28:01:43, 5.70s/it]
60%|█████▉ | 26249/43939 [42:19:40<27:57:31, 5.69s/it]
60%|█████▉ | 26250/43939 [42:19:45<28:02:40, 5.71s/it]
{'loss': 0.7773, 'grad_norm': 1.2156130075454712, 'learning_rate': 2.088193909602145e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.74, 'epoch': 2.39}
60%|█████▉ | 26250/43939 [42:19:45<28:02:40, 5.71s/it]
60%|█████▉ | 26251/43939 [42:19:51<27:59:27, 5.70s/it]
60%|█████▉ | 26252/43939 [42:19:57<27:58:24, 5.69s/it]
60%|█████▉ | 26253/43939 [42:20:02<27:56:11, 5.69s/it]
60%|█████▉ | 26254/43939 [42:20:08<28:01:09, 5.70s/it]
60%|█████▉ | 26255/43939 [42:20:14<28:03:03, 5.71s/it]
60%|█████▉ | 26256/43939 [42:20:19<28:00:07, 5.70s/it]
60%|█████▉ | 26257/43939 [42:20:25<28:03:34, 5.71s/it]
60%|█████▉ | 26258/43939 [42:20:31<28:04:44, 5.72s/it]
60%|█████▉ | 26259/43939 [42:20:37<27:58:58, 5.70s/it]
60%|█████▉ | 26260/43939 [42:20:42<27:59:14, 5.70s/it]
60%|█████▉ | 26261/43939 [42:20:48<28:00:30, 5.70s/it]
60%|█████▉ | 26262/43939 [42:20:54<28:01:07, 5.71s/it]
60%|█████▉ | 26263/43939 [42:20:59<28:01:17, 5.71s/it]
60%|█████▉ | 26264/43939 [42:21:05<27:58:05, 5.70s/it]
60%|█████▉ | 26265/43939 [42:21:11<28:14:34, 5.75s/it]
60%|█████▉ | 26266/43939 [42:21:17<28:11:36, 5.74s/it]
60%|█████▉ | 26267/43939 [42:21:22<28:04:22, 5.72s/it]
60%|█████▉ | 26268/43939 [42:21:28<28:03:02, 5.71s/it]
60%|█████▉ | 26269/43939 [42:21:34<28:01:05, 5.71s/it]
60%|█████▉ | 26270/43939 [42:21:40<28:04:49, 5.72s/it]
60%|█████▉ | 26271/43939 [42:21:45<28:02:19, 5.71s/it]
60%|█████▉ | 26272/43939 [42:21:51<27:58:41, 5.70s/it]
60%|█████▉ | 26273/43939 [42:21:57<27:57:16, 5.70s/it]
60%|█████▉ | 26274/43939 [42:22:02<27:55:09, 5.69s/it]
60%|█████▉ | 26275/43939 [42:22:08<27:52:59, 5.68s/it]
60%|█████▉ | 26276/43939 [42:22:14<27:51:57, 5.68s/it]
60%|█████▉ | 26277/43939 [42:22:19<27:53:52, 5.69s/it]
60%|█████▉ | 26278/43939 [42:22:25<28:01:13, 5.71s/it]
60%|█████▉ | 26279/43939 [42:22:31<27:59:22, 5.71s/it]
60%|█████▉ | 26280/43939 [42:22:36<28:00:23, 5.71s/it]
60%|█████▉ | 26281/43939 [42:22:42<27:59:19, 5.71s/it]
60%|█████▉ | 26282/43939 [42:22:48<27:59:35, 5.71s/it]
60%|█████▉ | 26283/43939 [42:22:54<27:56:53, 5.70s/it]
60%|█████▉ | 26284/43939 [42:22:59<27:56:08, 5.70s/it]
60%|█████▉ | 26285/43939 [42:23:05<27:58:09, 5.70s/it]
60%|█████▉ | 26286/43939 [42:23:11<28:17:53, 5.77s/it]
60%|█████▉ | 26287/43939 [42:23:17<28:09:48, 5.74s/it]
60%|█████▉ | 26288/43939 [42:23:22<28:05:48, 5.73s/it]
60%|█████▉ | 26289/43939 [42:23:28<28:00:10, 5.71s/it]
60%|█████▉ | 26290/43939 [42:23:34<28:00:25, 5.71s/it]
60%|█████▉ | 26291/43939 [42:23:39<28:00:07, 5.71s/it]
60%|█████▉ | 26292/43939 [42:23:45<27:57:03, 5.70s/it]
60%|█████▉ | 26293/43939 [42:23:51<28:00:31, 5.71s/it]
60%|█████▉ | 26294/43939 [42:23:56<27:59:38, 5.71s/it]
60%|█████▉ | 26295/43939 [42:24:02<28:00:57, 5.72s/it]
60%|█████▉ | 26296/43939 [42:24:08<28:00:39, 5.72s/it]
60%|█████▉ | 26297/43939 [42:24:14<28:00:51, 5.72s/it]
60%|█████▉ | 26298/43939 [42:24:19<27:55:22, 5.70s/it]
60%|█████▉ | 26299/43939 [42:24:25<27:57:03, 5.70s/it]
60%|█████▉ | 26300/43939 [42:24:31<27:56:42, 5.70s/it]
{'loss': 0.7978, 'grad_norm': 1.4776896238327026, 'learning_rate': 2.0784026460313808e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.34, 'epoch': 2.39}
60%|█████▉ | 26300/43939 [42:24:31<27:56:42, 5.70s/it]
60%|█████▉ | 26301/43939 [42:24:36<28:00:33, 5.72s/it]
60%|█████▉ | 26302/43939 [42:24:42<27:59:28, 5.71s/it]
60%|█████▉ | 26303/43939 [42:24:48<27:58:37, 5.71s/it]
60%|█████▉ | 26304/43939 [42:24:54<27:56:57, 5.71s/it]
60%|█████▉ | 26305/43939 [42:24:59<27:56:29, 5.70s/it]
60%|█████▉ | 26306/43939 [42:25:05<27:56:22, 5.70s/it]
60%|█████▉ | 26307/43939 [42:25:11<28:13:13, 5.76s/it]
60%|█████▉ | 26308/43939 [42:25:17<28:07:51, 5.74s/it]
60%|█████▉ | 26309/43939 [42:25:22<28:02:15, 5.73s/it]
60%|█████▉ | 26310/43939 [42:25:28<28:00:13, 5.72s/it]
60%|█████▉ | 26311/43939 [42:25:34<27:56:48, 5.71s/it]
60%|█████▉ | 26312/43939 [42:25:39<27:55:32, 5.70s/it]
60%|█████▉ | 26313/43939 [42:25:45<27:55:27, 5.70s/it]
60%|█████▉ | 26314/43939 [42:25:51<27:55:04, 5.70s/it]
60%|█████▉ | 26315/43939 [42:25:56<27:56:40, 5.71s/it]
60%|█████▉ | 26316/43939 [42:26:02<27:54:50, 5.70s/it]
60%|█████▉ | 26317/43939 [42:26:08<27:53:49, 5.70s/it]
60%|█████▉ | 26318/43939 [42:26:14<27:50:10, 5.69s/it]
60%|█████▉ | 26319/43939 [42:26:19<27:51:21, 5.69s/it]
60%|█████▉ | 26320/43939 [42:26:25<27:51:09, 5.69s/it]
60%|█████▉ | 26321/43939 [42:26:31<27:52:07, 5.69s/it]
60%|█████▉ | 26322/43939 [42:26:36<27:55:07, 5.71s/it]
60%|█████▉ | 26323/43939 [42:26:42<27:56:09, 5.71s/it]
60%|█████▉ | 26324/43939 [42:26:48<27:55:09, 5.71s/it]
60%|█████▉ | 26325/43939 [42:26:53<27:57:34, 5.71s/it]
60%|█████▉ | 26326/43939 [42:26:59<27:54:01, 5.70s/it]
60%|█████▉ | 26327/43939 [42:27:05<27:54:02, 5.70s/it]
60%|█████▉ | 26328/43939 [42:27:11<27:52:09, 5.70s/it]
60%|█████▉ | 26329/43939 [42:27:16<28:13:00, 5.77s/it]
60%|█████▉ | 26330/43939 [42:27:22<28:10:07, 5.76s/it]
60%|█████▉ | 26331/43939 [42:27:28<28:11:18, 5.76s/it]
60%|█████▉ | 26332/43939 [42:27:34<28:07:59, 5.75s/it]
60%|█████▉ | 26333/43939 [42:27:39<28:04:11, 5.74s/it]
60%|█████▉ | 26334/43939 [42:27:45<28:00:35, 5.73s/it]
60%|█████▉ | 26335/43939 [42:27:51<27:57:01, 5.72s/it]
60%|█████▉ | 26336/43939 [42:27:57<27:58:41, 5.72s/it]
60%|█████▉ | 26337/43939 [42:28:02<27:56:16, 5.71s/it]
60%|█████▉ | 26338/43939 [42:28:08<27:51:57, 5.70s/it]
60%|█████▉ | 26339/43939 [42:28:14<27:54:42, 5.71s/it]
60%|█████▉ | 26340/43939 [42:28:19<27:57:17, 5.72s/it]
60%|█████▉ | 26341/43939 [42:28:25<27:54:40, 5.71s/it]
60%|█████▉ | 26342/43939 [42:28:31<27:53:59, 5.71s/it]
60%|█████▉ | 26343/43939 [42:28:37<27:58:11, 5.72s/it]
60%|█████▉ | 26344/43939 [42:28:42<27:52:52, 5.70s/it]
60%|█████▉ | 26345/43939 [42:28:48<27:54:54, 5.71s/it]
60%|█████▉ | 26346/43939 [42:28:54<27:53:42, 5.71s/it]
60%|█████▉ | 26347/43939 [42:28:59<27:50:47, 5.70s/it]
60%|█████▉ | 26348/43939 [42:29:05<27:51:30, 5.70s/it]
60%|█████▉ | 26349/43939 [42:29:11<27:51:11, 5.70s/it]
60%|█████▉ | 26350/43939 [42:29:17<28:12:32, 5.77s/it]
{'loss': 0.7725, 'grad_norm': 1.4902770519256592, 'learning_rate': 2.0686180341496146e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 448.73, 'epoch': 2.4}
60%|█████▉ | 26350/43939 [42:29:17<28:12:32, 5.77s/it]
60%|█████▉ | 26351/43939 [42:29:22<28:06:05, 5.75s/it]
60%|█████▉ | 26352/43939 [42:29:28<28:00:59, 5.73s/it]
60%|█████▉ | 26353/43939 [42:29:34<28:01:14, 5.74s/it]
60%|█████▉ | 26354/43939 [42:29:40<27:59:05, 5.73s/it]
60%|█████▉ | 26355/43939 [42:29:45<27:56:39, 5.72s/it]
60%|█████▉ | 26356/43939 [42:29:51<27:55:57, 5.72s/it]
60%|█████▉ | 26357/43939 [42:29:57<27:53:20, 5.71s/it]
60%|█████▉ | 26358/43939 [42:30:02<27:52:33, 5.71s/it]
60%|█████▉ | 26359/43939 [42:30:08<27:53:08, 5.71s/it]
60%|█████▉ | 26360/43939 [42:30:14<27:54:50, 5.72s/it]
60%|█████▉ | 26361/43939 [42:30:20<27:57:47, 5.73s/it]
60%|█████▉ | 26362/43939 [42:30:25<27:57:56, 5.73s/it]
60%|█████▉ | 26363/43939 [42:30:31<27:56:38, 5.72s/it]
60%|██████ | 26364/43939 [42:30:37<27:57:54, 5.73s/it][2025-11-30 12:34:50,513] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.50s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:45, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.70s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.72s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:41, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9867295622825623, 'eval_runtime': 195.2766, 'eval_samples_per_second': 5.116, 'eval_steps_per_second': 0.369, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 2.4}
60%|██████ | 26364/43939 [42:33:52<27:57:54, 5.73s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
60%|██████ | 26365/43939 [42:33:58<314:10:35, 64.36s/it]
60%|██████ | 26366/43939 [42:34:04<228:14:34, 46.76s/it]
60%|██████ | 26367/43939 [42:34:09<168:06:39, 34.44s/it]
60%|██████ | 26368/43939 [42:34:15<126:02:07, 25.82s/it]
60%|██████ | 26369/43939 [42:34:21<96:32:47, 19.78s/it]
60%|██████ | 26370/43939 [42:34:27<76:19:41, 15.64s/it]
60%|██████ | 26371/43939 [42:34:32<61:48:23, 12.67s/it]
60%|██████ | 26372/43939 [42:34:38<51:35:08, 10.57s/it]
60%|██████ | 26373/43939 [42:34:44<44:28:17, 9.11s/it]
60%|██████ | 26374/43939 [42:34:49<39:29:32, 8.09s/it]
60%|██████ | 26375/43939 [42:34:55<36:00:37, 7.38s/it]
60%|██████ | 26376/43939 [42:35:01<33:34:01, 6.88s/it]
60%|██████ | 26377/43939 [42:35:07<31:54:46, 6.54s/it]
60%|██████ | 26378/43939 [42:35:12<30:45:05, 6.30s/it]
60%|██████ | 26379/43939 [42:35:18<29:48:44, 6.11s/it]
60%|██████ | 26380/43939 [42:35:24<29:15:06, 6.00s/it]
60%|██████ | 26381/43939 [42:35:29<28:48:41, 5.91s/it]
60%|██████ | 26382/43939 [42:35:35<28:27:54, 5.84s/it]
60%|██████ | 26383/43939 [42:35:41<28:12:23, 5.78s/it]
60%|██████ | 26384/43939 [42:35:46<28:02:14, 5.75s/it]
60%|██████ | 26385/43939 [42:35:52<27:55:49, 5.73s/it]
60%|██████ | 26386/43939 [42:35:58<27:49:53, 5.71s/it]
60%|██████ | 26387/43939 [42:36:04<27:49:04, 5.71s/it]
60%|██████ | 26388/43939 [42:36:09<27:52:48, 5.72s/it]
60%|██████ | 26389/43939 [42:36:15<27:49:30, 5.71s/it]
60%|██████ | 26390/43939 [42:36:21<27:49:12, 5.71s/it]
60%|██████ | 26391/43939 [42:36:27<28:10:56, 5.78s/it]
60%|██████ | 26392/43939 [42:36:32<28:02:59, 5.75s/it]
60%|██████ | 26393/43939 [42:36:38<27:55:04, 5.73s/it]
60%|██████ | 26394/43939 [42:36:44<27:53:13, 5.72s/it]
60%|██████ | 26395/43939 [42:36:49<27:54:21, 5.73s/it]
60%|██████ | 26396/43939 [42:36:55<27:52:32, 5.72s/it]
60%|██████ | 26397/43939 [42:37:01<27:52:10, 5.72s/it]
60%|██████ | 26398/43939 [42:37:07<27:48:03, 5.71s/it]
60%|██████ | 26399/43939 [42:37:12<27:45:49, 5.70s/it]
60%|██████ | 26400/43939 [42:37:18<27:46:50, 5.70s/it]
{'loss': 0.7874, 'grad_norm': 1.4182987213134766, 'learning_rate': 2.0588402283320937e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 356.84, 'epoch': 2.4}
60%|██████ | 26400/43939 [42:37:18<27:46:50, 5.70s/it]
60%|██████ | 26401/43939 [42:37:24<27:44:45, 5.70s/it]
60%|██████ | 26402/43939 [42:37:29<27:41:37, 5.68s/it]
60%|██████ | 26403/43939 [42:37:35<27:42:56, 5.69s/it]
60%|██████ | 26404/43939 [42:37:41<27:46:50, 5.70s/it]
60%|██████ | 26405/43939 [42:37:46<27:50:36, 5.72s/it]
60%|██████ | 26406/43939 [42:37:52<27:49:14, 5.71s/it]
60%|██████ | 26407/43939 [42:37:58<27:49:28, 5.71s/it]
60%|██████ | 26408/43939 [42:38:04<27:51:17, 5.72s/it]
60%|██████ | 26409/43939 [42:38:09<27:48:38, 5.71s/it]
60%|██████ | 26410/43939 [42:38:15<27:50:04, 5.72s/it]
60%|██████ | 26411/43939 [42:38:21<27:50:44, 5.72s/it]
60%|██████ | 26412/43939 [42:38:27<28:07:30, 5.78s/it]
60%|██████ | 26413/43939 [42:38:32<27:59:03, 5.75s/it]
60%|██████ | 26414/43939 [42:38:38<27:56:45, 5.74s/it]
60%|██████ | 26415/43939 [42:38:44<27:52:49, 5.73s/it]
60%|██████ | 26416/43939 [42:38:49<27:52:47, 5.73s/it]
60%|██████ | 26417/43939 [42:38:55<27:54:15, 5.73s/it]
60%|██████ | 26418/43939 [42:39:01<27:54:32, 5.73s/it]
60%|██████ | 26419/43939 [42:39:07<27:51:34, 5.72s/it]
60%|██████ | 26420/43939 [42:39:12<27:50:04, 5.72s/it]
60%|██████ | 26421/43939 [42:39:18<27:48:07, 5.71s/it]
60%|██████ | 26422/43939 [42:39:24<27:51:38, 5.73s/it]
60%|██████ | 26423/43939 [42:39:30<27:55:39, 5.74s/it]
60%|██████ | 26424/43939 [42:39:35<27:50:28, 5.72s/it]
60%|██████ | 26425/43939 [42:39:41<27:48:38, 5.72s/it]
60%|██████ | 26426/43939 [42:39:47<27:48:01, 5.71s/it]
60%|██████ | 26427/43939 [42:39:52<27:48:03, 5.72s/it]
60%|██████ | 26428/43939 [42:39:58<27:46:53, 5.71s/it]
60%|██████ | 26429/43939 [42:40:04<27:47:10, 5.71s/it]
60%|██████ | 26430/43939 [42:40:10<27:46:23, 5.71s/it]
60%|██████ | 26431/43939 [42:40:15<27:45:33, 5.71s/it]
60%|██████ | 26432/43939 [42:40:21<27:41:49, 5.70s/it]
60%|██████ | 26433/43939 [42:40:27<28:02:47, 5.77s/it]
60%|██████ | 26434/43939 [42:40:33<27:55:28, 5.74s/it]
60%|██████ | 26435/43939 [42:40:38<27:53:43, 5.74s/it]
60%|██████ | 26436/43939 [42:40:44<27:52:49, 5.73s/it]
60%|██████ | 26437/43939 [42:40:50<27:50:27, 5.73s/it]
60%|██████ | 26438/43939 [42:40:55<27:55:35, 5.74s/it]
60%|██████ | 26439/43939 [42:41:01<27:48:53, 5.72s/it]
60%|██████ | 26440/43939 [42:41:07<27:48:09, 5.72s/it]
60%|██████ | 26441/43939 [42:41:13<27:45:19, 5.71s/it]
60%|██████ | 26442/43939 [42:41:18<27:41:09, 5.70s/it]
60%|██████ | 26443/43939 [42:41:24<27:40:49, 5.70s/it]
60%|██████ | 26444/43939 [42:41:30<27:45:12, 5.71s/it]
60%|██████ | 26445/43939 [42:41:35<27:45:25, 5.71s/it]
60%|██████ | 26446/43939 [42:41:41<27:45:09, 5.71s/it]
60%|██████ | 26447/43939 [42:41:47<27:42:59, 5.70s/it]
60%|██████ | 26448/43939 [42:41:52<27:42:08, 5.70s/it]
60%|██████ | 26449/43939 [42:41:58<27:41:29, 5.70s/it]
60%|██████ | 26450/43939 [42:42:04<27:40:56, 5.70s/it]
{'loss': 0.7775, 'grad_norm': 1.3437907695770264, 'learning_rate': 2.0490693828466823e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.85, 'epoch': 2.41}
60%|██████ | 26450/43939 [42:42:04<27:40:56, 5.70s/it]
60%|██████ | 26451/43939 [42:42:10<27:41:05, 5.70s/it]
60%|██████ | 26452/43939 [42:42:15<27:42:13, 5.70s/it]
60%|██████ | 26453/43939 [42:42:21<27:40:51, 5.70s/it]
60%|██████ | 26454/43939 [42:42:27<27:40:15, 5.70s/it]
60%|██████ | 26455/43939 [42:42:33<28:04:02, 5.78s/it]
60%|██████ | 26456/43939 [42:42:38<27:56:46, 5.75s/it]
60%|██████ | 26457/43939 [42:42:44<27:56:12, 5.75s/it]
60%|██████ | 26458/43939 [42:42:50<27:49:29, 5.73s/it]
60%|██████ | 26459/43939 [42:42:55<27:48:15, 5.73s/it]
60%|██████ | 26460/43939 [42:43:01<27:48:26, 5.73s/it]
60%|██████ | 26461/43939 [42:43:07<27:45:54, 5.72s/it]
60%|██████ | 26462/43939 [42:43:13<27:41:33, 5.70s/it]
60%|██████ | 26463/43939 [42:43:18<27:42:35, 5.71s/it]
60%|██████ | 26464/43939 [42:43:24<27:40:59, 5.70s/it]
60%|██████ | 26465/43939 [42:43:30<27:38:29, 5.69s/it]
60%|██████ | 26466/43939 [42:43:35<27:39:08, 5.70s/it]
60%|██████ | 26467/43939 [42:43:41<27:38:48, 5.70s/it]
60%|██████ | 26468/43939 [42:43:47<27:38:29, 5.70s/it]
60%|██████ | 26469/43939 [42:43:52<27:36:47, 5.69s/it]
60%|██████ | 26470/43939 [42:43:58<27:37:52, 5.69s/it]
60%|██████ | 26471/43939 [42:44:04<27:38:14, 5.70s/it]
60%|██████ | 26472/43939 [42:44:10<27:44:19, 5.72s/it]
60%|██████ | 26473/43939 [42:44:15<27:39:34, 5.70s/it]
60%|██████ | 26474/43939 [42:44:21<27:39:37, 5.70s/it]
60%|██████ | 26475/43939 [42:44:27<27:40:17, 5.70s/it]
60%|██████ | 26476/43939 [42:44:32<27:40:35, 5.71s/it]
60%|██████ | 26477/43939 [42:44:38<27:58:47, 5.77s/it]
60%|██████ | 26478/43939 [42:44:44<27:50:25, 5.74s/it]
60%|██████ | 26479/43939 [42:44:50<27:47:35, 5.73s/it]
60%|██████ | 26480/43939 [42:44:55<27:42:38, 5.71s/it]
60%|██████ | 26481/43939 [42:45:01<27:41:31, 5.71s/it]
60%|██████ | 26482/43939 [42:45:07<27:42:21, 5.71s/it]
60%|██████ | 26483/43939 [42:45:12<27:41:26, 5.71s/it]
60%|██████ | 26484/43939 [42:45:18<27:39:39, 5.70s/it]
60%|██████ | 26485/43939 [42:45:24<27:40:08, 5.71s/it]
60%|██████ | 26486/43939 [42:45:30<27:40:08, 5.71s/it]
60%|██████ | 26487/43939 [42:45:35<27:43:19, 5.72s/it]
60%|██████ | 26488/43939 [42:45:41<27:47:06, 5.73s/it]
60%|██████ | 26489/43939 [42:45:47<27:44:01, 5.72s/it]
60%|██████ | 26490/43939 [42:45:52<27:42:11, 5.72s/it]
60%|██████ | 26491/43939 [42:45:58<27:45:03, 5.73s/it]
60%|██████ | 26492/43939 [42:46:04<27:42:49, 5.72s/it]
60%|██████ | 26493/43939 [42:46:10<27:41:21, 5.71s/it]
60%|██████ | 26494/43939 [42:46:15<27:37:22, 5.70s/it]
60%|██████ | 26495/43939 [42:46:21<27:36:51, 5.70s/it]
60%|██████ | 26496/43939 [42:46:27<27:38:38, 5.71s/it]
60%|██████ | 26497/43939 [42:46:32<27:34:39, 5.69s/it]
60%|██████ | 26498/43939 [42:46:38<27:52:33, 5.75s/it]
60%|██████ | 26499/43939 [42:46:44<27:47:39, 5.74s/it]
60%|██████ | 26500/43939 [42:46:50<27:45:01, 5.73s/it]
{'loss': 0.7888, 'grad_norm': 1.3449835777282715, 'learning_rate': 2.0393056518514288e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.83, 'epoch': 2.41}
60%|██████ | 26500/43939 [42:46:50<27:45:01, 5.73s/it]
60%|██████ | 26501/43939 [42:46:55<27:43:08, 5.72s/it]
60%|██████ | 26502/43939 [42:47:01<27:39:04, 5.71s/it]
60%|██████ | 26503/43939 [42:47:07<27:37:36, 5.70s/it]
60%|██████ | 26504/43939 [42:47:12<27:37:14, 5.70s/it]
60%|██████ | 26505/43939 [42:47:18<27:40:04, 5.71s/it]
60%|██████ | 26506/43939 [42:47:24<27:36:10, 5.70s/it]
60%|██████ | 26507/43939 [42:47:30<27:34:50, 5.70s/it]
60%|██████ | 26508/43939 [42:47:35<27:35:17, 5.70s/it]
60%|██████ | 26509/43939 [42:47:41<27:34:35, 5.70s/it]
60%|██████ | 26510/43939 [42:47:47<27:38:26, 5.71s/it]
60%|██████ | 26511/43939 [42:47:52<27:34:28, 5.70s/it]
60%|██████ | 26512/43939 [42:47:58<27:33:53, 5.69s/it]
60%|██████ | 26513/43939 [42:48:04<27:36:16, 5.70s/it]
60%|██████ | 26514/43939 [42:48:09<27:34:59, 5.70s/it]
60%|██████ | 26515/43939 [42:48:15<27:34:20, 5.70s/it]
60%|██████ | 26516/43939 [42:48:21<27:34:45, 5.70s/it]
60%|██████ | 26517/43939 [42:48:27<27:31:37, 5.69s/it]
60%|██████ | 26518/43939 [42:48:32<27:56:26, 5.77s/it]
60%|██████ | 26519/43939 [42:48:38<27:51:31, 5.76s/it]
60%|██████ | 26520/43939 [42:48:44<27:46:57, 5.74s/it]
60%|██████ | 26521/43939 [42:48:50<27:40:57, 5.72s/it]
60%|██████ | 26522/43939 [42:48:55<27:38:50, 5.71s/it]
60%|██████ | 26523/43939 [42:49:01<27:44:04, 5.73s/it]
60%|██████ | 26524/43939 [42:49:07<27:38:24, 5.71s/it]
60%|██████ | 26525/43939 [42:49:12<27:35:38, 5.70s/it]
60%|██████ | 26526/43939 [42:49:18<27:35:37, 5.70s/it]
60%|██████ | 26527/43939 [42:49:24<27:35:24, 5.70s/it]
60%|██████ | 26528/43939 [42:49:30<27:35:37, 5.71s/it]
60%|██████ | 26529/43939 [42:49:35<27:35:02, 5.70s/it]
60%|██████ | 26530/43939 [42:49:41<27:37:10, 5.71s/it]
60%|██████ | 26531/43939 [42:49:47<27:37:15, 5.71s/it]
60%|██████ | 26532/43939 [42:49:52<27:36:58, 5.71s/it]
60%|██████ | 26533/43939 [42:49:58<27:35:25, 5.71s/it]
60%|██████ | 26534/43939 [42:50:04<27:32:29, 5.70s/it]
60%|██████ | 26535/43939 [42:50:09<27:32:11, 5.70s/it]
60%|██████ | 26536/43939 [42:50:15<27:31:20, 5.69s/it]
60%|██████ | 26537/43939 [42:50:21<27:31:24, 5.69s/it]
60%|██████ | 26538/43939 [42:50:26<27:30:28, 5.69s/it]
60%|██████ | 26539/43939 [42:50:32<27:33:04, 5.70s/it]
60%|██████ | 26540/43939 [42:50:38<27:53:52, 5.77s/it]
60%|██████ | 26541/43939 [42:50:44<27:46:39, 5.75s/it]
60%|██████ | 26542/43939 [42:50:50<27:44:25, 5.74s/it]
60%|██████ | 26543/43939 [42:50:55<27:40:35, 5.73s/it]
60%|██████ | 26544/43939 [42:51:01<27:39:23, 5.72s/it]
60%|██████ | 26545/43939 [42:51:07<27:38:53, 5.72s/it]
60%|██████ | 26546/43939 [42:51:12<27:34:29, 5.71s/it]
60%|██████ | 26547/43939 [42:51:18<27:31:59, 5.70s/it]
60%|██████ | 26548/43939 [42:51:24<27:33:37, 5.71s/it]
60%|██████ | 26549/43939 [42:51:29<27:32:30, 5.70s/it]
60%|██████ | 26550/43939 [42:51:35<27:33:02, 5.70s/it]
{'loss': 0.7625, 'grad_norm': 1.2085769176483154, 'learning_rate': 2.0295491893921345e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.2, 'epoch': 2.42}
60%|██████ | 26550/43939 [42:51:35<27:33:02, 5.70s/it]
60%|██████ | 26551/43939 [42:51:41<27:33:13, 5.70s/it]
60%|██████ | 26552/43939 [42:51:47<27:31:45, 5.70s/it]
60%|██████ | 26553/43939 [42:51:52<27:31:16, 5.70s/it]
60%|██████ | 26554/43939 [42:51:58<27:30:59, 5.70s/it]
60%|██████ | 26555/43939 [42:52:04<27:28:26, 5.69s/it]
60%|██████ | 26556/43939 [42:52:09<27:32:58, 5.71s/it]
60%|██████ | 26557/43939 [42:52:15<27:34:44, 5.71s/it]
60%|██████ | 26558/43939 [42:52:21<27:32:03, 5.70s/it]
60%|██████ | 26559/43939 [42:52:27<27:33:16, 5.71s/it]
60%|██████ | 26560/43939 [42:52:32<27:33:00, 5.71s/it]
60%|██████ | 26561/43939 [42:52:38<28:01:26, 5.81s/it]
60%|██████ | 26562/43939 [42:52:44<27:49:01, 5.76s/it]
60%|██████ | 26563/43939 [42:52:50<27:47:56, 5.76s/it]
60%|██████ | 26564/43939 [42:52:55<27:43:29, 5.74s/it]
60%|██████ | 26565/43939 [42:53:01<27:39:42, 5.73s/it]
60%|██████ | 26566/43939 [42:53:07<27:35:57, 5.72s/it]
60%|██████ | 26567/43939 [42:53:13<27:37:18, 5.72s/it]
60%|██████ | 26568/43939 [42:53:18<27:30:50, 5.70s/it]
60%|██████ | 26569/43939 [42:53:24<27:31:26, 5.70s/it]
60%|██████ | 26570/43939 [42:53:30<27:29:34, 5.70s/it]
60%|██████ | 26571/43939 [42:53:35<27:31:03, 5.70s/it]
60%|██████ | 26572/43939 [42:53:41<27:31:02, 5.70s/it]
60%|██████ | 26573/43939 [42:53:47<27:32:32, 5.71s/it]
60%|██████ | 26574/43939 [42:53:52<27:29:06, 5.70s/it]
60%|██████ | 26575/43939 [42:53:58<27:27:45, 5.69s/it]
60%|██████ | 26576/43939 [42:54:04<27:27:21, 5.69s/it]
60%|██████ | 26577/43939 [42:54:09<27:24:13, 5.68s/it]
60%|██████ | 26578/43939 [42:54:15<27:20:59, 5.67s/it]
60%|██████ | 26579/43939 [42:54:21<27:19:59, 5.67s/it]
60%|██████ | 26580/43939 [42:54:26<27:19:55, 5.67s/it]
60%|██████ | 26581/43939 [42:54:32<27:23:04, 5.68s/it]
60%|██████ | 26582/43939 [42:54:38<27:47:49, 5.77s/it]
60%|██████ | 26583/43939 [42:54:44<27:42:31, 5.75s/it]
61%|██████ | 26584/43939 [42:54:49<27:40:07, 5.74s/it]
61%|██████ | 26585/43939 [42:54:55<27:36:17, 5.73s/it]
61%|██████ | 26586/43939 [42:55:01<27:36:00, 5.73s/it]
61%|██████ | 26587/43939 [42:55:07<27:36:08, 5.73s/it]
61%|██████ | 26588/43939 [42:55:12<27:32:09, 5.71s/it]
61%|██████ | 26589/43939 [42:55:18<27:27:50, 5.70s/it]
61%|██████ | 26590/43939 [42:55:24<27:33:15, 5.72s/it]
61%|██████ | 26591/43939 [42:55:29<27:31:33, 5.71s/it]
61%|██████ | 26592/43939 [42:55:35<27:28:41, 5.70s/it]
61%|██████ | 26593/43939 [42:55:41<27:29:51, 5.71s/it]
61%|██████ | 26594/43939 [42:55:47<27:34:25, 5.72s/it]
61%|██████ | 26595/43939 [42:55:52<27:31:59, 5.71s/it]
61%|██████ | 26596/43939 [42:55:58<27:28:27, 5.70s/it]
61%|██████ | 26597/43939 [42:56:04<27:29:30, 5.71s/it]
61%|██████ | 26598/43939 [42:56:09<27:31:57, 5.72s/it]
61%|██████ | 26599/43939 [42:56:15<27:28:25, 5.70s/it]
61%|██████ | 26600/43939 [42:56:21<27:26:51, 5.70s/it]
{'loss': 0.7863, 'grad_norm': 1.3855311870574951, 'learning_rate': 2.0198001493999215e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.99, 'epoch': 2.42}
61%|██████ | 26600/43939 [42:56:21<27:26:51, 5.70s/it]
61%|██████ | 26601/43939 [42:56:26<27:27:02, 5.70s/it]
61%|██████ | 26602/43939 [42:56:32<27:27:48, 5.70s/it]
61%|██████ | 26603/43939 [42:56:38<27:26:42, 5.70s/it]
61%|██████ | 26604/43939 [42:56:44<27:44:14, 5.76s/it]
61%|██████ | 26605/43939 [42:56:49<27:37:49, 5.74s/it]
61%|██████ | 26606/43939 [42:56:55<27:34:33, 5.73s/it]
61%|██████ | 26607/43939 [42:57:01<27:31:29, 5.72s/it]
61%|██████ | 26608/43939 [42:57:07<27:33:23, 5.72s/it]
61%|██████ | 26609/43939 [42:57:12<27:29:06, 5.71s/it]
61%|██████ | 26610/43939 [42:57:18<27:27:59, 5.71s/it]
61%|██████ | 26611/43939 [42:57:24<27:25:33, 5.70s/it]
61%|██████ | 26612/43939 [42:57:29<27:29:32, 5.71s/it]
61%|██████ | 26613/43939 [42:57:35<27:27:41, 5.71s/it]
61%|██████ | 26614/43939 [42:57:41<27:27:31, 5.71s/it]
61%|██████ | 26615/43939 [42:57:46<27:25:22, 5.70s/it]
61%|██████ | 26616/43939 [42:57:52<27:25:29, 5.70s/it]
61%|██████ | 26617/43939 [42:57:58<27:28:47, 5.71s/it]
61%|██████ | 26618/43939 [42:58:04<27:27:45, 5.71s/it]
61%|██████ | 26619/43939 [42:58:09<27:26:41, 5.70s/it]
61%|██████ | 26620/43939 [42:58:15<27:26:31, 5.70s/it]
61%|██████ | 26621/43939 [42:58:21<27:28:10, 5.71s/it]
61%|██████ | 26622/43939 [42:58:26<27:32:34, 5.73s/it]
61%|██████ | 26623/43939 [42:58:32<27:26:03, 5.70s/it]
61%|██████ | 26624/43939 [42:58:38<27:27:38, 5.71s/it]
61%|██████ | 26625/43939 [42:58:44<27:45:31, 5.77s/it]
61%|██████ | 26626/43939 [42:58:49<27:39:43, 5.75s/it]
61%|██████ | 26627/43939 [42:58:55<27:36:45, 5.74s/it]
61%|██████ | 26628/43939 [42:59:01<27:35:10, 5.74s/it]
61%|██████ | 26629/43939 [42:59:07<27:33:00, 5.73s/it]
61%|██████ | 26630/43939 [42:59:12<27:27:54, 5.71s/it]
61%|██████ | 26631/43939 [42:59:18<27:24:31, 5.70s/it]
61%|██████ | 26632/43939 [42:59:24<27:26:05, 5.71s/it]
61%|██████ | 26633/43939 [42:59:29<27:25:53, 5.71s/it]
61%|██████ | 26634/43939 [42:59:35<27:25:48, 5.71s/it]
61%|██████ | 26635/43939 [42:59:41<27:27:44, 5.71s/it]
61%|██████ | 26636/43939 [42:59:47<27:26:17, 5.71s/it]
61%|██████ | 26637/43939 [42:59:52<27:28:47, 5.72s/it]
61%|██████ | 26638/43939 [42:59:58<27:32:02, 5.73s/it]
61%|██████ | 26639/43939 [43:00:04<27:27:48, 5.71s/it]
61%|██████ | 26640/43939 [43:00:09<27:24:41, 5.70s/it]
61%|██████ | 26641/43939 [43:00:15<27:21:56, 5.70s/it]
61%|██████ | 26642/43939 [43:00:21<27:26:06, 5.71s/it]
61%|██████ | 26643/43939 [43:00:27<27:25:16, 5.71s/it]
61%|██████ | 26644/43939 [43:00:32<27:25:22, 5.71s/it]
61%|██████ | 26645/43939 [43:00:38<27:27:59, 5.72s/it]
61%|██████ | 26646/43939 [43:00:44<27:50:47, 5.80s/it]
61%|██████ | 26647/43939 [43:00:50<27:42:34, 5.77s/it]
61%|██████ | 26648/43939 [43:00:55<27:35:33, 5.74s/it]
61%|██████ | 26649/43939 [43:01:01<27:35:46, 5.75s/it]
61%|██████ | 26650/43939 [43:01:07<27:33:49, 5.74s/it]
{'loss': 0.779, 'grad_norm': 1.5885320901870728, 'learning_rate': 2.0100586856888054e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.83, 'epoch': 2.43}
61%|██████ | 26650/43939 [43:01:07<27:33:49, 5.74s/it]
61%|██████ | 26651/43939 [43:01:13<27:29:40, 5.73s/it]
61%|██████ | 26652/43939 [43:01:18<27:27:48, 5.72s/it]
61%|██████ | 26653/43939 [43:01:24<27:29:05, 5.72s/it]
61%|██████ | 26654/43939 [43:01:30<27:28:27, 5.72s/it]
61%|██████ | 26655/43939 [43:01:35<27:29:47, 5.73s/it]
61%|██████ | 26656/43939 [43:01:41<27:29:34, 5.73s/it]
61%|██████ | 26657/43939 [43:01:47<27:33:38, 5.74s/it]
61%|██████ | 26658/43939 [43:01:53<27:32:50, 5.74s/it]
61%|██████ | 26659/43939 [43:01:58<27:28:52, 5.73s/it]
61%|██████ | 26660/43939 [43:02:04<27:23:24, 5.71s/it]
61%|██████ | 26661/43939 [43:02:10<27:22:52, 5.71s/it]
61%|██████ | 26662/43939 [43:02:15<27:22:11, 5.70s/it]
61%|██████ | 26663/43939 [43:02:21<27:21:10, 5.70s/it]
61%|██████ | 26664/43939 [43:02:27<27:22:59, 5.71s/it]
61%|██████ | 26665/43939 [43:02:33<27:24:03, 5.71s/it]
61%|██████ | 26666/43939 [43:02:38<27:24:15, 5.71s/it]
61%|██████ | 26667/43939 [43:02:44<27:21:22, 5.70s/it]
61%|██████ | 26668/43939 [43:02:50<27:43:03, 5.78s/it]
61%|██████ | 26669/43939 [43:02:56<27:37:53, 5.76s/it]
61%|██████ | 26670/43939 [43:03:01<27:29:51, 5.73s/it]
61%|██████ | 26671/43939 [43:03:07<27:26:30, 5.72s/it]
61%|██████ | 26672/43939 [43:03:13<27:32:38, 5.74s/it]
61%|██████ | 26673/43939 [43:03:18<27:31:05, 5.74s/it]
61%|██████ | 26674/43939 [43:03:24<27:25:01, 5.72s/it]
61%|██████ | 26675/43939 [43:03:30<27:23:08, 5.71s/it]
61%|██████ | 26676/43939 [43:03:36<27:21:43, 5.71s/it]
61%|██████ | 26677/43939 [43:03:41<27:19:46, 5.70s/it]
61%|██████ | 26678/43939 [43:03:47<27:20:09, 5.70s/it]
61%|██████ | 26679/43939 [43:03:53<27:22:14, 5.71s/it]
61%|██████ | 26680/43939 [43:03:58<27:25:38, 5.72s/it]
61%|██████ | 26681/43939 [43:04:04<27:25:58, 5.72s/it]
61%|██████ | 26682/43939 [43:04:10<27:23:52, 5.72s/it]
61%|██████ | 26683/43939 [43:04:16<27:20:18, 5.70s/it]
61%|██████ | 26684/43939 [43:04:21<27:24:28, 5.72s/it]
61%|██████ | 26685/43939 [43:04:27<27:22:22, 5.71s/it]
61%|██████ | 26686/43939 [43:04:33<27:24:17, 5.72s/it]
61%|██████ | 26687/43939 [43:04:38<27:23:45, 5.72s/it]
61%|██████ | 26688/43939 [43:04:44<27:20:07, 5.70s/it]
61%|██████ | 26689/43939 [43:04:50<27:43:26, 5.79s/it]
61%|██████ | 26690/43939 [43:04:56<27:38:20, 5.77s/it]
61%|██████ | 26691/43939 [43:05:01<27:32:06, 5.75s/it]
61%|██████ | 26692/43939 [43:05:07<27:32:07, 5.75s/it]
61%|██████ | 26693/43939 [43:05:13<27:29:29, 5.74s/it]
61%|██████ | 26694/43939 [43:05:19<27:27:34, 5.73s/it]
61%|██████ | 26695/43939 [43:05:24<27:26:13, 5.73s/it]
61%|██████ | 26696/43939 [43:05:30<27:26:04, 5.73s/it]
61%|██████ | 26697/43939 [43:05:36<27:22:04, 5.71s/it]
61%|██████ | 26698/43939 [43:05:42<27:21:37, 5.71s/it]
61%|██████ | 26699/43939 [43:05:47<27:20:23, 5.71s/it]
61%|██████ | 26700/43939 [43:05:53<27:20:39, 5.71s/it]
{'loss': 0.7898, 'grad_norm': 1.4351317882537842, 'learning_rate': 2.0003249519532696e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.22, 'epoch': 2.43}
61%|██████ | 26700/43939 [43:05:53<27:20:39, 5.71s/it]
61%|██████ | 26701/43939 [43:05:59<27:18:37, 5.70s/it]
61%|██████ | 26702/43939 [43:06:04<27:18:23, 5.70s/it]
61%|██████ | 26703/43939 [43:06:10<27:16:06, 5.70s/it]
61%|██████ | 26704/43939 [43:06:16<27:21:16, 5.71s/it]
61%|██████ | 26705/43939 [43:06:22<27:26:41, 5.73s/it]
61%|██████ | 26706/43939 [43:06:27<27:25:42, 5.73s/it]
61%|██████ | 26707/43939 [43:06:33<27:22:07, 5.72s/it]
61%|██████ | 26708/43939 [43:06:39<27:21:28, 5.72s/it]
61%|██████ | 26709/43939 [43:06:44<27:18:07, 5.70s/it]
61%|██████ | 26710/43939 [43:06:50<27:17:36, 5.70s/it]
61%|██████ | 26711/43939 [43:06:56<27:38:31, 5.78s/it]
61%|██████ | 26712/43939 [43:07:02<27:31:36, 5.75s/it]
61%|██████ | 26713/43939 [43:07:07<27:26:26, 5.73s/it]
61%|██████ | 26714/43939 [43:07:13<27:26:01, 5.73s/it]
61%|██████ | 26715/43939 [43:07:19<27:25:54, 5.73s/it]
61%|██████ | 26716/43939 [43:07:25<27:27:42, 5.74s/it]
61%|██████ | 26717/43939 [43:07:30<27:25:17, 5.73s/it]
61%|██████ | 26718/43939 [43:07:36<27:22:30, 5.72s/it]
61%|██████ | 26719/43939 [43:07:42<27:21:21, 5.72s/it]
61%|██████ | 26720/43939 [43:07:47<27:17:11, 5.70s/it]
61%|██████ | 26721/43939 [43:07:53<27:14:34, 5.70s/it]
61%|██████ | 26722/43939 [43:07:59<27:13:57, 5.69s/it]
61%|██████ | 26723/43939 [43:08:04<27:15:57, 5.70s/it]
61%|██████ | 26724/43939 [43:08:10<27:12:59, 5.69s/it]
61%|██████ | 26725/43939 [43:08:16<27:15:13, 5.70s/it]
61%|██████ | 26726/43939 [43:08:22<27:14:14, 5.70s/it]
61%|██████ | 26727/43939 [43:08:27<27:15:49, 5.70s/it]
61%|██████ | 26728/43939 [43:08:33<27:16:18, 5.70s/it]
61%|██████ | 26729/43939 [43:08:39<27:15:56, 5.70s/it]
61%|██████ | 26730/43939 [43:08:44<27:19:35, 5.72s/it]
61%|██████ | 26731/43939 [43:08:50<27:18:31, 5.71s/it]
61%|██████ | 26732/43939 [43:08:56<27:34:22, 5.77s/it]
61%|██████ | 26733/43939 [43:09:02<27:27:11, 5.74s/it]
61%|██████ | 26734/43939 [43:09:07<27:18:20, 5.71s/it]
61%|██████ | 26735/43939 [43:09:13<27:21:17, 5.72s/it]
61%|██████ | 26736/43939 [43:09:19<27:19:30, 5.72s/it]
61%|██████ | 26737/43939 [43:09:25<27:20:48, 5.72s/it]
61%|██████ | 26738/43939 [43:09:30<27:17:34, 5.71s/it]
61%|██████ | 26739/43939 [43:09:36<27:15:39, 5.71s/it]
61%|██████ | 26740/43939 [43:09:42<27:14:21, 5.70s/it]
61%|██████ | 26741/43939 [43:09:47<27:19:27, 5.72s/it]
61%|██████ | 26742/43939 [43:09:53<27:17:57, 5.71s/it]
61%|██████ | 26743/43939 [43:09:59<27:17:30, 5.71s/it]
61%|██████ | 26744/43939 [43:10:04<27:13:24, 5.70s/it]
61%|██████ | 26745/43939 [43:10:10<27:14:56, 5.71s/it]
61%|██████ | 26746/43939 [43:10:16<27:15:47, 5.71s/it]
61%|██████ | 26747/43939 [43:10:22<27:10:09, 5.69s/it]
61%|██████ | 26748/43939 [43:10:27<27:13:18, 5.70s/it]
61%|██████ | 26749/43939 [43:10:33<27:09:39, 5.69s/it]
61%|██████ | 26750/43939 [43:10:39<27:10:42, 5.69s/it]
{'loss': 0.7752, 'grad_norm': 1.4788719415664673, 'learning_rate': 1.9905991017658364e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.65, 'epoch': 2.44}
61%|██████ | 26750/43939 [43:10:39<27:10:42, 5.69s/it]
61%|██████ | 26751/43939 [43:10:44<27:11:56, 5.70s/it]
61%|██████ | 26752/43939 [43:10:50<27:13:08, 5.70s/it]
61%|██████ | 26753/43939 [43:10:56<27:34:02, 5.77s/it]
61%|██████ | 26754/43939 [43:11:02<27:30:09, 5.76s/it]
61%|██████ | 26755/43939 [43:11:07<27:24:37, 5.74s/it]
61%|██████ | 26756/43939 [43:11:13<27:20:19, 5.73s/it]
61%|██████ | 26757/43939 [43:11:19<27:18:23, 5.72s/it]
61%|██████ | 26758/43939 [43:11:25<27:17:35, 5.72s/it]
61%|██████ | 26759/43939 [43:11:30<27:18:32, 5.72s/it]
61%|██████ | 26760/43939 [43:11:36<27:14:50, 5.71s/it]
61%|██████ | 26761/43939 [43:11:42<27:13:54, 5.71s/it]
61%|██████ | 26762/43939 [43:11:47<27:14:28, 5.71s/it]
61%|██████ | 26763/43939 [43:11:53<27:13:28, 5.71s/it]
61%|██████ | 26764/43939 [43:11:59<27:10:45, 5.70s/it]
61%|██████ | 26765/43939 [43:12:04<27:13:07, 5.71s/it]
61%|██████ | 26766/43939 [43:12:10<27:11:54, 5.70s/it]
61%|██████ | 26767/43939 [43:12:16<27:11:12, 5.70s/it]
61%|██████ | 26768/43939 [43:12:22<27:14:01, 5.71s/it]
61%|██████ | 26769/43939 [43:12:27<27:13:40, 5.71s/it]
61%|██████ | 26770/43939 [43:12:33<27:14:31, 5.71s/it]
61%|██████ | 26771/43939 [43:12:39<27:15:49, 5.72s/it]
61%|██████ | 26772/43939 [43:12:44<27:16:52, 5.72s/it]
61%|██████ | 26773/43939 [43:12:50<27:15:40, 5.72s/it]
61%|██████ | 26774/43939 [43:12:56<27:38:10, 5.80s/it]
61%|██████ | 26775/43939 [43:13:02<27:27:17, 5.76s/it]
61%|██████ | 26776/43939 [43:13:07<27:20:18, 5.73s/it]
61%|██████ | 26777/43939 [43:13:13<27:12:57, 5.71s/it]
61%|██████ | 26778/43939 [43:13:19<27:10:52, 5.70s/it]
61%|██████ | 26779/43939 [43:13:25<27:11:22, 5.70s/it]
61%|██████ | 26780/43939 [43:13:30<27:13:35, 5.71s/it]
61%|██████ | 26781/43939 [43:13:36<27:12:23, 5.71s/it]
61%|██████ | 26782/43939 [43:13:42<27:15:54, 5.72s/it]
61%|██████ | 26783/43939 [43:13:47<27:13:11, 5.71s/it]
61%|██████ | 26784/43939 [43:13:53<27:10:20, 5.70s/it]
61%|██████ | 26785/43939 [43:13:59<27:10:48, 5.70s/it]
61%|██████ | 26786/43939 [43:14:04<27:07:51, 5.69s/it]
61%|██████ | 26787/43939 [43:14:10<27:07:25, 5.69s/it]
61%|██████ | 26788/43939 [43:14:16<27:07:54, 5.69s/it]
61%|██████ | 26789/43939 [43:14:22<27:06:11, 5.69s/it]
61%|██████ | 26790/43939 [43:14:27<27:06:45, 5.69s/it]
61%|██████ | 26791/43939 [43:14:33<27:09:43, 5.70s/it]
61%|██████ | 26792/43939 [43:14:39<27:11:13, 5.71s/it]
61%|██████ | 26793/43939 [43:14:44<27:13:53, 5.72s/it]
61%|██████ | 26794/43939 [43:14:50<27:15:07, 5.72s/it]
61%|██████ | 26795/43939 [43:14:56<27:38:06, 5.80s/it]
61%|██████ | 26796/43939 [43:15:02<27:32:20, 5.78s/it]
61%|██████ | 26797/43939 [43:15:08<27:24:06, 5.75s/it]
61%|██████ | 26798/43939 [43:15:13<27:18:58, 5.74s/it]
61%|██████ | 26799/43939 [43:15:19<27:15:02, 5.72s/it]
61%|██████ | 26800/43939 [43:15:25<27:14:15, 5.72s/it]
{'loss': 0.777, 'grad_norm': 1.8284074068069458, 'learning_rate': 1.9808812885746492e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.51, 'epoch': 2.44}
61%|██████ | 26800/43939 [43:15:25<27:14:15, 5.72s/it]
61%|██████ | 26801/43939 [43:15:30<27:09:01, 5.70s/it]
61%|██████ | 26802/43939 [43:15:36<27:08:36, 5.70s/it]
61%|██████ | 26803/43939 [43:15:42<27:08:58, 5.70s/it]
61%|██████ | 26804/43939 [43:15:47<27:07:30, 5.70s/it]
61%|██████ | 26805/43939 [43:15:53<27:06:53, 5.70s/it]
61%|██████ | 26806/43939 [43:15:59<27:11:40, 5.71s/it]
61%|██████ | 26807/43939 [43:16:05<27:09:57, 5.71s/it]
61%|██████ | 26808/43939 [43:16:10<27:08:44, 5.70s/it]
61%|██████ | 26809/43939 [43:16:16<27:04:07, 5.69s/it]
61%|██████ | 26810/43939 [43:16:22<27:02:01, 5.68s/it]
61%|██████ | 26811/43939 [43:16:27<27:00:40, 5.68s/it]
61%|██████ | 26812/43939 [43:16:33<27:01:44, 5.68s/it]
61%|██████ | 26813/43939 [43:16:39<27:05:49, 5.70s/it]
61%|██████ | 26814/43939 [43:16:44<27:04:32, 5.69s/it]
61%|██████ | 26815/43939 [43:16:50<27:04:49, 5.69s/it]
61%|██████ | 26816/43939 [43:16:56<27:24:54, 5.76s/it]
61%|██████ | 26817/43939 [43:17:02<27:22:19, 5.76s/it]
61%|██████ | 26818/43939 [43:17:07<27:17:12, 5.74s/it]
61%|██████ | 26819/43939 [43:17:13<27:11:29, 5.72s/it]
61%|██████ | 26820/43939 [43:17:19<27:12:41, 5.72s/it]
61%|██████ | 26821/43939 [43:17:25<27:13:33, 5.73s/it]
61%|██████ | 26822/43939 [43:17:30<27:09:18, 5.71s/it]
61%|██████ | 26823/43939 [43:17:36<27:08:19, 5.71s/it]
61%|██████ | 26824/43939 [43:17:42<27:05:00, 5.70s/it]
61%|██████ | 26825/43939 [43:17:47<27:03:10, 5.69s/it]
61%|██████ | 26826/43939 [43:17:53<27:02:54, 5.69s/it]
61%|██████ | 26827/43939 [43:17:59<27:04:36, 5.70s/it]
61%|██████ | 26828/43939 [43:18:04<27:03:39, 5.69s/it]
61%|██████ | 26829/43939 [43:18:10<27:06:01, 5.70s/it]
61%|██████ | 26830/43939 [43:18:16<27:07:57, 5.71s/it]
61%|██████ | 26831/43939 [43:18:22<27:10:07, 5.72s/it]
61%|██████ | 26832/43939 [43:18:27<27:11:49, 5.72s/it]
61%|██████ | 26833/43939 [43:18:33<27:07:29, 5.71s/it]
61%|██████ | 26834/43939 [43:18:39<27:08:55, 5.71s/it]
61%|██████ | 26835/43939 [43:18:44<27:09:32, 5.72s/it]
61%|██████ | 26836/43939 [43:18:50<27:06:05, 5.70s/it]
61%|██████ | 26837/43939 [43:18:56<27:05:17, 5.70s/it]
61%|██████ | 26838/43939 [43:19:02<27:23:00, 5.76s/it]
61%|██████ | 26839/43939 [43:19:07<27:16:26, 5.74s/it]
61%|██████ | 26840/43939 [43:19:13<27:11:09, 5.72s/it]
61%|██████ | 26841/43939 [43:19:19<27:15:01, 5.74s/it]
61%|██████ | 26842/43939 [43:19:25<27:10:44, 5.72s/it]
61%|██████ | 26843/43939 [43:19:30<27:09:34, 5.72s/it]
61%|██████ | 26844/43939 [43:19:36<27:08:13, 5.71s/it]
61%|██████ | 26845/43939 [43:19:42<27:06:48, 5.71s/it]
61%|██████ | 26846/43939 [43:19:47<27:04:24, 5.70s/it]
61%|██████ | 26847/43939 [43:19:53<27:04:06, 5.70s/it]
61%|██████ | 26848/43939 [43:19:59<27:05:37, 5.71s/it]
61%|██████ | 26849/43939 [43:20:04<27:00:38, 5.69s/it]
61%|██████ | 26850/43939 [43:20:10<27:00:40, 5.69s/it]
{'loss': 0.7875, 'grad_norm': 1.3145912885665894, 'learning_rate': 1.9711716657010464e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.91, 'epoch': 2.44}
61%|██████ | 26850/43939 [43:20:10<27:00:40, 5.69s/it]
61%|██████ | 26851/43939 [43:20:16<27:00:28, 5.69s/it]
61%|██████ | 26852/43939 [43:20:21<27:00:12, 5.69s/it]
61%|██████ | 26853/43939 [43:20:27<27:00:36, 5.69s/it]
61%|██████ | 26854/43939 [43:20:33<27:00:08, 5.69s/it]
61%|██████ | 26855/43939 [43:20:39<26:59:55, 5.69s/it]
61%|██████ | 26856/43939 [43:20:44<27:01:45, 5.70s/it]
61%|██████ | 26857/43939 [43:20:50<27:03:50, 5.70s/it]
61%|██████ | 26858/43939 [43:20:56<27:03:10, 5.70s/it]
61%|██████ | 26859/43939 [43:21:02<27:24:27, 5.78s/it]
61%|██████ | 26860/43939 [43:21:07<27:19:36, 5.76s/it]
61%|██████ | 26861/43939 [43:21:13<27:13:36, 5.74s/it]
61%|██████ | 26862/43939 [43:21:19<27:10:03, 5.73s/it]
61%|██████ | 26863/43939 [43:21:24<27:06:14, 5.71s/it]
61%|██████ | 26864/43939 [43:21:30<27:07:10, 5.72s/it]
61%|██████ | 26865/43939 [43:21:36<27:07:29, 5.72s/it]
61%|██████ | 26866/43939 [43:21:42<27:07:26, 5.72s/it]
61%|██████ | 26867/43939 [43:21:47<27:06:48, 5.72s/it]
61%|██████ | 26868/43939 [43:21:53<27:05:01, 5.71s/it]
61%|██████ | 26869/43939 [43:21:59<27:04:47, 5.71s/it]
61%|██████ | 26870/43939 [43:22:04<27:04:47, 5.71s/it]
61%|██████ | 26871/43939 [43:22:10<27:02:11, 5.70s/it]
61%|██████ | 26872/43939 [43:22:16<26:59:59, 5.70s/it]
61%|██████ | 26873/43939 [43:22:21<27:00:26, 5.70s/it]
61%|██████ | 26874/43939 [43:22:27<27:00:53, 5.70s/it]
61%|██████ | 26875/43939 [43:22:33<26:59:06, 5.69s/it]
61%|██████ | 26876/43939 [43:22:39<27:00:33, 5.70s/it]
61%|██████ | 26877/43939 [43:22:44<27:01:46, 5.70s/it]
61%|██████ | 26878/43939 [43:22:50<27:00:15, 5.70s/it]
61%|██████ | 26879/43939 [43:22:56<26:59:18, 5.70s/it]
61%|██████ | 26880/43939 [43:23:02<27:16:49, 5.76s/it]
61%|██████ | 26881/43939 [43:23:07<27:09:49, 5.73s/it]
61%|██████ | 26882/43939 [43:23:13<27:06:20, 5.72s/it]
61%|██████ | 26883/43939 [43:23:19<27:05:10, 5.72s/it]
61%|██████ | 26884/43939 [43:23:24<27:04:36, 5.72s/it]
61%|██████ | 26885/43939 [43:23:30<27:01:59, 5.71s/it]
61%|██████ | 26886/43939 [43:23:36<27:04:28, 5.72s/it]
61%|██████ | 26887/43939 [43:23:42<27:07:23, 5.73s/it]
61%|██████ | 26888/43939 [43:23:47<27:04:51, 5.72s/it]
61%|██████ | 26889/43939 [43:23:53<27:04:09, 5.72s/it]
61%|██████ | 26890/43939 [43:23:59<27:04:21, 5.72s/it]
61%|██████ | 26891/43939 [43:24:04<27:01:01, 5.71s/it]
61%|██████ | 26892/43939 [43:24:10<27:01:11, 5.71s/it]
61%|██████ | 26893/43939 [43:24:16<26:59:52, 5.70s/it]
61%|██████ | 26894/43939 [43:24:21<26:58:48, 5.70s/it]
61%|██████ | 26895/43939 [43:24:27<26:59:44, 5.70s/it]
61%|██████ | 26896/43939 [43:24:33<26:57:49, 5.70s/it]
61%|██████ | 26897/43939 [43:24:39<27:00:46, 5.71s/it]
61%|██████ | 26898/43939 [43:24:44<27:02:49, 5.71s/it]
61%|██████ | 26899/43939 [43:24:50<27:02:42, 5.71s/it]
61%|██████ | 26900/43939 [43:24:56<27:02:08, 5.71s/it]
{'loss': 0.7768, 'grad_norm': 1.4024999141693115, 'learning_rate': 1.9614703863371473e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.1, 'epoch': 2.45}
61%|██████ | 26900/43939 [43:24:56<27:02:08, 5.71s/it]
61%|██████ | 26901/43939 [43:25:02<27:17:25, 5.77s/it]
61%|██████ | 26902/43939 [43:25:07<27:13:08, 5.75s/it]
61%|██████ | 26903/43939 [43:25:13<27:15:02, 5.76s/it]
61%|██████ | 26904/43939 [43:25:19<27:09:12, 5.74s/it]
61%|██████ | 26905/43939 [43:25:24<27:08:53, 5.74s/it]
61%|██████ | 26906/43939 [43:25:30<27:05:11, 5.72s/it]
61%|██████ | 26907/43939 [43:25:36<27:06:44, 5.73s/it]
61%|██████ | 26908/43939 [43:25:42<27:05:02, 5.72s/it]
61%|██████ | 26909/43939 [43:25:47<27:01:59, 5.71s/it]
61%|██████ | 26910/43939 [43:25:53<27:00:54, 5.71s/it]
61%|██████ | 26911/43939 [43:25:59<26:59:46, 5.71s/it]
61%|██████ | 26912/43939 [43:26:04<26:59:15, 5.71s/it]
61%|██████▏ | 26913/43939 [43:26:10<26:53:39, 5.69s/it]
61%|██████▏ | 26914/43939 [43:26:16<26:52:26, 5.68s/it]
61%|██████▏ | 26915/43939 [43:26:21<26:53:07, 5.69s/it]
61%|██████▏ | 26916/43939 [43:26:27<26:52:54, 5.68s/it]
61%|██████▏ | 26917/43939 [43:26:33<26:53:51, 5.69s/it]
61%|██████▏ | 26918/43939 [43:26:38<26:52:33, 5.68s/it]
61%|██████▏ | 26919/43939 [43:26:44<26:55:54, 5.70s/it]
61%|██████▏ | 26920/43939 [43:26:50<26:59:33, 5.71s/it]
61%|██████▏ | 26921/43939 [43:26:56<26:59:06, 5.71s/it]
61%|██████▏ | 26922/43939 [43:27:01<26:59:10, 5.71s/it]
61%|██████▏ | 26923/43939 [43:27:07<27:21:20, 5.79s/it]
61%|██████▏ | 26924/43939 [43:27:13<27:10:46, 5.75s/it]
61%|██████▏ | 26925/43939 [43:27:19<27:06:29, 5.74s/it]
61%|██████▏ | 26926/43939 [43:27:24<27:00:27, 5.71s/it]
61%|██████▏ | 26927/43939 [43:27:30<27:05:22, 5.73s/it]
61%|██████▏ | 26928/43939 [43:27:36<27:03:31, 5.73s/it]
61%|██████▏ | 26929/43939 [43:27:42<27:02:19, 5.72s/it]
61%|██████▏ | 26930/43939 [43:27:47<27:02:55, 5.72s/it]
61%|██████▏ | 26931/43939 [43:27:53<27:02:10, 5.72s/it]
61%|██████▏ | 26932/43939 [43:27:59<27:02:22, 5.72s/it]
61%|██████▏ | 26933/43939 [43:28:04<26:59:59, 5.72s/it]
61%|██████▏ | 26934/43939 [43:28:10<26:55:25, 5.70s/it]
61%|██████▏ | 26935/43939 [43:28:16<26:55:52, 5.70s/it]
61%|██████▏ | 26936/43939 [43:28:22<26:55:04, 5.70s/it]
61%|██████▏ | 26937/43939 [43:28:27<26:58:11, 5.71s/it]
61%|██████▏ | 26938/43939 [43:28:33<26:54:39, 5.70s/it]
61%|██████▏ | 26939/43939 [43:28:39<26:52:47, 5.69s/it]
61%|██████▏ | 26940/43939 [43:28:44<26:57:31, 5.71s/it]
61%|██████▏ | 26941/43939 [43:28:50<26:56:15, 5.71s/it]
61%|██████▏ | 26942/43939 [43:28:56<26:55:58, 5.70s/it]
61%|██████▏ | 26943/43939 [43:29:01<26:58:00, 5.71s/it]
61%|██████▏ | 26944/43939 [43:29:07<26:58:58, 5.72s/it]
61%|██████▏ | 26945/43939 [43:29:13<27:17:05, 5.78s/it]
61%|██████▏ | 26946/43939 [43:29:19<27:13:01, 5.77s/it]
61%|██████▏ | 26947/43939 [43:29:25<27:05:28, 5.74s/it]
61%|██████▏ | 26948/43939 [43:29:30<27:07:36, 5.75s/it]
61%|██████▏ | 26949/43939 [43:29:36<27:00:47, 5.72s/it]
61%|██████▏ | 26950/43939 [43:29:42<27:00:51, 5.72s/it]
{'loss': 0.7665, 'grad_norm': 1.2669835090637207, 'learning_rate': 1.9517776035434298e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.45, 'epoch': 2.45}
61%|██████▏ | 26950/43939 [43:29:42<27:00:51, 5.72s/it]
61%|██████▏ | 26951/43939 [43:29:47<26:57:13, 5.71s/it]
61%|██████▏ | 26952/43939 [43:29:53<26:54:44, 5.70s/it]
61%|██████▏ | 26953/43939 [43:29:59<26:53:29, 5.70s/it]
61%|██████▏ | 26954/43939 [43:30:04<26:54:36, 5.70s/it]
61%|██████▏ | 26955/43939 [43:30:10<26:54:26, 5.70s/it]
61%|██████▏ | 26956/43939 [43:30:16<26:53:28, 5.70s/it]
61%|██████▏ | 26957/43939 [43:30:22<26:51:14, 5.69s/it]
61%|██████▏ | 26958/43939 [43:30:27<26:49:43, 5.69s/it]
61%|██████▏ | 26959/43939 [43:30:33<26:46:11, 5.68s/it]
61%|██████▏ | 26960/43939 [43:30:39<26:47:50, 5.68s/it]
61%|██████▏ | 26961/43939 [43:30:44<26:49:10, 5.69s/it]
61%|██████▏ | 26962/43939 [43:30:50<26:49:03, 5.69s/it]
61%|██████▏ | 26963/43939 [43:30:56<26:51:23, 5.70s/it]
61%|██████▏ | 26964/43939 [43:31:01<26:55:40, 5.71s/it]
61%|██████▏ | 26965/43939 [43:31:07<26:50:49, 5.69s/it]
61%|██████▏ | 26966/43939 [43:31:13<27:07:50, 5.75s/it]
61%|██████▏ | 26967/43939 [43:31:19<27:05:38, 5.75s/it]
61%|██████▏ | 26968/43939 [43:31:24<27:01:34, 5.73s/it]
61%|██████▏ | 26969/43939 [43:31:30<26:59:15, 5.73s/it]
61%|██████▏ | 26970/43939 [43:31:36<26:54:08, 5.71s/it]
61%|██████▏ | 26971/43939 [43:31:41<26:54:25, 5.71s/it]
61%|██████▏ | 26972/43939 [43:31:47<26:53:06, 5.70s/it]
61%|██████▏ | 26973/43939 [43:31:53<26:53:47, 5.71s/it]
61%|██████▏ | 26974/43939 [43:31:59<26:54:32, 5.71s/it]
61%|██████▏ | 26975/43939 [43:32:04<26:56:32, 5.72s/it]
61%|██████▏ | 26976/43939 [43:32:10<26:55:06, 5.71s/it]
61%|██████▏ | 26977/43939 [43:32:16<26:54:49, 5.71s/it]
61%|██████▏ | 26978/43939 [43:32:21<26:55:11, 5.71s/it]
61%|██████▏ | 26979/43939 [43:32:27<26:53:39, 5.71s/it]
61%|██████▏ | 26980/43939 [43:32:33<26:52:17, 5.70s/it]
61%|██████▏ | 26981/43939 [43:32:39<26:49:26, 5.69s/it]
61%|██████▏ | 26982/43939 [43:32:44<26:48:08, 5.69s/it]
61%|██████▏ | 26983/43939 [43:32:50<26:47:10, 5.69s/it]
61%|██████▏ | 26984/43939 [43:32:56<26:46:54, 5.69s/it]
61%|██████▏ | 26985/43939 [43:33:01<26:50:09, 5.70s/it]
61%|██████▏ | 26986/43939 [43:33:07<26:54:20, 5.71s/it]
61%|██████▏ | 26987/43939 [43:33:13<27:11:52, 5.78s/it]
61%|██████▏ | 26988/43939 [43:33:19<27:03:22, 5.75s/it]
61%|██████▏ | 26989/43939 [43:33:24<27:00:59, 5.74s/it]
61%|██████▏ | 26990/43939 [43:33:30<26:57:27, 5.73s/it]
61%|██████▏ | 26991/43939 [43:33:36<26:55:50, 5.72s/it]
61%|██████▏ | 26992/43939 [43:33:41<26:53:22, 5.71s/it]
61%|██████▏ | 26993/43939 [43:33:47<26:56:44, 5.72s/it]
61%|██████▏ | 26994/43939 [43:33:53<27:00:05, 5.74s/it]
61%|██████▏ | 26995/43939 [43:33:59<26:58:05, 5.73s/it]
61%|██████▏ | 26996/43939 [43:34:04<26:55:45, 5.72s/it]
61%|██████▏ | 26997/43939 [43:34:10<26:54:28, 5.72s/it]
61%|██████▏ | 26998/43939 [43:34:16<26:55:20, 5.72s/it]
61%|██████▏ | 26999/43939 [43:34:22<26:57:08, 5.73s/it]
61%|██████▏ | 27000/43939 [43:34:27<26:57:03, 5.73s/it]
{'loss': 0.747, 'grad_norm': 1.4060825109481812, 'learning_rate': 1.9420934702463223e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.99, 'epoch': 2.46}
61%|██████▏ | 27000/43939 [43:34:27<26:57:03, 5.73s/it]
61%|██████▏ | 27001/43939 [43:34:33<26:56:58, 5.73s/it]
61%|██████▏ | 27002/43939 [43:34:39<26:55:30, 5.72s/it]
61%|██████▏ | 27003/43939 [43:34:44<26:55:06, 5.72s/it]
61%|██████▏ | 27004/43939 [43:34:50<26:53:24, 5.72s/it]
61%|██████▏ | 27005/43939 [43:34:56<26:52:32, 5.71s/it]
61%|██████▏ | 27006/43939 [43:35:02<26:51:58, 5.71s/it]
61%|██████▏ | 27007/43939 [43:35:07<26:51:25, 5.71s/it]
61%|██████▏ | 27008/43939 [43:35:13<27:11:13, 5.78s/it]
61%|██████▏ | 27009/43939 [43:35:19<27:06:24, 5.76s/it]
61%|██████▏ | 27010/43939 [43:35:25<27:00:20, 5.74s/it]
61%|██████▏ | 27011/43939 [43:35:30<26:58:08, 5.74s/it]
61%|██████▏ | 27012/43939 [43:35:36<26:57:01, 5.73s/it]
61%|██████▏ | 27013/43939 [43:35:42<26:54:36, 5.72s/it]
61%|██████▏ | 27014/43939 [43:35:47<26:51:37, 5.71s/it]
61%|██████▏ | 27015/43939 [43:35:53<26:46:51, 5.70s/it]
61%|██████▏ | 27016/43939 [43:35:59<26:48:28, 5.70s/it]
61%|██████▏ | 27017/43939 [43:36:05<26:45:44, 5.69s/it]
61%|██████▏ | 27018/43939 [43:36:10<26:46:03, 5.69s/it]
61%|██████▏ | 27019/43939 [43:36:16<26:47:32, 5.70s/it]
61%|██████▏ | 27020/43939 [43:36:22<26:48:15, 5.70s/it]
61%|██████▏ | 27021/43939 [43:36:27<26:46:09, 5.70s/it]
61%|██████▏ | 27022/43939 [43:36:33<26:47:39, 5.70s/it]
62%|██████▏ | 27023/43939 [43:36:39<26:47:04, 5.70s/it]
62%|██████▏ | 27024/43939 [43:36:44<26:51:39, 5.72s/it]
62%|██████▏ | 27025/43939 [43:36:50<26:50:08, 5.71s/it]
62%|██████▏ | 27026/43939 [43:36:56<26:46:40, 5.70s/it]
62%|██████▏ | 27027/43939 [43:37:02<26:47:10, 5.70s/it]
62%|██████▏ | 27028/43939 [43:37:07<26:45:23, 5.70s/it]
62%|██████▏ | 27029/43939 [43:37:13<27:11:39, 5.79s/it]
62%|██████▏ | 27030/43939 [43:37:19<27:07:55, 5.78s/it]
62%|██████▏ | 27031/43939 [43:37:25<27:00:06, 5.75s/it]
62%|██████▏ | 27032/43939 [43:37:30<26:57:22, 5.74s/it]
62%|██████▏ | 27033/43939 [43:37:36<26:51:09, 5.72s/it]
62%|██████▏ | 27034/43939 [43:37:42<26:50:45, 5.72s/it]
62%|██████▏ | 27035/43939 [43:37:48<26:49:30, 5.71s/it]
62%|██████▏ | 27036/43939 [43:37:53<26:47:28, 5.71s/it]
62%|██████▏ | 27037/43939 [43:37:59<26:46:48, 5.70s/it]
62%|██████▏ | 27038/43939 [43:38:05<26:45:48, 5.70s/it]
62%|██████▏ | 27039/43939 [43:38:10<26:43:24, 5.69s/it]
62%|██████▏ | 27040/43939 [43:38:16<26:42:42, 5.69s/it]
62%|██████▏ | 27041/43939 [43:38:22<26:42:18, 5.69s/it]
62%|██████▏ | 27042/43939 [43:38:27<26:47:29, 5.71s/it]
62%|██████▏ | 27043/43939 [43:38:33<26:47:36, 5.71s/it]
62%|██████▏ | 27044/43939 [43:38:39<26:49:47, 5.72s/it]
62%|██████▏ | 27045/43939 [43:38:45<26:50:05, 5.72s/it]
62%|██████▏ | 27046/43939 [43:38:50<26:48:51, 5.71s/it]
62%|██████▏ | 27047/43939 [43:38:56<26:51:36, 5.72s/it]
62%|██████▏ | 27048/43939 [43:39:02<26:48:31, 5.71s/it]
62%|██████▏ | 27049/43939 [43:39:07<26:46:14, 5.71s/it]
62%|██████▏ | 27050/43939 [43:39:13<27:01:54, 5.76s/it]
{'loss': 0.7568, 'grad_norm': 1.4920679330825806, 'learning_rate': 1.9324181392357837e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.28, 'epoch': 2.46}
62%|██████▏ | 27050/43939 [43:39:13<27:01:54, 5.76s/it]
62%|██████▏ | 27051/43939 [43:39:19<26:58:26, 5.75s/it]
62%|██████▏ | 27052/43939 [43:39:25<26:54:02, 5.73s/it]
62%|██████▏ | 27053/43939 [43:39:30<26:50:53, 5.72s/it]
62%|██████▏ | 27054/43939 [43:39:36<26:46:30, 5.71s/it]
62%|██████▏ | 27055/43939 [43:39:42<26:46:14, 5.71s/it]
62%|██████▏ | 27056/43939 [43:39:47<26:44:48, 5.70s/it]
62%|██████▏ | 27057/43939 [43:39:53<26:43:43, 5.70s/it]
62%|██████▏ | 27058/43939 [43:39:59<26:40:23, 5.69s/it]
62%|██████▏ | 27059/43939 [43:40:04<26:37:55, 5.68s/it]
62%|██████▏ | 27060/43939 [43:40:10<26:36:56, 5.68s/it]
62%|██████▏ | 27061/43939 [43:40:16<26:37:38, 5.68s/it]
62%|██████▏ | 27062/43939 [43:40:22<26:38:35, 5.68s/it]
62%|██████▏ | 27063/43939 [43:40:27<26:38:55, 5.68s/it]
62%|██████▏ | 27064/43939 [43:40:33<26:37:33, 5.68s/it]
62%|██████▏ | 27065/43939 [43:40:39<26:38:53, 5.69s/it]
62%|██████▏ | 27066/43939 [43:40:44<26:37:51, 5.68s/it]
62%|██████▏ | 27067/43939 [43:40:50<26:38:49, 5.69s/it]
62%|██████▏ | 27068/43939 [43:40:56<26:44:27, 5.71s/it]
62%|██████▏ | 27069/43939 [43:41:01<26:40:28, 5.69s/it]
62%|██████▏ | 27070/43939 [43:41:07<26:41:22, 5.70s/it]
62%|██████▏ | 27071/43939 [43:41:13<26:42:17, 5.70s/it]
62%|██████▏ | 27072/43939 [43:41:19<27:01:30, 5.77s/it]
62%|██████▏ | 27073/43939 [43:41:24<26:57:04, 5.75s/it]
62%|██████▏ | 27074/43939 [43:41:30<26:49:42, 5.73s/it]
62%|██████▏ | 27075/43939 [43:41:36<26:43:48, 5.71s/it]
62%|██████▏ | 27076/43939 [43:41:41<26:41:36, 5.70s/it]
62%|██████▏ | 27077/43939 [43:41:47<26:43:00, 5.70s/it]
62%|██████▏ | 27078/43939 [43:41:53<26:40:20, 5.69s/it]
62%|██████▏ | 27079/43939 [43:41:59<26:41:54, 5.70s/it]
62%|██████▏ | 27080/43939 [43:42:04<26:40:21, 5.70s/it]
62%|██████▏ | 27081/43939 [43:42:10<26:38:09, 5.69s/it]
62%|██████▏ | 27082/43939 [43:42:16<26:39:23, 5.69s/it]
62%|██████▏ | 27083/43939 [43:42:21<26:39:23, 5.69s/it]
62%|██████▏ | 27084/43939 [43:42:27<26:42:35, 5.70s/it]
62%|██████▏ | 27085/43939 [43:42:33<26:41:26, 5.70s/it]
62%|██████▏ | 27086/43939 [43:42:38<26:41:56, 5.70s/it]
62%|██████▏ | 27087/43939 [43:42:44<26:39:32, 5.69s/it]
62%|██████▏ | 27088/43939 [43:42:50<26:39:44, 5.70s/it]
62%|██████▏ | 27089/43939 [43:42:55<26:37:10, 5.69s/it]
62%|██████▏ | 27090/43939 [43:43:01<26:42:54, 5.71s/it]
62%|██████▏ | 27091/43939 [43:43:07<26:39:17, 5.70s/it]
62%|██████▏ | 27092/43939 [43:43:13<26:40:19, 5.70s/it]
62%|██████▏ | 27093/43939 [43:43:18<26:57:37, 5.76s/it]
62%|██████▏ | 27094/43939 [43:43:24<26:51:31, 5.74s/it]
62%|██████▏ | 27095/43939 [43:43:30<26:50:57, 5.74s/it]
62%|██████▏ | 27096/43939 [43:43:36<26:49:56, 5.74s/it]
62%|██████▏ | 27097/43939 [43:43:41<26:48:21, 5.73s/it]
62%|██████▏ | 27098/43939 [43:43:47<26:48:20, 5.73s/it]
62%|██████▏ | 27099/43939 [43:43:53<26:42:49, 5.71s/it]
62%|██████▏ | 27100/43939 [43:43:58<26:40:25, 5.70s/it]
{'loss': 0.7657, 'grad_norm': 1.3667105436325073, 'learning_rate': 1.922751763162898e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.79, 'epoch': 2.47}
62%|██████▏ | 27100/43939 [43:43:58<26:40:25, 5.70s/it]
62%|██████▏ | 27101/43939 [43:44:04<26:38:55, 5.70s/it]
62%|██████▏ | 27102/43939 [43:44:10<26:39:31, 5.70s/it]
62%|██████▏ | 27103/43939 [43:44:16<26:41:12, 5.71s/it]
62%|██████▏ | 27104/43939 [43:44:21<26:38:49, 5.70s/it]
62%|██████▏ | 27105/43939 [43:44:27<26:37:35, 5.69s/it]
62%|██████▏ | 27106/43939 [43:44:33<26:38:25, 5.70s/it]
62%|██████▏ | 27107/43939 [43:44:38<26:38:13, 5.70s/it]
62%|██████▏ | 27108/43939 [43:44:44<26:40:08, 5.70s/it]
62%|██████▏ | 27109/43939 [43:44:50<26:36:43, 5.69s/it]
62%|██████▏ | 27110/43939 [43:44:55<26:35:00, 5.69s/it]
62%|██████▏ | 27111/43939 [43:45:01<26:35:40, 5.69s/it]
62%|██████▏ | 27112/43939 [43:45:07<26:35:47, 5.69s/it]
62%|██████▏ | 27113/43939 [43:45:12<26:37:15, 5.70s/it]
62%|██████▏ | 27114/43939 [43:45:18<27:00:14, 5.78s/it]
62%|██████▏ | 27115/43939 [43:45:24<26:53:21, 5.75s/it]
62%|██████▏ | 27116/43939 [43:45:30<26:51:07, 5.75s/it]
62%|██████▏ | 27117/43939 [43:45:36<26:45:00, 5.72s/it]
62%|██████▏ | 27118/43939 [43:45:41<26:41:06, 5.71s/it]
62%|██████▏ | 27119/43939 [43:45:47<26:42:07, 5.72s/it]
62%|██████▏ | 27120/43939 [43:45:53<26:39:30, 5.71s/it]
62%|██████▏ | 27121/43939 [43:45:58<26:39:32, 5.71s/it]
62%|██████▏ | 27122/43939 [43:46:04<26:43:49, 5.72s/it]
62%|██████▏ | 27123/43939 [43:46:10<26:42:05, 5.72s/it]
62%|██████▏ | 27124/43939 [43:46:16<26:41:09, 5.71s/it]
62%|██████▏ | 27125/43939 [43:46:21<26:40:24, 5.71s/it]
62%|██████▏ | 27126/43939 [43:46:27<26:41:30, 5.72s/it]
62%|██████▏ | 27127/43939 [43:46:33<26:39:38, 5.71s/it]
62%|██████▏ | 27128/43939 [43:46:38<26:38:55, 5.71s/it]
62%|██████▏ | 27129/43939 [43:46:44<26:39:41, 5.71s/it]
62%|██████▏ | 27130/43939 [43:46:50<26:39:04, 5.71s/it]
62%|██████▏ | 27131/43939 [43:46:55<26:36:32, 5.70s/it]
62%|██████▏ | 27132/43939 [43:47:01<26:37:33, 5.70s/it]
62%|██████▏ | 27133/43939 [43:47:07<26:38:51, 5.71s/it]
62%|██████▏ | 27134/43939 [43:47:13<26:37:22, 5.70s/it]
62%|██████▏ | 27135/43939 [43:47:18<26:39:34, 5.71s/it]
62%|██████▏ | 27136/43939 [43:47:24<26:57:39, 5.78s/it]
62%|██████▏ | 27137/43939 [43:47:30<26:51:45, 5.76s/it]
62%|██████▏ | 27138/43939 [43:47:36<26:48:21, 5.74s/it]
62%|██████▏ | 27139/43939 [43:47:41<26:43:26, 5.73s/it]
62%|██████▏ | 27140/43939 [43:47:47<26:41:46, 5.72s/it]
62%|██████▏ | 27141/43939 [43:47:53<26:38:00, 5.71s/it]
62%|██████▏ | 27142/43939 [43:47:58<26:36:18, 5.70s/it]
62%|██████▏ | 27143/43939 [43:48:04<26:35:10, 5.70s/it]
62%|██████▏ | 27144/43939 [43:48:10<26:35:02, 5.70s/it]
62%|██████▏ | 27145/43939 [43:48:15<26:32:37, 5.69s/it]
62%|██████▏ | 27146/43939 [43:48:21<26:28:49, 5.68s/it]
62%|██████▏ | 27147/43939 [43:48:27<26:31:07, 5.69s/it]
62%|██████▏ | 27148/43939 [43:48:33<26:34:03, 5.70s/it]
62%|██████▏ | 27149/43939 [43:48:38<26:32:43, 5.69s/it]
62%|██████▏ | 27150/43939 [43:48:44<26:35:45, 5.70s/it]
{'loss': 0.7625, 'grad_norm': 1.24141526222229, 'learning_rate': 1.9130944945374628e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.58, 'epoch': 2.47}
62%|██████▏ | 27150/43939 [43:48:44<26:35:45, 5.70s/it]
62%|██████▏ | 27151/43939 [43:48:50<26:37:30, 5.71s/it]
62%|██████▏ | 27152/43939 [43:48:55<26:38:33, 5.71s/it]
62%|██████▏ | 27153/43939 [43:49:01<26:41:02, 5.72s/it]
62%|██████▏ | 27154/43939 [43:49:07<26:35:48, 5.70s/it]
62%|██████▏ | 27155/43939 [43:49:12<26:31:30, 5.69s/it]
62%|██████▏ | 27156/43939 [43:49:18<26:35:11, 5.70s/it]
62%|██████▏ | 27157/43939 [43:49:24<26:56:16, 5.78s/it]
62%|██████▏ | 27158/43939 [43:49:30<26:49:01, 5.75s/it]
62%|██████▏ | 27159/43939 [43:49:36<26:46:15, 5.74s/it]
62%|██████▏ | 27160/43939 [43:49:41<26:46:19, 5.74s/it]
62%|██████▏ | 27161/43939 [43:49:47<26:41:54, 5.73s/it]
62%|██████▏ | 27162/43939 [43:49:53<26:43:27, 5.73s/it]
62%|██████▏ | 27163/43939 [43:49:58<26:41:45, 5.73s/it]
62%|██████▏ | 27164/43939 [43:50:04<26:39:26, 5.72s/it]
62%|██████▏ | 27165/43939 [43:50:10<26:37:21, 5.71s/it]
62%|██████▏ | 27166/43939 [43:50:16<26:34:52, 5.71s/it]
62%|██████▏ | 27167/43939 [43:50:21<26:32:08, 5.70s/it]
62%|██████▏ | 27168/43939 [43:50:27<26:33:47, 5.70s/it]
62%|██████▏ | 27169/43939 [43:50:33<26:33:53, 5.70s/it]
62%|██████▏ | 27170/43939 [43:50:38<26:35:50, 5.71s/it]
62%|██████▏ | 27171/43939 [43:50:44<26:36:03, 5.71s/it]
62%|██████▏ | 27172/43939 [43:50:50<26:35:32, 5.71s/it]
62%|██████▏ | 27173/43939 [43:50:55<26:30:27, 5.69s/it]
62%|██████▏ | 27174/43939 [43:51:01<26:28:46, 5.69s/it]
62%|██████▏ | 27175/43939 [43:51:07<26:31:30, 5.70s/it]
62%|██████▏ | 27176/43939 [43:51:13<26:33:54, 5.71s/it]
62%|██████▏ | 27177/43939 [43:51:18<26:35:38, 5.71s/it]
62%|██████▏ | 27178/43939 [43:51:24<26:55:46, 5.78s/it]
62%|██████▏ | 27179/43939 [43:51:30<26:53:15, 5.78s/it]
62%|██████▏ | 27180/43939 [43:51:36<26:43:17, 5.74s/it]
62%|██████▏ | 27181/43939 [43:51:41<26:40:52, 5.73s/it]
62%|██████▏ | 27182/43939 [43:51:47<26:41:22, 5.73s/it]
62%|██████▏ | 27183/43939 [43:51:53<26:36:56, 5.72s/it]
62%|██████▏ | 27184/43939 [43:51:58<26:35:02, 5.71s/it]
62%|██████▏ | 27185/43939 [43:52:04<26:34:16, 5.71s/it]
62%|██████▏ | 27186/43939 [43:52:10<26:33:49, 5.71s/it]
62%|██████▏ | 27187/43939 [43:52:16<26:32:26, 5.70s/it]
62%|██████▏ | 27188/43939 [43:52:21<26:32:04, 5.70s/it]
62%|██████▏ | 27189/43939 [43:52:27<26:31:13, 5.70s/it]
62%|██████▏ | 27190/43939 [43:52:33<26:33:20, 5.71s/it]
62%|██████▏ | 27191/43939 [43:52:38<26:31:51, 5.70s/it]
62%|██████▏ | 27192/43939 [43:52:44<26:33:03, 5.71s/it]
62%|██████▏ | 27193/43939 [43:52:50<26:32:05, 5.70s/it]
62%|██████▏ | 27194/43939 [43:52:55<26:29:45, 5.70s/it]
62%|██████▏ | 27195/43939 [43:53:01<26:29:56, 5.70s/it]
62%|██████▏ | 27196/43939 [43:53:07<26:26:48, 5.69s/it]
62%|██████▏ | 27197/43939 [43:53:13<26:29:06, 5.70s/it]
62%|██████▏ | 27198/43939 [43:53:18<26:30:01, 5.70s/it]
62%|██████▏ | 27199/43939 [43:53:24<26:51:04, 5.77s/it]
62%|██████▏ | 27200/43939 [43:53:30<26:47:38, 5.76s/it]
{'loss': 0.7703, 'grad_norm': 1.4876346588134766, 'learning_rate': 1.9034464857255856e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.16, 'epoch': 2.48}
62%|██████▏ | 27200/43939 [43:53:30<26:47:38, 5.76s/it]
62%|██████▏ | 27201/43939 [43:53:36<26:44:58, 5.75s/it]
62%|██████▏ | 27202/43939 [43:53:41<26:39:39, 5.73s/it]
62%|██████▏ | 27203/43939 [43:53:47<26:37:04, 5.73s/it]
62%|██████▏ | 27204/43939 [43:53:53<26:35:10, 5.72s/it]
62%|██████▏ | 27205/43939 [43:53:59<26:37:44, 5.73s/it]
62%|██████▏ | 27206/43939 [43:54:04<26:34:59, 5.72s/it]
62%|██████▏ | 27207/43939 [43:54:10<26:32:52, 5.71s/it]
62%|██████▏ | 27208/43939 [43:54:16<26:32:27, 5.71s/it]
62%|██████▏ | 27209/43939 [43:54:21<26:37:23, 5.73s/it]
62%|██████▏ | 27210/43939 [43:54:27<26:33:52, 5.72s/it]
62%|██████▏ | 27211/43939 [43:54:33<26:30:20, 5.70s/it]
62%|██████▏ | 27212/43939 [43:54:38<26:27:52, 5.70s/it]
62%|██████▏ | 27213/43939 [43:54:44<26:30:09, 5.70s/it]
62%|██████▏ | 27214/43939 [43:54:50<26:30:11, 5.70s/it]
62%|██████▏ | 27215/43939 [43:54:56<26:28:43, 5.70s/it]
62%|██████▏ | 27216/43939 [43:55:01<26:29:01, 5.70s/it]
62%|██████▏ | 27217/43939 [43:55:07<26:26:27, 5.69s/it]
62%|██████▏ | 27218/43939 [43:55:13<26:23:32, 5.68s/it]
62%|██████▏ | 27219/43939 [43:55:18<26:23:14, 5.68s/it]
62%|██████▏ | 27220/43939 [43:55:24<26:23:12, 5.68s/it]
62%|██████▏ | 27221/43939 [43:55:30<26:41:57, 5.75s/it]
62%|██████▏ | 27222/43939 [43:55:36<26:38:52, 5.74s/it]
62%|██████▏ | 27223/43939 [43:55:41<26:37:54, 5.74s/it]
62%|██████▏ | 27224/43939 [43:55:47<26:33:35, 5.72s/it]
62%|██████▏ | 27225/43939 [43:55:53<26:32:25, 5.72s/it]
62%|██████▏ | 27226/43939 [43:55:58<26:30:00, 5.71s/it]
62%|██████▏ | 27227/43939 [43:56:04<26:27:44, 5.70s/it]
62%|██████▏ | 27228/43939 [43:56:10<26:28:48, 5.70s/it]
62%|██████▏ | 27229/43939 [43:56:15<26:29:31, 5.71s/it]
62%|██████▏ | 27230/43939 [43:56:21<26:25:43, 5.69s/it]
62%|██████▏ | 27231/43939 [43:56:27<26:25:27, 5.69s/it]
62%|██████▏ | 27232/43939 [43:56:33<26:30:18, 5.71s/it]
62%|██████▏ | 27233/43939 [43:56:38<26:27:25, 5.70s/it]
62%|██████▏ | 27234/43939 [43:56:44<26:25:50, 5.70s/it]
62%|██████▏ | 27235/43939 [43:56:50<26:25:52, 5.70s/it]
62%|██████▏ | 27236/43939 [43:56:55<26:28:00, 5.70s/it]
62%|██████▏ | 27237/43939 [43:57:01<26:27:53, 5.70s/it]
62%|██████▏ | 27238/43939 [43:57:07<26:31:15, 5.72s/it]
62%|██████▏ | 27239/43939 [43:57:13<26:30:44, 5.72s/it]
62%|██████▏ | 27240/43939 [43:57:18<26:30:05, 5.71s/it]
62%|██████▏ | 27241/43939 [43:57:24<26:25:37, 5.70s/it]
62%|██████▏ | 27242/43939 [43:57:30<26:45:42, 5.77s/it]
62%|██████▏ | 27243/43939 [43:57:36<26:37:49, 5.74s/it]
62%|██████▏ | 27244/43939 [43:57:41<26:31:38, 5.72s/it]
62%|██████▏ | 27245/43939 [43:57:47<26:30:03, 5.71s/it]
62%|██████▏ | 27246/43939 [43:57:53<26:28:33, 5.71s/it]
62%|██████▏ | 27247/43939 [43:57:58<26:30:10, 5.72s/it]
62%|██████▏ | 27248/43939 [43:58:04<26:25:51, 5.70s/it]
62%|██████▏ | 27249/43939 [43:58:10<26:26:43, 5.70s/it]
62%|██████▏ | 27250/43939 [43:58:15<26:25:40, 5.70s/it]
{'loss': 0.7652, 'grad_norm': 1.2360036373138428, 'learning_rate': 1.893807888947277e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.55, 'epoch': 2.48}
62%|██████▏ | 27250/43939 [43:58:15<26:25:40, 5.70s/it]
62%|██████▏ | 27251/43939 [43:58:21<26:32:06, 5.72s/it]
62%|██████▏ | 27252/43939 [43:58:27<26:27:35, 5.71s/it]
62%|██████▏ | 27253/43939 [43:58:33<26:23:46, 5.69s/it]
62%|██████▏ | 27254/43939 [43:58:38<26:26:17, 5.70s/it]
62%|██████▏ | 27255/43939 [43:58:44<26:28:43, 5.71s/it]
62%|██████▏ | 27256/43939 [43:58:50<26:26:15, 5.70s/it]
62%|██████▏ | 27257/43939 [43:58:55<26:28:29, 5.71s/it]
62%|██████▏ | 27258/43939 [43:59:01<26:27:15, 5.71s/it]
62%|██████▏ | 27259/43939 [43:59:07<26:25:14, 5.70s/it]
62%|██████▏ | 27260/43939 [43:59:12<26:24:18, 5.70s/it]
62%|██████▏ | 27261/43939 [43:59:18<26:22:59, 5.69s/it]
62%|██████▏ | 27262/43939 [43:59:24<26:23:28, 5.70s/it]
62%|██████▏ | 27263/43939 [43:59:30<26:40:58, 5.76s/it]
62%|██████▏ | 27264/43939 [43:59:36<26:39:32, 5.76s/it]
62%|██████▏ | 27265/43939 [43:59:41<26:31:56, 5.73s/it]
62%|██████▏ | 27266/43939 [43:59:47<26:26:38, 5.71s/it]
62%|██████▏ | 27267/43939 [43:59:53<26:26:46, 5.71s/it]
62%|██████▏ | 27268/43939 [43:59:58<26:24:28, 5.70s/it]
62%|██████▏ | 27269/43939 [44:00:04<26:28:20, 5.72s/it]
62%|██████▏ | 27270/43939 [44:00:10<26:24:46, 5.70s/it]
62%|██████▏ | 27271/43939 [44:00:15<26:29:52, 5.72s/it]
62%|██████▏ | 27272/43939 [44:00:21<26:29:14, 5.72s/it]
62%|██████▏ | 27273/43939 [44:00:27<26:26:52, 5.71s/it]
62%|██████▏ | 27274/43939 [44:00:33<26:24:19, 5.70s/it]
62%|██████▏ | 27275/43939 [44:00:38<26:27:08, 5.71s/it]
62%|██████▏ | 27276/43939 [44:00:44<26:26:06, 5.71s/it]
62%|██████▏ | 27277/43939 [44:00:50<26:24:17, 5.71s/it]
62%|██████▏ | 27278/43939 [44:00:55<26:28:18, 5.72s/it]
62%|██████▏ | 27279/43939 [44:01:01<26:27:53, 5.72s/it]
62%|██████▏ | 27280/43939 [44:01:07<26:25:46, 5.71s/it]
62%|██████▏ | 27281/43939 [44:01:13<26:22:50, 5.70s/it]
62%|██████▏ | 27282/43939 [44:01:18<26:23:43, 5.70s/it]
62%|██████▏ | 27283/43939 [44:01:24<26:23:55, 5.71s/it]
62%|██████▏ | 27284/43939 [44:01:30<26:24:09, 5.71s/it]
62%|██████▏ | 27285/43939 [44:01:36<26:46:59, 5.79s/it]
62%|██████▏ | 27286/43939 [44:01:41<26:39:05, 5.76s/it]
62%|██████▏ | 27287/43939 [44:01:47<26:31:52, 5.74s/it]
62%|██████▏ | 27288/43939 [44:01:53<26:27:30, 5.72s/it]
62%|██████▏ | 27289/43939 [44:01:58<26:30:55, 5.73s/it]
62%|██████▏ | 27290/43939 [44:02:04<26:31:12, 5.73s/it]
62%|██████▏ | 27291/43939 [44:02:10<26:28:08, 5.72s/it]
62%|██████▏ | 27292/43939 [44:02:16<26:26:38, 5.72s/it]
62%|██████▏ | 27293/43939 [44:02:21<26:31:34, 5.74s/it]
62%|██████▏ | 27294/43939 [44:02:27<26:26:50, 5.72s/it]
62%|██████▏ | 27295/43939 [44:02:33<26:23:37, 5.71s/it]
62%|██████▏ | 27296/43939 [44:02:38<26:27:24, 5.72s/it]
62%|██████▏ | 27297/43939 [44:02:44<26:24:13, 5.71s/it]
62%|██████▏ | 27298/43939 [44:02:50<26:23:23, 5.71s/it]
62%|██████▏ | 27299/43939 [44:02:56<26:24:25, 5.71s/it]
62%|██████▏ | 27300/43939 [44:03:01<26:20:16, 5.70s/it]
{'loss': 0.7767, 'grad_norm': 1.5152608156204224, 'learning_rate': 1.8841788562740528e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.47, 'epoch': 2.49}
62%|██████▏ | 27300/43939 [44:03:01<26:20:16, 5.70s/it]
62%|██████▏ | 27301/43939 [44:03:07<26:20:47, 5.70s/it]
62%|██████▏ | 27302/43939 [44:03:13<26:23:58, 5.71s/it]
62%|██████▏ | 27303/43939 [44:03:18<26:24:26, 5.71s/it]
62%|██████▏ | 27304/43939 [44:03:24<26:21:30, 5.70s/it]
62%|██████▏ | 27305/43939 [44:03:30<26:20:08, 5.70s/it]
62%|██████▏ | 27306/43939 [44:03:36<26:39:27, 5.77s/it]
62%|██████▏ | 27307/43939 [44:03:41<26:34:03, 5.75s/it]
62%|██████▏ | 27308/43939 [44:03:47<26:31:57, 5.74s/it]
62%|██████▏ | 27309/43939 [44:03:53<26:27:40, 5.73s/it]
62%|██████▏ | 27310/43939 [44:03:59<26:22:02, 5.71s/it]
62%|██████▏ | 27311/43939 [44:04:04<26:24:14, 5.72s/it]
62%|██████▏ | 27312/43939 [44:04:10<26:23:41, 5.71s/it]
62%|██████▏ | 27313/43939 [44:04:16<26:23:36, 5.71s/it]
62%|██████▏ | 27314/43939 [44:04:21<26:22:04, 5.71s/it]
62%|██████▏ | 27315/43939 [44:04:27<26:20:03, 5.70s/it]
62%|██████▏ | 27316/43939 [44:04:33<26:18:57, 5.70s/it]
62%|██████▏ | 27317/43939 [44:04:38<26:19:42, 5.70s/it]
62%|██████▏ | 27318/43939 [44:04:44<26:22:30, 5.71s/it]
62%|██████▏ | 27319/43939 [44:04:50<26:22:20, 5.71s/it]
62%|██████▏ | 27320/43939 [44:04:56<26:21:01, 5.71s/it]
62%|██████▏ | 27321/43939 [44:05:01<26:21:27, 5.71s/it]
62%|██████▏ | 27322/43939 [44:05:07<26:19:45, 5.70s/it]
62%|██████▏ | 27323/43939 [44:05:13<26:20:11, 5.71s/it]
62%|██████▏ | 27324/43939 [44:05:18<26:22:16, 5.71s/it]
62%|██████▏ | 27325/43939 [44:05:24<26:20:59, 5.71s/it]
62%|██████▏ | 27326/43939 [44:05:30<26:20:57, 5.71s/it]
62%|██████▏ | 27327/43939 [44:05:36<26:36:05, 5.76s/it]
62%|██████▏ | 27328/43939 [44:05:41<26:28:33, 5.74s/it]
62%|██████▏ | 27329/43939 [44:05:47<26:26:12, 5.73s/it]
62%|██████▏ | 27330/43939 [44:05:53<26:24:47, 5.73s/it]
62%|██████▏ | 27331/43939 [44:05:59<26:31:51, 5.75s/it]
62%|██████▏ | 27332/43939 [44:06:04<26:25:55, 5.73s/it]
62%|██████▏ | 27333/43939 [44:06:10<26:23:38, 5.72s/it]
62%|██████▏ | 27334/43939 [44:06:16<26:28:34, 5.74s/it]
62%|██████▏ | 27335/43939 [44:06:22<26:23:50, 5.72s/it]
62%|██████▏ | 27336/43939 [44:06:27<26:20:03, 5.71s/it]
62%|██████▏ | 27337/43939 [44:06:33<26:18:38, 5.71s/it]
62%|██████▏ | 27338/43939 [44:06:39<26:14:11, 5.69s/it]
62%|██████▏ | 27339/43939 [44:06:44<26:10:26, 5.68s/it]
62%|██████▏ | 27340/43939 [44:06:50<26:11:23, 5.68s/it]
62%|██████▏ | 27341/43939 [44:06:56<26:14:56, 5.69s/it]
62%|██████▏ | 27342/43939 [44:07:01<26:13:25, 5.69s/it]
62%|██████▏ | 27343/43939 [44:07:07<26:15:37, 5.70s/it]
62%|██████▏ | 27344/43939 [44:07:13<26:13:50, 5.69s/it]
62%|██████▏ | 27345/43939 [44:07:18<26:15:16, 5.70s/it]
62%|██████▏ | 27346/43939 [44:07:24<26:19:56, 5.71s/it]
62%|██████▏ | 27347/43939 [44:07:30<26:16:33, 5.70s/it]
62%|██████▏ | 27348/43939 [44:07:36<26:30:39, 5.75s/it]
62%|██████▏ | 27349/43939 [44:07:41<26:25:24, 5.73s/it]
62%|██████▏ | 27350/43939 [44:07:47<26:24:35, 5.73s/it]
{'loss': 0.7812, 'grad_norm': 1.4290846586227417, 'learning_rate': 1.8745595396265305e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.87, 'epoch': 2.49}
62%|██████▏ | 27350/43939 [44:07:47<26:24:35, 5.73s/it]
62%|██████▏ | 27351/43939 [44:07:53<26:26:16, 5.74s/it]
62%|██████▏ | 27352/43939 [44:07:59<26:25:52, 5.74s/it]
62%|██████▏ | 27353/43939 [44:08:04<26:23:53, 5.73s/it]
62%|██████▏ | 27354/43939 [44:08:10<26:20:00, 5.72s/it]
62%|██████▏ | 27355/43939 [44:08:16<26:18:59, 5.71s/it]
62%|██████▏ | 27356/43939 [44:08:21<26:15:17, 5.70s/it]
62%|██████▏ | 27357/43939 [44:08:27<26:13:44, 5.69s/it]
62%|██████▏ | 27358/43939 [44:08:33<26:12:43, 5.69s/it]
62%|██████▏ | 27359/43939 [44:08:38<26:12:38, 5.69s/it]
62%|██████▏ | 27360/43939 [44:08:44<26:10:46, 5.68s/it]
62%|██████▏ | 27361/43939 [44:08:50<26:09:53, 5.68s/it]
62%|██████▏ | 27362/43939 [44:08:55<26:10:34, 5.68s/it]
62%|██████▏ | 27363/43939 [44:09:01<26:08:49, 5.68s/it]
62%|██████▏ | 27364/43939 [44:09:07<26:07:51, 5.68s/it]
62%|██████▏ | 27365/43939 [44:09:12<26:07:33, 5.67s/it]
62%|██████▏ | 27366/43939 [44:09:18<26:08:56, 5.68s/it]
62%|██████▏ | 27367/43939 [44:09:24<26:08:30, 5.68s/it]
62%|██████▏ | 27368/43939 [44:09:30<26:10:59, 5.69s/it]
62%|██████▏ | 27369/43939 [44:09:35<26:12:15, 5.69s/it]
62%|██████▏ | 27370/43939 [44:09:41<26:39:31, 5.79s/it]
62%|██████▏ | 27371/43939 [44:09:47<26:33:07, 5.77s/it]
62%|██████▏ | 27372/43939 [44:09:53<26:24:14, 5.74s/it]
62%|██████▏ | 27373/43939 [44:09:58<26:20:46, 5.73s/it]
62%|██████▏ | 27374/43939 [44:10:04<26:21:20, 5.73s/it]
62%|██████▏ | 27375/43939 [44:10:10<26:24:11, 5.74s/it]
62%|██████▏ | 27376/43939 [44:10:16<26:21:05, 5.73s/it]
62%|██████▏ | 27377/43939 [44:10:21<26:19:20, 5.72s/it]
62%|██████▏ | 27378/43939 [44:10:27<26:18:24, 5.72s/it]
62%|██████▏ | 27379/43939 [44:10:33<26:13:57, 5.70s/it]
62%|██████▏ | 27380/43939 [44:10:38<26:12:34, 5.70s/it]
62%|██████▏ | 27381/43939 [44:10:44<26:09:58, 5.69s/it]
62%|██████▏ | 27382/43939 [44:10:50<26:08:31, 5.68s/it]
62%|██████▏ | 27383/43939 [44:10:55<26:10:04, 5.69s/it]
62%|██████▏ | 27384/43939 [44:11:01<26:10:34, 5.69s/it]
62%|██████▏ | 27385/43939 [44:11:07<26:12:20, 5.70s/it]
62%|██████▏ | 27386/43939 [44:11:12<26:12:02, 5.70s/it]
62%|██████▏ | 27387/43939 [44:11:18<26:14:50, 5.71s/it]
62%|██████▏ | 27388/43939 [44:11:24<26:13:59, 5.71s/it]
62%|██████▏ | 27389/43939 [44:11:30<26:13:11, 5.70s/it]
62%|██████▏ | 27390/43939 [44:11:35<26:17:34, 5.72s/it]
62%|██████▏ | 27391/43939 [44:11:41<26:17:10, 5.72s/it]
62%|██████▏ | 27392/43939 [44:11:47<26:32:58, 5.78s/it]
62%|██████▏ | 27393/43939 [44:11:53<26:31:22, 5.77s/it]
62%|██████▏ | 27394/43939 [44:11:58<26:26:03, 5.75s/it]
62%|██████▏ | 27395/43939 [44:12:04<26:22:27, 5.74s/it]
62%|██████▏ | 27396/43939 [44:12:10<26:19:39, 5.73s/it]
62%|██████▏ | 27397/43939 [44:12:16<26:20:41, 5.73s/it]
62%|██████▏ | 27398/43939 [44:12:21<26:17:12, 5.72s/it]
62%|██████▏ | 27399/43939 [44:12:27<26:12:32, 5.70s/it]
62%|██████▏ | 27400/43939 [44:12:33<26:10:26, 5.70s/it]
{'loss': 0.7831, 'grad_norm': 1.3766820430755615, 'learning_rate': 1.864950090772036e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.86, 'epoch': 2.49}
62%|██████▏ | 27400/43939 [44:12:33<26:10:26, 5.70s/it]
62%|██████▏ | 27401/43939 [44:12:38<26:09:55, 5.70s/it]
62%|██████▏ | 27402/43939 [44:12:44<26:08:49, 5.69s/it]
62%|██████▏ | 27403/43939 [44:12:50<26:11:02, 5.70s/it]
62%|██████▏ | 27404/43939 [44:12:55<26:13:22, 5.71s/it]
62%|██████▏ | 27405/43939 [44:13:01<26:17:38, 5.73s/it]
62%|██████▏ | 27406/43939 [44:13:07<26:14:11, 5.71s/it]
62%|██████▏ | 27407/43939 [44:13:13<26:13:09, 5.71s/it]
62%|██████▏ | 27408/43939 [44:13:18<26:15:34, 5.72s/it]
62%|██████▏ | 27409/43939 [44:13:24<26:16:17, 5.72s/it]
62%|██████▏ | 27410/43939 [44:13:30<26:14:46, 5.72s/it]
62%|██████▏ | 27411/43939 [44:13:35<26:12:13, 5.71s/it]
62%|██████▏ | 27412/43939 [44:13:41<26:11:55, 5.71s/it]
62%|██████▏ | 27413/43939 [44:13:47<26:29:46, 5.77s/it]
62%|██████▏ | 27414/43939 [44:13:53<26:26:09, 5.76s/it]
62%|██████▏ | 27415/43939 [44:13:59<26:24:06, 5.75s/it]
62%|██████▏ | 27416/43939 [44:14:04<26:20:17, 5.74s/it]
62%|██████▏ | 27417/43939 [44:14:10<26:20:06, 5.74s/it]
62%|██████▏ | 27418/43939 [44:14:16<26:14:36, 5.72s/it]
62%|██████▏ | 27419/43939 [44:14:21<26:17:48, 5.73s/it]
62%|██████▏ | 27420/43939 [44:14:27<26:17:31, 5.73s/it]
62%|██████▏ | 27421/43939 [44:14:33<26:12:40, 5.71s/it]
62%|██████▏ | 27422/43939 [44:14:39<26:12:24, 5.71s/it]
62%|██████▏ | 27423/43939 [44:14:44<26:10:38, 5.71s/it]
62%|██████▏ | 27424/43939 [44:14:50<26:10:10, 5.70s/it]
62%|██████▏ | 27425/43939 [44:14:56<26:12:14, 5.71s/it]
62%|██████▏ | 27426/43939 [44:15:01<26:09:58, 5.70s/it]
62%|██████▏ | 27427/43939 [44:15:07<26:05:22, 5.69s/it]
62%|██████▏ | 27428/43939 [44:15:13<26:05:17, 5.69s/it]
62%|██████▏ | 27429/43939 [44:15:18<26:07:21, 5.70s/it]
62%|██████▏ | 27430/43939 [44:15:24<26:09:00, 5.70s/it]
62%|██████▏ | 27431/43939 [44:15:30<26:12:54, 5.72s/it]
62%|██████▏ | 27432/43939 [44:15:36<26:12:39, 5.72s/it]
62%|██████▏ | 27433/43939 [44:15:41<26:15:59, 5.73s/it]
62%|██████▏ | 27434/43939 [44:15:47<26:31:30, 5.79s/it]
62%|██████▏ | 27435/43939 [44:15:53<26:24:29, 5.76s/it]
62%|██████▏ | 27436/43939 [44:15:59<26:20:00, 5.74s/it]
62%|██████▏ | 27437/43939 [44:16:04<26:15:27, 5.73s/it]
62%|██████▏ | 27438/43939 [44:16:10<26:13:49, 5.72s/it]
62%|██████▏ | 27439/43939 [44:16:16<26:09:45, 5.71s/it]
62%|██████▏ | 27440/43939 [44:16:21<26:09:03, 5.71s/it]
62%|██████▏ | 27441/43939 [44:16:27<26:06:35, 5.70s/it]
62%|██████▏ | 27442/43939 [44:16:33<26:08:50, 5.71s/it]
62%|██████▏ | 27443/43939 [44:16:39<26:09:05, 5.71s/it]
62%|██████▏ | 27444/43939 [44:16:44<26:07:24, 5.70s/it]
62%|██████▏ | 27445/43939 [44:16:50<26:06:00, 5.70s/it]
62%|██████▏ | 27446/43939 [44:16:56<26:09:26, 5.71s/it]
62%|██████▏ | 27447/43939 [44:17:01<26:09:49, 5.71s/it]
62%|██████▏ | 27448/43939 [44:17:07<26:11:25, 5.72s/it]
62%|██████▏ | 27449/43939 [44:17:13<26:09:25, 5.71s/it]
62%|██████▏ | 27450/43939 [44:17:19<26:10:54, 5.72s/it]
{'loss': 0.7738, 'grad_norm': 1.2925307750701904, 'learning_rate': 1.8553506613222056e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.61, 'epoch': 2.5}
62%|██████▏ | 27450/43939 [44:17:19<26:10:54, 5.72s/it]
62%|██████▏ | 27451/43939 [44:17:24<26:13:27, 5.73s/it]
62%|██████▏ | 27452/43939 [44:17:30<26:11:26, 5.72s/it]
62%|██████▏ | 27453/43939 [44:17:36<26:07:02, 5.70s/it]
62%|██████▏ | 27454/43939 [44:17:41<26:09:54, 5.71s/it]
62%|██████▏ | 27455/43939 [44:17:47<26:25:10, 5.77s/it]
62%|██████▏ | 27456/43939 [44:17:53<26:21:17, 5.76s/it]
62%|██████▏ | 27457/43939 [44:17:59<26:16:57, 5.74s/it]
62%|██████▏ | 27458/43939 [44:18:04<26:14:38, 5.73s/it]
62%|██████▏ | 27459/43939 [44:18:10<26:16:13, 5.74s/it]
62%|██████▏ | 27460/43939 [44:18:16<26:12:10, 5.72s/it]
62%|██████▏ | 27461/43939 [44:18:22<26:07:49, 5.71s/it]
63%|██████▎ | 27462/43939 [44:18:27<26:02:28, 5.69s/it]
63%|██████▎ | 27463/43939 [44:18:33<26:04:05, 5.70s/it]
63%|██████▎ | 27464/43939 [44:18:39<26:07:33, 5.71s/it]
63%|██████▎ | 27465/43939 [44:18:44<26:08:37, 5.71s/it][2025-11-30 14:22:58,203] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-27465
63%|██████▎ | 27466/43939 [44:18:52<29:04:38, 6.35s/it]
63%|██████▎ | 27467/43939 [44:18:58<28:08:27, 6.15s/it]
63%|██████▎ | 27468/43939 [44:19:04<27:29:28, 6.01s/it]
63%|██████▎ | 27469/43939 [44:19:09<27:05:49, 5.92s/it]
63%|██████▎ | 27470/43939 [44:19:15<26:44:53, 5.85s/it]
63%|██████▎ | 27471/43939 [44:19:21<26:30:20, 5.79s/it]
63%|██████▎ | 27472/43939 [44:19:26<26:23:23, 5.77s/it]
63%|██████▎ | 27473/43939 [44:19:32<26:23:33, 5.77s/it]
63%|██████▎ | 27474/43939 [44:19:38<26:20:18, 5.76s/it]
63%|██████▎ | 27475/43939 [44:19:44<26:10:41, 5.72s/it]
63%|██████▎ | 27476/43939 [44:19:49<26:23:25, 5.77s/it]
63%|██████▎ | 27477/43939 [44:19:55<26:16:29, 5.75s/it]
63%|██████▎ | 27478/43939 [44:20:01<26:15:17, 5.74s/it]
63%|██████▎ | 27479/43939 [44:20:07<26:11:47, 5.73s/it]
63%|██████▎ | 27480/43939 [44:20:12<26:12:40, 5.73s/it]
63%|██████▎ | 27481/43939 [44:20:18<26:17:23, 5.75s/it]
63%|██████▎ | 27482/43939 [44:20:24<26:12:31, 5.73s/it]
63%|██████▎ | 27483/43939 [44:20:29<26:09:11, 5.72s/it]
63%|██████▎ | 27484/43939 [44:20:35<26:10:37, 5.73s/it]
63%|██████▎ | 27485/43939 [44:20:41<26:08:11, 5.72s/it]
63%|██████▎ | 27486/43939 [44:20:47<26:03:22, 5.70s/it]
63%|██████▎ | 27487/43939 [44:20:52<26:02:02, 5.70s/it]
63%|██████▎ | 27488/43939 [44:20:58<26:02:18, 5.70s/it]
63%|██████▎ | 27489/43939 [44:21:04<26:03:17, 5.70s/it]
63%|██████▎ | 27490/43939 [44:21:09<26:06:40, 5.71s/it]
63%|██████▎ | 27491/43939 [44:21:15<26:03:26, 5.70s/it]
63%|██████▎ | 27492/43939 [44:21:21<26:05:41, 5.71s/it]
63%|██████▎ | 27493/43939 [44:21:26<26:03:06, 5.70s/it]
63%|██████▎ | 27494/43939 [44:21:32<26:01:40, 5.70s/it]
63%|██████▎ | 27495/43939 [44:21:38<26:02:45, 5.70s/it]
63%|██████▎ | 27496/43939 [44:21:44<26:00:56, 5.70s/it]
63%|██████▎ | 27497/43939 [44:21:49<26:20:06, 5.77s/it]
63%|██████▎ | 27498/43939 [44:21:55<26:12:42, 5.74s/it]
63%|██████▎ | 27499/43939 [44:22:01<26:08:21, 5.72s/it]
63%|██████▎ | 27500/43939 [44:22:07<26:06:41, 5.72s/it]
{'loss': 0.7921, 'grad_norm': 1.4461274147033691, 'learning_rate': 1.845761402730598e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.47, 'epoch': 2.5}
63%|██████▎ | 27500/43939 [44:22:07<26:06:41, 5.72s/it]
63%|██████▎ | 27501/43939 [44:22:12<26:05:48, 5.72s/it]
63%|██████▎ | 27502/43939 [44:22:18<26:04:48, 5.71s/it]
63%|██████▎ | 27503/43939 [44:22:24<26:08:30, 5.73s/it]
63%|██████▎ | 27504/43939 [44:22:29<26:10:18, 5.73s/it]
63%|██████▎ | 27505/43939 [44:22:35<26:08:17, 5.73s/it]
63%|██████▎ | 27506/43939 [44:22:41<26:06:02, 5.72s/it]
63%|██████▎ | 27507/43939 [44:22:47<26:03:01, 5.71s/it]
63%|██████▎ | 27508/43939 [44:22:52<26:02:40, 5.71s/it]
63%|██████▎ | 27509/43939 [44:22:58<26:02:25, 5.71s/it]
63%|██████▎ | 27510/43939 [44:23:04<26:00:07, 5.70s/it]
63%|██████▎ | 27511/43939 [44:23:09<26:01:01, 5.70s/it]
63%|██████▎ | 27512/43939 [44:23:15<25:59:04, 5.69s/it]
63%|██████▎ | 27513/43939 [44:23:21<26:00:02, 5.70s/it]
63%|██████▎ | 27514/43939 [44:23:26<25:59:44, 5.70s/it]
63%|██████▎ | 27515/43939 [44:23:32<26:02:36, 5.71s/it]
63%|██████▎ | 27516/43939 [44:23:38<26:03:49, 5.71s/it]
63%|██████▎ | 27517/43939 [44:23:44<26:03:55, 5.71s/it]
63%|██████▎ | 27518/43939 [44:23:50<26:20:01, 5.77s/it]
63%|██████▎ | 27519/43939 [44:23:55<26:14:36, 5.75s/it]
63%|██████▎ | 27520/43939 [44:24:01<26:09:23, 5.74s/it]
63%|██████▎ | 27521/43939 [44:24:07<26:07:09, 5.73s/it]
63%|██████▎ | 27522/43939 [44:24:12<26:03:28, 5.71s/it]
63%|██████▎ | 27523/43939 [44:24:18<26:02:22, 5.71s/it]
63%|██████▎ | 27524/43939 [44:24:24<26:01:20, 5.71s/it]
63%|██████▎ | 27525/43939 [44:24:29<25:59:20, 5.70s/it]
63%|██████▎ | 27526/43939 [44:24:35<26:01:14, 5.71s/it]
63%|██████▎ | 27527/43939 [44:24:41<25:59:00, 5.70s/it]
63%|██████▎ | 27528/43939 [44:24:46<25:58:08, 5.70s/it]
63%|██████▎ | 27529/43939 [44:24:52<25:59:15, 5.70s/it]
63%|██████▎ | 27530/43939 [44:24:58<25:58:17, 5.70s/it]
63%|██████▎ | 27531/43939 [44:25:04<26:00:40, 5.71s/it]
63%|██████▎ | 27532/43939 [44:25:09<25:58:31, 5.70s/it]
63%|██████▎ | 27533/43939 [44:25:15<25:58:50, 5.70s/it]
63%|██████▎ | 27534/43939 [44:25:21<26:00:52, 5.71s/it]
63%|██████▎ | 27535/43939 [44:25:26<25:59:17, 5.70s/it]
63%|██████▎ | 27536/43939 [44:25:32<25:57:52, 5.70s/it]
63%|██████▎ | 27537/43939 [44:25:38<26:02:43, 5.72s/it]
63%|██████▎ | 27538/43939 [44:25:44<26:01:50, 5.71s/it]
63%|██████▎ | 27539/43939 [44:25:50<26:22:33, 5.79s/it]
63%|██████▎ | 27540/43939 [44:25:55<26:14:25, 5.76s/it]
63%|██████▎ | 27541/43939 [44:26:01<26:10:00, 5.74s/it]
63%|██████▎ | 27542/43939 [44:26:07<26:07:36, 5.74s/it]
63%|██████▎ | 27543/43939 [44:26:12<26:05:51, 5.73s/it]
63%|██████▎ | 27544/43939 [44:26:18<26:02:00, 5.72s/it]
63%|██████▎ | 27545/43939 [44:26:24<26:01:50, 5.72s/it]
63%|██████▎ | 27546/43939 [44:26:29<25:58:18, 5.70s/it]
63%|██████▎ | 27547/43939 [44:26:35<25:58:26, 5.70s/it]
63%|██████▎ | 27548/43939 [44:26:41<26:00:51, 5.71s/it]
63%|██████▎ | 27549/43939 [44:26:47<26:00:49, 5.71s/it]
63%|██████▎ | 27550/43939 [44:26:52<26:01:30, 5.72s/it]
{'loss': 0.7685, 'grad_norm': 1.3631300926208496, 'learning_rate': 1.836182466290302e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.4, 'epoch': 2.51}
63%|██████▎ | 27550/43939 [44:26:52<26:01:30, 5.72s/it]
63%|██████▎ | 27551/43939 [44:26:58<26:04:37, 5.73s/it]
63%|██████▎ | 27552/43939 [44:27:04<26:05:31, 5.73s/it]
63%|██████▎ | 27553/43939 [44:27:10<26:03:08, 5.72s/it]
63%|██████▎ | 27554/43939 [44:27:15<26:04:17, 5.73s/it]
63%|██████▎ | 27555/43939 [44:27:21<26:02:33, 5.72s/it]
63%|██████▎ | 27556/43939 [44:27:27<25:58:32, 5.71s/it]
63%|██████▎ | 27557/43939 [44:27:32<25:58:10, 5.71s/it]
63%|██████▎ | 27558/43939 [44:27:38<25:57:10, 5.70s/it]
63%|██████▎ | 27559/43939 [44:27:44<25:56:40, 5.70s/it]
63%|██████▎ | 27560/43939 [44:27:49<25:55:20, 5.70s/it]
63%|██████▎ | 27561/43939 [44:27:55<26:13:38, 5.76s/it]
63%|██████▎ | 27562/43939 [44:28:01<26:12:55, 5.76s/it]
63%|██████▎ | 27563/43939 [44:28:07<26:08:16, 5.75s/it]
63%|██████▎ | 27564/43939 [44:28:13<26:04:35, 5.73s/it]
63%|██████▎ | 27565/43939 [44:28:18<26:04:29, 5.73s/it]
63%|██████▎ | 27566/43939 [44:28:24<26:05:09, 5.74s/it]
63%|██████▎ | 27567/43939 [44:28:30<26:03:38, 5.73s/it]
63%|██████▎ | 27568/43939 [44:28:35<26:00:13, 5.72s/it]
63%|██████▎ | 27569/43939 [44:28:41<25:57:18, 5.71s/it]
63%|██████▎ | 27570/43939 [44:28:47<25:55:51, 5.70s/it]
63%|██████▎ | 27571/43939 [44:28:52<25:56:22, 5.71s/it]
63%|██████▎ | 27572/43939 [44:28:58<25:58:26, 5.71s/it]
63%|██████▎ | 27573/43939 [44:29:04<26:00:14, 5.72s/it]
63%|██████▎ | 27574/43939 [44:29:10<26:03:30, 5.73s/it]
63%|██████▎ | 27575/43939 [44:29:15<26:02:37, 5.73s/it]
63%|██████▎ | 27576/43939 [44:29:21<25:57:31, 5.71s/it]
63%|██████▎ | 27577/43939 [44:29:27<25:55:15, 5.70s/it]
63%|██████▎ | 27578/43939 [44:29:32<25:54:07, 5.70s/it]
63%|██████▎ | 27579/43939 [44:29:38<25:51:16, 5.69s/it]
63%|██████▎ | 27580/43939 [44:29:44<25:52:20, 5.69s/it]
63%|██████▎ | 27581/43939 [44:29:50<25:53:56, 5.70s/it]
63%|██████▎ | 27582/43939 [44:29:55<26:10:59, 5.76s/it]
63%|██████▎ | 27583/43939 [44:30:01<26:05:35, 5.74s/it]
63%|██████▎ | 27584/43939 [44:30:07<26:00:39, 5.73s/it]
63%|██████▎ | 27585/43939 [44:30:13<25:58:32, 5.72s/it]
63%|██████▎ | 27586/43939 [44:30:18<25:55:38, 5.71s/it]
63%|██████▎ | 27587/43939 [44:30:24<25:54:59, 5.71s/it]
63%|██████▎ | 27588/43939 [44:30:30<25:53:47, 5.70s/it]
63%|██████▎ | 27589/43939 [44:30:35<25:54:42, 5.71s/it]
63%|██████▎ | 27590/43939 [44:30:41<25:53:45, 5.70s/it]
63%|██████▎ | 27591/43939 [44:30:47<25:55:16, 5.71s/it]
63%|██████▎ | 27592/43939 [44:30:52<25:52:57, 5.70s/it]
63%|██████▎ | 27593/43939 [44:30:58<25:51:51, 5.70s/it]
63%|██████▎ | 27594/43939 [44:31:04<25:54:35, 5.71s/it]
63%|██████▎ | 27595/43939 [44:31:10<25:54:33, 5.71s/it]
63%|██████▎ | 27596/43939 [44:31:15<25:53:10, 5.70s/it]
63%|██████▎ | 27597/43939 [44:31:21<25:50:33, 5.69s/it]
63%|██████▎ | 27598/43939 [44:31:27<25:48:53, 5.69s/it]
63%|██████▎ | 27599/43939 [44:31:32<25:52:46, 5.70s/it]
63%|██████▎ | 27600/43939 [44:31:38<25:57:21, 5.72s/it]
{'loss': 0.78, 'grad_norm': 1.1906591653823853, 'learning_rate': 1.8266140031315503e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.81, 'epoch': 2.51}
63%|██████▎ | 27600/43939 [44:31:38<25:57:21, 5.72s/it]
63%|██████▎ | 27601/43939 [44:31:44<25:55:52, 5.71s/it]
63%|██████▎ | 27602/43939 [44:31:50<25:54:02, 5.71s/it]
63%|██████▎ | 27603/43939 [44:31:55<26:14:19, 5.78s/it]
63%|██████▎ | 27604/43939 [44:32:01<26:06:17, 5.75s/it]
63%|██████▎ | 27605/43939 [44:32:07<26:03:45, 5.74s/it]
63%|██████▎ | 27606/43939 [44:32:13<26:02:23, 5.74s/it]
63%|██████▎ | 27607/43939 [44:32:18<25:58:21, 5.73s/it]
63%|██████▎ | 27608/43939 [44:32:24<25:57:27, 5.72s/it]
63%|██████▎ | 27609/43939 [44:32:30<25:55:17, 5.71s/it]
63%|██████▎ | 27610/43939 [44:32:35<25:54:56, 5.71s/it]
63%|██████▎ | 27611/43939 [44:32:41<25:54:50, 5.71s/it]
63%|██████▎ | 27612/43939 [44:32:47<25:50:58, 5.70s/it]
63%|██████▎ | 27613/43939 [44:32:52<25:50:43, 5.70s/it]
63%|██████▎ | 27614/43939 [44:32:58<25:48:51, 5.69s/it]
63%|██████▎ | 27615/43939 [44:33:04<25:49:24, 5.69s/it]
63%|██████▎ | 27616/43939 [44:33:10<25:49:46, 5.70s/it]
63%|██████▎ | 27617/43939 [44:33:15<25:50:38, 5.70s/it]
63%|██████▎ | 27618/43939 [44:33:21<25:49:29, 5.70s/it]
63%|██████▎ | 27619/43939 [44:33:27<25:50:54, 5.70s/it]
63%|██████▎ | 27620/43939 [44:33:32<25:51:58, 5.71s/it]
63%|██████▎ | 27621/43939 [44:33:38<25:49:59, 5.70s/it]
63%|██████▎ | 27622/43939 [44:33:44<25:51:39, 5.71s/it]
63%|██████▎ | 27623/43939 [44:33:50<25:55:34, 5.72s/it]
63%|██████▎ | 27624/43939 [44:33:56<26:18:12, 5.80s/it]
63%|██████▎ | 27625/43939 [44:34:01<26:07:58, 5.77s/it]
63%|██████▎ | 27626/43939 [44:34:07<26:00:09, 5.74s/it]
63%|██████▎ | 27627/43939 [44:34:13<25:55:58, 5.72s/it]
63%|██████▎ | 27628/43939 [44:34:18<25:54:21, 5.72s/it]
63%|██████▎ | 27629/43939 [44:34:24<25:52:46, 5.71s/it]
63%|██████▎ | 27630/43939 [44:34:30<25:49:40, 5.70s/it]
63%|██████▎ | 27631/43939 [44:34:35<25:49:06, 5.70s/it]
63%|██████▎ | 27632/43939 [44:34:41<25:50:00, 5.70s/it]
63%|██████▎ | 27633/43939 [44:34:47<25:50:25, 5.70s/it]
63%|██████▎ | 27634/43939 [44:34:52<25:47:54, 5.70s/it]
63%|██████▎ | 27635/43939 [44:34:58<25:45:27, 5.69s/it]
63%|██████▎ | 27636/43939 [44:35:04<25:47:02, 5.69s/it]
63%|██████▎ | 27637/43939 [44:35:10<25:49:37, 5.70s/it]
63%|██████▎ | 27638/43939 [44:35:15<25:57:23, 5.73s/it]
63%|██████▎ | 27639/43939 [44:35:21<25:55:31, 5.73s/it]
63%|██████▎ | 27640/43939 [44:35:27<25:56:23, 5.73s/it]
63%|██████▎ | 27641/43939 [44:35:33<25:56:48, 5.73s/it]
63%|██████▎ | 27642/43939 [44:35:38<25:56:10, 5.73s/it]
63%|██████▎ | 27643/43939 [44:35:44<25:54:25, 5.72s/it]
63%|██████▎ | 27644/43939 [44:35:50<25:52:00, 5.71s/it]
63%|██████▎ | 27645/43939 [44:35:55<25:51:53, 5.71s/it]
63%|██████▎ | 27646/43939 [44:36:01<26:07:34, 5.77s/it]
63%|██████▎ | 27647/43939 [44:36:07<26:01:33, 5.75s/it]
63%|██████▎ | 27648/43939 [44:36:13<26:03:44, 5.76s/it]
63%|██████▎ | 27649/43939 [44:36:18<26:00:18, 5.75s/it]
63%|██████▎ | 27650/43939 [44:36:24<26:00:43, 5.75s/it]
{'loss': 0.7766, 'grad_norm': 1.3514851331710815, 'learning_rate': 1.8170561642193345e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.14, 'epoch': 2.52}
63%|██████▎ | 27650/43939 [44:36:24<26:00:43, 5.75s/it]
63%|██████▎ | 27651/43939 [44:36:30<25:59:24, 5.74s/it]
63%|██████▎ | 27652/43939 [44:36:36<25:57:59, 5.74s/it]
63%|██████▎ | 27653/43939 [44:36:41<25:56:51, 5.74s/it]
63%|██████▎ | 27654/43939 [44:36:47<25:53:34, 5.72s/it]
63%|██████▎ | 27655/43939 [44:36:53<25:51:08, 5.72s/it]
63%|██████▎ | 27656/43939 [44:36:59<25:54:41, 5.73s/it]
63%|██████▎ | 27657/43939 [44:37:04<25:52:47, 5.72s/it]
63%|██████▎ | 27658/43939 [44:37:10<25:50:15, 5.71s/it]
63%|██████▎ | 27659/43939 [44:37:16<25:50:06, 5.71s/it]
63%|██████▎ | 27660/43939 [44:37:21<25:49:42, 5.71s/it]
63%|██████▎ | 27661/43939 [44:37:27<25:47:47, 5.71s/it]
63%|██████▎ | 27662/43939 [44:37:33<25:49:05, 5.71s/it]
63%|██████▎ | 27663/43939 [44:37:39<25:49:30, 5.71s/it]
63%|██████▎ | 27664/43939 [44:37:44<25:48:14, 5.71s/it]
63%|██████▎ | 27665/43939 [44:37:50<25:54:27, 5.73s/it]
63%|██████▎ | 27666/43939 [44:37:56<25:54:17, 5.73s/it]
63%|██████▎ | 27667/43939 [44:38:02<26:07:08, 5.78s/it]
63%|██████▎ | 27668/43939 [44:38:07<26:01:22, 5.76s/it]
63%|██████▎ | 27669/43939 [44:38:13<25:54:39, 5.73s/it]
63%|██████▎ | 27670/43939 [44:38:19<25:56:46, 5.74s/it]
63%|██████▎ | 27671/43939 [44:38:24<25:52:50, 5.73s/it]
63%|██████▎ | 27672/43939 [44:38:30<25:50:42, 5.72s/it]
63%|██████▎ | 27673/43939 [44:38:36<25:48:48, 5.71s/it]
63%|██████▎ | 27674/43939 [44:38:42<25:48:52, 5.71s/it]
63%|██████▎ | 27675/43939 [44:38:47<25:49:53, 5.72s/it]
63%|██████▎ | 27676/43939 [44:38:53<25:50:00, 5.72s/it]
63%|██████▎ | 27677/43939 [44:38:59<25:49:05, 5.72s/it]
63%|██████▎ | 27678/43939 [44:39:04<25:46:00, 5.70s/it]
63%|██████▎ | 27679/43939 [44:39:10<25:46:41, 5.71s/it]
63%|██████▎ | 27680/43939 [44:39:16<25:43:19, 5.70s/it]
63%|██████▎ | 27681/43939 [44:39:22<25:44:04, 5.70s/it]
63%|██████▎ | 27682/43939 [44:39:27<25:44:59, 5.70s/it]
63%|██████▎ | 27683/43939 [44:39:33<25:42:31, 5.69s/it]
63%|██████▎ | 27684/43939 [44:39:39<25:40:24, 5.69s/it]
63%|██████▎ | 27685/43939 [44:39:44<25:40:43, 5.69s/it]
63%|██████▎ | 27686/43939 [44:39:50<25:42:21, 5.69s/it]
63%|██████▎ | 27687/43939 [44:39:56<25:42:01, 5.69s/it]
63%|██████▎ | 27688/43939 [44:40:02<25:59:08, 5.76s/it]
63%|██████▎ | 27689/43939 [44:40:07<25:56:03, 5.75s/it]
63%|██████▎ | 27690/43939 [44:40:13<25:51:41, 5.73s/it]
63%|██████▎ | 27691/43939 [44:40:19<25:48:56, 5.72s/it]
63%|██████▎ | 27692/43939 [44:40:24<25:44:46, 5.70s/it]
63%|██████▎ | 27693/43939 [44:40:30<25:42:22, 5.70s/it]
63%|██████▎ | 27694/43939 [44:40:36<25:42:48, 5.70s/it]
63%|██████▎ | 27695/43939 [44:40:41<25:42:41, 5.70s/it]
63%|██████▎ | 27696/43939 [44:40:47<25:39:05, 5.69s/it]
63%|██████▎ | 27697/43939 [44:40:53<25:42:27, 5.70s/it]
63%|██████▎ | 27698/43939 [44:40:59<25:45:53, 5.71s/it]
63%|██████▎ | 27699/43939 [44:41:04<25:47:44, 5.72s/it]
63%|██████▎ | 27700/43939 [44:41:10<25:44:56, 5.71s/it]
{'loss': 0.7927, 'grad_norm': 1.5444172620773315, 'learning_rate': 1.807509100351024e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 505.9, 'epoch': 2.52}
63%|██████▎ | 27700/43939 [44:41:10<25:44:56, 5.71s/it]
63%|██████▎ | 27701/43939 [44:41:16<25:47:33, 5.72s/it]
63%|██████▎ | 27702/43939 [44:41:21<25:47:13, 5.72s/it]
63%|██████▎ | 27703/43939 [44:41:27<25:47:59, 5.72s/it]
63%|██████▎ | 27704/43939 [44:41:33<25:45:24, 5.71s/it]
63%|██████▎ | 27705/43939 [44:41:39<25:45:39, 5.71s/it]
63%|██████▎ | 27706/43939 [44:41:44<25:43:07, 5.70s/it]
63%|██████▎ | 27707/43939 [44:41:50<25:47:09, 5.72s/it]
63%|██████▎ | 27708/43939 [44:41:56<25:44:31, 5.71s/it]
63%|██████▎ | 27709/43939 [44:42:02<26:04:07, 5.78s/it]
63%|██████▎ | 27710/43939 [44:42:07<25:58:14, 5.76s/it]
63%|██████▎ | 27711/43939 [44:42:13<25:53:45, 5.74s/it]
63%|██████▎ | 27712/43939 [44:42:19<25:52:53, 5.74s/it]
63%|██████▎ | 27713/43939 [44:42:24<25:48:58, 5.73s/it]
63%|██████▎ | 27714/43939 [44:42:30<25:45:30, 5.72s/it]
63%|██████▎ | 27715/43939 [44:42:36<25:43:08, 5.71s/it]
63%|██████▎ | 27716/43939 [44:42:42<25:42:08, 5.70s/it]
63%|██████▎ | 27717/43939 [44:42:47<25:41:32, 5.70s/it]
63%|██████▎ | 27718/43939 [44:42:53<25:44:06, 5.71s/it]
63%|██████▎ | 27719/43939 [44:42:59<25:41:51, 5.70s/it]
63%|██████▎ | 27720/43939 [44:43:04<25:43:53, 5.71s/it]
63%|██████▎ | 27721/43939 [44:43:10<25:46:20, 5.72s/it]
63%|██████▎ | 27722/43939 [44:43:16<25:45:51, 5.72s/it]
63%|██████▎ | 27723/43939 [44:43:22<25:44:42, 5.72s/it]
63%|██████▎ | 27724/43939 [44:43:27<25:46:06, 5.72s/it]
63%|██████▎ | 27725/43939 [44:43:33<25:44:55, 5.72s/it]
63%|██████▎ | 27726/43939 [44:43:39<25:43:22, 5.71s/it]
63%|██████▎ | 27727/43939 [44:43:44<25:41:52, 5.71s/it]
63%|██████▎ | 27728/43939 [44:43:50<25:43:18, 5.71s/it]
63%|██████▎ | 27729/43939 [44:43:56<25:46:18, 5.72s/it]
63%|██████▎ | 27730/43939 [44:44:02<25:43:00, 5.71s/it]
63%|██████▎ | 27731/43939 [44:44:08<26:01:46, 5.78s/it]
63%|██████▎ | 27732/43939 [44:44:13<25:52:41, 5.75s/it]
63%|██████▎ | 27733/43939 [44:44:19<25:47:18, 5.73s/it]
63%|██████▎ | 27734/43939 [44:44:25<25:46:36, 5.73s/it]
63%|██████▎ | 27735/43939 [44:44:30<25:45:51, 5.72s/it]
63%|██████▎ | 27736/43939 [44:44:36<25:42:10, 5.71s/it]
63%|██████▎ | 27737/43939 [44:44:42<25:40:53, 5.71s/it]
63%|██████▎ | 27738/43939 [44:44:47<25:44:19, 5.72s/it]
63%|██████▎ | 27739/43939 [44:44:53<25:47:13, 5.73s/it]
63%|██████▎ | 27740/43939 [44:44:59<25:48:20, 5.73s/it]
63%|██████▎ | 27741/43939 [44:45:05<25:48:42, 5.74s/it]
63%|██████▎ | 27742/43939 [44:45:10<25:45:27, 5.72s/it]
63%|██████▎ | 27743/43939 [44:45:16<25:42:26, 5.71s/it]
63%|██████▎ | 27744/43939 [44:45:22<25:40:40, 5.71s/it]
63%|██████▎ | 27745/43939 [44:45:27<25:37:04, 5.69s/it]
63%|██████▎ | 27746/43939 [44:45:33<25:37:03, 5.70s/it]
63%|██████▎ | 27747/43939 [44:45:39<25:36:44, 5.69s/it]
63%|██████▎ | 27748/43939 [44:45:44<25:33:59, 5.68s/it]
63%|██████▎ | 27749/43939 [44:45:50<25:35:40, 5.69s/it]
63%|██████▎ | 27750/43939 [44:45:56<25:36:55, 5.70s/it]
{'loss': 0.7759, 'grad_norm': 1.380354642868042, 'learning_rate': 1.797972962153987e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.04, 'epoch': 2.53}
63%|██████▎ | 27750/43939 [44:45:56<25:36:55, 5.70s/it]
63%|██████▎ | 27751/43939 [44:46:02<25:35:32, 5.69s/it]
63%|██████▎ | 27752/43939 [44:46:07<25:34:18, 5.69s/it]
63%|██████▎ | 27753/43939 [44:46:13<25:52:52, 5.76s/it]
63%|██████▎ | 27754/43939 [44:46:19<25:49:02, 5.74s/it]
63%|██████▎ | 27755/43939 [44:46:25<25:45:10, 5.73s/it]
63%|██████▎ | 27756/43939 [44:46:30<25:41:09, 5.71s/it]
63%|██████▎ | 27757/43939 [44:46:36<25:38:37, 5.70s/it]
63%|██████▎ | 27758/43939 [44:46:42<25:35:46, 5.69s/it]
63%|██████▎ | 27759/43939 [44:46:47<25:34:18, 5.69s/it]
63%|██████▎ | 27760/43939 [44:46:53<25:35:09, 5.69s/it]
63%|██████▎ | 27761/43939 [44:46:59<25:39:50, 5.71s/it]
63%|██████▎ | 27762/43939 [44:47:04<25:38:47, 5.71s/it]
63%|██████▎ | 27763/43939 [44:47:10<25:39:09, 5.71s/it]
63%|██████▎ | 27764/43939 [44:47:16<25:34:40, 5.69s/it]
63%|██████▎ | 27765/43939 [44:47:21<25:31:55, 5.68s/it]
63%|██████▎ | 27766/43939 [44:47:27<25:32:29, 5.69s/it]
63%|██████▎ | 27767/43939 [44:47:33<25:31:54, 5.68s/it]
63%|██████▎ | 27768/43939 [44:47:39<25:32:10, 5.68s/it]
63%|██████▎ | 27769/43939 [44:47:44<25:34:17, 5.69s/it]
63%|██████▎ | 27770/43939 [44:47:50<25:33:07, 5.69s/it]
63%|██████▎ | 27771/43939 [44:47:56<25:33:46, 5.69s/it]
63%|██████▎ | 27772/43939 [44:48:01<25:38:13, 5.71s/it]
63%|██████▎ | 27773/43939 [44:48:07<25:38:09, 5.71s/it]
63%|██████▎ | 27774/43939 [44:48:13<25:50:07, 5.75s/it]
63%|██████▎ | 27775/43939 [44:48:19<25:45:13, 5.74s/it]
63%|██████▎ | 27776/43939 [44:48:24<25:42:53, 5.73s/it]
63%|██████▎ | 27777/43939 [44:48:30<25:38:04, 5.71s/it]
63%|██████▎ | 27778/43939 [44:48:36<25:39:30, 5.72s/it]
63%|██████▎ | 27779/43939 [44:48:41<25:37:01, 5.71s/it]
63%|██████▎ | 27780/43939 [44:48:47<25:37:10, 5.71s/it]
63%|██████▎ | 27781/43939 [44:48:53<25:36:10, 5.70s/it]
63%|██████▎ | 27782/43939 [44:48:58<25:33:03, 5.69s/it]
63%|██████▎ | 27783/43939 [44:49:04<25:32:54, 5.69s/it]
63%|██████▎ | 27784/43939 [44:49:10<25:35:54, 5.70s/it]
63%|██████▎ | 27785/43939 [44:49:16<25:31:36, 5.69s/it]
63%|██████▎ | 27786/43939 [44:49:21<25:33:14, 5.70s/it]
63%|██████▎ | 27787/43939 [44:49:27<25:32:52, 5.69s/it]
63%|██████▎ | 27788/43939 [44:49:33<25:35:12, 5.70s/it]
63%|██████▎ | 27789/43939 [44:49:38<25:34:36, 5.70s/it]
63%|██████▎ | 27790/43939 [44:49:44<25:33:05, 5.70s/it]
63%|██████▎ | 27791/43939 [44:49:50<25:39:12, 5.72s/it]
63%|██████▎ | 27792/43939 [44:49:56<25:36:17, 5.71s/it]
63%|██████▎ | 27793/43939 [44:50:01<25:38:48, 5.72s/it]
63%|██████▎ | 27794/43939 [44:50:07<25:38:00, 5.72s/it]
63%|██████▎ | 27795/43939 [44:50:13<25:56:23, 5.78s/it]
63%|██████▎ | 27796/43939 [44:50:19<25:48:57, 5.76s/it]
63%|██████▎ | 27797/43939 [44:50:24<25:44:19, 5.74s/it]
63%|██████▎ | 27798/43939 [44:50:30<25:48:29, 5.76s/it]
63%|██████▎ | 27799/43939 [44:50:36<25:46:55, 5.75s/it]
63%|██████▎ | 27800/43939 [44:50:42<25:44:37, 5.74s/it]
{'loss': 0.7711, 'grad_norm': 1.250873327255249, 'learning_rate': 1.7884479000832126e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.58, 'epoch': 2.53}
63%|██████▎ | 27800/43939 [44:50:42<25:44:37, 5.74s/it]
63%|██████▎ | 27801/43939 [44:50:47<25:40:32, 5.73s/it]
63%|██████▎ | 27802/43939 [44:50:53<25:40:12, 5.73s/it]
63%|██████▎ | 27803/43939 [44:50:59<25:40:59, 5.73s/it]
63%|██████▎ | 27804/43939 [44:51:04<25:38:58, 5.72s/it]
63%|██████▎ | 27805/43939 [44:51:10<25:38:26, 5.72s/it]
63%|██████▎ | 27806/43939 [44:51:16<25:36:41, 5.72s/it]
63%|██████▎ | 27807/43939 [44:51:22<25:35:18, 5.71s/it]
63%|██████▎ | 27808/43939 [44:51:27<25:35:58, 5.71s/it]
63%|██████▎ | 27809/43939 [44:51:33<25:36:27, 5.72s/it]
63%|██████▎ | 27810/43939 [44:51:39<25:37:26, 5.72s/it]
63%|██████▎ | 27811/43939 [44:51:44<25:38:28, 5.72s/it]
63%|██████▎ | 27812/43939 [44:51:50<25:38:24, 5.72s/it]
63%|██████▎ | 27813/43939 [44:51:56<25:34:57, 5.71s/it]
63%|██████▎ | 27814/43939 [44:52:02<25:38:22, 5.72s/it]
63%|██████▎ | 27815/43939 [44:52:07<25:36:49, 5.72s/it]
63%|██████▎ | 27816/43939 [44:52:13<25:55:36, 5.79s/it]
63%|██████▎ | 27817/43939 [44:52:19<25:53:41, 5.78s/it]
63%|██████▎ | 27818/43939 [44:52:25<25:49:02, 5.77s/it]
63%|██████▎ | 27819/43939 [44:52:30<25:45:10, 5.75s/it]
63%|██████▎ | 27820/43939 [44:52:36<25:46:36, 5.76s/it]
63%|██████▎ | 27821/43939 [44:52:42<25:42:14, 5.74s/it]
63%|██████▎ | 27822/43939 [44:52:48<25:39:59, 5.73s/it]
63%|██████▎ | 27823/43939 [44:52:53<25:39:59, 5.73s/it]
63%|██████▎ | 27824/43939 [44:52:59<25:35:51, 5.72s/it]
63%|██████▎ | 27825/43939 [44:53:05<25:31:40, 5.70s/it]
63%|██████▎ | 27826/43939 [44:53:10<25:31:48, 5.70s/it]
63%|██████▎ | 27827/43939 [44:53:16<25:30:49, 5.70s/it]
63%|██████▎ | 27828/43939 [44:53:22<25:30:41, 5.70s/it]
63%|██████▎ | 27829/43939 [44:53:28<25:30:04, 5.70s/it]
63%|██████▎ | 27830/43939 [44:53:33<25:30:44, 5.70s/it]
63%|██████▎ | 27831/43939 [44:53:39<25:30:50, 5.70s/it]
63%|██████▎ | 27832/43939 [44:53:45<25:34:36, 5.72s/it]
63%|██████▎ | 27833/43939 [44:53:50<25:33:05, 5.71s/it]
63%|██████▎ | 27834/43939 [44:53:56<25:31:12, 5.70s/it]
63%|██████▎ | 27835/43939 [44:54:02<25:31:03, 5.70s/it]
63%|██████▎ | 27836/43939 [44:54:07<25:30:57, 5.70s/it]
63%|██████▎ | 27837/43939 [44:54:13<25:48:46, 5.77s/it]
63%|██████▎ | 27838/43939 [44:54:19<25:42:00, 5.75s/it]
63%|██████▎ | 27839/43939 [44:54:25<25:37:36, 5.73s/it]
63%|██████▎ | 27840/43939 [44:54:30<25:32:52, 5.71s/it]
63%|██████▎ | 27841/43939 [44:54:36<25:30:01, 5.70s/it]
63%|██████▎ | 27842/43939 [44:54:42<25:30:10, 5.70s/it]
63%|██████▎ | 27843/43939 [44:54:48<25:27:02, 5.69s/it]
63%|██████▎ | 27844/43939 [44:54:53<25:27:23, 5.69s/it]
63%|██████▎ | 27845/43939 [44:54:59<25:29:03, 5.70s/it]
63%|██████▎ | 27846/43939 [44:55:05<25:34:13, 5.72s/it]
63%|██████▎ | 27847/43939 [44:55:10<25:32:09, 5.71s/it]
63%|██████▎ | 27848/43939 [44:55:16<25:28:16, 5.70s/it]
63%|██████▎ | 27849/43939 [44:55:22<25:26:01, 5.69s/it]
63%|██████▎ | 27850/43939 [44:55:27<25:30:19, 5.71s/it]
{'loss': 0.7771, 'grad_norm': 1.5400545597076416, 'learning_rate': 1.778934064418939e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.82, 'epoch': 2.54}
63%|██████▎ | 27850/43939 [44:55:27<25:30:19, 5.71s/it]
63%|██████▎ | 27851/43939 [44:55:33<25:30:32, 5.71s/it]
63%|██████▎ | 27852/43939 [44:55:39<25:29:00, 5.70s/it]
63%|██████▎ | 27853/43939 [44:55:45<25:29:12, 5.70s/it]
63%|██████▎ | 27854/43939 [44:55:50<25:29:47, 5.71s/it]
63%|██████▎ | 27855/43939 [44:55:56<25:32:05, 5.72s/it]
63%|██████▎ | 27856/43939 [44:56:02<25:29:26, 5.71s/it]
63%|██████▎ | 27857/43939 [44:56:07<25:27:47, 5.70s/it]
63%|██████▎ | 27858/43939 [44:56:13<25:51:11, 5.79s/it]
63%|██████▎ | 27859/43939 [44:56:19<25:40:57, 5.75s/it]
63%|██████▎ | 27860/43939 [44:56:25<25:37:32, 5.74s/it]
63%|██████▎ | 27861/43939 [44:56:30<25:33:44, 5.72s/it]
63%|██████▎ | 27862/43939 [44:56:36<25:32:18, 5.72s/it]
63%|██████▎ | 27863/43939 [44:56:42<25:32:23, 5.72s/it]
63%|██████▎ | 27864/43939 [44:56:48<25:30:18, 5.71s/it]
63%|██████▎ | 27865/43939 [44:56:53<25:31:24, 5.72s/it]
63%|██████▎ | 27866/43939 [44:56:59<25:30:33, 5.71s/it]
63%|██████▎ | 27867/43939 [44:57:05<25:27:23, 5.70s/it]
63%|██████▎ | 27868/43939 [44:57:10<25:30:10, 5.71s/it]
63%|██████▎ | 27869/43939 [44:57:16<25:31:57, 5.72s/it]
63%|██████▎ | 27870/43939 [44:57:22<25:25:33, 5.70s/it]
63%|██████▎ | 27871/43939 [44:57:28<25:28:31, 5.71s/it]
63%|██████▎ | 27872/43939 [44:57:33<25:26:18, 5.70s/it]
63%|██████▎ | 27873/43939 [44:57:39<25:27:16, 5.70s/it]
63%|██████▎ | 27874/43939 [44:57:45<25:26:56, 5.70s/it]
63%|██████▎ | 27875/43939 [44:57:50<25:23:22, 5.69s/it]
63%|██████▎ | 27876/43939 [44:57:56<25:25:11, 5.70s/it]
63%|██████▎ | 27877/43939 [44:58:02<25:25:19, 5.70s/it]
63%|██████▎ | 27878/43939 [44:58:07<25:24:13, 5.69s/it]
63%|██████▎ | 27879/43939 [44:58:13<25:26:33, 5.70s/it]
63%|██████▎ | 27880/43939 [44:58:19<25:43:05, 5.77s/it]
63%|██████▎ | 27881/43939 [44:58:25<25:38:44, 5.75s/it]
63%|██████▎ | 27882/43939 [44:58:30<25:36:14, 5.74s/it]
63%|██████▎ | 27883/43939 [44:58:36<25:34:05, 5.73s/it]
63%|██████▎ | 27884/43939 [44:58:42<25:32:00, 5.73s/it]
63%|██████▎ | 27885/43939 [44:58:48<25:29:47, 5.72s/it]
63%|██████▎ | 27886/43939 [44:58:53<25:28:06, 5.71s/it]
63%|██████▎ | 27887/43939 [44:58:59<25:28:29, 5.71s/it]
63%|██████▎ | 27888/43939 [44:59:05<25:26:27, 5.71s/it]
63%|██████▎ | 27889/43939 [44:59:10<25:26:19, 5.71s/it]
63%|██████▎ | 27890/43939 [44:59:16<25:25:56, 5.70s/it]
63%|██████▎ | 27891/43939 [44:59:22<25:20:52, 5.69s/it]
63%|██████▎ | 27892/43939 [44:59:27<25:19:24, 5.68s/it]
63%|██████▎ | 27893/43939 [44:59:33<25:20:26, 5.69s/it]
63%|██████▎ | 27894/43939 [44:59:39<25:21:06, 5.69s/it]
63%|██████▎ | 27895/43939 [44:59:45<25:24:36, 5.70s/it]
63%|██████▎ | 27896/43939 [44:59:50<25:20:58, 5.69s/it]
63%|██████▎ | 27897/43939 [44:59:56<25:19:03, 5.68s/it]
63%|██████▎ | 27898/43939 [45:00:02<25:20:00, 5.69s/it]
63%|██████▎ | 27899/43939 [45:00:07<25:20:03, 5.69s/it]
63%|██████▎ | 27900/43939 [45:00:13<25:18:45, 5.68s/it]
{'loss': 0.7701, 'grad_norm': 1.4404981136322021, 'learning_rate': 1.7694316052642808e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.42, 'epoch': 2.54}
63%|██████▎ | 27900/43939 [45:00:13<25:18:45, 5.68s/it]
63%|██████▎ | 27901/43939 [45:00:19<25:37:25, 5.75s/it]
64%|██████▎ | 27902/43939 [45:00:24<25:28:58, 5.72s/it]
64%|██████▎ | 27903/43939 [45:00:30<25:27:38, 5.72s/it]
64%|██████▎ | 27904/43939 [45:00:36<25:28:40, 5.72s/it]
64%|██████▎ | 27905/43939 [45:00:42<25:28:39, 5.72s/it]
64%|██████▎ | 27906/43939 [45:00:47<25:29:54, 5.73s/it]
64%|██████▎ | 27907/43939 [45:00:53<25:26:41, 5.71s/it]
64%|██████▎ | 27908/43939 [45:00:59<25:24:20, 5.71s/it]
64%|██████▎ | 27909/43939 [45:01:05<25:30:35, 5.73s/it]
64%|██████▎ | 27910/43939 [45:01:10<25:32:12, 5.74s/it]
64%|██████▎ | 27911/43939 [45:01:16<25:28:57, 5.72s/it]
64%|██████▎ | 27912/43939 [45:01:22<25:30:23, 5.73s/it]
64%|██████▎ | 27913/43939 [45:01:27<25:25:17, 5.71s/it]
64%|██████▎ | 27914/43939 [45:01:33<25:23:32, 5.70s/it]
64%|██████▎ | 27915/43939 [45:01:39<25:20:50, 5.69s/it]
64%|██████▎ | 27916/43939 [45:01:44<25:22:58, 5.70s/it]
64%|██████▎ | 27917/43939 [45:01:50<25:22:29, 5.70s/it]
64%|██████▎ | 27918/43939 [45:01:56<25:24:27, 5.71s/it]
64%|██████▎ | 27919/43939 [45:02:02<25:21:29, 5.70s/it]
64%|██████▎ | 27920/43939 [45:02:07<25:23:08, 5.71s/it]
64%|██████▎ | 27921/43939 [45:02:13<25:25:00, 5.71s/it]
64%|██████▎ | 27922/43939 [45:02:19<25:42:18, 5.78s/it]
64%|██████▎ | 27923/43939 [45:02:25<25:33:40, 5.75s/it]
64%|██████▎ | 27924/43939 [45:02:30<25:27:56, 5.72s/it]
64%|██████▎ | 27925/43939 [45:02:36<25:23:13, 5.71s/it]
64%|██████▎ | 27926/43939 [45:02:42<25:21:57, 5.70s/it]
64%|██████▎ | 27927/43939 [45:02:47<25:22:25, 5.70s/it]
64%|██████▎ | 27928/43939 [45:02:53<25:21:07, 5.70s/it]
64%|██████▎ | 27929/43939 [45:02:59<25:22:27, 5.71s/it]
64%|██████▎ | 27930/43939 [45:03:04<25:18:41, 5.69s/it]
64%|██████▎ | 27931/43939 [45:03:10<25:21:12, 5.70s/it]
64%|██████▎ | 27932/43939 [45:03:16<25:25:30, 5.72s/it]
64%|██████▎ | 27933/43939 [45:03:22<25:25:44, 5.72s/it]
64%|██████▎ | 27934/43939 [45:03:27<25:23:37, 5.71s/it]
64%|██████▎ | 27935/43939 [45:03:33<25:21:01, 5.70s/it]
64%|██████▎ | 27936/43939 [45:03:39<25:18:41, 5.69s/it]
64%|██████▎ | 27937/43939 [45:03:44<25:16:51, 5.69s/it]
64%|██████▎ | 27938/43939 [45:03:50<25:16:22, 5.69s/it]
64%|██████▎ | 27939/43939 [45:03:56<25:15:54, 5.68s/it]
64%|██████▎ | 27940/43939 [45:04:01<25:15:31, 5.68s/it]
64%|██████▎ | 27941/43939 [45:04:07<25:18:37, 5.70s/it]
64%|██████▎ | 27942/43939 [45:04:13<25:21:03, 5.71s/it]
64%|██████▎ | 27943/43939 [45:04:19<25:39:59, 5.78s/it]
64%|██████▎ | 27944/43939 [45:04:24<25:32:12, 5.75s/it]
64%|██████▎ | 27945/43939 [45:04:30<25:30:10, 5.74s/it]
64%|██████▎ | 27946/43939 [45:04:36<25:27:56, 5.73s/it]
64%|██████▎ | 27947/43939 [45:04:42<25:25:26, 5.72s/it]
64%|██████▎ | 27948/43939 [45:04:47<25:19:03, 5.70s/it]
64%|██████▎ | 27949/43939 [45:04:53<25:16:44, 5.69s/it]
64%|██████▎ | 27950/43939 [45:04:59<25:23:19, 5.72s/it]
{'loss': 0.767, 'grad_norm': 1.529981017112732, 'learning_rate': 1.7599406725428617e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.07, 'epoch': 2.54}
64%|██████▎ | 27950/43939 [45:04:59<25:23:19, 5.72s/it]
64%|██████▎ | 27951/43939 [45:05:04<25:20:10, 5.70s/it]
64%|██████▎ | 27952/43939 [45:05:10<25:18:21, 5.70s/it]
64%|██████▎ | 27953/43939 [45:05:16<25:15:37, 5.69s/it]
64%|██████▎ | 27954/43939 [45:05:21<25:16:48, 5.69s/it]
64%|██████▎ | 27955/43939 [45:05:27<25:21:03, 5.71s/it]
64%|██████▎ | 27956/43939 [45:05:33<25:20:32, 5.71s/it]
64%|██████▎ | 27957/43939 [45:05:39<25:18:33, 5.70s/it]
64%|██████▎ | 27958/43939 [45:05:44<25:19:18, 5.70s/it]
64%|██████▎ | 27959/43939 [45:05:50<25:17:38, 5.70s/it]
64%|██████▎ | 27960/43939 [45:05:56<25:17:29, 5.70s/it]
64%|██████▎ | 27961/43939 [45:06:01<25:15:30, 5.69s/it]
64%|██████▎ | 27962/43939 [45:06:07<25:18:05, 5.70s/it]
64%|██████▎ | 27963/43939 [45:06:13<25:18:55, 5.70s/it]
64%|██████▎ | 27964/43939 [45:06:18<25:18:39, 5.70s/it]
64%|██████▎ | 27965/43939 [45:06:24<25:38:43, 5.78s/it]
64%|██████▎ | 27966/43939 [45:06:30<25:31:50, 5.75s/it]
64%|██████▎ | 27967/43939 [45:06:36<25:26:39, 5.74s/it]
64%|██████▎ | 27968/43939 [45:06:41<25:22:11, 5.72s/it]
64%|██████▎ | 27969/43939 [45:06:47<25:21:47, 5.72s/it]
64%|██████▎ | 27970/43939 [45:06:53<25:19:43, 5.71s/it]
64%|██████▎ | 27971/43939 [45:06:59<25:18:19, 5.71s/it]
64%|██████▎ | 27972/43939 [45:07:04<25:19:48, 5.71s/it]
64%|██████▎ | 27973/43939 [45:07:10<25:18:01, 5.70s/it]
64%|██████▎ | 27974/43939 [45:07:16<25:16:47, 5.70s/it]
64%|██████▎ | 27975/43939 [45:07:21<25:13:58, 5.69s/it]
64%|██████▎ | 27976/43939 [45:07:27<25:15:03, 5.69s/it]
64%|██████▎ | 27977/43939 [45:07:33<25:14:55, 5.69s/it]
64%|██████▎ | 27978/43939 [45:07:38<25:16:19, 5.70s/it]
64%|██████▎ | 27979/43939 [45:07:44<25:18:03, 5.71s/it]
64%|██████▎ | 27980/43939 [45:07:50<25:19:34, 5.71s/it]
64%|██████▎ | 27981/43939 [45:07:56<25:17:43, 5.71s/it]
64%|██████▎ | 27982/43939 [45:08:01<25:15:59, 5.70s/it]
64%|██████▎ | 27983/43939 [45:08:07<25:13:31, 5.69s/it]
64%|██████▎ | 27984/43939 [45:08:13<25:16:53, 5.70s/it]
64%|██████▎ | 27985/43939 [45:08:18<25:19:29, 5.71s/it]
64%|██████▎ | 27986/43939 [45:08:24<25:19:46, 5.72s/it]
64%|██████▎ | 27987/43939 [45:08:30<25:39:41, 5.79s/it]
64%|██████▎ | 27988/43939 [45:08:36<25:35:34, 5.78s/it]
64%|██████▎ | 27989/43939 [45:08:42<25:28:37, 5.75s/it]
64%|██████▎ | 27990/43939 [45:08:47<25:24:41, 5.74s/it]
64%|██████▎ | 27991/43939 [45:08:53<25:22:47, 5.73s/it]
64%|██████▎ | 27992/43939 [45:08:59<25:18:52, 5.71s/it]
64%|██████▎ | 27993/43939 [45:09:04<25:18:32, 5.71s/it]
64%|██████▎ | 27994/43939 [45:09:10<25:17:34, 5.71s/it]
64%|██████▎ | 27995/43939 [45:09:16<25:17:30, 5.71s/it]
64%|██████▎ | 27996/43939 [45:09:21<25:16:58, 5.71s/it]
64%|██████▎ | 27997/43939 [45:09:27<25:17:15, 5.71s/it]
64%|██████▎ | 27998/43939 [45:09:33<25:15:53, 5.71s/it]
64%|██████▎ | 27999/43939 [45:09:39<25:16:30, 5.71s/it]
64%|██████▎ | 28000/43939 [45:09:44<25:17:17, 5.71s/it]
{'loss': 0.7719, 'grad_norm': 1.5745155811309814, 'learning_rate': 1.7504614159964493e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.73, 'epoch': 2.55}
64%|██████▎ | 28000/43939 [45:09:44<25:17:17, 5.71s/it]
64%|██████▎ | 28001/43939 [45:09:50<25:17:13, 5.71s/it]
64%|██████▎ | 28002/43939 [45:09:56<25:15:27, 5.71s/it]
64%|██████▎ | 28003/43939 [45:10:01<25:16:04, 5.71s/it]
64%|██████▎ | 28004/43939 [45:10:07<25:13:17, 5.70s/it]
64%|██████▎ | 28005/43939 [45:10:13<25:11:19, 5.69s/it]
64%|██████▎ | 28006/43939 [45:10:19<25:12:10, 5.69s/it]
64%|██████▎ | 28007/43939 [45:10:24<25:12:48, 5.70s/it]
64%|██████▎ | 28008/43939 [45:10:30<25:35:07, 5.78s/it]
64%|██████▎ | 28009/43939 [45:10:36<25:30:20, 5.76s/it]
64%|██████▎ | 28010/43939 [45:10:42<25:24:20, 5.74s/it]
64%|██████▎ | 28011/43939 [45:10:47<25:21:45, 5.73s/it]
64%|██████▍ | 28012/43939 [45:10:53<25:20:02, 5.73s/it]
64%|██████▍ | 28013/43939 [45:10:59<25:19:49, 5.73s/it]
64%|██████▍ | 28014/43939 [45:11:04<25:16:31, 5.71s/it]
64%|██████▍ | 28015/43939 [45:11:10<25:15:44, 5.71s/it]
64%|██████▍ | 28016/43939 [45:11:16<25:17:45, 5.72s/it]
64%|██████▍ | 28017/43939 [45:11:22<25:14:52, 5.71s/it]
64%|██████▍ | 28018/43939 [45:11:27<25:11:37, 5.70s/it]
64%|██████▍ | 28019/43939 [45:11:33<25:11:52, 5.70s/it]
64%|██████▍ | 28020/43939 [45:11:39<25:10:41, 5.69s/it]
64%|██████▍ | 28021/43939 [45:11:44<25:08:14, 5.69s/it]
64%|██████▍ | 28022/43939 [45:11:50<25:14:32, 5.71s/it]
64%|██████▍ | 28023/43939 [45:11:56<25:14:38, 5.71s/it]
64%|██████▍ | 28024/43939 [45:12:01<25:13:38, 5.71s/it]
64%|██████▍ | 28025/43939 [45:12:07<25:14:43, 5.71s/it]
64%|██████▍ | 28026/43939 [45:12:13<25:19:54, 5.73s/it]
64%|██████▍ | 28027/43939 [45:12:19<25:15:18, 5.71s/it]
64%|██████▍ | 28028/43939 [45:12:24<25:13:59, 5.71s/it]
64%|██████▍ | 28029/43939 [45:12:30<25:29:25, 5.77s/it]
64%|██████▍ | 28030/43939 [45:12:36<25:25:13, 5.75s/it]
64%|██████▍ | 28031/43939 [45:12:42<25:18:50, 5.73s/it]
64%|██████▍ | 28032/43939 [45:12:47<25:21:37, 5.74s/it]
64%|██████▍ | 28033/43939 [45:12:53<25:18:48, 5.73s/it]
64%|██████▍ | 28034/43939 [45:12:59<25:14:22, 5.71s/it]
64%|██████▍ | 28035/43939 [45:13:04<25:12:35, 5.71s/it]
64%|██████▍ | 28036/43939 [45:13:10<25:12:37, 5.71s/it]
64%|██████▍ | 28037/43939 [45:13:16<25:11:28, 5.70s/it]
64%|██████▍ | 28038/43939 [45:13:22<25:08:47, 5.69s/it]
64%|██████▍ | 28039/43939 [45:13:27<25:08:29, 5.69s/it]
64%|██████▍ | 28040/43939 [45:13:33<25:07:27, 5.69s/it]
64%|██████▍ | 28041/43939 [45:13:39<25:08:44, 5.69s/it]
64%|██████▍ | 28042/43939 [45:13:44<25:08:41, 5.69s/it]
64%|██████▍ | 28043/43939 [45:13:50<25:10:38, 5.70s/it]
64%|██████▍ | 28044/43939 [45:13:56<25:06:32, 5.69s/it]
64%|██████▍ | 28045/43939 [45:14:01<25:06:53, 5.69s/it]
64%|██████▍ | 28046/43939 [45:14:07<25:08:05, 5.69s/it]
64%|██████▍ | 28047/43939 [45:14:13<25:09:21, 5.70s/it]
64%|██████▍ | 28048/43939 [45:14:18<25:11:10, 5.71s/it]
64%|██████▍ | 28049/43939 [45:14:24<25:08:21, 5.70s/it]
64%|██████▍ | 28050/43939 [45:14:30<25:09:02, 5.70s/it]
{'loss': 0.7791, 'grad_norm': 1.2950514554977417, 'learning_rate': 1.7409939851825908e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.06, 'epoch': 2.55}
64%|██████▍ | 28050/43939 [45:14:30<25:09:02, 5.70s/it]
64%|██████▍ | 28051/43939 [45:14:36<25:27:24, 5.77s/it]
64%|██████▍ | 28052/43939 [45:14:42<25:21:28, 5.75s/it]
64%|██████▍ | 28053/43939 [45:14:47<25:14:18, 5.72s/it]
64%|██████▍ | 28054/43939 [45:14:53<25:13:24, 5.72s/it]
64%|██████▍ | 28055/43939 [45:14:59<25:21:49, 5.75s/it]
64%|██████▍ | 28056/43939 [45:15:04<25:18:48, 5.74s/it]
64%|██████▍ | 28057/43939 [45:15:10<25:18:05, 5.74s/it]
64%|██████▍ | 28058/43939 [45:15:16<25:15:27, 5.73s/it]
64%|██████▍ | 28059/43939 [45:15:22<25:18:00, 5.74s/it]
64%|██████▍ | 28060/43939 [45:15:27<25:17:17, 5.73s/it]
64%|██████▍ | 28061/43939 [45:15:33<25:21:04, 5.75s/it]
64%|██████▍ | 28062/43939 [45:15:39<25:19:02, 5.74s/it]
64%|██████▍ | 28063/43939 [45:15:45<25:18:09, 5.74s/it]
64%|██████▍ | 28064/43939 [45:15:50<25:16:17, 5.73s/it]
64%|██████▍ | 28065/43939 [45:15:56<25:13:22, 5.72s/it]
64%|██████▍ | 28066/43939 [45:16:02<25:13:58, 5.72s/it]
64%|██████▍ | 28067/43939 [45:16:07<25:16:30, 5.73s/it]
64%|██████▍ | 28068/43939 [45:16:13<25:12:49, 5.72s/it]
64%|██████▍ | 28069/43939 [45:16:19<25:07:59, 5.70s/it]
64%|██████▍ | 28070/43939 [45:16:25<25:07:52, 5.70s/it]
64%|██████▍ | 28071/43939 [45:16:30<25:08:41, 5.70s/it]
64%|██████▍ | 28072/43939 [45:16:36<25:07:34, 5.70s/it]
64%|██████▍ | 28073/43939 [45:16:42<25:25:59, 5.77s/it]
64%|██████▍ | 28074/43939 [45:16:48<25:18:04, 5.74s/it]
64%|██████▍ | 28075/43939 [45:16:53<25:15:18, 5.73s/it]
64%|██████▍ | 28076/43939 [45:16:59<25:10:13, 5.71s/it]
64%|██████▍ | 28077/43939 [45:17:05<25:11:53, 5.72s/it]
64%|██████▍ | 28078/43939 [45:17:10<25:10:03, 5.71s/it]
64%|██████▍ | 28079/43939 [45:17:16<25:08:49, 5.71s/it]
64%|██████▍ | 28080/43939 [45:17:22<25:06:55, 5.70s/it]
64%|██████▍ | 28081/43939 [45:17:27<25:05:55, 5.70s/it]
64%|██████▍ | 28082/43939 [45:17:33<25:05:09, 5.70s/it]
64%|██████▍ | 28083/43939 [45:17:39<25:05:13, 5.70s/it]
64%|██████▍ | 28084/43939 [45:17:44<25:02:59, 5.69s/it]
64%|██████▍ | 28085/43939 [45:17:50<25:06:39, 5.70s/it]
64%|██████▍ | 28086/43939 [45:17:56<25:07:29, 5.71s/it]
64%|██████▍ | 28087/43939 [45:18:02<25:07:19, 5.71s/it]
64%|██████▍ | 28088/43939 [45:18:07<25:09:37, 5.71s/it]
64%|██████▍ | 28089/43939 [45:18:13<25:08:35, 5.71s/it]
64%|██████▍ | 28090/43939 [45:18:19<25:07:44, 5.71s/it]
64%|██████▍ | 28091/43939 [45:18:24<25:06:25, 5.70s/it]
64%|██████▍ | 28092/43939 [45:18:30<25:09:11, 5.71s/it]
64%|██████▍ | 28093/43939 [45:18:36<25:08:14, 5.71s/it]
64%|██████▍ | 28094/43939 [45:18:42<25:22:48, 5.77s/it]
64%|██████▍ | 28095/43939 [45:18:47<25:16:38, 5.74s/it]
64%|██████▍ | 28096/43939 [45:18:53<25:11:51, 5.73s/it]
64%|██████▍ | 28097/43939 [45:18:59<25:12:06, 5.73s/it]
64%|██████▍ | 28098/43939 [45:19:05<25:10:40, 5.72s/it]
64%|██████▍ | 28099/43939 [45:19:10<25:12:30, 5.73s/it]
64%|██████▍ | 28100/43939 [45:19:16<25:08:27, 5.71s/it]
{'loss': 0.7746, 'grad_norm': 1.6050091981887817, 'learning_rate': 1.7315385294722555e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.37, 'epoch': 2.56}
64%|██████▍ | 28100/43939 [45:19:16<25:08:27, 5.71s/it]
64%|██████▍ | 28101/43939 [45:19:22<25:08:37, 5.72s/it]
64%|██████▍ | 28102/43939 [45:19:27<25:06:19, 5.71s/it]
64%|██████▍ | 28103/43939 [45:19:33<25:09:05, 5.72s/it]
64%|██████▍ | 28104/43939 [45:19:39<25:06:53, 5.71s/it]
64%|██████▍ | 28105/43939 [45:19:45<25:06:04, 5.71s/it]
64%|██████▍ | 28106/43939 [45:19:50<25:07:56, 5.71s/it]
64%|██████▍ | 28107/43939 [45:19:56<25:05:40, 5.71s/it]
64%|██████▍ | 28108/43939 [45:20:02<25:01:41, 5.69s/it]
64%|██████▍ | 28109/43939 [45:20:07<25:01:11, 5.69s/it]
64%|██████▍ | 28110/43939 [45:20:13<25:03:50, 5.70s/it]
64%|██████▍ | 28111/43939 [45:20:19<25:02:13, 5.69s/it]
64%|██████▍ | 28112/43939 [45:20:24<25:06:00, 5.71s/it]
64%|██████▍ | 28113/43939 [45:20:30<25:07:12, 5.71s/it]
64%|██████▍ | 28114/43939 [45:20:36<25:06:11, 5.71s/it]
64%|██████▍ | 28115/43939 [45:20:42<25:21:35, 5.77s/it]
64%|██████▍ | 28116/43939 [45:20:48<25:16:27, 5.75s/it]
64%|██████▍ | 28117/43939 [45:20:53<25:11:48, 5.73s/it]
64%|██████▍ | 28118/43939 [45:20:59<25:07:50, 5.72s/it]
64%|██████▍ | 28119/43939 [45:21:05<25:04:11, 5.70s/it]
64%|██████▍ | 28120/43939 [45:21:10<25:02:07, 5.70s/it]
64%|██████▍ | 28121/43939 [45:21:16<25:00:22, 5.69s/it]
64%|██████▍ | 28122/43939 [45:21:22<25:02:10, 5.70s/it]
64%|██████▍ | 28123/43939 [45:21:27<25:02:09, 5.70s/it]
64%|██████▍ | 28124/43939 [45:21:33<25:02:02, 5.70s/it]
64%|██████▍ | 28125/43939 [45:21:39<25:01:32, 5.70s/it]
64%|██████▍ | 28126/43939 [45:21:44<25:01:26, 5.70s/it]
64%|██████▍ | 28127/43939 [45:21:50<25:02:36, 5.70s/it]
64%|██████▍ | 28128/43939 [45:21:56<25:01:00, 5.70s/it]
64%|██████▍ | 28129/43939 [45:22:02<25:00:44, 5.70s/it]
64%|██████▍ | 28130/43939 [45:22:07<25:02:02, 5.70s/it]
64%|██████▍ | 28131/43939 [45:22:13<24:59:28, 5.69s/it]
64%|██████▍ | 28132/43939 [45:22:19<25:01:43, 5.70s/it]
64%|██████▍ | 28133/43939 [45:22:24<24:59:10, 5.69s/it]
64%|██████▍ | 28134/43939 [45:22:30<24:59:04, 5.69s/it]
64%|██████▍ | 28135/43939 [45:22:36<24:59:40, 5.69s/it]
64%|██████▍ | 28136/43939 [45:22:42<25:21:04, 5.78s/it]
64%|██████▍ | 28137/43939 [45:22:47<25:16:25, 5.76s/it]
64%|██████▍ | 28138/43939 [45:22:53<25:12:35, 5.74s/it]
64%|██████▍ | 28139/43939 [45:22:59<25:07:40, 5.73s/it]
64%|██████▍ | 28140/43939 [45:23:04<25:05:40, 5.72s/it]
64%|██████▍ | 28141/43939 [45:23:10<25:04:57, 5.72s/it]
64%|██████▍ | 28142/43939 [45:23:16<25:06:14, 5.72s/it]
64%|██████▍ | 28143/43939 [45:23:22<25:06:27, 5.72s/it]
64%|██████▍ | 28144/43939 [45:23:27<25:02:34, 5.71s/it]
64%|██████▍ | 28145/43939 [45:23:33<25:02:17, 5.71s/it]
64%|██████▍ | 28146/43939 [45:23:39<25:00:47, 5.70s/it]
64%|██████▍ | 28147/43939 [45:23:44<25:01:56, 5.71s/it]
64%|██████▍ | 28148/43939 [45:23:50<25:02:59, 5.71s/it]
64%|██████▍ | 28149/43939 [45:23:56<24:58:50, 5.70s/it]
64%|██████▍ | 28150/43939 [45:24:01<24:57:29, 5.69s/it]
{'loss': 0.7847, 'grad_norm': 1.3216919898986816, 'learning_rate': 1.722095198047477e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.94, 'epoch': 2.56}
64%|██████▍ | 28150/43939 [45:24:01<24:57:29, 5.69s/it]
64%|██████▍ | 28151/43939 [45:24:07<24:59:33, 5.70s/it]
64%|██████▍ | 28152/43939 [45:24:13<25:00:00, 5.70s/it]
64%|██████▍ | 28153/43939 [45:24:19<24:59:45, 5.70s/it]
64%|██████▍ | 28154/43939 [45:24:24<25:01:21, 5.71s/it]
64%|██████▍ | 28155/43939 [45:24:30<25:00:04, 5.70s/it]
64%|██████▍ | 28156/43939 [45:24:36<24:57:21, 5.69s/it]
64%|██████▍ | 28157/43939 [45:24:41<24:58:23, 5.70s/it]
64%|██████▍ | 28158/43939 [45:24:47<25:15:39, 5.76s/it]
64%|██████▍ | 28159/43939 [45:24:53<25:08:07, 5.73s/it]
64%|██████▍ | 28160/43939 [45:24:59<25:05:06, 5.72s/it]
64%|██████▍ | 28161/43939 [45:25:04<25:06:04, 5.73s/it]
64%|██████▍ | 28162/43939 [45:25:10<25:08:42, 5.74s/it]
64%|██████▍ | 28163/43939 [45:25:16<25:04:37, 5.72s/it]
64%|██████▍ | 28164/43939 [45:25:22<25:03:12, 5.72s/it]
64%|██████▍ | 28165/43939 [45:25:27<25:00:00, 5.71s/it]
64%|██████▍ | 28166/43939 [45:25:33<25:00:15, 5.71s/it]
64%|██████▍ | 28167/43939 [45:25:39<24:59:29, 5.70s/it]
64%|██████▍ | 28168/43939 [45:25:44<24:56:41, 5.69s/it]
64%|██████▍ | 28169/43939 [45:25:50<24:55:11, 5.69s/it]
64%|██████▍ | 28170/43939 [45:25:56<24:54:29, 5.69s/it]
64%|██████▍ | 28171/43939 [45:26:01<24:57:21, 5.70s/it]
64%|██████▍ | 28172/43939 [45:26:07<25:00:36, 5.71s/it]
64%|██████▍ | 28173/43939 [45:26:13<25:01:17, 5.71s/it]
64%|██████▍ | 28174/43939 [45:26:19<24:59:38, 5.71s/it]
64%|██████▍ | 28175/43939 [45:26:24<24:58:43, 5.70s/it]
64%|██████▍ | 28176/43939 [45:26:30<24:58:44, 5.70s/it]
64%|██████▍ | 28177/43939 [45:26:36<25:01:39, 5.72s/it]
64%|██████▍ | 28178/43939 [45:26:41<24:58:05, 5.70s/it]
64%|██████▍ | 28179/43939 [45:26:47<25:15:10, 5.77s/it]
64%|██████▍ | 28180/43939 [45:26:53<25:09:27, 5.75s/it]
64%|██████▍ | 28181/43939 [45:26:59<25:03:22, 5.72s/it]
64%|██████▍ | 28182/43939 [45:27:04<25:01:31, 5.72s/it]
64%|██████▍ | 28183/43939 [45:27:10<25:00:12, 5.71s/it]
64%|██████▍ | 28184/43939 [45:27:16<24:58:07, 5.71s/it]
64%|██████▍ | 28185/43939 [45:27:21<24:53:30, 5.69s/it]
64%|██████▍ | 28186/43939 [45:27:27<24:53:02, 5.69s/it]
64%|██████▍ | 28187/43939 [45:27:33<25:00:43, 5.72s/it]
64%|██████▍ | 28188/43939 [45:27:39<24:59:22, 5.71s/it]
64%|██████▍ | 28189/43939 [45:27:44<25:02:05, 5.72s/it]
64%|██████▍ | 28190/43939 [45:27:50<25:00:03, 5.71s/it]
64%|██████▍ | 28191/43939 [45:27:56<24:56:51, 5.70s/it]
64%|██████▍ | 28192/43939 [45:28:01<24:55:50, 5.70s/it]
64%|██████▍ | 28193/43939 [45:28:07<24:56:52, 5.70s/it]
64%|██████▍ | 28194/43939 [45:28:13<24:55:30, 5.70s/it]
64%|██████▍ | 28195/43939 [45:28:18<24:53:56, 5.69s/it]
64%|██████▍ | 28196/43939 [45:28:24<24:54:22, 5.70s/it]
64%|██████▍ | 28197/43939 [45:28:30<24:52:57, 5.69s/it]
64%|██████▍ | 28198/43939 [45:28:36<24:54:38, 5.70s/it]
64%|██████▍ | 28199/43939 [45:28:41<24:59:13, 5.71s/it]
64%|██████▍ | 28200/43939 [45:28:47<24:59:44, 5.72s/it]
{'loss': 0.758, 'grad_norm': 1.2275587320327759, 'learning_rate': 1.7126641398990013e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.97, 'epoch': 2.57}
64%|██████▍ | 28200/43939 [45:28:47<24:59:44, 5.72s/it]
64%|██████▍ | 28201/43939 [45:28:53<25:12:46, 5.77s/it]
64%|██████▍ | 28202/43939 [45:28:59<25:06:34, 5.74s/it]
64%|██████▍ | 28203/43939 [45:29:04<25:05:39, 5.74s/it]
64%|██████▍ | 28204/43939 [45:29:10<25:00:08, 5.72s/it]
64%|██████▍ | 28205/43939 [45:29:16<24:56:57, 5.71s/it]
64%|██████▍ | 28206/43939 [45:29:21<24:56:09, 5.71s/it]
64%|██████▍ | 28207/43939 [45:29:27<24:55:06, 5.70s/it]
64%|██████▍ | 28208/43939 [45:29:33<24:55:24, 5.70s/it]
64%|██████▍ | 28209/43939 [45:29:38<24:53:00, 5.69s/it]
64%|██████▍ | 28210/43939 [45:29:44<24:59:47, 5.72s/it]
64%|██████▍ | 28211/43939 [45:29:50<24:56:05, 5.71s/it]
64%|██████▍ | 28212/43939 [45:29:56<24:59:46, 5.72s/it]
64%|██████▍ | 28213/43939 [45:30:01<24:54:35, 5.70s/it]
64%|██████▍ | 28214/43939 [45:30:07<24:55:08, 5.70s/it]
64%|██████▍ | 28215/43939 [45:30:13<24:59:05, 5.72s/it]
64%|██████▍ | 28216/43939 [45:30:18<24:57:33, 5.71s/it]
64%|██████▍ | 28217/43939 [45:30:24<24:57:25, 5.71s/it]
64%|██████▍ | 28218/43939 [45:30:30<24:55:42, 5.71s/it]
64%|██████▍ | 28219/43939 [45:30:36<24:55:01, 5.71s/it]
64%|██████▍ | 28220/43939 [45:30:41<24:52:11, 5.70s/it]
64%|██████▍ | 28221/43939 [45:30:47<24:51:48, 5.69s/it]
64%|██████▍ | 28222/43939 [45:30:53<25:09:56, 5.76s/it]
64%|██████▍ | 28223/43939 [45:30:59<25:04:15, 5.74s/it]
64%|██████▍ | 28224/43939 [45:31:04<24:59:19, 5.72s/it]
64%|██████▍ | 28225/43939 [45:31:10<24:57:22, 5.72s/it]
64%|██████▍ | 28226/43939 [45:31:16<24:55:23, 5.71s/it]
64%|██████▍ | 28227/43939 [45:31:21<24:52:25, 5.70s/it]
64%|██████▍ | 28228/43939 [45:31:27<24:49:16, 5.69s/it]
64%|██████▍ | 28229/43939 [45:31:33<24:51:45, 5.70s/it]
64%|██████▍ | 28230/43939 [45:31:38<24:50:57, 5.69s/it]
64%|██████▍ | 28231/43939 [45:31:44<24:51:06, 5.70s/it]
64%|██████▍ | 28232/43939 [45:31:50<24:55:02, 5.71s/it]
64%|██████▍ | 28233/43939 [45:31:56<24:54:26, 5.71s/it]
64%|██████▍ | 28234/43939 [45:32:01<24:54:22, 5.71s/it]
64%|██████▍ | 28235/43939 [45:32:07<24:52:20, 5.70s/it]
64%|██████▍ | 28236/43939 [45:32:13<24:51:42, 5.70s/it]
64%|██████▍ | 28237/43939 [45:32:18<24:55:08, 5.71s/it]
64%|██████▍ | 28238/43939 [45:32:24<24:53:28, 5.71s/it]
64%|██████▍ | 28239/43939 [45:32:30<24:54:57, 5.71s/it]
64%|██████▍ | 28240/43939 [45:32:36<24:54:07, 5.71s/it]
64%|██████▍ | 28241/43939 [45:32:41<24:54:42, 5.71s/it]
64%|██████▍ | 28242/43939 [45:32:47<24:54:41, 5.71s/it]
64%|██████▍ | 28243/43939 [45:32:53<25:15:26, 5.79s/it]
64%|██████▍ | 28244/43939 [45:32:59<25:09:37, 5.77s/it]
64%|██████▍ | 28245/43939 [45:33:04<25:01:11, 5.74s/it]
64%|██████▍ | 28246/43939 [45:33:10<24:55:24, 5.72s/it]
64%|██████▍ | 28247/43939 [45:33:16<24:56:34, 5.72s/it]
64%|██████▍ | 28248/43939 [45:33:21<24:54:12, 5.71s/it]
64%|██████▍ | 28249/43939 [45:33:27<24:54:19, 5.71s/it]
64%|██████▍ | 28250/43939 [45:33:33<24:53:14, 5.71s/it]
{'loss': 0.7785, 'grad_norm': 1.3859919309616089, 'learning_rate': 1.703245503823932e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.57, 'epoch': 2.57}
64%|██████▍ | 28250/43939 [45:33:33<24:53:14, 5.71s/it]
64%|██████▍ | 28251/43939 [45:33:39<24:56:53, 5.72s/it]
64%|██████▍ | 28252/43939 [45:33:44<24:52:31, 5.71s/it]
64%|██████▍ | 28253/43939 [45:33:50<24:54:21, 5.72s/it]
64%|██████▍ | 28254/43939 [45:33:56<24:53:20, 5.71s/it]
64%|██████▍ | 28255/43939 [45:34:01<24:53:42, 5.71s/it]
64%|██████▍ | 28256/43939 [45:34:07<24:51:08, 5.70s/it]
64%|██████▍ | 28257/43939 [45:34:13<24:50:55, 5.70s/it]
64%|██████▍ | 28258/43939 [45:34:19<24:54:39, 5.72s/it]
64%|██████▍ | 28259/43939 [45:34:24<24:53:59, 5.72s/it]
64%|██████▍ | 28260/43939 [45:34:30<25:00:22, 5.74s/it]
64%|██████▍ | 28261/43939 [45:34:36<24:56:30, 5.73s/it]
64%|██████▍ | 28262/43939 [45:34:41<24:53:26, 5.72s/it]
64%|██████▍ | 28263/43939 [45:34:47<24:53:09, 5.72s/it]
64%|██████▍ | 28264/43939 [45:34:53<25:05:50, 5.76s/it]
64%|██████▍ | 28265/43939 [45:34:59<24:57:33, 5.73s/it]
64%|██████▍ | 28266/43939 [45:35:04<24:55:22, 5.72s/it]
64%|██████▍ | 28267/43939 [45:35:10<24:53:29, 5.72s/it]
64%|██████▍ | 28268/43939 [45:35:16<24:52:31, 5.71s/it]
64%|██████▍ | 28269/43939 [45:35:21<24:49:31, 5.70s/it]
64%|██████▍ | 28270/43939 [45:35:27<24:46:55, 5.69s/it]
64%|██████▍ | 28271/43939 [45:35:33<24:50:19, 5.71s/it]
64%|██████▍ | 28272/43939 [45:35:39<24:51:09, 5.71s/it]
64%|██████▍ | 28273/43939 [45:35:44<24:52:18, 5.72s/it]
64%|██████▍ | 28274/43939 [45:35:50<24:48:24, 5.70s/it]
64%|██████▍ | 28275/43939 [45:35:56<24:48:37, 5.70s/it]
64%|██████▍ | 28276/43939 [45:36:01<24:48:21, 5.70s/it]
64%|██████▍ | 28277/43939 [45:36:07<24:46:59, 5.70s/it]
64%|██████▍ | 28278/43939 [45:36:13<24:49:38, 5.71s/it]
64%|██████▍ | 28279/43939 [45:36:18<24:46:53, 5.70s/it]
64%|██████▍ | 28280/43939 [45:36:24<24:47:59, 5.70s/it]
64%|██████▍ | 28281/43939 [45:36:30<24:50:11, 5.71s/it]
64%|██████▍ | 28282/43939 [45:36:36<24:49:22, 5.71s/it]
64%|██████▍ | 28283/43939 [45:36:41<24:47:08, 5.70s/it]
64%|██████▍ | 28284/43939 [45:36:47<24:47:37, 5.70s/it]
64%|██████▍ | 28285/43939 [45:36:53<24:47:45, 5.70s/it]
64%|██████▍ | 28286/43939 [45:36:59<25:08:02, 5.78s/it]
64%|██████▍ | 28287/43939 [45:37:04<25:06:38, 5.78s/it]
64%|██████▍ | 28288/43939 [45:37:10<24:57:12, 5.74s/it]
64%|██████▍ | 28289/43939 [45:37:16<24:52:42, 5.72s/it]
64%|██████▍ | 28290/43939 [45:37:21<24:50:28, 5.71s/it]
64%|██████▍ | 28291/43939 [45:37:27<24:49:46, 5.71s/it]
64%|██████▍ | 28292/43939 [45:37:33<24:53:18, 5.73s/it]
64%|██████▍ | 28293/43939 [45:37:39<24:51:51, 5.72s/it]
64%|██████▍ | 28294/43939 [45:37:44<24:50:40, 5.72s/it]
64%|██████▍ | 28295/43939 [45:37:50<24:47:33, 5.71s/it]
64%|██████▍ | 28296/43939 [45:37:56<24:43:25, 5.69s/it]
64%|██████▍ | 28297/43939 [45:38:01<24:43:43, 5.69s/it]
64%|██████▍ | 28298/43939 [45:38:07<24:46:38, 5.70s/it]
64%|██████▍ | 28299/43939 [45:38:13<24:46:05, 5.70s/it]
64%|██████▍ | 28300/43939 [45:38:19<24:44:23, 5.69s/it]
{'loss': 0.7939, 'grad_norm': 1.4800488948822021, 'learning_rate': 1.6938394384233878e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 520.75, 'epoch': 2.58}
64%|██████▍ | 28300/43939 [45:38:19<24:44:23, 5.69s/it]
64%|██████▍ | 28301/43939 [45:38:24<24:49:58, 5.72s/it]
64%|██████▍ | 28302/43939 [45:38:30<24:45:42, 5.70s/it]
64%|██████▍ | 28303/43939 [45:38:36<24:44:15, 5.70s/it]
64%|██████▍ | 28304/43939 [45:38:41<24:41:31, 5.69s/it]
64%|██████▍ | 28305/43939 [45:38:47<24:42:34, 5.69s/it]
64%|██████▍ | 28306/43939 [45:38:53<24:41:47, 5.69s/it]
64%|██████▍ | 28307/43939 [45:38:59<25:01:56, 5.76s/it]
64%|██████▍ | 28308/43939 [45:39:04<24:57:05, 5.75s/it]
64%|██████▍ | 28309/43939 [45:39:10<24:50:30, 5.72s/it]
64%|██████▍ | 28310/43939 [45:39:16<24:45:34, 5.70s/it]
64%|██████▍ | 28311/43939 [45:39:21<24:44:33, 5.70s/it]
64%|██████▍ | 28312/43939 [45:39:27<24:43:06, 5.69s/it]
64%|██████▍ | 28313/43939 [45:39:33<24:41:48, 5.69s/it]
64%|██████▍ | 28314/43939 [45:39:38<24:46:52, 5.71s/it]
64%|██████▍ | 28315/43939 [45:39:44<24:49:01, 5.72s/it]
64%|██████▍ | 28316/43939 [45:39:50<24:49:19, 5.72s/it]
64%|██████▍ | 28317/43939 [45:39:56<24:49:01, 5.72s/it]
64%|██████▍ | 28318/43939 [45:40:01<24:48:37, 5.72s/it]
64%|██████▍ | 28319/43939 [45:40:07<24:44:23, 5.70s/it]
64%|██████▍ | 28320/43939 [45:40:13<24:45:52, 5.71s/it]
64%|██████▍ | 28321/43939 [45:40:18<24:44:30, 5.70s/it]
64%|██████▍ | 28322/43939 [45:40:24<24:40:42, 5.69s/it]
64%|██████▍ | 28323/43939 [45:40:30<24:41:52, 5.69s/it]
64%|██████▍ | 28324/43939 [45:40:35<24:41:22, 5.69s/it]
64%|██████▍ | 28325/43939 [45:40:41<24:42:29, 5.70s/it]
64%|██████▍ | 28326/43939 [45:40:47<24:39:58, 5.69s/it]
64%|██████▍ | 28327/43939 [45:40:53<24:39:31, 5.69s/it]
64%|██████▍ | 28328/43939 [45:40:58<25:01:17, 5.77s/it]
64%|██████▍ | 28329/43939 [45:41:04<24:55:35, 5.75s/it]
64%|██████▍ | 28330/43939 [45:41:10<24:55:55, 5.75s/it]
64%|██████▍ | 28331/43939 [45:41:16<24:50:04, 5.73s/it]
64%|██████▍ | 28332/43939 [45:41:21<24:47:19, 5.72s/it]
64%|██████▍ | 28333/43939 [45:41:27<24:47:03, 5.72s/it]
64%|██████▍ | 28334/43939 [45:41:33<24:49:12, 5.73s/it]
64%|██████▍ | 28335/43939 [45:41:38<24:47:04, 5.72s/it]
64%|██████▍ | 28336/43939 [45:41:44<24:43:26, 5.70s/it]
64%|██████▍ | 28337/43939 [45:41:50<24:44:05, 5.71s/it]
64%|██████▍ | 28338/43939 [45:41:56<24:42:17, 5.70s/it]
64%|██████▍ | 28339/43939 [45:42:01<24:43:15, 5.70s/it]
64%|██████▍ | 28340/43939 [45:42:07<24:42:21, 5.70s/it]
65%|██████▍ | 28341/43939 [45:42:13<24:39:40, 5.69s/it]
65%|██████▍ | 28342/43939 [45:42:18<24:40:33, 5.70s/it]
65%|██████▍ | 28343/43939 [45:42:24<24:38:47, 5.69s/it]
65%|██████▍ | 28344/43939 [45:42:30<24:38:13, 5.69s/it]
65%|██████▍ | 28345/43939 [45:42:35<24:38:02, 5.69s/it]
65%|██████▍ | 28346/43939 [45:42:41<24:38:24, 5.69s/it]
65%|██████▍ | 28347/43939 [45:42:47<24:38:19, 5.69s/it]
65%|██████▍ | 28348/43939 [45:42:52<24:38:25, 5.69s/it]
65%|██████▍ | 28349/43939 [45:42:58<24:36:04, 5.68s/it]
65%|██████▍ | 28350/43939 [45:43:04<24:59:42, 5.77s/it]
{'loss': 0.7654, 'grad_norm': 1.154043436050415, 'learning_rate': 1.6844460921001534e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 445.36, 'epoch': 2.58}
65%|██████▍ | 28350/43939 [45:43:04<24:59:42, 5.77s/it]
65%|██████▍ | 28351/43939 [45:43:10<24:54:09, 5.75s/it]
65%|██████▍ | 28352/43939 [45:43:16<24:54:23, 5.75s/it]
65%|██████▍ | 28353/43939 [45:43:21<24:49:24, 5.73s/it]
65%|██████▍ | 28354/43939 [45:43:27<24:45:17, 5.72s/it]
65%|██████▍ | 28355/43939 [45:43:33<24:44:00, 5.71s/it]
65%|██████▍ | 28356/43939 [45:43:38<24:39:52, 5.70s/it]
65%|██████▍ | 28357/43939 [45:43:44<24:40:23, 5.70s/it]
65%|██████▍ | 28358/43939 [45:43:50<24:39:54, 5.70s/it]
65%|██████▍ | 28359/43939 [45:43:55<24:40:01, 5.70s/it]
65%|██████▍ | 28360/43939 [45:44:01<24:43:42, 5.71s/it]
65%|██████▍ | 28361/43939 [45:44:07<24:46:19, 5.72s/it]
65%|██████▍ | 28362/43939 [45:44:13<24:45:30, 5.72s/it]
65%|██████▍ | 28363/43939 [45:44:18<24:40:38, 5.70s/it]
65%|██████▍ | 28364/43939 [45:44:24<24:43:58, 5.72s/it]
65%|██████▍ | 28365/43939 [45:44:30<24:41:55, 5.71s/it]
65%|██████▍ | 28366/43939 [45:44:35<24:45:52, 5.72s/it]
65%|██████▍ | 28367/43939 [45:44:41<24:44:52, 5.72s/it]
65%|██████▍ | 28368/43939 [45:44:47<24:40:07, 5.70s/it]
65%|██████▍ | 28369/43939 [45:44:52<24:35:59, 5.69s/it]
65%|██████▍ | 28370/43939 [45:44:58<24:37:11, 5.69s/it]
65%|██████▍ | 28371/43939 [45:45:04<24:55:37, 5.76s/it]
65%|██████▍ | 28372/43939 [45:45:10<24:48:41, 5.74s/it]
65%|██████▍ | 28373/43939 [45:45:16<24:48:29, 5.74s/it]
65%|██████▍ | 28374/43939 [45:45:21<24:44:12, 5.72s/it]
65%|██████▍ | 28375/43939 [45:45:27<24:46:49, 5.73s/it]
65%|██████▍ | 28376/43939 [45:45:33<24:48:52, 5.74s/it]
65%|██████▍ | 28377/43939 [45:45:38<24:46:15, 5.73s/it]
65%|██████▍ | 28378/43939 [45:45:44<24:42:39, 5.72s/it]
65%|██████▍ | 28379/43939 [45:45:50<24:46:09, 5.73s/it]
65%|██████▍ | 28380/43939 [45:45:56<24:41:44, 5.71s/it]
65%|██████▍ | 28381/43939 [45:46:01<24:38:45, 5.70s/it]
65%|██████▍ | 28382/43939 [45:46:07<24:38:47, 5.70s/it]
65%|██████▍ | 28383/43939 [45:46:13<24:36:09, 5.69s/it]
65%|██████▍ | 28384/43939 [45:46:18<24:41:03, 5.71s/it]
65%|██████▍ | 28385/43939 [45:46:24<24:38:34, 5.70s/it]
65%|██████▍ | 28386/43939 [45:46:30<24:38:34, 5.70s/it]
65%|██████▍ | 28387/43939 [45:46:35<24:40:49, 5.71s/it]
65%|██████▍ | 28388/43939 [45:46:41<24:39:50, 5.71s/it]
65%|██████▍ | 28389/43939 [45:46:47<24:41:25, 5.72s/it]
65%|██████▍ | 28390/43939 [45:46:53<24:39:10, 5.71s/it]
65%|██████▍ | 28391/43939 [45:46:58<24:39:07, 5.71s/it]
65%|██████▍ | 28392/43939 [45:47:04<24:55:10, 5.77s/it]
65%|██████▍ | 28393/43939 [45:47:10<24:48:09, 5.74s/it]
65%|██████▍ | 28394/43939 [45:47:16<24:43:28, 5.73s/it]
65%|██████▍ | 28395/43939 [45:47:21<24:42:25, 5.72s/it]
65%|██████▍ | 28396/43939 [45:47:27<24:37:48, 5.70s/it]
65%|██████▍ | 28397/43939 [45:47:33<24:34:00, 5.69s/it]
65%|██████▍ | 28398/43939 [45:47:38<24:34:28, 5.69s/it]
65%|██████▍ | 28399/43939 [45:47:44<24:40:23, 5.72s/it]
65%|██████▍ | 28400/43939 [45:47:50<24:36:01, 5.70s/it]
{'loss': 0.7612, 'grad_norm': 1.4552162885665894, 'learning_rate': 1.6750656130563415e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.73, 'epoch': 2.59}
65%|██████▍ | 28400/43939 [45:47:50<24:36:01, 5.70s/it]
65%|██████▍ | 28401/43939 [45:47:55<24:37:13, 5.70s/it]
65%|██████▍ | 28402/43939 [45:48:01<24:35:30, 5.70s/it]
65%|██████▍ | 28403/43939 [45:48:07<24:37:18, 5.71s/it]
65%|██████▍ | 28404/43939 [45:48:13<24:35:30, 5.70s/it]
65%|██████▍ | 28405/43939 [45:48:18<24:35:18, 5.70s/it]
65%|██████▍ | 28406/43939 [45:48:24<24:34:46, 5.70s/it]
65%|██████▍ | 28407/43939 [45:48:30<24:37:15, 5.71s/it]
65%|██████▍ | 28408/43939 [45:48:35<24:36:59, 5.71s/it]
65%|██████▍ | 28409/43939 [45:48:41<24:34:57, 5.70s/it]
65%|██████▍ | 28410/43939 [45:48:47<24:36:05, 5.70s/it]
65%|██████▍ | 28411/43939 [45:48:52<24:34:02, 5.70s/it]
65%|██████▍ | 28412/43939 [45:48:58<24:35:07, 5.70s/it]
65%|██████▍ | 28413/43939 [45:49:04<24:51:57, 5.77s/it]
65%|██████▍ | 28414/43939 [45:49:10<24:45:59, 5.74s/it]
65%|██████▍ | 28415/43939 [45:49:16<24:43:46, 5.73s/it]
65%|██████▍ | 28416/43939 [45:49:21<24:38:19, 5.71s/it]
65%|██████▍ | 28417/43939 [45:49:27<24:42:37, 5.73s/it]
65%|██████▍ | 28418/43939 [45:49:33<24:41:06, 5.73s/it]
65%|██████▍ | 28419/43939 [45:49:38<24:42:37, 5.73s/it]
65%|██████▍ | 28420/43939 [45:49:44<24:43:16, 5.73s/it]
65%|██████▍ | 28421/43939 [45:49:50<24:40:08, 5.72s/it]
65%|██████▍ | 28422/43939 [45:49:56<24:40:37, 5.73s/it]
65%|██████▍ | 28423/43939 [45:50:01<24:37:32, 5.71s/it]
65%|██████▍ | 28424/43939 [45:50:07<24:37:09, 5.71s/it]
65%|██████▍ | 28425/43939 [45:50:13<24:35:33, 5.71s/it]
65%|██████▍ | 28426/43939 [45:50:18<24:33:48, 5.70s/it]
65%|██████▍ | 28427/43939 [45:50:24<24:32:23, 5.70s/it]
65%|██████▍ | 28428/43939 [45:50:30<24:31:23, 5.69s/it]
65%|██████▍ | 28429/43939 [45:50:35<24:33:16, 5.70s/it]
65%|██████▍ | 28430/43939 [45:50:41<24:35:59, 5.71s/it]
65%|██████▍ | 28431/43939 [45:50:47<24:34:38, 5.71s/it]
65%|██████▍ | 28432/43939 [45:50:53<24:32:55, 5.70s/it]
65%|██████▍ | 28433/43939 [45:50:58<24:36:42, 5.71s/it]
65%|██████▍ | 28434/43939 [45:51:04<24:34:04, 5.70s/it]
65%|██████▍ | 28435/43939 [45:51:10<24:49:58, 5.77s/it]
65%|██████▍ | 28436/43939 [45:51:16<24:44:23, 5.74s/it]
65%|██████▍ | 28437/43939 [45:51:21<24:43:45, 5.74s/it]
65%|██████▍ | 28438/43939 [45:51:27<24:39:15, 5.73s/it]
65%|██████▍ | 28439/43939 [45:51:33<24:39:31, 5.73s/it]
65%|██████▍ | 28440/43939 [45:51:38<24:36:16, 5.71s/it]
65%|██████▍ | 28441/43939 [45:51:44<24:36:06, 5.71s/it]
65%|██████▍ | 28442/43939 [45:51:50<24:34:10, 5.71s/it]
65%|██████▍ | 28443/43939 [45:51:55<24:31:10, 5.70s/it]
65%|██████▍ | 28444/43939 [45:52:01<24:29:07, 5.69s/it]
65%|██████▍ | 28445/43939 [45:52:07<24:29:59, 5.69s/it]
65%|██████▍ | 28446/43939 [45:52:13<24:33:27, 5.71s/it]
65%|██████▍ | 28447/43939 [45:52:18<24:34:02, 5.71s/it]
65%|██████▍ | 28448/43939 [45:52:24<24:32:10, 5.70s/it]
65%|██████▍ | 28449/43939 [45:52:30<24:32:59, 5.71s/it]
65%|██████▍ | 28450/43939 [45:52:35<24:29:34, 5.69s/it]
{'loss': 0.7607, 'grad_norm': 1.4212393760681152, 'learning_rate': 1.6656981492910527e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.16, 'epoch': 2.59}
65%|██████▍ | 28450/43939 [45:52:35<24:29:34, 5.69s/it]
65%|██████▍ | 28451/43939 [45:52:41<24:30:58, 5.70s/it]
65%|██████▍ | 28452/43939 [45:52:47<24:29:52, 5.69s/it]
65%|██████▍ | 28453/43939 [45:52:53<24:32:55, 5.71s/it]
65%|██████▍ | 28454/43939 [45:52:58<24:33:13, 5.71s/it]
65%|██████▍ | 28455/43939 [45:53:04<24:31:58, 5.70s/it]
65%|██████▍ | 28456/43939 [45:53:10<24:51:57, 5.78s/it]
65%|██████▍ | 28457/43939 [45:53:16<24:44:45, 5.75s/it]
65%|██████▍ | 28458/43939 [45:53:21<24:39:56, 5.74s/it]
65%|██████▍ | 28459/43939 [45:53:27<24:36:32, 5.72s/it]
65%|██████▍ | 28460/43939 [45:53:33<24:35:01, 5.72s/it]
65%|██████▍ | 28461/43939 [45:53:38<24:33:46, 5.71s/it]
65%|██████▍ | 28462/43939 [45:53:44<24:37:25, 5.73s/it]
65%|██████▍ | 28463/43939 [45:53:50<24:33:27, 5.71s/it]
65%|██████▍ | 28464/43939 [45:53:56<24:32:17, 5.71s/it]
65%|██████▍ | 28465/43939 [45:54:01<24:28:47, 5.70s/it]
65%|██████▍ | 28466/43939 [45:54:07<24:30:02, 5.70s/it]
65%|██████▍ | 28467/43939 [45:54:13<24:30:51, 5.70s/it]
65%|██████▍ | 28468/43939 [45:54:18<24:32:39, 5.71s/it]
65%|██████▍ | 28469/43939 [45:54:24<24:28:43, 5.70s/it]
65%|██████▍ | 28470/43939 [45:54:30<24:28:00, 5.69s/it]
65%|██████▍ | 28471/43939 [45:54:35<24:26:33, 5.69s/it]
65%|██████▍ | 28472/43939 [45:54:41<24:25:43, 5.69s/it]
65%|██████▍ | 28473/43939 [45:54:47<24:27:01, 5.69s/it]
65%|██████▍ | 28474/43939 [45:54:52<24:31:16, 5.71s/it]
65%|██████▍ | 28475/43939 [45:54:58<24:33:33, 5.72s/it]
65%|██████▍ | 28476/43939 [45:55:04<24:32:07, 5.71s/it]
65%|██████▍ | 28477/43939 [45:55:10<24:47:58, 5.77s/it]
65%|██████▍ | 28478/43939 [45:55:16<24:45:16, 5.76s/it]
65%|██████▍ | 28479/43939 [45:55:21<24:45:06, 5.76s/it]
65%|██████▍ | 28480/43939 [45:55:27<24:40:06, 5.74s/it]
65%|██████▍ | 28481/43939 [45:55:33<24:37:08, 5.73s/it]
65%|██████▍ | 28482/43939 [45:55:38<24:33:10, 5.72s/it]
65%|██████▍ | 28483/43939 [45:55:44<24:34:00, 5.72s/it]
65%|██████▍ | 28484/43939 [45:55:50<24:30:15, 5.71s/it]
65%|██████▍ | 28485/43939 [45:55:56<24:29:18, 5.70s/it]
65%|██████▍ | 28486/43939 [45:56:01<24:26:47, 5.70s/it]
65%|██████▍ | 28487/43939 [45:56:07<24:25:39, 5.69s/it]
65%|██████▍ | 28488/43939 [45:56:13<24:26:43, 5.70s/it]
65%|██████▍ | 28489/43939 [45:56:18<24:26:27, 5.70s/it]
65%|██████▍ | 28490/43939 [45:56:24<24:27:14, 5.70s/it]
65%|██████▍ | 28491/43939 [45:56:30<24:26:34, 5.70s/it]
65%|██████▍ | 28492/43939 [45:56:35<24:25:22, 5.69s/it]
65%|██████▍ | 28493/43939 [45:56:41<24:24:14, 5.69s/it]
65%|██████▍ | 28494/43939 [45:56:47<24:25:14, 5.69s/it]
65%|██████▍ | 28495/43939 [45:56:52<24:26:14, 5.70s/it]
65%|██████▍ | 28496/43939 [45:56:58<24:28:49, 5.71s/it]
65%|██████▍ | 28497/43939 [45:57:04<24:36:24, 5.74s/it]
65%|██████▍ | 28498/43939 [45:57:10<24:49:36, 5.79s/it]
65%|██████▍ | 28499/43939 [45:57:16<24:43:47, 5.77s/it]
65%|██████▍ | 28500/43939 [45:57:21<24:40:50, 5.75s/it]
{'loss': 0.8062, 'grad_norm': 1.3629285097122192, 'learning_rate': 1.6563438485980415e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 523.95, 'epoch': 2.59}
65%|██████▍ | 28500/43939 [45:57:21<24:40:50, 5.75s/it]
65%|██████▍ | 28501/43939 [45:57:27<24:35:19, 5.73s/it]
65%|██████▍ | 28502/43939 [45:57:33<24:34:04, 5.73s/it]
65%|██████▍ | 28503/43939 [45:57:38<24:29:49, 5.71s/it]
65%|██████▍ | 28504/43939 [45:57:44<24:26:22, 5.70s/it]
65%|██████▍ | 28505/43939 [45:57:50<24:26:27, 5.70s/it]
65%|██████▍ | 28506/43939 [45:57:55<24:25:04, 5.70s/it]
65%|██████▍ | 28507/43939 [45:58:01<24:32:10, 5.72s/it]
65%|██████▍ | 28508/43939 [45:58:07<24:28:52, 5.71s/it]
65%|██████▍ | 28509/43939 [45:58:13<24:28:36, 5.71s/it]
65%|██████▍ | 28510/43939 [45:58:18<24:27:12, 5.71s/it]
65%|██████▍ | 28511/43939 [45:58:24<24:22:32, 5.69s/it]
65%|██████▍ | 28512/43939 [45:58:30<24:22:56, 5.69s/it]
65%|██████▍ | 28513/43939 [45:58:35<24:23:29, 5.69s/it]
65%|██████▍ | 28514/43939 [45:58:41<24:24:15, 5.70s/it]
65%|██████▍ | 28515/43939 [45:58:47<24:24:58, 5.70s/it]
65%|██████▍ | 28516/43939 [45:58:53<24:25:26, 5.70s/it]
65%|██████▍ | 28517/43939 [45:58:58<24:22:52, 5.69s/it]
65%|██████▍ | 28518/43939 [45:59:04<24:25:32, 5.70s/it]
65%|██████▍ | 28519/43939 [45:59:10<24:29:01, 5.72s/it]
65%|██████▍ | 28520/43939 [45:59:16<24:45:48, 5.78s/it]
65%|██████▍ | 28521/43939 [45:59:21<24:39:19, 5.76s/it]
65%|██████▍ | 28522/43939 [45:59:27<24:35:27, 5.74s/it]
65%|██████▍ | 28523/43939 [45:59:33<24:32:04, 5.73s/it]
65%|██████▍ | 28524/43939 [45:59:38<24:29:47, 5.72s/it]
65%|██████▍ | 28525/43939 [45:59:44<24:26:01, 5.71s/it]
65%|██████▍ | 28526/43939 [45:59:50<24:29:06, 5.72s/it]
65%|██████▍ | 28527/43939 [45:59:56<24:28:10, 5.72s/it]
65%|██████▍ | 28528/43939 [46:00:01<24:22:52, 5.70s/it]
65%|██████▍ | 28529/43939 [46:00:07<24:22:36, 5.69s/it]
65%|██████▍ | 28530/43939 [46:00:13<24:23:02, 5.70s/it]
65%|██████▍ | 28531/43939 [46:00:18<24:25:07, 5.71s/it]
65%|██████▍ | 28532/43939 [46:00:24<24:24:53, 5.70s/it]
65%|██████▍ | 28533/43939 [46:00:30<24:24:57, 5.71s/it]
65%|██████▍ | 28534/43939 [46:00:35<24:25:25, 5.71s/it]
65%|██████▍ | 28535/43939 [46:00:41<24:23:18, 5.70s/it]
65%|██████▍ | 28536/43939 [46:00:47<24:21:50, 5.69s/it]
65%|██████▍ | 28537/43939 [46:00:52<24:20:05, 5.69s/it]
65%|██████▍ | 28538/43939 [46:00:58<24:18:57, 5.68s/it]
65%|██████▍ | 28539/43939 [46:01:04<24:25:00, 5.71s/it]
65%|██████▍ | 28540/43939 [46:01:10<24:23:51, 5.70s/it]
65%|██████▍ | 28541/43939 [46:01:15<24:27:04, 5.72s/it]
65%|██████▍ | 28542/43939 [46:01:21<24:45:13, 5.79s/it]
65%|██████▍ | 28543/43939 [46:01:27<24:38:43, 5.76s/it]
65%|██████▍ | 28544/43939 [46:01:33<24:33:24, 5.74s/it]
65%|██████▍ | 28545/43939 [46:01:38<24:34:41, 5.75s/it]
65%|██████▍ | 28546/43939 [46:01:44<24:34:09, 5.75s/it]
65%|██████▍ | 28547/43939 [46:01:50<24:32:08, 5.74s/it]
65%|██████▍ | 28548/43939 [46:01:56<24:32:06, 5.74s/it]
65%|██████▍ | 28549/43939 [46:02:01<24:26:58, 5.72s/it]
65%|██████▍ | 28550/43939 [46:02:07<24:22:08, 5.70s/it]
{'loss': 0.781, 'grad_norm': 1.4396235942840576, 'learning_rate': 1.6470028585633833e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.82, 'epoch': 2.6}
65%|██████▍ | 28550/43939 [46:02:07<24:22:08, 5.70s/it]
65%|██████▍ | 28551/43939 [46:02:13<24:28:32, 5.73s/it]
65%|██████▍ | 28552/43939 [46:02:18<24:25:22, 5.71s/it]
65%|██████▍ | 28553/43939 [46:02:24<24:22:18, 5.70s/it]
65%|██████▍ | 28554/43939 [46:02:30<24:20:36, 5.70s/it]
65%|██████▍ | 28555/43939 [46:02:35<24:18:11, 5.69s/it]
65%|██████▍ | 28556/43939 [46:02:41<24:22:16, 5.70s/it]
65%|██████▍ | 28557/43939 [46:02:47<24:21:08, 5.70s/it]
65%|██████▍ | 28558/43939 [46:02:53<24:19:33, 5.69s/it]
65%|██████▍ | 28559/43939 [46:02:58<24:20:26, 5.70s/it]
65%|██████▍ | 28560/43939 [46:03:04<24:20:47, 5.70s/it]
65%|██████▌ | 28561/43939 [46:03:10<24:17:16, 5.69s/it][2025-11-30 16:07:23,472] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:45, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.72s/it][A
29%|██▉ | 21/72 [00:54<02:18, 2.72s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.72s/it][A
40%|████ | 29/72 [01:16<01:57, 2.72s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.72s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.72s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.72s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:48<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A{'eval_loss': 0.9801340699195862, 'eval_runtime': 197.5937, 'eval_samples_per_second': 5.056, 'eval_steps_per_second': 0.364, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 2.6}
65%|██████▌ | 28561/43939 [46:06:27<24:17:16, 5.69s/it]
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A
65%|██████▌ | 28562/43939 [46:06:33<278:02:23, 65.09s/it]
65%|██████▌ | 28563/43939 [46:06:39<201:56:40, 47.28s/it]
65%|██████▌ | 28564/43939 [46:06:45<148:39:33, 34.81s/it]
65%|██████▌ | 28565/43939 [46:06:51<111:23:29, 26.08s/it]
65%|██████▌ | 28566/43939 [46:06:56<85:16:58, 19.97s/it]
65%|██████▌ | 28567/43939 [46:07:02<66:59:13, 15.69s/it]
65%|██████▌ | 28568/43939 [46:07:08<54:14:59, 12.71s/it]
65%|██████▌ | 28569/43939 [46:07:13<45:17:06, 10.61s/it]
65%|██████▌ | 28570/43939 [46:07:19<39:00:33, 9.14s/it]
65%|██████▌ | 28571/43939 [46:07:25<34:37:04, 8.11s/it]
65%|██████▌ | 28572/43939 [46:07:31<31:35:08, 7.40s/it]
65%|██████▌ | 28573/43939 [46:07:36<29:23:51, 6.89s/it]
65%|██████▌ | 28574/43939 [46:07:42<27:52:50, 6.53s/it]
65%|██████▌ | 28575/43939 [46:07:48<26:48:53, 6.28s/it]
65%|██████▌ | 28576/43939 [46:07:53<26:02:51, 6.10s/it]
65%|██████▌ | 28577/43939 [46:07:59<25:28:00, 5.97s/it]
65%|██████▌ | 28578/43939 [46:08:05<25:06:06, 5.88s/it]
65%|██████▌ | 28579/43939 [46:08:10<24:47:44, 5.81s/it]
65%|██████▌ | 28580/43939 [46:08:16<24:38:08, 5.77s/it]
65%|██████▌ | 28581/43939 [46:08:22<24:32:04, 5.75s/it]
65%|██████▌ | 28582/43939 [46:08:27<24:27:17, 5.73s/it]
65%|██████▌ | 28583/43939 [46:08:33<24:23:49, 5.72s/it]
65%|██████▌ | 28584/43939 [46:08:39<24:43:29, 5.80s/it]
65%|██████▌ | 28585/43939 [46:08:45<24:36:21, 5.77s/it]
65%|██████▌ | 28586/43939 [46:08:50<24:31:37, 5.75s/it]
65%|██████▌ | 28587/43939 [46:08:56<24:27:58, 5.74s/it]
65%|██████▌ | 28588/43939 [46:09:02<24:30:26, 5.75s/it]
65%|██████▌ | 28589/43939 [46:09:08<24:29:54, 5.75s/it]
65%|██████▌ | 28590/43939 [46:09:13<24:26:53, 5.73s/it]
65%|██████▌ | 28591/43939 [46:09:19<24:25:04, 5.73s/it]
65%|██████▌ | 28592/43939 [46:09:25<24:21:56, 5.72s/it]
65%|██████▌ | 28593/43939 [46:09:30<24:21:08, 5.71s/it]
65%|██████▌ | 28594/43939 [46:09:36<24:20:05, 5.71s/it]
65%|██████▌ | 28595/43939 [46:09:42<24:22:18, 5.72s/it]
65%|██████▌ | 28596/43939 [46:09:48<24:25:07, 5.73s/it]
65%|██████▌ | 28597/43939 [46:09:53<24:27:26, 5.74s/it]
65%|██████▌ | 28598/43939 [46:09:59<24:25:52, 5.73s/it]
65%|██████▌ | 28599/43939 [46:10:05<24:24:39, 5.73s/it]
65%|██████▌ | 28600/43939 [46:10:11<24:20:39, 5.71s/it]
{'loss': 0.7819, 'grad_norm': 1.6099671125411987, 'learning_rate': 1.6376753265631474e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 377.83, 'epoch': 2.6}
65%|██████▌ | 28600/43939 [46:10:11<24:20:39, 5.71s/it]
65%|██████▌ | 28601/43939 [46:10:16<24:20:36, 5.71s/it]
65%|██████▌ | 28602/43939 [46:10:22<24:18:24, 5.71s/it]
65%|██████▌ | 28603/43939 [46:10:28<24:18:26, 5.71s/it]
65%|██████▌ | 28604/43939 [46:10:33<24:17:34, 5.70s/it]
65%|██████▌ | 28605/43939 [46:10:39<24:37:49, 5.78s/it]
65%|██████▌ | 28606/43939 [46:10:45<24:31:12, 5.76s/it]
65%|██████▌ | 28607/43939 [46:10:51<24:24:49, 5.73s/it]
65%|██████▌ | 28608/43939 [46:10:56<24:23:18, 5.73s/it]
65%|██████▌ | 28609/43939 [46:11:02<24:20:06, 5.71s/it]
65%|██████▌ | 28610/43939 [46:11:08<24:18:37, 5.71s/it]
65%|██████▌ | 28611/43939 [46:11:13<24:16:45, 5.70s/it]
65%|██████▌ | 28612/43939 [46:11:19<24:16:03, 5.70s/it]
65%|██████▌ | 28613/43939 [46:11:25<24:18:45, 5.71s/it]
65%|██████▌ | 28614/43939 [46:11:31<24:23:53, 5.73s/it]
65%|██████▌ | 28615/43939 [46:11:36<24:19:44, 5.72s/it]
65%|██████▌ | 28616/43939 [46:11:42<24:17:29, 5.71s/it]
65%|██████▌ | 28617/43939 [46:11:48<24:14:41, 5.70s/it]
65%|██████▌ | 28618/43939 [46:11:53<24:15:24, 5.70s/it]
65%|██████▌ | 28619/43939 [46:11:59<24:15:21, 5.70s/it]
65%|██████▌ | 28620/43939 [46:12:05<24:13:54, 5.69s/it]
65%|██████▌ | 28621/43939 [46:12:11<24:14:09, 5.70s/it]
65%|██████▌ | 28622/43939 [46:12:16<24:18:37, 5.71s/it]
65%|██████▌ | 28623/43939 [46:12:22<24:17:12, 5.71s/it]
65%|██████▌ | 28624/43939 [46:12:28<24:16:28, 5.71s/it]
65%|██████▌ | 28625/43939 [46:12:33<24:18:08, 5.71s/it]
65%|██████▌ | 28626/43939 [46:12:39<24:34:47, 5.78s/it]
65%|██████▌ | 28627/43939 [46:12:45<24:31:35, 5.77s/it]
65%|██████▌ | 28628/43939 [46:12:51<24:25:23, 5.74s/it]
65%|██████▌ | 28629/43939 [46:12:56<24:19:28, 5.72s/it]
65%|██████▌ | 28630/43939 [46:13:02<24:17:38, 5.71s/it]
65%|██████▌ | 28631/43939 [46:13:08<24:17:18, 5.71s/it]
65%|██████▌ | 28632/43939 [46:13:13<24:13:16, 5.70s/it]
65%|██████▌ | 28633/43939 [46:13:19<24:18:21, 5.72s/it]
65%|██████▌ | 28634/43939 [46:13:25<24:14:14, 5.70s/it]
65%|██████▌ | 28635/43939 [46:13:31<24:15:23, 5.71s/it]
65%|██████▌ | 28636/43939 [46:13:36<24:13:53, 5.70s/it]
65%|██████▌ | 28637/43939 [46:13:42<24:13:45, 5.70s/it]
65%|██████▌ | 28638/43939 [46:13:48<24:17:01, 5.71s/it]
65%|██████▌ | 28639/43939 [46:13:54<24:18:29, 5.72s/it]
65%|██████▌ | 28640/43939 [46:13:59<24:21:20, 5.73s/it]
65%|██████▌ | 28641/43939 [46:14:05<24:20:21, 5.73s/it]
65%|██████▌ | 28642/43939 [46:14:11<24:18:09, 5.72s/it]
65%|██████▌ | 28643/43939 [46:14:16<24:16:54, 5.71s/it]
65%|██████▌ | 28644/43939 [46:14:22<24:14:11, 5.70s/it]
65%|██████▌ | 28645/43939 [46:14:28<24:12:18, 5.70s/it]
65%|██████▌ | 28646/43939 [46:14:33<24:12:13, 5.70s/it]
65%|██████▌ | 28647/43939 [46:14:39<24:10:21, 5.69s/it]
65%|██████▌ | 28648/43939 [46:14:45<24:29:59, 5.77s/it]
65%|██████▌ | 28649/43939 [46:14:51<24:22:26, 5.74s/it]
65%|██████▌ | 28650/43939 [46:14:56<24:21:30, 5.74s/it]
{'loss': 0.7681, 'grad_norm': 1.198883056640625, 'learning_rate': 1.6283613997610696e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.68, 'epoch': 2.61}
65%|██████▌ | 28650/43939 [46:14:56<24:21:30, 5.74s/it]
65%|██████▌ | 28651/43939 [46:15:02<24:29:36, 5.77s/it]
65%|██████▌ | 28652/43939 [46:15:08<24:23:52, 5.75s/it]
65%|██████▌ | 28653/43939 [46:15:14<24:22:21, 5.74s/it]
65%|██████▌ | 28654/43939 [46:15:19<24:18:09, 5.72s/it]
65%|██████▌ | 28655/43939 [46:15:25<24:14:24, 5.71s/it]
65%|██████▌ | 28656/43939 [46:15:31<24:14:05, 5.71s/it]
65%|██████▌ | 28657/43939 [46:15:37<24:13:39, 5.71s/it]
65%|██████▌ | 28658/43939 [46:15:42<24:11:38, 5.70s/it]
65%|██████▌ | 28659/43939 [46:15:48<24:08:29, 5.69s/it]
65%|██████▌ | 28660/43939 [46:15:54<24:08:58, 5.69s/it]
65%|██████▌ | 28661/43939 [46:15:59<24:10:10, 5.70s/it]
65%|██████▌ | 28662/43939 [46:16:05<24:10:20, 5.70s/it]
65%|██████▌ | 28663/43939 [46:16:11<24:07:22, 5.68s/it]
65%|██████▌ | 28664/43939 [46:16:16<24:07:34, 5.69s/it]
65%|██████▌ | 28665/43939 [46:16:22<24:07:55, 5.69s/it]
65%|██████▌ | 28666/43939 [46:16:28<24:09:17, 5.69s/it]
65%|██████▌ | 28667/43939 [46:16:33<24:08:58, 5.69s/it]
65%|██████▌ | 28668/43939 [46:16:39<24:11:13, 5.70s/it]
65%|██████▌ | 28669/43939 [46:16:45<24:12:16, 5.71s/it]
65%|██████▌ | 28670/43939 [46:16:51<24:29:54, 5.78s/it]
65%|██████▌ | 28671/43939 [46:16:57<24:26:59, 5.76s/it]
65%|██████▌ | 28672/43939 [46:17:02<24:20:36, 5.74s/it]
65%|██████▌ | 28673/43939 [46:17:08<24:16:16, 5.72s/it]
65%|██████▌ | 28674/43939 [46:17:14<24:13:50, 5.71s/it]
65%|██████▌ | 28675/43939 [46:17:19<24:14:06, 5.72s/it]
65%|██████▌ | 28676/43939 [46:17:25<24:10:53, 5.70s/it]
65%|██████▌ | 28677/43939 [46:17:31<24:12:36, 5.71s/it]
65%|██████▌ | 28678/43939 [46:17:36<24:12:36, 5.71s/it]
65%|██████▌ | 28679/43939 [46:17:42<24:11:49, 5.71s/it]
65%|██████▌ | 28680/43939 [46:17:48<24:09:28, 5.70s/it]
65%|██████▌ | 28681/43939 [46:17:53<24:09:18, 5.70s/it]
65%|██████▌ | 28682/43939 [46:17:59<24:08:59, 5.70s/it]
65%|██████▌ | 28683/43939 [46:18:05<24:07:56, 5.69s/it]
65%|██████▌ | 28684/43939 [46:18:11<24:07:14, 5.69s/it]
65%|██████▌ | 28685/43939 [46:18:16<24:08:51, 5.70s/it]
65%|██████▌ | 28686/43939 [46:18:22<24:08:37, 5.70s/it]
65%|██████▌ | 28687/43939 [46:18:28<24:09:09, 5.70s/it]
65%|██████▌ | 28688/43939 [46:18:33<24:10:38, 5.71s/it]
65%|██████▌ | 28689/43939 [46:18:39<24:11:14, 5.71s/it]
65%|██████▌ | 28690/43939 [46:18:45<24:26:22, 5.77s/it]
65%|██████▌ | 28691/43939 [46:18:51<24:17:58, 5.74s/it]
65%|██████▌ | 28692/43939 [46:18:56<24:18:58, 5.74s/it]
65%|██████▌ | 28693/43939 [46:19:02<24:15:11, 5.73s/it]
65%|██████▌ | 28694/43939 [46:19:08<24:14:49, 5.73s/it]
65%|██████▌ | 28695/43939 [46:19:14<24:13:32, 5.72s/it]
65%|██████▌ | 28696/43939 [46:19:19<24:12:56, 5.72s/it]
65%|██████▌ | 28697/43939 [46:19:25<24:14:09, 5.72s/it]
65%|██████▌ | 28698/43939 [46:19:31<24:11:08, 5.71s/it]
65%|██████▌ | 28699/43939 [46:19:36<24:13:03, 5.72s/it]
65%|██████▌ | 28700/43939 [46:19:42<24:11:17, 5.71s/it]
{'loss': 0.7823, 'grad_norm': 1.4042035341262817, 'learning_rate': 1.619061225106234e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.33, 'epoch': 2.61}
65%|██████▌ | 28700/43939 [46:19:42<24:11:17, 5.71s/it]
65%|██████▌ | 28701/43939 [46:19:48<24:10:42, 5.71s/it]
65%|██████▌ | 28702/43939 [46:19:54<24:11:49, 5.72s/it]
65%|██████▌ | 28703/43939 [46:19:59<24:10:14, 5.71s/it]
65%|██████▌ | 28704/43939 [46:20:05<24:10:36, 5.71s/it]
65%|██████▌ | 28705/43939 [46:20:11<24:08:34, 5.71s/it]
65%|██████▌ | 28706/43939 [46:20:16<24:10:19, 5.71s/it]
65%|██████▌ | 28707/43939 [46:20:22<24:09:42, 5.71s/it]
65%|██████▌ | 28708/43939 [46:20:28<24:09:20, 5.71s/it]
65%|██████▌ | 28709/43939 [46:20:34<24:08:20, 5.71s/it]
65%|██████▌ | 28710/43939 [46:20:39<24:09:02, 5.71s/it]
65%|██████▌ | 28711/43939 [46:20:45<24:08:45, 5.71s/it]
65%|██████▌ | 28712/43939 [46:20:51<24:25:00, 5.77s/it]
65%|██████▌ | 28713/43939 [46:20:57<24:16:35, 5.74s/it]
65%|██████▌ | 28714/43939 [46:21:02<24:12:23, 5.72s/it]
65%|██████▌ | 28715/43939 [46:21:08<24:11:30, 5.72s/it]
65%|██████▌ | 28716/43939 [46:21:14<24:07:04, 5.70s/it]
65%|██████▌ | 28717/43939 [46:21:19<24:08:19, 5.71s/it]
65%|██████▌ | 28718/43939 [46:21:25<24:13:26, 5.73s/it]
65%|██████▌ | 28719/43939 [46:21:31<24:11:30, 5.72s/it]
65%|██████▌ | 28720/43939 [46:21:36<24:08:36, 5.71s/it]
65%|██████▌ | 28721/43939 [46:21:42<24:08:15, 5.71s/it]
65%|██████▌ | 28722/43939 [46:21:48<24:05:56, 5.70s/it]
65%|██████▌ | 28723/43939 [46:21:54<24:05:38, 5.70s/it]
65%|██████▌ | 28724/43939 [46:21:59<24:09:36, 5.72s/it]
65%|██████▌ | 28725/43939 [46:22:05<24:08:51, 5.71s/it]
65%|██████▌ | 28726/43939 [46:22:11<24:12:12, 5.73s/it]
65%|██████▌ | 28727/43939 [46:22:16<24:09:51, 5.72s/it]
65%|██████▌ | 28728/43939 [46:22:22<24:13:36, 5.73s/it]
65%|██████▌ | 28729/43939 [46:22:28<24:14:01, 5.74s/it]
65%|██████▌ | 28730/43939 [46:22:34<24:11:36, 5.73s/it]
65%|██████▌ | 28731/43939 [46:22:39<24:10:19, 5.72s/it]
65%|██████▌ | 28732/43939 [46:22:45<24:10:02, 5.72s/it]
65%|██████▌ | 28733/43939 [46:22:51<24:23:24, 5.77s/it]
65%|██████▌ | 28734/43939 [46:22:57<24:18:13, 5.75s/it]
65%|██████▌ | 28735/43939 [46:23:02<24:15:22, 5.74s/it]
65%|██████▌ | 28736/43939 [46:23:08<24:13:41, 5.74s/it]
65%|██████▌ | 28737/43939 [46:23:14<24:12:18, 5.73s/it]
65%|██████▌ | 28738/43939 [46:23:20<24:11:13, 5.73s/it]
65%|██████▌ | 28739/43939 [46:23:25<24:09:39, 5.72s/it]
65%|██████▌ | 28740/43939 [46:23:31<24:06:16, 5.71s/it]
65%|██████▌ | 28741/43939 [46:23:37<24:04:39, 5.70s/it]
65%|██████▌ | 28742/43939 [46:23:42<24:03:52, 5.70s/it]
65%|██████▌ | 28743/43939 [46:23:48<24:02:41, 5.70s/it]
65%|██████▌ | 28744/43939 [46:23:54<24:05:49, 5.71s/it]
65%|██████▌ | 28745/43939 [46:24:00<24:06:52, 5.71s/it]
65%|██████▌ | 28746/43939 [46:24:05<24:03:39, 5.70s/it]
65%|██████▌ | 28747/43939 [46:24:11<24:04:01, 5.70s/it]
65%|██████▌ | 28748/43939 [46:24:17<24:04:58, 5.71s/it]
65%|██████▌ | 28749/43939 [46:24:22<24:06:55, 5.72s/it]
65%|██████▌ | 28750/43939 [46:24:28<24:11:17, 5.73s/it]
{'loss': 0.7507, 'grad_norm': 1.3731398582458496, 'learning_rate': 1.6097749493307497e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 456.94, 'epoch': 2.62}
65%|██████▌ | 28750/43939 [46:24:28<24:11:17, 5.73s/it]
65%|██████▌ | 28751/43939 [46:24:34<24:08:33, 5.72s/it]
65%|██████▌ | 28752/43939 [46:24:39<24:04:27, 5.71s/it]
65%|██████▌ | 28753/43939 [46:24:45<24:02:36, 5.70s/it]
65%|██████▌ | 28754/43939 [46:24:51<24:03:09, 5.70s/it]
65%|██████▌ | 28755/43939 [46:24:57<24:17:18, 5.76s/it]
65%|██████▌ | 28756/43939 [46:25:02<24:13:35, 5.74s/it]
65%|██████▌ | 28757/43939 [46:25:08<24:12:28, 5.74s/it]
65%|██████▌ | 28758/43939 [46:25:14<24:08:43, 5.73s/it]
65%|██████▌ | 28759/43939 [46:25:20<24:08:21, 5.72s/it]
65%|██████▌ | 28760/43939 [46:25:25<24:04:45, 5.71s/it]
65%|██████▌ | 28761/43939 [46:25:31<24:01:18, 5.70s/it]
65%|██████▌ | 28762/43939 [46:25:37<24:03:40, 5.71s/it]
65%|██████▌ | 28763/43939 [46:25:42<24:02:49, 5.70s/it]
65%|██████▌ | 28764/43939 [46:25:48<24:04:32, 5.71s/it]
65%|██████▌ | 28765/43939 [46:25:54<24:04:12, 5.71s/it]
65%|██████▌ | 28766/43939 [46:26:00<24:05:00, 5.71s/it]
65%|██████▌ | 28767/43939 [46:26:05<24:05:33, 5.72s/it]
65%|██████▌ | 28768/43939 [46:26:11<24:04:12, 5.71s/it]
65%|██████▌ | 28769/43939 [46:26:17<24:00:36, 5.70s/it]
65%|██████▌ | 28770/43939 [46:26:22<24:00:45, 5.70s/it]
65%|██████▌ | 28771/43939 [46:26:28<24:02:36, 5.71s/it]
65%|██████▌ | 28772/43939 [46:26:34<24:04:57, 5.72s/it]
65%|██████▌ | 28773/43939 [46:26:40<24:03:24, 5.71s/it]
65%|██████▌ | 28774/43939 [46:26:45<24:00:23, 5.70s/it]
65%|██████▌ | 28775/43939 [46:26:51<24:10:04, 5.74s/it]
65%|██████▌ | 28776/43939 [46:26:57<24:27:24, 5.81s/it]
65%|██████▌ | 28777/43939 [46:27:03<24:15:57, 5.76s/it]
65%|██████▌ | 28778/43939 [46:27:08<24:07:21, 5.73s/it]
65%|██████▌ | 28779/43939 [46:27:14<24:03:52, 5.71s/it]
65%|██████▌ | 28780/43939 [46:27:20<24:03:54, 5.72s/it]
66%|██████▌ | 28781/43939 [46:27:25<24:01:26, 5.71s/it]
66%|██████▌ | 28782/43939 [46:27:31<24:01:37, 5.71s/it]
66%|██████▌ | 28783/43939 [46:27:37<23:58:58, 5.70s/it]
66%|██████▌ | 28784/43939 [46:27:42<23:59:13, 5.70s/it]
66%|██████▌ | 28785/43939 [46:27:48<24:00:24, 5.70s/it]
66%|██████▌ | 28786/43939 [46:27:54<23:59:27, 5.70s/it]
66%|██████▌ | 28787/43939 [46:28:00<23:59:17, 5.70s/it]
66%|██████▌ | 28788/43939 [46:28:05<24:00:15, 5.70s/it]
66%|██████▌ | 28789/43939 [46:28:11<24:02:53, 5.71s/it]
66%|██████▌ | 28790/43939 [46:28:17<24:01:17, 5.71s/it]
66%|██████▌ | 28791/43939 [46:28:22<23:59:08, 5.70s/it]
66%|██████▌ | 28792/43939 [46:28:28<24:03:56, 5.72s/it]
66%|██████▌ | 28793/43939 [46:28:34<24:03:53, 5.72s/it]
66%|██████▌ | 28794/43939 [46:28:40<24:02:00, 5.71s/it]
66%|██████▌ | 28795/43939 [46:28:45<24:01:46, 5.71s/it]
66%|██████▌ | 28796/43939 [46:28:51<24:02:30, 5.72s/it]
66%|██████▌ | 28797/43939 [46:28:57<24:00:42, 5.71s/it]
66%|██████▌ | 28798/43939 [46:29:03<24:18:01, 5.78s/it]
66%|██████▌ | 28799/43939 [46:29:08<24:13:35, 5.76s/it]
66%|██████▌ | 28800/43939 [46:29:14<24:07:38, 5.74s/it]
{'loss': 0.7715, 'grad_norm': 1.2420955896377563, 'learning_rate': 1.6005027189474396e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.3, 'epoch': 2.62}
66%|██████▌ | 28800/43939 [46:29:14<24:07:38, 5.74s/it]
66%|██████▌ | 28801/43939 [46:29:20<24:04:49, 5.73s/it]
66%|██████▌ | 28802/43939 [46:29:25<24:05:09, 5.73s/it]
66%|██████▌ | 28803/43939 [46:29:31<24:02:44, 5.72s/it]
66%|██████▌ | 28804/43939 [46:29:37<24:00:10, 5.71s/it]
66%|██████▌ | 28805/43939 [46:29:43<23:58:30, 5.70s/it]
66%|██████▌ | 28806/43939 [46:29:48<23:55:33, 5.69s/it]
66%|██████▌ | 28807/43939 [46:29:54<23:55:58, 5.69s/it]
66%|██████▌ | 28808/43939 [46:30:00<23:57:15, 5.70s/it]
66%|██████▌ | 28809/43939 [46:30:05<23:55:56, 5.69s/it]
66%|██████▌ | 28810/43939 [46:30:11<23:57:32, 5.70s/it]
66%|██████▌ | 28811/43939 [46:30:17<23:57:34, 5.70s/it]
66%|██████▌ | 28812/43939 [46:30:22<23:56:49, 5.70s/it]
66%|██████▌ | 28813/43939 [46:30:28<23:56:34, 5.70s/it]
66%|██████▌ | 28814/43939 [46:30:34<24:00:58, 5.72s/it]
66%|██████▌ | 28815/43939 [46:30:40<24:00:44, 5.72s/it]
66%|██████▌ | 28816/43939 [46:30:45<24:00:42, 5.72s/it]
66%|██████▌ | 28817/43939 [46:30:51<23:58:39, 5.71s/it]
66%|██████▌ | 28818/43939 [46:30:57<23:57:59, 5.71s/it]
66%|██████▌ | 28819/43939 [46:31:03<24:16:58, 5.78s/it]
66%|██████▌ | 28820/43939 [46:31:08<24:12:55, 5.77s/it]
66%|██████▌ | 28821/43939 [46:31:14<24:05:32, 5.74s/it]
66%|██████▌ | 28822/43939 [46:31:20<24:01:58, 5.72s/it]
66%|██████▌ | 28823/43939 [46:31:25<23:57:59, 5.71s/it]
66%|██████▌ | 28824/43939 [46:31:31<23:57:05, 5.70s/it]
66%|██████▌ | 28825/43939 [46:31:37<23:56:25, 5.70s/it]
66%|██████▌ | 28826/43939 [46:31:43<23:57:35, 5.71s/it]
66%|██████▌ | 28827/43939 [46:31:48<23:54:29, 5.70s/it]
66%|██████▌ | 28828/43939 [46:31:54<24:00:02, 5.72s/it]
66%|██████▌ | 28829/43939 [46:32:00<24:00:04, 5.72s/it]
66%|██████▌ | 28830/43939 [46:32:05<23:55:39, 5.70s/it]
66%|██████▌ | 28831/43939 [46:32:11<23:50:15, 5.68s/it]
66%|██████▌ | 28832/43939 [46:32:17<23:54:03, 5.70s/it]
66%|██████▌ | 28833/43939 [46:32:22<23:53:55, 5.70s/it]
66%|██████▌ | 28834/43939 [46:32:28<23:55:28, 5.70s/it]
66%|██████▌ | 28835/43939 [46:32:34<23:58:50, 5.72s/it]
66%|██████▌ | 28836/43939 [46:32:40<24:00:27, 5.72s/it]
66%|██████▌ | 28837/43939 [46:32:45<24:02:26, 5.73s/it]
66%|██████▌ | 28838/43939 [46:32:51<24:00:46, 5.72s/it]
66%|██████▌ | 28839/43939 [46:32:57<23:59:22, 5.72s/it]
66%|██████▌ | 28840/43939 [46:33:02<23:58:55, 5.72s/it]
66%|██████▌ | 28841/43939 [46:33:08<24:15:01, 5.78s/it]
66%|██████▌ | 28842/43939 [46:33:14<24:13:04, 5.77s/it]
66%|██████▌ | 28843/43939 [46:33:20<24:10:40, 5.77s/it]
66%|██████▌ | 28844/43939 [46:33:26<24:07:59, 5.76s/it]
66%|██████▌ | 28845/43939 [46:33:31<24:05:15, 5.75s/it]
66%|██████▌ | 28846/43939 [46:33:37<24:05:10, 5.75s/it]
66%|██████▌ | 28847/43939 [46:33:43<24:02:53, 5.74s/it]
66%|██████▌ | 28848/43939 [46:33:49<23:58:36, 5.72s/it]
66%|██████▌ | 28849/43939 [46:33:54<23:57:46, 5.72s/it]
66%|██████▌ | 28850/43939 [46:34:00<23:57:23, 5.72s/it]
{'loss': 0.7693, 'grad_norm': 1.4313956499099731, 'learning_rate': 1.5912446802475277e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 511.5, 'epoch': 2.63}
66%|██████▌ | 28850/43939 [46:34:00<23:57:23, 5.72s/it]
66%|██████▌ | 28851/43939 [46:34:06<23:54:10, 5.70s/it]
66%|██████▌ | 28852/43939 [46:34:11<23:52:17, 5.70s/it]
66%|██████▌ | 28853/43939 [46:34:17<23:52:58, 5.70s/it]
66%|██████▌ | 28854/43939 [46:34:23<23:52:14, 5.70s/it]
66%|██████▌ | 28855/43939 [46:34:28<23:52:11, 5.70s/it]
66%|██████▌ | 28856/43939 [46:34:34<23:51:23, 5.69s/it]
66%|██████▌ | 28857/43939 [46:34:40<23:49:59, 5.69s/it]
66%|██████▌ | 28858/43939 [46:34:45<23:51:21, 5.69s/it]
66%|██████▌ | 28859/43939 [46:34:51<23:49:32, 5.69s/it]
66%|██████▌ | 28860/43939 [46:34:57<23:48:54, 5.69s/it]
66%|██████▌ | 28861/43939 [46:35:03<24:06:36, 5.76s/it]
66%|██████▌ | 28862/43939 [46:35:09<24:08:58, 5.77s/it]
66%|██████▌ | 28863/43939 [46:35:14<24:06:39, 5.76s/it]
66%|██████▌ | 28864/43939 [46:35:20<24:05:54, 5.75s/it]
66%|██████▌ | 28865/43939 [46:35:26<24:00:59, 5.74s/it]
66%|██████▌ | 28866/43939 [46:35:31<24:01:25, 5.74s/it]
66%|██████▌ | 28867/43939 [46:35:37<23:59:09, 5.73s/it]
66%|██████▌ | 28868/43939 [46:35:43<23:56:32, 5.72s/it]
66%|██████▌ | 28869/43939 [46:35:49<23:58:07, 5.73s/it]
66%|██████▌ | 28870/43939 [46:35:54<23:56:26, 5.72s/it]
66%|██████▌ | 28871/43939 [46:36:00<23:53:20, 5.71s/it]
66%|██████▌ | 28872/43939 [46:36:06<23:52:17, 5.70s/it]
66%|██████▌ | 28873/43939 [46:36:11<23:48:31, 5.69s/it]
66%|██████▌ | 28874/43939 [46:36:17<23:50:13, 5.70s/it]
66%|██████▌ | 28875/43939 [46:36:23<23:50:42, 5.70s/it]
66%|██████▌ | 28876/43939 [46:36:28<23:50:11, 5.70s/it]
66%|██████▌ | 28877/43939 [46:36:34<23:49:27, 5.69s/it]
66%|██████▌ | 28878/43939 [46:36:40<23:51:15, 5.70s/it]
66%|██████▌ | 28879/43939 [46:36:46<23:49:50, 5.70s/it]
66%|██████▌ | 28880/43939 [46:36:51<23:48:55, 5.69s/it]
66%|██████▌ | 28881/43939 [46:36:57<23:49:46, 5.70s/it]
66%|██████▌ | 28882/43939 [46:37:03<23:49:07, 5.69s/it]
66%|██████▌ | 28883/43939 [46:37:09<24:06:47, 5.77s/it]
66%|██████▌ | 28884/43939 [46:37:14<24:00:04, 5.74s/it]
66%|██████▌ | 28885/43939 [46:37:20<23:59:51, 5.74s/it]
66%|██████▌ | 28886/43939 [46:37:26<23:58:01, 5.73s/it]
66%|██████▌ | 28887/43939 [46:37:31<24:01:37, 5.75s/it]
66%|██████▌ | 28888/43939 [46:37:37<23:57:45, 5.73s/it]
66%|██████▌ | 28889/43939 [46:37:43<23:56:09, 5.73s/it]
66%|██████▌ | 28890/43939 [46:37:49<23:51:44, 5.71s/it]
66%|██████▌ | 28891/43939 [46:37:54<23:51:38, 5.71s/it]
66%|██████▌ | 28892/43939 [46:38:00<23:51:27, 5.71s/it]
66%|██████▌ | 28893/43939 [46:38:06<23:52:57, 5.71s/it]
66%|██████▌ | 28894/43939 [46:38:11<23:54:59, 5.72s/it]
66%|██████▌ | 28895/43939 [46:38:17<23:55:13, 5.72s/it]
66%|██████▌ | 28896/43939 [46:38:23<23:54:49, 5.72s/it]
66%|██████▌ | 28897/43939 [46:38:29<23:49:33, 5.70s/it]
66%|██████▌ | 28898/43939 [46:38:34<23:50:38, 5.71s/it]
66%|██████▌ | 28899/43939 [46:38:40<23:48:14, 5.70s/it]
66%|██████▌ | 28900/43939 [46:38:46<23:48:02, 5.70s/it]
{'loss': 0.7713, 'grad_norm': 1.4873028993606567, 'learning_rate': 1.5820009792983303e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.58, 'epoch': 2.63}
66%|██████▌ | 28900/43939 [46:38:46<23:48:02, 5.70s/it]
66%|██████▌ | 28901/43939 [46:38:51<23:45:43, 5.69s/it]
66%|██████▌ | 28902/43939 [46:38:57<23:46:51, 5.69s/it]
66%|██████▌ | 28903/43939 [46:39:03<23:47:52, 5.70s/it]
66%|██████▌ | 28904/43939 [46:39:09<24:08:09, 5.78s/it]
66%|██████▌ | 28905/43939 [46:39:14<24:02:43, 5.76s/it]
66%|██████▌ | 28906/43939 [46:39:20<24:00:57, 5.75s/it]
66%|██████▌ | 28907/43939 [46:39:26<23:57:18, 5.74s/it]
66%|██████▌ | 28908/43939 [46:39:31<23:54:17, 5.73s/it]
66%|██████▌ | 28909/43939 [46:39:37<23:50:27, 5.71s/it]
66%|██████▌ | 28910/43939 [46:39:43<23:48:55, 5.70s/it]
66%|██████▌ | 28911/43939 [46:39:49<23:49:12, 5.71s/it]
66%|██████▌ | 28912/43939 [46:39:54<23:50:12, 5.71s/it]
66%|██████▌ | 28913/43939 [46:40:00<23:49:30, 5.71s/it]
66%|██████▌ | 28914/43939 [46:40:06<23:46:19, 5.70s/it]
66%|██████▌ | 28915/43939 [46:40:11<23:48:48, 5.71s/it]
66%|██████▌ | 28916/43939 [46:40:17<23:47:58, 5.70s/it]
66%|██████▌ | 28917/43939 [46:40:23<23:53:12, 5.72s/it]
66%|██████▌ | 28918/43939 [46:40:29<23:52:52, 5.72s/it]
66%|██████▌ | 28919/43939 [46:40:34<23:49:59, 5.71s/it]
66%|██████▌ | 28920/43939 [46:40:40<23:50:11, 5.71s/it]
66%|██████▌ | 28921/43939 [46:40:46<23:47:26, 5.70s/it]
66%|██████▌ | 28922/43939 [46:40:51<23:47:09, 5.70s/it]
66%|██████▌ | 28923/43939 [46:40:57<23:47:48, 5.71s/it]
66%|██████▌ | 28924/43939 [46:41:03<23:49:19, 5.71s/it]
66%|██████▌ | 28925/43939 [46:41:09<24:06:17, 5.78s/it]
66%|██████▌ | 28926/43939 [46:41:14<24:00:49, 5.76s/it]
66%|██████▌ | 28927/43939 [46:41:20<23:57:51, 5.75s/it]
66%|██████▌ | 28928/43939 [46:41:26<23:54:53, 5.74s/it]
66%|██████▌ | 28929/43939 [46:41:32<23:51:31, 5.72s/it]
66%|██████▌ | 28930/43939 [46:41:37<23:45:31, 5.70s/it]
66%|██████▌ | 28931/43939 [46:41:43<23:45:58, 5.70s/it]
66%|██████▌ | 28932/43939 [46:41:49<23:44:10, 5.69s/it]
66%|██████▌ | 28933/43939 [46:41:54<23:45:57, 5.70s/it]
66%|██████▌ | 28934/43939 [46:42:00<23:47:21, 5.71s/it]
66%|██████▌ | 28935/43939 [46:42:06<23:46:34, 5.70s/it]
66%|██████▌ | 28936/43939 [46:42:12<23:53:09, 5.73s/it]
66%|██████▌ | 28937/43939 [46:42:17<23:49:05, 5.72s/it]
66%|██████▌ | 28938/43939 [46:42:23<23:53:59, 5.74s/it]
66%|██████▌ | 28939/43939 [46:42:29<23:49:00, 5.72s/it]
66%|██████▌ | 28940/43939 [46:42:34<23:48:16, 5.71s/it]
66%|██████▌ | 28941/43939 [46:42:40<23:47:16, 5.71s/it]
66%|██████▌ | 28942/43939 [46:42:46<23:46:22, 5.71s/it]
66%|██████▌ | 28943/43939 [46:42:52<23:48:39, 5.72s/it]
66%|██████▌ | 28944/43939 [46:42:57<23:48:54, 5.72s/it]
66%|██████▌ | 28945/43939 [46:43:03<23:46:09, 5.71s/it]
66%|██████▌ | 28946/43939 [46:43:09<23:41:56, 5.69s/it]
66%|██████▌ | 28947/43939 [46:43:14<23:54:09, 5.74s/it]
66%|██████▌ | 28948/43939 [46:43:20<23:53:28, 5.74s/it]
66%|██████▌ | 28949/43939 [46:43:26<23:53:55, 5.74s/it]
66%|██████▌ | 28950/43939 [46:43:32<23:51:04, 5.73s/it]
{'loss': 0.7658, 'grad_norm': 1.6724083423614502, 'learning_rate': 1.5727717619409528e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.25, 'epoch': 2.64}
66%|██████▌ | 28950/43939 [46:43:32<23:51:04, 5.73s/it]
66%|██████▌ | 28951/43939 [46:43:37<23:49:54, 5.72s/it]
66%|██████▌ | 28952/43939 [46:43:43<23:47:51, 5.72s/it]
66%|██████▌ | 28953/43939 [46:43:49<23:50:21, 5.73s/it]
66%|██████▌ | 28954/43939 [46:43:54<23:49:52, 5.73s/it]
66%|██████▌ | 28955/43939 [46:44:00<23:48:34, 5.72s/it]
66%|██████▌ | 28956/43939 [46:44:06<23:45:43, 5.71s/it]
66%|██████▌ | 28957/43939 [46:44:12<23:44:39, 5.71s/it]
66%|██████▌ | 28958/43939 [46:44:17<23:41:19, 5.69s/it]
66%|██████▌ | 28959/43939 [46:44:23<23:42:36, 5.70s/it]
66%|██████▌ | 28960/43939 [46:44:29<23:40:14, 5.69s/it]
66%|██████▌ | 28961/43939 [46:44:34<23:42:31, 5.70s/it]
66%|██████▌ | 28962/43939 [46:44:40<23:38:37, 5.68s/it]
66%|██████▌ | 28963/43939 [46:44:46<23:41:46, 5.70s/it]
66%|██████▌ | 28964/43939 [46:44:51<23:42:48, 5.70s/it]
66%|██████▌ | 28965/43939 [46:44:57<23:42:42, 5.70s/it]
66%|██████▌ | 28966/43939 [46:45:03<23:43:50, 5.71s/it]
66%|██████▌ | 28967/43939 [46:45:09<23:40:38, 5.69s/it]
66%|██████▌ | 28968/43939 [46:45:14<24:02:16, 5.78s/it]
66%|██████▌ | 28969/43939 [46:45:20<24:02:45, 5.78s/it]
66%|██████▌ | 28970/43939 [46:45:26<23:54:47, 5.75s/it]
66%|██████▌ | 28971/43939 [46:45:32<23:49:24, 5.73s/it]
66%|██████▌ | 28972/43939 [46:45:37<23:47:52, 5.72s/it]
66%|██████▌ | 28973/43939 [46:45:43<23:48:52, 5.73s/it]
66%|██████▌ | 28974/43939 [46:45:49<23:47:49, 5.72s/it]
66%|██████▌ | 28975/43939 [46:45:54<23:45:05, 5.71s/it]
66%|██████▌ | 28976/43939 [46:46:00<23:45:04, 5.71s/it]
66%|██████▌ | 28977/43939 [46:46:06<23:46:26, 5.72s/it]
66%|██████▌ | 28978/43939 [46:46:12<23:43:07, 5.71s/it]
66%|██████▌ | 28979/43939 [46:46:17<23:43:33, 5.71s/it]
66%|██████▌ | 28980/43939 [46:46:23<23:44:15, 5.71s/it]
66%|██████▌ | 28981/43939 [46:46:29<23:42:09, 5.70s/it]
66%|██████▌ | 28982/43939 [46:46:34<23:41:42, 5.70s/it]
66%|██████▌ | 28983/43939 [46:46:40<23:39:42, 5.70s/it]
66%|██████▌ | 28984/43939 [46:46:46<23:37:28, 5.69s/it]
66%|██████▌ | 28985/43939 [46:46:51<23:39:20, 5.69s/it]
66%|██████▌ | 28986/43939 [46:46:57<23:40:20, 5.70s/it]
66%|██████▌ | 28987/43939 [46:47:03<23:40:58, 5.70s/it]
66%|██████▌ | 28988/43939 [46:47:09<23:43:18, 5.71s/it]
66%|██████▌ | 28989/43939 [46:47:15<23:55:59, 5.76s/it]
66%|██████▌ | 28990/43939 [46:47:20<23:52:14, 5.75s/it]
66%|██████▌ | 28991/43939 [46:47:26<23:50:11, 5.74s/it]
66%|██████▌ | 28992/43939 [46:47:32<23:47:26, 5.73s/it]
66%|██████▌ | 28993/43939 [46:47:37<23:49:42, 5.74s/it]
66%|██████▌ | 28994/43939 [46:47:43<23:46:35, 5.73s/it]
66%|██████▌ | 28995/43939 [46:47:49<23:42:42, 5.71s/it]
66%|██████▌ | 28996/43939 [46:47:55<23:42:14, 5.71s/it]
66%|██████▌ | 28997/43939 [46:48:00<23:40:54, 5.71s/it]
66%|██████▌ | 28998/43939 [46:48:06<23:41:01, 5.71s/it]
66%|██████▌ | 28999/43939 [46:48:12<23:41:55, 5.71s/it]
66%|██████▌ | 29000/43939 [46:48:17<23:38:32, 5.70s/it]
{'loss': 0.7635, 'grad_norm': 1.4013808965682983, 'learning_rate': 1.5635571737879874e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.29, 'epoch': 2.64}
66%|██████▌ | 29000/43939 [46:48:17<23:38:32, 5.70s/it]
66%|██████▌ | 29001/43939 [46:48:23<23:40:08, 5.70s/it]
66%|██████▌ | 29002/43939 [46:48:29<23:40:10, 5.70s/it]
66%|██████▌ | 29003/43939 [46:48:34<23:40:26, 5.71s/it]
66%|██████▌ | 29004/43939 [46:48:40<23:39:38, 5.70s/it]
66%|██████▌ | 29005/43939 [46:48:46<23:38:40, 5.70s/it]
66%|██████▌ | 29006/43939 [46:48:51<23:36:29, 5.69s/it]
66%|██████▌ | 29007/43939 [46:48:57<23:37:25, 5.70s/it]
66%|██████▌ | 29008/43939 [46:49:03<23:36:32, 5.69s/it]
66%|██████▌ | 29009/43939 [46:49:09<23:35:05, 5.69s/it]
66%|██████▌ | 29010/43939 [46:49:14<23:52:54, 5.76s/it]
66%|██████▌ | 29011/43939 [46:49:20<23:49:21, 5.75s/it]
66%|██████▌ | 29012/43939 [46:49:26<23:47:03, 5.74s/it]
66%|██████▌ | 29013/43939 [46:49:32<23:44:48, 5.73s/it]
66%|██████▌ | 29014/43939 [46:49:37<23:40:57, 5.71s/it]
66%|██████▌ | 29015/43939 [46:49:43<23:38:56, 5.70s/it]
66%|██████▌ | 29016/43939 [46:49:49<23:34:55, 5.69s/it]
66%|██████▌ | 29017/43939 [46:49:54<23:42:44, 5.72s/it]
66%|██████▌ | 29018/43939 [46:50:00<23:39:16, 5.71s/it]
66%|██████▌ | 29019/43939 [46:50:06<23:43:30, 5.72s/it]
66%|██████▌ | 29020/43939 [46:50:12<23:42:05, 5.72s/it]
66%|██████▌ | 29021/43939 [46:50:17<23:39:32, 5.71s/it]
66%|██████▌ | 29022/43939 [46:50:23<23:36:31, 5.70s/it]
66%|██████▌ | 29023/43939 [46:50:29<23:35:10, 5.69s/it]
66%|██████▌ | 29024/43939 [46:50:34<23:35:20, 5.69s/it]
66%|██████▌ | 29025/43939 [46:50:40<23:38:38, 5.71s/it]
66%|██████▌ | 29026/43939 [46:50:46<23:40:36, 5.72s/it]
66%|██████▌ | 29027/43939 [46:50:52<23:42:43, 5.72s/it]
66%|██████▌ | 29028/43939 [46:50:57<23:40:13, 5.71s/it]
66%|██████▌ | 29029/43939 [46:51:03<23:43:11, 5.73s/it]
66%|██████▌ | 29030/43939 [46:51:09<23:42:01, 5.72s/it]
66%|██████▌ | 29031/43939 [46:51:14<23:41:49, 5.72s/it]
66%|██████▌ | 29032/43939 [46:51:20<24:02:24, 5.81s/it]
66%|██████▌ | 29033/43939 [46:51:26<23:51:42, 5.76s/it]
66%|██████▌ | 29034/43939 [46:51:32<23:49:09, 5.75s/it]
66%|██████▌ | 29035/43939 [46:51:38<23:44:22, 5.73s/it]
66%|██████▌ | 29036/43939 [46:51:43<23:41:34, 5.72s/it]
66%|██████▌ | 29037/43939 [46:51:49<23:39:25, 5.72s/it]
66%|██████▌ | 29038/43939 [46:51:55<23:38:25, 5.71s/it]
66%|██████▌ | 29039/43939 [46:52:00<23:37:42, 5.71s/it]
66%|██████▌ | 29040/43939 [46:52:06<23:39:14, 5.72s/it]
66%|██████▌ | 29041/43939 [46:52:12<23:36:50, 5.71s/it]
66%|██████▌ | 29042/43939 [46:52:17<23:38:34, 5.71s/it]
66%|██████▌ | 29043/43939 [46:52:23<23:39:28, 5.72s/it]
66%|██████▌ | 29044/43939 [46:52:29<23:40:19, 5.72s/it]
66%|██████▌ | 29045/43939 [46:52:35<23:41:14, 5.73s/it]
66%|██████▌ | 29046/43939 [46:52:40<23:36:17, 5.71s/it]
66%|██████▌ | 29047/43939 [46:52:46<23:35:35, 5.70s/it]
66%|██████▌ | 29048/43939 [46:52:52<23:33:42, 5.70s/it]
66%|██████▌ | 29049/43939 [46:52:57<23:31:14, 5.69s/it]
66%|██████▌ | 29050/43939 [46:53:03<23:37:03, 5.71s/it]
{'loss': 0.78, 'grad_norm': 1.3551985025405884, 'learning_rate': 1.554357360221216e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.26, 'epoch': 2.64}
66%|██████▌ | 29050/43939 [46:53:03<23:37:03, 5.71s/it]
66%|██████▌ | 29051/43939 [46:53:09<23:44:12, 5.74s/it]
66%|██████▌ | 29052/43939 [46:53:15<23:42:25, 5.73s/it]
66%|██████▌ | 29053/43939 [46:53:21<23:58:13, 5.80s/it]
66%|██████▌ | 29054/43939 [46:53:26<23:49:11, 5.76s/it]
66%|██████▌ | 29055/43939 [46:53:32<23:48:29, 5.76s/it]
66%|██████▌ | 29056/43939 [46:53:38<23:41:27, 5.73s/it]
66%|██████▌ | 29057/43939 [46:53:43<23:41:04, 5.73s/it]
66%|██████▌ | 29058/43939 [46:53:49<23:35:52, 5.71s/it]
66%|██████▌ | 29059/43939 [46:53:55<23:38:24, 5.72s/it]
66%|██████▌ | 29060/43939 [46:54:00<23:36:21, 5.71s/it]
66%|██████▌ | 29061/43939 [46:54:06<23:34:45, 5.71s/it]
66%|██████▌ | 29062/43939 [46:54:12<23:33:28, 5.70s/it]
66%|██████▌ | 29063/43939 [46:54:18<23:38:26, 5.72s/it]
66%|██████▌ | 29064/43939 [46:54:23<23:34:49, 5.71s/it]
66%|██████▌ | 29065/43939 [46:54:29<23:35:28, 5.71s/it]
66%|██████▌ | 29066/43939 [46:54:35<23:34:37, 5.71s/it]
66%|██████▌ | 29067/43939 [46:54:40<23:32:33, 5.70s/it]
66%|██████▌ | 29068/43939 [46:54:46<23:33:29, 5.70s/it]
66%|██████▌ | 29069/43939 [46:54:52<23:32:35, 5.70s/it]
66%|██████▌ | 29070/43939 [46:54:58<23:34:17, 5.71s/it]
66%|██████▌ | 29071/43939 [46:55:03<23:33:21, 5.70s/it]
66%|██████▌ | 29072/43939 [46:55:09<23:35:13, 5.71s/it]
66%|██████▌ | 29073/43939 [46:55:15<23:31:28, 5.70s/it]
66%|██████▌ | 29074/43939 [46:55:21<23:48:16, 5.76s/it]
66%|██████▌ | 29075/43939 [46:55:26<23:41:15, 5.74s/it]
66%|██████▌ | 29076/43939 [46:55:32<23:39:03, 5.73s/it]
66%|██████▌ | 29077/43939 [46:55:38<23:37:07, 5.72s/it]
66%|██████▌ | 29078/43939 [46:55:43<23:34:51, 5.71s/it]
66%|██████▌ | 29079/43939 [46:55:49<23:34:28, 5.71s/it]
66%|██████▌ | 29080/43939 [46:55:55<23:33:41, 5.71s/it]
66%|██████▌ | 29081/43939 [46:56:00<23:33:24, 5.71s/it]
66%|██████▌ | 29082/43939 [46:56:06<23:30:06, 5.69s/it]
66%|██████▌ | 29083/43939 [46:56:12<23:29:47, 5.69s/it]
66%|██████▌ | 29084/43939 [46:56:18<23:32:48, 5.71s/it]
66%|██████▌ | 29085/43939 [46:56:23<23:33:24, 5.71s/it]
66%|██████▌ | 29086/43939 [46:56:29<23:33:32, 5.71s/it]
66%|██████▌ | 29087/43939 [46:56:35<23:33:02, 5.71s/it]
66%|██████▌ | 29088/43939 [46:56:40<23:31:51, 5.70s/it]
66%|██████▌ | 29089/43939 [46:56:46<23:35:17, 5.72s/it]
66%|██████▌ | 29090/43939 [46:56:52<23:35:42, 5.72s/it]
66%|██████▌ | 29091/43939 [46:56:57<23:30:17, 5.70s/it]
66%|██████▌ | 29092/43939 [46:57:03<23:31:11, 5.70s/it]
66%|██████▌ | 29093/43939 [46:57:09<23:28:48, 5.69s/it]
66%|██████▌ | 29094/43939 [46:57:15<23:30:38, 5.70s/it]
66%|██████▌ | 29095/43939 [46:57:21<23:46:44, 5.77s/it]
66%|██████▌ | 29096/43939 [46:57:26<23:43:31, 5.75s/it]
66%|██████▌ | 29097/43939 [46:57:32<23:37:30, 5.73s/it]
66%|██████▌ | 29098/43939 [46:57:38<23:36:32, 5.73s/it]
66%|██████▌ | 29099/43939 [46:57:43<23:33:49, 5.72s/it]
66%|██████▌ | 29100/43939 [46:57:49<23:34:05, 5.72s/it]
{'loss': 0.7927, 'grad_norm': 1.2868053913116455, 'learning_rate': 1.5451724663893164e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.48, 'epoch': 2.65}
66%|██████▌ | 29100/43939 [46:57:49<23:34:05, 5.72s/it]
66%|██████▌ | 29101/43939 [46:57:55<23:34:19, 5.72s/it]
66%|██████▌ | 29102/43939 [46:58:00<23:28:48, 5.70s/it]
66%|██████▌ | 29103/43939 [46:58:06<23:31:28, 5.71s/it]
66%|██████▌ | 29104/43939 [46:58:12<23:30:09, 5.70s/it]
66%|██████▌ | 29105/43939 [46:58:18<23:28:07, 5.70s/it]
66%|██████▌ | 29106/43939 [46:58:23<23:29:30, 5.70s/it]
66%|██████▌ | 29107/43939 [46:58:29<23:30:23, 5.71s/it]
66%|██████▌ | 29108/43939 [46:58:35<23:28:10, 5.70s/it]
66%|██████▌ | 29109/43939 [46:58:40<23:28:08, 5.70s/it]
66%|██████▋ | 29110/43939 [46:58:46<23:27:05, 5.69s/it]
66%|██████▋ | 29111/43939 [46:58:52<23:27:04, 5.69s/it]
66%|██████▋ | 29112/43939 [46:58:57<23:27:24, 5.70s/it]
66%|██████▋ | 29113/43939 [46:59:03<23:30:47, 5.71s/it]
66%|██████▋ | 29114/43939 [46:59:09<23:36:17, 5.73s/it]
66%|██████▋ | 29115/43939 [46:59:15<23:37:26, 5.74s/it]
66%|██████▋ | 29116/43939 [46:59:20<23:34:21, 5.72s/it]
66%|██████▋ | 29117/43939 [46:59:26<23:50:53, 5.79s/it]
66%|██████▋ | 29118/43939 [46:59:32<23:45:22, 5.77s/it]
66%|██████▋ | 29119/43939 [46:59:38<23:41:47, 5.76s/it]
66%|██████▋ | 29120/43939 [46:59:43<23:37:09, 5.74s/it]
66%|██████▋ | 29121/43939 [46:59:49<23:33:11, 5.72s/it]
66%|██████▋ | 29122/43939 [46:59:55<23:31:42, 5.72s/it]
66%|██████▋ | 29123/43939 [47:00:01<23:31:02, 5.71s/it]
66%|██████▋ | 29124/43939 [47:00:06<23:30:23, 5.71s/it]
66%|██████▋ | 29125/43939 [47:00:12<23:28:14, 5.70s/it]
66%|██████▋ | 29126/43939 [47:00:18<23:27:09, 5.70s/it]
66%|██████▋ | 29127/43939 [47:00:23<23:26:55, 5.70s/it]
66%|██████▋ | 29128/43939 [47:00:29<23:25:58, 5.70s/it]
66%|██████▋ | 29129/43939 [47:00:35<23:24:50, 5.69s/it]
66%|██████▋ | 29130/43939 [47:00:40<23:29:14, 5.71s/it]
66%|██████▋ | 29131/43939 [47:00:46<23:26:18, 5.70s/it]
66%|██████▋ | 29132/43939 [47:00:52<23:29:30, 5.71s/it]
66%|██████▋ | 29133/43939 [47:00:58<23:30:22, 5.72s/it]
66%|██████▋ | 29134/43939 [47:01:03<23:30:25, 5.72s/it]
66%|██████▋ | 29135/43939 [47:01:09<23:30:50, 5.72s/it]
66%|██████▋ | 29136/43939 [47:01:15<23:30:55, 5.72s/it]
66%|██████▋ | 29137/43939 [47:01:21<23:33:00, 5.73s/it]
66%|██████▋ | 29138/43939 [47:01:26<23:29:50, 5.72s/it]
66%|██████▋ | 29139/43939 [47:01:32<23:46:43, 5.78s/it]
66%|██████▋ | 29140/43939 [47:01:38<23:41:34, 5.76s/it]
66%|██████▋ | 29141/43939 [47:01:44<23:36:21, 5.74s/it]
66%|██████▋ | 29142/43939 [47:01:49<23:34:20, 5.73s/it]
66%|██████▋ | 29143/43939 [47:01:55<23:29:23, 5.72s/it]
66%|██████▋ | 29144/43939 [47:02:01<23:28:17, 5.71s/it]
66%|██████▋ | 29145/43939 [47:02:06<23:25:15, 5.70s/it]
66%|██████▋ | 29146/43939 [47:02:12<23:23:52, 5.69s/it]
66%|██████▋ | 29147/43939 [47:02:18<23:23:06, 5.69s/it]
66%|██████▋ | 29148/43939 [47:02:23<23:26:36, 5.71s/it]
66%|██████▋ | 29149/43939 [47:02:29<23:25:41, 5.70s/it]
66%|██████▋ | 29150/43939 [47:02:35<23:28:03, 5.71s/it]
{'loss': 0.7704, 'grad_norm': 1.7285215854644775, 'learning_rate': 1.5360026372055732e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.78, 'epoch': 2.65}
66%|██████▋ | 29150/43939 [47:02:35<23:28:03, 5.71s/it]
66%|██████▋ | 29151/43939 [47:02:41<23:27:33, 5.71s/it]
66%|██████▋ | 29152/43939 [47:02:46<23:26:31, 5.71s/it]
66%|██████▋ | 29153/43939 [47:02:52<23:24:15, 5.70s/it]
66%|██████▋ | 29154/43939 [47:02:58<23:25:01, 5.70s/it]
66%|██████▋ | 29155/43939 [47:03:03<23:24:51, 5.70s/it]
66%|██████▋ | 29156/43939 [47:03:09<23:22:10, 5.69s/it]
66%|██████▋ | 29157/43939 [47:03:15<23:21:03, 5.69s/it]
66%|██████▋ | 29158/43939 [47:03:20<23:21:31, 5.69s/it]
66%|██████▋ | 29159/43939 [47:03:26<23:23:08, 5.70s/it]
66%|██████▋ | 29160/43939 [47:03:32<23:40:06, 5.77s/it]
66%|██████▋ | 29161/43939 [47:03:38<23:36:25, 5.75s/it]
66%|██████▋ | 29162/43939 [47:03:43<23:32:51, 5.74s/it]
66%|██████▋ | 29163/43939 [47:03:49<23:28:42, 5.72s/it]
66%|██████▋ | 29164/43939 [47:03:55<23:29:12, 5.72s/it]
66%|██████▋ | 29165/43939 [47:04:01<23:28:25, 5.72s/it]
66%|██████▋ | 29166/43939 [47:04:06<23:30:11, 5.73s/it]
66%|██████▋ | 29167/43939 [47:04:12<23:28:23, 5.72s/it]
66%|██████▋ | 29168/43939 [47:04:18<23:26:09, 5.71s/it]
66%|██████▋ | 29169/43939 [47:04:23<23:28:38, 5.72s/it]
66%|██████▋ | 29170/43939 [47:04:29<23:27:03, 5.72s/it]
66%|██████▋ | 29171/43939 [47:04:35<23:24:32, 5.71s/it]
66%|██████▋ | 29172/43939 [47:04:41<23:23:06, 5.70s/it]
66%|██████▋ | 29173/43939 [47:04:46<23:24:41, 5.71s/it]
66%|██████▋ | 29174/43939 [47:04:52<23:20:55, 5.69s/it]
66%|██████▋ | 29175/43939 [47:04:58<23:21:25, 5.70s/it]
66%|██████▋ | 29176/43939 [47:05:03<23:25:06, 5.71s/it]
66%|██████▋ | 29177/43939 [47:05:09<23:26:12, 5.72s/it]
66%|██████▋ | 29178/43939 [47:05:15<23:24:57, 5.71s/it]
66%|██████▋ | 29179/43939 [47:05:20<23:22:14, 5.70s/it]
66%|██████▋ | 29180/43939 [47:05:26<23:21:55, 5.70s/it]
66%|██████▋ | 29181/43939 [47:05:32<23:37:14, 5.76s/it]
66%|██████▋ | 29182/43939 [47:05:38<23:32:28, 5.74s/it]
66%|██████▋ | 29183/43939 [47:05:43<23:28:45, 5.73s/it]
66%|██████▋ | 29184/43939 [47:05:49<23:24:55, 5.71s/it]
66%|██████▋ | 29185/43939 [47:05:55<23:24:01, 5.71s/it]
66%|██████▋ | 29186/43939 [47:06:01<23:22:44, 5.70s/it]
66%|██████▋ | 29187/43939 [47:06:06<23:21:03, 5.70s/it]
66%|██████▋ | 29188/43939 [47:06:12<23:19:15, 5.69s/it]
66%|██████▋ | 29189/43939 [47:06:18<23:20:52, 5.70s/it]
66%|██████▋ | 29190/43939 [47:06:23<23:20:43, 5.70s/it]
66%|██████▋ | 29191/43939 [47:06:29<23:18:56, 5.69s/it]
66%|██████▋ | 29192/43939 [47:06:35<23:17:58, 5.69s/it]
66%|██████▋ | 29193/43939 [47:06:40<23:17:18, 5.69s/it]
66%|██████▋ | 29194/43939 [47:06:46<23:17:55, 5.69s/it]
66%|██████▋ | 29195/43939 [47:06:52<23:16:54, 5.68s/it]
66%|██████▋ | 29196/43939 [47:06:57<23:20:46, 5.70s/it]
66%|██████▋ | 29197/43939 [47:07:03<23:20:31, 5.70s/it]
66%|██████▋ | 29198/43939 [47:07:09<23:20:19, 5.70s/it]
66%|██████▋ | 29199/43939 [47:07:15<23:23:32, 5.71s/it]
66%|██████▋ | 29200/43939 [47:07:20<23:25:07, 5.72s/it]
{'loss': 0.7576, 'grad_norm': 1.232396125793457, 'learning_rate': 1.5268480173455908e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 457.72, 'epoch': 2.66}
66%|██████▋ | 29200/43939 [47:07:20<23:25:07, 5.72s/it]
66%|██████▋ | 29201/43939 [47:07:26<23:26:44, 5.73s/it]
66%|██████▋ | 29202/43939 [47:07:32<23:42:28, 5.79s/it]
66%|██████▋ | 29203/43939 [47:07:38<23:37:24, 5.77s/it]
66%|██████▋ | 29204/43939 [47:07:43<23:34:08, 5.76s/it]
66%|██████▋ | 29205/43939 [47:07:49<23:29:55, 5.74s/it]
66%|██████▋ | 29206/43939 [47:07:55<23:26:52, 5.73s/it]
66%|██████▋ | 29207/43939 [47:08:01<23:27:33, 5.73s/it]
66%|██████▋ | 29208/43939 [47:08:06<23:23:12, 5.72s/it]
66%|██████▋ | 29209/43939 [47:08:12<23:24:17, 5.72s/it]
66%|██████▋ | 29210/43939 [47:08:18<23:24:41, 5.72s/it]
66%|██████▋ | 29211/43939 [47:08:23<23:24:26, 5.72s/it]
66%|██████▋ | 29212/43939 [47:08:29<23:24:22, 5.72s/it]
66%|██████▋ | 29213/43939 [47:08:35<23:23:56, 5.72s/it]
66%|██████▋ | 29214/43939 [47:08:41<23:23:14, 5.72s/it]
66%|██████▋ | 29215/43939 [47:08:46<23:19:40, 5.70s/it]
66%|██████▋ | 29216/43939 [47:08:52<23:22:47, 5.72s/it]
66%|██████▋ | 29217/43939 [47:08:58<23:23:22, 5.72s/it]
66%|██████▋ | 29218/43939 [47:09:03<23:25:02, 5.73s/it]
66%|██████▋ | 29219/43939 [47:09:09<23:24:33, 5.73s/it]
67%|██████▋ | 29220/43939 [47:09:15<23:23:07, 5.72s/it]
67%|██████▋ | 29221/43939 [47:09:21<23:20:03, 5.71s/it]
67%|██████▋ | 29222/43939 [47:09:26<23:22:29, 5.72s/it]
67%|██████▋ | 29223/43939 [47:09:32<23:43:48, 5.81s/it]
67%|██████▋ | 29224/43939 [47:09:38<23:35:16, 5.77s/it]
67%|██████▋ | 29225/43939 [47:09:44<23:33:00, 5.76s/it]
67%|██████▋ | 29226/43939 [47:09:50<23:31:04, 5.75s/it]
67%|██████▋ | 29227/43939 [47:09:55<23:27:03, 5.74s/it]
67%|██████▋ | 29228/43939 [47:10:01<23:27:22, 5.74s/it]
67%|██████▋ | 29229/43939 [47:10:07<23:23:52, 5.73s/it]
67%|██████▋ | 29230/43939 [47:10:12<23:24:11, 5.73s/it]
67%|██████▋ | 29231/43939 [47:10:18<23:22:08, 5.72s/it]
67%|██████▋ | 29232/43939 [47:10:24<23:19:07, 5.71s/it]
67%|██████▋ | 29233/43939 [47:10:29<23:18:37, 5.71s/it]
67%|██████▋ | 29234/43939 [47:10:35<23:19:08, 5.71s/it]
67%|██████▋ | 29235/43939 [47:10:41<23:16:29, 5.70s/it]
67%|██████▋ | 29236/43939 [47:10:47<23:18:13, 5.71s/it]
67%|██████▋ | 29237/43939 [47:10:52<23:19:10, 5.71s/it]
67%|██████▋ | 29238/43939 [47:10:58<23:18:19, 5.71s/it]
67%|██████▋ | 29239/43939 [47:11:04<23:15:34, 5.70s/it]
67%|██████▋ | 29240/43939 [47:11:09<23:16:59, 5.70s/it]
67%|██████▋ | 29241/43939 [47:11:15<23:14:59, 5.69s/it]
67%|██████▋ | 29242/43939 [47:11:21<23:13:26, 5.69s/it]
67%|██████▋ | 29243/43939 [47:11:26<23:14:39, 5.69s/it]
67%|██████▋ | 29244/43939 [47:11:32<23:30:32, 5.76s/it]
67%|██████▋ | 29245/43939 [47:11:38<23:26:33, 5.74s/it]
67%|██████▋ | 29246/43939 [47:11:44<23:22:30, 5.73s/it]
67%|██████▋ | 29247/43939 [47:11:50<23:23:54, 5.73s/it]
67%|██████▋ | 29248/43939 [47:11:55<23:23:00, 5.73s/it]
67%|██████▋ | 29249/43939 [47:12:01<23:18:28, 5.71s/it]
67%|██████▋ | 29250/43939 [47:12:07<23:22:40, 5.73s/it]
{'loss': 0.7793, 'grad_norm': 1.3689206838607788, 'learning_rate': 1.5177087512450109e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.38, 'epoch': 2.66}
67%|██████▋ | 29250/43939 [47:12:07<23:22:40, 5.73s/it]
67%|██████▋ | 29251/43939 [47:12:12<23:19:17, 5.72s/it]
67%|██████▋ | 29252/43939 [47:12:18<23:16:24, 5.70s/it]
67%|██████▋ | 29253/43939 [47:12:24<23:15:39, 5.70s/it]
67%|██████▋ | 29254/43939 [47:12:29<23:17:19, 5.71s/it]
67%|██████▋ | 29255/43939 [47:12:35<23:14:00, 5.70s/it]
67%|██████▋ | 29256/43939 [47:12:41<23:18:17, 5.71s/it]
67%|██████▋ | 29257/43939 [47:12:47<23:16:55, 5.71s/it]
67%|██████▋ | 29258/43939 [47:12:52<23:17:34, 5.71s/it]
67%|██████▋ | 29259/43939 [47:12:58<23:18:09, 5.71s/it]
67%|██████▋ | 29260/43939 [47:13:04<23:17:07, 5.71s/it]
67%|██████▋ | 29261/43939 [47:13:09<23:17:54, 5.71s/it]
67%|██████▋ | 29262/43939 [47:13:15<23:19:16, 5.72s/it]
67%|██████▋ | 29263/43939 [47:13:21<23:15:52, 5.71s/it]
67%|██████▋ | 29264/43939 [47:13:27<23:19:30, 5.72s/it]
67%|██████▋ | 29265/43939 [47:13:32<23:19:10, 5.72s/it]
67%|██████▋ | 29266/43939 [47:13:38<23:33:51, 5.78s/it]
67%|██████▋ | 29267/43939 [47:13:44<23:29:11, 5.76s/it]
67%|██████▋ | 29268/43939 [47:13:50<23:27:21, 5.76s/it]
67%|██████▋ | 29269/43939 [47:13:55<23:20:40, 5.73s/it]
67%|██████▋ | 29270/43939 [47:14:01<23:17:27, 5.72s/it]
67%|██████▋ | 29271/43939 [47:14:07<23:14:29, 5.70s/it]
67%|██████▋ | 29272/43939 [47:14:12<23:12:01, 5.69s/it]
67%|██████▋ | 29273/43939 [47:14:18<23:15:56, 5.71s/it]
67%|██████▋ | 29274/43939 [47:14:24<23:16:52, 5.72s/it]
67%|██████▋ | 29275/43939 [47:14:30<23:15:42, 5.71s/it]
67%|██████▋ | 29276/43939 [47:14:35<23:13:18, 5.70s/it]
67%|██████▋ | 29277/43939 [47:14:41<23:12:02, 5.70s/it]
67%|██████▋ | 29278/43939 [47:14:47<23:14:38, 5.71s/it]
67%|██████▋ | 29279/43939 [47:14:52<23:17:07, 5.72s/it]
67%|██████▋ | 29280/43939 [47:14:58<23:19:03, 5.73s/it]
67%|██████▋ | 29281/43939 [47:15:04<23:14:42, 5.71s/it]
67%|██████▋ | 29282/43939 [47:15:10<23:14:49, 5.71s/it]
67%|██████▋ | 29283/43939 [47:15:15<23:13:58, 5.71s/it]
67%|██████▋ | 29284/43939 [47:15:21<23:13:16, 5.70s/it]
67%|██████▋ | 29285/43939 [47:15:27<23:12:44, 5.70s/it]
67%|██████▋ | 29286/43939 [47:15:32<23:11:11, 5.70s/it]
67%|██████▋ | 29287/43939 [47:15:38<23:13:16, 5.71s/it]
67%|██████▋ | 29288/43939 [47:15:44<23:33:56, 5.79s/it]
67%|██████▋ | 29289/43939 [47:15:50<23:27:29, 5.76s/it]
67%|██████▋ | 29290/43939 [47:15:55<23:22:45, 5.75s/it]
67%|██████▋ | 29291/43939 [47:16:01<23:18:36, 5.73s/it]
67%|██████▋ | 29292/43939 [47:16:07<23:14:01, 5.71s/it]
67%|██████▋ | 29293/43939 [47:16:12<23:11:20, 5.70s/it]
67%|██████▋ | 29294/43939 [47:16:18<23:09:55, 5.69s/it]
67%|██████▋ | 29295/43939 [47:16:24<23:07:23, 5.68s/it]
67%|██████▋ | 29296/43939 [47:16:30<23:10:31, 5.70s/it]
67%|██████▋ | 29297/43939 [47:16:35<23:10:08, 5.70s/it]
67%|██████▋ | 29298/43939 [47:16:41<23:12:13, 5.71s/it]
67%|██████▋ | 29299/43939 [47:16:47<23:10:20, 5.70s/it]
67%|██████▋ | 29300/43939 [47:16:52<23:10:02, 5.70s/it]
{'loss': 0.7685, 'grad_norm': 1.5284311771392822, 'learning_rate': 1.508584983097233e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.96, 'epoch': 2.67}
67%|██████▋ | 29300/43939 [47:16:52<23:10:02, 5.70s/it]
67%|██████▋ | 29301/43939 [47:16:58<23:11:06, 5.70s/it]
67%|██████▋ | 29302/43939 [47:17:04<23:12:46, 5.71s/it]
67%|██████▋ | 29303/43939 [47:17:09<23:12:53, 5.71s/it]
67%|██████▋ | 29304/43939 [47:17:15<23:12:20, 5.71s/it]
67%|██████▋ | 29305/43939 [47:17:21<23:11:40, 5.71s/it]
67%|██████▋ | 29306/43939 [47:17:27<23:11:49, 5.71s/it]
67%|██████▋ | 29307/43939 [47:17:32<23:12:21, 5.71s/it]
67%|██████▋ | 29308/43939 [47:17:38<23:10:08, 5.70s/it]
67%|██████▋ | 29309/43939 [47:17:44<23:27:18, 5.77s/it]
67%|██████▋ | 29310/43939 [47:17:50<23:22:33, 5.75s/it]
67%|██████▋ | 29311/43939 [47:17:55<23:23:19, 5.76s/it]
67%|██████▋ | 29312/43939 [47:18:01<23:19:47, 5.74s/it]
67%|██████▋ | 29313/43939 [47:18:07<23:15:14, 5.72s/it]
67%|██████▋ | 29314/43939 [47:18:13<23:14:16, 5.72s/it]
67%|██████▋ | 29315/43939 [47:18:18<23:13:35, 5.72s/it]
67%|██████▋ | 29316/43939 [47:18:24<23:12:32, 5.71s/it]
67%|██████▋ | 29317/43939 [47:18:30<23:11:33, 5.71s/it]
67%|██████▋ | 29318/43939 [47:18:35<23:10:09, 5.70s/it]
67%|██████▋ | 29319/43939 [47:18:41<23:10:23, 5.71s/it]
67%|██████▋ | 29320/43939 [47:18:47<23:15:25, 5.73s/it]
67%|██████▋ | 29321/43939 [47:18:53<23:13:49, 5.72s/it]
67%|██████▋ | 29322/43939 [47:18:58<23:12:50, 5.72s/it]
67%|██████▋ | 29323/43939 [47:19:04<23:12:01, 5.71s/it]
67%|██████▋ | 29324/43939 [47:19:10<23:13:42, 5.72s/it]
67%|██████▋ | 29325/43939 [47:19:15<23:14:37, 5.73s/it]
67%|██████▋ | 29326/43939 [47:19:21<23:16:50, 5.74s/it]
67%|██████▋ | 29327/43939 [47:19:27<23:11:56, 5.72s/it]
67%|██████▋ | 29328/43939 [47:19:33<23:09:32, 5.71s/it]
67%|██████▋ | 29329/43939 [47:19:38<23:09:21, 5.71s/it]
67%|██████▋ | 29330/43939 [47:19:44<23:28:26, 5.78s/it]
67%|██████▋ | 29331/43939 [47:19:50<23:24:00, 5.77s/it]
67%|██████▋ | 29332/43939 [47:19:56<23:18:45, 5.75s/it]
67%|██████▋ | 29333/43939 [47:20:01<23:17:08, 5.74s/it]
67%|██████▋ | 29334/43939 [47:20:07<23:13:08, 5.72s/it]
67%|██████▋ | 29335/43939 [47:20:13<23:12:13, 5.72s/it]
67%|██████▋ | 29336/43939 [47:20:18<23:14:39, 5.73s/it]
67%|██████▋ | 29337/43939 [47:20:24<23:14:56, 5.73s/it]
67%|██████▋ | 29338/43939 [47:20:30<23:13:09, 5.72s/it]
67%|██████▋ | 29339/43939 [47:20:36<23:09:15, 5.71s/it]
67%|██████▋ | 29340/43939 [47:20:41<23:13:44, 5.73s/it]
67%|██████▋ | 29341/43939 [47:20:47<23:12:40, 5.72s/it]
67%|██████▋ | 29342/43939 [47:20:53<23:14:35, 5.73s/it]
67%|██████▋ | 29343/43939 [47:20:59<23:12:23, 5.72s/it]
67%|██████▋ | 29344/43939 [47:21:04<23:09:57, 5.71s/it]
67%|██████▋ | 29345/43939 [47:21:10<23:07:08, 5.70s/it]
67%|██████▋ | 29346/43939 [47:21:16<23:08:41, 5.71s/it]
67%|██████▋ | 29347/43939 [47:21:21<23:09:42, 5.71s/it]
67%|██████▋ | 29348/43939 [47:21:27<23:08:24, 5.71s/it]
67%|██████▋ | 29349/43939 [47:21:33<23:08:42, 5.71s/it]
67%|██████▋ | 29350/43939 [47:21:38<23:06:21, 5.70s/it]
{'loss': 0.7908, 'grad_norm': 1.5343505144119263, 'learning_rate': 1.4994768568511397e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 516.69, 'epoch': 2.67}
67%|██████▋ | 29350/43939 [47:21:38<23:06:21, 5.70s/it]
67%|██████▋ | 29351/43939 [47:21:44<23:09:44, 5.72s/it]
67%|██████▋ | 29352/43939 [47:21:50<23:24:57, 5.78s/it]
67%|██████▋ | 29353/43939 [47:21:56<23:19:56, 5.76s/it]
67%|██████▋ | 29354/43939 [47:22:02<23:16:25, 5.74s/it]
67%|██████▋ | 29355/43939 [47:22:07<23:14:25, 5.74s/it]
67%|██████▋ | 29356/43939 [47:22:13<23:12:00, 5.73s/it]
67%|██████▋ | 29357/43939 [47:22:19<23:09:05, 5.72s/it]
67%|██████▋ | 29358/43939 [47:22:24<23:10:11, 5.72s/it]
67%|██████▋ | 29359/43939 [47:22:30<23:09:19, 5.72s/it]
67%|██████▋ | 29360/43939 [47:22:36<23:08:08, 5.71s/it]
67%|██████▋ | 29361/43939 [47:22:42<23:07:58, 5.71s/it]
67%|██████▋ | 29362/43939 [47:22:47<23:06:59, 5.71s/it]
67%|██████▋ | 29363/43939 [47:22:53<23:06:47, 5.71s/it]
67%|██████▋ | 29364/43939 [47:22:59<23:08:01, 5.71s/it]
67%|██████▋ | 29365/43939 [47:23:04<23:06:21, 5.71s/it]
67%|██████▋ | 29366/43939 [47:23:10<23:05:50, 5.71s/it]
67%|██████▋ | 29367/43939 [47:23:16<23:04:38, 5.70s/it]
67%|██████▋ | 29368/43939 [47:23:21<23:00:44, 5.69s/it]
67%|██████▋ | 29369/43939 [47:23:27<23:05:41, 5.71s/it]
67%|██████▋ | 29370/43939 [47:23:33<23:04:22, 5.70s/it]
67%|██████▋ | 29371/43939 [47:23:39<23:03:54, 5.70s/it]
67%|██████▋ | 29372/43939 [47:23:44<23:04:39, 5.70s/it]
67%|██████▋ | 29373/43939 [47:23:50<23:07:38, 5.72s/it]
67%|██████▋ | 29374/43939 [47:23:56<23:25:29, 5.79s/it]
67%|██████▋ | 29375/43939 [47:24:02<23:17:21, 5.76s/it]
67%|██████▋ | 29376/43939 [47:24:07<23:12:11, 5.74s/it]
67%|██████▋ | 29377/43939 [47:24:13<23:09:57, 5.73s/it]
67%|██████▋ | 29378/43939 [47:24:19<23:11:11, 5.73s/it]
67%|██████▋ | 29379/43939 [47:24:25<23:10:55, 5.73s/it]
67%|██████▋ | 29380/43939 [47:24:30<23:09:22, 5.73s/it]
67%|██████▋ | 29381/43939 [47:24:36<23:08:32, 5.72s/it]
67%|██████▋ | 29382/43939 [47:24:42<23:06:44, 5.72s/it]
67%|██████▋ | 29383/43939 [47:24:47<23:02:44, 5.70s/it]
67%|██████▋ | 29384/43939 [47:24:53<23:03:21, 5.70s/it]
67%|██████▋ | 29385/43939 [47:24:59<23:03:51, 5.71s/it]
67%|██████▋ | 29386/43939 [47:25:04<23:01:35, 5.70s/it]
67%|██████▋ | 29387/43939 [47:25:10<23:01:11, 5.69s/it]
67%|██████▋ | 29388/43939 [47:25:16<23:02:37, 5.70s/it]
67%|██████▋ | 29389/43939 [47:25:21<23:01:34, 5.70s/it]
67%|██████▋ | 29390/43939 [47:25:27<23:00:59, 5.70s/it]
67%|██████▋ | 29391/43939 [47:25:33<23:01:16, 5.70s/it]
67%|██████▋ | 29392/43939 [47:25:39<22:59:14, 5.69s/it]
67%|██████▋ | 29393/43939 [47:25:44<23:01:42, 5.70s/it]
67%|██████▋ | 29394/43939 [47:25:50<22:58:57, 5.69s/it]
67%|██████▋ | 29395/43939 [47:25:56<23:16:53, 5.76s/it]
67%|██████▋ | 29396/43939 [47:26:02<23:17:08, 5.76s/it]
67%|██████▋ | 29397/43939 [47:26:07<23:13:47, 5.75s/it]
67%|██████▋ | 29398/43939 [47:26:13<23:10:59, 5.74s/it]
67%|██████▋ | 29399/43939 [47:26:19<23:06:24, 5.72s/it]
67%|██████▋ | 29400/43939 [47:26:25<23:08:53, 5.73s/it]
{'loss': 0.7841, 'grad_norm': 1.3912273645401, 'learning_rate': 1.490384516208827e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.58, 'epoch': 2.68}
67%|██████▋ | 29400/43939 [47:26:25<23:08:53, 5.73s/it]
67%|██████▋ | 29401/43939 [47:26:30<23:13:07, 5.75s/it]
67%|██████▋ | 29402/43939 [47:26:36<23:08:00, 5.73s/it]
67%|██████▋ | 29403/43939 [47:26:42<23:05:55, 5.72s/it]
67%|██████▋ | 29404/43939 [47:26:47<23:04:17, 5.71s/it]
67%|██████▋ | 29405/43939 [47:26:53<23:03:11, 5.71s/it]
67%|██████▋ | 29406/43939 [47:26:59<23:01:16, 5.70s/it]
67%|██████▋ | 29407/43939 [47:27:04<23:01:46, 5.71s/it]
67%|██████▋ | 29408/43939 [47:27:10<23:01:56, 5.71s/it]
67%|██████▋ | 29409/43939 [47:27:16<22:59:54, 5.70s/it]
67%|██████▋ | 29410/43939 [47:27:22<23:01:15, 5.70s/it]
67%|██████▋ | 29411/43939 [47:27:27<22:58:59, 5.70s/it]
67%|██████▋ | 29412/43939 [47:27:33<23:01:30, 5.71s/it]
67%|██████▋ | 29413/43939 [47:27:39<23:04:09, 5.72s/it]
67%|██████▋ | 29414/43939 [47:27:44<23:02:28, 5.71s/it]
67%|██████▋ | 29415/43939 [47:27:50<23:04:28, 5.72s/it]
67%|██████▋ | 29416/43939 [47:27:56<23:17:57, 5.78s/it]
67%|██████▋ | 29417/43939 [47:28:02<23:08:36, 5.74s/it]
67%|██████▋ | 29418/43939 [47:28:07<23:06:52, 5.73s/it]
67%|██████▋ | 29419/43939 [47:28:13<23:04:22, 5.72s/it]
67%|██████▋ | 29420/43939 [47:28:19<23:03:15, 5.72s/it]
67%|██████▋ | 29421/43939 [47:28:25<22:59:06, 5.70s/it]
67%|██████▋ | 29422/43939 [47:28:30<22:59:24, 5.70s/it]
67%|██████▋ | 29423/43939 [47:28:36<22:57:16, 5.69s/it]
67%|██████▋ | 29424/43939 [47:28:42<22:58:52, 5.70s/it]
67%|██████▋ | 29425/43939 [47:28:47<22:58:11, 5.70s/it]
67%|██████▋ | 29426/43939 [47:28:53<22:58:18, 5.70s/it]
67%|██████▋ | 29427/43939 [47:28:59<22:59:22, 5.70s/it]
67%|██████▋ | 29428/43939 [47:29:04<23:03:32, 5.72s/it]
67%|██████▋ | 29429/43939 [47:29:10<23:04:24, 5.72s/it]
67%|██████▋ | 29430/43939 [47:29:16<23:06:17, 5.73s/it]
67%|██████▋ | 29431/43939 [47:29:22<23:05:06, 5.73s/it]
67%|██████▋ | 29432/43939 [47:29:27<23:02:20, 5.72s/it]
67%|██████▋ | 29433/43939 [47:29:33<23:04:23, 5.73s/it]
67%|██████▋ | 29434/43939 [47:29:39<22:59:18, 5.71s/it]
67%|██████▋ | 29435/43939 [47:29:45<23:04:25, 5.73s/it]
67%|██████▋ | 29436/43939 [47:29:50<23:04:08, 5.73s/it]
67%|██████▋ | 29437/43939 [47:29:56<23:00:22, 5.71s/it]
67%|██████▋ | 29438/43939 [47:30:02<23:14:55, 5.77s/it]
67%|██████▋ | 29439/43939 [47:30:08<23:14:07, 5.77s/it]
67%|██████▋ | 29440/43939 [47:30:13<23:10:58, 5.76s/it]
67%|██████▋ | 29441/43939 [47:30:19<23:07:06, 5.74s/it]
67%|██████▋ | 29442/43939 [47:30:25<23:01:54, 5.72s/it]
67%|██████▋ | 29443/43939 [47:30:30<22:57:56, 5.70s/it]
67%|██████▋ | 29444/43939 [47:30:36<22:57:34, 5.70s/it]
67%|██████▋ | 29445/43939 [47:30:42<22:56:30, 5.70s/it]
67%|██████▋ | 29446/43939 [47:30:48<22:59:41, 5.71s/it]
67%|██████▋ | 29447/43939 [47:30:53<22:59:46, 5.71s/it]
67%|██████▋ | 29448/43939 [47:30:59<23:03:50, 5.73s/it]
67%|██████▋ | 29449/43939 [47:31:05<23:04:25, 5.73s/it]
67%|██████▋ | 29450/43939 [47:31:10<23:00:21, 5.72s/it]
{'loss': 0.7732, 'grad_norm': 1.5467596054077148, 'learning_rate': 1.4813081046233345e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.35, 'epoch': 2.68}
67%|██████▋ | 29450/43939 [47:31:10<23:00:21, 5.72s/it]
67%|██████▋ | 29451/43939 [47:31:16<23:00:30, 5.72s/it]
67%|██████▋ | 29452/43939 [47:31:22<23:00:05, 5.72s/it]
67%|██████▋ | 29453/43939 [47:31:28<22:58:55, 5.71s/it]
67%|██████▋ | 29454/43939 [47:31:33<22:55:48, 5.70s/it]
67%|██████▋ | 29455/43939 [47:31:39<23:02:16, 5.73s/it]
67%|██████▋ | 29456/43939 [47:31:45<23:00:34, 5.72s/it]
67%|██████▋ | 29457/43939 [47:31:50<23:01:22, 5.72s/it]
67%|██████▋ | 29458/43939 [47:31:56<22:59:15, 5.71s/it]
67%|██████▋ | 29459/43939 [47:32:02<22:57:03, 5.71s/it]
67%|██████▋ | 29460/43939 [47:32:08<23:13:40, 5.78s/it]
67%|██████▋ | 29461/43939 [47:32:13<23:08:38, 5.75s/it]
67%|██████▋ | 29462/43939 [47:32:19<23:05:39, 5.74s/it]
67%|██████▋ | 29463/43939 [47:32:25<23:01:07, 5.72s/it]
67%|██████▋ | 29464/43939 [47:32:31<23:02:03, 5.73s/it]
67%|██████▋ | 29465/43939 [47:32:36<23:03:10, 5.73s/it]
67%|██████▋ | 29466/43939 [47:32:42<23:00:36, 5.72s/it]
67%|██████▋ | 29467/43939 [47:32:48<22:59:46, 5.72s/it]
67%|██████▋ | 29468/43939 [47:32:53<22:59:37, 5.72s/it]
67%|██████▋ | 29469/43939 [47:32:59<22:59:05, 5.72s/it]
67%|██████▋ | 29470/43939 [47:33:05<22:55:42, 5.70s/it]
67%|██████▋ | 29471/43939 [47:33:11<22:57:03, 5.71s/it]
67%|██████▋ | 29472/43939 [47:33:16<22:59:24, 5.72s/it]
67%|██████▋ | 29473/43939 [47:33:22<22:59:28, 5.72s/it]
67%|██████▋ | 29474/43939 [47:33:28<22:57:15, 5.71s/it]
67%|██████▋ | 29475/43939 [47:33:33<22:57:20, 5.71s/it]
67%|██████▋ | 29476/43939 [47:33:39<22:57:43, 5.72s/it]
67%|██████▋ | 29477/43939 [47:33:45<22:57:35, 5.72s/it]
67%|██████▋ | 29478/43939 [47:33:51<22:56:10, 5.71s/it]
67%|██████▋ | 29479/43939 [47:33:56<22:56:45, 5.71s/it]
67%|██████▋ | 29480/43939 [47:34:02<22:55:21, 5.71s/it]
67%|██████▋ | 29481/43939 [47:34:08<23:19:03, 5.81s/it]
67%|██████▋ | 29482/43939 [47:34:14<23:12:01, 5.78s/it]
67%|██████▋ | 29483/43939 [47:34:19<23:05:10, 5.75s/it]
67%|██████▋ | 29484/43939 [47:34:25<23:00:24, 5.73s/it]
67%|██████▋ | 29485/43939 [47:34:31<22:57:20, 5.72s/it]
67%|██████▋ | 29486/43939 [47:34:37<22:56:48, 5.72s/it]
67%|██████▋ | 29487/43939 [47:34:42<22:57:29, 5.72s/it]
67%|██████▋ | 29488/43939 [47:34:48<22:54:34, 5.71s/it]
67%|██████▋ | 29489/43939 [47:34:54<22:52:49, 5.70s/it]
67%|██████▋ | 29490/43939 [47:34:59<22:51:28, 5.70s/it]
67%|██████▋ | 29491/43939 [47:35:05<22:50:02, 5.69s/it]
67%|██████▋ | 29492/43939 [47:35:11<22:50:24, 5.69s/it]
67%|██████▋ | 29493/43939 [47:35:16<22:48:44, 5.68s/it]
67%|██████▋ | 29494/43939 [47:35:22<22:49:07, 5.69s/it]
67%|██████▋ | 29495/43939 [47:35:28<22:51:03, 5.70s/it]
67%|██████▋ | 29496/43939 [47:35:33<22:47:44, 5.68s/it]
67%|██████▋ | 29497/43939 [47:35:39<22:49:54, 5.69s/it]
67%|██████▋ | 29498/43939 [47:35:45<22:51:24, 5.70s/it]
67%|██████▋ | 29499/43939 [47:35:51<22:52:31, 5.70s/it]
67%|██████▋ | 29500/43939 [47:35:56<22:53:01, 5.71s/it]
{'loss': 0.7824, 'grad_norm': 1.345490574836731, 'learning_rate': 1.4722477652963845e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.75, 'epoch': 2.69}
67%|██████▋ | 29500/43939 [47:35:56<22:53:01, 5.71s/it]
67%|██████▋ | 29501/43939 [47:36:02<22:55:03, 5.71s/it]
67%|██████▋ | 29502/43939 [47:36:08<23:10:39, 5.78s/it]
67%|██████▋ | 29503/43939 [47:36:14<23:03:25, 5.75s/it]
67%|██████▋ | 29504/43939 [47:36:19<22:58:10, 5.73s/it]
67%|██████▋ | 29505/43939 [47:36:25<22:55:04, 5.72s/it]
67%|██████▋ | 29506/43939 [47:36:31<22:53:24, 5.71s/it]
67%|██████▋ | 29507/43939 [47:36:36<22:53:15, 5.71s/it]
67%|██████▋ | 29508/43939 [47:36:42<22:54:56, 5.72s/it]
67%|██████▋ | 29509/43939 [47:36:48<22:53:39, 5.71s/it]
67%|██████▋ | 29510/43939 [47:36:54<22:54:59, 5.72s/it]
67%|██████▋ | 29511/43939 [47:36:59<22:54:59, 5.72s/it]
67%|██████▋ | 29512/43939 [47:37:05<22:54:57, 5.72s/it]
67%|██████▋ | 29513/43939 [47:37:11<22:53:12, 5.71s/it]
67%|██████▋ | 29514/43939 [47:37:16<22:53:03, 5.71s/it]
67%|██████▋ | 29515/43939 [47:37:22<22:52:53, 5.71s/it]
67%|██████▋ | 29516/43939 [47:37:28<22:51:28, 5.71s/it]
67%|██████▋ | 29517/43939 [47:37:33<22:51:58, 5.71s/it]
67%|██████▋ | 29518/43939 [47:37:39<22:52:28, 5.71s/it]
67%|██████▋ | 29519/43939 [47:37:45<22:56:50, 5.73s/it]
67%|██████▋ | 29520/43939 [47:37:51<22:55:28, 5.72s/it]
67%|██████▋ | 29521/43939 [47:37:56<22:52:42, 5.71s/it]
67%|██████▋ | 29522/43939 [47:38:02<22:48:50, 5.70s/it]
67%|██████▋ | 29523/43939 [47:38:08<23:05:27, 5.77s/it]
67%|██████▋ | 29524/43939 [47:38:14<23:00:31, 5.75s/it]
67%|██████▋ | 29525/43939 [47:38:19<22:56:47, 5.73s/it]
67%|██████▋ | 29526/43939 [47:38:25<22:54:58, 5.72s/it]
67%|██████▋ | 29527/43939 [47:38:31<22:52:49, 5.72s/it]
67%|██████▋ | 29528/43939 [47:38:36<22:51:30, 5.71s/it]
67%|██████▋ | 29529/43939 [47:38:42<22:49:36, 5.70s/it]
67%|██████▋ | 29530/43939 [47:38:48<22:53:24, 5.72s/it]
67%|██████▋ | 29531/43939 [47:38:54<22:51:07, 5.71s/it]
67%|██████▋ | 29532/43939 [47:38:59<22:50:34, 5.71s/it]
67%|██████▋ | 29533/43939 [47:39:05<22:51:05, 5.71s/it]
67%|██████▋ | 29534/43939 [47:39:11<22:49:57, 5.71s/it]
67%|██████▋ | 29535/43939 [47:39:16<22:51:18, 5.71s/it]
67%|██████▋ | 29536/43939 [47:39:22<22:49:27, 5.70s/it]
67%|██████▋ | 29537/43939 [47:39:28<22:51:37, 5.71s/it]
67%|██████▋ | 29538/43939 [47:39:34<22:48:43, 5.70s/it]
67%|██████▋ | 29539/43939 [47:39:39<22:45:57, 5.69s/it]
67%|██████▋ | 29540/43939 [47:39:45<22:45:24, 5.69s/it]
67%|██████▋ | 29541/43939 [47:39:51<22:46:30, 5.69s/it]
67%|██████▋ | 29542/43939 [47:39:56<22:49:12, 5.71s/it]
67%|██████▋ | 29543/43939 [47:40:02<22:53:13, 5.72s/it]
67%|██████▋ | 29544/43939 [47:40:08<23:06:00, 5.78s/it]
67%|██████▋ | 29545/43939 [47:40:14<23:04:12, 5.77s/it]
67%|██████▋ | 29546/43939 [47:40:19<23:00:44, 5.76s/it]
67%|██████▋ | 29547/43939 [47:40:25<22:57:17, 5.74s/it]
67%|██████▋ | 29548/43939 [47:40:31<22:53:14, 5.73s/it]
67%|██████▋ | 29549/43939 [47:40:37<22:55:29, 5.74s/it]
67%|██████▋ | 29550/43939 [47:40:42<22:52:59, 5.73s/it]
{'loss': 0.7539, 'grad_norm': 1.342452049255371, 'learning_rate': 1.4632036411761223e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.06, 'epoch': 2.69}
67%|██████▋ | 29550/43939 [47:40:42<22:52:59, 5.73s/it]
67%|██████▋ | 29551/43939 [47:40:48<22:51:24, 5.72s/it]
67%|██████▋ | 29552/43939 [47:40:54<22:51:51, 5.72s/it]
67%|██████▋ | 29553/43939 [47:40:59<22:47:31, 5.70s/it]
67%|██████▋ | 29554/43939 [47:41:05<22:46:48, 5.70s/it]
67%|██████▋ | 29555/43939 [47:41:11<22:50:05, 5.72s/it]
67%|██████▋ | 29556/43939 [47:41:17<22:48:54, 5.71s/it]
67%|██████▋ | 29557/43939 [47:41:22<22:47:34, 5.71s/it]
67%|██████▋ | 29558/43939 [47:41:28<22:47:51, 5.71s/it]
67%|██████▋ | 29559/43939 [47:41:34<22:47:04, 5.70s/it]
67%|██████▋ | 29560/43939 [47:41:39<22:46:24, 5.70s/it]
67%|██████▋ | 29561/43939 [47:41:45<22:46:47, 5.70s/it]
67%|██████▋ | 29562/43939 [47:41:51<22:44:52, 5.70s/it]
67%|██████▋ | 29563/43939 [47:41:56<22:44:46, 5.70s/it]
67%|██████▋ | 29564/43939 [47:42:02<22:45:11, 5.70s/it]
67%|██████▋ | 29565/43939 [47:42:08<23:03:26, 5.77s/it]
67%|██████▋ | 29566/43939 [47:42:14<23:01:05, 5.77s/it]
67%|██████▋ | 29567/43939 [47:42:20<22:55:44, 5.74s/it]
67%|██████▋ | 29568/43939 [47:42:25<22:50:23, 5.72s/it]
67%|██████▋ | 29569/43939 [47:42:31<22:48:15, 5.71s/it]
67%|██████▋ | 29570/43939 [47:42:37<22:46:46, 5.71s/it]
67%|██████▋ | 29571/43939 [47:42:42<22:44:00, 5.70s/it]
67%|██████▋ | 29572/43939 [47:42:48<22:47:13, 5.71s/it]
67%|██████▋ | 29573/43939 [47:42:54<22:45:35, 5.70s/it]
67%|██████▋ | 29574/43939 [47:42:59<22:47:57, 5.71s/it]
67%|██████▋ | 29575/43939 [47:43:05<22:47:47, 5.71s/it]
67%|██████▋ | 29576/43939 [47:43:11<22:47:40, 5.71s/it]
67%|██████▋ | 29577/43939 [47:43:17<22:46:39, 5.71s/it]
67%|██████▋ | 29578/43939 [47:43:22<22:45:17, 5.70s/it]
67%|██████▋ | 29579/43939 [47:43:28<22:42:23, 5.69s/it]
67%|██████▋ | 29580/43939 [47:43:34<22:42:50, 5.69s/it]
67%|██████▋ | 29581/43939 [47:43:39<22:43:56, 5.70s/it]
67%|██████▋ | 29582/43939 [47:43:45<22:43:08, 5.70s/it]
67%|██████▋ | 29583/43939 [47:43:51<22:40:57, 5.69s/it]
67%|██████▋ | 29584/43939 [47:43:56<22:43:50, 5.70s/it]
67%|██████▋ | 29585/43939 [47:44:02<22:48:50, 5.72s/it]
67%|██████▋ | 29586/43939 [47:44:08<22:45:33, 5.71s/it]
67%|██████▋ | 29587/43939 [47:44:14<23:02:34, 5.78s/it]
67%|██████▋ | 29588/43939 [47:44:20<22:56:44, 5.76s/it]
67%|██████▋ | 29589/43939 [47:44:25<22:51:46, 5.74s/it]
67%|██████▋ | 29590/43939 [47:44:31<22:47:08, 5.72s/it]
67%|██████▋ | 29591/43939 [47:44:37<22:43:49, 5.70s/it]
67%|██████▋ | 29592/43939 [47:44:42<22:43:14, 5.70s/it]
67%|██████▋ | 29593/43939 [47:44:48<22:43:33, 5.70s/it]
67%|██████▋ | 29594/43939 [47:44:54<22:40:29, 5.69s/it]
67%|██████▋ | 29595/43939 [47:44:59<22:41:46, 5.70s/it]
67%|██████▋ | 29596/43939 [47:45:05<22:40:54, 5.69s/it]
67%|██████▋ | 29597/43939 [47:45:11<22:42:25, 5.70s/it]
67%|██████▋ | 29598/43939 [47:45:16<22:42:24, 5.70s/it]
67%|██████▋ | 29599/43939 [47:45:22<22:44:04, 5.71s/it]
67%|██████▋ | 29600/43939 [47:45:28<22:43:30, 5.71s/it]
{'loss': 0.7648, 'grad_norm': 1.4391672611236572, 'learning_rate': 1.4541758749548581e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.41, 'epoch': 2.69}
67%|██████▋ | 29600/43939 [47:45:28<22:43:30, 5.71s/it]
67%|██████▋ | 29601/43939 [47:45:34<22:42:02, 5.70s/it]
67%|██████▋ | 29602/43939 [47:45:39<22:42:32, 5.70s/it]
67%|██████▋ | 29603/43939 [47:45:45<22:43:59, 5.71s/it]
67%|██████▋ | 29604/43939 [47:45:51<22:45:38, 5.72s/it]
67%|██████▋ | 29605/43939 [47:45:56<22:45:07, 5.71s/it]
67%|██████▋ | 29606/43939 [47:46:02<22:45:18, 5.72s/it]
67%|██████▋ | 29607/43939 [47:46:08<22:48:07, 5.73s/it]
67%|██████▋ | 29608/43939 [47:46:14<23:06:27, 5.80s/it]
67%|██████▋ | 29609/43939 [47:46:20<22:59:13, 5.77s/it]
67%|██████▋ | 29610/43939 [47:46:25<22:50:03, 5.74s/it]
67%|██████▋ | 29611/43939 [47:46:31<22:49:25, 5.73s/it]
67%|██████▋ | 29612/43939 [47:46:37<22:50:19, 5.74s/it]
67%|██████▋ | 29613/43939 [47:46:42<22:49:45, 5.74s/it]
67%|██████▋ | 29614/43939 [47:46:48<22:47:19, 5.73s/it]
67%|██████▋ | 29615/43939 [47:46:54<22:47:21, 5.73s/it]
67%|██████▋ | 29616/43939 [47:47:00<22:46:19, 5.72s/it]
67%|██████▋ | 29617/43939 [47:47:05<22:44:19, 5.72s/it]
67%|██████▋ | 29618/43939 [47:47:11<22:43:55, 5.71s/it]
67%|██████▋ | 29619/43939 [47:47:17<22:44:26, 5.72s/it]
67%|██████▋ | 29620/43939 [47:47:22<22:44:42, 5.72s/it]
67%|██████▋ | 29621/43939 [47:47:28<22:45:32, 5.72s/it]
67%|██████▋ | 29622/43939 [47:47:34<22:43:38, 5.71s/it]
67%|██████▋ | 29623/43939 [47:47:40<22:40:44, 5.70s/it]
67%|██████▋ | 29624/43939 [47:47:45<22:41:17, 5.71s/it]
67%|██████▋ | 29625/43939 [47:47:51<22:41:36, 5.71s/it]
67%|██████▋ | 29626/43939 [47:47:57<22:39:00, 5.70s/it]
67%|██████▋ | 29627/43939 [47:48:02<22:39:22, 5.70s/it]
67%|██████▋ | 29628/43939 [47:48:08<22:40:14, 5.70s/it]
67%|██████▋ | 29629/43939 [47:48:14<22:57:49, 5.78s/it]
67%|██████▋ | 29630/43939 [47:48:20<22:51:02, 5.75s/it]
67%|██████▋ | 29631/43939 [47:48:25<22:47:47, 5.74s/it]
67%|██████▋ | 29632/43939 [47:48:31<22:42:36, 5.71s/it]
67%|██████▋ | 29633/43939 [47:48:37<22:41:57, 5.71s/it]
67%|██████▋ | 29634/43939 [47:48:42<22:41:52, 5.71s/it]
67%|██████▋ | 29635/43939 [47:48:48<22:42:22, 5.71s/it]
67%|██████▋ | 29636/43939 [47:48:54<22:41:49, 5.71s/it]
67%|██████▋ | 29637/43939 [47:49:00<22:43:49, 5.72s/it]
67%|██████▋ | 29638/43939 [47:49:05<22:44:42, 5.73s/it]
67%|██████▋ | 29639/43939 [47:49:11<22:41:15, 5.71s/it]
67%|██████▋ | 29640/43939 [47:49:17<22:39:59, 5.71s/it]
67%|██████▋ | 29641/43939 [47:49:22<22:39:14, 5.70s/it]
67%|██████▋ | 29642/43939 [47:49:28<22:38:07, 5.70s/it]
67%|██████▋ | 29643/43939 [47:49:34<22:44:12, 5.73s/it]
67%|██████▋ | 29644/43939 [47:49:40<22:41:47, 5.72s/it]
67%|██████▋ | 29645/43939 [47:49:45<22:42:48, 5.72s/it]
67%|██████▋ | 29646/43939 [47:49:51<22:41:59, 5.72s/it]
67%|██████▋ | 29647/43939 [47:49:57<22:41:26, 5.72s/it]
67%|██████▋ | 29648/43939 [47:50:02<22:40:25, 5.71s/it]
67%|██████▋ | 29649/43939 [47:50:08<22:38:42, 5.70s/it]
67%|██████▋ | 29650/43939 [47:50:14<22:49:23, 5.75s/it]
{'loss': 0.7898, 'grad_norm': 1.5603262186050415, 'learning_rate': 1.4451646090668197e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.48, 'epoch': 2.7}
67%|██████▋ | 29650/43939 [47:50:14<22:49:23, 5.75s/it]
67%|██████▋ | 29651/43939 [47:50:20<22:44:51, 5.73s/it]
67%|██████▋ | 29652/43939 [47:50:25<22:44:56, 5.73s/it]
67%|██████▋ | 29653/43939 [47:50:31<22:41:56, 5.72s/it]
67%|██████▋ | 29654/43939 [47:50:37<22:39:22, 5.71s/it]
67%|██████▋ | 29655/43939 [47:50:42<22:36:56, 5.70s/it]
67%|██████▋ | 29656/43939 [47:50:48<22:37:15, 5.70s/it]
67%|██████▋ | 29657/43939 [47:50:54<22:39:24, 5.71s/it]
67%|██████▋ | 29658/43939 [47:51:00<22:37:21, 5.70s/it]
68%|██████▊ | 29659/43939 [47:51:05<22:37:45, 5.70s/it]
68%|██████▊ | 29660/43939 [47:51:11<22:37:36, 5.70s/it]
68%|██████▊ | 29661/43939 [47:51:17<22:40:58, 5.72s/it]
68%|██████▊ | 29662/43939 [47:51:22<22:40:46, 5.72s/it]
68%|██████▊ | 29663/43939 [47:51:28<22:39:26, 5.71s/it]
68%|██████▊ | 29664/43939 [47:51:34<22:38:19, 5.71s/it]
68%|██████▊ | 29665/43939 [47:51:40<22:37:24, 5.71s/it]
68%|██████▊ | 29666/43939 [47:51:45<22:39:19, 5.71s/it]
68%|██████▊ | 29667/43939 [47:51:51<22:38:30, 5.71s/it]
68%|██████▊ | 29668/43939 [47:51:57<22:39:21, 5.72s/it]
68%|██████▊ | 29669/43939 [47:52:02<22:36:36, 5.70s/it]
68%|██████▊ | 29670/43939 [47:52:08<22:37:28, 5.71s/it]
68%|██████▊ | 29671/43939 [47:52:14<22:39:30, 5.72s/it]
68%|██████▊ | 29672/43939 [47:52:20<22:52:51, 5.77s/it]
68%|██████▊ | 29673/43939 [47:52:25<22:47:19, 5.75s/it]
68%|██████▊ | 29674/43939 [47:52:31<22:40:55, 5.72s/it]
68%|██████▊ | 29675/43939 [47:52:37<22:41:56, 5.73s/it]
68%|██████▊ | 29676/43939 [47:52:43<22:44:12, 5.74s/it]
68%|██████▊ | 29677/43939 [47:52:48<22:38:27, 5.72s/it]
68%|██████▊ | 29678/43939 [47:52:54<22:41:03, 5.73s/it]
68%|██████▊ | 29679/43939 [47:53:00<22:38:56, 5.72s/it]
68%|██████▊ | 29680/43939 [47:53:05<22:36:01, 5.71s/it]
68%|██████▊ | 29681/43939 [47:53:11<22:34:03, 5.70s/it]
68%|██████▊ | 29682/43939 [47:53:17<22:35:04, 5.70s/it]
68%|██████▊ | 29683/43939 [47:53:23<22:34:01, 5.70s/it]
68%|██████▊ | 29684/43939 [47:53:28<22:32:32, 5.69s/it]
68%|██████▊ | 29685/43939 [47:53:34<22:34:52, 5.70s/it]
68%|██████▊ | 29686/43939 [47:53:40<22:35:07, 5.70s/it]
68%|██████▊ | 29687/43939 [47:53:45<22:34:27, 5.70s/it]
68%|██████▊ | 29688/43939 [47:53:51<22:33:51, 5.70s/it]
68%|██████▊ | 29689/43939 [47:53:57<22:33:22, 5.70s/it]
68%|██████▊ | 29690/43939 [47:54:02<22:32:17, 5.69s/it]
68%|██████▊ | 29691/43939 [47:54:08<22:32:10, 5.69s/it]
68%|██████▊ | 29692/43939 [47:54:14<22:29:45, 5.68s/it]
68%|██████▊ | 29693/43939 [47:54:20<22:46:49, 5.76s/it]
68%|██████▊ | 29694/43939 [47:54:25<22:42:43, 5.74s/it]
68%|██████▊ | 29695/43939 [47:54:31<22:36:52, 5.72s/it]
68%|██████▊ | 29696/43939 [47:54:37<22:35:02, 5.71s/it]
68%|██████▊ | 29697/43939 [47:54:42<22:34:07, 5.70s/it]
68%|██████▊ | 29698/43939 [47:54:48<22:32:34, 5.70s/it]
68%|██████▊ | 29699/43939 [47:54:54<22:31:40, 5.70s/it]
68%|██████▊ | 29700/43939 [47:54:59<22:30:12, 5.69s/it]
{'loss': 0.7741, 'grad_norm': 1.3951871395111084, 'learning_rate': 1.4361699856859023e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.26, 'epoch': 2.7}
68%|██████▊ | 29700/43939 [47:54:59<22:30:12, 5.69s/it]
68%|██████▊ | 29701/43939 [47:55:05<22:32:37, 5.70s/it]
68%|██████▊ | 29702/43939 [47:55:11<22:33:44, 5.71s/it]
68%|██████▊ | 29703/43939 [47:55:17<22:33:57, 5.71s/it]
68%|██████▊ | 29704/43939 [47:55:22<22:32:25, 5.70s/it]
68%|██████▊ | 29705/43939 [47:55:28<22:31:56, 5.70s/it]
68%|██████▊ | 29706/43939 [47:55:34<22:32:05, 5.70s/it]
68%|██████▊ | 29707/43939 [47:55:39<22:35:49, 5.72s/it]
68%|██████▊ | 29708/43939 [47:55:45<22:33:53, 5.71s/it]
68%|██████▊ | 29709/43939 [47:55:51<22:29:58, 5.69s/it]
68%|██████▊ | 29710/43939 [47:55:57<22:30:31, 5.69s/it]
68%|██████▊ | 29711/43939 [47:56:02<22:30:36, 5.70s/it]
68%|██████▊ | 29712/43939 [47:56:08<22:30:03, 5.69s/it]
68%|██████▊ | 29713/43939 [47:56:14<22:32:18, 5.70s/it]
68%|██████▊ | 29714/43939 [47:56:20<22:49:03, 5.77s/it]
68%|██████▊ | 29715/43939 [47:56:25<22:41:24, 5.74s/it]
68%|██████▊ | 29716/43939 [47:56:31<22:39:53, 5.74s/it]
68%|██████▊ | 29717/43939 [47:56:37<22:37:40, 5.73s/it]
68%|██████▊ | 29718/43939 [47:56:42<22:34:31, 5.71s/it]
68%|██████▊ | 29719/43939 [47:56:48<22:32:59, 5.71s/it]
68%|██████▊ | 29720/43939 [47:56:54<22:32:23, 5.71s/it]
68%|██████▊ | 29721/43939 [47:56:59<22:33:11, 5.71s/it]
68%|██████▊ | 29722/43939 [47:57:05<22:30:15, 5.70s/it]
68%|██████▊ | 29723/43939 [47:57:11<22:29:02, 5.69s/it]
68%|██████▊ | 29724/43939 [47:57:17<22:32:34, 5.71s/it]
68%|██████▊ | 29725/43939 [47:57:22<22:33:14, 5.71s/it]
68%|██████▊ | 29726/43939 [47:57:28<22:29:30, 5.70s/it]
68%|██████▊ | 29727/43939 [47:57:34<22:32:31, 5.71s/it]
68%|██████▊ | 29728/43939 [47:57:39<22:32:47, 5.71s/it]
68%|██████▊ | 29729/43939 [47:57:45<22:31:46, 5.71s/it]
68%|██████▊ | 29730/43939 [47:57:51<22:29:48, 5.70s/it]
68%|██████▊ | 29731/43939 [47:57:56<22:27:19, 5.69s/it]
68%|██████▊ | 29732/43939 [47:58:02<22:28:38, 5.70s/it]
68%|██████▊ | 29733/43939 [47:58:08<22:28:30, 5.70s/it]
68%|██████▊ | 29734/43939 [47:58:14<22:28:12, 5.69s/it]
68%|██████▊ | 29735/43939 [47:58:19<22:26:30, 5.69s/it]
68%|██████▊ | 29736/43939 [47:58:25<22:48:03, 5.78s/it]
68%|██████▊ | 29737/43939 [47:58:31<22:40:04, 5.75s/it]
68%|██████▊ | 29738/43939 [47:58:37<22:41:44, 5.75s/it]
68%|██████▊ | 29739/43939 [47:58:42<22:40:58, 5.75s/it]
68%|██████▊ | 29740/43939 [47:58:48<22:35:35, 5.73s/it]
68%|██████▊ | 29741/43939 [47:58:54<22:33:15, 5.72s/it]
68%|██████▊ | 29742/43939 [47:58:59<22:31:55, 5.71s/it]
68%|██████▊ | 29743/43939 [47:59:05<22:30:49, 5.71s/it]
68%|██████▊ | 29744/43939 [47:59:11<22:31:47, 5.71s/it]
68%|██████▊ | 29745/43939 [47:59:17<22:30:27, 5.71s/it]
68%|██████▊ | 29746/43939 [47:59:22<22:29:11, 5.70s/it]
68%|██████▊ | 29747/43939 [47:59:28<22:31:19, 5.71s/it]
68%|██████▊ | 29748/43939 [47:59:34<22:31:00, 5.71s/it]
68%|██████▊ | 29749/43939 [47:59:39<22:28:40, 5.70s/it]
68%|██████▊ | 29750/43939 [47:59:45<22:26:20, 5.69s/it]
{'loss': 0.7904, 'grad_norm': 1.4717990159988403, 'learning_rate': 1.427192146723427e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.01, 'epoch': 2.71}
68%|██████▊ | 29750/43939 [47:59:45<22:26:20, 5.69s/it]
68%|██████▊ | 29751/43939 [47:59:51<22:27:31, 5.70s/it]
68%|██████▊ | 29752/43939 [47:59:57<22:28:58, 5.71s/it]
68%|██████▊ | 29753/43939 [48:00:02<22:27:23, 5.70s/it]
68%|██████▊ | 29754/43939 [48:00:08<22:30:43, 5.71s/it]
68%|██████▊ | 29755/43939 [48:00:14<22:31:52, 5.72s/it]
68%|██████▊ | 29756/43939 [48:00:19<22:30:10, 5.71s/it]
68%|██████▊ | 29757/43939 [48:00:25<22:46:27, 5.78s/it]
68%|██████▊ | 29758/43939 [48:00:31<22:38:59, 5.75s/it]
68%|██████▊ | 29759/43939 [48:00:37<22:34:08, 5.73s/it]
68%|██████▊ | 29760/43939 [48:00:42<22:31:24, 5.72s/it]
68%|██████▊ | 29761/43939 [48:00:48<22:28:28, 5.71s/it]
68%|██████▊ | 29762/43939 [48:00:54<22:28:27, 5.71s/it]
68%|██████▊ | 29763/43939 [48:00:59<22:30:19, 5.72s/it]
68%|██████▊ | 29764/43939 [48:01:05<22:28:53, 5.71s/it]
68%|██████▊ | 29765/43939 [48:01:11<22:25:56, 5.70s/it]
68%|██████▊ | 29766/43939 [48:01:17<22:25:35, 5.70s/it]
68%|██████▊ | 29767/43939 [48:01:22<22:28:58, 5.71s/it]
68%|██████▊ | 29768/43939 [48:01:28<22:27:17, 5.70s/it]
68%|██████▊ | 29769/43939 [48:01:34<22:28:46, 5.71s/it]
68%|██████▊ | 29770/43939 [48:01:39<22:26:53, 5.70s/it]
68%|██████▊ | 29771/43939 [48:01:45<22:26:19, 5.70s/it]
68%|██████▊ | 29772/43939 [48:01:51<22:24:53, 5.70s/it]
68%|██████▊ | 29773/43939 [48:01:56<22:23:33, 5.69s/it]
68%|██████▊ | 29774/43939 [48:02:02<22:23:47, 5.69s/it]
68%|██████▊ | 29775/43939 [48:02:08<22:24:57, 5.70s/it]
68%|██████▊ | 29776/43939 [48:02:14<22:26:19, 5.70s/it]
68%|██████▊ | 29777/43939 [48:02:19<22:25:56, 5.70s/it]
68%|██████▊ | 29778/43939 [48:02:25<22:44:29, 5.78s/it]
68%|██████▊ | 29779/43939 [48:02:31<22:38:15, 5.76s/it]
68%|██████▊ | 29780/43939 [48:02:37<22:34:00, 5.74s/it]
68%|██████▊ | 29781/43939 [48:02:42<22:30:12, 5.72s/it]
68%|██████▊ | 29782/43939 [48:02:48<22:27:04, 5.71s/it]
68%|██████▊ | 29783/43939 [48:02:54<22:30:33, 5.72s/it]
68%|██████▊ | 29784/43939 [48:02:59<22:27:07, 5.71s/it]
68%|██████▊ | 29785/43939 [48:03:05<22:28:59, 5.72s/it]
68%|██████▊ | 29786/43939 [48:03:11<22:26:06, 5.71s/it]
68%|██████▊ | 29787/43939 [48:03:17<22:25:20, 5.70s/it]
68%|██████▊ | 29788/43939 [48:03:22<22:26:20, 5.71s/it]
68%|██████▊ | 29789/43939 [48:03:28<22:26:03, 5.71s/it]
68%|██████▊ | 29790/43939 [48:03:34<22:24:08, 5.70s/it]
68%|██████▊ | 29791/43939 [48:03:39<22:21:59, 5.69s/it]
68%|██████▊ | 29792/43939 [48:03:45<22:21:33, 5.69s/it]
68%|██████▊ | 29793/43939 [48:03:51<22:20:37, 5.69s/it]
68%|██████▊ | 29794/43939 [48:03:56<22:21:15, 5.69s/it]
68%|██████▊ | 29795/43939 [48:04:02<22:19:56, 5.68s/it]
68%|██████▊ | 29796/43939 [48:04:08<22:20:20, 5.69s/it]
68%|██████▊ | 29797/43939 [48:04:13<22:21:32, 5.69s/it]
68%|██████▊ | 29798/43939 [48:04:19<22:22:00, 5.69s/it]
68%|██████▊ | 29799/43939 [48:04:25<22:39:26, 5.77s/it]
68%|██████▊ | 29800/43939 [48:04:31<22:34:53, 5.75s/it]
{'loss': 0.7689, 'grad_norm': 1.3117893934249878, 'learning_rate': 1.4182312338259007e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.68, 'epoch': 2.71}
68%|██████▊ | 29800/43939 [48:04:31<22:34:53, 5.75s/it]
68%|██████▊ | 29801/43939 [48:04:37<22:34:30, 5.75s/it]
68%|██████▊ | 29802/43939 [48:04:42<22:30:23, 5.73s/it]
68%|██████▊ | 29803/43939 [48:04:48<22:27:52, 5.72s/it]
68%|██████▊ | 29804/43939 [48:04:54<22:28:49, 5.73s/it]
68%|██████▊ | 29805/43939 [48:04:59<22:31:25, 5.74s/it]
68%|██████▊ | 29806/43939 [48:05:05<22:28:09, 5.72s/it]
68%|██████▊ | 29807/43939 [48:05:11<22:29:11, 5.73s/it]
68%|██████▊ | 29808/43939 [48:05:17<22:24:45, 5.71s/it]
68%|██████▊ | 29809/43939 [48:05:22<22:23:36, 5.71s/it]
68%|██████▊ | 29810/43939 [48:05:28<22:22:37, 5.70s/it]
68%|██████▊ | 29811/43939 [48:05:34<22:21:09, 5.70s/it]
68%|██████▊ | 29812/43939 [48:05:39<22:24:25, 5.71s/it]
68%|██████▊ | 29813/43939 [48:05:45<22:24:10, 5.71s/it]
68%|██████▊ | 29814/43939 [48:05:51<22:25:23, 5.71s/it]
68%|██████▊ | 29815/43939 [48:05:56<22:23:51, 5.71s/it]
68%|██████▊ | 29816/43939 [48:06:02<22:25:04, 5.71s/it]
68%|██████▊ | 29817/43939 [48:06:08<22:26:10, 5.72s/it]
68%|██████▊ | 29818/43939 [48:06:14<22:25:12, 5.72s/it]
68%|██████▊ | 29819/43939 [48:06:19<22:24:05, 5.71s/it]
68%|██████▊ | 29820/43939 [48:06:25<22:24:29, 5.71s/it]
68%|██████▊ | 29821/43939 [48:06:31<22:42:53, 5.79s/it]
68%|██████▊ | 29822/43939 [48:06:37<22:37:39, 5.77s/it]
68%|██████▊ | 29823/43939 [48:06:42<22:32:08, 5.75s/it]
68%|██████▊ | 29824/43939 [48:06:48<22:27:43, 5.73s/it]
68%|██████▊ | 29825/43939 [48:06:54<22:26:00, 5.72s/it]
68%|██████▊ | 29826/43939 [48:07:00<22:23:05, 5.71s/it]
68%|██████▊ | 29827/43939 [48:07:05<22:21:47, 5.70s/it]
68%|██████▊ | 29828/43939 [48:07:11<22:19:40, 5.70s/it]
68%|██████▊ | 29829/43939 [48:07:17<22:17:26, 5.69s/it]
68%|██████▊ | 29830/43939 [48:07:22<22:17:40, 5.69s/it]
68%|██████▊ | 29831/43939 [48:07:28<22:18:50, 5.69s/it]
68%|██████▊ | 29832/43939 [48:07:34<22:18:58, 5.69s/it]
68%|██████▊ | 29833/43939 [48:07:39<22:21:17, 5.71s/it]
68%|██████▊ | 29834/43939 [48:07:45<22:20:20, 5.70s/it]
68%|██████▊ | 29835/43939 [48:07:51<22:19:34, 5.70s/it]
68%|██████▊ | 29836/43939 [48:07:57<22:21:54, 5.71s/it]
68%|██████▊ | 29837/43939 [48:08:02<22:20:51, 5.71s/it]
68%|██████▊ | 29838/43939 [48:08:08<22:22:47, 5.71s/it]
68%|██████▊ | 29839/43939 [48:08:14<22:20:31, 5.70s/it]
68%|██████▊ | 29840/43939 [48:08:19<22:22:33, 5.71s/it]
68%|██████▊ | 29841/43939 [48:08:25<22:20:57, 5.71s/it]
68%|██████▊ | 29842/43939 [48:08:31<22:21:47, 5.71s/it]
68%|██████▊ | 29843/43939 [48:08:37<22:38:24, 5.78s/it]
68%|██████▊ | 29844/43939 [48:08:42<22:33:53, 5.76s/it]
68%|██████▊ | 29845/43939 [48:08:48<22:30:13, 5.75s/it]
68%|██████▊ | 29846/43939 [48:08:54<22:23:33, 5.72s/it]
68%|██████▊ | 29847/43939 [48:08:59<22:21:32, 5.71s/it]
68%|██████▊ | 29848/43939 [48:09:05<22:20:12, 5.71s/it]
68%|██████▊ | 29849/43939 [48:09:11<22:22:26, 5.72s/it]
68%|██████▊ | 29850/43939 [48:09:17<22:20:41, 5.71s/it]
{'loss': 0.7682, 'grad_norm': 1.1606810092926025, 'learning_rate': 1.4092873883727829e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.78, 'epoch': 2.72}
68%|██████▊ | 29850/43939 [48:09:17<22:20:41, 5.71s/it]
68%|██████▊ | 29851/43939 [48:09:22<22:20:38, 5.71s/it]
68%|██████▊ | 29852/43939 [48:09:28<22:18:56, 5.70s/it]
68%|██████▊ | 29853/43939 [48:09:34<22:22:24, 5.72s/it]
68%|██████▊ | 29854/43939 [48:09:40<22:26:08, 5.73s/it]
68%|██████▊ | 29855/43939 [48:09:45<22:23:53, 5.73s/it]
68%|██████▊ | 29856/43939 [48:09:51<22:22:56, 5.72s/it]
68%|██████▊ | 29857/43939 [48:09:57<22:23:01, 5.72s/it]
68%|██████▊ | 29858/43939 [48:10:02<22:22:18, 5.72s/it]
68%|██████▊ | 29859/43939 [48:10:08<22:17:45, 5.70s/it]
68%|██████▊ | 29860/43939 [48:10:14<22:22:58, 5.72s/it]
68%|██████▊ | 29861/43939 [48:10:20<22:21:00, 5.72s/it]
68%|██████▊ | 29862/43939 [48:10:25<22:22:22, 5.72s/it]
68%|██████▊ | 29863/43939 [48:10:31<22:23:00, 5.72s/it]
68%|██████▊ | 29864/43939 [48:10:37<22:38:33, 5.79s/it]
68%|██████▊ | 29865/43939 [48:10:43<22:37:17, 5.79s/it]
68%|██████▊ | 29866/43939 [48:10:48<22:30:15, 5.76s/it]
68%|██████▊ | 29867/43939 [48:10:54<22:26:24, 5.74s/it]
68%|██████▊ | 29868/43939 [48:11:00<22:23:52, 5.73s/it]
68%|██████▊ | 29869/43939 [48:11:05<22:19:49, 5.71s/it]
68%|██████▊ | 29870/43939 [48:11:11<22:18:38, 5.71s/it]
68%|██████▊ | 29871/43939 [48:11:17<22:18:09, 5.71s/it]
68%|██████▊ | 29872/43939 [48:11:23<22:18:54, 5.71s/it]
68%|██████▊ | 29873/43939 [48:11:28<22:20:48, 5.72s/it]
68%|██████▊ | 29874/43939 [48:11:34<22:18:59, 5.71s/it]
68%|██████▊ | 29875/43939 [48:11:40<22:19:50, 5.72s/it]
68%|██████▊ | 29876/43939 [48:11:45<22:19:38, 5.72s/it]
68%|██████▊ | 29877/43939 [48:11:51<22:17:08, 5.71s/it]
68%|██████▊ | 29878/43939 [48:11:57<22:15:45, 5.70s/it]
68%|██████▊ | 29879/43939 [48:12:03<22:15:00, 5.70s/it]
68%|██████▊ | 29880/43939 [48:12:08<22:14:52, 5.70s/it]
68%|██████▊ | 29881/43939 [48:12:14<22:14:57, 5.70s/it]
68%|██████▊ | 29882/43939 [48:12:20<22:18:21, 5.71s/it]
68%|██████▊ | 29883/43939 [48:12:25<22:17:14, 5.71s/it]
68%|██████▊ | 29884/43939 [48:12:31<22:16:34, 5.71s/it]
68%|██████▊ | 29885/43939 [48:12:37<22:32:02, 5.77s/it]
68%|██████▊ | 29886/43939 [48:12:43<22:28:39, 5.76s/it]
68%|██████▊ | 29887/43939 [48:12:48<22:25:13, 5.74s/it]
68%|██████▊ | 29888/43939 [48:12:54<22:23:05, 5.74s/it]
68%|██████▊ | 29889/43939 [48:13:00<22:21:57, 5.73s/it]
68%|██████▊ | 29890/43939 [48:13:06<22:19:51, 5.72s/it]
68%|██████▊ | 29891/43939 [48:13:11<22:23:36, 5.74s/it]
68%|██████▊ | 29892/43939 [48:13:17<22:22:28, 5.73s/it]
68%|██████▊ | 29893/43939 [48:13:23<22:21:13, 5.73s/it]
68%|██████▊ | 29894/43939 [48:13:28<22:17:58, 5.72s/it]
68%|██████▊ | 29895/43939 [48:13:34<22:16:22, 5.71s/it]
68%|██████▊ | 29896/43939 [48:13:40<22:15:02, 5.70s/it]
68%|██████▊ | 29897/43939 [48:13:46<22:13:47, 5.70s/it]
68%|██████▊ | 29898/43939 [48:13:51<22:13:21, 5.70s/it]
68%|██████▊ | 29899/43939 [48:13:57<22:13:59, 5.70s/it]
68%|██████▊ | 29900/43939 [48:14:03<22:14:37, 5.70s/it]
{'loss': 0.7677, 'grad_norm': 1.2451289892196655, 'learning_rate': 1.4003607514742529e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.97, 'epoch': 2.72}
68%|██████▊ | 29900/43939 [48:14:03<22:14:37, 5.70s/it]
68%|██████▊ | 29901/43939 [48:14:08<22:13:18, 5.70s/it]
68%|██████▊ | 29902/43939 [48:14:14<22:14:10, 5.70s/it]
68%|██████▊ | 29903/43939 [48:14:20<22:14:45, 5.71s/it]
68%|██████▊ | 29904/43939 [48:14:25<22:14:10, 5.70s/it]
68%|██████▊ | 29905/43939 [48:14:31<22:14:07, 5.70s/it]
68%|██████▊ | 29906/43939 [48:14:37<22:12:27, 5.70s/it]
68%|██████▊ | 29907/43939 [48:14:43<22:25:03, 5.75s/it]
68%|██████▊ | 29908/43939 [48:14:49<22:25:18, 5.75s/it]
68%|██████▊ | 29909/43939 [48:14:54<22:21:37, 5.74s/it]
68%|██████▊ | 29910/43939 [48:15:00<22:21:02, 5.74s/it]
68%|██████▊ | 29911/43939 [48:15:06<22:18:13, 5.72s/it]
68%|██████▊ | 29912/43939 [48:15:11<22:16:37, 5.72s/it]
68%|██████▊ | 29913/43939 [48:15:17<22:15:18, 5.71s/it]
68%|██████▊ | 29914/43939 [48:15:23<22:15:19, 5.71s/it]
68%|██████▊ | 29915/43939 [48:15:28<22:13:38, 5.71s/it]
68%|██████▊ | 29916/43939 [48:15:34<22:12:41, 5.70s/it]
68%|██████▊ | 29917/43939 [48:15:40<22:11:11, 5.70s/it]
68%|██████▊ | 29918/43939 [48:15:46<22:11:32, 5.70s/it]
68%|██████▊ | 29919/43939 [48:15:51<22:12:46, 5.70s/it]
68%|██████▊ | 29920/43939 [48:15:57<22:11:15, 5.70s/it]
68%|██████▊ | 29921/43939 [48:16:03<22:12:31, 5.70s/it]
68%|██████▊ | 29922/43939 [48:16:08<22:16:26, 5.72s/it]
68%|██████▊ | 29923/43939 [48:16:14<22:13:42, 5.71s/it]
68%|██████▊ | 29924/43939 [48:16:20<22:15:58, 5.72s/it]
68%|██████▊ | 29925/43939 [48:16:26<22:14:03, 5.71s/it]
68%|██████▊ | 29926/43939 [48:16:31<22:13:26, 5.71s/it]
68%|██████▊ | 29927/43939 [48:16:37<22:16:24, 5.72s/it]
68%|██████▊ | 29928/43939 [48:16:43<22:30:35, 5.78s/it]
68%|██████▊ | 29929/43939 [48:16:49<22:25:57, 5.76s/it]
68%|██████▊ | 29930/43939 [48:16:54<22:19:39, 5.74s/it]
68%|██████▊ | 29931/43939 [48:17:00<22:17:36, 5.73s/it]
68%|██████▊ | 29932/43939 [48:17:06<22:16:21, 5.72s/it]
68%|██████▊ | 29933/43939 [48:17:11<22:18:25, 5.73s/it]
68%|██████▊ | 29934/43939 [48:17:17<22:18:51, 5.74s/it]
68%|██████▊ | 29935/43939 [48:17:23<22:23:38, 5.76s/it]
68%|██████▊ | 29936/43939 [48:17:29<22:21:14, 5.75s/it]
68%|██████▊ | 29937/43939 [48:17:34<22:17:17, 5.73s/it]
68%|██████▊ | 29938/43939 [48:17:40<22:15:08, 5.72s/it]
68%|██████▊ | 29939/43939 [48:17:46<22:14:01, 5.72s/it]
68%|██████▊ | 29940/43939 [48:17:52<22:13:37, 5.72s/it]
68%|██████▊ | 29941/43939 [48:17:57<22:14:23, 5.72s/it]
68%|██████▊ | 29942/43939 [48:18:03<22:11:07, 5.71s/it]
68%|██████▊ | 29943/43939 [48:18:09<22:11:14, 5.71s/it]
68%|██████▊ | 29944/43939 [48:18:14<22:11:13, 5.71s/it]
68%|██████▊ | 29945/43939 [48:18:20<22:08:39, 5.70s/it]
68%|██████▊ | 29946/43939 [48:18:26<22:06:21, 5.69s/it]
68%|██████▊ | 29947/43939 [48:18:31<22:09:38, 5.70s/it]
68%|██████▊ | 29948/43939 [48:18:37<22:10:07, 5.70s/it]
68%|██████▊ | 29949/43939 [48:18:43<22:27:20, 5.78s/it]
68%|██████▊ | 29950/43939 [48:18:49<22:23:01, 5.76s/it]
{'loss': 0.7496, 'grad_norm': 1.4051967859268188, 'learning_rate': 1.3914514639689841e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.44, 'epoch': 2.73}
68%|██████▊ | 29950/43939 [48:18:49<22:23:01, 5.76s/it]
68%|██████▊ | 29951/43939 [48:18:55<22:19:55, 5.75s/it]
68%|██████▊ | 29952/43939 [48:19:00<22:16:14, 5.73s/it]
68%|██████▊ | 29953/43939 [48:19:06<22:18:08, 5.74s/it]
68%|██████▊ | 29954/43939 [48:19:12<22:16:11, 5.73s/it]
68%|██████▊ | 29955/43939 [48:19:17<22:16:12, 5.73s/it]
68%|██████▊ | 29956/43939 [48:19:23<22:12:44, 5.72s/it]
68%|██████▊ | 29957/43939 [48:19:29<22:12:12, 5.72s/it]
68%|██████▊ | 29958/43939 [48:19:35<22:09:35, 5.71s/it]
68%|██████▊ | 29959/43939 [48:19:40<22:08:10, 5.70s/it]
68%|██████▊ | 29960/43939 [48:19:46<22:05:13, 5.69s/it]
68%|██████▊ | 29961/43939 [48:19:52<22:06:49, 5.70s/it]
68%|██████▊ | 29962/43939 [48:19:57<22:06:30, 5.69s/it]
68%|██████▊ | 29963/43939 [48:20:03<22:08:11, 5.70s/it]
68%|██████▊ | 29964/43939 [48:20:09<22:07:33, 5.70s/it]
68%|██████▊ | 29965/43939 [48:20:14<22:06:38, 5.70s/it]
68%|██████▊ | 29966/43939 [48:20:20<22:11:36, 5.72s/it]
68%|██████▊ | 29967/43939 [48:20:26<22:10:29, 5.71s/it]
68%|██████▊ | 29968/43939 [48:20:32<22:08:33, 5.71s/it]
68%|██████▊ | 29969/43939 [48:20:37<22:10:27, 5.71s/it]
68%|██████▊ | 29970/43939 [48:20:43<22:11:19, 5.72s/it]
68%|██████▊ | 29971/43939 [48:20:49<22:23:57, 5.77s/it]
68%|██████▊ | 29972/43939 [48:20:55<22:20:00, 5.76s/it]
68%|██████▊ | 29973/43939 [48:21:00<22:17:37, 5.75s/it]
68%|██████▊ | 29974/43939 [48:21:06<22:13:01, 5.73s/it]
68%|██████▊ | 29975/43939 [48:21:12<22:11:24, 5.72s/it]
68%|██████▊ | 29976/43939 [48:21:17<22:11:45, 5.72s/it]
68%|██████▊ | 29977/43939 [48:21:23<22:11:13, 5.72s/it]
68%|██████▊ | 29978/43939 [48:21:29<22:08:44, 5.71s/it]
68%|██████▊ | 29979/43939 [48:21:35<22:05:38, 5.70s/it]
68%|██████▊ | 29980/43939 [48:21:40<22:11:01, 5.72s/it]
68%|██████▊ | 29981/43939 [48:21:46<22:09:00, 5.71s/it]
68%|██████▊ | 29982/43939 [48:21:52<22:04:54, 5.70s/it]
68%|██████▊ | 29983/43939 [48:21:57<22:04:45, 5.70s/it]
68%|██████▊ | 29984/43939 [48:22:03<22:05:24, 5.70s/it]
68%|██████▊ | 29985/43939 [48:22:09<22:06:27, 5.70s/it]
68%|██████▊ | 29986/43939 [48:22:14<22:06:11, 5.70s/it]
68%|██████▊ | 29987/43939 [48:22:20<22:03:59, 5.69s/it]
68%|██████▊ | 29988/43939 [48:22:26<22:05:28, 5.70s/it]
68%|██████▊ | 29989/43939 [48:22:32<22:04:51, 5.70s/it]
68%|██████▊ | 29990/43939 [48:22:37<22:04:56, 5.70s/it]
68%|██████▊ | 29991/43939 [48:22:43<22:07:13, 5.71s/it]
68%|██████▊ | 29992/43939 [48:22:49<22:28:04, 5.80s/it]
68%|██████▊ | 29993/43939 [48:22:55<22:21:30, 5.77s/it]
68%|██████▊ | 29994/43939 [48:23:00<22:16:19, 5.75s/it]
68%|██████▊ | 29995/43939 [48:23:06<22:13:42, 5.74s/it]
68%|██████▊ | 29996/43939 [48:23:12<22:11:42, 5.73s/it]
68%|██████▊ | 29997/43939 [48:23:18<22:10:02, 5.72s/it]
68%|██████▊ | 29998/43939 [48:23:23<22:10:00, 5.72s/it]
68%|██████▊ | 29999/43939 [48:23:29<22:07:36, 5.71s/it]
68%|██████▊ | 30000/43939 [48:23:35<22:05:45, 5.71s/it]
{'loss': 0.7821, 'grad_norm': 1.65287446975708, 'learning_rate': 1.3825596664219254e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.98, 'epoch': 2.73}
68%|██████▊ | 30000/43939 [48:23:35<22:05:45, 5.71s/it]
68%|██████▊ | 30001/43939 [48:23:40<22:04:03, 5.70s/it]
68%|██████▊ | 30002/43939 [48:23:46<22:05:02, 5.70s/it]
68%|██████▊ | 30003/43939 [48:23:52<22:06:50, 5.71s/it]
68%|██████▊ | 30004/43939 [48:23:57<22:04:10, 5.70s/it]
68%|██████▊ | 30005/43939 [48:24:03<22:05:34, 5.71s/it]
68%|██████▊ | 30006/43939 [48:24:09<22:03:46, 5.70s/it]
68%|██████▊ | 30007/43939 [48:24:15<22:06:15, 5.71s/it]
68%|██████▊ | 30008/43939 [48:24:20<22:06:23, 5.71s/it]
68%|██████▊ | 30009/43939 [48:24:26<22:07:02, 5.72s/it]
68%|██████▊ | 30010/43939 [48:24:32<22:07:53, 5.72s/it]
68%|██████▊ | 30011/43939 [48:24:37<22:06:16, 5.71s/it]
68%|██████▊ | 30012/43939 [48:24:43<22:04:34, 5.71s/it]
68%|██████▊ | 30013/43939 [48:24:49<22:20:07, 5.77s/it]
68%|██████▊ | 30014/43939 [48:24:55<22:15:18, 5.75s/it]
68%|██████▊ | 30015/43939 [48:25:00<22:10:37, 5.73s/it]
68%|██████▊ | 30016/43939 [48:25:06<22:05:42, 5.71s/it]
68%|██████▊ | 30017/43939 [48:25:12<22:04:13, 5.71s/it]
68%|██████▊ | 30018/43939 [48:25:18<22:03:29, 5.70s/it]
68%|██████▊ | 30019/43939 [48:25:23<22:04:51, 5.71s/it]
68%|██████▊ | 30020/43939 [48:25:29<22:07:34, 5.72s/it]
68%|██████▊ | 30021/43939 [48:25:35<22:04:06, 5.71s/it]
68%|██████▊ | 30022/43939 [48:25:40<22:07:40, 5.72s/it]
68%|██████▊ | 30023/43939 [48:25:46<22:05:06, 5.71s/it]
68%|██████▊ | 30024/43939 [48:25:52<22:10:08, 5.74s/it]
68%|██████▊ | 30025/43939 [48:25:58<22:08:45, 5.73s/it]
68%|██████▊ | 30026/43939 [48:26:03<22:06:24, 5.72s/it]
68%|██████▊ | 30027/43939 [48:26:09<22:07:36, 5.73s/it]
68%|██████▊ | 30028/43939 [48:26:15<22:03:47, 5.71s/it]
68%|██████▊ | 30029/43939 [48:26:20<22:03:44, 5.71s/it]
68%|██████▊ | 30030/43939 [48:26:26<22:02:42, 5.71s/it]
68%|██████▊ | 30031/43939 [48:26:32<22:02:49, 5.71s/it]
68%|██████▊ | 30032/43939 [48:26:38<22:01:55, 5.70s/it]
68%|██████▊ | 30033/43939 [48:26:43<22:04:48, 5.72s/it]
68%|██████▊ | 30034/43939 [48:26:49<22:16:54, 5.77s/it]
68%|██████▊ | 30035/43939 [48:26:55<22:12:03, 5.75s/it]
68%|██████▊ | 30036/43939 [48:27:01<22:08:43, 5.73s/it]
68%|██████▊ | 30037/43939 [48:27:06<22:04:53, 5.72s/it]
68%|██████▊ | 30038/43939 [48:27:12<22:03:56, 5.71s/it]
68%|██████▊ | 30039/43939 [48:27:18<22:02:12, 5.71s/it]
68%|██████▊ | 30040/43939 [48:27:23<21:59:51, 5.70s/it]
68%|██████▊ | 30041/43939 [48:27:29<21:57:16, 5.69s/it]
68%|██████▊ | 30042/43939 [48:27:35<21:57:44, 5.69s/it]
68%|██████▊ | 30043/43939 [48:27:40<21:59:10, 5.70s/it]
68%|██████▊ | 30044/43939 [48:27:46<21:57:44, 5.69s/it]
68%|██████▊ | 30045/43939 [48:27:52<21:57:54, 5.69s/it]
68%|██████▊ | 30046/43939 [48:27:57<21:56:57, 5.69s/it]
68%|██████▊ | 30047/43939 [48:28:03<22:02:34, 5.71s/it]
68%|██████▊ | 30048/43939 [48:28:09<22:02:57, 5.71s/it]
68%|██████▊ | 30049/43939 [48:28:15<22:02:41, 5.71s/it]
68%|██████▊ | 30050/43939 [48:28:20<22:05:35, 5.73s/it]
{'loss': 0.7765, 'grad_norm': 1.2787915468215942, 'learning_rate': 1.3736854991220777e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.55, 'epoch': 2.74}
68%|██████▊ | 30050/43939 [48:28:20<22:05:35, 5.73s/it]
68%|██████▊ | 30051/43939 [48:28:26<22:06:43, 5.73s/it]
68%|██████▊ | 30052/43939 [48:28:32<22:01:43, 5.71s/it]
68%|██████▊ | 30053/43939 [48:28:38<22:01:46, 5.71s/it]
68%|██████▊ | 30054/43939 [48:28:43<22:04:58, 5.73s/it]
68%|██████▊ | 30055/43939 [48:28:49<22:17:14, 5.78s/it]
68%|██████▊ | 30056/43939 [48:28:55<22:13:05, 5.76s/it]
68%|██████▊ | 30057/43939 [48:29:01<22:08:42, 5.74s/it]
68%|██████▊ | 30058/43939 [48:29:06<22:04:43, 5.73s/it]
68%|██████▊ | 30059/43939 [48:29:12<21:58:44, 5.70s/it]
68%|██████▊ | 30060/43939 [48:29:18<21:58:59, 5.70s/it]
68%|██████▊ | 30061/43939 [48:29:23<22:03:42, 5.72s/it]
68%|██████▊ | 30062/43939 [48:29:29<22:00:19, 5.71s/it]
68%|██████▊ | 30063/43939 [48:29:35<21:59:27, 5.71s/it]
68%|██████▊ | 30064/43939 [48:29:40<21:58:56, 5.70s/it]
68%|██████▊ | 30065/43939 [48:29:46<22:01:03, 5.71s/it]
68%|██████▊ | 30066/43939 [48:29:52<22:00:52, 5.71s/it]
68%|██████▊ | 30067/43939 [48:29:58<22:02:01, 5.72s/it]
68%|██████▊ | 30068/43939 [48:30:03<22:01:55, 5.72s/it]
68%|██████▊ | 30069/43939 [48:30:09<22:02:02, 5.72s/it]
68%|██████▊ | 30070/43939 [48:30:15<22:01:52, 5.72s/it]
68%|██████▊ | 30071/43939 [48:30:21<21:59:30, 5.71s/it]
68%|██████▊ | 30072/43939 [48:30:26<21:59:50, 5.71s/it]
68%|██████▊ | 30073/43939 [48:30:32<22:00:03, 5.71s/it]
68%|██████▊ | 30074/43939 [48:30:38<21:58:39, 5.71s/it]
68%|██████▊ | 30075/43939 [48:30:43<21:55:53, 5.69s/it]
68%|██████▊ | 30076/43939 [48:30:49<21:54:49, 5.69s/it]
68%|██████▊ | 30077/43939 [48:30:55<22:12:25, 5.77s/it]
68%|██████▊ | 30078/43939 [48:31:01<22:07:47, 5.75s/it]
68%|██████▊ | 30079/43939 [48:31:06<22:06:36, 5.74s/it]
68%|██████▊ | 30080/43939 [48:31:12<22:03:00, 5.73s/it]
68%|██████▊ | 30081/43939 [48:31:18<21:59:48, 5.71s/it]
68%|██████▊ | 30082/43939 [48:31:23<22:01:06, 5.72s/it]
68%|██████▊ | 30083/43939 [48:31:29<22:01:54, 5.72s/it]
68%|██████▊ | 30084/43939 [48:31:35<22:01:21, 5.72s/it]
68%|██████▊ | 30085/43939 [48:31:41<22:01:26, 5.72s/it]
68%|██████▊ | 30086/43939 [48:31:46<22:00:26, 5.72s/it]
68%|██████▊ | 30087/43939 [48:31:52<22:00:02, 5.72s/it]
68%|██████▊ | 30088/43939 [48:31:58<21:59:21, 5.72s/it]
68%|██████▊ | 30089/43939 [48:32:03<21:56:03, 5.70s/it]
68%|██████▊ | 30090/43939 [48:32:09<21:56:39, 5.70s/it]
68%|██████▊ | 30091/43939 [48:32:15<21:54:43, 5.70s/it]
68%|██████▊ | 30092/43939 [48:32:21<21:55:17, 5.70s/it]
68%|██████▊ | 30093/43939 [48:32:26<21:58:16, 5.71s/it]
68%|██████▊ | 30094/43939 [48:32:32<22:00:14, 5.72s/it]
68%|██████▊ | 30095/43939 [48:32:38<21:59:21, 5.72s/it]
68%|██████▊ | 30096/43939 [48:32:44<22:03:24, 5.74s/it]
68%|██████▊ | 30097/43939 [48:32:49<21:58:19, 5.71s/it]
68%|██████▊ | 30098/43939 [48:32:55<22:11:17, 5.77s/it]
69%|██████▊ | 30099/43939 [48:33:01<22:09:31, 5.76s/it]
69%|██████▊ | 30100/43939 [48:33:07<22:05:29, 5.75s/it]
{'loss': 0.777, 'grad_norm': 1.6088179349899292, 'learning_rate': 1.364829102080284e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.44, 'epoch': 2.74}
69%|██████▊ | 30100/43939 [48:33:07<22:05:29, 5.75s/it]
69%|██████▊ | 30101/43939 [48:33:12<22:01:36, 5.73s/it]
69%|██████▊ | 30102/43939 [48:33:18<21:59:26, 5.72s/it]
69%|██████▊ | 30103/43939 [48:33:24<21:57:17, 5.71s/it]
69%|██████▊ | 30104/43939 [48:33:29<21:58:20, 5.72s/it]
69%|██████▊ | 30105/43939 [48:33:35<21:55:07, 5.70s/it]
69%|██████▊ | 30106/43939 [48:33:41<21:55:08, 5.70s/it]
69%|██████▊ | 30107/43939 [48:33:46<21:55:06, 5.70s/it]
69%|██████▊ | 30108/43939 [48:33:52<21:55:37, 5.71s/it]
69%|██████▊ | 30109/43939 [48:33:58<21:58:30, 5.72s/it]
69%|██████▊ | 30110/43939 [48:34:04<21:57:07, 5.71s/it]
69%|██████▊ | 30111/43939 [48:34:09<21:55:19, 5.71s/it]
69%|██████▊ | 30112/43939 [48:34:15<21:56:37, 5.71s/it]
69%|██████▊ | 30113/43939 [48:34:21<21:55:03, 5.71s/it]
69%|██████▊ | 30114/43939 [48:34:26<21:51:45, 5.69s/it]
69%|██████▊ | 30115/43939 [48:34:32<21:53:06, 5.70s/it]
69%|██████▊ | 30116/43939 [48:34:38<21:54:02, 5.70s/it]
69%|██████▊ | 30117/43939 [48:34:44<21:58:50, 5.72s/it]
69%|██████▊ | 30118/43939 [48:34:49<21:59:14, 5.73s/it]
69%|██████▊ | 30119/43939 [48:34:55<22:14:54, 5.80s/it]
69%|██████▊ | 30120/43939 [48:35:01<22:09:28, 5.77s/it]
69%|██████▊ | 30121/43939 [48:35:07<22:07:10, 5.76s/it]
69%|██████▊ | 30122/43939 [48:35:12<22:01:56, 5.74s/it]
69%|██████▊ | 30123/43939 [48:35:18<22:00:16, 5.73s/it]
69%|██████▊ | 30124/43939 [48:35:24<22:00:36, 5.74s/it]
69%|██████▊ | 30125/43939 [48:35:30<21:58:10, 5.73s/it]
69%|██████▊ | 30126/43939 [48:35:35<21:56:07, 5.72s/it]
69%|██████▊ | 30127/43939 [48:35:41<21:55:30, 5.71s/it]
69%|██████▊ | 30128/43939 [48:35:47<21:54:13, 5.71s/it]
69%|██████▊ | 30129/43939 [48:35:52<21:56:06, 5.72s/it]
69%|██████▊ | 30130/43939 [48:35:58<21:55:58, 5.72s/it]
69%|██████▊ | 30131/43939 [48:36:04<21:52:09, 5.70s/it]
69%|██████▊ | 30132/43939 [48:36:10<21:53:30, 5.71s/it]
69%|██████▊ | 30133/43939 [48:36:15<21:51:38, 5.70s/it]
69%|██████▊ | 30134/43939 [48:36:21<21:49:09, 5.69s/it]
69%|██████▊ | 30135/43939 [48:36:27<21:53:53, 5.71s/it]
69%|██████▊ | 30136/43939 [48:36:32<21:53:43, 5.71s/it]
69%|██████▊ | 30137/43939 [48:36:38<21:51:04, 5.70s/it]
69%|██████▊ | 30138/43939 [48:36:44<21:56:01, 5.72s/it]
69%|██████▊ | 30139/43939 [48:36:49<21:54:18, 5.71s/it]
69%|██████▊ | 30140/43939 [48:36:55<21:53:27, 5.71s/it]
69%|██████▊ | 30141/43939 [48:37:01<22:12:33, 5.79s/it]
69%|██████▊ | 30142/43939 [48:37:07<22:08:37, 5.78s/it]
69%|██████▊ | 30143/43939 [48:37:13<22:02:55, 5.75s/it]
69%|██████▊ | 30144/43939 [48:37:18<21:59:45, 5.74s/it]
69%|██████▊ | 30145/43939 [48:37:24<21:56:00, 5.72s/it]
69%|██████▊ | 30146/43939 [48:37:30<21:54:04, 5.72s/it]
69%|██████▊ | 30147/43939 [48:37:35<21:51:58, 5.71s/it]
69%|██████▊ | 30148/43939 [48:37:41<21:51:28, 5.71s/it]
69%|██████▊ | 30149/43939 [48:37:47<21:49:32, 5.70s/it]
69%|██████▊ | 30150/43939 [48:37:52<21:46:57, 5.69s/it]
{'loss': 0.7788, 'grad_norm': 1.7011981010437012, 'learning_rate': 1.35599061502702e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.22, 'epoch': 2.74}
69%|██████▊ | 30150/43939 [48:37:52<21:46:57, 5.69s/it]
69%|██████▊ | 30151/43939 [48:37:58<21:48:36, 5.69s/it]
69%|██████▊ | 30152/43939 [48:38:04<21:49:17, 5.70s/it]
69%|██████▊ | 30153/43939 [48:38:10<21:50:14, 5.70s/it]
69%|██████▊ | 30154/43939 [48:38:15<21:49:30, 5.70s/it]
69%|██████▊ | 30155/43939 [48:38:21<21:48:23, 5.70s/it]
69%|██████▊ | 30156/43939 [48:38:27<21:48:22, 5.70s/it]
69%|██████▊ | 30157/43939 [48:38:32<21:47:51, 5.69s/it]
69%|██████▊ | 30158/43939 [48:38:38<21:50:38, 5.71s/it]
69%|██████▊ | 30159/43939 [48:38:44<21:49:41, 5.70s/it]
69%|██████▊ | 30160/43939 [48:38:49<21:49:02, 5.70s/it]
69%|██████▊ | 30161/43939 [48:38:55<21:49:51, 5.70s/it]
69%|██████▊ | 30162/43939 [48:39:01<21:46:31, 5.69s/it]
69%|██████▊ | 30163/43939 [48:39:07<22:02:19, 5.76s/it]
69%|██████▊ | 30164/43939 [48:39:12<21:57:10, 5.74s/it]
69%|██████▊ | 30165/43939 [48:39:18<21:52:49, 5.72s/it]
69%|██████▊ | 30166/43939 [48:39:24<21:49:35, 5.71s/it]
69%|██████▊ | 30167/43939 [48:39:29<21:49:11, 5.70s/it]
69%|██████▊ | 30168/43939 [48:39:35<21:46:36, 5.69s/it]
69%|██████▊ | 30169/43939 [48:39:41<21:45:21, 5.69s/it]
69%|██████▊ | 30170/43939 [48:39:47<21:50:00, 5.71s/it]
69%|██████▊ | 30171/43939 [48:39:52<21:46:33, 5.69s/it]
69%|██████▊ | 30172/43939 [48:39:58<21:46:13, 5.69s/it]
69%|██████▊ | 30173/43939 [48:40:04<21:44:24, 5.69s/it]
69%|██████▊ | 30174/43939 [48:40:09<21:44:58, 5.69s/it]
69%|██████▊ | 30175/43939 [48:40:15<21:46:09, 5.69s/it]
69%|██████▊ | 30176/43939 [48:40:21<21:51:38, 5.72s/it]
69%|██████▊ | 30177/43939 [48:40:26<21:49:44, 5.71s/it]
69%|██████▊ | 30178/43939 [48:40:32<21:51:18, 5.72s/it]
69%|██████▊ | 30179/43939 [48:40:38<21:48:53, 5.71s/it]
69%|██████▊ | 30180/43939 [48:40:44<21:46:52, 5.70s/it]
69%|██████▊ | 30181/43939 [48:40:49<21:46:33, 5.70s/it]
69%|██████▊ | 30182/43939 [48:40:55<21:47:02, 5.70s/it]
69%|██████▊ | 30183/43939 [48:41:01<21:45:54, 5.70s/it]
69%|██████▊ | 30184/43939 [48:41:07<22:13:54, 5.82s/it]
69%|██████▊ | 30185/43939 [48:41:12<22:05:24, 5.78s/it]
69%|██████▊ | 30186/43939 [48:41:18<21:59:33, 5.76s/it]
69%|██████▊ | 30187/43939 [48:41:24<21:55:38, 5.74s/it]
69%|██████▊ | 30188/43939 [48:41:30<21:55:54, 5.74s/it]
69%|██████▊ | 30189/43939 [48:41:35<21:53:42, 5.73s/it]
69%|██████▊ | 30190/43939 [48:41:41<21:49:38, 5.72s/it]
69%|██████▊ | 30191/43939 [48:41:47<21:45:25, 5.70s/it]
69%|██████▊ | 30192/43939 [48:41:52<21:45:30, 5.70s/it]
69%|██████▊ | 30193/43939 [48:41:58<21:45:01, 5.70s/it]
69%|██████▊ | 30194/43939 [48:42:04<21:45:22, 5.70s/it]
69%|██████▊ | 30195/43939 [48:42:09<21:44:49, 5.70s/it]
69%|██████▊ | 30196/43939 [48:42:15<21:45:32, 5.70s/it]
69%|██████▊ | 30197/43939 [48:42:21<21:45:15, 5.70s/it]
69%|██████▊ | 30198/43939 [48:42:27<21:46:00, 5.70s/it]
69%|██████▊ | 30199/43939 [48:42:32<21:44:28, 5.70s/it]
69%|██████▊ | 30200/43939 [48:42:38<21:46:57, 5.71s/it]
{'loss': 0.7916, 'grad_norm': 1.3469337224960327, 'learning_rate': 1.3471701774101892e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.97, 'epoch': 2.75}
69%|██████▊ | 30200/43939 [48:42:38<21:46:57, 5.71s/it]
69%|██████▊ | 30201/43939 [48:42:44<21:47:59, 5.71s/it]
69%|██████▊ | 30202/43939 [48:42:49<21:45:21, 5.70s/it]
69%|██████▊ | 30203/43939 [48:42:55<21:48:15, 5.71s/it]
69%|██████▊ | 30204/43939 [48:43:01<21:46:00, 5.71s/it]
69%|██████▊ | 30205/43939 [48:43:07<21:58:37, 5.76s/it]
69%|██████▊ | 30206/43939 [48:43:12<21:54:52, 5.74s/it]
69%|██████▊ | 30207/43939 [48:43:18<21:52:19, 5.73s/it]
69%|██████▊ | 30208/43939 [48:43:24<21:57:45, 5.76s/it]
69%|██████▉ | 30209/43939 [48:43:30<21:57:53, 5.76s/it]
69%|██████▉ | 30210/43939 [48:43:35<21:54:10, 5.74s/it]
69%|██████▉ | 30211/43939 [48:43:41<21:49:43, 5.72s/it]
69%|██████▉ | 30212/43939 [48:43:47<21:48:48, 5.72s/it]
69%|██████▉ | 30213/43939 [48:43:53<21:50:39, 5.73s/it]
69%|██████▉ | 30214/43939 [48:43:58<21:48:57, 5.72s/it]
69%|██████▉ | 30215/43939 [48:44:04<21:48:00, 5.72s/it]
69%|██████▉ | 30216/43939 [48:44:10<21:45:17, 5.71s/it]
69%|██████▉ | 30217/43939 [48:44:15<21:43:18, 5.70s/it]
69%|██████▉ | 30218/43939 [48:44:21<21:42:43, 5.70s/it]
69%|██████▉ | 30219/43939 [48:44:27<21:42:42, 5.70s/it]
69%|██████▉ | 30220/43939 [48:44:32<21:40:13, 5.69s/it]
69%|██████▉ | 30221/43939 [48:44:38<21:39:21, 5.68s/it]
69%|██████▉ | 30222/43939 [48:44:44<21:37:53, 5.68s/it]
69%|██████▉ | 30223/43939 [48:44:49<21:38:37, 5.68s/it]
69%|██████▉ | 30224/43939 [48:44:55<21:42:34, 5.70s/it]
69%|██████▉ | 30225/43939 [48:45:01<21:46:32, 5.72s/it]
69%|██████▉ | 30226/43939 [48:45:07<21:58:57, 5.77s/it]
69%|██████▉ | 30227/43939 [48:45:12<21:55:05, 5.75s/it]
69%|██████▉ | 30228/43939 [48:45:18<21:50:05, 5.73s/it]
69%|██████▉ | 30229/43939 [48:45:24<21:47:24, 5.72s/it]
69%|██████▉ | 30230/43939 [48:45:30<21:43:03, 5.70s/it]
69%|██████▉ | 30231/43939 [48:45:35<21:43:08, 5.70s/it]
69%|██████▉ | 30232/43939 [48:45:41<21:41:53, 5.70s/it]
69%|██████▉ | 30233/43939 [48:45:47<21:41:31, 5.70s/it]
69%|██████▉ | 30234/43939 [48:45:52<21:42:19, 5.70s/it]
69%|██████▉ | 30235/43939 [48:45:58<21:42:37, 5.70s/it]
69%|██████▉ | 30236/43939 [48:46:04<21:43:47, 5.71s/it]
69%|██████▉ | 30237/43939 [48:46:09<21:41:57, 5.70s/it]
69%|██████▉ | 30238/43939 [48:46:15<21:45:59, 5.72s/it]
69%|██████▉ | 30239/43939 [48:46:21<21:45:48, 5.72s/it]
69%|██████▉ | 30240/43939 [48:46:27<21:43:41, 5.71s/it]
69%|██████▉ | 30241/43939 [48:46:32<21:46:12, 5.72s/it]
69%|██████▉ | 30242/43939 [48:46:38<21:46:05, 5.72s/it]
69%|██████▉ | 30243/43939 [48:46:44<21:43:53, 5.71s/it]
69%|██████▉ | 30244/43939 [48:46:50<21:45:32, 5.72s/it]
69%|██████▉ | 30245/43939 [48:46:55<21:47:10, 5.73s/it]
69%|██████▉ | 30246/43939 [48:47:01<21:44:03, 5.71s/it]
69%|██████▉ | 30247/43939 [48:47:07<21:58:47, 5.78s/it]
69%|██████▉ | 30248/43939 [48:47:13<21:55:33, 5.77s/it]
69%|██████▉ | 30249/43939 [48:47:18<21:50:46, 5.74s/it]
69%|██████▉ | 30250/43939 [48:47:24<21:46:20, 5.73s/it]
{'loss': 0.7767, 'grad_norm': 1.2775756120681763, 'learning_rate': 1.3383679283929223e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.9, 'epoch': 2.75}
69%|██████▉ | 30250/43939 [48:47:24<21:46:20, 5.73s/it]
69%|██████▉ | 30251/43939 [48:47:30<21:43:52, 5.72s/it]
69%|██████▉ | 30252/43939 [48:47:35<21:43:04, 5.71s/it]
69%|██████▉ | 30253/43939 [48:47:41<21:41:02, 5.70s/it]
69%|██████▉ | 30254/43939 [48:47:47<21:41:22, 5.71s/it]
69%|██████▉ | 30255/43939 [48:47:52<21:40:01, 5.70s/it]
69%|██████▉ | 30256/43939 [48:47:58<21:41:09, 5.71s/it]
69%|██████▉ | 30257/43939 [48:48:04<21:40:23, 5.70s/it]
69%|██████▉ | 30258/43939 [48:48:10<21:40:56, 5.71s/it]
69%|██████▉ | 30259/43939 [48:48:15<21:40:11, 5.70s/it]
69%|██████▉ | 30260/43939 [48:48:21<21:40:03, 5.70s/it]
69%|██████▉ | 30261/43939 [48:48:27<21:41:59, 5.71s/it]
69%|██████▉ | 30262/43939 [48:48:32<21:42:39, 5.71s/it]
69%|██████▉ | 30263/43939 [48:48:38<21:41:38, 5.71s/it]
69%|██████▉ | 30264/43939 [48:48:44<21:43:34, 5.72s/it]
69%|██████▉ | 30265/43939 [48:48:50<21:40:21, 5.71s/it]
69%|██████▉ | 30266/43939 [48:48:55<21:40:23, 5.71s/it]
69%|██████▉ | 30267/43939 [48:49:01<21:40:28, 5.71s/it]
69%|██████▉ | 30268/43939 [48:49:07<21:41:09, 5.71s/it]
69%|██████▉ | 30269/43939 [48:49:13<21:55:57, 5.78s/it]
69%|██████▉ | 30270/43939 [48:49:18<21:50:34, 5.75s/it]
69%|██████▉ | 30271/43939 [48:49:24<21:50:35, 5.75s/it]
69%|██████▉ | 30272/43939 [48:49:30<21:48:08, 5.74s/it]
69%|██████▉ | 30273/43939 [48:49:35<21:45:08, 5.73s/it]
69%|██████▉ | 30274/43939 [48:49:41<21:41:55, 5.72s/it]
69%|██████▉ | 30275/43939 [48:49:47<21:39:43, 5.71s/it]
69%|██████▉ | 30276/43939 [48:49:53<21:40:48, 5.71s/it]
69%|██████▉ | 30277/43939 [48:49:58<21:38:29, 5.70s/it]
69%|██████▉ | 30278/43939 [48:50:04<21:39:02, 5.71s/it]
69%|██████▉ | 30279/43939 [48:50:10<21:38:24, 5.70s/it]
69%|██████▉ | 30280/43939 [48:50:15<21:39:12, 5.71s/it]
69%|██████▉ | 30281/43939 [48:50:21<21:44:18, 5.73s/it]
69%|██████▉ | 30282/43939 [48:50:27<21:43:25, 5.73s/it]
69%|██████▉ | 30283/43939 [48:50:33<21:44:47, 5.73s/it]
69%|██████▉ | 30284/43939 [48:50:38<21:41:08, 5.72s/it]
69%|██████▉ | 30285/43939 [48:50:44<21:39:30, 5.71s/it]
69%|██████▉ | 30286/43939 [48:50:50<21:39:54, 5.71s/it]
69%|██████▉ | 30287/43939 [48:50:55<21:42:33, 5.72s/it]
69%|██████▉ | 30288/43939 [48:51:01<21:43:11, 5.73s/it]
69%|██████▉ | 30289/43939 [48:51:07<21:43:33, 5.73s/it]
69%|██████▉ | 30290/43939 [48:51:13<22:02:14, 5.81s/it]
69%|██████▉ | 30291/43939 [48:51:19<21:54:42, 5.78s/it]
69%|██████▉ | 30292/43939 [48:51:24<21:50:23, 5.76s/it]
69%|██████▉ | 30293/43939 [48:51:30<21:46:28, 5.74s/it]
69%|██████▉ | 30294/43939 [48:51:36<21:43:07, 5.73s/it]
69%|██████▉ | 30295/43939 [48:51:41<21:40:55, 5.72s/it]
69%|██████▉ | 30296/43939 [48:51:47<21:40:57, 5.72s/it]
69%|██████▉ | 30297/43939 [48:51:53<21:38:29, 5.71s/it]
69%|██████▉ | 30298/43939 [48:51:59<21:40:50, 5.72s/it]
69%|██████▉ | 30299/43939 [48:52:04<21:40:21, 5.72s/it]
69%|██████▉ | 30300/43939 [48:52:10<21:41:17, 5.72s/it]
{'loss': 0.7575, 'grad_norm': 1.406716227531433, 'learning_rate': 1.3295840068513832e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.82, 'epoch': 2.76}
69%|██████▉ | 30300/43939 [48:52:10<21:41:17, 5.72s/it]
69%|██████▉ | 30301/43939 [48:52:16<21:37:56, 5.71s/it]
69%|██████▉ | 30302/43939 [48:52:22<21:40:21, 5.72s/it]
69%|██████▉ | 30303/43939 [48:52:27<21:39:55, 5.72s/it]
69%|██████▉ | 30304/43939 [48:52:33<21:36:55, 5.71s/it]
69%|██████▉ | 30305/43939 [48:52:39<21:34:33, 5.70s/it]
69%|██████▉ | 30306/43939 [48:52:44<21:34:25, 5.70s/it]
69%|██████▉ | 30307/43939 [48:52:50<21:35:28, 5.70s/it]
69%|██████▉ | 30308/43939 [48:52:56<21:35:54, 5.70s/it]
69%|██████▉ | 30309/43939 [48:53:01<21:36:33, 5.71s/it]
69%|██████▉ | 30310/43939 [48:53:07<21:38:18, 5.72s/it]
69%|██████▉ | 30311/43939 [48:53:13<21:39:38, 5.72s/it]
69%|██████▉ | 30312/43939 [48:53:19<21:55:08, 5.79s/it]
69%|██████▉ | 30313/43939 [48:53:25<21:51:39, 5.78s/it]
69%|██████▉ | 30314/43939 [48:53:30<21:46:37, 5.75s/it]
69%|██████▉ | 30315/43939 [48:53:36<21:42:25, 5.74s/it]
69%|██████▉ | 30316/43939 [48:53:42<21:39:49, 5.72s/it]
69%|██████▉ | 30317/43939 [48:53:47<21:38:29, 5.72s/it]
69%|██████▉ | 30318/43939 [48:53:53<21:36:38, 5.71s/it]
69%|██████▉ | 30319/43939 [48:53:59<21:34:55, 5.70s/it]
69%|██████▉ | 30320/43939 [48:54:04<21:34:16, 5.70s/it]
69%|██████▉ | 30321/43939 [48:54:10<21:33:39, 5.70s/it]
69%|██████▉ | 30322/43939 [48:54:16<21:33:44, 5.70s/it]
69%|██████▉ | 30323/43939 [48:54:22<21:30:21, 5.69s/it]
69%|██████▉ | 30324/43939 [48:54:27<21:31:55, 5.69s/it]
69%|██████▉ | 30325/43939 [48:54:33<21:32:36, 5.70s/it]
69%|██████▉ | 30326/43939 [48:54:39<21:32:18, 5.70s/it]
69%|██████▉ | 30327/43939 [48:54:44<21:37:19, 5.72s/it]
69%|██████▉ | 30328/43939 [48:54:50<21:35:31, 5.71s/it]
69%|██████▉ | 30329/43939 [48:54:56<21:36:47, 5.72s/it]
69%|██████▉ | 30330/43939 [48:55:02<21:35:45, 5.71s/it]
69%|██████▉ | 30331/43939 [48:55:07<21:36:20, 5.72s/it]
69%|██████▉ | 30332/43939 [48:55:13<21:33:54, 5.71s/it]
69%|██████▉ | 30333/43939 [48:55:19<21:50:02, 5.78s/it]
69%|██████▉ | 30334/43939 [48:55:25<21:46:27, 5.76s/it]
69%|██████▉ | 30335/43939 [48:55:30<21:41:55, 5.74s/it]
69%|██████▉ | 30336/43939 [48:55:36<21:38:15, 5.73s/it]
69%|██████▉ | 30337/43939 [48:55:42<21:35:22, 5.71s/it]
69%|██████▉ | 30338/43939 [48:55:47<21:34:22, 5.71s/it]
69%|██████▉ | 30339/43939 [48:55:53<21:35:46, 5.72s/it]
69%|██████▉ | 30340/43939 [48:55:59<21:31:17, 5.70s/it]
69%|██████▉ | 30341/43939 [48:56:04<21:30:54, 5.70s/it]
69%|██████▉ | 30342/43939 [48:56:10<21:28:37, 5.69s/it]
69%|██████▉ | 30343/43939 [48:56:16<21:27:25, 5.68s/it]
69%|██████▉ | 30344/43939 [48:56:21<21:28:19, 5.69s/it]
69%|██████▉ | 30345/43939 [48:56:27<21:30:04, 5.69s/it]
69%|██████▉ | 30346/43939 [48:56:33<21:30:02, 5.69s/it]
69%|██████▉ | 30347/43939 [48:56:39<21:32:09, 5.70s/it]
69%|██████▉ | 30348/43939 [48:56:44<21:32:27, 5.71s/it]
69%|██████▉ | 30349/43939 [48:56:50<21:28:38, 5.69s/it]
69%|██████▉ | 30350/43939 [48:56:56<21:29:36, 5.69s/it]
{'loss': 0.778, 'grad_norm': 1.6327651739120483, 'learning_rate': 1.3208185513725758e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.54, 'epoch': 2.76}
69%|██████▉ | 30350/43939 [48:56:56<21:29:36, 5.69s/it]
69%|██████▉ | 30351/43939 [48:57:01<21:31:49, 5.70s/it]
69%|██████▉ | 30352/43939 [48:57:07<21:31:18, 5.70s/it]
69%|██████▉ | 30353/43939 [48:57:13<21:29:43, 5.70s/it]
69%|██████▉ | 30354/43939 [48:57:18<21:29:58, 5.70s/it]
69%|██████▉ | 30355/43939 [48:57:24<21:44:00, 5.76s/it]
69%|██████▉ | 30356/43939 [48:57:30<21:37:30, 5.73s/it]
69%|██████▉ | 30357/43939 [48:57:36<21:34:38, 5.72s/it]
69%|██████▉ | 30358/43939 [48:57:41<21:33:28, 5.71s/it]
69%|██████▉ | 30359/43939 [48:57:47<21:33:54, 5.72s/it]
69%|██████▉ | 30360/43939 [48:57:53<21:32:28, 5.71s/it]
69%|██████▉ | 30361/43939 [48:57:59<21:30:23, 5.70s/it]
69%|██████▉ | 30362/43939 [48:58:04<21:34:39, 5.72s/it]
69%|██████▉ | 30363/43939 [48:58:10<21:30:10, 5.70s/it]
69%|██████▉ | 30364/43939 [48:58:16<21:30:02, 5.70s/it]
69%|██████▉ | 30365/43939 [48:58:21<21:29:56, 5.70s/it]
69%|██████▉ | 30366/43939 [48:58:27<21:29:11, 5.70s/it]
69%|██████▉ | 30367/43939 [48:58:33<21:29:14, 5.70s/it]
69%|██████▉ | 30368/43939 [48:58:38<21:27:44, 5.69s/it]
69%|██████▉ | 30369/43939 [48:58:44<21:27:13, 5.69s/it]
69%|██████▉ | 30370/43939 [48:58:50<21:29:02, 5.70s/it]
69%|██████▉ | 30371/43939 [48:58:56<21:27:13, 5.69s/it]
69%|██████▉ | 30372/43939 [48:59:01<21:27:17, 5.69s/it]
69%|██████▉ | 30373/43939 [48:59:07<21:27:10, 5.69s/it]
69%|██████▉ | 30374/43939 [48:59:13<21:28:25, 5.70s/it]
69%|██████▉ | 30375/43939 [48:59:18<21:30:24, 5.71s/it]
69%|██████▉ | 30376/43939 [48:59:24<21:43:46, 5.77s/it]
69%|██████▉ | 30377/43939 [48:59:30<21:40:36, 5.75s/it]
69%|██████▉ | 30378/43939 [48:59:36<21:35:50, 5.73s/it]
69%|██████▉ | 30379/43939 [48:59:41<21:34:36, 5.73s/it]
69%|██████▉ | 30380/43939 [48:59:47<21:32:04, 5.72s/it]
69%|██████▉ | 30381/43939 [48:59:53<21:30:19, 5.71s/it]
69%|██████▉ | 30382/43939 [48:59:58<21:27:47, 5.70s/it]
69%|██████▉ | 30383/43939 [49:00:04<21:28:45, 5.70s/it]
69%|██████▉ | 30384/43939 [49:00:10<21:25:44, 5.69s/it]
69%|██████▉ | 30385/43939 [49:00:16<21:28:00, 5.70s/it]
69%|██████▉ | 30386/43939 [49:00:21<21:29:54, 5.71s/it]
69%|██████▉ | 30387/43939 [49:00:27<21:30:11, 5.71s/it]
69%|██████▉ | 30388/43939 [49:00:33<21:26:58, 5.70s/it]
69%|██████▉ | 30389/43939 [49:00:38<21:26:13, 5.70s/it]
69%|██████▉ | 30390/43939 [49:00:44<21:27:14, 5.70s/it]
69%|██████▉ | 30391/43939 [49:00:50<21:23:20, 5.68s/it]
69%|██████▉ | 30392/43939 [49:00:55<21:25:22, 5.69s/it]
69%|██████▉ | 30393/43939 [49:01:01<21:28:39, 5.71s/it]
69%|██████▉ | 30394/43939 [49:01:07<21:28:13, 5.71s/it]
69%|██████▉ | 30395/43939 [49:01:13<21:24:43, 5.69s/it]
69%|██████▉ | 30396/43939 [49:01:18<21:24:16, 5.69s/it]
69%|██████▉ | 30397/43939 [49:01:24<21:25:43, 5.70s/it]
69%|██████▉ | 30398/43939 [49:01:30<21:40:14, 5.76s/it]
69%|██████▉ | 30399/43939 [49:01:36<21:39:36, 5.76s/it]
69%|██████▉ | 30400/43939 [49:01:41<21:31:47, 5.72s/it]
{'loss': 0.7744, 'grad_norm': 1.525422215461731, 'learning_rate': 1.3120717002521588e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.67, 'epoch': 2.77}
69%|██████▉ | 30400/43939 [49:01:41<21:31:47, 5.72s/it]
69%|██████▉ | 30401/43939 [49:01:47<21:27:24, 5.71s/it]
69%|██████▉ | 30402/43939 [49:01:53<21:25:46, 5.70s/it]
69%|██████▉ | 30403/43939 [49:01:58<21:25:40, 5.70s/it]
69%|██████▉ | 30404/43939 [49:02:04<21:27:25, 5.71s/it]
69%|██████▉ | 30405/43939 [49:02:10<21:27:45, 5.71s/it]
69%|██████▉ | 30406/43939 [49:02:15<21:26:35, 5.70s/it]
69%|██████▉ | 30407/43939 [49:02:21<21:26:04, 5.70s/it]
69%|██████▉ | 30408/43939 [49:02:27<21:25:23, 5.70s/it]
69%|██████▉ | 30409/43939 [49:02:33<21:26:52, 5.71s/it]
69%|██████▉ | 30410/43939 [49:02:38<21:28:10, 5.71s/it]
69%|██████▉ | 30411/43939 [49:02:44<21:26:57, 5.71s/it]
69%|██████▉ | 30412/43939 [49:02:50<21:24:29, 5.70s/it]
69%|██████▉ | 30413/43939 [49:02:55<21:24:43, 5.70s/it]
69%|██████▉ | 30414/43939 [49:03:01<21:24:58, 5.70s/it]
69%|██████▉ | 30415/43939 [49:03:07<21:28:21, 5.72s/it]
69%|██████▉ | 30416/43939 [49:03:12<21:25:03, 5.70s/it]
69%|██████▉ | 30417/43939 [49:03:18<21:27:11, 5.71s/it]
69%|██████▉ | 30418/43939 [49:03:24<21:25:53, 5.71s/it]
69%|██████▉ | 30419/43939 [49:03:30<21:40:45, 5.77s/it]
69%|██████▉ | 30420/43939 [49:03:35<21:35:49, 5.75s/it]
69%|██████▉ | 30421/43939 [49:03:41<21:34:47, 5.75s/it]
69%|██████▉ | 30422/43939 [49:03:47<21:32:50, 5.74s/it]
69%|██████▉ | 30423/43939 [49:03:53<21:25:23, 5.71s/it]
69%|██████▉ | 30424/43939 [49:03:58<21:24:15, 5.70s/it]
69%|██████▉ | 30425/43939 [49:04:04<21:23:31, 5.70s/it]
69%|██████▉ | 30426/43939 [49:04:10<21:27:12, 5.72s/it]
69%|██████▉ | 30427/43939 [49:04:15<21:25:28, 5.71s/it]
69%|██████▉ | 30428/43939 [49:04:21<21:25:10, 5.71s/it]
69%|██████▉ | 30429/43939 [49:04:27<21:26:48, 5.71s/it]
69%|██████▉ | 30430/43939 [49:04:33<21:24:57, 5.71s/it]
69%|██████▉ | 30431/43939 [49:04:38<21:21:57, 5.69s/it]
69%|██████▉ | 30432/43939 [49:04:44<21:19:22, 5.68s/it]
69%|██████▉ | 30433/43939 [49:04:50<21:22:48, 5.70s/it]
69%|██████▉ | 30434/43939 [49:04:55<21:20:17, 5.69s/it]
69%|██████▉ | 30435/43939 [49:05:01<21:22:11, 5.70s/it]
69%|██████▉ | 30436/43939 [49:05:07<21:21:15, 5.69s/it]
69%|██████▉ | 30437/43939 [49:05:12<21:20:04, 5.69s/it]
69%|██████▉ | 30438/43939 [49:05:18<21:23:20, 5.70s/it]
69%|██████▉ | 30439/43939 [49:05:24<21:23:59, 5.71s/it]
69%|██████▉ | 30440/43939 [49:05:30<21:46:41, 5.81s/it]
69%|██████▉ | 30441/43939 [49:05:36<21:39:47, 5.78s/it]
69%|██████▉ | 30442/43939 [49:05:41<21:36:30, 5.76s/it]
69%|██████▉ | 30443/43939 [49:05:47<21:30:35, 5.74s/it]
69%|██████▉ | 30444/43939 [49:05:53<21:29:11, 5.73s/it]
69%|██████▉ | 30445/43939 [49:05:58<21:27:49, 5.73s/it]
69%|██████▉ | 30446/43939 [49:06:04<21:25:03, 5.71s/it]
69%|██████▉ | 30447/43939 [49:06:10<21:26:19, 5.72s/it]
69%|██████▉ | 30448/43939 [49:06:16<21:29:55, 5.74s/it]
69%|██████▉ | 30449/43939 [49:06:21<21:29:57, 5.74s/it]
69%|██████▉ | 30450/43939 [49:06:27<21:28:10, 5.73s/it]
{'loss': 0.7763, 'grad_norm': 1.5094884634017944, 'learning_rate': 1.3033435914922637e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.77, 'epoch': 2.77}
69%|██████▉ | 30450/43939 [49:06:27<21:28:10, 5.73s/it]
69%|██████▉ | 30451/43939 [49:06:33<21:25:03, 5.72s/it]
69%|██████▉ | 30452/43939 [49:06:38<21:26:48, 5.72s/it]
69%|██████▉ | 30453/43939 [49:06:44<21:25:23, 5.72s/it]
69%|██████▉ | 30454/43939 [49:06:50<21:26:10, 5.72s/it]
69%|██████▉ | 30455/43939 [49:06:56<21:23:43, 5.71s/it]
69%|██████▉ | 30456/43939 [49:07:01<21:21:30, 5.70s/it]
69%|██████▉ | 30457/43939 [49:07:07<21:24:47, 5.72s/it]
69%|██████▉ | 30458/43939 [49:07:13<21:24:51, 5.72s/it]
69%|██████▉ | 30459/43939 [49:07:18<21:20:53, 5.70s/it]
69%|██████▉ | 30460/43939 [49:07:24<21:20:32, 5.70s/it]
69%|██████▉ | 30461/43939 [49:07:30<21:21:27, 5.70s/it]
69%|██████▉ | 30462/43939 [49:07:36<21:36:25, 5.77s/it]
69%|██████▉ | 30463/43939 [49:07:41<21:32:51, 5.76s/it]
69%|██████▉ | 30464/43939 [49:07:47<21:27:06, 5.73s/it]
69%|██████▉ | 30465/43939 [49:07:53<21:25:23, 5.72s/it]
69%|██████▉ | 30466/43939 [49:07:59<21:22:13, 5.71s/it]
69%|██████▉ | 30467/43939 [49:08:04<21:21:17, 5.71s/it]
69%|██████▉ | 30468/43939 [49:08:10<21:23:49, 5.72s/it]
69%|██████▉ | 30469/43939 [49:08:16<21:25:09, 5.72s/it]
69%|██████▉ | 30470/43939 [49:08:21<21:27:03, 5.73s/it]
69%|██████▉ | 30471/43939 [49:08:27<21:22:40, 5.71s/it]
69%|██████▉ | 30472/43939 [49:08:33<21:21:03, 5.71s/it]
69%|██████▉ | 30473/43939 [49:08:39<21:19:46, 5.70s/it]
69%|██████▉ | 30474/43939 [49:08:44<21:18:57, 5.70s/it]
69%|██████▉ | 30475/43939 [49:08:50<21:18:58, 5.70s/it]
69%|██████▉ | 30476/43939 [49:08:56<21:18:10, 5.70s/it]
69%|██████▉ | 30477/43939 [49:09:01<21:15:37, 5.69s/it]
69%|██████▉ | 30478/43939 [49:09:07<21:15:36, 5.69s/it]
69%|██████▉ | 30479/43939 [49:09:13<21:16:56, 5.69s/it]
69%|██████▉ | 30480/43939 [49:09:18<21:21:32, 5.71s/it]
69%|██████▉ | 30481/43939 [49:09:24<21:21:16, 5.71s/it]
69%|██████▉ | 30482/43939 [49:09:30<21:22:18, 5.72s/it]
69%|██████▉ | 30483/43939 [49:09:36<21:19:18, 5.70s/it]
69%|██████▉ | 30484/43939 [49:09:41<21:35:11, 5.78s/it]
69%|██████▉ | 30485/43939 [49:09:47<21:28:29, 5.75s/it]
69%|██████▉ | 30486/43939 [49:09:53<21:26:11, 5.74s/it]
69%|██████▉ | 30487/43939 [49:09:59<21:24:43, 5.73s/it]
69%|██████▉ | 30488/43939 [49:10:04<21:23:25, 5.72s/it]
69%|██████▉ | 30489/43939 [49:10:10<21:23:02, 5.72s/it]
69%|██████▉ | 30490/43939 [49:10:16<21:23:01, 5.72s/it]
69%|██████▉ | 30491/43939 [49:10:21<21:24:24, 5.73s/it]
69%|██████▉ | 30492/43939 [49:10:27<21:23:45, 5.73s/it]
69%|██████▉ | 30493/43939 [49:10:33<21:22:42, 5.72s/it]
69%|██████▉ | 30494/43939 [49:10:39<21:21:05, 5.72s/it]
69%|██████▉ | 30495/43939 [49:10:44<21:18:16, 5.70s/it]
69%|██████▉ | 30496/43939 [49:10:50<21:17:30, 5.70s/it]
69%|██████▉ | 30497/43939 [49:10:56<21:17:08, 5.70s/it]
69%|██████▉ | 30498/43939 [49:11:01<21:18:12, 5.71s/it]
69%|██████▉ | 30499/43939 [49:11:07<21:18:45, 5.71s/it]
69%|██████▉ | 30500/43939 [49:11:13<21:17:21, 5.70s/it]
{'loss': 0.773, 'grad_norm': 1.7871735095977783, 'learning_rate': 1.2946343627993166e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.31, 'epoch': 2.78}
69%|██████▉ | 30500/43939 [49:11:13<21:17:21, 5.70s/it]
69%|██████▉ | 30501/43939 [49:11:19<21:18:53, 5.71s/it]
69%|██████▉ | 30502/43939 [49:11:24<21:19:01, 5.71s/it]
69%|██████▉ | 30503/43939 [49:11:30<21:18:37, 5.71s/it]
69%|██████▉ | 30504/43939 [49:11:36<21:17:52, 5.71s/it]
69%|██████▉ | 30505/43939 [49:11:42<21:31:02, 5.77s/it]
69%|██████▉ | 30506/43939 [49:11:47<21:27:39, 5.75s/it]
69%|██████▉ | 30507/43939 [49:11:53<21:24:26, 5.74s/it]
69%|██████▉ | 30508/43939 [49:11:59<21:24:42, 5.74s/it]
69%|██████▉ | 30509/43939 [49:12:04<21:22:00, 5.73s/it]
69%|██████▉ | 30510/43939 [49:12:10<21:21:26, 5.73s/it]
69%|██████▉ | 30511/43939 [49:12:16<21:20:59, 5.72s/it]
69%|██████▉ | 30512/43939 [49:12:22<21:20:33, 5.72s/it]
69%|██████▉ | 30513/43939 [49:12:27<21:22:39, 5.73s/it]
69%|██████▉ | 30514/43939 [49:12:33<21:19:50, 5.72s/it]
69%|██████▉ | 30515/43939 [49:12:39<21:17:12, 5.71s/it]
69%|██████▉ | 30516/43939 [49:12:44<21:15:39, 5.70s/it]
69%|██████▉ | 30517/43939 [49:12:50<21:15:44, 5.70s/it]
69%|██████▉ | 30518/43939 [49:12:56<21:17:43, 5.71s/it]
69%|██████▉ | 30519/43939 [49:13:02<21:16:31, 5.71s/it]
69%|██████▉ | 30520/43939 [49:13:07<21:18:56, 5.72s/it]
69%|██████▉ | 30521/43939 [49:13:13<21:14:51, 5.70s/it]
69%|██████▉ | 30522/43939 [49:13:19<21:14:30, 5.70s/it]
69%|██████▉ | 30523/43939 [49:13:24<21:16:30, 5.71s/it]
69%|██████▉ | 30524/43939 [49:13:30<21:15:05, 5.70s/it]
69%|██████▉ | 30525/43939 [49:13:36<21:12:28, 5.69s/it]
69%|██████▉ | 30526/43939 [49:13:42<21:29:48, 5.77s/it]
69%|██████▉ | 30527/43939 [49:13:47<21:29:42, 5.77s/it]
69%|██████▉ | 30528/43939 [49:13:53<21:26:04, 5.75s/it]
69%|██████▉ | 30529/43939 [49:13:59<21:24:47, 5.75s/it]
69%|██████▉ | 30530/43939 [49:14:05<21:20:36, 5.73s/it]
69%|██████▉ | 30531/43939 [49:14:10<21:19:58, 5.73s/it]
69%|██████▉ | 30532/43939 [49:14:16<21:20:35, 5.73s/it]
69%|██████▉ | 30533/43939 [49:14:22<21:16:24, 5.71s/it]
69%|██████▉ | 30534/43939 [49:14:27<21:13:58, 5.70s/it]
69%|██████▉ | 30535/43939 [49:14:33<21:15:01, 5.71s/it]
69%|██████▉ | 30536/43939 [49:14:39<21:11:40, 5.69s/it]
69%|██████▉ | 30537/43939 [49:14:44<21:13:45, 5.70s/it]
70%|██████▉ | 30538/43939 [49:14:50<21:11:40, 5.69s/it]
70%|██████▉ | 30539/43939 [49:14:56<21:11:48, 5.69s/it]
70%|██████▉ | 30540/43939 [49:15:02<21:14:32, 5.71s/it]
70%|██████▉ | 30541/43939 [49:15:07<21:16:25, 5.72s/it]
70%|██████▉ | 30542/43939 [49:15:13<21:14:19, 5.71s/it]
70%|██████▉ | 30543/43939 [49:15:19<21:12:10, 5.70s/it]
70%|██████▉ | 30544/43939 [49:15:24<21:09:41, 5.69s/it]
70%|██████▉ | 30545/43939 [49:15:30<21:13:29, 5.70s/it]
70%|██████▉ | 30546/43939 [49:15:36<21:15:33, 5.71s/it]
70%|██████▉ | 30547/43939 [49:15:42<21:33:09, 5.79s/it]
70%|██████▉ | 30548/43939 [49:15:48<21:28:03, 5.77s/it]
70%|██████▉ | 30549/43939 [49:15:53<21:22:59, 5.75s/it]
70%|██████▉ | 30550/43939 [49:15:59<21:17:50, 5.73s/it]
{'loss': 0.7703, 'grad_norm': 1.6897590160369873, 'learning_rate': 1.2859441515818656e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.29, 'epoch': 2.78}
70%|██████▉ | 30550/43939 [49:15:59<21:17:50, 5.73s/it]
70%|██████▉ | 30551/43939 [49:16:05<21:20:37, 5.74s/it]
70%|██████▉ | 30552/43939 [49:16:10<21:19:43, 5.74s/it]
70%|██████▉ | 30553/43939 [49:16:16<21:16:39, 5.72s/it]
70%|██████▉ | 30554/43939 [49:16:22<21:19:28, 5.74s/it]
70%|██████▉ | 30555/43939 [49:16:28<21:18:00, 5.73s/it]
70%|██████▉ | 30556/43939 [49:16:33<21:15:51, 5.72s/it]
70%|██████▉ | 30557/43939 [49:16:39<21:14:41, 5.72s/it]
70%|██████▉ | 30558/43939 [49:16:45<21:14:45, 5.72s/it]
70%|██████▉ | 30559/43939 [49:16:50<21:11:40, 5.70s/it]
70%|██████▉ | 30560/43939 [49:16:56<21:14:10, 5.71s/it]
70%|██████▉ | 30561/43939 [49:17:02<21:12:45, 5.71s/it]
70%|██████▉ | 30562/43939 [49:17:07<21:11:30, 5.70s/it]
70%|██████▉ | 30563/43939 [49:17:13<21:14:22, 5.72s/it]
70%|██████▉ | 30564/43939 [49:17:19<21:12:20, 5.71s/it]
70%|██████▉ | 30565/43939 [49:17:25<21:16:26, 5.73s/it]
70%|██████▉ | 30566/43939 [49:17:30<21:19:34, 5.74s/it]
70%|██████▉ | 30567/43939 [49:17:36<21:17:06, 5.73s/it]
70%|██████▉ | 30568/43939 [49:17:42<21:28:36, 5.78s/it]
70%|██████▉ | 30569/43939 [49:17:48<21:26:46, 5.77s/it]
70%|██████▉ | 30570/43939 [49:17:54<21:20:37, 5.75s/it]
70%|██████▉ | 30571/43939 [49:17:59<21:17:12, 5.73s/it]
70%|██████▉ | 30572/43939 [49:18:05<21:12:32, 5.71s/it]
70%|██████▉ | 30573/43939 [49:18:11<21:08:39, 5.69s/it]
70%|██████▉ | 30574/43939 [49:18:16<21:10:01, 5.70s/it]
70%|██████▉ | 30575/43939 [49:18:22<21:07:14, 5.69s/it]
70%|██████▉ | 30576/43939 [49:18:28<21:10:10, 5.70s/it]
70%|██████▉ | 30577/43939 [49:18:33<21:09:27, 5.70s/it]
70%|██████▉ | 30578/43939 [49:18:39<21:12:07, 5.71s/it]
70%|██████▉ | 30579/43939 [49:18:45<21:13:36, 5.72s/it]
70%|██████▉ | 30580/43939 [49:18:50<21:10:17, 5.71s/it]
70%|██████▉ | 30581/43939 [49:18:56<21:06:55, 5.69s/it]
70%|██████▉ | 30582/43939 [49:19:02<21:07:29, 5.69s/it]
70%|██████▉ | 30583/43939 [49:19:08<21:07:30, 5.69s/it]
70%|██████▉ | 30584/43939 [49:19:13<21:09:59, 5.71s/it]
70%|██████▉ | 30585/43939 [49:19:19<21:13:38, 5.72s/it]
70%|██████▉ | 30586/43939 [49:19:25<21:12:29, 5.72s/it]
70%|██████▉ | 30587/43939 [49:19:30<21:12:11, 5.72s/it]
70%|██████▉ | 30588/43939 [49:19:36<21:11:32, 5.71s/it]
70%|██████▉ | 30589/43939 [49:19:42<21:26:15, 5.78s/it]
70%|██████▉ | 30590/43939 [49:19:48<21:16:56, 5.74s/it]
70%|██████▉ | 30591/43939 [49:19:53<21:15:57, 5.74s/it]
70%|██████▉ | 30592/43939 [49:19:59<21:11:47, 5.72s/it]
70%|██████▉ | 30593/43939 [49:20:05<21:10:47, 5.71s/it]
70%|██████▉ | 30594/43939 [49:20:11<21:08:47, 5.70s/it]
70%|██████▉ | 30595/43939 [49:20:16<21:08:51, 5.71s/it]
70%|██████▉ | 30596/43939 [49:20:22<21:06:26, 5.69s/it]
70%|██████▉ | 30597/43939 [49:20:28<21:06:56, 5.70s/it]
70%|██████▉ | 30598/43939 [49:20:33<21:08:00, 5.70s/it]
70%|██████▉ | 30599/43939 [49:20:39<21:10:39, 5.72s/it]
70%|██████▉ | 30600/43939 [49:20:45<21:12:18, 5.72s/it]
{'loss': 0.765, 'grad_norm': 1.4181755781173706, 'learning_rate': 1.2772730949484168e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.15, 'epoch': 2.79}
70%|██████▉ | 30600/43939 [49:20:45<21:12:18, 5.72s/it]
70%|██████▉ | 30601/43939 [49:20:51<21:12:44, 5.73s/it]
70%|██████▉ | 30602/43939 [49:20:56<21:09:36, 5.71s/it]
70%|██████▉ | 30603/43939 [49:21:02<21:09:35, 5.71s/it]
70%|██████▉ | 30604/43939 [49:21:08<21:12:25, 5.73s/it]
70%|██████▉ | 30605/43939 [49:21:13<21:14:21, 5.73s/it]
70%|██████▉ | 30606/43939 [49:21:19<21:09:51, 5.71s/it]
70%|██████▉ | 30607/43939 [49:21:25<21:06:59, 5.70s/it]
70%|██████▉ | 30608/43939 [49:21:30<21:05:42, 5.70s/it]
70%|██████▉ | 30609/43939 [49:21:36<21:05:50, 5.70s/it]
70%|██████▉ | 30610/43939 [49:21:42<21:05:26, 5.70s/it]
70%|██████▉ | 30611/43939 [49:21:48<21:18:02, 5.75s/it]
70%|██████▉ | 30612/43939 [49:21:54<21:18:23, 5.76s/it]
70%|██████▉ | 30613/43939 [49:21:59<21:15:34, 5.74s/it]
70%|██████▉ | 30614/43939 [49:22:05<21:11:35, 5.73s/it]
70%|██████▉ | 30615/43939 [49:22:11<21:10:02, 5.72s/it]
70%|██████▉ | 30616/43939 [49:22:16<21:08:46, 5.71s/it]
70%|██████▉ | 30617/43939 [49:22:22<21:12:25, 5.73s/it]
70%|██████▉ | 30618/43939 [49:22:28<21:09:39, 5.72s/it]
70%|██████▉ | 30619/43939 [49:22:33<21:05:10, 5.70s/it]
70%|██████▉ | 30620/43939 [49:22:39<21:05:23, 5.70s/it]
70%|██████▉ | 30621/43939 [49:22:45<21:07:27, 5.71s/it]
70%|██████▉ | 30622/43939 [49:22:51<21:09:49, 5.72s/it]
70%|██████▉ | 30623/43939 [49:22:56<21:05:51, 5.70s/it]
70%|██████▉ | 30624/43939 [49:23:02<21:06:20, 5.71s/it]
70%|██████▉ | 30625/43939 [49:23:08<21:10:17, 5.72s/it]
70%|██████▉ | 30626/43939 [49:23:13<21:07:18, 5.71s/it]
70%|██████▉ | 30627/43939 [49:23:19<21:04:39, 5.70s/it]
70%|██████▉ | 30628/43939 [49:23:25<21:05:05, 5.70s/it]
70%|██████▉ | 30629/43939 [49:23:30<21:02:52, 5.69s/it]
70%|██████▉ | 30630/43939 [49:23:36<21:03:52, 5.70s/it]
70%|██████▉ | 30631/43939 [49:23:42<21:04:20, 5.70s/it]
70%|██████▉ | 30632/43939 [49:23:48<21:17:10, 5.76s/it]
70%|██████▉ | 30633/43939 [49:23:54<21:13:04, 5.74s/it]
70%|██████▉ | 30634/43939 [49:23:59<21:11:11, 5.73s/it]
70%|██████▉ | 30635/43939 [49:24:05<21:13:53, 5.75s/it]
70%|██████▉ | 30636/43939 [49:24:11<21:11:07, 5.73s/it]
70%|██████▉ | 30637/43939 [49:24:16<21:09:21, 5.73s/it]
70%|██████▉ | 30638/43939 [49:24:22<21:07:23, 5.72s/it]
70%|██████▉ | 30639/43939 [49:24:28<21:03:44, 5.70s/it]
70%|██████▉ | 30640/43939 [49:24:33<21:05:41, 5.71s/it]
70%|██████▉ | 30641/43939 [49:24:39<21:01:35, 5.69s/it]
70%|██████▉ | 30642/43939 [49:24:45<21:02:02, 5.69s/it]
70%|██████▉ | 30643/43939 [49:24:51<21:01:50, 5.69s/it]
70%|██████▉ | 30644/43939 [49:24:56<21:02:50, 5.70s/it]
70%|██████▉ | 30645/43939 [49:25:02<21:04:48, 5.71s/it]
70%|██████▉ | 30646/43939 [49:25:08<21:03:55, 5.70s/it]
70%|██████▉ | 30647/43939 [49:25:13<21:03:21, 5.70s/it]
70%|██████▉ | 30648/43939 [49:25:19<21:02:52, 5.70s/it]
70%|██████▉ | 30649/43939 [49:25:25<21:00:23, 5.69s/it]
70%|██████▉ | 30650/43939 [49:25:30<21:03:17, 5.70s/it]
{'loss': 0.7762, 'grad_norm': 1.358834147453308, 'learning_rate': 1.2686213297052638e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.07, 'epoch': 2.79}
70%|██████▉ | 30650/43939 [49:25:30<21:03:17, 5.70s/it]
70%|██████▉ | 30651/43939 [49:25:36<21:03:50, 5.71s/it]
70%|██████▉ | 30652/43939 [49:25:42<21:03:39, 5.71s/it]
70%|██████▉ | 30653/43939 [49:25:48<21:21:03, 5.79s/it]
70%|██████▉ | 30654/43939 [49:25:54<21:17:49, 5.77s/it]
70%|██████▉ | 30655/43939 [49:25:59<21:14:13, 5.76s/it]
70%|██████▉ | 30656/43939 [49:26:05<21:10:19, 5.74s/it]
70%|██████▉ | 30657/43939 [49:26:11<21:09:55, 5.74s/it]
70%|██████▉ | 30658/43939 [49:26:16<21:07:50, 5.73s/it]
70%|██████▉ | 30659/43939 [49:26:22<21:04:59, 5.72s/it]
70%|██████▉ | 30660/43939 [49:26:28<21:04:31, 5.71s/it]
70%|██████▉ | 30661/43939 [49:26:34<21:06:40, 5.72s/it]
70%|██████▉ | 30662/43939 [49:26:39<21:03:01, 5.71s/it]
70%|██████▉ | 30663/43939 [49:26:45<21:02:07, 5.70s/it]
70%|██████▉ | 30664/43939 [49:26:51<21:00:43, 5.70s/it]
70%|██████▉ | 30665/43939 [49:26:56<21:00:40, 5.70s/it]
70%|██████▉ | 30666/43939 [49:27:02<21:00:36, 5.70s/it]
70%|██████▉ | 30667/43939 [49:27:08<21:00:27, 5.70s/it]
70%|██████▉ | 30668/43939 [49:27:13<21:01:36, 5.70s/it]
70%|██████▉ | 30669/43939 [49:27:19<21:02:35, 5.71s/it]
70%|██████▉ | 30670/43939 [49:27:25<21:02:43, 5.71s/it]
70%|██████▉ | 30671/43939 [49:27:31<20:58:34, 5.69s/it]
70%|██████▉ | 30672/43939 [49:27:36<20:59:49, 5.70s/it]
70%|██████▉ | 30673/43939 [49:27:42<20:59:25, 5.70s/it]
70%|██████▉ | 30674/43939 [49:27:48<21:11:36, 5.75s/it]
70%|██████▉ | 30675/43939 [49:27:54<21:11:46, 5.75s/it]
70%|██████▉ | 30676/43939 [49:27:59<21:08:26, 5.74s/it]
70%|██████▉ | 30677/43939 [49:28:05<21:05:20, 5.72s/it]
70%|██████▉ | 30678/43939 [49:28:11<21:05:38, 5.73s/it]
70%|██████▉ | 30679/43939 [49:28:16<21:04:54, 5.72s/it]
70%|██████▉ | 30680/43939 [49:28:22<21:04:32, 5.72s/it]
70%|██████▉ | 30681/43939 [49:28:28<21:04:52, 5.72s/it]
70%|██████▉ | 30682/43939 [49:28:34<21:05:53, 5.73s/it]
70%|██████▉ | 30683/43939 [49:28:39<21:03:06, 5.72s/it]
70%|██████▉ | 30684/43939 [49:28:45<21:01:27, 5.71s/it]
70%|██████▉ | 30685/43939 [49:28:51<21:00:26, 5.71s/it]
70%|██████▉ | 30686/43939 [49:28:56<21:01:13, 5.71s/it]
70%|██████▉ | 30687/43939 [49:29:02<21:05:22, 5.73s/it]
70%|██████▉ | 30688/43939 [49:29:08<21:02:43, 5.72s/it]
70%|██████▉ | 30689/43939 [49:29:14<21:03:27, 5.72s/it]
70%|██████▉ | 30690/43939 [49:29:19<21:03:17, 5.72s/it]
70%|██████▉ | 30691/43939 [49:29:25<21:01:44, 5.71s/it]
70%|██████▉ | 30692/43939 [49:29:31<21:01:46, 5.71s/it]
70%|██████▉ | 30693/43939 [49:29:37<21:05:25, 5.73s/it]
70%|██████▉ | 30694/43939 [49:29:42<21:04:50, 5.73s/it]
70%|██████▉ | 30695/43939 [49:29:48<21:06:02, 5.74s/it]
70%|██████▉ | 30696/43939 [49:29:54<21:23:25, 5.81s/it]
70%|██████▉ | 30697/43939 [49:30:00<21:16:59, 5.79s/it]
70%|██████▉ | 30698/43939 [49:30:05<21:12:06, 5.76s/it]
70%|██████▉ | 30699/43939 [49:30:11<21:08:31, 5.75s/it]
70%|██████▉ | 30700/43939 [49:30:17<21:04:49, 5.73s/it]
{'loss': 0.7621, 'grad_norm': 1.68776273727417, 'learning_rate': 1.2599889923543349e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.27, 'epoch': 2.79}
70%|██████▉ | 30700/43939 [49:30:17<21:04:49, 5.73s/it]
70%|██████▉ | 30701/43939 [49:30:23<21:01:25, 5.72s/it]
70%|██████▉ | 30702/43939 [49:30:28<20:58:55, 5.71s/it]
70%|██████▉ | 30703/43939 [49:30:34<21:01:10, 5.72s/it]
70%|██████▉ | 30704/43939 [49:30:40<21:02:27, 5.72s/it]
70%|██████▉ | 30705/43939 [49:30:45<21:00:10, 5.71s/it]
70%|██████▉ | 30706/43939 [49:30:51<21:02:43, 5.73s/it]
70%|██████▉ | 30707/43939 [49:30:57<21:01:06, 5.72s/it]
70%|██████▉ | 30708/43939 [49:31:03<21:03:04, 5.73s/it]
70%|██████▉ | 30709/43939 [49:31:08<21:02:37, 5.73s/it]
70%|██████▉ | 30710/43939 [49:31:14<21:02:58, 5.73s/it]
70%|██████▉ | 30711/43939 [49:31:20<20:59:59, 5.72s/it]
70%|██████▉ | 30712/43939 [49:31:25<20:58:57, 5.71s/it]
70%|██████▉ | 30713/43939 [49:31:31<20:59:23, 5.71s/it]
70%|██████▉ | 30714/43939 [49:31:37<20:59:24, 5.71s/it]
70%|██████▉ | 30715/43939 [49:31:43<20:58:25, 5.71s/it]
70%|██████▉ | 30716/43939 [49:31:48<20:57:57, 5.71s/it]
70%|██████▉ | 30717/43939 [49:31:54<20:56:43, 5.70s/it]
70%|██████▉ | 30718/43939 [49:32:00<21:13:10, 5.78s/it]
70%|██████▉ | 30719/43939 [49:32:06<21:09:38, 5.76s/it]
70%|██████▉ | 30720/43939 [49:32:11<21:05:15, 5.74s/it]
70%|██████▉ | 30721/43939 [49:32:17<21:03:30, 5.74s/it]
70%|██████▉ | 30722/43939 [49:32:23<21:02:45, 5.73s/it]
70%|██████▉ | 30723/43939 [49:32:29<21:06:21, 5.75s/it]
70%|██████▉ | 30724/43939 [49:32:34<21:03:11, 5.74s/it]
70%|██████▉ | 30725/43939 [49:32:40<20:59:21, 5.72s/it]
70%|██████▉ | 30726/43939 [49:32:46<20:57:27, 5.71s/it]
70%|██████▉ | 30727/43939 [49:32:51<20:58:22, 5.71s/it]
70%|██████▉ | 30728/43939 [49:32:57<20:56:11, 5.71s/it]
70%|██████▉ | 30729/43939 [49:33:03<20:55:33, 5.70s/it]
70%|██████▉ | 30730/43939 [49:33:08<20:55:41, 5.70s/it]
70%|██████▉ | 30731/43939 [49:33:14<20:54:40, 5.70s/it]
70%|██████▉ | 30732/43939 [49:33:20<20:56:26, 5.71s/it]
70%|██████▉ | 30733/43939 [49:33:26<20:56:03, 5.71s/it]
70%|██████▉ | 30734/43939 [49:33:31<20:55:50, 5.71s/it]
70%|██████▉ | 30735/43939 [49:33:37<20:55:19, 5.70s/it]
70%|██████▉ | 30736/43939 [49:33:43<20:53:03, 5.69s/it]
70%|██████▉ | 30737/43939 [49:33:48<20:50:35, 5.68s/it]
70%|██████▉ | 30738/43939 [49:33:54<20:58:33, 5.72s/it]
70%|██████▉ | 30739/43939 [49:34:00<21:15:17, 5.80s/it]
70%|██████▉ | 30740/43939 [49:34:06<21:07:12, 5.76s/it]
70%|██████▉ | 30741/43939 [49:34:11<21:01:45, 5.74s/it]
70%|██████▉ | 30742/43939 [49:34:17<21:00:13, 5.73s/it]
70%|██████▉ | 30743/43939 [49:34:23<20:57:34, 5.72s/it]
70%|██████▉ | 30744/43939 [49:34:29<20:57:24, 5.72s/it]
70%|██████▉ | 30745/43939 [49:34:34<21:00:03, 5.73s/it]
70%|██████▉ | 30746/43939 [49:34:40<20:58:07, 5.72s/it]
70%|██████▉ | 30747/43939 [49:34:46<20:57:33, 5.72s/it]
70%|██████▉ | 30748/43939 [49:34:51<20:59:13, 5.73s/it]
70%|██████▉ | 30749/43939 [49:34:57<20:58:44, 5.73s/it]
70%|██████▉ | 30750/43939 [49:35:03<20:57:09, 5.72s/it]
{'loss': 0.7722, 'grad_norm': 1.5909497737884521, 'learning_rate': 1.2513762190910372e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.41, 'epoch': 2.8}
70%|██████▉ | 30750/43939 [49:35:03<20:57:09, 5.72s/it]
70%|██████▉ | 30751/43939 [49:35:09<20:56:38, 5.72s/it]
70%|██████▉ | 30752/43939 [49:35:14<20:53:45, 5.70s/it]
70%|██████▉ | 30753/43939 [49:35:20<20:51:42, 5.70s/it]
70%|██████▉ | 30754/43939 [49:35:26<20:52:58, 5.70s/it]
70%|██████▉ | 30755/43939 [49:35:31<20:49:50, 5.69s/it]
70%|██████▉ | 30756/43939 [49:35:37<20:49:58, 5.69s/it]
70%|██████▉ | 30757/43939 [49:35:43<20:49:11, 5.69s/it]
70%|███████ | 30758/43939 [49:35:48<20:49:30, 5.69s/it][2025-11-30 19:40:02,211] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:12, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:34, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:30<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:26, 2.78s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.76s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.76s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.75s/it][A
62%|██████▎ | 45/72 [02:00<01:14, 2.74s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.74s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:22<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.72s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.72s/it][A
85%|████████▍ | 61/72 [02:43<00:29, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.72s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.72s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A{'eval_loss': 0.978504478931427, 'eval_runtime': 199.068, 'eval_samples_per_second': 5.018, 'eval_steps_per_second': 0.362, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 2.8}
70%|███████ | 30758/43939 [49:39:07<20:49:30, 5.69s/it]
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A
70%|███████ | 30759/43939 [49:39:13<239:37:33, 65.45s/it]
70%|███████ | 30760/43939 [49:39:19<174:17:55, 47.61s/it]
70%|███████ | 30761/43939 [49:39:25<128:15:15, 35.04s/it]
70%|███████ | 30762/43939 [49:39:31<96:01:50, 26.24s/it]
70%|███████ | 30763/43939 [49:39:36<73:27:34, 20.07s/it]
70%|███████ | 30764/43939 [49:39:42<57:40:16, 15.76s/it]
70%|███████ | 30765/43939 [49:39:48<46:34:29, 12.73s/it]
70%|███████ | 30766/43939 [49:39:53<38:52:12, 10.62s/it]
70%|███████ | 30767/43939 [49:39:59<33:29:37, 9.15s/it]
70%|███████ | 30768/43939 [49:40:05<29:41:33, 8.12s/it]
70%|███████ | 30769/43939 [49:40:11<27:05:15, 7.40s/it]
70%|███████ | 30770/43939 [49:40:16<25:11:42, 6.89s/it]
70%|███████ | 30771/43939 [49:40:22<23:54:18, 6.54s/it]
70%|███████ | 30772/43939 [49:40:28<22:56:42, 6.27s/it]
70%|███████ | 30773/43939 [49:40:33<22:19:07, 6.10s/it]
70%|███████ | 30774/43939 [49:40:39<21:54:26, 5.99s/it]
70%|███████ | 30775/43939 [49:40:45<21:35:19, 5.90s/it]
70%|███████ | 30776/43939 [49:40:50<21:22:25, 5.85s/it]
70%|███████ | 30777/43939 [49:40:56<21:13:55, 5.81s/it]
70%|███████ | 30778/43939 [49:41:02<21:06:53, 5.78s/it]
70%|███████ | 30779/43939 [49:41:08<21:01:07, 5.75s/it]
70%|███████ | 30780/43939 [49:41:13<20:56:08, 5.73s/it]
70%|███████ | 30781/43939 [49:41:19<21:08:07, 5.78s/it]
70%|███████ | 30782/43939 [49:41:25<21:03:17, 5.76s/it]
70%|███████ | 30783/43939 [49:41:31<20:57:15, 5.73s/it]
70%|███████ | 30784/43939 [49:41:36<20:54:30, 5.72s/it]
70%|███████ | 30785/43939 [49:41:42<20:53:16, 5.72s/it]
70%|███████ | 30786/43939 [49:41:48<20:50:47, 5.71s/it]
70%|███████ | 30787/43939 [49:41:53<20:50:03, 5.70s/it]
70%|███████ | 30788/43939 [49:41:59<20:51:55, 5.71s/it]
70%|███████ | 30789/43939 [49:42:05<20:52:05, 5.71s/it]
70%|███████ | 30790/43939 [49:42:11<20:53:53, 5.72s/it]
70%|███████ | 30791/43939 [49:42:16<20:54:23, 5.72s/it]
70%|███████ | 30792/43939 [49:42:22<20:51:12, 5.71s/it]
70%|███████ | 30793/43939 [49:42:28<20:51:55, 5.71s/it]
70%|███████ | 30794/43939 [49:42:33<20:55:24, 5.73s/it]
70%|███████ | 30795/43939 [49:42:39<20:53:16, 5.72s/it]
70%|███████ | 30796/43939 [49:42:45<20:51:46, 5.71s/it]
70%|███████ | 30797/43939 [49:42:51<20:51:02, 5.71s/it]
70%|███████ | 30798/43939 [49:42:56<20:48:59, 5.70s/it]
70%|███████ | 30799/43939 [49:43:02<20:52:33, 5.72s/it]
70%|███████ | 30800/43939 [49:43:08<20:52:56, 5.72s/it]
{'loss': 0.7839, 'grad_norm': 1.3685954809188843, 'learning_rate': 1.2427831458021094e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 430.53, 'epoch': 2.8}
70%|███████ | 30800/43939 [49:43:08<20:52:56, 5.72s/it]
70%|███████ | 30801/43939 [49:43:13<20:52:49, 5.72s/it]
70%|███████ | 30802/43939 [49:43:19<21:07:36, 5.79s/it]
70%|███████ | 30803/43939 [49:43:25<21:01:54, 5.76s/it]
70%|███████ | 30804/43939 [49:43:31<20:58:25, 5.75s/it]
70%|███████ | 30805/43939 [49:43:36<20:55:04, 5.73s/it]
70%|███████ | 30806/43939 [49:43:42<20:54:40, 5.73s/it]
70%|███████ | 30807/43939 [49:43:48<20:53:48, 5.73s/it]
70%|███████ | 30808/43939 [49:43:54<20:55:35, 5.74s/it]
70%|███████ | 30809/43939 [49:43:59<20:52:26, 5.72s/it]
70%|███████ | 30810/43939 [49:44:05<20:52:52, 5.73s/it]
70%|███████ | 30811/43939 [49:44:11<20:52:34, 5.72s/it]
70%|███████ | 30812/43939 [49:44:17<20:49:43, 5.71s/it]
70%|███████ | 30813/43939 [49:44:22<20:51:17, 5.72s/it]
70%|███████ | 30814/43939 [49:44:28<20:51:08, 5.72s/it]
70%|███████ | 30815/43939 [49:44:34<20:49:05, 5.71s/it]
70%|███████ | 30816/43939 [49:44:39<20:49:11, 5.71s/it]
70%|███████ | 30817/43939 [49:44:45<20:46:29, 5.70s/it]
70%|███████ | 30818/43939 [49:44:51<20:46:32, 5.70s/it]
70%|███████ | 30819/43939 [49:44:56<20:44:47, 5.69s/it]
70%|███████ | 30820/43939 [49:45:02<20:47:22, 5.70s/it]
70%|███████ | 30821/43939 [49:45:08<20:45:37, 5.70s/it]
70%|███████ | 30822/43939 [49:45:14<20:47:17, 5.71s/it]
70%|███████ | 30823/43939 [49:45:20<21:04:13, 5.78s/it]
70%|███████ | 30824/43939 [49:45:25<20:59:58, 5.76s/it]
70%|███████ | 30825/43939 [49:45:31<20:55:57, 5.75s/it]
70%|███████ | 30826/43939 [49:45:37<20:54:10, 5.74s/it]
70%|███████ | 30827/43939 [49:45:42<20:54:06, 5.74s/it]
70%|███████ | 30828/43939 [49:45:48<20:53:27, 5.74s/it]
70%|███████ | 30829/43939 [49:45:54<20:50:59, 5.73s/it]
70%|███████ | 30830/43939 [49:46:00<20:49:03, 5.72s/it]
70%|███████ | 30831/43939 [49:46:05<20:50:32, 5.72s/it]
70%|███████ | 30832/43939 [49:46:11<20:49:14, 5.72s/it]
70%|███████ | 30833/43939 [49:46:17<20:45:29, 5.70s/it]
70%|███████ | 30834/43939 [49:46:22<20:44:05, 5.70s/it]
70%|███████ | 30835/43939 [49:46:28<20:42:36, 5.69s/it]
70%|███████ | 30836/43939 [49:46:34<20:41:11, 5.68s/it]
70%|███████ | 30837/43939 [49:46:39<20:42:25, 5.69s/it]
70%|███████ | 30838/43939 [49:46:45<20:43:32, 5.70s/it]
70%|███████ | 30839/43939 [49:46:51<20:43:21, 5.69s/it]
70%|███████ | 30840/43939 [49:46:57<20:46:42, 5.71s/it]
70%|███████ | 30841/43939 [49:47:02<20:46:06, 5.71s/it]
70%|███████ | 30842/43939 [49:47:08<20:44:46, 5.70s/it]
70%|███████ | 30843/43939 [49:47:14<20:44:31, 5.70s/it]
70%|███████ | 30844/43939 [49:47:20<21:00:29, 5.78s/it]
70%|███████ | 30845/43939 [49:47:25<20:54:53, 5.75s/it]
70%|███████ | 30846/43939 [49:47:31<20:52:20, 5.74s/it]
70%|███████ | 30847/43939 [49:47:37<20:50:46, 5.73s/it]
70%|███████ | 30848/43939 [49:47:42<20:48:02, 5.72s/it]
70%|███████ | 30849/43939 [49:47:48<20:49:22, 5.73s/it]
70%|███████ | 30850/43939 [49:47:54<20:51:25, 5.74s/it]
{'loss': 0.7794, 'grad_norm': 1.3834000825881958, 'learning_rate': 1.2342099080634754e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.33, 'epoch': 2.81}
70%|███████ | 30850/43939 [49:47:54<20:51:25, 5.74s/it]
70%|███████ | 30851/43939 [49:48:00<20:46:27, 5.71s/it]
70%|███████ | 30852/43939 [49:48:05<20:45:55, 5.71s/it]
70%|███████ | 30853/43939 [49:48:11<20:43:16, 5.70s/it]
70%|███████ | 30854/43939 [49:48:17<20:41:37, 5.69s/it]
70%|███████ | 30855/43939 [49:48:22<20:45:41, 5.71s/it]
70%|███████ | 30856/43939 [49:48:28<20:44:23, 5.71s/it]
70%|███████ | 30857/43939 [49:48:34<20:45:45, 5.71s/it]
70%|███████ | 30858/43939 [49:48:40<20:47:02, 5.72s/it]
70%|███████ | 30859/43939 [49:48:45<20:46:23, 5.72s/it]
70%|███████ | 30860/43939 [49:48:51<20:46:41, 5.72s/it]
70%|███████ | 30861/43939 [49:48:57<20:44:36, 5.71s/it]
70%|███████ | 30862/43939 [49:49:02<20:44:47, 5.71s/it]
70%|███████ | 30863/43939 [49:49:08<20:43:56, 5.71s/it]
70%|███████ | 30864/43939 [49:49:14<20:45:26, 5.72s/it]
70%|███████ | 30865/43939 [49:49:20<21:02:58, 5.80s/it]
70%|███████ | 30866/43939 [49:49:25<20:55:45, 5.76s/it]
70%|███████ | 30867/43939 [49:49:31<20:50:00, 5.74s/it]
70%|███████ | 30868/43939 [49:49:37<20:49:17, 5.73s/it]
70%|███████ | 30869/43939 [49:49:43<20:46:11, 5.72s/it]
70%|███████ | 30870/43939 [49:49:48<20:42:57, 5.71s/it]
70%|███████ | 30871/43939 [49:49:54<20:40:26, 5.70s/it]
70%|███████ | 30872/43939 [49:50:00<20:41:55, 5.70s/it]
70%|███████ | 30873/43939 [49:50:05<20:43:16, 5.71s/it]
70%|███████ | 30874/43939 [49:50:11<20:43:40, 5.71s/it]
70%|███████ | 30875/43939 [49:50:17<20:39:47, 5.69s/it]
70%|███████ | 30876/43939 [49:50:22<20:38:06, 5.69s/it]
70%|███████ | 30877/43939 [49:50:28<20:37:47, 5.69s/it]
70%|███████ | 30878/43939 [49:50:34<20:40:54, 5.70s/it]
70%|███████ | 30879/43939 [49:50:40<20:41:13, 5.70s/it]
70%|███████ | 30880/43939 [49:50:45<20:40:51, 5.70s/it]
70%|███████ | 30881/43939 [49:50:51<20:39:24, 5.69s/it]
70%|███████ | 30882/43939 [49:50:57<20:38:42, 5.69s/it]
70%|███████ | 30883/43939 [49:51:02<20:40:32, 5.70s/it]
70%|███████ | 30884/43939 [49:51:08<20:40:09, 5.70s/it]
70%|███████ | 30885/43939 [49:51:14<20:38:50, 5.69s/it]
70%|███████ | 30886/43939 [49:51:19<20:39:13, 5.70s/it]
70%|███████ | 30887/43939 [49:51:25<20:50:49, 5.75s/it]
70%|███████ | 30888/43939 [49:51:31<20:49:38, 5.75s/it]
70%|███████ | 30889/43939 [49:51:37<20:45:44, 5.73s/it]
70%|███████ | 30890/43939 [49:51:42<20:46:41, 5.73s/it]
70%|███████ | 30891/43939 [49:51:48<20:44:27, 5.72s/it]
70%|███████ | 30892/43939 [49:51:54<20:42:06, 5.71s/it]
70%|███████ | 30893/43939 [49:52:00<20:42:09, 5.71s/it]
70%|███████ | 30894/43939 [49:52:05<20:39:39, 5.70s/it]
70%|███████ | 30895/43939 [49:52:11<20:41:10, 5.71s/it]
70%|███████ | 30896/43939 [49:52:17<20:39:15, 5.70s/it]
70%|███████ | 30897/43939 [49:52:22<20:41:18, 5.71s/it]
70%|███████ | 30898/43939 [49:52:28<20:39:52, 5.70s/it]
70%|███████ | 30899/43939 [49:52:34<20:38:05, 5.70s/it]
70%|███████ | 30900/43939 [49:52:39<20:40:38, 5.71s/it]
{'loss': 0.7918, 'grad_norm': 1.3825865983963013, 'learning_rate': 1.2256566411381074e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.53, 'epoch': 2.81}
70%|███████ | 30900/43939 [49:52:39<20:40:38, 5.71s/it]
70%|███████ | 30901/43939 [49:52:45<20:38:27, 5.70s/it]
70%|███████ | 30902/43939 [49:52:51<20:38:26, 5.70s/it]
70%|███████ | 30903/43939 [49:52:57<20:38:42, 5.70s/it]
70%|███████ | 30904/43939 [49:53:02<20:38:14, 5.70s/it]
70%|███████ | 30905/43939 [49:53:08<20:33:52, 5.68s/it]
70%|███████ | 30906/43939 [49:53:14<20:36:16, 5.69s/it]
70%|███████ | 30907/43939 [49:53:19<20:35:56, 5.69s/it]
70%|███████ | 30908/43939 [49:53:25<20:55:24, 5.78s/it]
70%|███████ | 30909/43939 [49:53:31<20:49:17, 5.75s/it]
70%|███████ | 30910/43939 [49:53:37<20:46:10, 5.74s/it]
70%|███████ | 30911/43939 [49:53:42<20:45:58, 5.74s/it]
70%|███████ | 30912/43939 [49:53:48<20:44:12, 5.73s/it]
70%|███████ | 30913/43939 [49:53:54<20:41:46, 5.72s/it]
70%|███████ | 30914/43939 [49:53:59<20:39:20, 5.71s/it]
70%|███████ | 30915/43939 [49:54:05<20:39:44, 5.71s/it]
70%|███████ | 30916/43939 [49:54:11<20:38:28, 5.71s/it]
70%|███████ | 30917/43939 [49:54:17<20:37:29, 5.70s/it]
70%|███████ | 30918/43939 [49:54:22<20:36:58, 5.70s/it]
70%|███████ | 30919/43939 [49:54:28<20:38:43, 5.71s/it]
70%|███████ | 30920/43939 [49:54:34<20:35:41, 5.69s/it]
70%|███████ | 30921/43939 [49:54:39<20:35:10, 5.69s/it]
70%|███████ | 30922/43939 [49:54:45<20:36:38, 5.70s/it]
70%|███████ | 30923/43939 [49:54:51<20:39:51, 5.72s/it]
70%|███████ | 30924/43939 [49:54:57<20:40:37, 5.72s/it]
70%|███████ | 30925/43939 [49:55:02<20:40:20, 5.72s/it]
70%|███████ | 30926/43939 [49:55:08<20:36:50, 5.70s/it]
70%|███████ | 30927/43939 [49:55:14<20:36:04, 5.70s/it]
70%|███████ | 30928/43939 [49:55:19<20:34:52, 5.69s/it]
70%|███████ | 30929/43939 [49:55:25<20:51:55, 5.77s/it]
70%|███████ | 30930/43939 [49:55:31<20:45:19, 5.74s/it]
70%|███████ | 30931/43939 [49:55:37<20:42:04, 5.73s/it]
70%|███████ | 30932/43939 [49:55:42<20:38:36, 5.71s/it]
70%|███████ | 30933/43939 [49:55:48<20:35:53, 5.70s/it]
70%|███████ | 30934/43939 [49:55:54<20:36:26, 5.70s/it]
70%|███████ | 30935/43939 [49:55:59<20:40:50, 5.73s/it]
70%|███████ | 30936/43939 [49:56:05<20:39:24, 5.72s/it]
70%|███████ | 30937/43939 [49:56:11<20:39:23, 5.72s/it]
70%|███████ | 30938/43939 [49:56:17<20:38:44, 5.72s/it]
70%|███████ | 30939/43939 [49:56:22<20:38:33, 5.72s/it]
70%|███████ | 30940/43939 [49:56:28<20:36:00, 5.71s/it]
70%|███████ | 30941/43939 [49:56:34<20:35:39, 5.70s/it]
70%|███████ | 30942/43939 [49:56:39<20:32:46, 5.69s/it]
70%|███████ | 30943/43939 [49:56:45<20:33:17, 5.69s/it]
70%|███████ | 30944/43939 [49:56:51<20:32:25, 5.69s/it]
70%|███████ | 30945/43939 [49:56:56<20:33:37, 5.70s/it]
70%|███████ | 30946/43939 [49:57:02<20:37:00, 5.71s/it]
70%|███████ | 30947/43939 [49:57:08<20:35:49, 5.71s/it]
70%|███████ | 30948/43939 [49:57:14<20:33:20, 5.70s/it]
70%|███████ | 30949/43939 [49:57:19<20:33:34, 5.70s/it]
70%|███████ | 30950/43939 [49:57:25<20:46:01, 5.76s/it]
{'loss': 0.7689, 'grad_norm': 1.6372039318084717, 'learning_rate': 1.2171234799738914e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 454.37, 'epoch': 2.82}
70%|███████ | 30950/43939 [49:57:25<20:46:01, 5.76s/it]
70%|███████ | 30951/43939 [49:57:31<20:41:44, 5.74s/it]
70%|███████ | 30952/43939 [49:57:37<20:36:33, 5.71s/it]
70%|███████ | 30953/43939 [49:57:42<20:32:56, 5.70s/it]
70%|███████ | 30954/43939 [49:57:48<20:36:12, 5.71s/it]
70%|███████ | 30955/43939 [49:57:54<20:35:20, 5.71s/it]
70%|███████ | 30956/43939 [49:57:59<20:34:21, 5.70s/it]
70%|███████ | 30957/43939 [49:58:05<20:32:10, 5.69s/it]
70%|███████ | 30958/43939 [49:58:11<20:33:54, 5.70s/it]
70%|███████ | 30959/43939 [49:58:16<20:32:31, 5.70s/it]
70%|███████ | 30960/43939 [49:58:22<20:30:04, 5.69s/it]
70%|███████ | 30961/43939 [49:58:28<20:30:23, 5.69s/it]
70%|███████ | 30962/43939 [49:58:33<20:32:24, 5.70s/it]
70%|███████ | 30963/43939 [49:58:39<20:33:15, 5.70s/it]
70%|███████ | 30964/43939 [49:58:45<20:33:25, 5.70s/it]
70%|███████ | 30965/43939 [49:58:51<20:34:17, 5.71s/it]
70%|███████ | 30966/43939 [49:58:56<20:35:05, 5.71s/it]
70%|███████ | 30967/43939 [49:59:02<20:35:25, 5.71s/it]
70%|███████ | 30968/43939 [49:59:08<20:36:14, 5.72s/it]
70%|███████ | 30969/43939 [49:59:13<20:34:00, 5.71s/it]
70%|███████ | 30970/43939 [49:59:19<20:33:24, 5.71s/it]
70%|███████ | 30971/43939 [49:59:25<20:33:19, 5.71s/it]
70%|███████ | 30972/43939 [49:59:31<20:45:47, 5.76s/it]
70%|███████ | 30973/43939 [49:59:36<20:43:21, 5.75s/it]
70%|███████ | 30974/43939 [49:59:42<20:45:33, 5.76s/it]
70%|███████ | 30975/43939 [49:59:48<20:39:45, 5.74s/it]
70%|███████ | 30976/43939 [49:59:54<20:36:34, 5.72s/it]
71%|███████ | 30977/43939 [49:59:59<20:35:59, 5.72s/it]
71%|███████ | 30978/43939 [50:00:05<20:37:22, 5.73s/it]
71%|███████ | 30979/43939 [50:00:11<20:33:35, 5.71s/it]
71%|███████ | 30980/43939 [50:00:17<20:35:18, 5.72s/it]
71%|███████ | 30981/43939 [50:00:22<20:32:27, 5.71s/it]
71%|███████ | 30982/43939 [50:00:28<20:30:43, 5.70s/it]
71%|███████ | 30983/43939 [50:00:34<20:29:38, 5.69s/it]
71%|███████ | 30984/43939 [50:00:39<20:31:57, 5.71s/it]
71%|███████ | 30985/43939 [50:00:45<20:30:25, 5.70s/it]
71%|███████ | 30986/43939 [50:00:51<20:29:01, 5.69s/it]
71%|███████ | 30987/43939 [50:00:56<20:29:52, 5.70s/it]
71%|███████ | 30988/43939 [50:01:02<20:30:12, 5.70s/it]
71%|███████ | 30989/43939 [50:01:08<20:29:33, 5.70s/it]
71%|███████ | 30990/43939 [50:01:13<20:28:04, 5.69s/it]
71%|███████ | 30991/43939 [50:01:19<20:27:33, 5.69s/it]
71%|███████ | 30992/43939 [50:01:25<20:25:36, 5.68s/it]
71%|███████ | 30993/43939 [50:01:30<20:26:39, 5.69s/it]
71%|███████ | 30994/43939 [50:01:36<20:41:42, 5.76s/it]
71%|███████ | 30995/43939 [50:01:42<20:38:21, 5.74s/it]
71%|███████ | 30996/43939 [50:01:48<20:34:43, 5.72s/it]
71%|███████ | 30997/43939 [50:01:53<20:34:03, 5.72s/it]
71%|███████ | 30998/43939 [50:01:59<20:31:12, 5.71s/it]
71%|███████ | 30999/43939 [50:02:05<20:31:06, 5.71s/it]
71%|███████ | 31000/43939 [50:02:11<20:29:49, 5.70s/it]
{'loss': 0.7569, 'grad_norm': 1.580826997756958, 'learning_rate': 1.2086105592014974e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.76, 'epoch': 2.82}
71%|███████ | 31000/43939 [50:02:11<20:29:49, 5.70s/it]
71%|███████ | 31001/43939 [50:02:16<20:30:17, 5.71s/it]
71%|███████ | 31002/43939 [50:02:22<20:30:00, 5.70s/it]
71%|███████ | 31003/43939 [50:02:28<20:31:54, 5.71s/it]
71%|███████ | 31004/43939 [50:02:34<20:36:16, 5.73s/it]
71%|███████ | 31005/43939 [50:02:39<20:32:19, 5.72s/it]
71%|███████ | 31006/43939 [50:02:45<20:30:15, 5.71s/it]
71%|███████ | 31007/43939 [50:02:51<20:32:16, 5.72s/it]
71%|███████ | 31008/43939 [50:02:56<20:30:47, 5.71s/it]
71%|███████ | 31009/43939 [50:03:02<20:29:07, 5.70s/it]
71%|███████ | 31010/43939 [50:03:08<20:27:21, 5.70s/it]
71%|███████ | 31011/43939 [50:03:13<20:25:47, 5.69s/it]
71%|███████ | 31012/43939 [50:03:19<20:30:46, 5.71s/it]
71%|███████ | 31013/43939 [50:03:25<20:30:31, 5.71s/it]
71%|███████ | 31014/43939 [50:03:31<20:28:49, 5.70s/it]
71%|███████ | 31015/43939 [50:03:36<20:43:34, 5.77s/it]
71%|███████ | 31016/43939 [50:03:42<20:43:26, 5.77s/it]
71%|███████ | 31017/43939 [50:03:48<20:36:36, 5.74s/it]
71%|███████ | 31018/43939 [50:03:54<20:34:45, 5.73s/it]
71%|███████ | 31019/43939 [50:03:59<20:30:59, 5.72s/it]
71%|███████ | 31020/43939 [50:04:05<20:31:03, 5.72s/it]
71%|███████ | 31021/43939 [50:04:11<20:28:48, 5.71s/it]
71%|███████ | 31022/43939 [50:04:16<20:26:49, 5.70s/it]
71%|███████ | 31023/43939 [50:04:22<20:26:33, 5.70s/it]
71%|███████ | 31024/43939 [50:04:28<20:26:59, 5.70s/it]
71%|███████ | 31025/43939 [50:04:33<20:27:50, 5.70s/it]
71%|███████ | 31026/43939 [50:04:39<20:28:18, 5.71s/it]
71%|███████ | 31027/43939 [50:04:45<20:28:33, 5.71s/it]
71%|███████ | 31028/43939 [50:04:51<20:27:46, 5.71s/it]
71%|███████ | 31029/43939 [50:04:56<20:25:33, 5.70s/it]
71%|███████ | 31030/43939 [50:05:02<20:25:48, 5.70s/it]
71%|███████ | 31031/43939 [50:05:08<20:27:52, 5.71s/it]
71%|███████ | 31032/43939 [50:05:13<20:27:03, 5.70s/it]
71%|███████ | 31033/43939 [50:05:19<20:25:34, 5.70s/it]
71%|███████ | 31034/43939 [50:05:25<20:26:25, 5.70s/it]
71%|███████ | 31035/43939 [50:05:31<20:26:58, 5.71s/it]
71%|███████ | 31036/43939 [50:05:36<20:24:41, 5.69s/it]
71%|███████ | 31037/43939 [50:05:42<20:39:04, 5.76s/it]
71%|███████ | 31038/43939 [50:05:48<20:36:09, 5.75s/it]
71%|███████ | 31039/43939 [50:05:54<20:34:53, 5.74s/it]
71%|███████ | 31040/43939 [50:05:59<20:31:42, 5.73s/it]
71%|███████ | 31041/43939 [50:06:05<20:29:41, 5.72s/it]
71%|███████ | 31042/43939 [50:06:11<20:26:21, 5.71s/it]
71%|███████ | 31043/43939 [50:06:16<20:27:24, 5.71s/it]
71%|███████ | 31044/43939 [50:06:22<20:26:00, 5.70s/it]
71%|███████ | 31045/43939 [50:06:28<20:26:06, 5.71s/it]
71%|███████ | 31046/43939 [50:06:33<20:27:37, 5.71s/it]
71%|███████ | 31047/43939 [50:06:39<20:29:49, 5.72s/it]
71%|███████ | 31048/43939 [50:06:45<20:33:29, 5.74s/it]
71%|███████ | 31049/43939 [50:06:51<20:30:34, 5.73s/it]
71%|███████ | 31050/43939 [50:06:56<20:30:30, 5.73s/it]
{'loss': 0.758, 'grad_norm': 1.1925673484802246, 'learning_rate': 1.2001180131322553e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.36, 'epoch': 2.83}
71%|███████ | 31050/43939 [50:06:56<20:30:30, 5.73s/it]
71%|███████ | 31051/43939 [50:07:02<20:29:02, 5.72s/it]
71%|███████ | 31052/43939 [50:07:08<20:25:26, 5.71s/it]
71%|███████ | 31053/43939 [50:07:13<20:22:43, 5.69s/it]
71%|███████ | 31054/43939 [50:07:19<20:23:18, 5.70s/it]
71%|███████ | 31055/43939 [50:07:25<20:25:27, 5.71s/it]
71%|███████ | 31056/43939 [50:07:31<20:27:24, 5.72s/it]
71%|███████ | 31057/43939 [50:07:36<20:25:23, 5.71s/it]
71%|███████ | 31058/43939 [50:07:42<20:42:45, 5.79s/it]
71%|███████ | 31059/43939 [50:07:48<20:39:29, 5.77s/it]
71%|███████ | 31060/43939 [50:07:54<20:32:56, 5.74s/it]
71%|███████ | 31061/43939 [50:07:59<20:34:44, 5.75s/it]
71%|███████ | 31062/43939 [50:08:05<20:31:52, 5.74s/it]
71%|███████ | 31063/43939 [50:08:11<20:32:44, 5.74s/it]
71%|███████ | 31064/43939 [50:08:17<20:27:25, 5.72s/it]
71%|███████ | 31065/43939 [50:08:22<20:30:27, 5.73s/it]
71%|███████ | 31066/43939 [50:08:28<20:26:22, 5.72s/it]
71%|███████ | 31067/43939 [50:08:34<20:32:36, 5.75s/it]
71%|███████ | 31068/43939 [50:08:40<20:33:59, 5.75s/it]
71%|███████ | 31069/43939 [50:08:45<20:34:15, 5.75s/it]
71%|███████ | 31070/43939 [50:08:51<20:30:41, 5.74s/it]
71%|███████ | 31071/43939 [50:08:57<20:28:32, 5.73s/it]
71%|███████ | 31072/43939 [50:09:02<20:24:29, 5.71s/it]
71%|███████ | 31073/43939 [50:09:08<20:25:28, 5.71s/it]
71%|███████ | 31074/43939 [50:09:14<20:28:08, 5.73s/it]
71%|███████ | 31075/43939 [50:09:20<20:24:51, 5.71s/it]
71%|███████ | 31076/43939 [50:09:25<20:27:41, 5.73s/it]
71%|███████ | 31077/43939 [50:09:31<20:26:30, 5.72s/it]
71%|███████ | 31078/43939 [50:09:37<20:25:03, 5.72s/it]
71%|███████ | 31079/43939 [50:09:42<20:22:44, 5.70s/it]
71%|███████ | 31080/43939 [50:09:48<20:37:00, 5.77s/it]
71%|███████ | 31081/43939 [50:09:54<20:36:48, 5.77s/it]
71%|███████ | 31082/43939 [50:10:00<20:30:11, 5.74s/it]
71%|███████ | 31083/43939 [50:10:06<20:27:06, 5.73s/it]
71%|███████ | 31084/43939 [50:10:11<20:29:58, 5.74s/it]
71%|███████ | 31085/43939 [50:10:17<20:27:41, 5.73s/it]
71%|███████ | 31086/43939 [50:10:23<20:27:38, 5.73s/it]
71%|███████ | 31087/43939 [50:10:28<20:23:54, 5.71s/it]
71%|███████ | 31088/43939 [50:10:34<20:24:14, 5.72s/it]
71%|███████ | 31089/43939 [50:10:40<20:21:23, 5.70s/it]
71%|███████ | 31090/43939 [50:10:46<20:25:20, 5.72s/it]
71%|███████ | 31091/43939 [50:10:51<20:23:30, 5.71s/it]
71%|███████ | 31092/43939 [50:10:57<20:23:47, 5.72s/it]
71%|███████ | 31093/43939 [50:11:03<20:20:05, 5.70s/it]
71%|███████ | 31094/43939 [50:11:08<20:20:10, 5.70s/it]
71%|███████ | 31095/43939 [50:11:14<20:23:47, 5.72s/it]
71%|███████ | 31096/43939 [50:11:20<20:24:07, 5.72s/it]
71%|███████ | 31097/43939 [50:11:26<20:24:20, 5.72s/it]
71%|███████ | 31098/43939 [50:11:31<20:21:52, 5.71s/it]
71%|███████ | 31099/43939 [50:11:37<20:22:04, 5.71s/it]
71%|███████ | 31100/43939 [50:11:43<20:36:10, 5.78s/it]
{'loss': 0.7612, 'grad_norm': 1.757616639137268, 'learning_rate': 1.1916459757560367e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.46, 'epoch': 2.83}
71%|███████ | 31100/43939 [50:11:43<20:36:10, 5.78s/it]
71%|███████ | 31101/43939 [50:11:49<20:31:53, 5.76s/it]
71%|███████ | 31102/43939 [50:11:54<20:28:58, 5.74s/it]
71%|███████ | 31103/43939 [50:12:00<20:25:29, 5.73s/it]
71%|███████ | 31104/43939 [50:12:06<20:23:48, 5.72s/it]
71%|███████ | 31105/43939 [50:12:11<20:21:47, 5.71s/it]
71%|███████ | 31106/43939 [50:12:17<20:19:15, 5.70s/it]
71%|███████ | 31107/43939 [50:12:23<20:17:02, 5.69s/it]
71%|███████ | 31108/43939 [50:12:28<20:19:54, 5.70s/it]
71%|███████ | 31109/43939 [50:12:34<20:17:38, 5.69s/it]
71%|███████ | 31110/43939 [50:12:40<20:17:39, 5.69s/it]
71%|███████ | 31111/43939 [50:12:46<20:17:53, 5.70s/it]
71%|███████ | 31112/43939 [50:12:51<20:16:46, 5.69s/it]
71%|███████ | 31113/43939 [50:12:57<20:17:24, 5.70s/it]
71%|███████ | 31114/43939 [50:13:03<20:18:52, 5.70s/it]
71%|███████ | 31115/43939 [50:13:08<20:22:25, 5.72s/it]
71%|███████ | 31116/43939 [50:13:14<20:20:11, 5.71s/it]
71%|███████ | 31117/43939 [50:13:20<20:19:31, 5.71s/it]
71%|███████ | 31118/43939 [50:13:25<20:19:03, 5.70s/it]
71%|███████ | 31119/43939 [50:13:31<20:20:51, 5.71s/it]
71%|███████ | 31120/43939 [50:13:37<20:19:50, 5.71s/it]
71%|███████ | 31121/43939 [50:13:43<20:16:44, 5.70s/it]
71%|███████ | 31122/43939 [50:13:48<20:29:30, 5.76s/it]
71%|███████ | 31123/43939 [50:13:54<20:25:11, 5.74s/it]
71%|███████ | 31124/43939 [50:14:00<20:19:30, 5.71s/it]
71%|███████ | 31125/43939 [50:14:06<20:17:00, 5.70s/it]
71%|███████ | 31126/43939 [50:14:11<20:18:16, 5.70s/it]
71%|███████ | 31127/43939 [50:14:17<20:18:12, 5.70s/it]
71%|███████ | 31128/43939 [50:14:23<20:20:06, 5.71s/it]
71%|███████ | 31129/43939 [50:14:28<20:20:24, 5.72s/it]
71%|███████ | 31130/43939 [50:14:34<20:18:06, 5.71s/it]
71%|███████ | 31131/43939 [50:14:40<20:17:10, 5.70s/it]
71%|███████ | 31132/43939 [50:14:45<20:17:44, 5.71s/it]
71%|███████ | 31133/43939 [50:14:51<20:20:58, 5.72s/it]
71%|███████ | 31134/43939 [50:14:57<20:19:26, 5.71s/it]
71%|███████ | 31135/43939 [50:15:03<20:20:12, 5.72s/it]
71%|███████ | 31136/43939 [50:15:08<20:19:29, 5.72s/it]
71%|███████ | 31137/43939 [50:15:14<20:18:17, 5.71s/it]
71%|███████ | 31138/43939 [50:15:20<20:17:57, 5.71s/it]
71%|███████ | 31139/43939 [50:15:25<20:17:25, 5.71s/it]
71%|███████ | 31140/43939 [50:15:31<20:17:32, 5.71s/it]
71%|███████ | 31141/43939 [50:15:37<20:15:32, 5.70s/it]
71%|███████ | 31142/43939 [50:15:43<20:13:30, 5.69s/it]
71%|███████ | 31143/43939 [50:15:48<20:11:54, 5.68s/it]
71%|███████ | 31144/43939 [50:15:54<20:28:21, 5.76s/it]
71%|███████ | 31145/43939 [50:16:00<20:26:36, 5.75s/it]
71%|███████ | 31146/43939 [50:16:06<20:24:37, 5.74s/it]
71%|███████ | 31147/43939 [50:16:11<20:20:43, 5.73s/it]
71%|███████ | 31148/43939 [50:16:17<20:17:42, 5.71s/it]
71%|███████ | 31149/43939 [50:16:23<20:16:52, 5.71s/it]
71%|███████ | 31150/43939 [50:16:28<20:17:33, 5.71s/it]
{'loss': 0.7944, 'grad_norm': 1.388952612876892, 'learning_rate': 1.1831945807391394e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.87, 'epoch': 2.84}
71%|███████ | 31150/43939 [50:16:28<20:17:33, 5.71s/it]
71%|███████ | 31151/43939 [50:16:34<20:18:01, 5.71s/it]
71%|███████ | 31152/43939 [50:16:40<20:15:32, 5.70s/it]
71%|███████ | 31153/43939 [50:16:45<20:15:18, 5.70s/it]
71%|███████ | 31154/43939 [50:16:51<20:17:19, 5.71s/it]
71%|███████ | 31155/43939 [50:16:57<20:19:51, 5.73s/it]
71%|███████ | 31156/43939 [50:17:03<20:20:13, 5.73s/it]
71%|███████ | 31157/43939 [50:17:08<20:17:38, 5.72s/it]
71%|███████ | 31158/43939 [50:17:14<20:16:30, 5.71s/it]
71%|███████ | 31159/43939 [50:17:20<20:16:15, 5.71s/it]
71%|███████ | 31160/43939 [50:17:25<20:13:19, 5.70s/it]
71%|███████ | 31161/43939 [50:17:31<20:14:02, 5.70s/it]
71%|███████ | 31162/43939 [50:17:37<20:14:17, 5.70s/it]
71%|███████ | 31163/43939 [50:17:43<20:15:22, 5.71s/it]
71%|███████ | 31164/43939 [50:17:48<20:15:07, 5.71s/it]
71%|███████ | 31165/43939 [50:17:54<20:28:41, 5.77s/it]
71%|███████ | 31166/43939 [50:18:00<20:25:14, 5.76s/it]
71%|███████ | 31167/43939 [50:18:06<20:22:53, 5.74s/it]
71%|███████ | 31168/43939 [50:18:11<20:20:39, 5.73s/it]
71%|███████ | 31169/43939 [50:18:17<20:16:50, 5.72s/it]
71%|███████ | 31170/43939 [50:18:23<20:12:43, 5.70s/it]
71%|███████ | 31171/43939 [50:18:28<20:11:43, 5.69s/it]
71%|███████ | 31172/43939 [50:18:34<20:15:05, 5.71s/it]
71%|███████ | 31173/43939 [50:18:40<20:15:40, 5.71s/it]
71%|███████ | 31174/43939 [50:18:46<20:15:18, 5.71s/it]
71%|███████ | 31175/43939 [50:18:51<20:14:26, 5.71s/it]
71%|███████ | 31176/43939 [50:18:57<20:12:39, 5.70s/it]
71%|███████ | 31177/43939 [50:19:03<20:10:29, 5.69s/it]
71%|███████ | 31178/43939 [50:19:08<20:15:14, 5.71s/it]
71%|███████ | 31179/43939 [50:19:14<20:19:04, 5.73s/it]
71%|███████ | 31180/43939 [50:19:20<20:16:49, 5.72s/it]
71%|███████ | 31181/43939 [50:19:26<20:17:56, 5.73s/it]
71%|███████ | 31182/43939 [50:19:31<20:15:11, 5.72s/it]
71%|███████ | 31183/43939 [50:19:37<20:14:14, 5.71s/it]
71%|███████ | 31184/43939 [50:19:43<20:13:44, 5.71s/it]
71%|███████ | 31185/43939 [50:19:48<20:16:06, 5.72s/it]
71%|███████ | 31186/43939 [50:19:54<20:14:36, 5.71s/it]
71%|███████ | 31187/43939 [50:20:00<20:27:02, 5.77s/it]
71%|███████ | 31188/43939 [50:20:06<20:21:11, 5.75s/it]
71%|███████ | 31189/43939 [50:20:11<20:18:34, 5.73s/it]
71%|███████ | 31190/43939 [50:20:17<20:15:45, 5.72s/it]
71%|███████ | 31191/43939 [50:20:23<20:16:09, 5.72s/it]
71%|███████ | 31192/43939 [50:20:29<20:17:43, 5.73s/it]
71%|███████ | 31193/43939 [50:20:34<20:14:57, 5.72s/it]
71%|███████ | 31194/43939 [50:20:40<20:12:10, 5.71s/it]
71%|███████ | 31195/43939 [50:20:46<20:13:05, 5.71s/it]
71%|███████ | 31196/43939 [50:20:51<20:14:46, 5.72s/it]
71%|███████ | 31197/43939 [50:20:57<20:18:17, 5.74s/it]
71%|███████ | 31198/43939 [50:21:03<20:15:36, 5.72s/it]
71%|███████ | 31199/43939 [50:21:09<20:14:03, 5.72s/it]
71%|███████ | 31200/43939 [50:21:14<20:13:08, 5.71s/it]
{'loss': 0.754, 'grad_norm': 1.3602991104125977, 'learning_rate': 1.1747639614221823e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.19, 'epoch': 2.84}
71%|███████ | 31200/43939 [50:21:14<20:13:08, 5.71s/it]
71%|███████ | 31201/43939 [50:21:20<20:13:32, 5.72s/it]
71%|███████ | 31202/43939 [50:21:26<20:13:38, 5.72s/it]
71%|███████ | 31203/43939 [50:21:31<20:13:21, 5.72s/it]
71%|███████ | 31204/43939 [50:21:37<20:13:08, 5.72s/it]
71%|███████ | 31205/43939 [50:21:43<20:11:58, 5.71s/it]
71%|███████ | 31206/43939 [50:21:49<20:12:06, 5.71s/it]
71%|███████ | 31207/43939 [50:21:54<20:10:40, 5.71s/it]
71%|███████ | 31208/43939 [50:22:00<20:25:03, 5.77s/it]
71%|███████ | 31209/43939 [50:22:06<20:20:03, 5.75s/it]
71%|███████ | 31210/43939 [50:22:12<20:16:49, 5.74s/it]
71%|███████ | 31211/43939 [50:22:17<20:14:23, 5.72s/it]
71%|███████ | 31212/43939 [50:22:23<20:11:44, 5.71s/it]
71%|███████ | 31213/43939 [50:22:29<20:12:32, 5.72s/it]
71%|███████ | 31214/43939 [50:22:34<20:11:32, 5.71s/it]
71%|███████ | 31215/43939 [50:22:40<20:09:21, 5.70s/it]
71%|███████ | 31216/43939 [50:22:46<20:12:31, 5.72s/it]
71%|███████ | 31217/43939 [50:22:52<20:11:09, 5.71s/it]
71%|███████ | 31218/43939 [50:22:57<20:12:20, 5.72s/it]
71%|███████ | 31219/43939 [50:23:03<20:10:14, 5.71s/it]
71%|███████ | 31220/43939 [50:23:09<20:12:20, 5.72s/it]
71%|███████ | 31221/43939 [50:23:14<20:13:46, 5.73s/it]
71%|███████ | 31222/43939 [50:23:20<20:14:38, 5.73s/it]
71%|███████ | 31223/43939 [50:23:26<20:13:17, 5.72s/it]
71%|███████ | 31224/43939 [50:23:32<20:09:43, 5.71s/it]
71%|███████ | 31225/43939 [50:23:37<20:08:14, 5.70s/it]
71%|███████ | 31226/43939 [50:23:43<20:08:51, 5.71s/it]
71%|███████ | 31227/43939 [50:23:49<20:08:46, 5.71s/it]
71%|███████ | 31228/43939 [50:23:54<20:11:59, 5.72s/it]
71%|███████ | 31229/43939 [50:24:00<20:11:39, 5.72s/it]
71%|███████ | 31230/43939 [50:24:06<20:23:22, 5.78s/it]
71%|███████ | 31231/43939 [50:24:12<20:18:57, 5.76s/it]
71%|███████ | 31232/43939 [50:24:18<20:17:22, 5.75s/it]
71%|███████ | 31233/43939 [50:24:23<20:11:37, 5.72s/it]
71%|███████ | 31234/43939 [50:24:29<20:10:18, 5.72s/it]
71%|███████ | 31235/43939 [50:24:35<20:10:08, 5.72s/it]
71%|███████ | 31236/43939 [50:24:40<20:08:35, 5.71s/it]
71%|███████ | 31237/43939 [50:24:46<20:08:51, 5.71s/it]
71%|███████ | 31238/43939 [50:24:52<20:06:14, 5.70s/it]
71%|███████ | 31239/43939 [50:24:57<20:04:37, 5.69s/it]
71%|███████ | 31240/43939 [50:25:03<20:04:17, 5.69s/it]
71%|███████ | 31241/43939 [50:25:09<20:07:59, 5.71s/it]
71%|███████ | 31242/43939 [50:25:15<20:09:04, 5.71s/it]
71%|███████ | 31243/43939 [50:25:20<20:08:42, 5.71s/it]
71%|███████ | 31244/43939 [50:25:26<20:07:56, 5.71s/it]
71%|███████ | 31245/43939 [50:25:32<20:13:10, 5.73s/it]
71%|███████ | 31246/43939 [50:25:37<20:11:03, 5.72s/it]
71%|███████ | 31247/43939 [50:25:43<20:07:36, 5.71s/it]
71%|███████ | 31248/43939 [50:25:49<20:06:21, 5.70s/it]
71%|███████ | 31249/43939 [50:25:54<20:06:17, 5.70s/it]
71%|███████ | 31250/43939 [50:26:00<20:05:44, 5.70s/it]
{'loss': 0.7804, 'grad_norm': 1.4546005725860596, 'learning_rate': 1.1663542508179955e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.82, 'epoch': 2.84}
71%|███████ | 31250/43939 [50:26:00<20:05:44, 5.70s/it]
71%|███████ | 31251/43939 [50:26:06<20:20:11, 5.77s/it]
71%|███████ | 31252/43939 [50:26:12<20:19:40, 5.77s/it]
71%|███████ | 31253/43939 [50:26:18<20:15:31, 5.75s/it]
71%|███████ | 31254/43939 [50:26:23<20:11:57, 5.73s/it]
71%|███████ | 31255/43939 [50:26:29<20:10:46, 5.73s/it]
71%|███████ | 31256/43939 [50:26:35<20:11:11, 5.73s/it]
71%|███████ | 31257/43939 [50:26:40<20:09:11, 5.72s/it]
71%|███████ | 31258/43939 [50:26:46<20:08:26, 5.72s/it]
71%|███████ | 31259/43939 [50:26:52<20:05:01, 5.70s/it]
71%|███████ | 31260/43939 [50:26:58<20:09:07, 5.72s/it]
71%|███████ | 31261/43939 [50:27:03<20:07:58, 5.72s/it]
71%|███████ | 31262/43939 [50:27:09<20:08:02, 5.72s/it]
71%|███████ | 31263/43939 [50:27:15<20:07:02, 5.71s/it]
71%|███████ | 31264/43939 [50:27:20<20:07:37, 5.72s/it]
71%|███████ | 31265/43939 [50:27:26<20:06:52, 5.71s/it]
71%|███████ | 31266/43939 [50:27:32<20:07:52, 5.72s/it]
71%|███████ | 31267/43939 [50:27:38<20:10:50, 5.73s/it]
71%|███████ | 31268/43939 [50:27:43<20:07:17, 5.72s/it]
71%|███████ | 31269/43939 [50:27:49<20:08:20, 5.72s/it]
71%|███████ | 31270/43939 [50:27:55<20:08:09, 5.72s/it]
71%|███████ | 31271/43939 [50:28:00<20:06:45, 5.72s/it]
71%|███████ | 31272/43939 [50:28:06<20:06:14, 5.71s/it]
71%|███████ | 31273/43939 [50:28:12<20:18:34, 5.77s/it]
71%|███████ | 31274/43939 [50:28:18<20:12:48, 5.75s/it]
71%|███████ | 31275/43939 [50:28:24<20:13:57, 5.75s/it]
71%|███████ | 31276/43939 [50:28:29<20:09:12, 5.73s/it]
71%|███████ | 31277/43939 [50:28:35<20:06:23, 5.72s/it]
71%|███████ | 31278/43939 [50:28:41<20:05:29, 5.71s/it]
71%|███████ | 31279/43939 [50:28:46<20:07:57, 5.72s/it]
71%|███████ | 31280/43939 [50:28:52<20:05:38, 5.71s/it]
71%|███████ | 31281/43939 [50:28:58<20:06:53, 5.72s/it]
71%|███████ | 31282/43939 [50:29:03<20:04:33, 5.71s/it]
71%|███████ | 31283/43939 [50:29:09<20:03:07, 5.70s/it]
71%|███████ | 31284/43939 [50:29:15<20:05:02, 5.71s/it]
71%|███████ | 31285/43939 [50:29:21<20:04:27, 5.71s/it]
71%|███████ | 31286/43939 [50:29:26<20:05:07, 5.71s/it]
71%|███████ | 31287/43939 [50:29:32<20:03:56, 5.71s/it]
71%|███████ | 31288/43939 [50:29:38<20:03:10, 5.71s/it]
71%|███████ | 31289/43939 [50:29:43<20:01:40, 5.70s/it]
71%|███████ | 31290/43939 [50:29:49<20:03:17, 5.71s/it]
71%|███████ | 31291/43939 [50:29:55<19:59:15, 5.69s/it]
71%|███████ | 31292/43939 [50:30:00<19:58:05, 5.68s/it]
71%|███████ | 31293/43939 [50:30:06<20:01:08, 5.70s/it]
71%|███████ | 31294/43939 [50:30:12<20:13:51, 5.76s/it]
71%|███████ | 31295/43939 [50:30:18<20:09:23, 5.74s/it]
71%|███████ | 31296/43939 [50:30:23<20:06:39, 5.73s/it]
71%|███████ | 31297/43939 [50:30:29<20:04:42, 5.72s/it]
71%|███████ | 31298/43939 [50:30:35<20:04:30, 5.72s/it]
71%|███████ | 31299/43939 [50:30:41<20:03:14, 5.71s/it]
71%|███████ | 31300/43939 [50:30:46<20:02:41, 5.71s/it]
{'loss': 0.7746, 'grad_norm': 1.4353296756744385, 'learning_rate': 1.157965581609527e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.3, 'epoch': 2.85}
71%|███████ | 31300/43939 [50:30:46<20:02:41, 5.71s/it]
71%|███████ | 31301/43939 [50:30:52<20:02:27, 5.71s/it]
71%|███████ | 31302/43939 [50:30:58<20:02:11, 5.71s/it]
71%|███████ | 31303/43939 [50:31:03<20:01:20, 5.70s/it]
71%|███████ | 31304/43939 [50:31:09<20:01:14, 5.70s/it]
71%|███████ | 31305/43939 [50:31:15<20:00:38, 5.70s/it]
71%|███████ | 31306/43939 [50:31:21<20:03:30, 5.72s/it]
71%|███████▏ | 31307/43939 [50:31:26<20:01:39, 5.71s/it]
71%|███████▏ | 31308/43939 [50:31:32<20:00:07, 5.70s/it]
71%|███████▏ | 31309/43939 [50:31:38<20:00:27, 5.70s/it]
71%|███████▏ | 31310/43939 [50:31:43<20:00:17, 5.70s/it]
71%|███████▏ | 31311/43939 [50:31:49<20:02:59, 5.72s/it]
71%|███████▏ | 31312/43939 [50:31:55<20:02:37, 5.71s/it]
71%|███████▏ | 31313/43939 [50:32:01<20:03:24, 5.72s/it]
71%|███████▏ | 31314/43939 [50:32:06<20:02:07, 5.71s/it]
71%|███████▏ | 31315/43939 [50:32:12<20:01:41, 5.71s/it]
71%|███████▏ | 31316/43939 [50:32:18<20:14:59, 5.78s/it]
71%|███████▏ | 31317/43939 [50:32:24<20:13:54, 5.77s/it]
71%|███████▏ | 31318/43939 [50:32:29<20:09:36, 5.75s/it]
71%|███████▏ | 31319/43939 [50:32:35<20:06:55, 5.74s/it]
71%|███████▏ | 31320/43939 [50:32:41<20:04:04, 5.73s/it]
71%|███████▏ | 31321/43939 [50:32:46<20:00:45, 5.71s/it]
71%|███████▏ | 31322/43939 [50:32:52<19:58:07, 5.70s/it]
71%|███████▏ | 31323/43939 [50:32:58<20:00:54, 5.71s/it]
71%|███████▏ | 31324/43939 [50:33:04<20:02:35, 5.72s/it]
71%|███████▏ | 31325/43939 [50:33:09<20:05:34, 5.73s/it]
71%|███████▏ | 31326/43939 [50:33:15<20:05:17, 5.73s/it]
71%|███████▏ | 31327/43939 [50:33:21<20:03:10, 5.72s/it]
71%|███████▏ | 31328/43939 [50:33:26<20:03:15, 5.72s/it]
71%|███████▏ | 31329/43939 [50:33:32<20:01:58, 5.72s/it]
71%|███████▏ | 31330/43939 [50:33:38<20:04:32, 5.73s/it]
71%|███████▏ | 31331/43939 [50:33:44<20:03:06, 5.73s/it]
71%|███████▏ | 31332/43939 [50:33:49<20:01:59, 5.72s/it]
71%|███████▏ | 31333/43939 [50:33:55<20:00:34, 5.71s/it]
71%|███████▏ | 31334/43939 [50:34:01<20:01:59, 5.72s/it]
71%|███████▏ | 31335/43939 [50:34:06<20:00:56, 5.72s/it]
71%|███████▏ | 31336/43939 [50:34:12<20:14:55, 5.78s/it]
71%|███████▏ | 31337/43939 [50:34:18<20:07:28, 5.75s/it]
71%|███████▏ | 31338/43939 [50:34:24<20:03:01, 5.73s/it]
71%|███████▏ | 31339/43939 [50:34:29<19:59:00, 5.71s/it]
71%|███████▏ | 31340/43939 [50:34:35<20:02:04, 5.72s/it]
71%|███████▏ | 31341/43939 [50:34:41<19:59:34, 5.71s/it]
71%|███████▏ | 31342/43939 [50:34:47<20:01:47, 5.72s/it]
71%|███████▏ | 31343/43939 [50:34:52<20:01:17, 5.72s/it]
71%|███████▏ | 31344/43939 [50:34:58<19:59:46, 5.72s/it]
71%|███████▏ | 31345/43939 [50:35:04<19:59:10, 5.71s/it]
71%|███████▏ | 31346/43939 [50:35:09<19:56:43, 5.70s/it]
71%|███████▏ | 31347/43939 [50:35:15<19:57:59, 5.71s/it]
71%|███████▏ | 31348/43939 [50:35:21<19:57:08, 5.70s/it]
71%|███████▏ | 31349/43939 [50:35:27<19:57:33, 5.71s/it]
71%|███████▏ | 31350/43939 [50:35:32<19:57:26, 5.71s/it]
{'loss': 0.7636, 'grad_norm': 1.5844627618789673, 'learning_rate': 1.1495980861477456e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.15, 'epoch': 2.85}
71%|███████▏ | 31350/43939 [50:35:32<19:57:26, 5.71s/it]
71%|███████▏ | 31351/43939 [50:35:38<20:00:09, 5.72s/it]
71%|███████▏ | 31352/43939 [50:35:44<20:03:37, 5.74s/it]
71%|███████▏ | 31353/43939 [50:35:50<20:02:58, 5.73s/it]
71%|███████▏ | 31354/43939 [50:35:55<20:00:13, 5.72s/it]
71%|███████▏ | 31355/43939 [50:36:01<19:59:20, 5.72s/it]
71%|███████▏ | 31356/43939 [50:36:07<19:58:37, 5.72s/it]
71%|███████▏ | 31357/43939 [50:36:12<20:00:18, 5.72s/it]
71%|███████▏ | 31358/43939 [50:36:18<20:16:57, 5.80s/it]
71%|███████▏ | 31359/43939 [50:36:24<20:09:59, 5.77s/it]
71%|███████▏ | 31360/43939 [50:36:30<20:04:59, 5.75s/it]
71%|███████▏ | 31361/43939 [50:36:35<20:01:33, 5.73s/it]
71%|███████▏ | 31362/43939 [50:36:41<20:01:54, 5.73s/it]
71%|███████▏ | 31363/43939 [50:36:47<19:59:51, 5.72s/it]
71%|███████▏ | 31364/43939 [50:36:53<19:57:56, 5.72s/it]
71%|███████▏ | 31365/43939 [50:36:58<19:55:04, 5.70s/it]
71%|███████▏ | 31366/43939 [50:37:04<19:55:41, 5.71s/it]
71%|███████▏ | 31367/43939 [50:37:10<19:52:12, 5.69s/it]
71%|███████▏ | 31368/43939 [50:37:15<19:53:59, 5.70s/it]
71%|███████▏ | 31369/43939 [50:37:21<19:54:40, 5.70s/it]
71%|███████▏ | 31370/43939 [50:37:27<19:56:38, 5.71s/it]
71%|███████▏ | 31371/43939 [50:37:32<19:54:32, 5.70s/it]
71%|███████▏ | 31372/43939 [50:37:38<19:57:01, 5.72s/it]
71%|███████▏ | 31373/43939 [50:37:44<19:53:57, 5.70s/it]
71%|███████▏ | 31374/43939 [50:37:50<19:55:49, 5.71s/it]
71%|███████▏ | 31375/43939 [50:37:55<19:58:25, 5.72s/it]
71%|███████▏ | 31376/43939 [50:38:01<19:57:48, 5.72s/it]
71%|███████▏ | 31377/43939 [50:38:07<19:55:49, 5.71s/it]
71%|███████▏ | 31378/43939 [50:38:12<19:54:01, 5.70s/it]
71%|███████▏ | 31379/43939 [50:38:18<20:08:38, 5.77s/it]
71%|███████▏ | 31380/43939 [50:38:24<20:05:00, 5.76s/it]
71%|███████▏ | 31381/43939 [50:38:30<20:01:02, 5.74s/it]
71%|███████▏ | 31382/43939 [50:38:36<19:57:56, 5.72s/it]
71%|███████▏ | 31383/43939 [50:38:41<19:57:20, 5.72s/it]
71%|███████▏ | 31384/43939 [50:38:47<19:56:19, 5.72s/it]
71%|███████▏ | 31385/43939 [50:38:53<19:55:49, 5.72s/it]
71%|███████▏ | 31386/43939 [50:38:58<19:54:45, 5.71s/it]
71%|███████▏ | 31387/43939 [50:39:04<19:51:04, 5.69s/it]
71%|███████▏ | 31388/43939 [50:39:10<19:56:20, 5.72s/it]
71%|███████▏ | 31389/43939 [50:39:16<19:57:10, 5.72s/it]
71%|███████▏ | 31390/43939 [50:39:21<19:55:40, 5.72s/it]
71%|███████▏ | 31391/43939 [50:39:27<19:52:51, 5.70s/it]
71%|███████▏ | 31392/43939 [50:39:33<19:56:39, 5.72s/it]
71%|███████▏ | 31393/43939 [50:39:38<19:56:40, 5.72s/it]
71%|███████▏ | 31394/43939 [50:39:44<19:55:58, 5.72s/it]
71%|███████▏ | 31395/43939 [50:39:50<19:54:40, 5.71s/it]
71%|███████▏ | 31396/43939 [50:39:55<19:53:44, 5.71s/it]
71%|███████▏ | 31397/43939 [50:40:01<19:53:18, 5.71s/it]
71%|███████▏ | 31398/43939 [50:40:07<19:54:28, 5.71s/it]
71%|███████▏ | 31399/43939 [50:40:13<19:54:54, 5.72s/it]
71%|███████▏ | 31400/43939 [50:40:19<20:11:07, 5.80s/it]
{'loss': 0.7765, 'grad_norm': 1.4013336896896362, 'learning_rate': 1.1412518964495567e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.55, 'epoch': 2.86}
71%|███████▏ | 31400/43939 [50:40:19<20:11:07, 5.80s/it]
71%|███████▏ | 31401/43939 [50:40:24<20:06:19, 5.77s/it]
71%|███████▏ | 31402/43939 [50:40:30<20:00:56, 5.75s/it]
71%|███████▏ | 31403/43939 [50:40:36<20:03:11, 5.76s/it]
71%|███████▏ | 31404/43939 [50:40:42<19:58:06, 5.73s/it]
71%|███████▏ | 31405/43939 [50:40:47<19:56:04, 5.73s/it]
71%|███████▏ | 31406/43939 [50:40:53<19:58:29, 5.74s/it]
71%|███████▏ | 31407/43939 [50:40:59<19:55:40, 5.72s/it]
71%|███████▏ | 31408/43939 [50:41:04<19:55:03, 5.72s/it]
71%|███████▏ | 31409/43939 [50:41:10<19:55:58, 5.73s/it]
71%|███████▏ | 31410/43939 [50:41:16<19:54:23, 5.72s/it]
71%|███████▏ | 31411/43939 [50:41:22<19:54:10, 5.72s/it]
71%|███████▏ | 31412/43939 [50:41:27<19:52:26, 5.71s/it]
71%|███████▏ | 31413/43939 [50:41:33<19:52:34, 5.71s/it]
71%|███████▏ | 31414/43939 [50:41:39<19:52:25, 5.71s/it]
71%|███████▏ | 31415/43939 [50:41:44<19:52:24, 5.71s/it]
71%|███████▏ | 31416/43939 [50:41:50<19:54:18, 5.72s/it]
72%|███████▏ | 31417/43939 [50:41:56<19:54:45, 5.72s/it]
72%|███████▏ | 31418/43939 [50:42:02<19:53:34, 5.72s/it]
72%|███████▏ | 31419/43939 [50:42:07<19:55:11, 5.73s/it]
72%|███████▏ | 31420/43939 [50:42:13<19:52:22, 5.71s/it]
72%|███████▏ | 31421/43939 [50:42:19<19:51:24, 5.71s/it]
72%|███████▏ | 31422/43939 [50:42:25<20:08:14, 5.79s/it]
72%|███████▏ | 31423/43939 [50:42:30<20:04:27, 5.77s/it]
72%|███████▏ | 31424/43939 [50:42:36<19:59:44, 5.75s/it]
72%|███████▏ | 31425/43939 [50:42:42<19:56:00, 5.73s/it]
72%|███████▏ | 31426/43939 [50:42:48<19:58:19, 5.75s/it]
72%|███████▏ | 31427/43939 [50:42:53<19:55:06, 5.73s/it]
72%|███████▏ | 31428/43939 [50:42:59<19:54:34, 5.73s/it]
72%|███████▏ | 31429/43939 [50:43:05<19:53:09, 5.72s/it]
72%|███████▏ | 31430/43939 [50:43:10<19:52:53, 5.72s/it]
72%|███████▏ | 31431/43939 [50:43:16<19:51:25, 5.72s/it]
72%|███████▏ | 31432/43939 [50:43:22<19:51:25, 5.72s/it]
72%|███████▏ | 31433/43939 [50:43:28<19:50:59, 5.71s/it]
72%|███████▏ | 31434/43939 [50:43:33<19:50:05, 5.71s/it]
72%|███████▏ | 31435/43939 [50:43:39<19:49:08, 5.71s/it]
72%|███████▏ | 31436/43939 [50:43:45<19:52:01, 5.72s/it]
72%|███████▏ | 31437/43939 [50:43:50<19:49:26, 5.71s/it]
72%|███████▏ | 31438/43939 [50:43:56<19:46:35, 5.70s/it]
72%|███████▏ | 31439/43939 [50:44:02<19:46:26, 5.69s/it]
72%|███████▏ | 31440/43939 [50:44:07<19:46:11, 5.69s/it]
72%|███████▏ | 31441/43939 [50:44:13<19:46:43, 5.70s/it]
72%|███████▏ | 31442/43939 [50:44:19<19:47:47, 5.70s/it]
72%|███████▏ | 31443/43939 [50:44:25<19:48:10, 5.71s/it]
72%|███████▏ | 31444/43939 [50:44:31<20:07:39, 5.80s/it]
72%|███████▏ | 31445/43939 [50:44:36<19:59:57, 5.76s/it]
72%|███████▏ | 31446/43939 [50:44:42<19:59:14, 5.76s/it]
72%|███████▏ | 31447/43939 [50:44:48<19:55:31, 5.74s/it]
72%|███████▏ | 31448/43939 [50:44:53<19:55:39, 5.74s/it]
72%|███████▏ | 31449/43939 [50:44:59<19:57:25, 5.75s/it]
72%|███████▏ | 31450/43939 [50:45:05<19:53:21, 5.73s/it]
{'loss': 0.7962, 'grad_norm': 1.7948715686798096, 'learning_rate': 1.132927144195715e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 525.9, 'epoch': 2.86}
72%|███████▏ | 31450/43939 [50:45:05<19:53:21, 5.73s/it]
72%|███████▏ | 31451/43939 [50:45:11<19:51:41, 5.73s/it]
72%|███████▏ | 31452/43939 [50:45:16<19:51:21, 5.72s/it]
72%|███████▏ | 31453/43939 [50:45:22<19:46:37, 5.70s/it]
72%|███████▏ | 31454/43939 [50:45:28<19:46:26, 5.70s/it]
72%|███████▏ | 31455/43939 [50:45:33<19:48:23, 5.71s/it]
72%|███████▏ | 31456/43939 [50:45:39<19:48:03, 5.71s/it]
72%|███████▏ | 31457/43939 [50:45:45<19:51:34, 5.73s/it]
72%|███████▏ | 31458/43939 [50:45:51<19:50:06, 5.72s/it]
72%|███████▏ | 31459/43939 [50:45:56<19:49:33, 5.72s/it]
72%|███████▏ | 31460/43939 [50:46:02<19:46:49, 5.71s/it]
72%|███████▏ | 31461/43939 [50:46:08<19:48:46, 5.72s/it]
72%|███████▏ | 31462/43939 [50:46:13<19:45:51, 5.70s/it]
72%|███████▏ | 31463/43939 [50:46:19<19:47:30, 5.71s/it]
72%|███████▏ | 31464/43939 [50:46:25<19:48:35, 5.72s/it]
72%|███████▏ | 31465/43939 [50:46:31<20:06:22, 5.80s/it]
72%|███████▏ | 31466/43939 [50:46:37<20:00:10, 5.77s/it]
72%|███████▏ | 31467/43939 [50:46:42<19:57:37, 5.76s/it]
72%|███████▏ | 31468/43939 [50:46:48<19:53:21, 5.74s/it]
72%|███████▏ | 31469/43939 [50:46:54<19:50:22, 5.73s/it]
72%|███████▏ | 31470/43939 [50:46:59<19:48:57, 5.72s/it]
72%|███████▏ | 31471/43939 [50:47:05<19:47:49, 5.72s/it]
72%|███████▏ | 31472/43939 [50:47:11<19:46:18, 5.71s/it]
72%|███████▏ | 31473/43939 [50:47:17<19:47:04, 5.71s/it]
72%|███████▏ | 31474/43939 [50:47:22<19:46:59, 5.71s/it]
72%|███████▏ | 31475/43939 [50:47:28<19:45:36, 5.71s/it]
72%|███████▏ | 31476/43939 [50:47:34<19:42:46, 5.69s/it]
72%|███████▏ | 31477/43939 [50:47:39<19:43:04, 5.70s/it]
72%|███████▏ | 31478/43939 [50:47:45<19:43:21, 5.70s/it]
72%|███████▏ | 31479/43939 [50:47:51<19:45:14, 5.71s/it]
72%|███████▏ | 31480/43939 [50:47:56<19:47:27, 5.72s/it]
72%|███████▏ | 31481/43939 [50:48:02<19:44:22, 5.70s/it]
72%|███████▏ | 31482/43939 [50:48:08<19:43:47, 5.70s/it]
72%|███████▏ | 31483/43939 [50:48:14<19:45:19, 5.71s/it]
72%|███████▏ | 31484/43939 [50:48:19<19:43:53, 5.70s/it]
72%|███████▏ | 31485/43939 [50:48:25<19:42:23, 5.70s/it]
72%|███████▏ | 31486/43939 [50:48:31<20:00:07, 5.78s/it]
72%|███████▏ | 31487/43939 [50:48:37<19:53:47, 5.75s/it]
72%|███████▏ | 31488/43939 [50:48:42<19:49:28, 5.73s/it]
72%|███████▏ | 31489/43939 [50:48:48<19:46:27, 5.72s/it]
72%|███████▏ | 31490/43939 [50:48:54<19:45:54, 5.72s/it]
72%|███████▏ | 31491/43939 [50:48:59<19:49:04, 5.73s/it]
72%|███████▏ | 31492/43939 [50:49:05<19:46:37, 5.72s/it]
72%|███████▏ | 31493/43939 [50:49:11<19:49:12, 5.73s/it]
72%|███████▏ | 31494/43939 [50:49:17<19:46:51, 5.72s/it]
72%|███████▏ | 31495/43939 [50:49:22<19:50:10, 5.74s/it]
72%|███████▏ | 31496/43939 [50:49:28<19:47:48, 5.73s/it]
72%|███████▏ | 31497/43939 [50:49:34<19:44:47, 5.71s/it]
72%|███████▏ | 31498/43939 [50:49:39<19:42:23, 5.70s/it]
72%|███████▏ | 31499/43939 [50:49:45<19:41:53, 5.70s/it]
72%|███████▏ | 31500/43939 [50:49:51<19:38:35, 5.68s/it]
{'loss': 0.7898, 'grad_norm': 1.7212830781936646, 'learning_rate': 1.1246239607287513e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 526.86, 'epoch': 2.87}
72%|███████▏ | 31500/43939 [50:49:51<19:38:35, 5.68s/it]
72%|███████▏ | 31501/43939 [50:49:57<19:41:13, 5.70s/it]
72%|███████▏ | 31502/43939 [50:50:02<19:42:55, 5.71s/it]
72%|███████▏ | 31503/43939 [50:50:08<19:42:45, 5.71s/it]
72%|███████▏ | 31504/43939 [50:50:14<19:42:55, 5.71s/it]
72%|███████▏ | 31505/43939 [50:50:19<19:42:13, 5.70s/it]
72%|███████▏ | 31506/43939 [50:50:25<19:42:01, 5.70s/it]
72%|███████▏ | 31507/43939 [50:50:31<19:42:39, 5.71s/it]
72%|███████▏ | 31508/43939 [50:50:37<20:02:21, 5.80s/it]
72%|███████▏ | 31509/43939 [50:50:42<19:52:18, 5.76s/it]
72%|███████▏ | 31510/43939 [50:50:48<19:46:33, 5.73s/it]
72%|███████▏ | 31511/43939 [50:50:54<19:43:47, 5.72s/it]
72%|███████▏ | 31512/43939 [50:50:59<19:41:50, 5.71s/it]
72%|███████▏ | 31513/43939 [50:51:05<19:39:05, 5.69s/it]
72%|███████▏ | 31514/43939 [50:51:11<19:42:11, 5.71s/it]
72%|███████▏ | 31515/43939 [50:51:17<19:40:28, 5.70s/it]
72%|███████▏ | 31516/43939 [50:51:22<19:41:18, 5.71s/it]
72%|███████▏ | 31517/43939 [50:51:28<19:40:51, 5.70s/it]
72%|███████▏ | 31518/43939 [50:51:34<19:42:41, 5.71s/it]
72%|███████▏ | 31519/43939 [50:51:39<19:41:45, 5.71s/it]
72%|███████▏ | 31520/43939 [50:51:45<19:40:39, 5.70s/it]
72%|███████▏ | 31521/43939 [50:51:51<19:38:04, 5.69s/it]
72%|███████▏ | 31522/43939 [50:51:56<19:39:05, 5.70s/it]
72%|███████▏ | 31523/43939 [50:52:02<19:40:57, 5.71s/it]
72%|███████▏ | 31524/43939 [50:52:08<19:42:00, 5.71s/it]
72%|███████▏ | 31525/43939 [50:52:14<19:43:39, 5.72s/it]
72%|███████▏ | 31526/43939 [50:52:19<19:42:33, 5.72s/it]
72%|███████▏ | 31527/43939 [50:52:25<19:42:41, 5.72s/it]
72%|███████▏ | 31528/43939 [50:52:31<19:42:54, 5.72s/it]
72%|███████▏ | 31529/43939 [50:52:37<19:41:09, 5.71s/it]
72%|███████▏ | 31530/43939 [50:52:42<19:52:02, 5.76s/it]
72%|███████▏ | 31531/43939 [50:52:48<19:51:12, 5.76s/it]
72%|███████▏ | 31532/43939 [50:52:54<19:45:14, 5.73s/it]
72%|███████▏ | 31533/43939 [50:53:00<19:43:47, 5.73s/it]
72%|███████▏ | 31534/43939 [50:53:05<19:43:08, 5.72s/it]
72%|███████▏ | 31535/43939 [50:53:11<19:43:51, 5.73s/it]
72%|███████▏ | 31536/43939 [50:53:17<19:40:39, 5.71s/it]
72%|███████▏ | 31537/43939 [50:53:22<19:38:13, 5.70s/it]
72%|███████▏ | 31538/43939 [50:53:28<19:36:35, 5.69s/it]
72%|███████▏ | 31539/43939 [50:53:34<19:36:26, 5.69s/it]
72%|███████▏ | 31540/43939 [50:53:39<19:35:26, 5.69s/it]
72%|███████▏ | 31541/43939 [50:53:45<19:36:53, 5.70s/it]
72%|███████▏ | 31542/43939 [50:53:51<19:36:07, 5.69s/it]
72%|███████▏ | 31543/43939 [50:53:56<19:35:03, 5.69s/it]
72%|███████▏ | 31544/43939 [50:54:02<19:34:56, 5.69s/it]
72%|███████▏ | 31545/43939 [50:54:08<19:34:17, 5.68s/it]
72%|███████▏ | 31546/43939 [50:54:13<19:32:59, 5.68s/it]
72%|███████▏ | 31547/43939 [50:54:19<19:37:54, 5.70s/it]
72%|███████▏ | 31548/43939 [50:54:25<19:43:48, 5.73s/it]
72%|███████▏ | 31549/43939 [50:54:31<19:42:02, 5.72s/it]
72%|███████▏ | 31550/43939 [50:54:36<19:42:40, 5.73s/it]
{'loss': 0.7584, 'grad_norm': 1.5194088220596313, 'learning_rate': 1.116342477050896e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.35, 'epoch': 2.87}
72%|███████▏ | 31550/43939 [50:54:37<19:42:40, 5.73s/it]
72%|███████▏ | 31551/43939 [50:54:42<19:56:02, 5.79s/it]
72%|███████▏ | 31552/43939 [50:54:48<19:51:25, 5.77s/it]
72%|███████▏ | 31553/43939 [50:54:54<19:47:14, 5.75s/it]
72%|███████▏ | 31554/43939 [50:55:00<19:44:48, 5.74s/it]
72%|███████▏ | 31555/43939 [50:55:05<19:41:28, 5.72s/it]
72%|███████▏ | 31556/43939 [50:55:11<19:39:20, 5.71s/it]
72%|███████▏ | 31557/43939 [50:55:17<19:37:30, 5.71s/it]
72%|███████▏ | 31558/43939 [50:55:22<19:35:19, 5.70s/it]
72%|███████▏ | 31559/43939 [50:55:28<19:36:57, 5.70s/it]
72%|███████▏ | 31560/43939 [50:55:34<19:37:11, 5.71s/it]
72%|███████▏ | 31561/43939 [50:55:39<19:33:36, 5.69s/it]
72%|███████▏ | 31562/43939 [50:55:45<19:34:43, 5.69s/it]
72%|███████▏ | 31563/43939 [50:55:51<19:34:03, 5.69s/it]
72%|███████▏ | 31564/43939 [50:55:57<19:35:39, 5.70s/it]
72%|███████▏ | 31565/43939 [50:56:02<19:35:19, 5.70s/it]
72%|███████▏ | 31566/43939 [50:56:08<19:34:39, 5.70s/it]
72%|███████▏ | 31567/43939 [50:56:14<19:33:58, 5.69s/it]
72%|███████▏ | 31568/43939 [50:56:19<19:34:08, 5.69s/it]
72%|███████▏ | 31569/43939 [50:56:25<19:34:25, 5.70s/it]
72%|███████▏ | 31570/43939 [50:56:31<19:33:46, 5.69s/it]
72%|███████▏ | 31571/43939 [50:56:36<19:34:57, 5.70s/it]
72%|███████▏ | 31572/43939 [50:56:42<19:48:18, 5.77s/it]
72%|███████▏ | 31573/43939 [50:56:48<19:42:37, 5.74s/it]
72%|███████▏ | 31574/43939 [50:56:54<19:39:54, 5.73s/it]
72%|███████▏ | 31575/43939 [50:56:59<19:39:35, 5.72s/it]
72%|███████▏ | 31576/43939 [50:57:05<19:36:44, 5.71s/it]
72%|███████▏ | 31577/43939 [50:57:11<19:35:49, 5.71s/it]
72%|███████▏ | 31578/43939 [50:57:16<19:36:05, 5.71s/it]
72%|███████▏ | 31579/43939 [50:57:22<19:38:20, 5.72s/it]
72%|███████▏ | 31580/43939 [50:57:28<19:34:43, 5.70s/it]
72%|███████▏ | 31581/43939 [50:57:34<19:35:29, 5.71s/it]
72%|███████▏ | 31582/43939 [50:57:39<19:34:33, 5.70s/it]
72%|███████▏ | 31583/43939 [50:57:45<19:33:41, 5.70s/it]
72%|███████▏ | 31584/43939 [50:57:51<19:35:36, 5.71s/it]
72%|███████▏ | 31585/43939 [50:57:56<19:33:37, 5.70s/it]
72%|███████▏ | 31586/43939 [50:58:02<19:34:08, 5.70s/it]
72%|███████▏ | 31587/43939 [50:58:08<19:33:12, 5.70s/it]
72%|███████▏ | 31588/43939 [50:58:13<19:32:14, 5.69s/it]
72%|███████▏ | 31589/43939 [50:58:19<19:31:20, 5.69s/it]
72%|███████▏ | 31590/43939 [50:58:25<19:32:03, 5.69s/it]
72%|███████▏ | 31591/43939 [50:58:31<19:31:46, 5.69s/it]
72%|███████▏ | 31592/43939 [50:58:36<19:32:28, 5.70s/it]
72%|███████▏ | 31593/43939 [50:58:42<19:48:02, 5.77s/it]
72%|███████▏ | 31594/43939 [50:58:48<19:45:05, 5.76s/it]
72%|███████▏ | 31595/43939 [50:58:54<19:40:00, 5.74s/it]
72%|███████▏ | 31596/43939 [50:58:59<19:36:59, 5.72s/it]
72%|███████▏ | 31597/43939 [50:59:05<19:36:29, 5.72s/it]
72%|███████▏ | 31598/43939 [50:59:11<19:36:39, 5.72s/it]
72%|███████▏ | 31599/43939 [50:59:16<19:36:11, 5.72s/it]
72%|███████▏ | 31600/43939 [50:59:22<19:35:51, 5.72s/it]
{'loss': 0.7799, 'grad_norm': 1.4639698266983032, 'learning_rate': 1.1080828238220157e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.55, 'epoch': 2.88}
72%|███████▏ | 31600/43939 [50:59:22<19:35:51, 5.72s/it]
72%|███████▏ | 31601/43939 [50:59:28<19:39:25, 5.74s/it]
72%|███████▏ | 31602/43939 [50:59:34<19:38:05, 5.73s/it]
72%|███████▏ | 31603/43939 [50:59:39<19:37:59, 5.73s/it]
72%|███████▏ | 31604/43939 [50:59:45<19:33:13, 5.71s/it]
72%|███████▏ | 31605/43939 [50:59:51<19:31:07, 5.70s/it]
72%|███████▏ | 31606/43939 [50:59:56<19:31:26, 5.70s/it]
72%|███████▏ | 31607/43939 [51:00:02<19:31:16, 5.70s/it]
72%|███████▏ | 31608/43939 [51:00:08<19:31:54, 5.70s/it]
72%|███████▏ | 31609/43939 [51:00:14<19:30:15, 5.69s/it]
72%|███████▏ | 31610/43939 [51:00:19<19:29:10, 5.69s/it]
72%|███████▏ | 31611/43939 [51:00:25<19:30:10, 5.70s/it]
72%|███████▏ | 31612/43939 [51:00:31<19:29:17, 5.69s/it]
72%|███████▏ | 31613/43939 [51:00:36<19:32:55, 5.71s/it]
72%|███████▏ | 31614/43939 [51:00:42<19:32:34, 5.71s/it]
72%|███████▏ | 31615/43939 [51:00:48<19:44:49, 5.77s/it]
72%|███████▏ | 31616/43939 [51:00:54<19:39:36, 5.74s/it]
72%|███████▏ | 31617/43939 [51:00:59<19:39:49, 5.75s/it]
72%|███████▏ | 31618/43939 [51:01:05<19:37:06, 5.73s/it]
72%|███████▏ | 31619/43939 [51:01:11<19:33:44, 5.72s/it]
72%|███████▏ | 31620/43939 [51:01:17<19:34:22, 5.72s/it]
72%|███████▏ | 31621/43939 [51:01:22<19:35:01, 5.72s/it]
72%|███████▏ | 31622/43939 [51:01:28<19:32:46, 5.71s/it]
72%|███████▏ | 31623/43939 [51:01:34<19:29:50, 5.70s/it]
72%|███████▏ | 31624/43939 [51:01:39<19:32:27, 5.71s/it]
72%|███████▏ | 31625/43939 [51:01:45<19:34:44, 5.72s/it]
72%|███████▏ | 31626/43939 [51:01:51<19:32:58, 5.72s/it]
72%|███████▏ | 31627/43939 [51:01:57<19:33:06, 5.72s/it]
72%|███████▏ | 31628/43939 [51:02:02<19:32:39, 5.72s/it]
72%|███████▏ | 31629/43939 [51:02:08<19:31:11, 5.71s/it]
72%|███████▏ | 31630/43939 [51:02:14<19:30:21, 5.70s/it]
72%|███████▏ | 31631/43939 [51:02:19<19:30:13, 5.70s/it]
72%|███████▏ | 31632/43939 [51:02:25<19:29:25, 5.70s/it]
72%|███████▏ | 31633/43939 [51:02:31<19:30:24, 5.71s/it]
72%|███████▏ | 31634/43939 [51:02:36<19:31:15, 5.71s/it]
72%|███████▏ | 31635/43939 [51:02:42<19:28:05, 5.70s/it]
72%|███████▏ | 31636/43939 [51:02:48<19:42:24, 5.77s/it]
72%|███████▏ | 31637/43939 [51:02:54<19:44:36, 5.78s/it]
72%|███████▏ | 31638/43939 [51:03:00<19:39:03, 5.75s/it]
72%|███████▏ | 31639/43939 [51:03:05<19:34:43, 5.73s/it]
72%|███████▏ | 31640/43939 [51:03:11<19:34:16, 5.73s/it]
72%|███████▏ | 31641/43939 [51:03:17<19:33:42, 5.73s/it]
72%|███████▏ | 31642/43939 [51:03:22<19:31:36, 5.72s/it]
72%|███████▏ | 31643/43939 [51:03:28<19:31:01, 5.71s/it]
72%|███████▏ | 31644/43939 [51:03:34<19:28:38, 5.70s/it]
72%|███████▏ | 31645/43939 [51:03:39<19:27:10, 5.70s/it]
72%|███████▏ | 31646/43939 [51:03:45<19:26:22, 5.69s/it]
72%|███████▏ | 31647/43939 [51:03:51<19:27:45, 5.70s/it]
72%|███████▏ | 31648/43939 [51:03:56<19:25:28, 5.69s/it]
72%|███████▏ | 31649/43939 [51:04:02<19:23:50, 5.68s/it]
72%|███████▏ | 31650/43939 [51:04:08<19:23:45, 5.68s/it]
{'loss': 0.7683, 'grad_norm': 1.4781426191329956, 'learning_rate': 1.0998451313575503e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.53, 'epoch': 2.88}
72%|███████▏ | 31650/43939 [51:04:08<19:23:45, 5.68s/it]
72%|███████▏ | 31651/43939 [51:04:14<19:24:50, 5.69s/it]
72%|███████▏ | 31652/43939 [51:04:19<19:24:52, 5.69s/it]
72%|███████▏ | 31653/43939 [51:04:25<19:24:38, 5.69s/it]
72%|███████▏ | 31654/43939 [51:04:31<19:26:33, 5.70s/it]
72%|███████▏ | 31655/43939 [51:04:36<19:26:34, 5.70s/it]
72%|███████▏ | 31656/43939 [51:04:42<19:27:49, 5.70s/it]
72%|███████▏ | 31657/43939 [51:04:48<19:40:54, 5.77s/it]
72%|███████▏ | 31658/43939 [51:04:54<19:35:09, 5.74s/it]
72%|███████▏ | 31659/43939 [51:04:59<19:32:01, 5.73s/it]
72%|███████▏ | 31660/43939 [51:05:05<19:30:06, 5.72s/it]
72%|███████▏ | 31661/43939 [51:05:11<19:29:53, 5.72s/it]
72%|███████▏ | 31662/43939 [51:05:16<19:27:07, 5.70s/it]
72%|███████▏ | 31663/43939 [51:05:22<19:25:09, 5.69s/it]
72%|███████▏ | 31664/43939 [51:05:28<19:24:39, 5.69s/it]
72%|███████▏ | 31665/43939 [51:05:33<19:25:32, 5.70s/it]
72%|███████▏ | 31666/43939 [51:05:39<19:26:38, 5.70s/it]
72%|███████▏ | 31667/43939 [51:05:45<19:24:07, 5.69s/it]
72%|███████▏ | 31668/43939 [51:05:51<19:24:51, 5.70s/it]
72%|███████▏ | 31669/43939 [51:05:56<19:25:25, 5.70s/it]
72%|███████▏ | 31670/43939 [51:06:02<19:24:12, 5.69s/it]
72%|███████▏ | 31671/43939 [51:06:08<19:24:01, 5.69s/it]
72%|███████▏ | 31672/43939 [51:06:13<19:22:24, 5.69s/it]
72%|███████▏ | 31673/43939 [51:06:19<19:22:47, 5.69s/it]
72%|███████▏ | 31674/43939 [51:06:25<19:21:48, 5.68s/it]
72%|███████▏ | 31675/43939 [51:06:30<19:25:25, 5.70s/it]
72%|███████▏ | 31676/43939 [51:06:36<19:27:50, 5.71s/it]
72%|███████▏ | 31677/43939 [51:06:42<19:28:15, 5.72s/it]
72%|███████▏ | 31678/43939 [51:06:48<19:28:14, 5.72s/it]
72%|███████▏ | 31679/43939 [51:06:54<19:41:48, 5.78s/it]
72%|███████▏ | 31680/43939 [51:06:59<19:36:38, 5.76s/it]
72%|███████▏ | 31681/43939 [51:07:05<19:32:25, 5.74s/it]
72%|███████▏ | 31682/43939 [51:07:11<19:33:32, 5.74s/it]
72%|███████▏ | 31683/43939 [51:07:16<19:28:03, 5.72s/it]
72%|███████▏ | 31684/43939 [51:07:22<19:26:05, 5.71s/it]
72%|███████▏ | 31685/43939 [51:07:28<19:24:17, 5.70s/it]
72%|███████▏ | 31686/43939 [51:07:33<19:25:18, 5.71s/it]
72%|███████▏ | 31687/43939 [51:07:39<19:27:15, 5.72s/it]
72%|███████▏ | 31688/43939 [51:07:45<19:31:57, 5.74s/it]
72%|███████▏ | 31689/43939 [51:07:51<19:31:04, 5.74s/it]
72%|███████▏ | 31690/43939 [51:07:56<19:30:46, 5.73s/it]
72%|███████▏ | 31691/43939 [51:08:02<19:27:12, 5.72s/it]
72%|███████▏ | 31692/43939 [51:08:08<19:24:11, 5.70s/it]
72%|███████▏ | 31693/43939 [51:08:14<19:27:12, 5.72s/it]
72%|███████▏ | 31694/43939 [51:08:19<19:25:40, 5.71s/it]
72%|███████▏ | 31695/43939 [51:08:25<19:25:17, 5.71s/it]
72%|███████▏ | 31696/43939 [51:08:31<19:25:53, 5.71s/it]
72%|███████▏ | 31697/43939 [51:08:36<19:23:54, 5.70s/it]
72%|███████▏ | 31698/43939 [51:08:42<19:23:41, 5.70s/it]
72%|███████▏ | 31699/43939 [51:08:48<19:23:50, 5.71s/it]
72%|███████▏ | 31700/43939 [51:08:53<19:23:27, 5.70s/it]
{'loss': 0.774, 'grad_norm': 1.3627039194107056, 'learning_rate': 1.0916295296264565e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.56, 'epoch': 2.89}
72%|███████▏ | 31700/43939 [51:08:53<19:23:27, 5.70s/it]
72%|███████▏ | 31701/43939 [51:08:59<19:36:56, 5.77s/it]
72%|███████▏ | 31702/43939 [51:09:05<19:32:39, 5.75s/it]
72%|███████▏ | 31703/43939 [51:09:11<19:31:37, 5.75s/it]
72%|███████▏ | 31704/43939 [51:09:17<19:31:54, 5.75s/it]
72%|███████▏ | 31705/43939 [51:09:22<19:28:50, 5.73s/it]
72%|███████▏ | 31706/43939 [51:09:28<19:28:34, 5.73s/it]
72%|███████▏ | 31707/43939 [51:09:34<19:26:49, 5.72s/it]
72%|███████▏ | 31708/43939 [51:09:39<19:25:15, 5.72s/it]
72%|███████▏ | 31709/43939 [51:09:45<19:24:18, 5.71s/it]
72%|███████▏ | 31710/43939 [51:09:51<19:25:05, 5.72s/it]
72%|███████▏ | 31711/43939 [51:09:57<19:26:40, 5.72s/it]
72%|███████▏ | 31712/43939 [51:10:02<19:26:23, 5.72s/it]
72%|███████▏ | 31713/43939 [51:10:08<19:23:12, 5.71s/it]
72%|███████▏ | 31714/43939 [51:10:14<19:23:33, 5.71s/it]
72%|███████▏ | 31715/43939 [51:10:19<19:23:04, 5.71s/it]
72%|███████▏ | 31716/43939 [51:10:25<19:22:49, 5.71s/it]
72%|███████▏ | 31717/43939 [51:10:31<19:22:35, 5.71s/it]
72%|███████▏ | 31718/43939 [51:10:37<19:23:24, 5.71s/it]
72%|███████▏ | 31719/43939 [51:10:42<19:22:31, 5.71s/it]
72%|███████▏ | 31720/43939 [51:10:48<19:22:06, 5.71s/it]
72%|███████▏ | 31721/43939 [51:10:54<19:39:00, 5.79s/it]
72%|███████▏ | 31722/43939 [51:11:00<19:31:36, 5.75s/it]
72%|███████▏ | 31723/43939 [51:11:05<19:28:15, 5.74s/it]
72%|███████▏ | 31724/43939 [51:11:11<19:26:08, 5.73s/it]
72%|███████▏ | 31725/43939 [51:11:17<19:24:58, 5.72s/it]
72%|███████▏ | 31726/43939 [51:11:22<19:22:43, 5.71s/it]
72%|███████▏ | 31727/43939 [51:11:28<19:23:34, 5.72s/it]
72%|███████▏ | 31728/43939 [51:11:34<19:22:16, 5.71s/it]
72%|███████▏ | 31729/43939 [51:11:40<19:23:35, 5.72s/it]
72%|███████▏ | 31730/43939 [51:11:45<19:25:12, 5.73s/it]
72%|███████▏ | 31731/43939 [51:11:51<19:23:01, 5.72s/it]
72%|███████▏ | 31732/43939 [51:11:57<19:20:53, 5.71s/it]
72%|███████▏ | 31733/43939 [51:12:02<19:20:38, 5.71s/it]
72%|███████▏ | 31734/43939 [51:12:08<19:21:12, 5.71s/it]
72%|███████▏ | 31735/43939 [51:12:14<19:23:22, 5.72s/it]
72%|███████▏ | 31736/43939 [51:12:20<19:19:57, 5.70s/it]
72%|███████▏ | 31737/43939 [51:12:25<19:18:04, 5.69s/it]
72%|███████▏ | 31738/43939 [51:12:31<19:18:46, 5.70s/it]
72%|███████▏ | 31739/43939 [51:12:37<19:17:15, 5.69s/it]
72%|███████▏ | 31740/43939 [51:12:42<19:20:55, 5.71s/it]
72%|███████▏ | 31741/43939 [51:12:48<19:20:11, 5.71s/it]
72%|███████▏ | 31742/43939 [51:12:54<19:19:44, 5.71s/it]
72%|███████▏ | 31743/43939 [51:13:00<19:37:36, 5.79s/it]
72%|███████▏ | 31744/43939 [51:13:05<19:31:21, 5.76s/it]
72%|███████▏ | 31745/43939 [51:13:11<19:29:09, 5.75s/it]
72%|███████▏ | 31746/43939 [51:13:17<19:27:09, 5.74s/it]
72%|███████▏ | 31747/43939 [51:13:23<19:26:43, 5.74s/it]
72%|███████▏ | 31748/43939 [51:13:28<19:25:12, 5.73s/it]
72%|███████▏ | 31749/43939 [51:13:34<19:21:22, 5.72s/it]
72%|███████▏ | 31750/43939 [51:13:40<19:22:19, 5.72s/it]
{'loss': 0.765, 'grad_norm': 1.423780083656311, 'learning_rate': 1.083436148249157e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.8, 'epoch': 2.89}
72%|███████▏ | 31750/43939 [51:13:40<19:22:19, 5.72s/it]
72%|███████▏ | 31751/43939 [51:13:45<19:21:25, 5.72s/it]
72%|███████▏ | 31752/43939 [51:13:51<19:20:15, 5.71s/it]
72%|███████▏ | 31753/43939 [51:13:57<19:21:46, 5.72s/it]
72%|███████▏ | 31754/43939 [51:14:03<19:23:24, 5.73s/it]
72%|███████▏ | 31755/43939 [51:14:08<19:24:44, 5.74s/it]
72%|███████▏ | 31756/43939 [51:14:14<19:20:32, 5.72s/it]
72%|███████▏ | 31757/43939 [51:14:20<19:19:21, 5.71s/it]
72%|███████▏ | 31758/43939 [51:14:25<19:19:28, 5.71s/it]
72%|███████▏ | 31759/43939 [51:14:31<19:21:47, 5.72s/it]
72%|███████▏ | 31760/43939 [51:14:37<19:25:14, 5.74s/it]
72%|███████▏ | 31761/43939 [51:14:43<19:21:26, 5.72s/it]
72%|███████▏ | 31762/43939 [51:14:48<19:24:16, 5.74s/it]
72%|███████▏ | 31763/43939 [51:14:54<19:21:51, 5.73s/it]
72%|███████▏ | 31764/43939 [51:15:00<19:38:59, 5.81s/it]
72%|███████▏ | 31765/43939 [51:15:06<19:35:34, 5.79s/it]
72%|███████▏ | 31766/43939 [51:15:12<19:29:23, 5.76s/it]
72%|███████▏ | 31767/43939 [51:15:17<19:27:52, 5.76s/it]
72%|███████▏ | 31768/43939 [51:15:23<19:23:34, 5.74s/it]
72%|███████▏ | 31769/43939 [51:15:29<19:20:14, 5.72s/it]
72%|███████▏ | 31770/43939 [51:15:34<19:20:03, 5.72s/it]
72%|███████▏ | 31771/43939 [51:15:40<19:16:35, 5.70s/it]
72%|███████▏ | 31772/43939 [51:15:46<19:15:13, 5.70s/it]
72%|███████▏ | 31773/43939 [51:15:51<19:15:12, 5.70s/it]
72%|███████▏ | 31774/43939 [51:15:57<19:16:32, 5.70s/it]
72%|███████▏ | 31775/43939 [51:16:03<19:16:26, 5.70s/it]
72%|███████▏ | 31776/43939 [51:16:09<19:16:29, 5.70s/it]
72%|███████▏ | 31777/43939 [51:16:14<19:21:51, 5.73s/it]
72%|███████▏ | 31778/43939 [51:16:20<19:17:49, 5.71s/it]
72%|███████▏ | 31779/43939 [51:16:26<19:17:51, 5.71s/it]
72%|███████▏ | 31780/43939 [51:16:31<19:17:06, 5.71s/it]
72%|███████▏ | 31781/43939 [51:16:37<19:16:01, 5.71s/it]
72%|███████▏ | 31782/43939 [51:16:43<19:14:34, 5.70s/it]
72%|███████▏ | 31783/43939 [51:16:49<19:14:08, 5.70s/it]
72%|███████▏ | 31784/43939 [51:16:54<19:14:59, 5.70s/it]
72%|███████▏ | 31785/43939 [51:17:00<19:29:47, 5.77s/it]
72%|███████▏ | 31786/43939 [51:17:06<19:24:58, 5.75s/it]
72%|███████▏ | 31787/43939 [51:17:12<19:22:36, 5.74s/it]
72%|███████▏ | 31788/43939 [51:17:17<19:20:17, 5.73s/it]
72%|███████▏ | 31789/43939 [51:17:23<19:18:48, 5.72s/it]
72%|███████▏ | 31790/43939 [51:17:29<19:16:29, 5.71s/it]
72%|███████▏ | 31791/43939 [51:17:34<19:16:54, 5.71s/it]
72%|███████▏ | 31792/43939 [51:17:40<19:16:20, 5.71s/it]
72%|███████▏ | 31793/43939 [51:17:46<19:15:08, 5.71s/it]
72%|███████▏ | 31794/43939 [51:17:52<19:14:20, 5.70s/it]
72%|███████▏ | 31795/43939 [51:17:57<19:13:14, 5.70s/it]
72%|███████▏ | 31796/43939 [51:18:03<19:14:06, 5.70s/it]
72%|███████▏ | 31797/43939 [51:18:09<19:10:42, 5.69s/it]
72%|███████▏ | 31798/43939 [51:18:14<19:14:51, 5.71s/it]
72%|███████▏ | 31799/43939 [51:18:20<19:14:11, 5.70s/it]
72%|███████▏ | 31800/43939 [51:18:26<19:12:26, 5.70s/it]
{'loss': 0.7756, 'grad_norm': 1.3814677000045776, 'learning_rate': 1.0752651164954991e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.12, 'epoch': 2.89}
72%|███████▏ | 31800/43939 [51:18:26<19:12:26, 5.70s/it]
72%|███████▏ | 31801/43939 [51:18:31<19:12:55, 5.70s/it]
72%|███████▏ | 31802/43939 [51:18:37<19:12:29, 5.70s/it]
72%|███████▏ | 31803/43939 [51:18:43<19:10:41, 5.69s/it]
72%|███████▏ | 31804/43939 [51:18:48<19:09:29, 5.68s/it]
72%|███████▏ | 31805/43939 [51:18:54<19:10:46, 5.69s/it]
72%|███████▏ | 31806/43939 [51:19:00<19:11:26, 5.69s/it]
72%|███████▏ | 31807/43939 [51:19:06<19:27:50, 5.78s/it]
72%|███████▏ | 31808/43939 [51:19:12<19:27:35, 5.77s/it]
72%|███████▏ | 31809/43939 [51:19:17<19:23:52, 5.76s/it]
72%|███████▏ | 31810/43939 [51:19:23<19:22:11, 5.75s/it]
72%|███████▏ | 31811/43939 [51:19:29<19:19:56, 5.74s/it]
72%|███████▏ | 31812/43939 [51:19:34<19:16:28, 5.72s/it]
72%|███████▏ | 31813/43939 [51:19:40<19:18:13, 5.73s/it]
72%|███████▏ | 31814/43939 [51:19:46<19:19:50, 5.74s/it]
72%|███████▏ | 31815/43939 [51:19:52<19:16:06, 5.72s/it]
72%|███████▏ | 31816/43939 [51:19:57<19:16:02, 5.72s/it]
72%|███████▏ | 31817/43939 [51:20:03<19:16:39, 5.73s/it]
72%|███████▏ | 31818/43939 [51:20:09<19:13:22, 5.71s/it]
72%|███████▏ | 31819/43939 [51:20:14<19:12:14, 5.70s/it]
72%|███████▏ | 31820/43939 [51:20:20<19:12:38, 5.71s/it]
72%|███████▏ | 31821/43939 [51:20:26<19:16:02, 5.72s/it]
72%|███████▏ | 31822/43939 [51:20:32<19:14:47, 5.72s/it]
72%|███████▏ | 31823/43939 [51:20:37<19:11:15, 5.70s/it]
72%|███████▏ | 31824/43939 [51:20:43<19:10:51, 5.70s/it]
72%|███████▏ | 31825/43939 [51:20:49<19:11:20, 5.70s/it]
72%|███████▏ | 31826/43939 [51:20:54<19:10:21, 5.70s/it]
72%|███████▏ | 31827/43939 [51:21:00<19:10:44, 5.70s/it]
72%|███████▏ | 31828/43939 [51:21:06<19:26:16, 5.78s/it]
72%|███████▏ | 31829/43939 [51:21:12<19:23:46, 5.77s/it]
72%|███████▏ | 31830/43939 [51:21:17<19:19:07, 5.74s/it]
72%|███████▏ | 31831/43939 [51:21:23<19:15:59, 5.73s/it]
72%|███████▏ | 31832/43939 [51:21:29<19:13:36, 5.72s/it]
72%|███████▏ | 31833/43939 [51:21:35<19:12:57, 5.71s/it]
72%|███████▏ | 31834/43939 [51:21:40<19:14:35, 5.72s/it]
72%|███████▏ | 31835/43939 [51:21:46<19:12:02, 5.71s/it]
72%|███████▏ | 31836/43939 [51:21:52<19:09:43, 5.70s/it]
72%|███████▏ | 31837/43939 [51:21:57<19:10:22, 5.70s/it]
72%|███████▏ | 31838/43939 [51:22:03<19:12:32, 5.71s/it]
72%|███████▏ | 31839/43939 [51:22:09<19:14:07, 5.72s/it]
72%|███████▏ | 31840/43939 [51:22:15<19:11:13, 5.71s/it]
72%|███████▏ | 31841/43939 [51:22:20<19:12:30, 5.72s/it]
72%|███████▏ | 31842/43939 [51:22:26<19:10:52, 5.71s/it]
72%|███████▏ | 31843/43939 [51:22:32<19:10:46, 5.71s/it]
72%|███████▏ | 31844/43939 [51:22:37<19:09:01, 5.70s/it]
72%|███████▏ | 31845/43939 [51:22:43<19:08:12, 5.70s/it]
72%|███████▏ | 31846/43939 [51:22:49<19:09:50, 5.71s/it]
72%|███████▏ | 31847/43939 [51:22:54<19:09:40, 5.70s/it]
72%|███████▏ | 31848/43939 [51:23:00<19:09:43, 5.71s/it]
72%|███████▏ | 31849/43939 [51:23:06<19:08:27, 5.70s/it]
72%|███████▏ | 31850/43939 [51:23:12<19:23:15, 5.77s/it]
{'loss': 0.7894, 'grad_norm': 1.5840386152267456, 'learning_rate': 1.0671165632827097e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.78, 'epoch': 2.9}
72%|███████▏ | 31850/43939 [51:23:12<19:23:15, 5.77s/it]
72%|███████▏ | 31851/43939 [51:23:18<19:20:07, 5.76s/it]
72%|███████▏ | 31852/43939 [51:23:23<19:17:21, 5.75s/it]
72%|███████▏ | 31853/43939 [51:23:29<19:13:57, 5.73s/it]
72%|███████▏ | 31854/43939 [51:23:35<19:12:36, 5.72s/it]
72%|███████▏ | 31855/43939 [51:23:40<19:10:32, 5.71s/it]
73%|███████▎ | 31856/43939 [51:23:46<19:09:55, 5.71s/it]
73%|███████▎ | 31857/43939 [51:23:52<19:11:02, 5.72s/it]
73%|███████▎ | 31858/43939 [51:23:57<19:10:07, 5.71s/it]
73%|███████▎ | 31859/43939 [51:24:03<19:08:29, 5.70s/it]
73%|███████▎ | 31860/43939 [51:24:09<19:08:10, 5.70s/it]
73%|███████▎ | 31861/43939 [51:24:15<19:07:20, 5.70s/it]
73%|███████▎ | 31862/43939 [51:24:20<19:04:10, 5.68s/it]
73%|███████▎ | 31863/43939 [51:24:26<19:06:52, 5.70s/it]
73%|███████▎ | 31864/43939 [51:24:32<19:05:34, 5.69s/it]
73%|███████▎ | 31865/43939 [51:24:37<19:06:17, 5.70s/it]
73%|███████▎ | 31866/43939 [51:24:43<19:08:05, 5.71s/it]
73%|███████▎ | 31867/43939 [51:24:49<19:06:27, 5.70s/it]
73%|███████▎ | 31868/43939 [51:24:54<19:07:55, 5.71s/it]
73%|███████▎ | 31869/43939 [51:25:00<19:07:30, 5.70s/it]
73%|███████▎ | 31870/43939 [51:25:06<19:06:01, 5.70s/it]
73%|███████▎ | 31871/43939 [51:25:12<19:24:19, 5.79s/it]
73%|███████▎ | 31872/43939 [51:25:18<19:20:36, 5.77s/it]
73%|███████▎ | 31873/43939 [51:25:23<19:18:09, 5.76s/it]
73%|███████▎ | 31874/43939 [51:25:29<19:14:36, 5.74s/it]
73%|███████▎ | 31875/43939 [51:25:35<19:11:08, 5.73s/it]
73%|███████▎ | 31876/43939 [51:25:40<19:10:00, 5.72s/it]
73%|███████▎ | 31877/43939 [51:25:46<19:06:43, 5.70s/it]
73%|███████▎ | 31878/43939 [51:25:52<19:04:52, 5.70s/it]
73%|███████▎ | 31879/43939 [51:25:57<19:04:52, 5.70s/it]
73%|███████▎ | 31880/43939 [51:26:03<19:03:36, 5.69s/it]
73%|███████▎ | 31881/43939 [51:26:09<19:05:29, 5.70s/it]
73%|███████▎ | 31882/43939 [51:26:14<19:03:25, 5.69s/it]
73%|███████▎ | 31883/43939 [51:26:20<19:04:18, 5.69s/it]
73%|███████▎ | 31884/43939 [51:26:26<19:05:30, 5.70s/it]
73%|███████▎ | 31885/43939 [51:26:32<19:05:08, 5.70s/it]
73%|███████▎ | 31886/43939 [51:26:37<19:05:04, 5.70s/it]
73%|███████▎ | 31887/43939 [51:26:43<19:06:15, 5.71s/it]
73%|███████▎ | 31888/43939 [51:26:49<19:07:07, 5.71s/it]
73%|███████▎ | 31889/43939 [51:26:54<19:04:49, 5.70s/it]
73%|███████▎ | 31890/43939 [51:27:00<19:03:07, 5.69s/it]
73%|███████▎ | 31891/43939 [51:27:06<19:03:10, 5.69s/it]
73%|███████▎ | 31892/43939 [51:27:12<19:15:13, 5.75s/it]
73%|███████▎ | 31893/43939 [51:27:17<19:12:32, 5.74s/it]
73%|███████▎ | 31894/43939 [51:27:23<19:08:58, 5.72s/it]
73%|███████▎ | 31895/43939 [51:27:29<19:09:35, 5.73s/it]
73%|███████▎ | 31896/43939 [51:27:35<19:08:34, 5.72s/it]
73%|███████▎ | 31897/43939 [51:27:40<19:07:09, 5.72s/it]
73%|███████▎ | 31898/43939 [51:27:46<19:06:00, 5.71s/it]
73%|███████▎ | 31899/43939 [51:27:52<19:06:23, 5.71s/it]
73%|███████▎ | 31900/43939 [51:27:57<19:06:05, 5.71s/it]
{'loss': 0.7566, 'grad_norm': 1.2030837535858154, 'learning_rate': 1.0589906171733641e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.87, 'epoch': 2.9}
73%|███████▎ | 31900/43939 [51:27:57<19:06:05, 5.71s/it]
73%|███████▎ | 31901/43939 [51:28:03<19:07:16, 5.72s/it]
73%|███████▎ | 31902/43939 [51:28:09<19:08:03, 5.72s/it]
73%|███████▎ | 31903/43939 [51:28:15<19:06:15, 5.71s/it]
73%|███████▎ | 31904/43939 [51:28:20<19:05:15, 5.71s/it]
73%|███████▎ | 31905/43939 [51:28:26<19:06:29, 5.72s/it]
73%|███████▎ | 31906/43939 [51:28:32<19:05:31, 5.71s/it]
73%|███████▎ | 31907/43939 [51:28:37<19:03:09, 5.70s/it]
73%|███████▎ | 31908/43939 [51:28:43<19:02:27, 5.70s/it]
73%|███████▎ | 31909/43939 [51:28:49<19:01:59, 5.70s/it]
73%|███████▎ | 31910/43939 [51:28:54<19:01:02, 5.69s/it]
73%|███████▎ | 31911/43939 [51:29:00<19:01:26, 5.69s/it]
73%|███████▎ | 31912/43939 [51:29:06<19:02:17, 5.70s/it]
73%|███████▎ | 31913/43939 [51:29:12<19:05:32, 5.72s/it]
73%|███████▎ | 31914/43939 [51:29:17<19:19:10, 5.78s/it]
73%|███████▎ | 31915/43939 [51:29:23<19:13:49, 5.76s/it]
73%|███████▎ | 31916/43939 [51:29:29<19:09:38, 5.74s/it]
73%|███████▎ | 31917/43939 [51:29:35<19:06:40, 5.72s/it]
73%|███████▎ | 31918/43939 [51:29:40<19:04:20, 5.71s/it]
73%|███████▎ | 31919/43939 [51:29:46<19:03:07, 5.71s/it]
73%|███████▎ | 31920/43939 [51:29:52<19:03:08, 5.71s/it]
73%|███████▎ | 31921/43939 [51:29:57<19:01:24, 5.70s/it]
73%|███████▎ | 31922/43939 [51:30:03<19:00:07, 5.69s/it]
73%|███████▎ | 31923/43939 [51:30:09<19:01:02, 5.70s/it]
73%|███████▎ | 31924/43939 [51:30:14<19:01:05, 5.70s/it]
73%|███████▎ | 31925/43939 [51:30:20<19:02:00, 5.70s/it]
73%|███████▎ | 31926/43939 [51:30:26<18:59:44, 5.69s/it]
73%|███████▎ | 31927/43939 [51:30:32<19:00:10, 5.70s/it]
73%|███████▎ | 31928/43939 [51:30:37<19:01:31, 5.70s/it]
73%|███████▎ | 31929/43939 [51:30:43<18:59:46, 5.69s/it]
73%|███████▎ | 31930/43939 [51:30:49<19:01:50, 5.70s/it]
73%|███████▎ | 31931/43939 [51:30:54<19:05:05, 5.72s/it]
73%|███████▎ | 31932/43939 [51:31:00<19:03:34, 5.71s/it]
73%|███████▎ | 31933/43939 [51:31:06<19:02:16, 5.71s/it]
73%|███████▎ | 31934/43939 [51:31:11<19:01:27, 5.70s/it]
73%|███████▎ | 31935/43939 [51:31:17<19:00:42, 5.70s/it]
73%|███████▎ | 31936/43939 [51:31:23<19:16:59, 5.78s/it]
73%|███████▎ | 31937/43939 [51:31:29<19:13:44, 5.77s/it]
73%|███████▎ | 31938/43939 [51:31:35<19:09:44, 5.75s/it]
73%|███████▎ | 31939/43939 [51:31:40<19:05:49, 5.73s/it]
73%|███████▎ | 31940/43939 [51:31:46<19:03:52, 5.72s/it]
73%|███████▎ | 31941/43939 [51:31:52<19:02:40, 5.71s/it]
73%|███████▎ | 31942/43939 [51:31:57<19:00:44, 5.71s/it]
73%|███████▎ | 31943/43939 [51:32:03<19:01:27, 5.71s/it]
73%|███████▎ | 31944/43939 [51:32:09<19:03:16, 5.72s/it]
73%|███████▎ | 31945/43939 [51:32:15<19:03:23, 5.72s/it]
73%|███████▎ | 31946/43939 [51:32:20<19:00:53, 5.71s/it]
73%|███████▎ | 31947/43939 [51:32:26<18:58:53, 5.70s/it]
73%|███████▎ | 31948/43939 [51:32:32<18:57:50, 5.69s/it]
73%|███████▎ | 31949/43939 [51:32:37<18:58:35, 5.70s/it]
73%|███████▎ | 31950/43939 [51:32:43<18:57:24, 5.69s/it]
{'loss': 0.7636, 'grad_norm': 1.4285019636154175, 'learning_rate': 1.0508874063733582e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.3, 'epoch': 2.91}
73%|███████▎ | 31950/43939 [51:32:43<18:57:24, 5.69s/it]
73%|███████▎ | 31951/43939 [51:32:49<18:56:11, 5.69s/it]
73%|███████▎ | 31952/43939 [51:32:54<18:56:12, 5.69s/it]
73%|███████▎ | 31953/43939 [51:33:00<18:59:56, 5.71s/it]
73%|███████▎ | 31954/43939 [51:33:06<19:04:13, 5.73s/it]
73%|███████▎ | 31955/43939 [51:33:12<19:02:15, 5.72s/it]
73%|███████▎ | 31956/43939 [51:33:17<19:02:23, 5.72s/it]
73%|███████▎ | 31957/43939 [51:33:23<19:00:53, 5.71s/it]
73%|███████▎ | 31958/43939 [51:33:29<19:13:18, 5.78s/it]
73%|███████▎ | 31959/43939 [51:33:35<19:07:24, 5.75s/it]
73%|███████▎ | 31960/43939 [51:33:40<19:04:14, 5.73s/it]
73%|███████▎ | 31961/43939 [51:33:46<19:03:04, 5.73s/it]
73%|███████▎ | 31962/43939 [51:33:52<19:03:09, 5.73s/it]
73%|███████▎ | 31963/43939 [51:33:57<19:01:18, 5.72s/it]
73%|███████▎ | 31964/43939 [51:34:03<19:02:33, 5.72s/it]
73%|███████▎ | 31965/43939 [51:34:09<19:02:41, 5.73s/it]
73%|███████▎ | 31966/43939 [51:34:15<19:00:45, 5.72s/it]
73%|███████▎ | 31967/43939 [51:34:20<19:02:13, 5.72s/it]
73%|███████▎ | 31968/43939 [51:34:26<18:59:44, 5.71s/it]
73%|███████▎ | 31969/43939 [51:34:32<18:59:42, 5.71s/it]
73%|███████▎ | 31970/43939 [51:34:37<18:58:34, 5.71s/it]
73%|███████▎ | 31971/43939 [51:34:43<18:56:47, 5.70s/it]
73%|███████▎ | 31972/43939 [51:34:49<18:54:17, 5.69s/it]
73%|███████▎ | 31973/43939 [51:34:54<18:55:59, 5.70s/it]
73%|███████▎ | 31974/43939 [51:35:00<18:55:47, 5.70s/it]
73%|███████▎ | 31975/43939 [51:35:06<18:54:46, 5.69s/it]
73%|███████▎ | 31976/43939 [51:35:12<18:55:33, 5.70s/it]
73%|███████▎ | 31977/43939 [51:35:17<18:53:11, 5.68s/it]
73%|███████▎ | 31978/43939 [51:35:23<19:09:46, 5.77s/it]
73%|███████▎ | 31979/43939 [51:35:29<19:05:20, 5.75s/it]
73%|███████▎ | 31980/43939 [51:35:35<19:04:12, 5.74s/it]
73%|███████▎ | 31981/43939 [51:35:40<19:02:33, 5.73s/it]
73%|███████▎ | 31982/43939 [51:35:46<19:01:29, 5.73s/it]
73%|███████▎ | 31983/43939 [51:35:52<18:58:43, 5.71s/it]
73%|███████▎ | 31984/43939 [51:35:57<18:59:44, 5.72s/it]
73%|███████▎ | 31985/43939 [51:36:03<19:02:13, 5.73s/it]
73%|███████▎ | 31986/43939 [51:36:09<19:02:53, 5.74s/it]
73%|███████▎ | 31987/43939 [51:36:15<18:59:59, 5.72s/it]
73%|███████▎ | 31988/43939 [51:36:20<19:00:31, 5.73s/it]
73%|███████▎ | 31989/43939 [51:36:26<18:57:10, 5.71s/it]
73%|███████▎ | 31990/43939 [51:36:32<18:57:19, 5.71s/it]
73%|███████▎ | 31991/43939 [51:36:37<18:56:59, 5.71s/it]
73%|███████▎ | 31992/43939 [51:36:43<18:58:08, 5.72s/it]
73%|███████▎ | 31993/43939 [51:36:49<18:58:22, 5.72s/it]
73%|███████▎ | 31994/43939 [51:36:55<18:59:11, 5.72s/it]
73%|███████▎ | 31995/43939 [51:37:00<19:00:53, 5.73s/it]
73%|███████▎ | 31996/43939 [51:37:06<19:02:11, 5.74s/it]
73%|███████▎ | 31997/43939 [51:37:12<18:58:13, 5.72s/it]
73%|███████▎ | 31998/43939 [51:37:17<18:55:14, 5.70s/it]
73%|███████▎ | 31999/43939 [51:37:23<18:57:53, 5.72s/it]
73%|███████▎ | 32000/43939 [51:37:29<19:13:51, 5.80s/it]
{'loss': 0.7785, 'grad_norm': 1.4058661460876465, 'learning_rate': 1.0428070587298846e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.42, 'epoch': 2.91}
73%|███████▎ | 32000/43939 [51:37:29<19:13:51, 5.80s/it]
73%|███████▎ | 32001/43939 [51:37:35<19:10:36, 5.78s/it]
73%|███████▎ | 32002/43939 [51:37:41<19:06:54, 5.76s/it]
73%|███████▎ | 32003/43939 [51:37:46<19:05:13, 5.76s/it]
73%|███████▎ | 32004/43939 [51:37:52<19:03:12, 5.75s/it]
73%|███████▎ | 32005/43939 [51:37:58<19:00:28, 5.73s/it]
73%|███████▎ | 32006/43939 [51:38:04<18:58:09, 5.72s/it]
73%|███████▎ | 32007/43939 [51:38:09<18:55:26, 5.71s/it]
73%|███████▎ | 32008/43939 [51:38:15<18:56:28, 5.72s/it]
73%|███████▎ | 32009/43939 [51:38:21<18:53:14, 5.70s/it]
73%|███████▎ | 32010/43939 [51:38:26<18:57:35, 5.72s/it]
73%|███████▎ | 32011/43939 [51:38:32<18:55:17, 5.71s/it]
73%|███████▎ | 32012/43939 [51:38:38<18:54:42, 5.71s/it]
73%|███████▎ | 32013/43939 [51:38:43<18:54:49, 5.71s/it]
73%|███████▎ | 32014/43939 [51:38:49<18:54:01, 5.71s/it]
73%|███████▎ | 32015/43939 [51:38:55<18:52:54, 5.70s/it]
73%|███████▎ | 32016/43939 [51:39:01<18:50:52, 5.69s/it]
73%|███████▎ | 32017/43939 [51:39:06<18:49:25, 5.68s/it]
73%|███████▎ | 32018/43939 [51:39:12<18:52:41, 5.70s/it]
73%|███████▎ | 32019/43939 [51:39:18<18:50:43, 5.69s/it]
73%|███████▎ | 32020/43939 [51:39:23<18:51:17, 5.69s/it]
73%|███████▎ | 32021/43939 [51:39:29<19:05:59, 5.77s/it]
73%|███████▎ | 32022/43939 [51:39:35<19:03:10, 5.76s/it]
73%|███████▎ | 32023/43939 [51:39:41<19:02:14, 5.75s/it]
73%|███████▎ | 32024/43939 [51:39:46<19:00:16, 5.74s/it]
73%|███████▎ | 32025/43939 [51:39:52<18:57:29, 5.73s/it]
73%|███████▎ | 32026/43939 [51:39:58<18:57:33, 5.73s/it]
73%|███████▎ | 32027/43939 [51:40:04<18:56:43, 5.73s/it]
73%|███████▎ | 32028/43939 [51:40:09<18:55:08, 5.72s/it]
73%|███████▎ | 32029/43939 [51:40:15<18:55:30, 5.72s/it]
73%|███████▎ | 32030/43939 [51:40:21<18:52:05, 5.70s/it]
73%|███████▎ | 32031/43939 [51:40:26<18:51:25, 5.70s/it]
73%|███████▎ | 32032/43939 [51:40:32<18:50:52, 5.70s/it]
73%|███████▎ | 32033/43939 [51:40:38<18:54:50, 5.72s/it]
73%|███████▎ | 32034/43939 [51:40:44<18:52:16, 5.71s/it]
73%|███████▎ | 32035/43939 [51:40:49<18:54:57, 5.72s/it]
73%|███████▎ | 32036/43939 [51:40:55<18:56:07, 5.73s/it]
73%|███████▎ | 32037/43939 [51:41:01<18:52:31, 5.71s/it]
73%|███████▎ | 32038/43939 [51:41:06<18:54:34, 5.72s/it]
73%|███████▎ | 32039/43939 [51:41:12<18:52:52, 5.71s/it]
73%|███████▎ | 32040/43939 [51:41:18<18:51:47, 5.71s/it]
73%|███████▎ | 32041/43939 [51:41:23<18:49:38, 5.70s/it]
73%|███████▎ | 32042/43939 [51:41:29<19:07:06, 5.79s/it]
73%|███████▎ | 32043/43939 [51:41:35<19:00:58, 5.75s/it]
73%|███████▎ | 32044/43939 [51:41:41<18:57:10, 5.74s/it]
73%|███████▎ | 32045/43939 [51:41:47<18:53:54, 5.72s/it]
73%|███████▎ | 32046/43939 [51:41:52<18:53:43, 5.72s/it]
73%|███████▎ | 32047/43939 [51:41:58<18:54:50, 5.73s/it]
73%|███████▎ | 32048/43939 [51:42:04<18:54:01, 5.72s/it]
73%|███████▎ | 32049/43939 [51:42:09<18:52:33, 5.72s/it]
73%|███████▎ | 32050/43939 [51:42:15<18:52:47, 5.72s/it]
{'loss': 0.7675, 'grad_norm': 1.3887815475463867, 'learning_rate': 1.034749701729416e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.89, 'epoch': 2.92}
73%|███████▎ | 32050/43939 [51:42:15<18:52:47, 5.72s/it]
73%|███████▎ | 32051/43939 [51:42:21<18:51:36, 5.71s/it]
73%|███████▎ | 32052/43939 [51:42:27<18:53:27, 5.72s/it]
73%|███████▎ | 32053/43939 [51:42:32<18:53:11, 5.72s/it]
73%|███████▎ | 32054/43939 [51:42:38<18:54:28, 5.73s/it]
73%|███████▎ | 32055/43939 [51:42:44<18:52:25, 5.72s/it]
73%|███████▎ | 32056/43939 [51:42:49<18:50:36, 5.71s/it]
73%|███████▎ | 32057/43939 [51:42:55<18:50:51, 5.71s/it]
73%|███████▎ | 32058/43939 [51:43:01<18:49:01, 5.70s/it]
73%|███████▎ | 32059/43939 [51:43:07<18:51:07, 5.71s/it]
73%|███████▎ | 32060/43939 [51:43:12<18:50:27, 5.71s/it]
73%|███████▎ | 32061/43939 [51:43:18<18:49:59, 5.71s/it]
73%|███████▎ | 32062/43939 [51:43:24<18:49:00, 5.70s/it]
73%|███████▎ | 32063/43939 [51:43:30<19:04:12, 5.78s/it]
73%|███████▎ | 32064/43939 [51:43:35<19:00:55, 5.76s/it]
73%|███████▎ | 32065/43939 [51:43:41<18:55:05, 5.74s/it]
73%|███████▎ | 32066/43939 [51:43:47<18:51:14, 5.72s/it]
73%|███████▎ | 32067/43939 [51:43:52<18:51:51, 5.72s/it]
73%|███████▎ | 32068/43939 [51:43:58<18:50:55, 5.72s/it]
73%|███████▎ | 32069/43939 [51:44:04<18:48:37, 5.70s/it]
73%|███████▎ | 32070/43939 [51:44:10<18:49:11, 5.71s/it]
73%|███████▎ | 32071/43939 [51:44:15<18:47:16, 5.70s/it]
73%|███████▎ | 32072/43939 [51:44:21<18:47:53, 5.70s/it]
73%|███████▎ | 32073/43939 [51:44:27<18:48:35, 5.71s/it]
73%|███████▎ | 32074/43939 [51:44:32<18:48:13, 5.71s/it]
73%|███████▎ | 32075/43939 [51:44:38<18:46:04, 5.69s/it]
73%|███████▎ | 32076/43939 [51:44:44<18:44:41, 5.69s/it]
73%|███████▎ | 32077/43939 [51:44:49<18:47:11, 5.70s/it]
73%|███████▎ | 32078/43939 [51:44:55<18:47:10, 5.70s/it]
73%|███████▎ | 32079/43939 [51:45:01<18:47:38, 5.70s/it]
73%|███████▎ | 32080/43939 [51:45:06<18:44:55, 5.69s/it]
73%|███████▎ | 32081/43939 [51:45:12<18:46:01, 5.70s/it]
73%|███████▎ | 32082/43939 [51:45:18<18:46:41, 5.70s/it]
73%|███████▎ | 32083/43939 [51:45:24<18:45:42, 5.70s/it]
73%|███████▎ | 32084/43939 [51:45:29<18:44:42, 5.69s/it]
73%|███████▎ | 32085/43939 [51:45:35<18:59:16, 5.77s/it]
73%|███████▎ | 32086/43939 [51:45:41<18:59:36, 5.77s/it]
73%|███████▎ | 32087/43939 [51:45:47<18:54:01, 5.74s/it]
73%|███████▎ | 32088/43939 [51:45:52<18:51:42, 5.73s/it]
73%|███████▎ | 32089/43939 [51:45:58<18:47:52, 5.71s/it]
73%|███████▎ | 32090/43939 [51:46:04<18:45:32, 5.70s/it]
73%|███████▎ | 32091/43939 [51:46:09<18:44:52, 5.70s/it]
73%|███████▎ | 32092/43939 [51:46:15<18:46:56, 5.71s/it]
73%|███████▎ | 32093/43939 [51:46:21<18:44:22, 5.69s/it]
73%|███████▎ | 32094/43939 [51:46:26<18:43:32, 5.69s/it]
73%|███████▎ | 32095/43939 [51:46:32<18:45:37, 5.70s/it]
73%|███████▎ | 32096/43939 [51:46:38<18:45:46, 5.70s/it]
73%|███████▎ | 32097/43939 [51:46:44<18:49:33, 5.72s/it]
73%|███████▎ | 32098/43939 [51:46:49<18:46:23, 5.71s/it]
73%|███████▎ | 32099/43939 [51:46:55<18:44:29, 5.70s/it]
73%|███████▎ | 32100/43939 [51:47:01<18:44:33, 5.70s/it]
{'loss': 0.7545, 'grad_norm': 1.4016015529632568, 'learning_rate': 1.0267154624956937e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.9, 'epoch': 2.92}
73%|███████▎ | 32100/43939 [51:47:01<18:44:33, 5.70s/it]
73%|███████▎ | 32101/43939 [51:47:06<18:45:56, 5.71s/it]
73%|███████▎ | 32102/43939 [51:47:12<18:44:55, 5.70s/it]
73%|███████▎ | 32103/43939 [51:47:18<18:46:30, 5.71s/it]
73%|███████▎ | 32104/43939 [51:47:24<18:45:19, 5.71s/it]
73%|███████▎ | 32105/43939 [51:47:29<18:44:19, 5.70s/it]
73%|███████▎ | 32106/43939 [51:47:35<18:45:43, 5.71s/it]
73%|███████▎ | 32107/43939 [51:47:41<18:59:13, 5.78s/it]
73%|███████▎ | 32108/43939 [51:47:47<18:55:47, 5.76s/it]
73%|███████▎ | 32109/43939 [51:47:52<18:52:00, 5.74s/it]
73%|███████▎ | 32110/43939 [51:47:58<18:48:55, 5.73s/it]
73%|███████▎ | 32111/43939 [51:48:04<18:47:25, 5.72s/it]
73%|███████▎ | 32112/43939 [51:48:09<18:45:18, 5.71s/it]
73%|███████▎ | 32113/43939 [51:48:15<18:45:35, 5.71s/it]
73%|███████▎ | 32114/43939 [51:48:21<18:45:31, 5.71s/it]
73%|███████▎ | 32115/43939 [51:48:27<18:44:30, 5.71s/it]
73%|███████▎ | 32116/43939 [51:48:32<18:43:56, 5.70s/it]
73%|███████▎ | 32117/43939 [51:48:38<18:43:44, 5.70s/it]
73%|███████▎ | 32118/43939 [51:48:44<18:41:16, 5.69s/it]
73%|███████▎ | 32119/43939 [51:48:49<18:41:43, 5.69s/it]
73%|███████▎ | 32120/43939 [51:48:55<18:40:29, 5.69s/it]
73%|███████▎ | 32121/43939 [51:49:01<18:41:37, 5.69s/it]
73%|███████▎ | 32122/43939 [51:49:06<18:40:52, 5.69s/it]
73%|███████▎ | 32123/43939 [51:49:12<18:38:51, 5.68s/it]
73%|███████▎ | 32124/43939 [51:49:18<18:39:26, 5.68s/it]
73%|███████▎ | 32125/43939 [51:49:23<18:39:45, 5.69s/it]
73%|███████▎ | 32126/43939 [51:49:29<18:37:16, 5.67s/it]
73%|███████▎ | 32127/43939 [51:49:35<18:40:09, 5.69s/it]
73%|███████▎ | 32128/43939 [51:49:41<18:54:02, 5.76s/it]
73%|███████▎ | 32129/43939 [51:49:46<18:48:27, 5.73s/it]
73%|███████▎ | 32130/43939 [51:49:52<18:45:57, 5.72s/it]
73%|███████▎ | 32131/43939 [51:49:58<18:45:32, 5.72s/it]
73%|███████▎ | 32132/43939 [51:50:04<18:45:51, 5.72s/it]
73%|███████▎ | 32133/43939 [51:50:09<18:44:51, 5.72s/it]
73%|███████▎ | 32134/43939 [51:50:15<18:43:10, 5.71s/it]
73%|███████▎ | 32135/43939 [51:50:21<18:42:31, 5.71s/it]
73%|███████▎ | 32136/43939 [51:50:26<18:41:51, 5.70s/it]
73%|███████▎ | 32137/43939 [51:50:32<18:38:45, 5.69s/it]
73%|███████▎ | 32138/43939 [51:50:38<18:38:23, 5.69s/it]
73%|███████▎ | 32139/43939 [51:50:43<18:40:16, 5.70s/it]
73%|███████▎ | 32140/43939 [51:50:49<18:44:07, 5.72s/it]
73%|███████▎ | 32141/43939 [51:50:55<18:42:59, 5.71s/it]
73%|███████▎ | 32142/43939 [51:51:01<18:43:12, 5.71s/it]
73%|███████▎ | 32143/43939 [51:51:06<18:46:03, 5.73s/it]
73%|███████▎ | 32144/43939 [51:51:12<18:45:15, 5.72s/it]
73%|███████▎ | 32145/43939 [51:51:18<18:44:33, 5.72s/it]
73%|███████▎ | 32146/43939 [51:51:23<18:43:47, 5.72s/it]
73%|███████▎ | 32147/43939 [51:51:29<18:46:26, 5.73s/it]
73%|███████▎ | 32148/43939 [51:51:35<18:45:49, 5.73s/it]
73%|███████▎ | 32149/43939 [51:51:41<18:55:55, 5.78s/it]
73%|███████▎ | 32150/43939 [51:51:47<18:51:22, 5.76s/it]
{'loss': 0.7635, 'grad_norm': 1.3880101442337036, 'learning_rate': 1.0187044677877216e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.32, 'epoch': 2.93}
73%|███████▎ | 32150/43939 [51:51:47<18:51:22, 5.76s/it]
73%|███████▎ | 32151/43939 [51:51:52<18:46:35, 5.73s/it]
73%|███████▎ | 32152/43939 [51:51:58<18:44:07, 5.72s/it]
73%|███████▎ | 32153/43939 [51:52:04<18:41:35, 5.71s/it]
73%|███████▎ | 32154/43939 [51:52:09<18:38:44, 5.70s/it]
73%|███████▎ | 32155/43939 [51:52:15<18:36:36, 5.69s/it]
73%|███████▎ | 32156/43939 [51:52:21<18:38:45, 5.70s/it]
73%|███████▎ | 32157/43939 [51:52:26<18:39:42, 5.70s/it]
73%|███████▎ | 32158/43939 [51:52:32<18:40:04, 5.70s/it]
73%|███████▎ | 32159/43939 [51:52:38<18:39:44, 5.70s/it]
73%|███████▎ | 32160/43939 [51:52:43<18:41:49, 5.71s/it]
73%|███████▎ | 32161/43939 [51:52:49<18:42:48, 5.72s/it]
73%|███████▎ | 32162/43939 [51:52:55<18:38:37, 5.70s/it]
73%|███████▎ | 32163/43939 [51:53:01<18:40:56, 5.71s/it]
73%|███████▎ | 32164/43939 [51:53:06<18:39:58, 5.71s/it]
73%|███████▎ | 32165/43939 [51:53:12<18:40:36, 5.71s/it]
73%|███████▎ | 32166/43939 [51:53:18<18:41:57, 5.72s/it]
73%|███████▎ | 32167/43939 [51:53:23<18:37:52, 5.70s/it]
73%|███████▎ | 32168/43939 [51:53:29<18:39:00, 5.70s/it]
73%|███████▎ | 32169/43939 [51:53:35<18:37:17, 5.70s/it]
73%|███████▎ | 32170/43939 [51:53:41<18:49:15, 5.76s/it]
73%|███████▎ | 32171/43939 [51:53:46<18:47:00, 5.75s/it]
73%|███████▎ | 32172/43939 [51:53:52<18:42:29, 5.72s/it]
73%|███████▎ | 32173/43939 [51:53:58<18:42:42, 5.73s/it]
73%|███████▎ | 32174/43939 [51:54:04<18:42:43, 5.73s/it]
73%|███████▎ | 32175/43939 [51:54:09<18:39:07, 5.71s/it]
73%|███████▎ | 32176/43939 [51:54:15<18:39:29, 5.71s/it]
73%|███████▎ | 32177/43939 [51:54:21<18:39:15, 5.71s/it]
73%|███████▎ | 32178/43939 [51:54:26<18:39:13, 5.71s/it]
73%|███████▎ | 32179/43939 [51:54:32<18:39:05, 5.71s/it]
73%|███████▎ | 32180/43939 [51:54:38<18:38:36, 5.71s/it]
73%|███████▎ | 32181/43939 [51:54:43<18:37:15, 5.70s/it]
73%|███████▎ | 32182/43939 [51:54:49<18:37:05, 5.70s/it]
73%|███████▎ | 32183/43939 [51:54:55<18:36:04, 5.70s/it]
73%|███████▎ | 32184/43939 [51:55:01<18:36:18, 5.70s/it]
73%|███████▎ | 32185/43939 [51:55:06<18:37:50, 5.71s/it]
73%|███████▎ | 32186/43939 [51:55:12<18:34:54, 5.69s/it]
73%|███████▎ | 32187/43939 [51:55:18<18:35:24, 5.69s/it]
73%|███████▎ | 32188/43939 [51:55:23<18:34:51, 5.69s/it]
73%|███████▎ | 32189/43939 [51:55:29<18:36:26, 5.70s/it]
73%|███████▎ | 32190/43939 [51:55:35<18:35:09, 5.69s/it]
73%|███████▎ | 32191/43939 [51:55:40<18:35:06, 5.70s/it]
73%|███████▎ | 32192/43939 [51:55:46<18:47:20, 5.76s/it]
73%|███████▎ | 32193/43939 [51:55:52<18:43:52, 5.74s/it]
73%|███████▎ | 32194/43939 [51:55:58<18:41:43, 5.73s/it]
73%|███████▎ | 32195/43939 [51:56:03<18:42:05, 5.73s/it]
73%|███████▎ | 32196/43939 [51:56:09<18:40:30, 5.73s/it]
73%|███████▎ | 32197/43939 [51:56:15<18:41:31, 5.73s/it]
73%|███████▎ | 32198/43939 [51:56:21<18:38:29, 5.72s/it]
73%|███████▎ | 32199/43939 [51:56:26<18:35:28, 5.70s/it]
73%|███████▎ | 32200/43939 [51:56:32<18:36:50, 5.71s/it]
{'loss': 0.7569, 'grad_norm': 1.4154095649719238, 'learning_rate': 1.0107168439977669e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 453.4, 'epoch': 2.93}
73%|███████▎ | 32200/43939 [51:56:32<18:36:50, 5.71s/it]
73%|███████▎ | 32201/43939 [51:56:38<18:36:44, 5.71s/it]
73%|███████▎ | 32202/43939 [51:56:43<18:38:44, 5.72s/it]
73%|███████▎ | 32203/43939 [51:56:49<18:35:42, 5.70s/it]
73%|███████▎ | 32204/43939 [51:56:55<18:36:25, 5.71s/it]
73%|███████▎ | 32205/43939 [51:57:00<18:33:23, 5.69s/it]
73%|███████▎ | 32206/43939 [51:57:06<18:34:29, 5.70s/it]
73%|███████▎ | 32207/43939 [51:57:12<18:35:16, 5.70s/it]
73%|███████▎ | 32208/43939 [51:57:18<18:33:16, 5.69s/it]
73%|███████▎ | 32209/43939 [51:57:23<18:33:14, 5.69s/it]
73%|███████▎ | 32210/43939 [51:57:29<18:33:33, 5.70s/it]
73%|███████▎ | 32211/43939 [51:57:35<18:36:21, 5.71s/it]
73%|███████▎ | 32212/43939 [51:57:41<18:49:02, 5.78s/it]
73%|███████▎ | 32213/43939 [51:57:46<18:43:19, 5.75s/it]
73%|███████▎ | 32214/43939 [51:57:52<18:41:16, 5.74s/it]
73%|███████▎ | 32215/43939 [51:57:58<18:38:35, 5.72s/it]
73%|███████▎ | 32216/43939 [51:58:03<18:38:41, 5.73s/it]
73%|███████▎ | 32217/43939 [51:58:09<18:37:33, 5.72s/it]
73%|███████▎ | 32218/43939 [51:58:15<18:38:24, 5.73s/it]
73%|███████▎ | 32219/43939 [51:58:21<18:37:37, 5.72s/it]
73%|███████▎ | 32220/43939 [51:58:26<18:33:53, 5.70s/it]
73%|███████▎ | 32221/43939 [51:58:32<18:36:17, 5.72s/it]
73%|███████▎ | 32222/43939 [51:58:38<18:35:04, 5.71s/it]
73%|███████▎ | 32223/43939 [51:58:43<18:30:31, 5.69s/it]
73%|███████▎ | 32224/43939 [51:58:49<18:32:28, 5.70s/it]
73%|███████▎ | 32225/43939 [51:58:55<18:31:33, 5.69s/it]
73%|███████▎ | 32226/43939 [51:59:00<18:31:04, 5.69s/it]
73%|███████▎ | 32227/43939 [51:59:06<18:31:05, 5.69s/it]
73%|███████▎ | 32228/43939 [51:59:12<18:33:06, 5.70s/it]
73%|███████▎ | 32229/43939 [51:59:18<18:31:40, 5.70s/it]
73%|███████▎ | 32230/43939 [51:59:23<18:32:16, 5.70s/it]
73%|███████▎ | 32231/43939 [51:59:29<18:30:36, 5.69s/it]
73%|███████▎ | 32232/43939 [51:59:35<18:29:23, 5.69s/it]
73%|███████▎ | 32233/43939 [51:59:40<18:28:46, 5.68s/it]
73%|███████▎ | 32234/43939 [51:59:46<18:43:34, 5.76s/it]
73%|███████▎ | 32235/43939 [51:59:52<18:40:02, 5.74s/it]
73%|███████▎ | 32236/43939 [51:59:58<18:36:32, 5.72s/it]
73%|███████▎ | 32237/43939 [52:00:03<18:35:35, 5.72s/it]
73%|███████▎ | 32238/43939 [52:00:09<18:33:43, 5.71s/it]
73%|███████▎ | 32239/43939 [52:00:15<18:32:09, 5.70s/it]
73%|███████▎ | 32240/43939 [52:00:20<18:34:09, 5.71s/it]
73%|███████▎ | 32241/43939 [52:00:26<18:32:34, 5.71s/it]
73%|███████▎ | 32242/43939 [52:00:32<18:33:31, 5.71s/it]
73%|███████▎ | 32243/43939 [52:00:38<18:35:45, 5.72s/it]
73%|███████▎ | 32244/43939 [52:00:43<18:34:27, 5.72s/it]
73%|███████▎ | 32245/43939 [52:00:49<18:33:22, 5.71s/it]
73%|███████▎ | 32246/43939 [52:00:55<18:30:18, 5.70s/it]
73%|███████▎ | 32247/43939 [52:01:00<18:29:04, 5.69s/it]
73%|███████▎ | 32248/43939 [52:01:06<18:30:38, 5.70s/it]
73%|███████▎ | 32249/43939 [52:01:12<18:32:59, 5.71s/it]
73%|███████▎ | 32250/43939 [52:01:17<18:30:34, 5.70s/it]
{'loss': 0.7684, 'grad_norm': 1.4633469581604004, 'learning_rate': 1.0027527171493654e-05, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.0, 'epoch': 2.94}
73%|███████▎ | 32250/43939 [52:01:17<18:30:34, 5.70s/it]
73%|███████▎ | 32251/43939 [52:01:23<18:30:55, 5.70s/it]
73%|███████▎ | 32252/43939 [52:01:29<18:28:28, 5.69s/it]
73%|███████▎ | 32253/43939 [52:01:35<18:33:49, 5.72s/it]
73%|███████▎ | 32254/43939 [52:01:40<18:31:50, 5.71s/it]
73%|███████▎ | 32255/43939 [52:01:46<18:45:22, 5.78s/it]
73%|███████▎ | 32256/43939 [52:01:52<18:41:22, 5.76s/it]
73%|███████▎ | 32257/43939 [52:01:58<18:39:24, 5.75s/it]
73%|███████▎ | 32258/43939 [52:02:03<18:39:09, 5.75s/it]
73%|███████▎ | 32259/43939 [52:02:09<18:36:04, 5.73s/it]
73%|███████▎ | 32260/43939 [52:02:15<18:34:09, 5.72s/it]
73%|███████▎ | 32261/43939 [52:02:21<18:34:25, 5.73s/it]
73%|███████▎ | 32262/43939 [52:02:26<18:30:33, 5.71s/it]
73%|███████▎ | 32263/43939 [52:02:32<18:29:36, 5.70s/it]
73%|███████▎ | 32264/43939 [52:02:38<18:31:44, 5.71s/it]
73%|███████▎ | 32265/43939 [52:02:43<18:31:25, 5.71s/it]
73%|███████▎ | 32266/43939 [52:02:49<18:32:44, 5.72s/it]
73%|███████▎ | 32267/43939 [52:02:55<18:30:00, 5.71s/it]
73%|███████▎ | 32268/43939 [52:03:00<18:29:30, 5.70s/it]
73%|███████▎ | 32269/43939 [52:03:06<18:27:29, 5.69s/it]
73%|███████▎ | 32270/43939 [52:03:12<18:27:33, 5.69s/it]
73%|███████▎ | 32271/43939 [52:03:18<18:25:11, 5.68s/it]
73%|███████▎ | 32272/43939 [52:03:23<18:26:40, 5.69s/it]
73%|███████▎ | 32273/43939 [52:03:29<18:26:22, 5.69s/it]
73%|███████▎ | 32274/43939 [52:03:35<18:27:03, 5.69s/it]
73%|███████▎ | 32275/43939 [52:03:40<18:27:02, 5.69s/it]
73%|███████▎ | 32276/43939 [52:03:46<18:40:07, 5.76s/it]
73%|███████▎ | 32277/43939 [52:03:52<18:35:40, 5.74s/it]
73%|███████▎ | 32278/43939 [52:03:58<18:31:05, 5.72s/it]
73%|███████▎ | 32279/43939 [52:04:03<18:30:44, 5.72s/it]
73%|███████▎ | 32280/43939 [52:04:09<18:29:33, 5.71s/it]
73%|███████▎ | 32281/43939 [52:04:15<18:28:49, 5.71s/it]
73%|███████▎ | 32282/43939 [52:04:20<18:30:49, 5.72s/it]
73%|███████▎ | 32283/43939 [52:04:26<18:29:13, 5.71s/it]
73%|███████▎ | 32284/43939 [52:04:32<18:28:10, 5.70s/it]
73%|███████▎ | 32285/43939 [52:04:38<18:28:33, 5.71s/it]
73%|███████▎ | 32286/43939 [52:04:43<18:29:19, 5.71s/it]
73%|███████▎ | 32287/43939 [52:04:49<18:28:52, 5.71s/it]
73%|███████▎ | 32288/43939 [52:04:55<18:27:23, 5.70s/it]
73%|███████▎ | 32289/43939 [52:05:00<18:27:25, 5.70s/it]
73%|███████▎ | 32290/43939 [52:05:06<18:26:18, 5.70s/it]
73%|███████▎ | 32291/43939 [52:05:12<18:29:27, 5.71s/it]
73%|███████▎ | 32292/43939 [52:05:18<18:29:11, 5.71s/it]
73%|███████▎ | 32293/43939 [52:05:23<18:31:21, 5.73s/it]
73%|███████▎ | 32294/43939 [52:05:29<18:28:44, 5.71s/it]
73%|███████▎ | 32295/43939 [52:05:35<18:28:30, 5.71s/it]
74%|███████▎ | 32296/43939 [52:05:40<18:28:34, 5.71s/it]
74%|███████▎ | 32297/43939 [52:05:46<18:39:08, 5.77s/it]
74%|███████▎ | 32298/43939 [52:05:52<18:35:29, 5.75s/it]
74%|███████▎ | 32299/43939 [52:05:58<18:32:12, 5.73s/it]
74%|███████▎ | 32300/43939 [52:06:03<18:27:48, 5.71s/it]
{'loss': 0.7619, 'grad_norm': 1.5587745904922485, 'learning_rate': 9.948122128953335e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.6, 'epoch': 2.94}
74%|███████▎ | 32300/43939 [52:06:03<18:27:48, 5.71s/it]
74%|███████▎ | 32301/43939 [52:06:09<18:27:02, 5.71s/it]
74%|███████▎ | 32302/43939 [52:06:15<18:26:19, 5.70s/it]
74%|███████▎ | 32303/43939 [52:06:20<18:28:49, 5.72s/it]
74%|███████▎ | 32304/43939 [52:06:26<18:26:51, 5.71s/it]
74%|███████▎ | 32305/43939 [52:06:32<18:24:49, 5.70s/it]
74%|███████▎ | 32306/43939 [52:06:38<18:23:33, 5.69s/it]
74%|███████▎ | 32307/43939 [52:06:43<18:23:46, 5.69s/it]
74%|███████▎ | 32308/43939 [52:06:49<18:25:22, 5.70s/it]
74%|███████▎ | 32309/43939 [52:06:55<18:23:28, 5.69s/it]
74%|███████▎ | 32310/43939 [52:07:00<18:22:56, 5.69s/it]
74%|███████▎ | 32311/43939 [52:07:06<18:23:55, 5.70s/it]
74%|███████▎ | 32312/43939 [52:07:12<18:26:02, 5.71s/it]
74%|███████▎ | 32313/43939 [52:07:17<18:24:55, 5.70s/it]
74%|███████▎ | 32314/43939 [52:07:23<18:24:42, 5.70s/it]
74%|███████▎ | 32315/43939 [52:07:29<18:25:56, 5.71s/it]
74%|███████▎ | 32316/43939 [52:07:35<18:23:53, 5.70s/it]
74%|███████▎ | 32317/43939 [52:07:40<18:23:53, 5.70s/it]
74%|███████▎ | 32318/43939 [52:07:46<18:23:03, 5.70s/it]
74%|███████▎ | 32319/43939 [52:07:52<18:41:35, 5.79s/it]
74%|███████▎ | 32320/43939 [52:07:58<18:37:10, 5.77s/it]
74%|███████▎ | 32321/43939 [52:08:03<18:31:35, 5.74s/it]
74%|███████▎ | 32322/43939 [52:08:09<18:29:42, 5.73s/it]
74%|███████▎ | 32323/43939 [52:08:15<18:30:07, 5.73s/it]
74%|███████▎ | 32324/43939 [52:08:20<18:26:15, 5.71s/it]
74%|███████▎ | 32325/43939 [52:08:26<18:27:29, 5.72s/it]
74%|███████▎ | 32326/43939 [52:08:32<18:27:17, 5.72s/it]
74%|███████▎ | 32327/43939 [52:08:38<18:28:19, 5.73s/it]
74%|███████▎ | 32328/43939 [52:08:43<18:29:09, 5.73s/it]
74%|███████▎ | 32329/43939 [52:08:49<18:28:28, 5.73s/it]
74%|███████▎ | 32330/43939 [52:08:55<18:26:15, 5.72s/it]
74%|███████▎ | 32331/43939 [52:09:00<18:24:44, 5.71s/it]
74%|███████▎ | 32332/43939 [52:09:06<18:24:34, 5.71s/it]
74%|███████▎ | 32333/43939 [52:09:12<18:27:19, 5.72s/it]
74%|███████▎ | 32334/43939 [52:09:18<18:23:49, 5.71s/it]
74%|███████▎ | 32335/43939 [52:09:23<18:24:22, 5.71s/it]
74%|███████▎ | 32336/43939 [52:09:29<18:22:45, 5.70s/it]
74%|███████▎ | 32337/43939 [52:09:35<18:19:54, 5.69s/it]
74%|███████▎ | 32338/43939 [52:09:40<18:22:27, 5.70s/it]
74%|███████▎ | 32339/43939 [52:09:46<18:22:34, 5.70s/it]
74%|███████▎ | 32340/43939 [52:09:52<18:20:53, 5.69s/it]
74%|███████▎ | 32341/43939 [52:09:58<18:34:54, 5.77s/it]
74%|███████▎ | 32342/43939 [52:10:03<18:31:37, 5.75s/it]
74%|███████▎ | 32343/43939 [52:10:09<18:30:47, 5.75s/it]
74%|███████▎ | 32344/43939 [52:10:15<18:28:21, 5.74s/it]
74%|███████▎ | 32345/43939 [52:10:21<18:26:33, 5.73s/it]
74%|███████▎ | 32346/43939 [52:10:26<18:25:04, 5.72s/it]
74%|███████▎ | 32347/43939 [52:10:32<18:23:12, 5.71s/it]
74%|███████▎ | 32348/43939 [52:10:38<18:24:08, 5.72s/it]
74%|███████▎ | 32349/43939 [52:10:43<18:25:40, 5.72s/it]
74%|███████▎ | 32350/43939 [52:10:49<18:26:37, 5.73s/it]
{'loss': 0.7431, 'grad_norm': 1.40702223777771, 'learning_rate': 9.868954565157854e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.39, 'epoch': 2.94}
74%|███████▎ | 32350/43939 [52:10:49<18:26:37, 5.73s/it]
74%|███████▎ | 32351/43939 [52:10:55<18:25:47, 5.73s/it]
74%|███████▎ | 32352/43939 [52:11:01<18:25:48, 5.73s/it]
74%|███████▎ | 32353/43939 [52:11:06<18:23:05, 5.71s/it]
74%|███████▎ | 32354/43939 [52:11:12<18:22:57, 5.71s/it]
74%|███████▎ | 32355/43939 [52:11:18<18:22:40, 5.71s/it]
74%|███████▎ | 32356/43939 [52:11:23<18:21:38, 5.71s/it]
74%|███████▎ | 32357/43939 [52:11:29<18:20:52, 5.70s/it]
74%|███████▎ | 32358/43939 [52:11:35<18:19:52, 5.70s/it]
74%|███████▎ | 32359/43939 [52:11:41<18:21:04, 5.71s/it]
74%|███████▎ | 32360/43939 [52:11:46<18:21:14, 5.71s/it]
74%|███████▎ | 32361/43939 [52:11:52<18:21:28, 5.71s/it]
74%|███████▎ | 32362/43939 [52:11:58<18:33:39, 5.77s/it]
74%|███████▎ | 32363/43939 [52:12:04<18:30:27, 5.76s/it]
74%|███████▎ | 32364/43939 [52:12:09<18:30:26, 5.76s/it]
74%|███████▎ | 32365/43939 [52:12:15<18:28:49, 5.75s/it]
74%|███████▎ | 32366/43939 [52:12:21<18:28:45, 5.75s/it]
74%|███████▎ | 32367/43939 [52:12:26<18:23:51, 5.72s/it]
74%|███████▎ | 32368/43939 [52:12:32<18:23:13, 5.72s/it]
74%|███████▎ | 32369/43939 [52:12:38<18:22:12, 5.72s/it]
74%|███████▎ | 32370/43939 [52:12:44<18:21:09, 5.71s/it]
74%|███████▎ | 32371/43939 [52:12:49<18:21:18, 5.71s/it]
74%|███████▎ | 32372/43939 [52:12:55<18:20:53, 5.71s/it]
74%|███████▎ | 32373/43939 [52:13:01<18:19:43, 5.70s/it]
74%|███████▎ | 32374/43939 [52:13:06<18:19:05, 5.70s/it]
74%|███████▎ | 32375/43939 [52:13:12<18:18:38, 5.70s/it]
74%|███████▎ | 32376/43939 [52:13:18<18:19:54, 5.71s/it]
74%|███████▎ | 32377/43939 [52:13:24<18:18:52, 5.70s/it]
74%|███████▎ | 32378/43939 [52:13:29<18:17:46, 5.70s/it]
74%|███████▎ | 32379/43939 [52:13:35<18:22:49, 5.72s/it]
74%|███████▎ | 32380/43939 [52:13:41<18:22:20, 5.72s/it]
74%|███████▎ | 32381/43939 [52:13:46<18:19:22, 5.71s/it]
74%|███████▎ | 32382/43939 [52:13:52<18:21:22, 5.72s/it]
74%|███████▎ | 32383/43939 [52:13:58<18:33:25, 5.78s/it]
74%|███████▎ | 32384/43939 [52:14:04<18:29:06, 5.76s/it]
74%|███████▎ | 32385/43939 [52:14:09<18:26:14, 5.74s/it]
74%|███████▎ | 32386/43939 [52:14:15<18:26:02, 5.74s/it]
74%|███████▎ | 32387/43939 [52:14:21<18:22:58, 5.73s/it]
74%|███████▎ | 32388/43939 [52:14:27<18:21:17, 5.72s/it]
74%|███████▎ | 32389/43939 [52:14:32<18:22:47, 5.73s/it]
74%|███████▎ | 32390/43939 [52:14:38<18:20:05, 5.72s/it]
74%|███████▎ | 32391/43939 [52:14:44<18:21:05, 5.72s/it]
74%|███████▎ | 32392/43939 [52:14:50<18:21:01, 5.72s/it]
74%|███████▎ | 32393/43939 [52:14:55<18:22:13, 5.73s/it]
74%|███████▎ | 32394/43939 [52:15:01<18:22:20, 5.73s/it]
74%|███████▎ | 32395/43939 [52:15:07<18:20:47, 5.72s/it]
74%|███████▎ | 32396/43939 [52:15:12<18:20:41, 5.72s/it]
74%|███████▎ | 32397/43939 [52:15:18<18:19:31, 5.72s/it]
74%|███████▎ | 32398/43939 [52:15:24<18:18:19, 5.71s/it]
74%|███████▎ | 32399/43939 [52:15:30<18:19:08, 5.71s/it]
74%|███████▎ | 32400/43939 [52:15:35<18:18:32, 5.71s/it]
{'loss': 0.7828, 'grad_norm': 1.499337077140808, 'learning_rate': 9.790025729161583e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.84, 'epoch': 2.95}
74%|███████▎ | 32400/43939 [52:15:35<18:18:32, 5.71s/it]
74%|███████▎ | 32401/43939 [52:15:41<18:17:37, 5.71s/it]
74%|███████▎ | 32402/43939 [52:15:47<18:16:24, 5.70s/it]
74%|███████▎ | 32403/43939 [52:15:52<18:22:55, 5.74s/it]
74%|███████▎ | 32404/43939 [52:15:58<18:36:44, 5.81s/it]
74%|███████▎ | 32405/43939 [52:16:04<18:31:30, 5.78s/it]
74%|███████▍ | 32406/43939 [52:16:10<18:25:49, 5.75s/it]
74%|███████▍ | 32407/43939 [52:16:16<18:22:21, 5.74s/it]
74%|███████▍ | 32408/43939 [52:16:21<18:20:41, 5.73s/it]
74%|███████▍ | 32409/43939 [52:16:27<18:17:09, 5.71s/it]
74%|███████▍ | 32410/43939 [52:16:33<18:19:52, 5.72s/it]
74%|███████▍ | 32411/43939 [52:16:38<18:19:09, 5.72s/it]
74%|███████▍ | 32412/43939 [52:16:44<18:19:28, 5.72s/it]
74%|███████▍ | 32413/43939 [52:16:50<18:17:09, 5.71s/it]
74%|███████▍ | 32414/43939 [52:16:56<18:18:49, 5.72s/it]
74%|███████▍ | 32415/43939 [52:17:01<18:18:35, 5.72s/it]
74%|███████▍ | 32416/43939 [52:17:07<18:16:30, 5.71s/it]
74%|███████▍ | 32417/43939 [52:17:13<18:15:23, 5.70s/it]
74%|███████▍ | 32418/43939 [52:17:18<18:16:09, 5.71s/it]
74%|███████▍ | 32419/43939 [52:17:24<18:14:21, 5.70s/it]
74%|███████▍ | 32420/43939 [52:17:30<18:15:28, 5.71s/it]
74%|███████▍ | 32421/43939 [52:17:35<18:17:09, 5.72s/it]
74%|███████▍ | 32422/43939 [52:17:41<18:19:00, 5.73s/it]
74%|███████▍ | 32423/43939 [52:17:47<18:18:07, 5.72s/it]
74%|███████▍ | 32424/43939 [52:17:53<18:15:35, 5.71s/it]
74%|███████▍ | 32425/43939 [52:17:59<18:35:40, 5.81s/it]
74%|███████▍ | 32426/43939 [52:18:04<18:28:44, 5.78s/it]
74%|███████▍ | 32427/43939 [52:18:10<18:25:35, 5.76s/it]
74%|███████▍ | 32428/43939 [52:18:16<18:19:52, 5.73s/it]
74%|███████▍ | 32429/43939 [52:18:21<18:16:20, 5.72s/it]
74%|███████▍ | 32430/43939 [52:18:27<18:15:45, 5.71s/it]
74%|███████▍ | 32431/43939 [52:18:33<18:16:07, 5.71s/it]
74%|███████▍ | 32432/43939 [52:18:39<18:12:43, 5.70s/it]
74%|███████▍ | 32433/43939 [52:18:44<18:12:23, 5.70s/it]
74%|███████▍ | 32434/43939 [52:18:50<18:11:37, 5.69s/it]
74%|███████▍ | 32435/43939 [52:18:56<18:13:52, 5.71s/it]
74%|███████▍ | 32436/43939 [52:19:01<18:13:12, 5.70s/it]
74%|███████▍ | 32437/43939 [52:19:07<18:15:03, 5.71s/it]
74%|███████▍ | 32438/43939 [52:19:13<18:14:25, 5.71s/it]
74%|███████▍ | 32439/43939 [52:19:18<18:11:02, 5.69s/it]
74%|███████▍ | 32440/43939 [52:19:24<18:13:41, 5.71s/it]
74%|███████▍ | 32441/43939 [52:19:30<18:09:39, 5.69s/it]
74%|███████▍ | 32442/43939 [52:19:36<18:11:39, 5.70s/it]
74%|███████▍ | 32443/43939 [52:19:41<18:14:06, 5.71s/it]
74%|███████▍ | 32444/43939 [52:19:47<18:14:40, 5.71s/it]
74%|███████▍ | 32445/43939 [52:19:53<18:13:30, 5.71s/it]
74%|███████▍ | 32446/43939 [52:19:59<18:25:50, 5.77s/it]
74%|███████▍ | 32447/43939 [52:20:04<18:23:26, 5.76s/it]
74%|███████▍ | 32448/43939 [52:20:10<18:19:14, 5.74s/it]
74%|███████▍ | 32449/43939 [52:20:16<18:15:02, 5.72s/it]
74%|███████▍ | 32450/43939 [52:20:21<18:12:25, 5.71s/it]
{'loss': 0.7842, 'grad_norm': 1.3095389604568481, 'learning_rate': 9.711336866252386e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 518.91, 'epoch': 2.95}
74%|███████▍ | 32450/43939 [52:20:21<18:12:25, 5.71s/it]
74%|███████▍ | 32451/43939 [52:20:27<18:09:17, 5.69s/it]
74%|███████▍ | 32452/43939 [52:20:33<18:09:05, 5.69s/it]
74%|███████▍ | 32453/43939 [52:20:38<18:13:35, 5.71s/it]
74%|███████▍ | 32454/43939 [52:20:44<18:12:50, 5.71s/it]
74%|███████▍ | 32455/43939 [52:20:50<18:09:45, 5.69s/it]
74%|███████▍ | 32456/43939 [52:20:56<18:08:40, 5.69s/it]
74%|███████▍ | 32457/43939 [52:21:01<18:11:05, 5.70s/it]
74%|███████▍ | 32458/43939 [52:21:07<18:11:05, 5.70s/it]
74%|███████▍ | 32459/43939 [52:21:13<18:11:45, 5.71s/it]
74%|███████▍ | 32460/43939 [52:21:18<18:09:49, 5.70s/it]
74%|███████▍ | 32461/43939 [52:21:24<18:08:31, 5.69s/it]
74%|███████▍ | 32462/43939 [52:21:30<18:10:00, 5.70s/it]
74%|███████▍ | 32463/43939 [52:21:35<18:07:41, 5.69s/it]
74%|███████▍ | 32464/43939 [52:21:41<18:08:24, 5.69s/it]
74%|███████▍ | 32465/43939 [52:21:47<18:12:27, 5.71s/it]
74%|███████▍ | 32466/43939 [52:21:53<18:11:15, 5.71s/it]
74%|███████▍ | 32467/43939 [52:21:58<18:07:34, 5.69s/it]
74%|███████▍ | 32468/43939 [52:22:04<18:24:38, 5.78s/it]
74%|███████▍ | 32469/43939 [52:22:10<18:22:46, 5.77s/it]
74%|███████▍ | 32470/43939 [52:22:16<18:20:12, 5.76s/it]
74%|███████▍ | 32471/43939 [52:22:21<18:16:48, 5.74s/it]
74%|███████▍ | 32472/43939 [52:22:27<18:12:38, 5.72s/it]
74%|███████▍ | 32473/43939 [52:22:33<18:11:54, 5.71s/it]
74%|███████▍ | 32474/43939 [52:22:38<18:10:56, 5.71s/it]
74%|███████▍ | 32475/43939 [52:22:44<18:10:49, 5.71s/it]
74%|███████▍ | 32476/43939 [52:22:50<18:10:36, 5.71s/it]
74%|███████▍ | 32477/43939 [52:22:56<18:09:57, 5.71s/it]
74%|███████▍ | 32478/43939 [52:23:01<18:09:29, 5.70s/it]
74%|███████▍ | 32479/43939 [52:23:07<18:09:12, 5.70s/it]
74%|███████▍ | 32480/43939 [52:23:13<18:07:20, 5.69s/it]
74%|███████▍ | 32481/43939 [52:23:18<18:07:29, 5.69s/it]
74%|███████▍ | 32482/43939 [52:23:24<18:08:05, 5.70s/it]
74%|███████▍ | 32483/43939 [52:23:30<18:08:36, 5.70s/it]
74%|███████▍ | 32484/43939 [52:23:35<18:09:02, 5.70s/it]
74%|███████▍ | 32485/43939 [52:23:41<18:10:16, 5.71s/it]
74%|███████▍ | 32486/43939 [52:23:47<18:09:19, 5.71s/it]
74%|███████▍ | 32487/43939 [52:23:53<18:07:06, 5.70s/it]
74%|███████▍ | 32488/43939 [52:23:58<18:06:55, 5.70s/it]
74%|███████▍ | 32489/43939 [52:24:04<18:05:38, 5.69s/it]
74%|███████▍ | 32490/43939 [52:24:10<18:20:48, 5.77s/it]
74%|███████▍ | 32491/43939 [52:24:16<18:18:26, 5.76s/it]
74%|███████▍ | 32492/43939 [52:24:21<18:14:21, 5.74s/it]
74%|███████▍ | 32493/43939 [52:24:27<18:14:03, 5.74s/it]
74%|███████▍ | 32494/43939 [52:24:33<18:12:33, 5.73s/it]
74%|███████▍ | 32495/43939 [52:24:38<18:12:24, 5.73s/it]
74%|███████▍ | 32496/43939 [52:24:44<18:12:29, 5.73s/it]
74%|███████▍ | 32497/43939 [52:24:50<18:10:56, 5.72s/it]
74%|███████▍ | 32498/43939 [52:24:56<18:10:06, 5.72s/it]
74%|███████▍ | 32499/43939 [52:25:01<18:07:40, 5.70s/it]
74%|███████▍ | 32500/43939 [52:25:07<18:06:18, 5.70s/it]
{'loss': 0.7597, 'grad_norm': 1.4443596601486206, 'learning_rate': 9.632889217931984e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.51, 'epoch': 2.96}
74%|███████▍ | 32500/43939 [52:25:07<18:06:18, 5.70s/it]
74%|███████▍ | 32501/43939 [52:25:13<18:06:19, 5.70s/it]
74%|███████▍ | 32502/43939 [52:25:18<18:07:43, 5.71s/it]
74%|███████▍ | 32503/43939 [52:25:24<18:06:32, 5.70s/it]
74%|███████▍ | 32504/43939 [52:25:30<18:09:14, 5.72s/it]
74%|███████▍ | 32505/43939 [52:25:36<18:08:59, 5.71s/it]
74%|███████▍ | 32506/43939 [52:25:41<18:08:54, 5.71s/it]
74%|███████▍ | 32507/43939 [52:25:47<18:08:13, 5.71s/it]
74%|███████▍ | 32508/43939 [52:25:53<18:07:15, 5.71s/it]
74%|███████▍ | 32509/43939 [52:25:58<18:06:33, 5.70s/it]
74%|███████▍ | 32510/43939 [52:26:04<18:06:42, 5.71s/it]
74%|███████▍ | 32511/43939 [52:26:10<18:18:53, 5.77s/it]
74%|███████▍ | 32512/43939 [52:26:16<18:17:32, 5.76s/it]
74%|███████▍ | 32513/43939 [52:26:21<18:14:54, 5.75s/it]
74%|███████▍ | 32514/43939 [52:26:27<18:12:11, 5.74s/it]
74%|███████▍ | 32515/43939 [52:26:33<18:08:51, 5.72s/it]
74%|███████▍ | 32516/43939 [52:26:38<18:07:58, 5.71s/it]
74%|███████▍ | 32517/43939 [52:26:44<18:06:00, 5.70s/it]
74%|███████▍ | 32518/43939 [52:26:50<18:07:01, 5.71s/it]
74%|███████▍ | 32519/43939 [52:26:56<18:08:13, 5.72s/it]
74%|███████▍ | 32520/43939 [52:27:01<18:06:20, 5.71s/it]
74%|███████▍ | 32521/43939 [52:27:07<18:07:01, 5.71s/it]
74%|███████▍ | 32522/43939 [52:27:13<18:06:36, 5.71s/it]
74%|███████▍ | 32523/43939 [52:27:18<18:04:41, 5.70s/it]
74%|███████▍ | 32524/43939 [52:27:24<18:05:23, 5.71s/it]
74%|███████▍ | 32525/43939 [52:27:30<18:06:36, 5.71s/it]
74%|███████▍ | 32526/43939 [52:27:36<18:06:20, 5.71s/it]
74%|███████▍ | 32527/43939 [52:27:41<18:03:40, 5.70s/it]
74%|███████▍ | 32528/43939 [52:27:47<18:06:03, 5.71s/it]
74%|███████▍ | 32529/43939 [52:27:53<18:06:09, 5.71s/it]
74%|███████▍ | 32530/43939 [52:27:58<18:03:09, 5.70s/it]
74%|███████▍ | 32531/43939 [52:28:04<18:02:50, 5.70s/it]
74%|███████▍ | 32532/43939 [52:28:10<18:16:11, 5.77s/it]
74%|███████▍ | 32533/43939 [52:28:16<18:12:07, 5.75s/it]
74%|███████▍ | 32534/43939 [52:28:21<18:08:07, 5.72s/it]
74%|███████▍ | 32535/43939 [52:28:27<18:06:06, 5.71s/it]
74%|███████▍ | 32536/43939 [52:28:33<18:08:42, 5.73s/it]
74%|███████▍ | 32537/43939 [52:28:38<18:04:29, 5.71s/it]
74%|███████▍ | 32538/43939 [52:28:44<18:04:22, 5.71s/it]
74%|███████▍ | 32539/43939 [52:28:50<18:05:35, 5.71s/it]
74%|███████▍ | 32540/43939 [52:28:56<18:04:38, 5.71s/it]
74%|███████▍ | 32541/43939 [52:29:01<18:04:36, 5.71s/it]
74%|███████▍ | 32542/43939 [52:29:07<18:02:33, 5.70s/it]
74%|███████▍ | 32543/43939 [52:29:13<18:03:52, 5.71s/it]
74%|███████▍ | 32544/43939 [52:29:18<18:04:12, 5.71s/it]
74%|███████▍ | 32545/43939 [52:29:24<18:04:43, 5.71s/it]
74%|███████▍ | 32546/43939 [52:29:30<18:07:11, 5.73s/it]
74%|███████▍ | 32547/43939 [52:29:36<18:07:13, 5.73s/it]
74%|███████▍ | 32548/43939 [52:29:41<18:07:10, 5.73s/it]
74%|███████▍ | 32549/43939 [52:29:47<18:05:19, 5.72s/it]
74%|███████▍ | 32550/43939 [52:29:53<18:09:10, 5.74s/it]
{'loss': 0.7748, 'grad_norm': 1.2316476106643677, 'learning_rate': 9.554684021896388e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 511.13, 'epoch': 2.96}
74%|███████▍ | 32550/43939 [52:29:53<18:09:10, 5.74s/it]
74%|███████▍ | 32551/43939 [52:29:59<18:05:55, 5.72s/it]
74%|███████▍ | 32552/43939 [52:30:04<18:04:52, 5.72s/it]
74%|███████▍ | 32553/43939 [52:30:10<18:16:43, 5.78s/it]
74%|███████▍ | 32554/43939 [52:30:16<18:14:47, 5.77s/it]
74%|███████▍ | 32555/43939 [52:30:22<18:13:41, 5.76s/it]
74%|███████▍ | 32556/43939 [52:30:27<18:09:50, 5.74s/it]
74%|███████▍ | 32557/43939 [52:30:33<18:06:46, 5.73s/it]
74%|███████▍ | 32558/43939 [52:30:39<18:06:48, 5.73s/it]
74%|███████▍ | 32559/43939 [52:30:44<18:05:43, 5.72s/it]
74%|███████▍ | 32560/43939 [52:30:50<18:02:59, 5.71s/it]
74%|███████▍ | 32561/43939 [52:30:56<18:01:50, 5.70s/it]
74%|███████▍ | 32562/43939 [52:31:02<17:59:29, 5.69s/it]
74%|███████▍ | 32563/43939 [52:31:07<18:00:29, 5.70s/it]
74%|███████▍ | 32564/43939 [52:31:13<17:58:10, 5.69s/it]
74%|███████▍ | 32565/43939 [52:31:19<18:00:41, 5.70s/it]
74%|███████▍ | 32566/43939 [52:31:24<17:58:16, 5.69s/it]
74%|███████▍ | 32567/43939 [52:31:30<18:00:23, 5.70s/it]
74%|███████▍ | 32568/43939 [52:31:36<18:00:26, 5.70s/it]
74%|███████▍ | 32569/43939 [52:31:41<18:01:12, 5.71s/it]
74%|███████▍ | 32570/43939 [52:31:47<17:59:16, 5.70s/it]
74%|███████▍ | 32571/43939 [52:31:53<18:02:38, 5.71s/it]
74%|███████▍ | 32572/43939 [52:31:59<18:04:20, 5.72s/it]
74%|███████▍ | 32573/43939 [52:32:04<18:02:59, 5.72s/it]
74%|███████▍ | 32574/43939 [52:32:10<18:19:13, 5.80s/it]
74%|███████▍ | 32575/43939 [52:32:16<18:12:04, 5.77s/it]
74%|███████▍ | 32576/43939 [52:32:22<18:11:34, 5.76s/it]
74%|███████▍ | 32577/43939 [52:32:28<18:10:26, 5.76s/it]
74%|███████▍ | 32578/43939 [52:32:33<18:07:15, 5.74s/it]
74%|███████▍ | 32579/43939 [52:32:39<18:05:17, 5.73s/it]
74%|███████▍ | 32580/43939 [52:32:45<18:04:21, 5.73s/it]
74%|███████▍ | 32581/43939 [52:32:50<18:02:24, 5.72s/it]
74%|███████▍ | 32582/43939 [52:32:56<18:00:35, 5.71s/it]
74%|███████▍ | 32583/43939 [52:33:02<18:02:18, 5.72s/it]
74%|███████▍ | 32584/43939 [52:33:07<18:01:27, 5.71s/it]
74%|███████▍ | 32585/43939 [52:33:13<18:00:09, 5.71s/it]
74%|███████▍ | 32586/43939 [52:33:19<17:59:37, 5.71s/it]
74%|███████▍ | 32587/43939 [52:33:25<17:58:49, 5.70s/it]
74%|███████▍ | 32588/43939 [52:33:30<18:00:53, 5.71s/it]
74%|███████▍ | 32589/43939 [52:33:36<17:58:38, 5.70s/it]
74%|███████▍ | 32590/43939 [52:33:42<17:59:05, 5.70s/it]
74%|███████▍ | 32591/43939 [52:33:47<17:58:50, 5.70s/it]
74%|███████▍ | 32592/43939 [52:33:53<17:58:10, 5.70s/it]
74%|███████▍ | 32593/43939 [52:33:59<17:56:32, 5.69s/it]
74%|███████▍ | 32594/43939 [52:34:04<17:56:57, 5.70s/it]
74%|███████▍ | 32595/43939 [52:34:10<18:10:54, 5.77s/it]
74%|███████▍ | 32596/43939 [52:34:16<18:07:28, 5.75s/it]
74%|███████▍ | 32597/43939 [52:34:22<18:06:44, 5.75s/it]
74%|███████▍ | 32598/43939 [52:34:28<18:07:25, 5.75s/it]
74%|███████▍ | 32599/43939 [52:34:33<18:04:14, 5.74s/it]
74%|███████▍ | 32600/43939 [52:34:39<18:01:54, 5.72s/it]
{'loss': 0.7907, 'grad_norm': 1.3004258871078491, 'learning_rate': 9.47672251201634e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 509.13, 'epoch': 2.97}
74%|███████▍ | 32600/43939 [52:34:39<18:01:54, 5.72s/it]
74%|███████▍ | 32601/43939 [52:34:45<18:03:39, 5.73s/it]
74%|███████▍ | 32602/43939 [52:34:50<18:00:57, 5.72s/it]
74%|███████▍ | 32603/43939 [52:34:56<17:59:42, 5.71s/it]
74%|███████▍ | 32604/43939 [52:35:02<17:56:55, 5.70s/it]
74%|███████▍ | 32605/43939 [52:35:08<17:56:01, 5.70s/it]
74%|███████▍ | 32606/43939 [52:35:13<17:54:39, 5.69s/it]
74%|███████▍ | 32607/43939 [52:35:19<17:54:37, 5.69s/it]
74%|███████▍ | 32608/43939 [52:35:25<17:57:08, 5.70s/it]
74%|███████▍ | 32609/43939 [52:35:30<17:56:46, 5.70s/it]
74%|███████▍ | 32610/43939 [52:35:36<17:56:37, 5.70s/it]
74%|███████▍ | 32611/43939 [52:35:42<17:54:53, 5.69s/it]
74%|███████▍ | 32612/43939 [52:35:47<17:54:44, 5.69s/it]
74%|███████▍ | 32613/43939 [52:35:53<17:52:52, 5.68s/it]
74%|███████▍ | 32614/43939 [52:35:59<17:53:14, 5.69s/it]
74%|███████▍ | 32615/43939 [52:36:04<17:54:26, 5.69s/it]
74%|███████▍ | 32616/43939 [52:36:10<17:55:24, 5.70s/it]
74%|███████▍ | 32617/43939 [52:36:16<18:10:04, 5.78s/it]
74%|███████▍ | 32618/43939 [52:36:22<18:06:38, 5.76s/it]
74%|███████▍ | 32619/43939 [52:36:28<18:06:10, 5.76s/it]
74%|███████▍ | 32620/43939 [52:36:33<18:03:50, 5.75s/it]
74%|███████▍ | 32621/43939 [52:36:39<18:00:47, 5.73s/it]
74%|███████▍ | 32622/43939 [52:36:45<18:00:20, 5.73s/it]
74%|███████▍ | 32623/43939 [52:36:50<17:59:47, 5.73s/it]
74%|███████▍ | 32624/43939 [52:36:56<17:58:36, 5.72s/it]
74%|███████▍ | 32625/43939 [52:37:02<17:57:42, 5.72s/it]
74%|███████▍ | 32626/43939 [52:37:08<17:58:26, 5.72s/it]
74%|███████▍ | 32627/43939 [52:37:13<18:01:33, 5.74s/it]
74%|███████▍ | 32628/43939 [52:37:19<17:59:30, 5.73s/it]
74%|███████▍ | 32629/43939 [52:37:25<17:57:56, 5.72s/it]
74%|███████▍ | 32630/43939 [52:37:30<17:57:21, 5.72s/it]
74%|███████▍ | 32631/43939 [52:37:36<17:56:50, 5.71s/it]
74%|███████▍ | 32632/43939 [52:37:42<17:55:02, 5.70s/it]
74%|███████▍ | 32633/43939 [52:37:48<17:53:43, 5.70s/it]
74%|███████▍ | 32634/43939 [52:37:53<17:56:49, 5.72s/it]
74%|███████▍ | 32635/43939 [52:37:59<17:56:44, 5.72s/it]
74%|███████▍ | 32636/43939 [52:38:05<17:57:06, 5.72s/it]
74%|███████▍ | 32637/43939 [52:38:10<17:55:19, 5.71s/it]
74%|███████▍ | 32638/43939 [52:38:16<17:56:24, 5.71s/it]
74%|███████▍ | 32639/43939 [52:38:22<18:11:03, 5.79s/it]
74%|███████▍ | 32640/43939 [52:38:28<18:06:31, 5.77s/it]
74%|███████▍ | 32641/43939 [52:38:34<18:04:28, 5.76s/it]
74%|███████▍ | 32642/43939 [52:38:39<18:00:54, 5.74s/it]
74%|███████▍ | 32643/43939 [52:38:45<17:59:06, 5.73s/it]
74%|███████▍ | 32644/43939 [52:38:51<17:58:35, 5.73s/it]
74%|███████▍ | 32645/43939 [52:38:56<18:00:35, 5.74s/it]
74%|███████▍ | 32646/43939 [52:39:02<18:00:41, 5.74s/it]
74%|███████▍ | 32647/43939 [52:39:08<17:59:46, 5.74s/it]
74%|███████▍ | 32648/43939 [52:39:14<17:54:59, 5.71s/it]
74%|███████▍ | 32649/43939 [52:39:19<17:54:28, 5.71s/it]
74%|███████▍ | 32650/43939 [52:39:25<17:56:10, 5.72s/it]
{'loss': 0.7514, 'grad_norm': 1.1468687057495117, 'learning_rate': 9.399005918317866e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 477.26, 'epoch': 2.97}
74%|███████▍ | 32650/43939 [52:39:25<17:56:10, 5.72s/it]
74%|███████▍ | 32651/43939 [52:39:31<17:56:30, 5.72s/it]
74%|███████▍ | 32652/43939 [52:39:36<17:55:54, 5.72s/it]
74%|███████▍ | 32653/43939 [52:39:42<17:55:16, 5.72s/it]
74%|███████▍ | 32654/43939 [52:39:48<17:53:48, 5.71s/it]
74%|███████▍ | 32655/43939 [52:39:54<17:55:07, 5.72s/it]
74%|███████▍ | 32656/43939 [52:39:59<17:55:30, 5.72s/it]
74%|███████▍ | 32657/43939 [52:40:05<17:53:11, 5.71s/it]
74%|███████▍ | 32658/43939 [52:40:11<17:52:00, 5.70s/it]
74%|███████▍ | 32659/43939 [52:40:16<17:51:05, 5.70s/it]
74%|███████▍ | 32660/43939 [52:40:22<18:04:37, 5.77s/it]
74%|███████▍ | 32661/43939 [52:40:28<17:59:37, 5.74s/it]
74%|███████▍ | 32662/43939 [52:40:34<17:56:17, 5.73s/it]
74%|███████▍ | 32663/43939 [52:40:39<17:55:38, 5.72s/it]
74%|███████▍ | 32664/43939 [52:40:45<17:57:01, 5.73s/it]
74%|███████▍ | 32665/43939 [52:40:51<17:52:26, 5.71s/it]
74%|███████▍ | 32666/43939 [52:40:57<17:52:21, 5.71s/it]
74%|███████▍ | 32667/43939 [52:41:02<17:53:25, 5.71s/it]
74%|███████▍ | 32668/43939 [52:41:08<17:50:10, 5.70s/it]
74%|███████▍ | 32669/43939 [52:41:14<17:54:27, 5.72s/it]
74%|███████▍ | 32670/43939 [52:41:19<17:54:59, 5.72s/it]
74%|███████▍ | 32671/43939 [52:41:25<17:53:37, 5.72s/it]
74%|███████▍ | 32672/43939 [52:41:31<17:51:25, 5.71s/it]
74%|███████▍ | 32673/43939 [52:41:36<17:48:34, 5.69s/it]
74%|███████▍ | 32674/43939 [52:41:42<17:48:28, 5.69s/it]
74%|███████▍ | 32675/43939 [52:41:48<17:49:10, 5.70s/it]
74%|███████▍ | 32676/43939 [52:41:54<17:53:20, 5.72s/it]
74%|███████▍ | 32677/43939 [52:41:59<17:53:31, 5.72s/it]
74%|███████▍ | 32678/43939 [52:42:05<17:52:26, 5.71s/it]
74%|███████▍ | 32679/43939 [52:42:11<17:51:14, 5.71s/it]
74%|███████▍ | 32680/43939 [52:42:16<17:51:51, 5.71s/it]
74%|███████▍ | 32681/43939 [52:42:22<18:04:33, 5.78s/it]
74%|███████▍ | 32682/43939 [52:42:28<17:58:54, 5.75s/it]
74%|███████▍ | 32683/43939 [52:42:34<17:55:18, 5.73s/it]
74%|███████▍ | 32684/43939 [52:42:39<17:53:14, 5.72s/it]
74%|███████▍ | 32685/43939 [52:42:45<17:50:12, 5.71s/it]
74%|███████▍ | 32686/43939 [52:42:51<17:48:13, 5.70s/it]
74%|███████▍ | 32687/43939 [52:42:57<17:48:57, 5.70s/it]
74%|███████▍ | 32688/43939 [52:43:02<17:49:43, 5.70s/it]
74%|███████▍ | 32689/43939 [52:43:08<17:51:19, 5.71s/it]
74%|███████▍ | 32690/43939 [52:43:14<17:51:36, 5.72s/it]
74%|███████▍ | 32691/43939 [52:43:19<17:51:53, 5.72s/it]
74%|███████▍ | 32692/43939 [52:43:25<17:51:56, 5.72s/it]
74%|███████▍ | 32693/43939 [52:43:31<17:50:45, 5.71s/it]
74%|███████▍ | 32694/43939 [52:43:37<17:51:41, 5.72s/it]
74%|███████▍ | 32695/43939 [52:43:42<17:53:04, 5.73s/it]
74%|███████▍ | 32696/43939 [52:43:48<17:54:32, 5.73s/it]
74%|███████▍ | 32697/43939 [52:43:54<17:51:34, 5.72s/it]
74%|███████▍ | 32698/43939 [52:43:59<17:50:06, 5.71s/it]
74%|███████▍ | 32699/43939 [52:44:05<17:48:23, 5.70s/it]
74%|███████▍ | 32700/43939 [52:44:11<17:47:51, 5.70s/it]
{'loss': 0.7709, 'grad_norm': 1.4267300367355347, 'learning_rate': 9.321535466962861e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.3, 'epoch': 2.98}
74%|███████▍ | 32700/43939 [52:44:11<17:47:51, 5.70s/it]
74%|███████▍ | 32701/43939 [52:44:17<17:47:50, 5.70s/it]
74%|███████▍ | 32702/43939 [52:44:22<18:02:00, 5.78s/it]
74%|███████▍ | 32703/43939 [52:44:28<18:00:08, 5.77s/it]
74%|███████▍ | 32704/43939 [52:44:34<17:57:23, 5.75s/it]
74%|███████▍ | 32705/43939 [52:44:40<17:53:37, 5.73s/it]
74%|███████▍ | 32706/43939 [52:44:45<17:50:55, 5.72s/it]
74%|███████▍ | 32707/43939 [52:44:51<17:50:21, 5.72s/it]
74%|███████▍ | 32708/43939 [52:44:57<17:50:18, 5.72s/it]
74%|███████▍ | 32709/43939 [52:45:02<17:51:25, 5.72s/it]
74%|███████▍ | 32710/43939 [52:45:08<17:51:57, 5.73s/it]
74%|███████▍ | 32711/43939 [52:45:14<17:50:09, 5.72s/it]
74%|███████▍ | 32712/43939 [52:45:20<17:50:15, 5.72s/it]
74%|███████▍ | 32713/43939 [52:45:25<17:48:01, 5.71s/it]
74%|███████▍ | 32714/43939 [52:45:31<17:45:47, 5.70s/it]
74%|███████▍ | 32715/43939 [52:45:37<17:47:29, 5.71s/it]
74%|███████▍ | 32716/43939 [52:45:42<17:46:27, 5.70s/it]
74%|███████▍ | 32717/43939 [52:45:48<17:46:41, 5.70s/it]
74%|███████▍ | 32718/43939 [52:45:54<17:47:11, 5.71s/it]
74%|███████▍ | 32719/43939 [52:46:00<17:46:59, 5.71s/it]
74%|███████▍ | 32720/43939 [52:46:05<17:46:20, 5.70s/it]
74%|███████▍ | 32721/43939 [52:46:11<17:45:11, 5.70s/it]
74%|███████▍ | 32722/43939 [52:46:17<17:42:58, 5.69s/it]
74%|███████▍ | 32723/43939 [52:46:23<18:01:03, 5.78s/it]
74%|███████▍ | 32724/43939 [52:46:28<17:58:53, 5.77s/it]
74%|███████▍ | 32725/43939 [52:46:34<17:54:23, 5.75s/it]
74%|███████▍ | 32726/43939 [52:46:40<17:52:15, 5.74s/it]
74%|███████▍ | 32727/43939 [52:46:45<17:51:39, 5.73s/it]
74%|███████▍ | 32728/43939 [52:46:51<17:51:24, 5.73s/it]
74%|███████▍ | 32729/43939 [52:46:57<17:47:49, 5.72s/it]
74%|███████▍ | 32730/43939 [52:47:03<17:47:02, 5.71s/it]
74%|███████▍ | 32731/43939 [52:47:08<17:46:09, 5.71s/it]
74%|███████▍ | 32732/43939 [52:47:14<17:45:42, 5.71s/it]
74%|███████▍ | 32733/43939 [52:47:20<17:45:25, 5.70s/it]
74%|███████▍ | 32734/43939 [52:47:25<17:46:10, 5.71s/it]
75%|███████▍ | 32735/43939 [52:47:31<17:45:01, 5.70s/it]
75%|███████▍ | 32736/43939 [52:47:37<17:43:21, 5.70s/it]
75%|███████▍ | 32737/43939 [52:47:42<17:42:16, 5.69s/it]
75%|███████▍ | 32738/43939 [52:47:48<17:41:11, 5.68s/it]
75%|███████▍ | 32739/43939 [52:47:54<17:41:44, 5.69s/it]
75%|███████▍ | 32740/43939 [52:48:00<17:43:31, 5.70s/it]
75%|███████▍ | 32741/43939 [52:48:05<17:45:10, 5.71s/it]
75%|███████▍ | 32742/43939 [52:48:11<17:45:53, 5.71s/it]
75%|███████▍ | 32743/43939 [52:48:17<17:46:11, 5.71s/it]
75%|███████▍ | 32744/43939 [52:48:23<17:58:10, 5.78s/it]
75%|███████▍ | 32745/43939 [52:48:28<17:53:13, 5.75s/it]
75%|███████▍ | 32746/43939 [52:48:34<17:49:35, 5.73s/it]
75%|███████▍ | 32747/43939 [52:48:40<17:47:51, 5.72s/it]
75%|███████▍ | 32748/43939 [52:48:45<17:48:54, 5.73s/it]
75%|███████▍ | 32749/43939 [52:48:51<17:47:19, 5.72s/it]
75%|███████▍ | 32750/43939 [52:48:57<17:45:18, 5.71s/it]
{'loss': 0.7552, 'grad_norm': 1.6295465230941772, 'learning_rate': 9.244312380229753e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.54, 'epoch': 2.98}
75%|███████▍ | 32750/43939 [52:48:57<17:45:18, 5.71s/it]
75%|███████▍ | 32751/43939 [52:49:03<17:44:37, 5.71s/it]
75%|███████▍ | 32752/43939 [52:49:08<17:44:52, 5.71s/it]
75%|███████▍ | 32753/43939 [52:49:14<17:45:23, 5.71s/it]
75%|███████▍ | 32754/43939 [52:49:20<17:45:29, 5.72s/it]
75%|███████▍ | 32755/43939 [52:49:25<17:45:09, 5.71s/it]
75%|███████▍ | 32756/43939 [52:49:31<17:45:48, 5.72s/it]
75%|███████▍ | 32757/43939 [52:49:37<17:44:39, 5.71s/it]
75%|███████▍ | 32758/43939 [52:49:43<17:43:51, 5.71s/it]
75%|███████▍ | 32759/43939 [52:49:48<17:43:21, 5.71s/it]
75%|███████▍ | 32760/43939 [52:49:54<17:42:50, 5.70s/it]
75%|███████▍ | 32761/43939 [52:50:00<17:41:36, 5.70s/it]
75%|███████▍ | 32762/43939 [52:50:05<17:41:01, 5.70s/it]
75%|███████▍ | 32763/43939 [52:50:11<17:42:33, 5.70s/it]
75%|███████▍ | 32764/43939 [52:50:17<17:43:02, 5.71s/it]
75%|███████▍ | 32765/43939 [52:50:22<17:41:09, 5.70s/it]
75%|███████▍ | 32766/43939 [52:50:28<17:53:15, 5.76s/it]
75%|███████▍ | 32767/43939 [52:50:34<17:50:42, 5.75s/it]
75%|███████▍ | 32768/43939 [52:50:40<17:48:36, 5.74s/it]
75%|███████▍ | 32769/43939 [52:50:45<17:45:59, 5.73s/it]
75%|███████▍ | 32770/43939 [52:50:51<17:44:23, 5.72s/it]
75%|███████▍ | 32771/43939 [52:50:57<17:43:15, 5.71s/it]
75%|███████▍ | 32772/43939 [52:51:03<17:41:34, 5.70s/it]
75%|███████▍ | 32773/43939 [52:51:08<17:41:01, 5.70s/it]
75%|███████▍ | 32774/43939 [52:51:14<17:44:16, 5.72s/it]
75%|███████▍ | 32775/43939 [52:51:20<17:41:43, 5.71s/it]
75%|███████▍ | 32776/43939 [52:51:25<17:46:01, 5.73s/it]
75%|███████▍ | 32777/43939 [52:51:31<17:44:19, 5.72s/it]
75%|███████▍ | 32778/43939 [52:51:37<17:42:24, 5.71s/it]
75%|███████▍ | 32779/43939 [52:51:43<17:42:37, 5.71s/it]
75%|███████▍ | 32780/43939 [52:51:48<17:42:55, 5.72s/it]
75%|███████▍ | 32781/43939 [52:51:54<17:41:09, 5.71s/it]
75%|███████▍ | 32782/43939 [52:52:00<17:40:43, 5.70s/it]
75%|███████▍ | 32783/43939 [52:52:05<17:40:04, 5.70s/it]
75%|███████▍ | 32784/43939 [52:52:11<17:39:17, 5.70s/it]
75%|███████▍ | 32785/43939 [52:52:17<17:39:28, 5.70s/it]
75%|███████▍ | 32786/43939 [52:52:23<17:42:22, 5.72s/it]
75%|███████▍ | 32787/43939 [52:52:28<17:42:40, 5.72s/it]
75%|███████▍ | 32788/43939 [52:52:34<17:52:45, 5.77s/it]
75%|███████▍ | 32789/43939 [52:52:40<17:50:43, 5.76s/it]
75%|███████▍ | 32790/43939 [52:52:46<17:47:40, 5.75s/it]
75%|███████▍ | 32791/43939 [52:52:51<17:43:46, 5.73s/it]
75%|███████▍ | 32792/43939 [52:52:57<17:42:48, 5.72s/it]
75%|███████▍ | 32793/43939 [52:53:03<17:41:52, 5.72s/it]
75%|███████▍ | 32794/43939 [52:53:08<17:40:28, 5.71s/it]
75%|███████▍ | 32795/43939 [52:53:14<17:38:38, 5.70s/it]
75%|███████▍ | 32796/43939 [52:53:20<17:41:04, 5.71s/it]
75%|███████▍ | 32797/43939 [52:53:26<17:41:47, 5.72s/it]
75%|███████▍ | 32798/43939 [52:53:31<17:39:43, 5.71s/it]
75%|███████▍ | 32799/43939 [52:53:37<17:39:55, 5.71s/it]
75%|███████▍ | 32800/43939 [52:53:43<17:37:51, 5.70s/it]
{'loss': 0.7695, 'grad_norm': 1.4877276420593262, 'learning_rate': 9.167337876494206e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.82, 'epoch': 2.99}
75%|███████▍ | 32800/43939 [52:53:43<17:37:51, 5.70s/it]
75%|███████▍ | 32801/43939 [52:53:48<17:38:42, 5.70s/it]
75%|███████▍ | 32802/43939 [52:53:54<17:38:52, 5.70s/it]
75%|███████▍ | 32803/43939 [52:54:00<17:37:15, 5.70s/it]
75%|███████▍ | 32804/43939 [52:54:05<17:37:29, 5.70s/it]
75%|███████▍ | 32805/43939 [52:54:11<17:37:21, 5.70s/it]
75%|███████▍ | 32806/43939 [52:54:17<17:37:37, 5.70s/it]
75%|███████▍ | 32807/43939 [52:54:23<17:37:58, 5.70s/it]
75%|███████▍ | 32808/43939 [52:54:28<17:36:49, 5.70s/it]
75%|███████▍ | 32809/43939 [52:54:34<17:48:40, 5.76s/it]
75%|███████▍ | 32810/43939 [52:54:40<17:47:03, 5.75s/it]
75%|███████▍ | 32811/43939 [52:54:46<17:45:31, 5.75s/it]
75%|███████▍ | 32812/43939 [52:54:51<17:44:49, 5.74s/it]
75%|███████▍ | 32813/43939 [52:54:57<17:41:15, 5.72s/it]
75%|███████▍ | 32814/43939 [52:55:03<17:42:48, 5.73s/it]
75%|███████▍ | 32815/43939 [52:55:08<17:39:53, 5.72s/it]
75%|███████▍ | 32816/43939 [52:55:14<17:39:38, 5.72s/it]
75%|███████▍ | 32817/43939 [52:55:20<17:39:01, 5.71s/it]
75%|███████▍ | 32818/43939 [52:55:26<17:41:06, 5.72s/it]
75%|███████▍ | 32819/43939 [52:55:31<17:41:24, 5.73s/it]
75%|███████▍ | 32820/43939 [52:55:37<17:39:52, 5.72s/it]
75%|███████▍ | 32821/43939 [52:55:43<17:38:12, 5.71s/it]
75%|███████▍ | 32822/43939 [52:55:48<17:35:49, 5.70s/it]
75%|███████▍ | 32823/43939 [52:55:54<17:36:24, 5.70s/it]
75%|███████▍ | 32824/43939 [52:56:00<17:37:35, 5.71s/it]
75%|███████▍ | 32825/43939 [52:56:06<17:36:46, 5.71s/it]
75%|███████▍ | 32826/43939 [52:56:11<17:36:50, 5.71s/it]
75%|███████▍ | 32827/43939 [52:56:17<17:38:20, 5.71s/it]
75%|███████▍ | 32828/43939 [52:56:23<17:39:13, 5.72s/it]
75%|███████▍ | 32829/43939 [52:56:28<17:39:46, 5.72s/it]
75%|███████▍ | 32830/43939 [52:56:34<17:37:18, 5.71s/it]
75%|███████▍ | 32831/43939 [52:56:40<17:51:14, 5.79s/it]
75%|███████▍ | 32832/43939 [52:56:46<17:46:59, 5.76s/it]
75%|███████▍ | 32833/43939 [52:56:51<17:43:46, 5.75s/it]
75%|███████▍ | 32834/43939 [52:56:57<17:40:17, 5.73s/it]
75%|███████▍ | 32835/43939 [52:57:03<17:38:57, 5.72s/it]
75%|███████▍ | 32836/43939 [52:57:09<17:40:34, 5.73s/it]
75%|███████▍ | 32837/43939 [52:57:14<17:39:39, 5.73s/it]
75%|███████▍ | 32838/43939 [52:57:20<17:36:41, 5.71s/it]
75%|███████▍ | 32839/43939 [52:57:26<17:35:45, 5.71s/it]
75%|███████▍ | 32840/43939 [52:57:31<17:34:01, 5.70s/it]
75%|███████▍ | 32841/43939 [52:57:37<17:34:12, 5.70s/it]
75%|███████▍ | 32842/43939 [52:57:43<17:36:02, 5.71s/it]
75%|███████▍ | 32843/43939 [52:57:49<17:33:35, 5.70s/it]
75%|███████▍ | 32844/43939 [52:57:54<17:35:39, 5.71s/it]
75%|███████▍ | 32845/43939 [52:58:00<17:33:32, 5.70s/it]
75%|███████▍ | 32846/43939 [52:58:06<17:35:44, 5.71s/it]
75%|███████▍ | 32847/43939 [52:58:11<17:35:14, 5.71s/it]
75%|███████▍ | 32848/43939 [52:58:17<17:35:00, 5.71s/it]
75%|███████▍ | 32849/43939 [52:58:23<17:32:39, 5.70s/it]
75%|███████▍ | 32850/43939 [52:58:28<17:31:39, 5.69s/it]
{'loss': 0.7687, 'grad_norm': 1.6292316913604736, 'learning_rate': 9.0906131702099e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.45, 'epoch': 2.99}
75%|███████▍ | 32850/43939 [52:58:28<17:31:39, 5.69s/it]
75%|███████▍ | 32851/43939 [52:58:34<17:34:52, 5.71s/it]
75%|███████▍ | 32852/43939 [52:58:40<17:47:44, 5.78s/it]
75%|███████▍ | 32853/43939 [52:58:46<17:43:30, 5.76s/it]
75%|███████▍ | 32854/43939 [52:58:52<17:40:09, 5.74s/it]
75%|███████▍ | 32855/43939 [52:58:57<17:37:18, 5.72s/it]
75%|███████▍ | 32856/43939 [52:59:03<17:33:41, 5.70s/it]
75%|███████▍ | 32857/43939 [52:59:09<17:32:36, 5.70s/it]
75%|███████▍ | 32858/43939 [52:59:14<17:33:24, 5.70s/it]
75%|███████▍ | 32859/43939 [52:59:20<17:33:57, 5.71s/it]
75%|███████▍ | 32860/43939 [52:59:26<17:36:07, 5.72s/it]
75%|███████▍ | 32861/43939 [52:59:31<17:35:42, 5.72s/it]
75%|███████▍ | 32862/43939 [52:59:37<17:36:53, 5.72s/it]
75%|███████▍ | 32863/43939 [52:59:43<17:32:41, 5.70s/it]
75%|███████▍ | 32864/43939 [52:59:48<17:31:02, 5.69s/it]
75%|███████▍ | 32865/43939 [52:59:54<17:31:06, 5.69s/it]
75%|███████▍ | 32866/43939 [53:00:00<17:33:49, 5.71s/it]
75%|███████▍ | 32867/43939 [53:00:06<17:32:33, 5.70s/it]
75%|███████▍ | 32868/43939 [53:00:11<17:31:37, 5.70s/it]
75%|███████▍ | 32869/43939 [53:00:17<17:33:36, 5.71s/it]
75%|███████▍ | 32870/43939 [53:00:23<17:33:18, 5.71s/it]
75%|███████▍ | 32871/43939 [53:00:28<17:31:14, 5.70s/it]
75%|███████▍ | 32872/43939 [53:00:34<17:30:41, 5.70s/it]
75%|███████▍ | 32873/43939 [53:00:40<17:41:27, 5.76s/it]
75%|███████▍ | 32874/43939 [53:00:46<17:39:53, 5.75s/it]
75%|███████▍ | 32875/43939 [53:00:51<17:37:38, 5.74s/it]
75%|███████▍ | 32876/43939 [53:00:57<17:34:02, 5.72s/it]
75%|███████▍ | 32877/43939 [53:01:03<17:31:58, 5.71s/it]
75%|███████▍ | 32878/43939 [53:01:09<17:31:43, 5.71s/it]
75%|███████▍ | 32879/43939 [53:01:14<17:30:44, 5.70s/it]
75%|███████▍ | 32880/43939 [53:01:20<17:29:10, 5.69s/it]
75%|███████▍ | 32881/43939 [53:01:26<17:29:23, 5.69s/it]
75%|███████▍ | 32882/43939 [53:01:31<17:31:01, 5.70s/it]
75%|███████▍ | 32883/43939 [53:01:37<17:31:17, 5.71s/it]
75%|███████▍ | 32884/43939 [53:01:43<17:29:05, 5.69s/it]
75%|███████▍ | 32885/43939 [53:01:48<17:29:54, 5.70s/it]
75%|███████▍ | 32886/43939 [53:01:54<17:27:44, 5.69s/it]
75%|███████▍ | 32887/43939 [53:02:00<17:27:44, 5.69s/it]
75%|███████▍ | 32888/43939 [53:02:05<17:27:43, 5.69s/it]
75%|███████▍ | 32889/43939 [53:02:11<17:29:48, 5.70s/it]
75%|███████▍ | 32890/43939 [53:02:17<17:29:23, 5.70s/it]
75%|███████▍ | 32891/43939 [53:02:23<17:27:39, 5.69s/it]
75%|███████▍ | 32892/43939 [53:02:28<17:27:33, 5.69s/it]
75%|███████▍ | 32893/43939 [53:02:34<17:29:47, 5.70s/it]
75%|███████▍ | 32894/43939 [53:02:40<17:41:33, 5.77s/it]
75%|███████▍ | 32895/43939 [53:02:46<17:36:54, 5.74s/it]
75%|███████▍ | 32896/43939 [53:02:51<17:33:26, 5.72s/it]
75%|███████▍ | 32897/43939 [53:02:57<17:31:35, 5.71s/it]
75%|███████▍ | 32898/43939 [53:03:03<17:30:40, 5.71s/it]
75%|███████▍ | 32899/43939 [53:03:08<17:31:00, 5.71s/it]
75%|███████▍ | 32900/43939 [53:03:14<17:30:54, 5.71s/it]
{'loss': 0.7623, 'grad_norm': 1.3835020065307617, 'learning_rate': 9.014139471889379e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 466.51, 'epoch': 2.99}
75%|███████▍ | 32900/43939 [53:03:14<17:30:54, 5.71s/it]
75%|███████▍ | 32901/43939 [53:03:20<17:30:10, 5.71s/it]
75%|███████▍ | 32902/43939 [53:03:25<17:31:15, 5.71s/it]
75%|███████▍ | 32903/43939 [53:03:31<17:33:22, 5.73s/it]
75%|███████▍ | 32904/43939 [53:03:37<17:31:30, 5.72s/it]
75%|███████▍ | 32905/43939 [53:03:43<17:30:30, 5.71s/it]
75%|███████▍ | 32906/43939 [53:03:48<17:29:09, 5.71s/it]
75%|███████▍ | 32907/43939 [53:03:54<17:29:52, 5.71s/it]
75%|███████▍ | 32908/43939 [53:04:00<17:29:00, 5.71s/it]
75%|███████▍ | 32909/43939 [53:04:05<17:27:44, 5.70s/it]
75%|███████▍ | 32910/43939 [53:04:11<17:28:10, 5.70s/it]
75%|███████▍ | 32911/43939 [53:04:17<17:26:23, 5.69s/it]
75%|███████▍ | 32912/43939 [53:04:22<17:25:29, 5.69s/it]
75%|███████▍ | 32913/43939 [53:04:28<17:24:19, 5.68s/it]
75%|███████▍ | 32914/43939 [53:04:34<17:26:04, 5.69s/it]
75%|███████▍ | 32915/43939 [53:04:40<17:25:48, 5.69s/it]
75%|███████▍ | 32916/43939 [53:04:45<17:37:11, 5.75s/it]
75%|███████▍ | 32917/43939 [53:04:51<17:34:40, 5.74s/it]
75%|███████▍ | 32918/43939 [53:04:57<17:32:02, 5.73s/it]
75%|███████▍ | 32919/43939 [53:05:03<17:30:43, 5.72s/it]
75%|███████▍ | 32920/43939 [53:05:08<17:30:10, 5.72s/it]
75%|███████▍ | 32921/43939 [53:05:14<17:30:39, 5.72s/it]
75%|███████▍ | 32922/43939 [53:05:20<17:27:54, 5.71s/it]
75%|███████▍ | 32923/43939 [53:05:25<17:26:22, 5.70s/it]
75%|███████▍ | 32924/43939 [53:05:31<17:24:52, 5.69s/it]
75%|███████▍ | 32925/43939 [53:05:37<17:25:32, 5.70s/it]
75%|███████▍ | 32926/43939 [53:05:42<17:27:42, 5.71s/it]
75%|███████▍ | 32927/43939 [53:05:48<17:28:17, 5.71s/it]
75%|███████▍ | 32928/43939 [53:05:54<17:25:01, 5.69s/it]
75%|███████▍ | 32929/43939 [53:06:00<17:26:45, 5.70s/it]
75%|███████▍ | 32930/43939 [53:06:05<17:31:45, 5.73s/it]
75%|███████▍ | 32931/43939 [53:06:11<17:29:24, 5.72s/it]
75%|███████▍ | 32932/43939 [53:06:17<17:27:47, 5.71s/it]
75%|███████▍ | 32933/43939 [53:06:22<17:28:05, 5.71s/it]
75%|███████▍ | 32934/43939 [53:06:28<17:25:47, 5.70s/it]
75%|███████▍ | 32935/43939 [53:06:34<17:25:10, 5.70s/it]
75%|███████▍ | 32936/43939 [53:06:40<17:25:01, 5.70s/it]
75%|███████▍ | 32937/43939 [53:06:45<17:28:43, 5.72s/it]
75%|███████▍ | 32938/43939 [53:06:51<17:40:08, 5.78s/it]
75%|███████▍ | 32939/43939 [53:06:57<17:38:07, 5.77s/it]
75%|███████▍ | 32940/43939 [53:07:03<17:32:42, 5.74s/it]
75%|███████▍ | 32941/43939 [53:07:08<17:32:40, 5.74s/it]
75%|███████▍ | 32942/43939 [53:07:14<17:29:03, 5.72s/it]
75%|███████▍ | 32943/43939 [53:07:20<17:28:14, 5.72s/it]
75%|███████▍ | 32944/43939 [53:07:25<17:27:20, 5.72s/it]
75%|███████▍ | 32945/43939 [53:07:31<17:29:44, 5.73s/it]
75%|███████▍ | 32946/43939 [53:07:37<17:27:52, 5.72s/it]
75%|███████▍ | 32947/43939 [53:07:43<17:29:41, 5.73s/it]
75%|███████▍ | 32948/43939 [53:07:48<17:27:48, 5.72s/it]
75%|███████▍ | 32949/43939 [53:07:54<17:25:37, 5.71s/it]
75%|███████▍ | 32950/43939 [53:08:00<17:25:26, 5.71s/it]
{'loss': 0.7704, 'grad_norm': 1.5445491075515747, 'learning_rate': 8.93791798808494e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.35, 'epoch': 3.0}
75%|███████▍ | 32950/43939 [53:08:00<17:25:26, 5.71s/it]
75%|███████▍ | 32951/43939 [53:08:05<17:24:59, 5.71s/it]
75%|███████▍ | 32952/43939 [53:08:11<17:24:07, 5.70s/it]
75%|███████▍ | 32953/43939 [53:08:17<17:24:16, 5.70s/it]
75%|███████▍ | 32954/43939 [53:08:23<17:23:24, 5.70s/it]
75%|███████▌ | 32955/43939 [53:08:27<15:46:31, 5.17s/it][2025-11-30 23:12:40,336] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.50s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:45, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:34, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:11, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<01:00, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.73s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 0.9735714197158813, 'eval_runtime': 196.1801, 'eval_samples_per_second': 5.092, 'eval_steps_per_second': 0.367, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 3.0}
75%|███████▌ | 32955/43939 [53:11:43<15:46:31, 5.17s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
75%|███████▌ | 32956/43939 [53:11:51<197:52:09, 64.86s/it]
75%|███████▌ | 32957/43939 [53:11:56<143:43:10, 47.11s/it]
75%|███████▌ | 32958/43939 [53:12:02<106:04:41, 34.78s/it][2025-11-30 23:16:16,165] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-32958
75%|███████▌ | 32959/43939 [53:12:10<81:21:01, 26.67s/it]
75%|███████▌ | 32960/43939 [53:12:16<62:11:20, 20.39s/it]
75%|███████▌ | 32961/43939 [53:12:22<48:45:34, 15.99s/it]
75%|███████▌ | 32962/43939 [53:12:27<39:18:29, 12.89s/it]
75%|███████▌ | 32963/43939 [53:12:33<32:42:55, 10.73s/it]
75%|███████▌ | 32964/43939 [53:12:39<28:06:07, 9.22s/it]
75%|███████▌ | 32965/43939 [53:12:44<24:57:09, 8.19s/it]
75%|███████▌ | 32966/43939 [53:12:50<22:39:25, 7.43s/it]
75%|███████▌ | 32967/43939 [53:12:56<21:04:08, 6.91s/it]
75%|███████▌ | 32968/43939 [53:13:01<19:59:12, 6.56s/it]
75%|███████▌ | 32969/43939 [53:13:07<19:12:22, 6.30s/it]
75%|███████▌ | 32970/43939 [53:13:13<18:40:02, 6.13s/it]
75%|███████▌ | 32971/43939 [53:13:19<18:17:50, 6.01s/it]
75%|███████▌ | 32972/43939 [53:13:24<18:02:05, 5.92s/it]
75%|███████▌ | 32973/43939 [53:13:30<17:50:46, 5.86s/it]
75%|███████▌ | 32974/43939 [53:13:36<17:42:37, 5.81s/it]
75%|███████▌ | 32975/43939 [53:13:42<17:37:34, 5.79s/it]
75%|███████▌ | 32976/43939 [53:13:47<17:35:32, 5.78s/it]
75%|███████▌ | 32977/43939 [53:13:53<17:31:52, 5.76s/it]
75%|███████▌ | 32978/43939 [53:13:59<17:43:38, 5.82s/it]
75%|███████▌ | 32979/43939 [53:14:05<17:37:03, 5.79s/it]
75%|███████▌ | 32980/43939 [53:14:10<17:31:51, 5.76s/it]
75%|███████▌ | 32981/43939 [53:14:16<17:30:54, 5.75s/it]
75%|███████▌ | 32982/43939 [53:14:22<17:28:05, 5.74s/it]
75%|███████▌ | 32983/43939 [53:14:27<17:25:45, 5.73s/it]
75%|███████▌ | 32984/43939 [53:14:33<17:23:16, 5.71s/it]
75%|███████▌ | 32985/43939 [53:14:39<17:22:07, 5.71s/it]
75%|███████▌ | 32986/43939 [53:14:45<17:21:18, 5.70s/it]
75%|███████▌ | 32987/43939 [53:14:50<17:20:21, 5.70s/it]
75%|███████▌ | 32988/43939 [53:14:56<17:18:30, 5.69s/it]
75%|███████▌ | 32989/43939 [53:15:02<17:17:32, 5.69s/it]
75%|███████▌ | 32990/43939 [53:15:07<17:17:09, 5.68s/it]
75%|███████▌ | 32991/43939 [53:15:13<17:16:02, 5.68s/it]
75%|███████▌ | 32992/43939 [53:15:19<17:18:58, 5.69s/it]
75%|███████▌ | 32993/43939 [53:15:24<17:17:54, 5.69s/it]
75%|███████▌ | 32994/43939 [53:15:30<17:18:30, 5.69s/it]
75%|███████▌ | 32995/43939 [53:15:36<17:18:55, 5.70s/it]
75%|███████▌ | 32996/43939 [53:15:41<17:19:06, 5.70s/it]
75%|███████▌ | 32997/43939 [53:15:47<17:18:57, 5.70s/it]
75%|███████▌ | 32998/43939 [53:15:53<17:17:32, 5.69s/it]
75%|███████▌ | 32999/43939 [53:15:59<17:18:26, 5.70s/it]
75%|███████▌ | 33000/43939 [53:16:04<17:31:29, 5.77s/it]
{'loss': 0.6531, 'grad_norm': 1.3530285358428955, 'learning_rate': 8.861949921369619e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 415.57, 'epoch': 3.0}
75%|███████▌ | 33000/43939 [53:16:04<17:31:29, 5.77s/it]
75%|███████▌ | 33001/43939 [53:16:10<17:30:01, 5.76s/it]
75%|███████▌ | 33002/43939 [53:16:16<17:29:21, 5.76s/it]
75%|███████▌ | 33003/43939 [53:16:22<17:24:51, 5.73s/it]
75%|███████▌ | 33004/43939 [53:16:27<17:23:58, 5.73s/it]
75%|███████▌ | 33005/43939 [53:16:33<17:20:34, 5.71s/it]
75%|███████▌ | 33006/43939 [53:16:39<17:18:45, 5.70s/it]
75%|███████▌ | 33007/43939 [53:16:44<17:20:25, 5.71s/it]
75%|███████▌ | 33008/43939 [53:16:50<17:23:09, 5.73s/it]
75%|███████▌ | 33009/43939 [53:16:56<17:22:50, 5.72s/it]
75%|███████▌ | 33010/43939 [53:17:02<17:21:57, 5.72s/it]
75%|███████▌ | 33011/43939 [53:17:07<17:19:20, 5.71s/it]
75%|███████▌ | 33012/43939 [53:17:13<17:18:09, 5.70s/it]
75%|███████▌ | 33013/43939 [53:17:19<17:17:15, 5.70s/it]
75%|███████▌ | 33014/43939 [53:17:24<17:15:29, 5.69s/it]
75%|███████▌ | 33015/43939 [53:17:30<17:16:06, 5.69s/it]
75%|███████▌ | 33016/43939 [53:17:36<17:16:24, 5.69s/it]
75%|███████▌ | 33017/43939 [53:17:41<17:17:08, 5.70s/it]
75%|███████▌ | 33018/43939 [53:17:47<17:20:36, 5.72s/it]
75%|███████▌ | 33019/43939 [53:17:53<17:18:01, 5.70s/it]
75%|███████▌ | 33020/43939 [53:17:59<17:17:45, 5.70s/it]
75%|███████▌ | 33021/43939 [53:18:05<17:30:19, 5.77s/it]
75%|███████▌ | 33022/43939 [53:18:10<17:27:38, 5.76s/it]
75%|███████▌ | 33023/43939 [53:18:16<17:23:12, 5.73s/it]
75%|███████▌ | 33024/43939 [53:18:22<17:21:37, 5.73s/it]
75%|███████▌ | 33025/43939 [53:18:27<17:21:35, 5.73s/it]
75%|███████▌ | 33026/43939 [53:18:33<17:21:37, 5.73s/it]
75%|███████▌ | 33027/43939 [53:18:39<17:21:00, 5.72s/it]
75%|███████▌ | 33028/43939 [53:18:44<17:18:36, 5.71s/it]
75%|███████▌ | 33029/43939 [53:18:50<17:18:44, 5.71s/it]
75%|███████▌ | 33030/43939 [53:18:56<17:16:51, 5.70s/it]
75%|███████▌ | 33031/43939 [53:19:02<17:16:30, 5.70s/it]
75%|███████▌ | 33032/43939 [53:19:07<17:17:41, 5.71s/it]
75%|███████▌ | 33033/43939 [53:19:13<17:19:11, 5.72s/it]
75%|███████▌ | 33034/43939 [53:19:19<17:21:16, 5.73s/it]
75%|███████▌ | 33035/43939 [53:19:25<17:20:47, 5.73s/it]
75%|███████▌ | 33036/43939 [53:19:30<17:18:01, 5.71s/it]
75%|███████▌ | 33037/43939 [53:19:36<17:17:39, 5.71s/it]
75%|███████▌ | 33038/43939 [53:19:42<17:16:53, 5.71s/it]
75%|███████▌ | 33039/43939 [53:19:47<17:17:27, 5.71s/it]
75%|███████▌ | 33040/43939 [53:19:53<17:18:30, 5.72s/it]
75%|███████▌ | 33041/43939 [53:19:59<17:17:37, 5.71s/it]
75%|███████▌ | 33042/43939 [53:20:04<17:18:39, 5.72s/it]
75%|███████▌ | 33043/43939 [53:20:10<17:29:33, 5.78s/it]
75%|███████▌ | 33044/43939 [53:20:16<17:26:19, 5.76s/it]
75%|███████▌ | 33045/43939 [53:20:22<17:23:12, 5.75s/it]
75%|███████▌ | 33046/43939 [53:20:28<17:21:28, 5.74s/it]
75%|███████▌ | 33047/43939 [53:20:33<17:18:43, 5.72s/it]
75%|███████▌ | 33048/43939 [53:20:39<17:17:13, 5.71s/it]
75%|███████▌ | 33049/43939 [53:20:45<17:16:11, 5.71s/it]
75%|███████▌ | 33050/43939 [53:20:50<17:15:17, 5.70s/it]
{'loss': 0.662, 'grad_norm': 1.3419058322906494, 'learning_rate': 8.786236470318187e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.58, 'epoch': 3.01}
75%|███████▌ | 33050/43939 [53:20:50<17:15:17, 5.70s/it]
75%|███████▌ | 33051/43939 [53:20:56<17:15:50, 5.71s/it]
75%|███████▌ | 33052/43939 [53:21:02<17:16:35, 5.71s/it]
75%|███████▌ | 33053/43939 [53:21:08<17:19:16, 5.73s/it]
75%|███████▌ | 33054/43939 [53:21:13<17:18:00, 5.72s/it]
75%|███████▌ | 33055/43939 [53:21:19<17:15:45, 5.71s/it]
75%|███████▌ | 33056/43939 [53:21:25<17:18:17, 5.72s/it]
75%|███████▌ | 33057/43939 [53:21:30<17:16:20, 5.71s/it]
75%|███████▌ | 33058/43939 [53:21:36<17:15:56, 5.71s/it]
75%|███████▌ | 33059/43939 [53:21:42<17:15:59, 5.71s/it]
75%|███████▌ | 33060/43939 [53:21:47<17:13:26, 5.70s/it]
75%|███████▌ | 33061/43939 [53:21:53<17:13:57, 5.70s/it]
75%|███████▌ | 33062/43939 [53:21:59<17:12:44, 5.70s/it]
75%|███████▌ | 33063/43939 [53:22:05<17:25:48, 5.77s/it]
75%|███████▌ | 33064/43939 [53:22:10<17:22:08, 5.75s/it]
75%|███████▌ | 33065/43939 [53:22:16<17:19:48, 5.74s/it]
75%|███████▌ | 33066/43939 [53:22:22<17:18:15, 5.73s/it]
75%|███████▌ | 33067/43939 [53:22:28<17:19:00, 5.73s/it]
75%|███████▌ | 33068/43939 [53:22:33<17:15:51, 5.72s/it]
75%|███████▌ | 33069/43939 [53:22:39<17:12:58, 5.70s/it]
75%|███████▌ | 33070/43939 [53:22:45<17:11:36, 5.69s/it]
75%|███████▌ | 33071/43939 [53:22:50<17:11:37, 5.70s/it]
75%|███████▌ | 33072/43939 [53:22:56<17:10:46, 5.69s/it]
75%|███████▌ | 33073/43939 [53:23:02<17:14:32, 5.71s/it]
75%|███████▌ | 33074/43939 [53:23:08<17:14:34, 5.71s/it]
75%|███████▌ | 33075/43939 [53:23:13<17:15:48, 5.72s/it]
75%|███████▌ | 33076/43939 [53:23:19<17:16:29, 5.72s/it]
75%|███████▌ | 33077/43939 [53:23:25<17:14:06, 5.71s/it]
75%|███████▌ | 33078/43939 [53:23:30<17:12:54, 5.71s/it]
75%|███████▌ | 33079/43939 [53:23:36<17:11:00, 5.70s/it]
75%|███████▌ | 33080/43939 [53:23:42<17:11:46, 5.70s/it]
75%|███████▌ | 33081/43939 [53:23:47<17:13:22, 5.71s/it]
75%|███████▌ | 33082/43939 [53:23:53<17:16:10, 5.73s/it]
75%|███████▌ | 33083/43939 [53:23:59<17:15:51, 5.73s/it]
75%|███████▌ | 33084/43939 [53:24:05<17:13:22, 5.71s/it]
75%|███████▌ | 33085/43939 [53:24:11<17:27:31, 5.79s/it]
75%|███████▌ | 33086/43939 [53:24:16<17:23:15, 5.77s/it]
75%|███████▌ | 33087/43939 [53:24:22<17:20:19, 5.75s/it]
75%|███████▌ | 33088/43939 [53:24:28<17:16:03, 5.73s/it]
75%|███████▌ | 33089/43939 [53:24:33<17:13:25, 5.71s/it]
75%|███████▌ | 33090/43939 [53:24:39<17:13:55, 5.72s/it]
75%|███████▌ | 33091/43939 [53:24:45<17:12:54, 5.71s/it]
75%|███████▌ | 33092/43939 [53:24:51<17:11:42, 5.71s/it]
75%|███████▌ | 33093/43939 [53:24:56<17:09:41, 5.70s/it]
75%|███████▌ | 33094/43939 [53:25:02<17:09:44, 5.70s/it]
75%|███████▌ | 33095/43939 [53:25:08<17:11:32, 5.71s/it]
75%|███████▌ | 33096/43939 [53:25:13<17:12:31, 5.71s/it]
75%|███████▌ | 33097/43939 [53:25:19<17:10:45, 5.70s/it]
75%|███████▌ | 33098/43939 [53:25:25<17:10:13, 5.70s/it]
75%|███████▌ | 33099/43939 [53:25:30<17:10:47, 5.71s/it]
75%|███████▌ | 33100/43939 [53:25:36<17:13:27, 5.72s/it]
{'loss': 0.6681, 'grad_norm': 1.4862751960754395, 'learning_rate': 8.710778829488255e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.86, 'epoch': 3.01}
75%|███████▌ | 33100/43939 [53:25:36<17:13:27, 5.72s/it]
75%|███████▌ | 33101/43939 [53:25:42<17:12:51, 5.72s/it]
75%|███████▌ | 33102/43939 [53:25:48<17:12:06, 5.71s/it]
75%|███████▌ | 33103/43939 [53:25:53<17:11:50, 5.71s/it]
75%|███████▌ | 33104/43939 [53:25:59<17:10:59, 5.71s/it]
75%|███████▌ | 33105/43939 [53:26:05<17:13:37, 5.72s/it]
75%|███████▌ | 33106/43939 [53:26:11<17:25:26, 5.79s/it]
75%|███████▌ | 33107/43939 [53:26:16<17:20:15, 5.76s/it]
75%|███████▌ | 33108/43939 [53:26:22<17:16:56, 5.74s/it]
75%|███████▌ | 33109/43939 [53:26:28<17:14:51, 5.73s/it]
75%|███████▌ | 33110/43939 [53:26:34<17:13:33, 5.73s/it]
75%|███████▌ | 33111/43939 [53:26:39<17:11:30, 5.72s/it]
75%|███████▌ | 33112/43939 [53:26:45<17:09:24, 5.70s/it]
75%|███████▌ | 33113/43939 [53:26:51<17:09:35, 5.71s/it]
75%|███████▌ | 33114/43939 [53:26:56<17:09:36, 5.71s/it]
75%|███████▌ | 33115/43939 [53:27:02<17:11:37, 5.72s/it]
75%|███████▌ | 33116/43939 [53:27:08<17:11:44, 5.72s/it]
75%|███████▌ | 33117/43939 [53:27:14<17:11:35, 5.72s/it]
75%|███████▌ | 33118/43939 [53:27:19<17:10:47, 5.72s/it]
75%|███████▌ | 33119/43939 [53:27:25<17:10:08, 5.71s/it]
75%|███████▌ | 33120/43939 [53:27:31<17:07:34, 5.70s/it]
75%|███████▌ | 33121/43939 [53:27:36<17:05:22, 5.69s/it]
75%|███████▌ | 33122/43939 [53:27:42<17:03:04, 5.67s/it]
75%|███████▌ | 33123/43939 [53:27:48<17:04:24, 5.68s/it]
75%|███████▌ | 33124/43939 [53:27:53<17:04:03, 5.68s/it]
75%|███████▌ | 33125/43939 [53:27:59<17:03:28, 5.68s/it]
75%|███████▌ | 33126/43939 [53:28:05<17:04:41, 5.69s/it]
75%|███████▌ | 33127/43939 [53:28:11<17:19:26, 5.77s/it]
75%|███████▌ | 33128/43939 [53:28:16<17:17:10, 5.76s/it]
75%|███████▌ | 33129/43939 [53:28:22<17:14:25, 5.74s/it]
75%|███████▌ | 33130/43939 [53:28:28<17:14:21, 5.74s/it]
75%|███████▌ | 33131/43939 [53:28:34<17:11:11, 5.72s/it]
75%|███████▌ | 33132/43939 [53:28:39<17:08:52, 5.71s/it]
75%|███████▌ | 33133/43939 [53:28:45<17:08:30, 5.71s/it]
75%|███████▌ | 33134/43939 [53:28:51<17:06:55, 5.70s/it]
75%|███████▌ | 33135/43939 [53:28:56<17:04:24, 5.69s/it]
75%|███████▌ | 33136/43939 [53:29:02<17:04:51, 5.69s/it]
75%|███████▌ | 33137/43939 [53:29:08<17:05:19, 5.70s/it]
75%|███████▌ | 33138/43939 [53:29:13<17:11:21, 5.73s/it]
75%|███████▌ | 33139/43939 [53:29:19<17:09:22, 5.72s/it]
75%|███████▌ | 33140/43939 [53:29:25<17:06:48, 5.71s/it]
75%|███████▌ | 33141/43939 [53:29:30<17:05:26, 5.70s/it]
75%|███████▌ | 33142/43939 [53:29:36<17:05:38, 5.70s/it]
75%|███████▌ | 33143/43939 [53:29:42<17:06:22, 5.70s/it]
75%|███████▌ | 33144/43939 [53:29:48<17:05:46, 5.70s/it]
75%|███████▌ | 33145/43939 [53:29:53<17:07:57, 5.71s/it]
75%|███████▌ | 33146/43939 [53:29:59<17:08:23, 5.72s/it]
75%|███████▌ | 33147/43939 [53:30:05<17:07:41, 5.71s/it]
75%|███████▌ | 33148/43939 [53:30:11<17:20:13, 5.78s/it]
75%|███████▌ | 33149/43939 [53:30:16<17:14:50, 5.75s/it]
75%|███████▌ | 33150/43939 [53:30:22<17:11:44, 5.74s/it]
{'loss': 0.6357, 'grad_norm': 1.7250821590423584, 'learning_rate': 8.635578189401436e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.61, 'epoch': 3.02}
75%|███████▌ | 33150/43939 [53:30:22<17:11:44, 5.74s/it]
75%|███████▌ | 33151/43939 [53:30:28<17:09:13, 5.72s/it]
75%|███████▌ | 33152/43939 [53:30:34<17:07:54, 5.72s/it]
75%|███████▌ | 33153/43939 [53:30:39<17:07:35, 5.72s/it]
75%|███████▌ | 33154/43939 [53:30:45<17:06:34, 5.71s/it]
75%|███████▌ | 33155/43939 [53:30:51<17:08:27, 5.72s/it]
75%|███████▌ | 33156/43939 [53:30:56<17:05:42, 5.71s/it]
75%|███████▌ | 33157/43939 [53:31:02<17:04:19, 5.70s/it]
75%|███████▌ | 33158/43939 [53:31:08<17:02:56, 5.69s/it]
75%|███████▌ | 33159/43939 [53:31:13<17:03:19, 5.70s/it]
75%|███████▌ | 33160/43939 [53:31:19<17:01:56, 5.69s/it]
75%|███████▌ | 33161/43939 [53:31:25<17:00:28, 5.68s/it]
75%|███████▌ | 33162/43939 [53:31:30<17:04:26, 5.70s/it]
75%|███████▌ | 33163/43939 [53:31:36<17:03:00, 5.70s/it]
75%|███████▌ | 33164/43939 [53:31:42<17:04:01, 5.70s/it]
75%|███████▌ | 33165/43939 [53:31:48<17:03:00, 5.70s/it]
75%|███████▌ | 33166/43939 [53:31:53<17:01:51, 5.69s/it]
75%|███████▌ | 33167/43939 [53:31:59<17:03:12, 5.70s/it]
75%|███████▌ | 33168/43939 [53:32:05<17:02:34, 5.70s/it]
75%|███████▌ | 33169/43939 [53:32:10<17:03:19, 5.70s/it]
75%|███████▌ | 33170/43939 [53:32:16<17:16:16, 5.77s/it]
75%|███████▌ | 33171/43939 [53:32:22<17:11:05, 5.75s/it]
75%|███████▌ | 33172/43939 [53:32:28<17:10:26, 5.74s/it]
75%|███████▌ | 33173/43939 [53:32:33<17:10:12, 5.74s/it]
76%|███████▌ | 33174/43939 [53:32:39<17:05:54, 5.72s/it]
76%|███████▌ | 33175/43939 [53:32:45<17:04:50, 5.71s/it]
76%|███████▌ | 33176/43939 [53:32:51<17:05:06, 5.71s/it]
76%|███████▌ | 33177/43939 [53:32:56<17:08:38, 5.73s/it]
76%|███████▌ | 33178/43939 [53:33:02<17:08:24, 5.73s/it]
76%|███████▌ | 33179/43939 [53:33:08<17:05:26, 5.72s/it]
76%|███████▌ | 33180/43939 [53:33:13<17:04:12, 5.71s/it]
76%|███████▌ | 33181/43939 [53:33:19<17:01:49, 5.70s/it]
76%|███████▌ | 33182/43939 [53:33:25<17:04:45, 5.72s/it]
76%|███████▌ | 33183/43939 [53:33:31<17:05:24, 5.72s/it]
76%|███████▌ | 33184/43939 [53:33:36<17:04:04, 5.71s/it]
76%|███████▌ | 33185/43939 [53:33:42<17:03:38, 5.71s/it]
76%|███████▌ | 33186/43939 [53:33:48<17:05:47, 5.72s/it]
76%|███████▌ | 33187/43939 [53:33:53<17:04:05, 5.71s/it]
76%|███████▌ | 33188/43939 [53:33:59<17:04:26, 5.72s/it]
76%|███████▌ | 33189/43939 [53:34:05<17:04:36, 5.72s/it]
76%|███████▌ | 33190/43939 [53:34:11<17:02:27, 5.71s/it]
76%|███████▌ | 33191/43939 [53:34:16<17:01:42, 5.70s/it]
76%|███████▌ | 33192/43939 [53:34:22<17:15:01, 5.78s/it]
76%|███████▌ | 33193/43939 [53:34:28<17:13:37, 5.77s/it]
76%|███████▌ | 33194/43939 [53:34:34<17:09:37, 5.75s/it]
76%|███████▌ | 33195/43939 [53:34:39<17:07:38, 5.74s/it]
76%|███████▌ | 33196/43939 [53:34:45<17:04:53, 5.72s/it]
76%|███████▌ | 33197/43939 [53:34:51<17:02:38, 5.71s/it]
76%|███████▌ | 33198/43939 [53:34:57<17:06:26, 5.73s/it]
76%|███████▌ | 33199/43939 [53:35:02<17:04:24, 5.72s/it]
76%|███████▌ | 33200/43939 [53:35:08<17:03:48, 5.72s/it]
{'loss': 0.6431, 'grad_norm': 1.5229028463363647, 'learning_rate': 8.560635736524541e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.77, 'epoch': 3.02}
76%|███████▌ | 33200/43939 [53:35:08<17:03:48, 5.72s/it]
76%|███████▌ | 33201/43939 [53:35:14<17:02:14, 5.71s/it]
76%|███████▌ | 33202/43939 [53:35:19<17:02:22, 5.71s/it]
76%|███████▌ | 33203/43939 [53:35:25<17:00:33, 5.70s/it]
76%|███████▌ | 33204/43939 [53:35:31<17:00:40, 5.70s/it]
76%|███████▌ | 33205/43939 [53:35:36<17:01:42, 5.71s/it]
76%|███████▌ | 33206/43939 [53:35:42<16:59:15, 5.70s/it]
76%|███████▌ | 33207/43939 [53:35:48<17:00:00, 5.70s/it]
76%|███████▌ | 33208/43939 [53:35:54<16:58:28, 5.69s/it]
76%|███████▌ | 33209/43939 [53:35:59<16:58:39, 5.70s/it]
76%|███████▌ | 33210/43939 [53:36:05<17:01:59, 5.72s/it]
76%|███████▌ | 33211/43939 [53:36:11<17:01:00, 5.71s/it]
76%|███████▌ | 33212/43939 [53:36:16<17:01:24, 5.71s/it]
76%|███████▌ | 33213/43939 [53:36:22<17:12:57, 5.78s/it]
76%|███████▌ | 33214/43939 [53:36:28<17:09:06, 5.76s/it]
76%|███████▌ | 33215/43939 [53:36:34<17:07:39, 5.75s/it]
76%|███████▌ | 33216/43939 [53:36:40<17:06:20, 5.74s/it]
76%|███████▌ | 33217/43939 [53:36:45<17:03:49, 5.73s/it]
76%|███████▌ | 33218/43939 [53:36:51<17:02:11, 5.72s/it]
76%|███████▌ | 33219/43939 [53:36:57<17:00:11, 5.71s/it]
76%|███████▌ | 33220/43939 [53:37:02<17:01:05, 5.72s/it]
76%|███████▌ | 33221/43939 [53:37:08<17:01:37, 5.72s/it]
76%|███████▌ | 33222/43939 [53:37:14<16:58:45, 5.70s/it]
76%|███████▌ | 33223/43939 [53:37:19<16:59:06, 5.71s/it]
76%|███████▌ | 33224/43939 [53:37:25<16:57:34, 5.70s/it]
76%|███████▌ | 33225/43939 [53:37:31<16:57:58, 5.70s/it]
76%|███████▌ | 33226/43939 [53:37:37<16:58:11, 5.70s/it]
76%|███████▌ | 33227/43939 [53:37:42<16:57:51, 5.70s/it]
76%|███████▌ | 33228/43939 [53:37:48<16:59:05, 5.71s/it]
76%|███████▌ | 33229/43939 [53:37:54<16:57:47, 5.70s/it]
76%|███████▌ | 33230/43939 [53:37:59<16:57:38, 5.70s/it]
76%|███████▌ | 33231/43939 [53:38:05<16:59:02, 5.71s/it]
76%|███████▌ | 33232/43939 [53:38:11<16:59:51, 5.72s/it]
76%|███████▌ | 33233/43939 [53:38:16<16:56:43, 5.70s/it]
76%|███████▌ | 33234/43939 [53:38:22<16:56:39, 5.70s/it]
76%|███████▌ | 33235/43939 [53:38:28<17:06:37, 5.75s/it]
76%|███████▌ | 33236/43939 [53:38:34<17:04:23, 5.74s/it]
76%|███████▌ | 33237/43939 [53:38:39<17:02:01, 5.73s/it]
76%|███████▌ | 33238/43939 [53:38:45<16:59:34, 5.72s/it]
76%|███████▌ | 33239/43939 [53:38:51<16:59:05, 5.71s/it]
76%|███████▌ | 33240/43939 [53:38:57<16:59:45, 5.72s/it]
76%|███████▌ | 33241/43939 [53:39:02<16:58:52, 5.71s/it]
76%|███████▌ | 33242/43939 [53:39:08<16:57:30, 5.71s/it]
76%|███████▌ | 33243/43939 [53:39:14<16:56:02, 5.70s/it]
76%|███████▌ | 33244/43939 [53:39:19<16:56:36, 5.70s/it]
76%|███████▌ | 33245/43939 [53:39:25<16:57:56, 5.71s/it]
76%|███████▌ | 33246/43939 [53:39:31<16:58:38, 5.72s/it]
76%|███████▌ | 33247/43939 [53:39:37<17:00:33, 5.73s/it]
76%|███████▌ | 33248/43939 [53:39:42<16:57:21, 5.71s/it]
76%|███████▌ | 33249/43939 [53:39:48<16:58:18, 5.72s/it]
76%|███████▌ | 33250/43939 [53:39:54<16:55:10, 5.70s/it]
{'loss': 0.66, 'grad_norm': 1.6300897598266602, 'learning_rate': 8.485952653250877e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.31, 'epoch': 3.03}
76%|███████▌ | 33250/43939 [53:39:54<16:55:10, 5.70s/it]
76%|███████▌ | 33251/43939 [53:39:59<16:55:53, 5.70s/it]
76%|███████▌ | 33252/43939 [53:40:05<16:55:22, 5.70s/it]
76%|███████▌ | 33253/43939 [53:40:11<16:54:44, 5.70s/it]
76%|███████▌ | 33254/43939 [53:40:16<16:52:48, 5.69s/it]
76%|███████▌ | 33255/43939 [53:40:22<16:54:28, 5.70s/it]
76%|███████▌ | 33256/43939 [53:40:28<17:08:07, 5.77s/it]
76%|███████▌ | 33257/43939 [53:40:34<17:05:25, 5.76s/it]
76%|███████▌ | 33258/43939 [53:40:39<17:01:23, 5.74s/it]
76%|███████▌ | 33259/43939 [53:40:45<16:58:56, 5.72s/it]
76%|███████▌ | 33260/43939 [53:40:51<17:00:28, 5.73s/it]
76%|███████▌ | 33261/43939 [53:40:57<16:58:47, 5.72s/it]
76%|███████▌ | 33262/43939 [53:41:02<16:59:03, 5.73s/it]
76%|███████▌ | 33263/43939 [53:41:08<16:56:18, 5.71s/it]
76%|███████▌ | 33264/43939 [53:41:14<16:54:50, 5.70s/it]
76%|███████▌ | 33265/43939 [53:41:19<16:55:34, 5.71s/it]
76%|███████▌ | 33266/43939 [53:41:25<16:55:02, 5.71s/it]
76%|███████▌ | 33267/43939 [53:41:31<16:57:13, 5.72s/it]
76%|███████▌ | 33268/43939 [53:41:37<16:55:11, 5.71s/it]
76%|███████▌ | 33269/43939 [53:41:42<16:55:22, 5.71s/it]
76%|███████▌ | 33270/43939 [53:41:48<16:54:51, 5.71s/it]
76%|███████▌ | 33271/43939 [53:41:54<16:54:53, 5.71s/it]
76%|███████▌ | 33272/43939 [53:41:59<16:57:06, 5.72s/it]
76%|███████▌ | 33273/43939 [53:42:05<16:54:37, 5.71s/it]
76%|███████▌ | 33274/43939 [53:42:11<16:54:59, 5.71s/it]
76%|███████▌ | 33275/43939 [53:42:17<16:55:38, 5.71s/it]
76%|███████▌ | 33276/43939 [53:42:22<16:52:16, 5.70s/it]
76%|███████▌ | 33277/43939 [53:42:28<16:52:52, 5.70s/it]
76%|███████▌ | 33278/43939 [53:42:34<17:02:21, 5.75s/it]
76%|███████▌ | 33279/43939 [53:42:40<17:01:38, 5.75s/it]
76%|███████▌ | 33280/43939 [53:42:45<16:59:43, 5.74s/it]
76%|███████▌ | 33281/43939 [53:42:51<16:58:29, 5.73s/it]
76%|███████▌ | 33282/43939 [53:42:57<16:57:08, 5.73s/it]
76%|███████▌ | 33283/43939 [53:43:02<16:55:29, 5.72s/it]
76%|███████▌ | 33284/43939 [53:43:08<16:54:01, 5.71s/it]
76%|███████▌ | 33285/43939 [53:43:14<16:58:03, 5.73s/it]
76%|███████▌ | 33286/43939 [53:43:20<16:56:15, 5.72s/it]
76%|███████▌ | 33287/43939 [53:43:25<16:58:28, 5.74s/it]
76%|███████▌ | 33288/43939 [53:43:31<16:57:10, 5.73s/it]
76%|███████▌ | 33289/43939 [53:43:37<16:56:20, 5.73s/it]
76%|███████▌ | 33290/43939 [53:43:42<16:53:40, 5.71s/it]
76%|███████▌ | 33291/43939 [53:43:48<16:51:43, 5.70s/it]
76%|███████▌ | 33292/43939 [53:43:54<16:53:26, 5.71s/it]
76%|███████▌ | 33293/43939 [53:44:00<16:53:11, 5.71s/it]
76%|███████▌ | 33294/43939 [53:44:05<16:53:53, 5.71s/it]
76%|███████▌ | 33295/43939 [53:44:11<16:52:45, 5.71s/it]
76%|███████▌ | 33296/43939 [53:44:17<16:51:16, 5.70s/it]
76%|███████▌ | 33297/43939 [53:44:22<16:51:44, 5.70s/it]
76%|███████▌ | 33298/43939 [53:44:28<17:03:07, 5.77s/it]
76%|███████▌ | 33299/43939 [53:44:34<16:59:31, 5.75s/it]
76%|███████▌ | 33300/43939 [53:44:40<16:57:07, 5.74s/it]
{'loss': 0.6632, 'grad_norm': 1.4617377519607544, 'learning_rate': 8.411530117881587e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.02, 'epoch': 3.03}
76%|███████▌ | 33300/43939 [53:44:40<16:57:07, 5.74s/it]
76%|███████▌ | 33301/43939 [53:44:45<16:55:57, 5.73s/it]
76%|███████▌ | 33302/43939 [53:44:51<16:57:45, 5.74s/it]
76%|███████▌ | 33303/43939 [53:44:57<16:56:43, 5.74s/it]
76%|███████▌ | 33304/43939 [53:45:03<16:54:20, 5.72s/it]
76%|███████▌ | 33305/43939 [53:45:08<16:55:10, 5.73s/it]
76%|███████▌ | 33306/43939 [53:45:14<16:58:20, 5.75s/it]
76%|███████▌ | 33307/43939 [53:45:20<16:56:27, 5.74s/it]
76%|███████▌ | 33308/43939 [53:45:26<16:53:38, 5.72s/it]
76%|███████▌ | 33309/43939 [53:45:31<16:52:02, 5.71s/it]
76%|███████▌ | 33310/43939 [53:45:37<16:51:06, 5.71s/it]
76%|███████▌ | 33311/43939 [53:45:43<16:51:30, 5.71s/it]
76%|███████▌ | 33312/43939 [53:45:48<16:50:03, 5.70s/it]
76%|███████▌ | 33313/43939 [53:45:54<16:47:41, 5.69s/it]
76%|███████▌ | 33314/43939 [53:46:00<16:48:37, 5.70s/it]
76%|███████▌ | 33315/43939 [53:46:06<16:54:20, 5.73s/it]
76%|███████▌ | 33316/43939 [53:46:11<16:54:10, 5.73s/it]
76%|███████▌ | 33317/43939 [53:46:17<16:52:26, 5.72s/it]
76%|███████▌ | 33318/43939 [53:46:23<16:53:09, 5.72s/it]
76%|███████▌ | 33319/43939 [53:46:28<16:52:35, 5.72s/it]
76%|███████▌ | 33320/43939 [53:46:34<17:03:20, 5.78s/it]
76%|███████▌ | 33321/43939 [53:46:40<17:00:24, 5.77s/it]
76%|███████▌ | 33322/43939 [53:46:46<16:55:18, 5.74s/it]
76%|███████▌ | 33323/43939 [53:46:51<16:52:29, 5.72s/it]
76%|███████▌ | 33324/43939 [53:46:57<16:52:41, 5.72s/it]
76%|███████▌ | 33325/43939 [53:47:03<16:50:22, 5.71s/it]
76%|███████▌ | 33326/43939 [53:47:09<16:49:29, 5.71s/it]
76%|███████▌ | 33327/43939 [53:47:14<16:49:47, 5.71s/it]
76%|███████▌ | 33328/43939 [53:47:20<16:48:16, 5.70s/it]
76%|███████▌ | 33329/43939 [53:47:26<16:48:48, 5.70s/it]
76%|███████▌ | 33330/43939 [53:47:31<16:47:23, 5.70s/it]
76%|███████▌ | 33331/43939 [53:47:37<16:49:20, 5.71s/it]
76%|███████▌ | 33332/43939 [53:47:43<16:50:16, 5.71s/it]
76%|███████▌ | 33333/43939 [53:47:48<16:48:42, 5.71s/it]
76%|███████▌ | 33334/43939 [53:47:54<16:50:29, 5.72s/it]
76%|███████▌ | 33335/43939 [53:48:00<16:47:46, 5.70s/it]
76%|███████▌ | 33336/43939 [53:48:06<16:49:31, 5.71s/it]
76%|███████▌ | 33337/43939 [53:48:11<16:47:49, 5.70s/it]
76%|███████▌ | 33338/43939 [53:48:17<16:44:43, 5.69s/it]
76%|███████▌ | 33339/43939 [53:48:23<16:45:28, 5.69s/it]
76%|███████▌ | 33340/43939 [53:48:28<16:43:35, 5.68s/it]
76%|███████▌ | 33341/43939 [53:48:34<16:45:56, 5.70s/it]
76%|███████▌ | 33342/43939 [53:48:40<17:00:23, 5.78s/it]
76%|███████▌ | 33343/43939 [53:48:46<16:56:10, 5.75s/it]
76%|███████▌ | 33344/43939 [53:48:51<16:53:17, 5.74s/it]
76%|███████▌ | 33345/43939 [53:48:57<16:51:07, 5.73s/it]
76%|███████▌ | 33346/43939 [53:49:03<16:48:34, 5.71s/it]
76%|███████▌ | 33347/43939 [53:49:08<16:48:07, 5.71s/it]
76%|███████▌ | 33348/43939 [53:49:14<16:48:48, 5.72s/it]
76%|███████▌ | 33349/43939 [53:49:20<16:48:28, 5.71s/it]
76%|███████▌ | 33350/43939 [53:49:26<16:47:28, 5.71s/it]
{'loss': 0.6502, 'grad_norm': 1.6737926006317139, 'learning_rate': 8.337369304607057e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.89, 'epoch': 3.04}
76%|███████▌ | 33350/43939 [53:49:26<16:47:28, 5.71s/it]
76%|███████▌ | 33351/43939 [53:49:31<16:48:00, 5.71s/it]
76%|███████▌ | 33352/43939 [53:49:37<16:47:42, 5.71s/it]
76%|███████▌ | 33353/43939 [53:49:43<16:47:03, 5.71s/it]
76%|███████▌ | 33354/43939 [53:49:48<16:46:27, 5.71s/it]
76%|███████▌ | 33355/43939 [53:49:54<16:44:47, 5.70s/it]
76%|███████▌ | 33356/43939 [53:50:00<16:44:48, 5.70s/it]
76%|███████▌ | 33357/43939 [53:50:05<16:44:27, 5.70s/it]
76%|███████▌ | 33358/43939 [53:50:11<16:46:42, 5.71s/it]
76%|███████▌ | 33359/43939 [53:50:17<16:47:04, 5.71s/it]
76%|███████▌ | 33360/43939 [53:50:23<16:47:43, 5.72s/it]
76%|███████▌ | 33361/43939 [53:50:28<16:48:07, 5.72s/it]
76%|███████▌ | 33362/43939 [53:50:34<16:45:10, 5.70s/it]
76%|███████▌ | 33363/43939 [53:50:40<16:57:32, 5.77s/it]
76%|███████▌ | 33364/43939 [53:50:46<16:58:19, 5.78s/it]
76%|███████▌ | 33365/43939 [53:50:52<16:55:08, 5.76s/it]
76%|███████▌ | 33366/43939 [53:50:57<16:53:52, 5.75s/it]
76%|███████▌ | 33367/43939 [53:51:03<16:52:02, 5.74s/it]
76%|███████▌ | 33368/43939 [53:51:09<16:49:59, 5.73s/it]
76%|███████▌ | 33369/43939 [53:51:14<16:48:47, 5.73s/it]
76%|███████▌ | 33370/43939 [53:51:20<16:46:51, 5.72s/it]
76%|███████▌ | 33371/43939 [53:51:26<16:46:54, 5.72s/it]
76%|███████▌ | 33372/43939 [53:51:31<16:43:59, 5.70s/it]
76%|███████▌ | 33373/43939 [53:51:37<16:43:19, 5.70s/it]
76%|███████▌ | 33374/43939 [53:51:43<16:44:11, 5.70s/it]
76%|███████▌ | 33375/43939 [53:51:49<16:45:10, 5.71s/it]
76%|███████▌ | 33376/43939 [53:51:54<16:43:17, 5.70s/it]
76%|███████▌ | 33377/43939 [53:52:00<16:44:20, 5.71s/it]
76%|███████▌ | 33378/43939 [53:52:06<16:46:01, 5.72s/it]
76%|███████▌ | 33379/43939 [53:52:11<16:45:53, 5.72s/it]
76%|███████▌ | 33380/43939 [53:52:17<16:44:52, 5.71s/it]
76%|███████▌ | 33381/43939 [53:52:23<16:44:18, 5.71s/it]
76%|███████▌ | 33382/43939 [53:52:29<16:43:23, 5.70s/it]
76%|███████▌ | 33383/43939 [53:52:34<16:46:13, 5.72s/it]
76%|███████▌ | 33384/43939 [53:52:40<16:57:27, 5.78s/it]
76%|███████▌ | 33385/43939 [53:52:46<16:57:14, 5.78s/it]
76%|███████▌ | 33386/43939 [53:52:52<16:53:23, 5.76s/it]
76%|███████▌ | 33387/43939 [53:52:57<16:49:38, 5.74s/it]
76%|███████▌ | 33388/43939 [53:53:03<16:50:04, 5.74s/it]
76%|███████▌ | 33389/43939 [53:53:09<16:46:42, 5.73s/it]
76%|███████▌ | 33390/43939 [53:53:15<16:44:04, 5.71s/it]
76%|███████▌ | 33391/43939 [53:53:20<16:43:09, 5.71s/it]
76%|███████▌ | 33392/43939 [53:53:26<16:45:43, 5.72s/it]
76%|███████▌ | 33393/43939 [53:53:32<16:42:51, 5.71s/it]
76%|███████▌ | 33394/43939 [53:53:37<16:43:14, 5.71s/it]
76%|███████▌ | 33395/43939 [53:53:43<16:43:22, 5.71s/it]
76%|███████▌ | 33396/43939 [53:53:49<16:40:19, 5.69s/it]
76%|███████▌ | 33397/43939 [53:53:54<16:41:02, 5.70s/it]
76%|███████▌ | 33398/43939 [53:54:00<16:41:49, 5.70s/it]
76%|███████▌ | 33399/43939 [53:54:06<16:44:05, 5.72s/it]
76%|███████▌ | 33400/43939 [53:54:12<16:44:50, 5.72s/it]
{'loss': 0.6685, 'grad_norm': 1.4772462844848633, 'learning_rate': 8.263471383488392e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.74, 'epoch': 3.04}
76%|███████▌ | 33400/43939 [53:54:12<16:44:50, 5.72s/it]
76%|███████▌ | 33401/43939 [53:54:17<16:44:49, 5.72s/it]
76%|███████▌ | 33402/43939 [53:54:23<16:44:19, 5.72s/it]
76%|███████▌ | 33403/43939 [53:54:29<16:45:52, 5.73s/it]
76%|███████▌ | 33404/43939 [53:54:34<16:42:48, 5.71s/it]
76%|███████▌ | 33405/43939 [53:54:40<16:40:41, 5.70s/it]
76%|███████▌ | 33406/43939 [53:54:46<16:51:19, 5.76s/it]
76%|███████▌ | 33407/43939 [53:54:52<16:49:50, 5.75s/it]
76%|███████▌ | 33408/43939 [53:54:57<16:44:54, 5.73s/it]
76%|███████▌ | 33409/43939 [53:55:03<16:44:24, 5.72s/it]
76%|███████▌ | 33410/43939 [53:55:09<16:43:51, 5.72s/it]
76%|███████▌ | 33411/43939 [53:55:15<16:41:35, 5.71s/it]
76%|███████▌ | 33412/43939 [53:55:20<16:42:17, 5.71s/it]
76%|███████▌ | 33413/43939 [53:55:26<16:41:03, 5.71s/it]
76%|███████▌ | 33414/43939 [53:55:32<16:43:50, 5.72s/it]
76%|███████▌ | 33415/43939 [53:55:37<16:42:02, 5.71s/it]
76%|███████▌ | 33416/43939 [53:55:43<16:40:44, 5.71s/it]
76%|███████▌ | 33417/43939 [53:55:49<16:40:36, 5.71s/it]
76%|███████▌ | 33418/43939 [53:55:55<16:42:02, 5.71s/it]
76%|███████▌ | 33419/43939 [53:56:00<16:40:58, 5.71s/it]
76%|███████▌ | 33420/43939 [53:56:06<16:41:12, 5.71s/it]
76%|███████▌ | 33421/43939 [53:56:12<16:39:58, 5.70s/it]
76%|███████▌ | 33422/43939 [53:56:17<16:37:41, 5.69s/it]
76%|███████▌ | 33423/43939 [53:56:23<16:38:21, 5.70s/it]
76%|███████▌ | 33424/43939 [53:56:29<16:37:32, 5.69s/it]
76%|███████▌ | 33425/43939 [53:56:34<16:37:49, 5.69s/it]
76%|███████▌ | 33426/43939 [53:56:40<16:36:24, 5.69s/it]
76%|███████▌ | 33427/43939 [53:56:46<16:35:23, 5.68s/it]
76%|███████▌ | 33428/43939 [53:56:52<16:48:59, 5.76s/it]
76%|███████▌ | 33429/43939 [53:56:57<16:45:57, 5.74s/it]
76%|███████▌ | 33430/43939 [53:57:03<16:45:13, 5.74s/it]
76%|███████▌ | 33431/43939 [53:57:09<16:40:52, 5.71s/it]
76%|███████▌ | 33432/43939 [53:57:14<16:40:18, 5.71s/it]
76%|███████▌ | 33433/43939 [53:57:20<16:39:54, 5.71s/it]
76%|███████▌ | 33434/43939 [53:57:26<16:38:51, 5.71s/it]
76%|███████▌ | 33435/43939 [53:57:32<16:40:14, 5.71s/it]
76%|███████▌ | 33436/43939 [53:57:37<16:41:24, 5.72s/it]
76%|███████▌ | 33437/43939 [53:57:43<16:42:28, 5.73s/it]
76%|███████▌ | 33438/43939 [53:57:49<16:38:44, 5.71s/it]
76%|███████▌ | 33439/43939 [53:57:54<16:37:45, 5.70s/it]
76%|███████▌ | 33440/43939 [53:58:00<16:38:29, 5.71s/it]
76%|███████▌ | 33441/43939 [53:58:06<16:38:21, 5.71s/it]
76%|███████▌ | 33442/43939 [53:58:12<16:40:10, 5.72s/it]
76%|███████▌ | 33443/43939 [53:58:17<16:41:03, 5.72s/it]
76%|███████▌ | 33444/43939 [53:58:23<16:39:49, 5.72s/it]
76%|███████▌ | 33445/43939 [53:58:29<16:40:48, 5.72s/it]
76%|███████▌ | 33446/43939 [53:58:34<16:39:17, 5.71s/it]
76%|███████▌ | 33447/43939 [53:58:40<16:37:43, 5.71s/it]
76%|███████▌ | 33448/43939 [53:58:46<16:37:44, 5.71s/it]
76%|███████▌ | 33449/43939 [53:58:52<16:47:55, 5.77s/it]
76%|███████▌ | 33450/43939 [53:58:57<16:42:50, 5.74s/it]
{'loss': 0.6599, 'grad_norm': 1.6019269227981567, 'learning_rate': 8.189837520438952e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 495.54, 'epoch': 3.05}
76%|███████▌ | 33450/43939 [53:58:57<16:42:50, 5.74s/it]
76%|███████▌ | 33451/43939 [53:59:03<16:44:54, 5.75s/it]
76%|███████▌ | 33452/43939 [53:59:09<16:42:26, 5.74s/it]
76%|███████▌ | 33453/43939 [53:59:15<16:39:01, 5.72s/it]
76%|███████▌ | 33454/43939 [53:59:20<16:36:23, 5.70s/it]
76%|███████▌ | 33455/43939 [53:59:26<16:38:03, 5.71s/it]
76%|███████▌ | 33456/43939 [53:59:32<16:36:47, 5.71s/it]
76%|███████▌ | 33457/43939 [53:59:37<16:35:21, 5.70s/it]
76%|███████▌ | 33458/43939 [53:59:43<16:34:58, 5.70s/it]
76%|███████▌ | 33459/43939 [53:59:49<16:34:05, 5.69s/it]
76%|███████▌ | 33460/43939 [53:59:54<16:35:14, 5.70s/it]
76%|███████▌ | 33461/43939 [54:00:00<16:35:46, 5.70s/it]
76%|███████▌ | 33462/43939 [54:00:06<16:35:36, 5.70s/it]
76%|███████▌ | 33463/43939 [54:00:12<16:34:36, 5.70s/it]
76%|███████▌ | 33464/43939 [54:00:17<16:35:21, 5.70s/it]
76%|███████▌ | 33465/43939 [54:00:23<16:36:35, 5.71s/it]
76%|███████▌ | 33466/43939 [54:00:29<16:37:47, 5.72s/it]
76%|███████▌ | 33467/43939 [54:00:34<16:38:30, 5.72s/it]
76%|███████▌ | 33468/43939 [54:00:40<16:38:40, 5.72s/it]
76%|███████▌ | 33469/43939 [54:00:46<16:38:10, 5.72s/it]
76%|███████▌ | 33470/43939 [54:00:52<16:49:49, 5.79s/it]
76%|███████▌ | 33471/43939 [54:00:58<16:44:20, 5.76s/it]
76%|███████▌ | 33472/43939 [54:01:03<16:40:19, 5.73s/it]
76%|███████▌ | 33473/43939 [54:01:09<16:40:13, 5.73s/it]
76%|███████▌ | 33474/43939 [54:01:15<16:40:00, 5.73s/it]
76%|███████▌ | 33475/43939 [54:01:20<16:39:35, 5.73s/it]
76%|███████▌ | 33476/43939 [54:01:26<16:35:56, 5.71s/it]
76%|███████▌ | 33477/43939 [54:01:32<16:35:58, 5.71s/it]
76%|███████▌ | 33478/43939 [54:01:37<16:34:51, 5.71s/it]
76%|███████▌ | 33479/43939 [54:01:43<16:33:37, 5.70s/it]
76%|███████▌ | 33480/43939 [54:01:49<16:31:45, 5.69s/it]
76%|███████▌ | 33481/43939 [54:01:55<16:32:23, 5.69s/it]
76%|███████▌ | 33482/43939 [54:02:00<16:30:41, 5.68s/it]
76%|███████▌ | 33483/43939 [54:02:06<16:28:47, 5.67s/it]
76%|███████▌ | 33484/43939 [54:02:12<16:30:38, 5.69s/it]
76%|███████▌ | 33485/43939 [54:02:17<16:31:48, 5.69s/it]
76%|███████▌ | 33486/43939 [54:02:23<16:31:44, 5.69s/it]
76%|███████▌ | 33487/43939 [54:02:29<16:31:04, 5.69s/it]
76%|███████▌ | 33488/43939 [54:02:34<16:31:31, 5.69s/it]
76%|███████▌ | 33489/43939 [54:02:40<16:30:40, 5.69s/it]
76%|███████▌ | 33490/43939 [54:02:46<16:33:33, 5.71s/it]
76%|███████▌ | 33491/43939 [54:02:51<16:32:54, 5.70s/it]
76%|███████▌ | 33492/43939 [54:02:57<16:43:35, 5.76s/it]
76%|███████▌ | 33493/43939 [54:03:03<16:40:04, 5.74s/it]
76%|███████▌ | 33494/43939 [54:03:09<16:35:20, 5.72s/it]
76%|███████▌ | 33495/43939 [54:03:14<16:35:40, 5.72s/it]
76%|███████▌ | 33496/43939 [54:03:20<16:38:02, 5.73s/it]
76%|███████▌ | 33497/43939 [54:03:26<16:34:35, 5.71s/it]
76%|███████▌ | 33498/43939 [54:03:32<16:32:34, 5.70s/it]
76%|███████▌ | 33499/43939 [54:03:37<16:33:31, 5.71s/it]
76%|███████▌ | 33500/43939 [54:03:43<16:32:45, 5.71s/it]
{'loss': 0.6568, 'grad_norm': 1.7322639226913452, 'learning_rate': 8.116468877205969e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.57, 'epoch': 3.05}
76%|███████▌ | 33500/43939 [54:03:43<16:32:45, 5.71s/it]
76%|███████▌ | 33501/43939 [54:03:49<16:33:21, 5.71s/it]
76%|███████▌ | 33502/43939 [54:03:54<16:32:34, 5.71s/it]
76%|███████▌ | 33503/43939 [54:04:00<16:33:42, 5.71s/it]
76%|███████▋ | 33504/43939 [54:04:06<16:33:53, 5.71s/it]
76%|███████▋ | 33505/43939 [54:04:12<16:31:43, 5.70s/it]
76%|███████▋ | 33506/43939 [54:04:17<16:33:20, 5.71s/it]
76%|███████▋ | 33507/43939 [54:04:23<16:33:03, 5.71s/it]
76%|███████▋ | 33508/43939 [54:04:29<16:33:23, 5.71s/it]
76%|███████▋ | 33509/43939 [54:04:34<16:32:26, 5.71s/it]
76%|███████▋ | 33510/43939 [54:04:40<16:30:58, 5.70s/it]
76%|███████▋ | 33511/43939 [54:04:46<16:33:00, 5.71s/it]
76%|███████▋ | 33512/43939 [54:04:52<16:33:32, 5.72s/it]
76%|███████▋ | 33513/43939 [54:04:57<16:34:59, 5.73s/it]
76%|███████▋ | 33514/43939 [54:05:03<16:49:39, 5.81s/it]
76%|███████▋ | 33515/43939 [54:05:09<16:45:06, 5.79s/it]
76%|███████▋ | 33516/43939 [54:05:15<16:41:49, 5.77s/it]
76%|███████▋ | 33517/43939 [54:05:20<16:37:24, 5.74s/it]
76%|███████▋ | 33518/43939 [54:05:26<16:36:38, 5.74s/it]
76%|███████▋ | 33519/43939 [54:05:32<16:34:29, 5.73s/it]
76%|███████▋ | 33520/43939 [54:05:38<16:32:23, 5.71s/it]
76%|███████▋ | 33521/43939 [54:05:43<16:30:11, 5.70s/it]
76%|███████▋ | 33522/43939 [54:05:49<16:29:58, 5.70s/it]
76%|███████▋ | 33523/43939 [54:05:55<16:31:14, 5.71s/it]
76%|███████▋ | 33524/43939 [54:06:00<16:31:12, 5.71s/it]
76%|███████▋ | 33525/43939 [54:06:06<16:28:50, 5.70s/it]
76%|███████▋ | 33526/43939 [54:06:12<16:31:42, 5.71s/it]
76%|███████▋ | 33527/43939 [54:06:17<16:30:56, 5.71s/it]
76%|███████▋ | 33528/43939 [54:06:23<16:30:04, 5.71s/it]
76%|███████▋ | 33529/43939 [54:06:29<16:29:41, 5.70s/it]
76%|███████▋ | 33530/43939 [54:06:35<16:29:20, 5.70s/it]
76%|███████▋ | 33531/43939 [54:06:40<16:27:00, 5.69s/it]
76%|███████▋ | 33532/43939 [54:06:46<16:27:14, 5.69s/it]
76%|███████▋ | 33533/43939 [54:06:52<16:26:37, 5.69s/it]
76%|███████▋ | 33534/43939 [54:06:57<16:29:39, 5.71s/it]
76%|███████▋ | 33535/43939 [54:07:03<16:43:01, 5.78s/it]
76%|███████▋ | 33536/43939 [54:07:09<16:36:06, 5.75s/it]
76%|███████▋ | 33537/43939 [54:07:15<16:34:51, 5.74s/it]
76%|███████▋ | 33538/43939 [54:07:20<16:34:01, 5.73s/it]
76%|███████▋ | 33539/43939 [54:07:26<16:30:24, 5.71s/it]
76%|███████▋ | 33540/43939 [54:07:32<16:28:53, 5.71s/it]
76%|███████▋ | 33541/43939 [54:07:37<16:28:03, 5.70s/it]
76%|███████▋ | 33542/43939 [54:07:43<16:29:24, 5.71s/it]
76%|███████▋ | 33543/43939 [54:07:49<16:28:25, 5.70s/it]
76%|███████▋ | 33544/43939 [54:07:55<16:28:19, 5.70s/it]
76%|███████▋ | 33545/43939 [54:08:00<16:26:35, 5.70s/it]
76%|███████▋ | 33546/43939 [54:08:06<16:24:38, 5.68s/it]
76%|███████▋ | 33547/43939 [54:08:12<16:23:34, 5.68s/it]
76%|███████▋ | 33548/43939 [54:08:17<16:23:40, 5.68s/it]
76%|███████▋ | 33549/43939 [54:08:23<16:23:07, 5.68s/it]
76%|███████▋ | 33550/43939 [54:08:29<16:25:36, 5.69s/it]
{'loss': 0.6588, 'grad_norm': 1.6173838376998901, 'learning_rate': 8.043366611352193e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.71, 'epoch': 3.05}
76%|███████▋ | 33550/43939 [54:08:29<16:25:36, 5.69s/it]
76%|███████▋ | 33551/43939 [54:08:34<16:26:15, 5.70s/it]
76%|███████▋ | 33552/43939 [54:08:40<16:25:09, 5.69s/it]
76%|███████▋ | 33553/43939 [54:08:46<16:24:44, 5.69s/it]
76%|███████▋ | 33554/43939 [54:08:51<16:23:51, 5.68s/it]
76%|███████▋ | 33555/43939 [54:08:57<16:23:49, 5.68s/it]
76%|███████▋ | 33556/43939 [54:09:03<16:23:12, 5.68s/it]
76%|███████▋ | 33557/43939 [54:09:09<16:35:14, 5.75s/it]
76%|███████▋ | 33558/43939 [54:09:14<16:32:42, 5.74s/it]
76%|███████▋ | 33559/43939 [54:09:20<16:29:46, 5.72s/it]
76%|███████▋ | 33560/43939 [54:09:26<16:28:02, 5.71s/it]
76%|███████▋ | 33561/43939 [54:09:31<16:27:15, 5.71s/it]
76%|███████▋ | 33562/43939 [54:09:37<16:27:52, 5.71s/it]
76%|███████▋ | 33563/43939 [54:09:43<16:26:24, 5.70s/it]
76%|███████▋ | 33564/43939 [54:09:49<16:25:18, 5.70s/it]
76%|███████▋ | 33565/43939 [54:09:54<16:24:59, 5.70s/it]
76%|███████▋ | 33566/43939 [54:10:00<16:23:32, 5.69s/it]
76%|███████▋ | 33567/43939 [54:10:06<16:24:36, 5.70s/it]
76%|███████▋ | 33568/43939 [54:10:11<16:24:15, 5.69s/it]
76%|███████▋ | 33569/43939 [54:10:17<16:25:13, 5.70s/it]
76%|███████▋ | 33570/43939 [54:10:23<16:28:26, 5.72s/it]
76%|███████▋ | 33571/43939 [54:10:29<16:28:58, 5.72s/it]
76%|███████▋ | 33572/43939 [54:10:34<16:25:34, 5.70s/it]
76%|███████▋ | 33573/43939 [54:10:40<16:23:55, 5.70s/it]
76%|███████▋ | 33574/43939 [54:10:46<16:24:03, 5.70s/it]
76%|███████▋ | 33575/43939 [54:10:51<16:22:49, 5.69s/it]
76%|███████▋ | 33576/43939 [54:10:57<16:21:46, 5.68s/it]
76%|███████▋ | 33577/43939 [54:11:03<16:22:17, 5.69s/it]
76%|███████▋ | 33578/43939 [54:11:09<16:35:46, 5.77s/it]
76%|███████▋ | 33579/43939 [54:11:14<16:32:49, 5.75s/it]
76%|███████▋ | 33580/43939 [54:11:20<16:31:23, 5.74s/it]
76%|███████▋ | 33581/43939 [54:11:26<16:30:43, 5.74s/it]
76%|███████▋ | 33582/43939 [54:11:31<16:27:35, 5.72s/it]
76%|███████▋ | 33583/43939 [54:11:37<16:24:46, 5.71s/it]
76%|███████▋ | 33584/43939 [54:11:43<16:24:39, 5.71s/it]
76%|███████▋ | 33585/43939 [54:11:49<16:25:36, 5.71s/it]
76%|███████▋ | 33586/43939 [54:11:54<16:26:45, 5.72s/it]
76%|███████▋ | 33587/43939 [54:12:00<16:25:47, 5.71s/it]
76%|███████▋ | 33588/43939 [54:12:06<16:25:43, 5.71s/it]
76%|███████▋ | 33589/43939 [54:12:11<16:24:56, 5.71s/it]
76%|███████▋ | 33590/43939 [54:12:17<16:26:03, 5.72s/it]
76%|███████▋ | 33591/43939 [54:12:23<16:25:17, 5.71s/it]
76%|███████▋ | 33592/43939 [54:12:28<16:24:07, 5.71s/it]
76%|███████▋ | 33593/43939 [54:12:34<16:25:22, 5.71s/it]
76%|███████▋ | 33594/43939 [54:12:40<16:23:07, 5.70s/it]
76%|███████▋ | 33595/43939 [54:12:46<16:21:31, 5.69s/it]
76%|███████▋ | 33596/43939 [54:12:51<16:22:15, 5.70s/it]
76%|███████▋ | 33597/43939 [54:12:57<16:21:29, 5.69s/it]
76%|███████▋ | 33598/43939 [54:13:03<16:21:03, 5.69s/it]
76%|███████▋ | 33599/43939 [54:13:09<16:33:41, 5.77s/it]
76%|███████▋ | 33600/43939 [54:13:14<16:28:25, 5.74s/it]
{'loss': 0.6461, 'grad_norm': 1.5913090705871582, 'learning_rate': 7.970531876237672e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.56, 'epoch': 3.06}
76%|███████▋ | 33600/43939 [54:13:14<16:28:25, 5.74s/it]
76%|███████▋ | 33601/43939 [54:13:20<16:27:19, 5.73s/it]
76%|███████▋ | 33602/43939 [54:13:26<16:25:57, 5.72s/it]
76%|███████▋ | 33603/43939 [54:13:31<16:24:31, 5.72s/it]
76%|███████▋ | 33604/43939 [54:13:37<16:26:28, 5.73s/it]
76%|███████▋ | 33605/43939 [54:13:43<16:24:02, 5.71s/it]
76%|███████▋ | 33606/43939 [54:13:49<16:25:10, 5.72s/it]
76%|███████▋ | 33607/43939 [54:13:54<16:24:52, 5.72s/it]
76%|███████▋ | 33608/43939 [54:14:00<16:24:22, 5.72s/it]
76%|███████▋ | 33609/43939 [54:14:06<16:22:37, 5.71s/it]
76%|███████▋ | 33610/43939 [54:14:11<16:25:57, 5.73s/it]
76%|███████▋ | 33611/43939 [54:14:17<16:24:07, 5.72s/it]
76%|███████▋ | 33612/43939 [54:14:23<16:24:06, 5.72s/it]
76%|███████▋ | 33613/43939 [54:14:29<16:23:11, 5.71s/it]
77%|███████▋ | 33614/43939 [54:14:34<16:21:41, 5.70s/it]
77%|███████▋ | 33615/43939 [54:14:40<16:20:02, 5.70s/it]
77%|███████▋ | 33616/43939 [54:14:46<16:18:51, 5.69s/it]
77%|███████▋ | 33617/43939 [54:14:51<16:18:35, 5.69s/it]
77%|███████▋ | 33618/43939 [54:14:57<16:19:10, 5.69s/it]
77%|███████▋ | 33619/43939 [54:15:03<16:20:32, 5.70s/it]
77%|███████▋ | 33620/43939 [54:15:09<16:33:06, 5.77s/it]
77%|███████▋ | 33621/43939 [54:15:14<16:27:36, 5.74s/it]
77%|███████▋ | 33622/43939 [54:15:20<16:24:36, 5.73s/it]
77%|███████▋ | 33623/43939 [54:15:26<16:25:24, 5.73s/it]
77%|███████▋ | 33624/43939 [54:15:31<16:22:44, 5.72s/it]
77%|███████▋ | 33625/43939 [54:15:37<16:23:03, 5.72s/it]
77%|███████▋ | 33626/43939 [54:15:43<16:20:55, 5.71s/it]
77%|███████▋ | 33627/43939 [54:15:49<16:21:22, 5.71s/it]
77%|███████▋ | 33628/43939 [54:15:54<16:22:16, 5.72s/it]
77%|███████▋ | 33629/43939 [54:16:00<16:23:12, 5.72s/it]
77%|███████▋ | 33630/43939 [54:16:06<16:21:39, 5.71s/it]
77%|███████▋ | 33631/43939 [54:16:11<16:18:41, 5.70s/it]
77%|███████▋ | 33632/43939 [54:16:17<16:19:50, 5.70s/it]
77%|███████▋ | 33633/43939 [54:16:23<16:19:48, 5.70s/it]
77%|███████▋ | 33634/43939 [54:16:28<16:19:49, 5.70s/it]
77%|███████▋ | 33635/43939 [54:16:34<16:18:23, 5.70s/it]
77%|███████▋ | 33636/43939 [54:16:40<16:18:45, 5.70s/it]
77%|███████▋ | 33637/43939 [54:16:46<16:20:54, 5.71s/it]
77%|███████▋ | 33638/43939 [54:16:51<16:21:39, 5.72s/it]
77%|███████▋ | 33639/43939 [54:16:57<16:22:26, 5.72s/it]
77%|███████▋ | 33640/43939 [54:17:03<16:19:25, 5.71s/it]
77%|███████▋ | 33641/43939 [54:17:08<16:18:22, 5.70s/it]
77%|███████▋ | 33642/43939 [54:17:14<16:30:42, 5.77s/it]
77%|███████▋ | 33643/43939 [54:17:20<16:26:44, 5.75s/it]
77%|███████▋ | 33644/43939 [54:17:26<16:25:16, 5.74s/it]
77%|███████▋ | 33645/43939 [54:17:32<16:25:49, 5.75s/it]
77%|███████▋ | 33646/43939 [54:17:37<16:23:29, 5.73s/it]
77%|███████▋ | 33647/43939 [54:17:43<16:23:09, 5.73s/it]
77%|███████▋ | 33648/43939 [54:17:49<16:20:43, 5.72s/it]
77%|███████▋ | 33649/43939 [54:17:54<16:19:27, 5.71s/it]
77%|███████▋ | 33650/43939 [54:18:00<16:20:56, 5.72s/it]
{'loss': 0.6573, 'grad_norm': 1.3040716648101807, 'learning_rate': 7.897965821001504e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.3, 'epoch': 3.06}
77%|███████▋ | 33650/43939 [54:18:00<16:20:56, 5.72s/it]
77%|███████▋ | 33651/43939 [54:18:06<16:20:36, 5.72s/it]
77%|███████▋ | 33652/43939 [54:18:12<16:21:16, 5.72s/it]
77%|███████▋ | 33653/43939 [54:18:17<16:19:57, 5.72s/it]
77%|███████▋ | 33654/43939 [54:18:23<16:18:56, 5.71s/it]
77%|███████▋ | 33655/43939 [54:18:29<16:17:21, 5.70s/it]
77%|███████▋ | 33656/43939 [54:18:34<16:16:22, 5.70s/it]
77%|███████▋ | 33657/43939 [54:18:40<16:17:12, 5.70s/it]
77%|███████▋ | 33658/43939 [54:18:46<16:19:42, 5.72s/it]
77%|███████▋ | 33659/43939 [54:18:51<16:19:08, 5.71s/it]
77%|███████▋ | 33660/43939 [54:18:57<16:18:10, 5.71s/it]
77%|███████▋ | 33661/43939 [54:19:03<16:15:25, 5.69s/it]
77%|███████▋ | 33662/43939 [54:19:09<16:16:24, 5.70s/it]
77%|███████▋ | 33663/43939 [54:19:15<16:29:19, 5.78s/it]
77%|███████▋ | 33664/43939 [54:19:20<16:27:33, 5.77s/it]
77%|███████▋ | 33665/43939 [54:19:26<16:23:00, 5.74s/it]
77%|███████▋ | 33666/43939 [54:19:32<16:20:52, 5.73s/it]
77%|███████▋ | 33667/43939 [54:19:37<16:18:21, 5.71s/it]
77%|███████▋ | 33668/43939 [54:19:43<16:17:35, 5.71s/it]
77%|███████▋ | 33669/43939 [54:19:49<16:17:26, 5.71s/it]
77%|███████▋ | 33670/43939 [54:19:54<16:18:12, 5.72s/it]
77%|███████▋ | 33671/43939 [54:20:00<16:19:04, 5.72s/it]
77%|███████▋ | 33672/43939 [54:20:06<16:19:42, 5.73s/it]
77%|███████▋ | 33673/43939 [54:20:12<16:18:53, 5.72s/it]
77%|███████▋ | 33674/43939 [54:20:17<16:18:10, 5.72s/it]
77%|███████▋ | 33675/43939 [54:20:23<16:16:34, 5.71s/it]
77%|███████▋ | 33676/43939 [54:20:29<16:14:24, 5.70s/it]
77%|███████▋ | 33677/43939 [54:20:34<16:14:13, 5.70s/it]
77%|███████▋ | 33678/43939 [54:20:40<16:13:47, 5.69s/it]
77%|███████▋ | 33679/43939 [54:20:46<16:15:10, 5.70s/it]
77%|███████▋ | 33680/43939 [54:20:52<16:16:29, 5.71s/it]
77%|███████▋ | 33681/43939 [54:20:57<16:17:28, 5.72s/it]
77%|███████▋ | 33682/43939 [54:21:03<16:16:00, 5.71s/it]
77%|███████▋ | 33683/43939 [54:21:09<16:16:19, 5.71s/it]
77%|███████▋ | 33684/43939 [54:21:15<16:30:27, 5.79s/it]
77%|███████▋ | 33685/43939 [54:21:20<16:26:14, 5.77s/it]
77%|███████▋ | 33686/43939 [54:21:26<16:24:12, 5.76s/it]
77%|███████▋ | 33687/43939 [54:21:32<16:18:41, 5.73s/it]
77%|███████▋ | 33688/43939 [54:21:37<16:17:45, 5.72s/it]
77%|███████▋ | 33689/43939 [54:21:43<16:15:24, 5.71s/it]
77%|███████▋ | 33690/43939 [54:21:49<16:16:01, 5.71s/it]
77%|███████▋ | 33691/43939 [54:21:55<16:15:41, 5.71s/it]
77%|███████▋ | 33692/43939 [54:22:00<16:15:58, 5.71s/it]
77%|███████▋ | 33693/43939 [54:22:06<16:17:30, 5.72s/it]
77%|███████▋ | 33694/43939 [54:22:12<16:15:51, 5.72s/it]
77%|███████▋ | 33695/43939 [54:22:18<16:17:11, 5.72s/it]
77%|███████▋ | 33696/43939 [54:22:23<16:15:24, 5.71s/it]
77%|███████▋ | 33697/43939 [54:22:29<16:15:09, 5.71s/it]
77%|███████▋ | 33698/43939 [54:22:35<16:14:47, 5.71s/it]
77%|███████▋ | 33699/43939 [54:22:40<16:15:36, 5.72s/it]
77%|███████▋ | 33700/43939 [54:22:46<16:13:13, 5.70s/it]
{'loss': 0.6716, 'grad_norm': 1.6446518898010254, 'learning_rate': 7.825669590543738e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.55, 'epoch': 3.07}
77%|███████▋ | 33700/43939 [54:22:46<16:13:13, 5.70s/it]
77%|███████▋ | 33701/43939 [54:22:52<16:15:30, 5.72s/it]
77%|███████▋ | 33702/43939 [54:22:57<16:14:37, 5.71s/it]
77%|███████▋ | 33703/43939 [54:23:03<16:13:12, 5.70s/it]
77%|███████▋ | 33704/43939 [54:23:09<16:13:03, 5.70s/it]
77%|███████▋ | 33705/43939 [54:23:15<16:13:43, 5.71s/it]
77%|███████▋ | 33706/43939 [54:23:21<16:24:51, 5.77s/it]
77%|███████▋ | 33707/43939 [54:23:26<16:20:47, 5.75s/it]
77%|███████▋ | 33708/43939 [54:23:32<16:17:23, 5.73s/it]
77%|███████▋ | 33709/43939 [54:23:38<16:16:01, 5.72s/it]
77%|███████▋ | 33710/43939 [54:23:43<16:15:07, 5.72s/it]
77%|███████▋ | 33711/43939 [54:23:49<16:15:07, 5.72s/it]
77%|███████▋ | 33712/43939 [54:23:55<16:13:31, 5.71s/it]
77%|███████▋ | 33713/43939 [54:24:00<16:12:32, 5.71s/it]
77%|███████▋ | 33714/43939 [54:24:06<16:11:07, 5.70s/it]
77%|███████▋ | 33715/43939 [54:24:12<16:12:49, 5.71s/it]
77%|███████▋ | 33716/43939 [54:24:18<16:13:47, 5.72s/it]
77%|███████▋ | 33717/43939 [54:24:23<16:11:44, 5.70s/it]
77%|███████▋ | 33718/43939 [54:24:29<16:11:28, 5.70s/it]
77%|███████▋ | 33719/43939 [54:24:35<16:13:01, 5.71s/it]
77%|███████▋ | 33720/43939 [54:24:40<16:12:28, 5.71s/it]
77%|███████▋ | 33721/43939 [54:24:46<16:11:07, 5.70s/it]
77%|███████▋ | 33722/43939 [54:24:52<16:10:50, 5.70s/it]
77%|███████▋ | 33723/43939 [54:24:57<16:11:21, 5.70s/it]
77%|███████▋ | 33724/43939 [54:25:03<16:11:16, 5.70s/it]
77%|███████▋ | 33725/43939 [54:25:09<16:11:07, 5.70s/it]
77%|███████▋ | 33726/43939 [54:25:15<16:12:32, 5.71s/it]
77%|███████▋ | 33727/43939 [54:25:21<16:24:16, 5.78s/it]
77%|███████▋ | 33728/43939 [54:25:26<16:19:16, 5.75s/it]
77%|███████▋ | 33729/43939 [54:25:32<16:14:19, 5.73s/it]
77%|███████▋ | 33730/43939 [54:25:38<16:13:27, 5.72s/it]
77%|███████▋ | 33731/43939 [54:25:43<16:14:07, 5.73s/it]
77%|███████▋ | 33732/43939 [54:25:49<16:11:29, 5.71s/it]
77%|███████▋ | 33733/43939 [54:25:55<16:13:11, 5.72s/it]
77%|███████▋ | 33734/43939 [54:26:00<16:11:57, 5.71s/it]
77%|███████▋ | 33735/43939 [54:26:06<16:10:04, 5.70s/it]
77%|███████▋ | 33736/43939 [54:26:12<16:09:45, 5.70s/it]
77%|███████▋ | 33737/43939 [54:26:18<16:10:08, 5.71s/it]
77%|███████▋ | 33738/43939 [54:26:23<16:09:35, 5.70s/it]
77%|███████▋ | 33739/43939 [54:26:29<16:08:23, 5.70s/it]
77%|███████▋ | 33740/43939 [54:26:35<16:08:13, 5.70s/it]
77%|███████▋ | 33741/43939 [54:26:40<16:12:30, 5.72s/it]
77%|███████▋ | 33742/43939 [54:26:46<16:11:41, 5.72s/it]
77%|███████▋ | 33743/43939 [54:26:52<16:12:19, 5.72s/it]
77%|███████▋ | 33744/43939 [54:26:58<16:12:28, 5.72s/it]
77%|███████▋ | 33745/43939 [54:27:03<16:12:28, 5.72s/it]
77%|███████▋ | 33746/43939 [54:27:09<16:11:11, 5.72s/it]
77%|███████▋ | 33747/43939 [54:27:15<16:13:44, 5.73s/it]
77%|███████▋ | 33748/43939 [54:27:21<16:13:16, 5.73s/it]
77%|███████▋ | 33749/43939 [54:27:26<16:24:49, 5.80s/it]
77%|███████▋ | 33750/43939 [54:27:32<16:18:16, 5.76s/it]
{'loss': 0.6368, 'grad_norm': 1.7018547058105469, 'learning_rate': 7.753644325507306e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.4, 'epoch': 3.07}
77%|███████▋ | 33750/43939 [54:27:32<16:18:16, 5.76s/it]
77%|███████▋ | 33751/43939 [54:27:38<16:14:26, 5.74s/it]
77%|███████▋ | 33752/43939 [54:27:44<16:12:57, 5.73s/it]
77%|███████▋ | 33753/43939 [54:27:49<16:14:04, 5.74s/it]
77%|███████▋ | 33754/43939 [54:27:55<16:11:25, 5.72s/it]
77%|███████▋ | 33755/43939 [54:28:01<16:10:19, 5.72s/it]
77%|███████▋ | 33756/43939 [54:28:06<16:08:00, 5.70s/it]
77%|███████▋ | 33757/43939 [54:28:12<16:07:59, 5.70s/it]
77%|███████▋ | 33758/43939 [54:28:18<16:07:51, 5.70s/it]
77%|███████▋ | 33759/43939 [54:28:23<16:07:05, 5.70s/it]
77%|███████▋ | 33760/43939 [54:28:29<16:07:40, 5.70s/it]
77%|███████▋ | 33761/43939 [54:28:35<16:06:45, 5.70s/it]
77%|███████▋ | 33762/43939 [54:28:41<16:09:08, 5.71s/it]
77%|███████▋ | 33763/43939 [54:28:46<16:08:56, 5.71s/it]
77%|███████▋ | 33764/43939 [54:28:52<16:09:12, 5.72s/it]
77%|███████▋ | 33765/43939 [54:28:58<16:07:16, 5.70s/it]
77%|███████▋ | 33766/43939 [54:29:03<16:07:03, 5.70s/it]
77%|███████▋ | 33767/43939 [54:29:09<16:10:18, 5.72s/it]
77%|███████▋ | 33768/43939 [54:29:15<16:09:07, 5.72s/it]
77%|███████▋ | 33769/43939 [54:29:21<16:08:16, 5.71s/it]
77%|███████▋ | 33770/43939 [54:29:26<16:06:57, 5.71s/it]
77%|███████▋ | 33771/43939 [54:29:32<16:16:55, 5.76s/it]
77%|███████▋ | 33772/43939 [54:29:38<16:13:36, 5.75s/it]
77%|███████▋ | 33773/43939 [54:29:44<16:10:32, 5.73s/it]
77%|███████▋ | 33774/43939 [54:29:49<16:08:33, 5.72s/it]
77%|███████▋ | 33775/43939 [54:29:55<16:10:43, 5.73s/it]
77%|███████▋ | 33776/43939 [54:30:01<16:10:13, 5.73s/it]
77%|███████▋ | 33777/43939 [54:30:06<16:10:16, 5.73s/it]
77%|███████▋ | 33778/43939 [54:30:12<16:09:47, 5.73s/it]
77%|███████▋ | 33779/43939 [54:30:18<16:08:31, 5.72s/it]
77%|███████▋ | 33780/43939 [54:30:24<16:08:13, 5.72s/it]
77%|███████▋ | 33781/43939 [54:30:29<16:07:32, 5.71s/it]
77%|███████▋ | 33782/43939 [54:30:35<16:06:37, 5.71s/it]
77%|███████▋ | 33783/43939 [54:30:41<16:10:45, 5.74s/it]
77%|███████▋ | 33784/43939 [54:30:46<16:07:25, 5.72s/it]
77%|███████▋ | 33785/43939 [54:30:52<16:06:42, 5.71s/it]
77%|███████▋ | 33786/43939 [54:30:58<16:06:35, 5.71s/it]
77%|███████▋ | 33787/43939 [54:31:04<16:04:40, 5.70s/it]
77%|███████▋ | 33788/43939 [54:31:09<16:02:53, 5.69s/it]
77%|███████▋ | 33789/43939 [54:31:15<16:06:10, 5.71s/it]
77%|███████▋ | 33790/43939 [54:31:21<16:06:12, 5.71s/it]
77%|███████▋ | 33791/43939 [54:31:26<16:07:07, 5.72s/it]
77%|███████▋ | 33792/43939 [54:31:32<16:18:59, 5.79s/it]
77%|███████▋ | 33793/43939 [54:31:38<16:12:12, 5.75s/it]
77%|███████▋ | 33794/43939 [54:31:44<16:13:54, 5.76s/it]
77%|███████▋ | 33795/43939 [54:31:50<16:09:57, 5.74s/it]
77%|███████▋ | 33796/43939 [54:31:55<16:08:02, 5.73s/it]
77%|███████▋ | 33797/43939 [54:32:01<16:07:05, 5.72s/it]
77%|███████▋ | 33798/43939 [54:32:07<16:04:56, 5.71s/it]
77%|███████▋ | 33799/43939 [54:32:12<16:05:10, 5.71s/it]
77%|███████▋ | 33800/43939 [54:32:18<16:04:43, 5.71s/it]
{'loss': 0.6798, 'grad_norm': 1.6681090593338013, 'learning_rate': 7.681891162260015e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.25, 'epoch': 3.08}
77%|███████▋ | 33800/43939 [54:32:18<16:04:43, 5.71s/it]
77%|███████▋ | 33801/43939 [54:32:24<16:03:29, 5.70s/it]
77%|███████▋ | 33802/43939 [54:32:29<16:04:55, 5.71s/it]
77%|███████▋ | 33803/43939 [54:32:35<16:04:04, 5.71s/it]
77%|███████▋ | 33804/43939 [54:32:41<16:04:18, 5.71s/it]
77%|███████▋ | 33805/43939 [54:32:47<16:04:20, 5.71s/it]
77%|███████▋ | 33806/43939 [54:32:52<16:02:29, 5.70s/it]
77%|███████▋ | 33807/43939 [54:32:58<16:01:40, 5.69s/it]
77%|███████▋ | 33808/43939 [54:33:04<16:03:06, 5.70s/it]
77%|███████▋ | 33809/43939 [54:33:09<16:03:00, 5.70s/it]
77%|███████▋ | 33810/43939 [54:33:15<16:03:19, 5.71s/it]
77%|███████▋ | 33811/43939 [54:33:21<16:01:53, 5.70s/it]
77%|███████▋ | 33812/43939 [54:33:26<16:02:16, 5.70s/it]
77%|███████▋ | 33813/43939 [54:33:32<16:03:07, 5.71s/it]
77%|███████▋ | 33814/43939 [54:33:38<16:11:42, 5.76s/it]
77%|███████▋ | 33815/43939 [54:33:44<16:09:25, 5.75s/it]
77%|███████▋ | 33816/43939 [54:33:50<16:10:39, 5.75s/it]
77%|███████▋ | 33817/43939 [54:33:55<16:07:18, 5.73s/it]
77%|███████▋ | 33818/43939 [54:34:01<16:09:33, 5.75s/it]
77%|███████▋ | 33819/43939 [54:34:07<16:06:56, 5.73s/it]
77%|███████▋ | 33820/43939 [54:34:12<16:08:03, 5.74s/it]
77%|███████▋ | 33821/43939 [54:34:18<16:07:17, 5.74s/it]
77%|███████▋ | 33822/43939 [54:34:24<16:05:15, 5.72s/it]
77%|███████▋ | 33823/43939 [54:34:30<16:02:08, 5.71s/it]
77%|███████▋ | 33824/43939 [54:34:35<16:03:48, 5.72s/it]
77%|███████▋ | 33825/43939 [54:34:41<16:03:15, 5.71s/it]
77%|███████▋ | 33826/43939 [54:34:47<16:01:57, 5.71s/it]
77%|███████▋ | 33827/43939 [54:34:52<16:01:36, 5.71s/it]
77%|███████▋ | 33828/43939 [54:34:58<15:58:53, 5.69s/it]
77%|███████▋ | 33829/43939 [54:35:04<16:00:23, 5.70s/it]
77%|███████▋ | 33830/43939 [54:35:09<16:00:27, 5.70s/it]
77%|███████▋ | 33831/43939 [54:35:15<16:00:55, 5.70s/it]
77%|███████▋ | 33832/43939 [54:35:21<16:01:09, 5.71s/it]
77%|███████▋ | 33833/43939 [54:35:27<16:02:08, 5.71s/it]
77%|███████▋ | 33834/43939 [54:35:32<15:58:36, 5.69s/it]
77%|███████▋ | 33835/43939 [54:35:38<16:12:49, 5.78s/it]
77%|███████▋ | 33836/43939 [54:35:44<16:08:27, 5.75s/it]
77%|███████▋ | 33837/43939 [54:35:50<16:05:34, 5.73s/it]
77%|███████▋ | 33838/43939 [54:35:55<16:04:57, 5.73s/it]
77%|███████▋ | 33839/43939 [54:36:01<16:02:08, 5.72s/it]
77%|███████▋ | 33840/43939 [54:36:07<15:59:49, 5.70s/it]
77%|███████▋ | 33841/43939 [54:36:12<15:57:31, 5.69s/it]
77%|███████▋ | 33842/43939 [54:36:18<16:00:58, 5.71s/it]
77%|███████▋ | 33843/43939 [54:36:24<15:59:13, 5.70s/it]
77%|███████▋ | 33844/43939 [54:36:29<15:57:10, 5.69s/it]
77%|███████▋ | 33845/43939 [54:36:35<15:58:58, 5.70s/it]
77%|███████▋ | 33846/43939 [54:36:41<15:59:11, 5.70s/it]
77%|███████▋ | 33847/43939 [54:36:47<16:00:44, 5.71s/it]
77%|███████▋ | 33848/43939 [54:36:52<16:02:04, 5.72s/it]
77%|███████▋ | 33849/43939 [54:36:58<16:00:51, 5.71s/it]
77%|███████▋ | 33850/43939 [54:37:04<16:01:45, 5.72s/it]
{'loss': 0.6555, 'grad_norm': 1.5188624858856201, 'learning_rate': 7.610411232876638e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.0, 'epoch': 3.08}
77%|███████▋ | 33850/43939 [54:37:04<16:01:45, 5.72s/it]
77%|███████▋ | 33851/43939 [54:37:10<16:01:10, 5.72s/it]
77%|███████▋ | 33852/43939 [54:37:15<16:00:39, 5.71s/it]
77%|███████▋ | 33853/43939 [54:37:21<15:58:27, 5.70s/it]
77%|███████▋ | 33854/43939 [54:37:27<16:00:19, 5.71s/it]
77%|███████▋ | 33855/43939 [54:37:32<16:00:59, 5.72s/it]
77%|███████▋ | 33856/43939 [54:37:38<15:59:56, 5.71s/it]
77%|███████▋ | 33857/43939 [54:37:44<16:09:47, 5.77s/it]
77%|███████▋ | 33858/43939 [54:37:50<16:07:22, 5.76s/it]
77%|███████▋ | 33859/43939 [54:37:55<16:04:01, 5.74s/it]
77%|███████▋ | 33860/43939 [54:38:01<16:02:36, 5.73s/it]
77%|███████▋ | 33861/43939 [54:38:07<16:02:10, 5.73s/it]
77%|███████▋ | 33862/43939 [54:38:13<16:00:45, 5.72s/it]
77%|███████▋ | 33863/43939 [54:38:18<15:58:16, 5.71s/it]
77%|███████▋ | 33864/43939 [54:38:24<15:58:27, 5.71s/it]
77%|███████▋ | 33865/43939 [54:38:30<15:57:41, 5.70s/it]
77%|███████▋ | 33866/43939 [54:38:35<15:57:19, 5.70s/it]
77%|███████▋ | 33867/43939 [54:38:41<15:54:30, 5.69s/it]
77%|███████▋ | 33868/43939 [54:38:47<15:55:31, 5.69s/it]
77%|███████▋ | 33869/43939 [54:38:52<15:57:58, 5.71s/it]
77%|███████▋ | 33870/43939 [54:38:58<15:57:19, 5.70s/it]
77%|███████▋ | 33871/43939 [54:39:04<15:54:59, 5.69s/it]
77%|███████▋ | 33872/43939 [54:39:09<15:55:43, 5.70s/it]
77%|███████▋ | 33873/43939 [54:39:15<15:58:24, 5.71s/it]
77%|███████▋ | 33874/43939 [54:39:21<16:01:06, 5.73s/it]
77%|███████▋ | 33875/43939 [54:39:27<16:00:15, 5.72s/it]
77%|███████▋ | 33876/43939 [54:39:32<15:59:53, 5.72s/it]
77%|███████▋ | 33877/43939 [54:39:38<15:58:54, 5.72s/it]
77%|███████▋ | 33878/43939 [54:39:44<16:10:45, 5.79s/it]
77%|███████▋ | 33879/43939 [54:39:50<16:04:34, 5.75s/it]
77%|███████▋ | 33880/43939 [54:39:55<16:01:17, 5.73s/it]
77%|███████▋ | 33881/43939 [54:40:01<16:00:12, 5.73s/it]
77%|███████▋ | 33882/43939 [54:40:07<15:58:15, 5.72s/it]
77%|███████▋ | 33883/43939 [54:40:13<15:58:02, 5.72s/it]
77%|███████▋ | 33884/43939 [54:40:18<15:56:13, 5.71s/it]
77%|███████▋ | 33885/43939 [54:40:24<15:59:38, 5.73s/it]
77%|███████▋ | 33886/43939 [54:40:30<15:57:40, 5.72s/it]
77%|███████▋ | 33887/43939 [54:40:35<15:56:47, 5.71s/it]
77%|███████▋ | 33888/43939 [54:40:41<15:56:17, 5.71s/it]
77%|███████▋ | 33889/43939 [54:40:47<15:54:48, 5.70s/it]
77%|███████▋ | 33890/43939 [54:40:53<15:55:59, 5.71s/it]
77%|███████▋ | 33891/43939 [54:40:58<15:56:27, 5.71s/it]
77%|███████▋ | 33892/43939 [54:41:04<15:56:12, 5.71s/it]
77%|███████▋ | 33893/43939 [54:41:10<15:56:53, 5.72s/it]
77%|███████▋ | 33894/43939 [54:41:15<15:56:39, 5.71s/it]
77%|███████▋ | 33895/43939 [54:41:21<15:57:49, 5.72s/it]
77%|███████▋ | 33896/43939 [54:41:27<15:56:28, 5.71s/it]
77%|███████▋ | 33897/43939 [54:41:33<15:56:50, 5.72s/it]
77%|███████▋ | 33898/43939 [54:41:38<15:56:01, 5.71s/it]
77%|███████▋ | 33899/43939 [54:41:44<15:54:48, 5.71s/it]
77%|███████▋ | 33900/43939 [54:41:50<16:05:22, 5.77s/it]
{'loss': 0.6636, 'grad_norm': 1.7600923776626587, 'learning_rate': 7.539205665121035e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.09, 'epoch': 3.09}
77%|███████▋ | 33900/43939 [54:41:50<16:05:22, 5.77s/it]
77%|███████▋ | 33901/43939 [54:41:56<16:01:25, 5.75s/it]
77%|███████▋ | 33902/43939 [54:42:01<16:00:44, 5.74s/it]
77%|███████▋ | 33903/43939 [54:42:07<15:59:41, 5.74s/it]
77%|███████▋ | 33904/43939 [54:42:13<15:56:51, 5.72s/it]
77%|███████▋ | 33905/43939 [54:42:18<15:55:06, 5.71s/it]
77%|███████▋ | 33906/43939 [54:42:24<15:53:56, 5.70s/it]
77%|███████▋ | 33907/43939 [54:42:30<15:52:20, 5.70s/it]
77%|███████▋ | 33908/43939 [54:42:35<15:53:57, 5.71s/it]
77%|███████▋ | 33909/43939 [54:42:41<15:53:59, 5.71s/it]
77%|███████▋ | 33910/43939 [54:42:47<15:57:43, 5.73s/it]
77%|███████▋ | 33911/43939 [54:42:53<15:57:50, 5.73s/it]
77%|███████▋ | 33912/43939 [54:42:58<15:59:28, 5.74s/it]
77%|███████▋ | 33913/43939 [54:43:04<15:56:49, 5.73s/it]
77%|███████▋ | 33914/43939 [54:43:10<15:55:34, 5.72s/it]
77%|███████▋ | 33915/43939 [54:43:16<15:57:15, 5.73s/it]
77%|███████▋ | 33916/43939 [54:43:21<15:56:10, 5.72s/it]
77%|███████▋ | 33917/43939 [54:43:27<15:55:44, 5.72s/it]
77%|███████▋ | 33918/43939 [54:43:33<15:52:43, 5.70s/it]
77%|███████▋ | 33919/43939 [54:43:38<15:52:42, 5.70s/it]
77%|███████▋ | 33920/43939 [54:43:44<15:50:53, 5.69s/it]
77%|███████▋ | 33921/43939 [54:43:50<16:04:42, 5.78s/it]
77%|███████▋ | 33922/43939 [54:43:56<16:00:55, 5.76s/it]
77%|███████▋ | 33923/43939 [54:44:01<15:58:55, 5.74s/it]
77%|███████▋ | 33924/43939 [54:44:07<15:56:43, 5.73s/it]
77%|███████▋ | 33925/43939 [54:44:13<15:57:28, 5.74s/it]
77%|███████▋ | 33926/43939 [54:44:19<15:54:53, 5.72s/it]
77%|███████▋ | 33927/43939 [54:44:24<15:53:56, 5.72s/it]
77%|███████▋ | 33928/43939 [54:44:30<15:51:34, 5.70s/it]
77%|███████▋ | 33929/43939 [54:44:36<15:52:50, 5.71s/it]
77%|███████▋ | 33930/43939 [54:44:41<15:53:07, 5.71s/it]
77%|███████▋ | 33931/43939 [54:44:47<15:57:07, 5.74s/it]
77%|███████▋ | 33932/43939 [54:44:53<15:57:08, 5.74s/it]
77%|███████▋ | 33933/43939 [54:44:59<15:56:35, 5.74s/it]
77%|███████▋ | 33934/43939 [54:45:04<15:56:54, 5.74s/it]
77%|███████▋ | 33935/43939 [54:45:10<15:54:11, 5.72s/it]
77%|███████▋ | 33936/43939 [54:45:16<15:55:11, 5.73s/it]
77%|███████▋ | 33937/43939 [54:45:22<15:55:21, 5.73s/it]
77%|███████▋ | 33938/43939 [54:45:27<15:55:36, 5.73s/it]
77%|███████▋ | 33939/43939 [54:45:33<15:52:39, 5.72s/it]
77%|███████▋ | 33940/43939 [54:45:39<15:55:16, 5.73s/it]
77%|███████▋ | 33941/43939 [54:45:44<15:53:12, 5.72s/it]
77%|███████▋ | 33942/43939 [54:45:50<15:51:01, 5.71s/it]
77%|███████▋ | 33943/43939 [54:45:56<16:02:02, 5.77s/it]
77%|███████▋ | 33944/43939 [54:46:02<15:58:30, 5.75s/it]
77%|███████▋ | 33945/43939 [54:46:08<15:55:47, 5.74s/it]
77%|███████▋ | 33946/43939 [54:46:13<15:55:39, 5.74s/it]
77%|███████▋ | 33947/43939 [54:46:19<15:52:16, 5.72s/it]
77%|███████▋ | 33948/43939 [54:46:25<15:50:36, 5.71s/it]
77%|███████▋ | 33949/43939 [54:46:30<15:50:07, 5.71s/it]
77%|███████▋ | 33950/43939 [54:46:36<15:50:25, 5.71s/it]
{'loss': 0.6543, 'grad_norm': 1.9836052656173706, 'learning_rate': 7.468275582428369e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 524.05, 'epoch': 3.09}
77%|███████▋ | 33950/43939 [54:46:36<15:50:25, 5.71s/it]
77%|███████▋ | 33951/43939 [54:46:42<15:49:48, 5.71s/it]
77%|███████▋ | 33952/43939 [54:46:47<15:50:57, 5.71s/it]
77%|███████▋ | 33953/43939 [54:46:53<15:49:06, 5.70s/it]
77%|███████▋ | 33954/43939 [54:46:59<15:50:12, 5.71s/it]
77%|███████▋ | 33955/43939 [54:47:05<15:49:56, 5.71s/it]
77%|███████▋ | 33956/43939 [54:47:10<15:50:27, 5.71s/it]
77%|███████▋ | 33957/43939 [54:47:16<15:49:41, 5.71s/it]
77%|███████▋ | 33958/43939 [54:47:22<15:48:17, 5.70s/it]
77%|███████▋ | 33959/43939 [54:47:27<15:48:11, 5.70s/it]
77%|███████▋ | 33960/43939 [54:47:33<15:48:16, 5.70s/it]
77%|███████▋ | 33961/43939 [54:47:39<15:50:12, 5.71s/it]
77%|███████▋ | 33962/43939 [54:47:45<15:49:33, 5.71s/it]
77%|███████▋ | 33963/43939 [54:47:50<15:59:50, 5.77s/it]
77%|███████▋ | 33964/43939 [54:47:56<15:58:16, 5.76s/it]
77%|███████▋ | 33965/43939 [54:48:02<15:54:28, 5.74s/it]
77%|███████▋ | 33966/43939 [54:48:08<15:52:02, 5.73s/it]
77%|███████▋ | 33967/43939 [54:48:13<15:50:35, 5.72s/it]
77%|███████▋ | 33968/43939 [54:48:19<15:49:32, 5.71s/it]
77%|███████▋ | 33969/43939 [54:48:25<15:50:50, 5.72s/it]
77%|███████▋ | 33970/43939 [54:48:30<15:49:23, 5.71s/it]
77%|███████▋ | 33971/43939 [54:48:36<15:48:27, 5.71s/it]
77%|███████▋ | 33972/43939 [54:48:42<15:49:05, 5.71s/it]
77%|███████▋ | 33973/43939 [54:48:48<15:49:46, 5.72s/it]
77%|███████▋ | 33974/43939 [54:48:53<15:47:55, 5.71s/it]
77%|███████▋ | 33975/43939 [54:48:59<15:46:37, 5.70s/it]
77%|███████▋ | 33976/43939 [54:49:05<15:49:06, 5.72s/it]
77%|███████▋ | 33977/43939 [54:49:10<15:47:57, 5.71s/it]
77%|███████▋ | 33978/43939 [54:49:16<15:47:25, 5.71s/it]
77%|███████▋ | 33979/43939 [54:49:22<15:46:34, 5.70s/it]
77%|███████▋ | 33980/43939 [54:49:28<15:48:39, 5.72s/it]
77%|███████▋ | 33981/43939 [54:49:33<15:46:26, 5.70s/it]
77%|███████▋ | 33982/43939 [54:49:39<15:48:45, 5.72s/it]
77%|███████▋ | 33983/43939 [54:49:45<15:46:47, 5.71s/it]
77%|███████▋ | 33984/43939 [54:49:50<15:43:35, 5.69s/it]
77%|███████▋ | 33985/43939 [54:49:56<15:56:17, 5.76s/it]
77%|███████▋ | 33986/43939 [54:50:02<15:52:14, 5.74s/it]
77%|███████▋ | 33987/43939 [54:50:08<15:52:19, 5.74s/it]
77%|███████▋ | 33988/43939 [54:50:13<15:49:18, 5.72s/it]
77%|███████▋ | 33989/43939 [54:50:19<15:47:58, 5.72s/it]
77%|███████▋ | 33990/43939 [54:50:25<15:46:44, 5.71s/it]
77%|███████▋ | 33991/43939 [54:50:30<15:46:36, 5.71s/it]
77%|███████▋ | 33992/43939 [54:50:36<15:47:47, 5.72s/it]
77%|███████▋ | 33993/43939 [54:50:42<15:46:32, 5.71s/it]
77%|███████▋ | 33994/43939 [54:50:48<15:45:39, 5.71s/it]
77%|███████▋ | 33995/43939 [54:50:53<15:45:40, 5.71s/it]
77%|███████▋ | 33996/43939 [54:50:59<15:45:46, 5.71s/it]
77%|███████▋ | 33997/43939 [54:51:05<15:44:20, 5.70s/it]
77%|███████▋ | 33998/43939 [54:51:10<15:44:45, 5.70s/it]
77%|███████▋ | 33999/43939 [54:51:16<15:45:37, 5.71s/it]
77%|███████▋ | 34000/43939 [54:51:22<15:46:12, 5.71s/it]
{'loss': 0.6395, 'grad_norm': 1.8480523824691772, 'learning_rate': 7.3976221038873775e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 445.77, 'epoch': 3.1}
77%|███████▋ | 34000/43939 [54:51:22<15:46:12, 5.71s/it]
77%|███████▋ | 34001/43939 [54:51:27<15:44:32, 5.70s/it]
77%|███████▋ | 34002/43939 [54:51:33<15:45:13, 5.71s/it]
77%|███████▋ | 34003/43939 [54:51:39<15:46:06, 5.71s/it]
77%|███████▋ | 34004/43939 [54:51:45<15:46:43, 5.72s/it]
77%|███████▋ | 34005/43939 [54:51:50<15:47:35, 5.72s/it]
77%|███████▋ | 34006/43939 [54:51:56<15:45:44, 5.71s/it]
77%|███████▋ | 34007/43939 [54:52:02<15:56:21, 5.78s/it]
77%|███████▋ | 34008/43939 [54:52:08<15:53:03, 5.76s/it]
77%|███████▋ | 34009/43939 [54:52:13<15:48:39, 5.73s/it]
77%|███████▋ | 34010/43939 [54:52:19<15:47:58, 5.73s/it]
77%|███████▋ | 34011/43939 [54:52:25<15:46:47, 5.72s/it]
77%|███████▋ | 34012/43939 [54:52:30<15:44:09, 5.71s/it]
77%|███████▋ | 34013/43939 [54:52:36<15:44:18, 5.71s/it]
77%|███████▋ | 34014/43939 [54:52:42<15:43:29, 5.70s/it]
77%|███████▋ | 34015/43939 [54:52:48<15:44:02, 5.71s/it]
77%|███████▋ | 34016/43939 [54:52:53<15:42:32, 5.70s/it]
77%|███████▋ | 34017/43939 [54:52:59<15:42:29, 5.70s/it]
77%|███████▋ | 34018/43939 [54:53:05<15:43:19, 5.71s/it]
77%|███████▋ | 34019/43939 [54:53:10<15:46:24, 5.72s/it]
77%|███████▋ | 34020/43939 [54:53:16<15:44:26, 5.71s/it]
77%|███████▋ | 34021/43939 [54:53:22<15:44:55, 5.72s/it]
77%|███████▋ | 34022/43939 [54:53:28<15:42:55, 5.70s/it]
77%|███████▋ | 34023/43939 [54:53:33<15:43:35, 5.71s/it]
77%|███████▋ | 34024/43939 [54:53:39<15:44:41, 5.72s/it]
77%|███████▋ | 34025/43939 [54:53:45<15:42:28, 5.70s/it]
77%|███████▋ | 34026/43939 [54:53:50<15:40:24, 5.69s/it]
77%|███████▋ | 34027/43939 [54:53:56<15:43:27, 5.71s/it]
77%|███████▋ | 34028/43939 [54:54:02<15:52:02, 5.76s/it]
77%|███████▋ | 34029/43939 [54:54:08<15:49:04, 5.75s/it]
77%|███████▋ | 34030/43939 [54:54:13<15:47:28, 5.74s/it]
77%|███████▋ | 34031/43939 [54:54:19<15:45:27, 5.73s/it]
77%|███████▋ | 34032/43939 [54:54:25<15:45:56, 5.73s/it]
77%|███████▋ | 34033/43939 [54:54:31<15:45:04, 5.72s/it]
77%|███████▋ | 34034/43939 [54:54:36<15:42:41, 5.71s/it]
77%|███████▋ | 34035/43939 [54:54:42<15:43:25, 5.72s/it]
77%|███████▋ | 34036/43939 [54:54:48<15:42:21, 5.71s/it]
77%|███████▋ | 34037/43939 [54:54:53<15:42:54, 5.71s/it]
77%|███████▋ | 34038/43939 [54:54:59<15:42:18, 5.71s/it]
77%|███████▋ | 34039/43939 [54:55:05<15:41:49, 5.71s/it]
77%|███████▋ | 34040/43939 [54:55:11<15:42:34, 5.71s/it]
77%|███████▋ | 34041/43939 [54:55:16<15:42:12, 5.71s/it]
77%|███████▋ | 34042/43939 [54:55:22<15:40:15, 5.70s/it]
77%|███████▋ | 34043/43939 [54:55:28<15:39:48, 5.70s/it]
77%|███████▋ | 34044/43939 [54:55:33<15:39:55, 5.70s/it]
77%|███████▋ | 34045/43939 [54:55:39<15:39:12, 5.70s/it]
77%|███████▋ | 34046/43939 [54:55:45<15:37:15, 5.68s/it]
77%|███████▋ | 34047/43939 [54:55:50<15:37:59, 5.69s/it]
77%|███████▋ | 34048/43939 [54:55:56<15:36:03, 5.68s/it]
77%|███████▋ | 34049/43939 [54:56:02<15:37:03, 5.68s/it]
77%|███████▋ | 34050/43939 [54:56:08<15:47:10, 5.75s/it]
{'loss': 0.635, 'grad_norm': 1.5933022499084473, 'learning_rate': 7.327246344222716e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.74, 'epoch': 3.1}
77%|███████▋ | 34050/43939 [54:56:08<15:47:10, 5.75s/it]
77%|███████▋ | 34051/43939 [54:56:13<15:46:18, 5.74s/it]
77%|███████▋ | 34052/43939 [54:56:19<15:42:16, 5.72s/it]
78%|███████▊ | 34053/43939 [54:56:25<15:39:49, 5.70s/it]
78%|███████▊ | 34054/43939 [54:56:30<15:39:44, 5.70s/it]
78%|███████▊ | 34055/43939 [54:56:36<15:39:09, 5.70s/it]
78%|███████▊ | 34056/43939 [54:56:42<15:38:52, 5.70s/it]
78%|███████▊ | 34057/43939 [54:56:47<15:37:21, 5.69s/it]
78%|███████▊ | 34058/43939 [54:56:53<15:36:35, 5.69s/it]
78%|███████▊ | 34059/43939 [54:56:59<15:37:14, 5.69s/it]
78%|███████▊ | 34060/43939 [54:57:05<15:39:35, 5.71s/it]
78%|███████▊ | 34061/43939 [54:57:10<15:41:24, 5.72s/it]
78%|███████▊ | 34062/43939 [54:57:16<15:41:29, 5.72s/it]
78%|███████▊ | 34063/43939 [54:57:22<15:40:56, 5.72s/it]
78%|███████▊ | 34064/43939 [54:57:27<15:39:19, 5.71s/it]
78%|███████▊ | 34065/43939 [54:57:33<15:38:34, 5.70s/it]
78%|███████▊ | 34066/43939 [54:57:39<15:37:28, 5.70s/it]
78%|███████▊ | 34067/43939 [54:57:44<15:38:09, 5.70s/it]
78%|███████▊ | 34068/43939 [54:57:50<15:37:25, 5.70s/it]
78%|███████▊ | 34069/43939 [54:57:56<15:38:38, 5.71s/it]
78%|███████▊ | 34070/43939 [54:58:02<15:37:51, 5.70s/it]
78%|███████▊ | 34071/43939 [54:58:07<15:47:14, 5.76s/it]
78%|███████▊ | 34072/43939 [54:58:13<15:45:07, 5.75s/it]
78%|███████▊ | 34073/43939 [54:58:19<15:42:55, 5.73s/it]
78%|███████▊ | 34074/43939 [54:58:25<15:40:56, 5.72s/it]
78%|███████▊ | 34075/43939 [54:58:30<15:37:36, 5.70s/it]
78%|███████▊ | 34076/43939 [54:58:36<15:36:40, 5.70s/it]
78%|███████▊ | 34077/43939 [54:58:42<15:39:01, 5.71s/it]
78%|███████▊ | 34078/43939 [54:58:47<15:38:24, 5.71s/it]
78%|███████▊ | 34079/43939 [54:58:53<15:37:44, 5.71s/it]
78%|███████▊ | 34080/43939 [54:58:59<15:36:39, 5.70s/it]
78%|███████▊ | 34081/43939 [54:59:05<15:38:32, 5.71s/it]
78%|███████▊ | 34082/43939 [54:59:10<15:37:20, 5.71s/it]
78%|███████▊ | 34083/43939 [54:59:16<15:38:22, 5.71s/it]
78%|███████▊ | 34084/43939 [54:59:22<15:38:22, 5.71s/it]
78%|███████▊ | 34085/43939 [54:59:27<15:37:35, 5.71s/it]
78%|███████▊ | 34086/43939 [54:59:33<15:37:31, 5.71s/it]
78%|███████▊ | 34087/43939 [54:59:39<15:38:06, 5.71s/it]
78%|███████▊ | 34088/43939 [54:59:44<15:37:33, 5.71s/it]
78%|███████▊ | 34089/43939 [54:59:50<15:38:00, 5.71s/it]
78%|███████▊ | 34090/43939 [54:59:56<15:35:33, 5.70s/it]
78%|███████▊ | 34091/43939 [55:00:02<15:37:09, 5.71s/it]
78%|███████▊ | 34092/43939 [55:00:07<15:37:27, 5.71s/it]
78%|███████▊ | 34093/43939 [55:00:13<15:48:22, 5.78s/it]
78%|███████▊ | 34094/43939 [55:00:19<15:46:03, 5.77s/it]
78%|███████▊ | 34095/43939 [55:00:25<15:42:31, 5.74s/it]
78%|███████▊ | 34096/43939 [55:00:30<15:40:35, 5.73s/it]
78%|███████▊ | 34097/43939 [55:00:36<15:39:39, 5.73s/it]
78%|███████▊ | 34098/43939 [55:00:42<15:41:36, 5.74s/it]
78%|███████▊ | 34099/43939 [55:00:48<15:40:16, 5.73s/it]
78%|███████▊ | 34100/43939 [55:00:53<15:40:16, 5.73s/it]
{'loss': 0.6424, 'grad_norm': 1.4891561269760132, 'learning_rate': 7.2571494137773754e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 467.57, 'epoch': 3.1}
78%|███████▊ | 34100/43939 [55:00:53<15:40:16, 5.73s/it]
78%|███████▊ | 34101/43939 [55:00:59<15:39:41, 5.73s/it]
78%|███████▊ | 34102/43939 [55:01:05<15:39:07, 5.73s/it]
78%|███████▊ | 34103/43939 [55:01:11<15:39:45, 5.73s/it]
78%|███████▊ | 34104/43939 [55:01:16<15:42:41, 5.75s/it]
78%|███████▊ | 34105/43939 [55:01:22<15:39:17, 5.73s/it]
78%|███████▊ | 34106/43939 [55:01:28<15:37:29, 5.72s/it]
78%|███████▊ | 34107/43939 [55:01:33<15:36:48, 5.72s/it]
78%|███████▊ | 34108/43939 [55:01:39<15:33:06, 5.69s/it]
78%|███████▊ | 34109/43939 [55:01:45<15:34:34, 5.70s/it]
78%|███████▊ | 34110/43939 [55:01:51<15:36:23, 5.72s/it]
78%|███████▊ | 34111/43939 [55:01:56<15:35:15, 5.71s/it]
78%|███████▊ | 34112/43939 [55:02:02<15:37:01, 5.72s/it]
78%|███████▊ | 34113/43939 [55:02:08<15:36:43, 5.72s/it]
78%|███████▊ | 34114/43939 [55:02:14<15:47:33, 5.79s/it]
78%|███████▊ | 34115/43939 [55:02:19<15:43:04, 5.76s/it]
78%|███████▊ | 34116/43939 [55:02:25<15:40:50, 5.75s/it]
78%|███████▊ | 34117/43939 [55:02:31<15:37:47, 5.73s/it]
78%|███████▊ | 34118/43939 [55:02:36<15:38:17, 5.73s/it]
78%|███████▊ | 34119/43939 [55:02:42<15:35:43, 5.72s/it]
78%|███████▊ | 34120/43939 [55:02:48<15:34:22, 5.71s/it]
78%|███████▊ | 34121/43939 [55:02:53<15:31:27, 5.69s/it]
78%|███████▊ | 34122/43939 [55:02:59<15:31:29, 5.69s/it]
78%|███████▊ | 34123/43939 [55:03:05<15:31:25, 5.69s/it]
78%|███████▊ | 34124/43939 [55:03:11<15:30:05, 5.69s/it]
78%|███████▊ | 34125/43939 [55:03:16<15:30:33, 5.69s/it]
78%|███████▊ | 34126/43939 [55:03:22<15:33:50, 5.71s/it]
78%|███████▊ | 34127/43939 [55:03:28<15:32:51, 5.70s/it]
78%|███████▊ | 34128/43939 [55:03:33<15:30:27, 5.69s/it]
78%|███████▊ | 34129/43939 [55:03:39<15:31:40, 5.70s/it]
78%|███████▊ | 34130/43939 [55:03:45<15:31:41, 5.70s/it]
78%|███████▊ | 34131/43939 [55:03:50<15:31:35, 5.70s/it]
78%|███████▊ | 34132/43939 [55:03:56<15:31:56, 5.70s/it]
78%|███████▊ | 34133/43939 [55:04:02<15:32:28, 5.71s/it]
78%|███████▊ | 34134/43939 [55:04:08<15:31:40, 5.70s/it]
78%|███████▊ | 34135/43939 [55:04:13<15:29:09, 5.69s/it]
78%|███████▊ | 34136/43939 [55:04:19<15:43:00, 5.77s/it]
78%|███████▊ | 34137/43939 [55:04:25<15:37:27, 5.74s/it]
78%|███████▊ | 34138/43939 [55:04:31<15:35:58, 5.73s/it]
78%|███████▊ | 34139/43939 [55:04:36<15:32:59, 5.71s/it]
78%|███████▊ | 34140/43939 [55:04:42<15:32:42, 5.71s/it]
78%|███████▊ | 34141/43939 [55:04:48<15:30:51, 5.70s/it]
78%|███████▊ | 34142/43939 [55:04:53<15:29:56, 5.70s/it]
78%|███████▊ | 34143/43939 [55:04:59<15:32:12, 5.71s/it]
78%|███████▊ | 34144/43939 [55:05:05<15:31:21, 5.71s/it]
78%|███████▊ | 34145/43939 [55:05:10<15:31:49, 5.71s/it]
78%|███████▊ | 34146/43939 [55:05:16<15:33:55, 5.72s/it]
78%|███████▊ | 34147/43939 [55:05:22<15:31:46, 5.71s/it]
78%|███████▊ | 34148/43939 [55:05:28<15:30:32, 5.70s/it]
78%|███████▊ | 34149/43939 [55:05:33<15:29:38, 5.70s/it]
78%|███████▊ | 34150/43939 [55:05:39<15:32:05, 5.71s/it]
{'loss': 0.6589, 'grad_norm': 1.3638157844543457, 'learning_rate': 7.187332418495152e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.34, 'epoch': 3.11}
78%|███████▊ | 34150/43939 [55:05:39<15:32:05, 5.71s/it]
78%|███████▊ | 34151/43939 [55:05:45<15:31:34, 5.71s/it]
78%|███████▊ | 34152/43939 [55:05:50<15:30:59, 5.71s/it]
78%|███████▊ | 34153/43939 [55:05:56<15:30:55, 5.71s/it]
78%|███████▊ | 34154/43939 [55:06:02<15:29:43, 5.70s/it]
78%|███████▊ | 34155/43939 [55:06:08<15:29:12, 5.70s/it]
78%|███████▊ | 34156/43939 [55:06:13<15:28:52, 5.70s/it]
78%|███████▊ | 34157/43939 [55:06:19<15:38:39, 5.76s/it]
78%|███████▊ | 34158/43939 [55:06:25<15:35:35, 5.74s/it]
78%|███████▊ | 34159/43939 [55:06:31<15:34:09, 5.73s/it]
78%|███████▊ | 34160/43939 [55:06:36<15:34:19, 5.73s/it]
78%|███████▊ | 34161/43939 [55:06:42<15:32:30, 5.72s/it]
78%|███████▊ | 34162/43939 [55:06:48<15:32:19, 5.72s/it]
78%|███████▊ | 34163/43939 [55:06:53<15:32:05, 5.72s/it]
78%|███████▊ | 34164/43939 [55:06:59<15:32:02, 5.72s/it]
78%|███████▊ | 34165/43939 [55:07:05<15:30:25, 5.71s/it]
78%|███████▊ | 34166/43939 [55:07:10<15:29:21, 5.71s/it]
78%|███████▊ | 34167/43939 [55:07:16<15:28:30, 5.70s/it]
78%|███████▊ | 34168/43939 [55:07:22<15:27:53, 5.70s/it]
78%|███████▊ | 34169/43939 [55:07:28<15:28:27, 5.70s/it]
78%|███████▊ | 34170/43939 [55:07:33<15:26:34, 5.69s/it]
78%|███████▊ | 34171/43939 [55:07:39<15:27:53, 5.70s/it]
78%|███████▊ | 34172/43939 [55:07:45<15:27:51, 5.70s/it]
78%|███████▊ | 34173/43939 [55:07:50<15:27:54, 5.70s/it]
78%|███████▊ | 34174/43939 [55:07:56<15:29:21, 5.71s/it]
78%|███████▊ | 34175/43939 [55:08:02<15:27:28, 5.70s/it]
78%|███████▊ | 34176/43939 [55:08:07<15:26:09, 5.69s/it]
78%|███████▊ | 34177/43939 [55:08:13<15:26:36, 5.70s/it]
78%|███████▊ | 34178/43939 [55:08:19<15:28:23, 5.71s/it]
78%|███████▊ | 34179/43939 [55:08:25<15:38:56, 5.77s/it]
78%|███████▊ | 34180/43939 [55:08:31<15:36:51, 5.76s/it]
78%|███████▊ | 34181/43939 [55:08:36<15:33:47, 5.74s/it]
78%|███████▊ | 34182/43939 [55:08:42<15:31:56, 5.73s/it]
78%|███████▊ | 34183/43939 [55:08:48<15:28:39, 5.71s/it]
78%|███████▊ | 34184/43939 [55:08:53<15:28:43, 5.71s/it]
78%|███████▊ | 34185/43939 [55:08:59<15:25:29, 5.69s/it]
78%|███████▊ | 34186/43939 [55:09:05<15:24:26, 5.69s/it]
78%|███████▊ | 34187/43939 [55:09:10<15:25:53, 5.70s/it]
78%|███████▊ | 34188/43939 [55:09:16<15:26:13, 5.70s/it]
78%|███████▊ | 34189/43939 [55:09:22<15:26:17, 5.70s/it]
78%|███████▊ | 34190/43939 [55:09:27<15:25:26, 5.70s/it]
78%|███████▊ | 34191/43939 [55:09:33<15:26:42, 5.70s/it]
78%|███████▊ | 34192/43939 [55:09:39<15:26:30, 5.70s/it]
78%|███████▊ | 34193/43939 [55:09:45<15:29:47, 5.72s/it]
78%|███████▊ | 34194/43939 [55:09:50<15:30:32, 5.73s/it]
78%|███████▊ | 34195/43939 [55:09:56<15:27:44, 5.71s/it]
78%|███████▊ | 34196/43939 [55:10:02<15:31:22, 5.74s/it]
78%|███████▊ | 34197/43939 [55:10:08<15:31:17, 5.74s/it]
78%|███████▊ | 34198/43939 [55:10:13<15:29:16, 5.72s/it]
78%|███████▊ | 34199/43939 [55:10:19<15:28:42, 5.72s/it]
78%|███████▊ | 34200/43939 [55:10:25<15:41:49, 5.80s/it]
{'loss': 0.6617, 'grad_norm': 1.347619891166687, 'learning_rate': 7.117796459903228e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.39, 'epoch': 3.11}
78%|███████▊ | 34200/43939 [55:10:25<15:41:49, 5.80s/it]
78%|███████▊ | 34201/43939 [55:10:31<15:38:50, 5.78s/it]
78%|███████▊ | 34202/43939 [55:10:36<15:34:25, 5.76s/it]
78%|███████▊ | 34203/43939 [55:10:42<15:31:45, 5.74s/it]
78%|███████▊ | 34204/43939 [55:10:48<15:29:50, 5.73s/it]
78%|███████▊ | 34205/43939 [55:10:54<15:28:15, 5.72s/it]
78%|███████▊ | 34206/43939 [55:10:59<15:28:01, 5.72s/it]
78%|███████▊ | 34207/43939 [55:11:05<15:27:06, 5.72s/it]
78%|███████▊ | 34208/43939 [55:11:11<15:25:08, 5.70s/it]
78%|███████▊ | 34209/43939 [55:11:16<15:23:49, 5.70s/it]
78%|███████▊ | 34210/43939 [55:11:22<15:22:45, 5.69s/it]
78%|███████▊ | 34211/43939 [55:11:28<15:23:30, 5.70s/it]
78%|███████▊ | 34212/43939 [55:11:33<15:25:21, 5.71s/it]
78%|███████▊ | 34213/43939 [55:11:39<15:25:26, 5.71s/it]
78%|███████▊ | 34214/43939 [55:11:45<15:23:23, 5.70s/it]
78%|███████▊ | 34215/43939 [55:11:51<15:22:31, 5.69s/it]
78%|███████▊ | 34216/43939 [55:11:56<15:26:53, 5.72s/it]
78%|███████▊ | 34217/43939 [55:12:02<15:26:03, 5.72s/it]
78%|███████▊ | 34218/43939 [55:12:08<15:25:53, 5.71s/it]
78%|███████▊ | 34219/43939 [55:12:13<15:24:56, 5.71s/it]
78%|███████▊ | 34220/43939 [55:12:19<15:23:10, 5.70s/it]
78%|███████▊ | 34221/43939 [55:12:25<15:23:10, 5.70s/it]
78%|███████▊ | 34222/43939 [55:12:31<15:35:28, 5.78s/it]
78%|███████▊ | 34223/43939 [55:12:36<15:30:48, 5.75s/it]
78%|███████▊ | 34224/43939 [55:12:42<15:29:15, 5.74s/it]
78%|███████▊ | 34225/43939 [55:12:48<15:27:42, 5.73s/it]
78%|███████▊ | 34226/43939 [55:12:54<15:26:45, 5.72s/it]
78%|███████▊ | 34227/43939 [55:12:59<15:24:45, 5.71s/it]
78%|███████▊ | 34228/43939 [55:13:05<15:24:06, 5.71s/it]
78%|███████▊ | 34229/43939 [55:13:11<15:25:08, 5.72s/it]
78%|███████▊ | 34230/43939 [55:13:16<15:26:54, 5.73s/it]
78%|███████▊ | 34231/43939 [55:13:22<15:26:07, 5.72s/it]
78%|███████▊ | 34232/43939 [55:13:28<15:25:51, 5.72s/it]
78%|███████▊ | 34233/43939 [55:13:34<15:25:38, 5.72s/it]
78%|███████▊ | 34234/43939 [55:13:39<15:24:01, 5.71s/it]
78%|███████▊ | 34235/43939 [55:13:45<15:23:21, 5.71s/it]
78%|███████▊ | 34236/43939 [55:13:51<15:22:47, 5.71s/it]
78%|███████▊ | 34237/43939 [55:13:56<15:22:44, 5.71s/it]
78%|███████▊ | 34238/43939 [55:14:02<15:23:05, 5.71s/it]
78%|███████▊ | 34239/43939 [55:14:08<15:21:04, 5.70s/it]
78%|███████▊ | 34240/43939 [55:14:13<15:21:12, 5.70s/it]
78%|███████▊ | 34241/43939 [55:14:19<15:21:13, 5.70s/it]
78%|███████▊ | 34242/43939 [55:14:25<15:20:12, 5.69s/it]
78%|███████▊ | 34243/43939 [55:14:31<15:32:19, 5.77s/it]
78%|███████▊ | 34244/43939 [55:14:37<15:29:42, 5.75s/it]
78%|███████▊ | 34245/43939 [55:14:42<15:26:46, 5.74s/it]
78%|███████▊ | 34246/43939 [55:14:48<15:27:54, 5.74s/it]
78%|███████▊ | 34247/43939 [55:14:54<15:25:24, 5.73s/it]
78%|███████▊ | 34248/43939 [55:14:59<15:23:13, 5.72s/it]
78%|███████▊ | 34249/43939 [55:15:05<15:23:26, 5.72s/it]
78%|███████▊ | 34250/43939 [55:15:11<15:23:16, 5.72s/it]
{'loss': 0.6414, 'grad_norm': 1.770125150680542, 'learning_rate': 7.048542635094752e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.02, 'epoch': 3.12}
78%|███████▊ | 34250/43939 [55:15:11<15:23:16, 5.72s/it]
78%|███████▊ | 34251/43939 [55:15:17<15:26:49, 5.74s/it]
78%|███████▊ | 34252/43939 [55:15:22<15:26:01, 5.74s/it]
78%|███████▊ | 34253/43939 [55:15:28<15:23:20, 5.72s/it]
78%|███████▊ | 34254/43939 [55:15:34<15:22:40, 5.72s/it]
78%|███████▊ | 34255/43939 [55:15:39<15:20:44, 5.70s/it]
78%|███████▊ | 34256/43939 [55:15:45<15:20:29, 5.70s/it]
78%|███████▊ | 34257/43939 [55:15:51<15:18:28, 5.69s/it]
78%|███████▊ | 34258/43939 [55:15:56<15:19:34, 5.70s/it]
78%|███████▊ | 34259/43939 [55:16:02<15:19:10, 5.70s/it]
78%|███████▊ | 34260/43939 [55:16:08<15:18:50, 5.70s/it]
78%|███████▊ | 34261/43939 [55:16:14<15:20:20, 5.71s/it]
78%|███████▊ | 34262/43939 [55:16:19<15:19:09, 5.70s/it]
78%|███████▊ | 34263/43939 [55:16:25<15:19:28, 5.70s/it]
78%|███████▊ | 34264/43939 [55:16:31<15:31:25, 5.78s/it]
78%|███████▊ | 34265/43939 [55:16:37<15:27:18, 5.75s/it]
78%|███████▊ | 34266/43939 [55:16:42<15:28:01, 5.76s/it]
78%|███████▊ | 34267/43939 [55:16:48<15:26:39, 5.75s/it]
78%|███████▊ | 34268/43939 [55:16:54<15:23:23, 5.73s/it]
78%|███████▊ | 34269/43939 [55:16:59<15:21:42, 5.72s/it]
78%|███████▊ | 34270/43939 [55:17:05<15:21:00, 5.72s/it]
78%|███████▊ | 34271/43939 [55:17:11<15:18:19, 5.70s/it]
78%|███████▊ | 34272/43939 [55:17:17<15:17:56, 5.70s/it]
78%|███████▊ | 34273/43939 [55:17:22<15:18:19, 5.70s/it]
78%|███████▊ | 34274/43939 [55:17:28<15:16:59, 5.69s/it]
78%|███████▊ | 34275/43939 [55:17:34<15:16:07, 5.69s/it]
78%|███████▊ | 34276/43939 [55:17:39<15:17:09, 5.69s/it]
78%|███████▊ | 34277/43939 [55:17:45<15:19:51, 5.71s/it]
78%|███████▊ | 34278/43939 [55:17:51<15:18:31, 5.70s/it]
78%|███████▊ | 34279/43939 [55:17:56<15:19:32, 5.71s/it]
78%|███████▊ | 34280/43939 [55:18:02<15:18:25, 5.71s/it]
78%|███████▊ | 34281/43939 [55:18:08<15:18:41, 5.71s/it]
78%|███████▊ | 34282/43939 [55:18:14<15:17:52, 5.70s/it]
78%|███████▊ | 34283/43939 [55:18:19<15:17:00, 5.70s/it]
78%|███████▊ | 34284/43939 [55:18:25<15:17:22, 5.70s/it]
78%|███████▊ | 34285/43939 [55:18:31<15:26:40, 5.76s/it]
78%|███████▊ | 34286/43939 [55:18:37<15:23:22, 5.74s/it]
78%|███████▊ | 34287/43939 [55:18:42<15:20:50, 5.72s/it]
78%|███████▊ | 34288/43939 [55:18:48<15:20:00, 5.72s/it]
78%|███████▊ | 34289/43939 [55:18:54<15:18:50, 5.71s/it]
78%|███████▊ | 34290/43939 [55:18:59<15:19:20, 5.72s/it]
78%|███████▊ | 34291/43939 [55:19:05<15:19:44, 5.72s/it]
78%|███████▊ | 34292/43939 [55:19:11<15:17:55, 5.71s/it]
78%|███████▊ | 34293/43939 [55:19:17<15:17:26, 5.71s/it]
78%|███████▊ | 34294/43939 [55:19:22<15:18:05, 5.71s/it]
78%|███████▊ | 34295/43939 [55:19:28<15:20:07, 5.72s/it]
78%|███████▊ | 34296/43939 [55:19:34<15:19:13, 5.72s/it]
78%|███████▊ | 34297/43939 [55:19:39<15:17:13, 5.71s/it]
78%|███████▊ | 34298/43939 [55:19:45<15:15:34, 5.70s/it]
78%|███████▊ | 34299/43939 [55:19:51<15:15:51, 5.70s/it]
78%|███████▊ | 34300/43939 [55:19:56<15:15:36, 5.70s/it]
{'loss': 0.6603, 'grad_norm': 1.34881591796875, 'learning_rate': 6.979572036711548e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.63, 'epoch': 3.12}
78%|███████▊ | 34300/43939 [55:19:56<15:15:36, 5.70s/it]
78%|███████▊ | 34301/43939 [55:20:02<15:16:48, 5.71s/it]
78%|███████▊ | 34302/43939 [55:20:08<15:16:45, 5.71s/it]
78%|███████▊ | 34303/43939 [55:20:14<15:15:32, 5.70s/it]
78%|███████▊ | 34304/43939 [55:20:19<15:14:18, 5.69s/it]
78%|███████▊ | 34305/43939 [55:20:25<15:13:27, 5.69s/it]
78%|███████▊ | 34306/43939 [55:20:31<15:15:01, 5.70s/it]
78%|███████▊ | 34307/43939 [55:20:37<15:24:16, 5.76s/it]
78%|███████▊ | 34308/43939 [55:20:42<15:20:50, 5.74s/it]
78%|███████▊ | 34309/43939 [55:20:48<15:18:23, 5.72s/it]
78%|███████▊ | 34310/43939 [55:20:54<15:17:43, 5.72s/it]
78%|███████▊ | 34311/43939 [55:20:59<15:16:37, 5.71s/it]
78%|███████▊ | 34312/43939 [55:21:05<15:18:03, 5.72s/it]
78%|███████▊ | 34313/43939 [55:21:11<15:17:23, 5.72s/it]
78%|███████▊ | 34314/43939 [55:21:16<15:17:05, 5.72s/it]
78%|███████▊ | 34315/43939 [55:21:22<15:16:13, 5.71s/it]
78%|███████▊ | 34316/43939 [55:21:28<15:16:58, 5.72s/it]
78%|███████▊ | 34317/43939 [55:21:34<15:13:45, 5.70s/it]
78%|███████▊ | 34318/43939 [55:21:39<15:13:36, 5.70s/it]
78%|███████▊ | 34319/43939 [55:21:45<15:13:20, 5.70s/it]
78%|███████▊ | 34320/43939 [55:21:51<15:13:17, 5.70s/it]
78%|███████▊ | 34321/43939 [55:21:56<15:15:13, 5.71s/it]
78%|███████▊ | 34322/43939 [55:22:02<15:16:00, 5.71s/it]
78%|███████▊ | 34323/43939 [55:22:08<15:15:34, 5.71s/it]
78%|███████▊ | 34324/43939 [55:22:14<15:16:46, 5.72s/it]
78%|███████▊ | 34325/43939 [55:22:19<15:17:23, 5.73s/it]
78%|███████▊ | 34326/43939 [55:22:25<15:14:51, 5.71s/it]
78%|███████▊ | 34327/43939 [55:22:31<15:14:09, 5.71s/it]
78%|███████▊ | 34328/43939 [55:22:36<15:13:49, 5.70s/it]
78%|███████▊ | 34329/43939 [55:22:42<15:21:45, 5.75s/it]
78%|███████▊ | 34330/43939 [55:22:48<15:17:56, 5.73s/it]
78%|███████▊ | 34331/43939 [55:22:54<15:15:01, 5.71s/it]
78%|███████▊ | 34332/43939 [55:22:59<15:14:06, 5.71s/it]
78%|███████▊ | 34333/43939 [55:23:05<15:15:34, 5.72s/it]
78%|███████▊ | 34334/43939 [55:23:11<15:16:27, 5.72s/it]
78%|███████▊ | 34335/43939 [55:23:16<15:15:05, 5.72s/it]
78%|███████▊ | 34336/43939 [55:23:22<15:13:49, 5.71s/it]
78%|███████▊ | 34337/43939 [55:23:28<15:13:17, 5.71s/it]
78%|███████▊ | 34338/43939 [55:23:34<15:13:25, 5.71s/it]
78%|███████▊ | 34339/43939 [55:23:39<15:11:01, 5.69s/it]
78%|███████▊ | 34340/43939 [55:23:45<15:10:31, 5.69s/it]
78%|███████▊ | 34341/43939 [55:23:51<15:11:48, 5.70s/it]
78%|███████▊ | 34342/43939 [55:23:56<15:10:31, 5.69s/it]
78%|███████▊ | 34343/43939 [55:24:02<15:10:22, 5.69s/it]
78%|███████▊ | 34344/43939 [55:24:08<15:12:22, 5.71s/it]
78%|███████▊ | 34345/43939 [55:24:14<15:14:06, 5.72s/it]
78%|███████▊ | 34346/43939 [55:24:19<15:13:25, 5.71s/it]
78%|███████▊ | 34347/43939 [55:24:25<15:12:37, 5.71s/it]
78%|███████▊ | 34348/43939 [55:24:31<15:13:48, 5.72s/it]
78%|███████▊ | 34349/43939 [55:24:36<15:11:29, 5.70s/it]
78%|███████▊ | 34350/43939 [55:24:42<15:23:07, 5.78s/it]
{'loss': 0.6562, 'grad_norm': 1.9866487979888916, 'learning_rate': 6.9108857529268965e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.14, 'epoch': 3.13}
78%|███████▊ | 34350/43939 [55:24:42<15:23:07, 5.78s/it]
78%|███████▊ | 34351/43939 [55:24:48<15:19:22, 5.75s/it]
78%|███████▊ | 34352/43939 [55:24:54<15:17:09, 5.74s/it]
78%|███████▊ | 34353/43939 [55:24:59<15:13:57, 5.72s/it]
78%|███████▊ | 34354/43939 [55:25:05<15:10:05, 5.70s/it]
78%|███████▊ | 34355/43939 [55:25:11<15:09:20, 5.69s/it]
78%|███████▊ | 34356/43939 [55:25:16<15:08:50, 5.69s/it]
78%|███████▊ | 34357/43939 [55:25:22<15:08:54, 5.69s/it]
78%|███████▊ | 34358/43939 [55:25:28<15:10:42, 5.70s/it]
78%|███████▊ | 34359/43939 [55:25:34<15:13:38, 5.72s/it]
78%|███████▊ | 34360/43939 [55:25:39<15:12:07, 5.71s/it]
78%|███████▊ | 34361/43939 [55:25:45<15:10:37, 5.70s/it]
78%|███████▊ | 34362/43939 [55:25:51<15:12:10, 5.71s/it]
78%|███████▊ | 34363/43939 [55:25:56<15:11:48, 5.71s/it]
78%|███████▊ | 34364/43939 [55:26:02<15:10:35, 5.71s/it]
78%|███████▊ | 34365/43939 [55:26:08<15:10:20, 5.71s/it]
78%|███████▊ | 34366/43939 [55:26:13<15:08:58, 5.70s/it]
78%|███████▊ | 34367/43939 [55:26:19<15:07:37, 5.69s/it]
78%|███████▊ | 34368/43939 [55:26:25<15:07:46, 5.69s/it]
78%|███████▊ | 34369/43939 [55:26:30<15:07:43, 5.69s/it]
78%|███████▊ | 34370/43939 [55:26:36<15:09:14, 5.70s/it]
78%|███████▊ | 34371/43939 [55:26:42<15:12:12, 5.72s/it]
78%|███████▊ | 34372/43939 [55:26:48<15:22:56, 5.79s/it]
78%|███████▊ | 34373/43939 [55:26:54<15:18:00, 5.76s/it]
78%|███████▊ | 34374/43939 [55:26:59<15:15:07, 5.74s/it]
78%|███████▊ | 34375/43939 [55:27:05<15:12:49, 5.73s/it]
78%|███████▊ | 34376/43939 [55:27:11<15:11:06, 5.72s/it]
78%|███████▊ | 34377/43939 [55:27:16<15:09:50, 5.71s/it]
78%|███████▊ | 34378/43939 [55:27:22<15:10:22, 5.71s/it]
78%|███████▊ | 34379/43939 [55:27:28<15:10:07, 5.71s/it]
78%|███████▊ | 34380/43939 [55:27:34<15:11:38, 5.72s/it]
78%|███████▊ | 34381/43939 [55:27:39<15:10:49, 5.72s/it]
78%|███████▊ | 34382/43939 [55:27:45<15:11:04, 5.72s/it]
78%|███████▊ | 34383/43939 [55:27:51<15:11:03, 5.72s/it]
78%|███████▊ | 34384/43939 [55:27:56<15:11:08, 5.72s/it]
78%|███████▊ | 34385/43939 [55:28:02<15:11:22, 5.72s/it]
78%|███████▊ | 34386/43939 [55:28:08<15:09:45, 5.71s/it]
78%|███████▊ | 34387/43939 [55:28:14<15:09:52, 5.72s/it]
78%|███████▊ | 34388/43939 [55:28:19<15:09:28, 5.71s/it]
78%|███████▊ | 34389/43939 [55:28:25<15:08:37, 5.71s/it]
78%|███████▊ | 34390/43939 [55:28:31<15:06:45, 5.70s/it]
78%|███████▊ | 34391/43939 [55:28:36<15:08:18, 5.71s/it]
78%|███████▊ | 34392/43939 [55:28:42<15:07:45, 5.71s/it]
78%|███████▊ | 34393/43939 [55:28:48<15:20:07, 5.78s/it]
78%|███████▊ | 34394/43939 [55:28:54<15:15:36, 5.76s/it]
78%|███████▊ | 34395/43939 [55:28:59<15:14:02, 5.75s/it]
78%|███████▊ | 34396/43939 [55:29:05<15:13:10, 5.74s/it]
78%|███████▊ | 34397/43939 [55:29:11<15:10:47, 5.73s/it]
78%|███████▊ | 34398/43939 [55:29:17<15:08:05, 5.71s/it]
78%|███████▊ | 34399/43939 [55:29:22<15:06:26, 5.70s/it]
78%|███████▊ | 34400/43939 [55:29:28<15:08:10, 5.71s/it]
{'loss': 0.6569, 'grad_norm': 1.886730670928955, 'learning_rate': 6.842484867428323e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.25, 'epoch': 3.13}
78%|███████▊ | 34400/43939 [55:29:28<15:08:10, 5.71s/it]
78%|███████▊ | 34401/43939 [55:29:34<15:07:01, 5.71s/it]
78%|███████▊ | 34402/43939 [55:29:39<15:06:42, 5.70s/it]
78%|███████▊ | 34403/43939 [55:29:45<15:08:31, 5.72s/it]
78%|███████▊ | 34404/43939 [55:29:51<15:06:45, 5.71s/it]
78%|███████▊ | 34405/43939 [55:29:57<15:07:55, 5.71s/it]
78%|███████▊ | 34406/43939 [55:30:02<15:07:54, 5.71s/it]
78%|███████▊ | 34407/43939 [55:30:08<15:07:24, 5.71s/it]
78%|███████▊ | 34408/43939 [55:30:14<15:06:48, 5.71s/it]
78%|███████▊ | 34409/43939 [55:30:19<15:05:45, 5.70s/it]
78%|███████▊ | 34410/43939 [55:30:25<15:07:25, 5.71s/it]
78%|███████▊ | 34411/43939 [55:30:31<15:06:35, 5.71s/it]
78%|███████▊ | 34412/43939 [55:30:36<15:06:21, 5.71s/it]
78%|███████▊ | 34413/43939 [55:30:42<15:06:30, 5.71s/it]
78%|███████▊ | 34414/43939 [55:30:48<15:18:33, 5.79s/it]
78%|███████▊ | 34415/43939 [55:30:54<15:15:04, 5.76s/it]
78%|███████▊ | 34416/43939 [55:31:00<15:11:42, 5.74s/it]
78%|███████▊ | 34417/43939 [55:31:05<15:08:25, 5.72s/it]
78%|███████▊ | 34418/43939 [55:31:11<15:07:18, 5.72s/it]
78%|███████▊ | 34419/43939 [55:31:17<15:05:26, 5.71s/it]
78%|███████▊ | 34420/43939 [55:31:22<15:06:32, 5.71s/it]
78%|███████▊ | 34421/43939 [55:31:28<15:05:48, 5.71s/it]
78%|███████▊ | 34422/43939 [55:31:34<15:04:46, 5.70s/it]
78%|███████▊ | 34423/43939 [55:31:39<15:05:34, 5.71s/it]
78%|███████▊ | 34424/43939 [55:31:45<15:07:01, 5.72s/it]
78%|███████▊ | 34425/43939 [55:31:51<15:06:12, 5.71s/it]
78%|███████▊ | 34426/43939 [55:31:57<15:05:42, 5.71s/it]
78%|███████▊ | 34427/43939 [55:32:02<15:06:09, 5.72s/it]
78%|███████▊ | 34428/43939 [55:32:08<15:05:29, 5.71s/it]
78%|███████▊ | 34429/43939 [55:32:14<15:03:32, 5.70s/it]
78%|███████▊ | 34430/43939 [55:32:19<15:03:26, 5.70s/it]
78%|███████▊ | 34431/43939 [55:32:25<15:05:07, 5.71s/it]
78%|███████▊ | 34432/43939 [55:32:31<15:04:18, 5.71s/it]
78%|███████▊ | 34433/43939 [55:32:37<15:03:21, 5.70s/it]
78%|███████▊ | 34434/43939 [55:32:42<15:03:54, 5.71s/it]
78%|███████▊ | 34435/43939 [55:32:48<15:15:56, 5.78s/it]
78%|███████▊ | 34436/43939 [55:32:54<15:13:46, 5.77s/it]
78%|███████▊ | 34437/43939 [55:33:00<15:09:57, 5.75s/it]
78%|███████▊ | 34438/43939 [55:33:05<15:09:10, 5.74s/it]
78%|███████▊ | 34439/43939 [55:33:11<15:04:53, 5.72s/it]
78%|███████▊ | 34440/43939 [55:33:17<15:03:59, 5.71s/it]
78%|███████▊ | 34441/43939 [55:33:22<15:02:26, 5.70s/it]
78%|███████▊ | 34442/43939 [55:33:28<15:00:50, 5.69s/it]
78%|███████▊ | 34443/43939 [55:33:34<15:00:49, 5.69s/it]
78%|███████▊ | 34444/43939 [55:33:39<15:00:54, 5.69s/it]
78%|███████▊ | 34445/43939 [55:33:45<15:00:12, 5.69s/it]
78%|███████▊ | 34446/43939 [55:33:51<15:02:43, 5.71s/it]
78%|███████▊ | 34447/43939 [55:33:57<15:02:38, 5.71s/it]
78%|███████▊ | 34448/43939 [55:34:02<15:01:09, 5.70s/it]
78%|███████▊ | 34449/43939 [55:34:08<15:02:37, 5.71s/it]
78%|███████▊ | 34450/43939 [55:34:14<15:02:29, 5.71s/it]
{'loss': 0.6639, 'grad_norm': 1.8228403329849243, 'learning_rate': 6.774370459400542e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.22, 'epoch': 3.14}
78%|███████▊ | 34450/43939 [55:34:14<15:02:29, 5.71s/it]
78%|███████▊ | 34451/43939 [55:34:19<15:03:48, 5.72s/it]
78%|███████▊ | 34452/43939 [55:34:25<15:02:54, 5.71s/it]
78%|███████▊ | 34453/43939 [55:34:31<15:02:43, 5.71s/it]
78%|███████▊ | 34454/43939 [55:34:37<15:01:09, 5.70s/it]
78%|███████▊ | 34455/43939 [55:34:42<15:01:58, 5.71s/it]
78%|███████▊ | 34456/43939 [55:34:48<15:02:05, 5.71s/it]
78%|███████▊ | 34457/43939 [55:34:54<15:11:43, 5.77s/it]
78%|███████▊ | 34458/43939 [55:35:00<15:09:04, 5.75s/it]
78%|███████▊ | 34459/43939 [55:35:05<15:09:01, 5.75s/it]
78%|███████▊ | 34460/43939 [55:35:11<15:08:38, 5.75s/it]
78%|███████▊ | 34461/43939 [55:35:17<15:07:46, 5.75s/it]
78%|███████▊ | 34462/43939 [55:35:23<15:03:44, 5.72s/it]
78%|███████▊ | 34463/43939 [55:35:28<15:02:28, 5.71s/it]
78%|███████▊ | 34464/43939 [55:35:34<15:01:26, 5.71s/it]
78%|███████▊ | 34465/43939 [55:35:40<15:00:40, 5.70s/it]
78%|███████▊ | 34466/43939 [55:35:45<15:00:20, 5.70s/it]
78%|███████▊ | 34467/43939 [55:35:51<15:00:42, 5.71s/it]
78%|███████▊ | 34468/43939 [55:35:57<14:59:48, 5.70s/it]
78%|███████▊ | 34469/43939 [55:36:02<14:59:47, 5.70s/it]
78%|███████▊ | 34470/43939 [55:36:08<14:59:58, 5.70s/it]
78%|███████▊ | 34471/43939 [55:36:14<14:58:01, 5.69s/it]
78%|███████▊ | 34472/43939 [55:36:19<14:58:25, 5.69s/it]
78%|███████▊ | 34473/43939 [55:36:25<14:58:52, 5.70s/it]
78%|███████▊ | 34474/43939 [55:36:31<14:57:42, 5.69s/it]
78%|███████▊ | 34475/43939 [55:36:37<14:56:59, 5.69s/it]
78%|███████▊ | 34476/43939 [55:36:42<14:57:16, 5.69s/it]
78%|███████▊ | 34477/43939 [55:36:48<14:57:58, 5.69s/it]
78%|███████▊ | 34478/43939 [55:36:54<15:13:26, 5.79s/it]
78%|███████▊ | 34479/43939 [55:37:00<15:07:12, 5.75s/it]
78%|███████▊ | 34480/43939 [55:37:05<15:04:48, 5.74s/it]
78%|███████▊ | 34481/43939 [55:37:11<15:02:07, 5.72s/it]
78%|███████▊ | 34482/43939 [55:37:17<15:00:14, 5.71s/it]
78%|███████▊ | 34483/43939 [55:37:22<15:00:59, 5.72s/it]
78%|███████▊ | 34484/43939 [55:37:28<14:58:24, 5.70s/it]
78%|███████▊ | 34485/43939 [55:37:34<14:59:55, 5.71s/it]
78%|███████▊ | 34486/43939 [55:37:40<15:01:16, 5.72s/it]
78%|███████▊ | 34487/43939 [55:37:45<15:01:26, 5.72s/it]
78%|███████▊ | 34488/43939 [55:37:51<15:00:44, 5.72s/it]
78%|███████▊ | 34489/43939 [55:37:57<15:00:18, 5.72s/it]
78%|███████▊ | 34490/43939 [55:38:02<15:00:57, 5.72s/it]
78%|███████▊ | 34491/43939 [55:38:08<15:02:47, 5.73s/it]
78%|███████▊ | 34492/43939 [55:38:14<15:02:35, 5.73s/it]
79%|███████▊ | 34493/43939 [55:38:20<15:01:45, 5.73s/it]
79%|███████▊ | 34494/43939 [55:38:25<14:59:46, 5.72s/it]
79%|███████▊ | 34495/43939 [55:38:31<14:59:31, 5.71s/it]
79%|███████▊ | 34496/43939 [55:38:37<14:59:47, 5.72s/it]
79%|███████▊ | 34497/43939 [55:38:42<14:58:12, 5.71s/it]
79%|███████▊ | 34498/43939 [55:38:48<14:57:20, 5.70s/it]
79%|███████▊ | 34499/43939 [55:38:54<15:06:50, 5.76s/it]
79%|███████▊ | 34500/43939 [55:39:00<15:02:55, 5.74s/it]
{'loss': 0.6439, 'grad_norm': 1.59607994556427, 'learning_rate': 6.706543603508403e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.81, 'epoch': 3.14}
79%|███████▊ | 34500/43939 [55:39:00<15:02:55, 5.74s/it]
79%|███████▊ | 34501/43939 [55:39:05<15:02:36, 5.74s/it]
79%|███████▊ | 34502/43939 [55:39:11<15:00:57, 5.73s/it]
79%|███████▊ | 34503/43939 [55:39:17<14:58:30, 5.71s/it]
79%|███████▊ | 34504/43939 [55:39:23<14:58:44, 5.72s/it]
79%|███████▊ | 34505/43939 [55:39:28<14:57:35, 5.71s/it]
79%|███████▊ | 34506/43939 [55:39:34<14:57:25, 5.71s/it]
79%|███████▊ | 34507/43939 [55:39:40<14:57:29, 5.71s/it]
79%|███████▊ | 34508/43939 [55:39:45<14:58:01, 5.71s/it]
79%|███████▊ | 34509/43939 [55:39:51<14:58:37, 5.72s/it]
79%|███████▊ | 34510/43939 [55:39:57<14:58:30, 5.72s/it]
79%|███████▊ | 34511/43939 [55:40:03<14:58:52, 5.72s/it]
79%|███████▊ | 34512/43939 [55:40:08<14:58:16, 5.72s/it]
79%|███████▊ | 34513/43939 [55:40:14<14:59:23, 5.72s/it]
79%|███████▊ | 34514/43939 [55:40:20<14:59:21, 5.73s/it]
79%|███████▊ | 34515/43939 [55:40:25<14:59:03, 5.72s/it]
79%|███████▊ | 34516/43939 [55:40:31<14:56:20, 5.71s/it]
79%|███████▊ | 34517/43939 [55:40:37<14:56:06, 5.71s/it]
79%|███████▊ | 34518/43939 [55:40:43<14:55:06, 5.70s/it]
79%|███████▊ | 34519/43939 [55:40:48<14:54:08, 5.70s/it]
79%|███████▊ | 34520/43939 [55:40:54<15:05:21, 5.77s/it]
79%|███████▊ | 34521/43939 [55:41:00<15:01:25, 5.74s/it]
79%|███████▊ | 34522/43939 [55:41:06<15:00:11, 5.74s/it]
79%|███████▊ | 34523/43939 [55:41:11<14:59:18, 5.73s/it]
79%|███████▊ | 34524/43939 [55:41:17<14:59:48, 5.73s/it]
79%|███████▊ | 34525/43939 [55:41:23<14:57:06, 5.72s/it]
79%|███████▊ | 34526/43939 [55:41:28<14:58:57, 5.73s/it]
79%|███████▊ | 34527/43939 [55:41:34<14:56:48, 5.72s/it]
79%|███████▊ | 34528/43939 [55:41:40<14:55:00, 5.71s/it]
79%|███████▊ | 34529/43939 [55:41:46<14:54:48, 5.71s/it]
79%|███████▊ | 34530/43939 [55:41:51<14:55:22, 5.71s/it]
79%|███████▊ | 34531/43939 [55:41:57<14:55:40, 5.71s/it]
79%|███████▊ | 34532/43939 [55:42:03<14:53:52, 5.70s/it]
79%|███████▊ | 34533/43939 [55:42:08<14:53:06, 5.70s/it]
79%|███████▊ | 34534/43939 [55:42:14<14:53:30, 5.70s/it]
79%|███████▊ | 34535/43939 [55:42:20<14:53:19, 5.70s/it]
79%|███████▊ | 34536/43939 [55:42:25<14:53:56, 5.70s/it]
79%|███████▊ | 34537/43939 [55:42:31<14:54:15, 5.71s/it]
79%|███████▊ | 34538/43939 [55:42:37<14:55:31, 5.72s/it]
79%|███████▊ | 34539/43939 [55:42:43<14:56:37, 5.72s/it]
79%|███████▊ | 34540/43939 [55:42:48<14:57:46, 5.73s/it]
79%|███████▊ | 34541/43939 [55:42:54<14:56:57, 5.73s/it]
79%|███████▊ | 34542/43939 [55:43:00<15:07:46, 5.80s/it]
79%|███████▊ | 34543/43939 [55:43:06<15:05:03, 5.78s/it]
79%|███████▊ | 34544/43939 [55:43:12<15:02:43, 5.77s/it]
79%|███████▊ | 34545/43939 [55:43:17<15:00:56, 5.75s/it]
79%|███████▊ | 34546/43939 [55:43:23<15:01:09, 5.76s/it]
79%|███████▊ | 34547/43939 [55:43:29<14:59:20, 5.75s/it]
79%|███████▊ | 34548/43939 [55:43:34<14:56:57, 5.73s/it]
79%|███████▊ | 34549/43939 [55:43:40<14:58:23, 5.74s/it]
79%|███████▊ | 34550/43939 [55:43:46<14:54:03, 5.71s/it]
{'loss': 0.6514, 'grad_norm': 1.8505710363388062, 'learning_rate': 6.639005369879955e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.72, 'epoch': 3.15}
79%|███████▊ | 34550/43939 [55:43:46<14:54:03, 5.71s/it]
79%|███████▊ | 34551/43939 [55:43:52<14:55:07, 5.72s/it]
79%|███████▊ | 34552/43939 [55:43:57<14:53:51, 5.71s/it]
79%|███████▊ | 34553/43939 [55:44:03<14:52:25, 5.70s/it]
79%|███████▊ | 34554/43939 [55:44:09<14:51:08, 5.70s/it]
79%|███████▊ | 34555/43939 [55:44:14<14:52:55, 5.71s/it]
79%|███████▊ | 34556/43939 [55:44:20<14:53:14, 5.71s/it]
79%|███████▊ | 34557/43939 [55:44:26<14:52:00, 5.70s/it]
79%|███████▊ | 34558/43939 [55:44:32<14:51:54, 5.70s/it]
79%|███████▊ | 34559/43939 [55:44:37<14:51:33, 5.70s/it]
79%|███████▊ | 34560/43939 [55:44:43<14:51:31, 5.70s/it]
79%|███████▊ | 34561/43939 [55:44:49<14:51:43, 5.71s/it]
79%|███████▊ | 34562/43939 [55:44:54<14:51:46, 5.71s/it]
79%|███████▊ | 34563/43939 [55:45:00<14:52:41, 5.71s/it]
79%|███████▊ | 34564/43939 [55:45:06<15:03:38, 5.78s/it]
79%|███████▊ | 34565/43939 [55:45:12<14:58:31, 5.75s/it]
79%|███████▊ | 34566/43939 [55:45:17<14:56:28, 5.74s/it]
79%|███████▊ | 34567/43939 [55:45:23<14:52:51, 5.72s/it]
79%|███████▊ | 34568/43939 [55:45:29<14:50:32, 5.70s/it]
79%|███████▊ | 34569/43939 [55:45:34<14:50:18, 5.70s/it]
79%|███████▊ | 34570/43939 [55:45:40<14:49:09, 5.69s/it]
79%|███████▊ | 34571/43939 [55:45:46<14:49:05, 5.69s/it]
79%|███████▊ | 34572/43939 [55:45:52<14:49:17, 5.70s/it]
79%|███████▊ | 34573/43939 [55:45:57<14:49:32, 5.70s/it]
79%|███████▊ | 34574/43939 [55:46:03<14:49:20, 5.70s/it]
79%|███████▊ | 34575/43939 [55:46:09<14:49:26, 5.70s/it]
79%|███████▊ | 34576/43939 [55:46:14<14:49:33, 5.70s/it]
79%|███████▊ | 34577/43939 [55:46:20<14:49:15, 5.70s/it]
79%|███████▊ | 34578/43939 [55:46:26<14:49:25, 5.70s/it]
79%|███████▊ | 34579/43939 [55:46:31<14:47:42, 5.69s/it]
79%|███████▊ | 34580/43939 [55:46:37<14:50:41, 5.71s/it]
79%|███████▊ | 34581/43939 [55:46:43<14:50:20, 5.71s/it]
79%|███████▊ | 34582/43939 [55:46:49<14:50:49, 5.71s/it]
79%|███████▊ | 34583/43939 [55:46:54<14:52:26, 5.72s/it]
79%|███████▊ | 34584/43939 [55:47:00<14:51:10, 5.72s/it]
79%|███████▊ | 34585/43939 [55:47:06<14:59:27, 5.77s/it]
79%|███████▊ | 34586/43939 [55:47:12<14:59:52, 5.77s/it]
79%|███████▊ | 34587/43939 [55:47:17<14:57:39, 5.76s/it]
79%|███████▊ | 34588/43939 [55:47:23<14:59:44, 5.77s/it]
79%|███████▊ | 34589/43939 [55:47:29<14:56:19, 5.75s/it]
79%|███████▊ | 34590/43939 [55:47:35<14:54:02, 5.74s/it]
79%|███████▊ | 34591/43939 [55:47:40<14:51:43, 5.72s/it]
79%|███████▊ | 34592/43939 [55:47:46<14:51:48, 5.72s/it]
79%|███████▊ | 34593/43939 [55:47:52<14:51:23, 5.72s/it]
79%|███████▊ | 34594/43939 [55:47:57<14:50:34, 5.72s/it]
79%|███████▊ | 34595/43939 [55:48:03<14:50:22, 5.72s/it]
79%|███████▊ | 34596/43939 [55:48:09<14:50:56, 5.72s/it]
79%|███████▊ | 34597/43939 [55:48:15<14:48:20, 5.71s/it]
79%|███████▊ | 34598/43939 [55:48:20<14:48:55, 5.71s/it]
79%|███████▊ | 34599/43939 [55:48:26<14:48:04, 5.70s/it]
79%|███████▊ | 34600/43939 [55:48:32<14:47:41, 5.70s/it]
{'loss': 0.6657, 'grad_norm': 1.8522517681121826, 'learning_rate': 6.571756824089539e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.77, 'epoch': 3.15}
79%|███████▊ | 34600/43939 [55:48:32<14:47:41, 5.70s/it]
79%|███████▊ | 34601/43939 [55:48:37<14:46:35, 5.70s/it]
79%|███████▉ | 34602/43939 [55:48:43<14:47:45, 5.70s/it]
79%|███████▉ | 34603/43939 [55:48:49<14:48:31, 5.71s/it]
79%|███████▉ | 34604/43939 [55:48:55<14:47:58, 5.71s/it]
79%|███████▉ | 34605/43939 [55:49:00<14:48:30, 5.71s/it]
79%|███████▉ | 34606/43939 [55:49:06<14:48:47, 5.71s/it]
79%|███████▉ | 34607/43939 [55:49:12<14:57:05, 5.77s/it]
79%|███████▉ | 34608/43939 [55:49:18<14:55:09, 5.76s/it]
79%|███████▉ | 34609/43939 [55:49:23<14:52:35, 5.74s/it]
79%|███████▉ | 34610/43939 [55:49:29<14:51:10, 5.73s/it]
79%|███████▉ | 34611/43939 [55:49:35<14:50:51, 5.73s/it]
79%|███████▉ | 34612/43939 [55:49:40<14:51:13, 5.73s/it]
79%|███████▉ | 34613/43939 [55:49:46<14:49:36, 5.72s/it]
79%|███████▉ | 34614/43939 [55:49:52<14:48:00, 5.71s/it]
79%|███████▉ | 34615/43939 [55:49:58<14:47:05, 5.71s/it]
79%|███████▉ | 34616/43939 [55:50:03<14:46:34, 5.71s/it]
79%|███████▉ | 34617/43939 [55:50:09<14:47:05, 5.71s/it]
79%|███████▉ | 34618/43939 [55:50:15<14:47:03, 5.71s/it]
79%|███████▉ | 34619/43939 [55:50:20<14:47:38, 5.71s/it]
79%|███████▉ | 34620/43939 [55:50:26<14:49:36, 5.73s/it]
79%|███████▉ | 34621/43939 [55:50:32<14:47:08, 5.71s/it]
79%|███████▉ | 34622/43939 [55:50:38<14:47:40, 5.72s/it]
79%|███████▉ | 34623/43939 [55:50:43<14:45:29, 5.70s/it]
79%|███████▉ | 34624/43939 [55:50:49<14:46:38, 5.71s/it]
79%|███████▉ | 34625/43939 [55:50:55<14:45:21, 5.70s/it]
79%|███████▉ | 34626/43939 [55:51:00<14:44:23, 5.70s/it]
79%|███████▉ | 34627/43939 [55:51:06<14:42:30, 5.69s/it]
79%|███████▉ | 34628/43939 [55:51:12<14:55:06, 5.77s/it]
79%|███████▉ | 34629/43939 [55:51:18<14:52:57, 5.75s/it]
79%|███████▉ | 34630/43939 [55:51:23<14:50:35, 5.74s/it]
79%|███████▉ | 34631/43939 [55:51:29<14:49:29, 5.73s/it]
79%|███████▉ | 34632/43939 [55:51:35<14:46:54, 5.72s/it]
79%|███████▉ | 34633/43939 [55:51:40<14:45:55, 5.71s/it]
79%|███████▉ | 34634/43939 [55:51:46<14:46:37, 5.72s/it]
79%|███████▉ | 34635/43939 [55:51:52<14:46:11, 5.71s/it]
79%|███████▉ | 34636/43939 [55:51:58<14:44:47, 5.71s/it]
79%|███████▉ | 34637/43939 [55:52:03<14:45:08, 5.71s/it]
79%|███████▉ | 34638/43939 [55:52:09<14:45:13, 5.71s/it]
79%|███████▉ | 34639/43939 [55:52:15<14:45:31, 5.71s/it]
79%|███████▉ | 34640/43939 [55:52:20<14:43:50, 5.70s/it]
79%|███████▉ | 34641/43939 [55:52:26<14:45:16, 5.71s/it]
79%|███████▉ | 34642/43939 [55:52:32<14:45:12, 5.71s/it]
79%|███████▉ | 34643/43939 [55:52:38<14:44:53, 5.71s/it]
79%|███████▉ | 34644/43939 [55:52:43<14:45:31, 5.72s/it]
79%|███████▉ | 34645/43939 [55:52:49<14:45:45, 5.72s/it]
79%|███████▉ | 34646/43939 [55:52:55<14:45:10, 5.72s/it]
79%|███████▉ | 34647/43939 [55:53:01<14:48:45, 5.74s/it]
79%|███████▉ | 34648/43939 [55:53:06<14:47:11, 5.73s/it]
79%|███████▉ | 34649/43939 [55:53:12<14:48:30, 5.74s/it]
79%|███████▉ | 34650/43939 [55:53:18<14:59:26, 5.81s/it]
{'loss': 0.6696, 'grad_norm': 1.5501015186309814, 'learning_rate': 6.504799027141001e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 524.56, 'epoch': 3.15}
79%|███████▉ | 34650/43939 [55:53:18<14:59:26, 5.81s/it]
79%|███████▉ | 34651/43939 [55:53:24<14:54:51, 5.78s/it]
79%|███████▉ | 34652/43939 [55:53:29<14:53:33, 5.77s/it]
79%|███████▉ | 34653/43939 [55:53:35<14:51:00, 5.76s/it]
79%|███████▉ | 34654/43939 [55:53:41<14:51:33, 5.76s/it]
79%|███████▉ | 34655/43939 [55:53:47<14:48:40, 5.74s/it]
79%|███████▉ | 34656/43939 [55:53:52<14:48:54, 5.75s/it]
79%|███████▉ | 34657/43939 [55:53:58<14:50:52, 5.76s/it]
79%|███████▉ | 34658/43939 [55:54:04<14:47:51, 5.74s/it]
79%|███████▉ | 34659/43939 [55:54:10<14:46:34, 5.73s/it]
79%|███████▉ | 34660/43939 [55:54:15<14:46:18, 5.73s/it]
79%|███████▉ | 34661/43939 [55:54:21<14:45:57, 5.73s/it]
79%|███████▉ | 34662/43939 [55:54:27<14:44:56, 5.72s/it]
79%|███████▉ | 34663/43939 [55:54:32<14:43:30, 5.71s/it]
79%|███████▉ | 34664/43939 [55:54:38<14:44:06, 5.72s/it]
79%|███████▉ | 34665/43939 [55:54:44<14:43:58, 5.72s/it]
79%|███████▉ | 34666/43939 [55:54:50<14:41:48, 5.71s/it]
79%|███████▉ | 34667/43939 [55:54:55<14:40:14, 5.70s/it]
79%|███████▉ | 34668/43939 [55:55:01<14:40:21, 5.70s/it]
79%|███████▉ | 34669/43939 [55:55:07<14:40:25, 5.70s/it]
79%|███████▉ | 34670/43939 [55:55:13<14:51:57, 5.77s/it]
79%|███████▉ | 34671/43939 [55:55:18<14:47:43, 5.75s/it]
79%|███████▉ | 34672/43939 [55:55:24<14:47:14, 5.74s/it]
79%|███████▉ | 34673/43939 [55:55:30<14:48:57, 5.76s/it]
79%|███████▉ | 34674/43939 [55:55:36<14:47:32, 5.75s/it]
79%|███████▉ | 34675/43939 [55:55:41<14:47:07, 5.75s/it]
79%|███████▉ | 34676/43939 [55:55:47<14:44:34, 5.73s/it]
79%|███████▉ | 34677/43939 [55:55:53<14:43:02, 5.72s/it]
79%|███████▉ | 34678/43939 [55:55:58<14:42:34, 5.72s/it]
79%|███████▉ | 34679/43939 [55:56:04<14:41:48, 5.71s/it]
79%|███████▉ | 34680/43939 [55:56:10<14:41:54, 5.71s/it]
79%|███████▉ | 34681/43939 [55:56:16<14:40:54, 5.71s/it]
79%|███████▉ | 34682/43939 [55:56:21<14:40:18, 5.71s/it]
79%|███████▉ | 34683/43939 [55:56:27<14:38:18, 5.69s/it]
79%|███████▉ | 34684/43939 [55:56:33<14:37:15, 5.69s/it]
79%|███████▉ | 34685/43939 [55:56:38<14:38:59, 5.70s/it]
79%|███████▉ | 34686/43939 [55:56:44<14:39:16, 5.70s/it]
79%|███████▉ | 34687/43939 [55:56:50<14:38:02, 5.69s/it]
79%|███████▉ | 34688/43939 [55:56:55<14:38:24, 5.70s/it]
79%|███████▉ | 34689/43939 [55:57:01<14:37:05, 5.69s/it]
79%|███████▉ | 34690/43939 [55:57:07<14:39:15, 5.70s/it]
79%|███████▉ | 34691/43939 [55:57:12<14:38:08, 5.70s/it]
79%|███████▉ | 34692/43939 [55:57:18<14:50:22, 5.78s/it]
79%|███████▉ | 34693/43939 [55:57:24<14:50:36, 5.78s/it]
79%|███████▉ | 34694/43939 [55:57:30<14:49:25, 5.77s/it]
79%|███████▉ | 34695/43939 [55:57:36<14:45:55, 5.75s/it]
79%|███████▉ | 34696/43939 [55:57:41<14:43:30, 5.74s/it]
79%|███████▉ | 34697/43939 [55:57:47<14:41:49, 5.72s/it]
79%|███████▉ | 34698/43939 [55:57:53<14:40:02, 5.71s/it]
79%|███████▉ | 34699/43939 [55:57:58<14:41:17, 5.72s/it]
79%|███████▉ | 34700/43939 [55:58:04<14:39:48, 5.71s/it]
{'loss': 0.675, 'grad_norm': 1.4875781536102295, 'learning_rate': 6.438133035450936e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 520.96, 'epoch': 3.16}
79%|███████▉ | 34700/43939 [55:58:04<14:39:48, 5.71s/it]
79%|███████▉ | 34701/43939 [55:58:10<14:38:05, 5.70s/it]
79%|███████▉ | 34702/43939 [55:58:16<14:37:06, 5.70s/it]
79%|███████▉ | 34703/43939 [55:58:21<14:38:53, 5.71s/it]
79%|███████▉ | 34704/43939 [55:58:27<14:38:00, 5.70s/it]
79%|███████▉ | 34705/43939 [55:58:33<14:38:52, 5.71s/it]
79%|███████▉ | 34706/43939 [55:58:38<14:38:42, 5.71s/it]
79%|███████▉ | 34707/43939 [55:58:44<14:38:12, 5.71s/it]
79%|███████▉ | 34708/43939 [55:58:50<14:37:44, 5.71s/it]
79%|███████▉ | 34709/43939 [55:58:56<14:37:41, 5.71s/it]
79%|███████▉ | 34710/43939 [55:59:01<14:39:21, 5.72s/it]
79%|███████▉ | 34711/43939 [55:59:07<14:43:14, 5.74s/it]
79%|███████▉ | 34712/43939 [55:59:13<14:42:51, 5.74s/it]
79%|███████▉ | 34713/43939 [55:59:19<14:54:43, 5.82s/it]
79%|███████▉ | 34714/43939 [55:59:25<14:51:15, 5.80s/it]
79%|███████▉ | 34715/43939 [55:59:30<14:47:57, 5.78s/it]
79%|███████▉ | 34716/43939 [55:59:36<14:44:35, 5.75s/it]
79%|███████▉ | 34717/43939 [55:59:42<14:41:38, 5.74s/it]
79%|███████▉ | 34718/43939 [55:59:47<14:42:14, 5.74s/it]
79%|███████▉ | 34719/43939 [55:59:53<14:39:03, 5.72s/it]
79%|███████▉ | 34720/43939 [55:59:59<14:38:24, 5.72s/it]
79%|███████▉ | 34721/43939 [56:00:05<14:39:24, 5.72s/it]
79%|███████▉ | 34722/43939 [56:00:10<14:37:41, 5.71s/it]
79%|███████▉ | 34723/43939 [56:00:16<14:37:53, 5.72s/it]
79%|███████▉ | 34724/43939 [56:00:22<14:34:47, 5.70s/it]
79%|███████▉ | 34725/43939 [56:00:27<14:34:40, 5.70s/it]
79%|███████▉ | 34726/43939 [56:00:33<14:37:21, 5.71s/it]
79%|███████▉ | 34727/43939 [56:00:39<14:37:07, 5.71s/it]
79%|███████▉ | 34728/43939 [56:00:44<14:35:34, 5.70s/it]
79%|███████▉ | 34729/43939 [56:00:50<14:36:22, 5.71s/it]
79%|███████▉ | 34730/43939 [56:00:56<14:34:18, 5.70s/it]
79%|███████▉ | 34731/43939 [56:01:02<14:35:52, 5.71s/it]
79%|███████▉ | 34732/43939 [56:01:07<14:37:54, 5.72s/it]
79%|███████▉ | 34733/43939 [56:01:13<14:36:51, 5.71s/it]
79%|███████▉ | 34734/43939 [56:01:19<14:46:41, 5.78s/it]
79%|███████▉ | 34735/43939 [56:01:25<14:44:08, 5.76s/it]
79%|███████▉ | 34736/43939 [56:01:30<14:43:28, 5.76s/it]
79%|███████▉ | 34737/43939 [56:01:36<14:40:53, 5.74s/it]
79%|███████▉ | 34738/43939 [56:01:42<14:37:48, 5.72s/it]
79%|███████▉ | 34739/43939 [56:01:47<14:35:14, 5.71s/it]
79%|███████▉ | 34740/43939 [56:01:53<14:33:23, 5.70s/it]
79%|███████▉ | 34741/43939 [56:01:59<14:32:14, 5.69s/it]
79%|███████▉ | 34742/43939 [56:02:05<14:31:34, 5.69s/it]
79%|███████▉ | 34743/43939 [56:02:10<14:31:01, 5.68s/it]
79%|███████▉ | 34744/43939 [56:02:16<14:31:22, 5.69s/it]
79%|███████▉ | 34745/43939 [56:02:22<14:32:39, 5.69s/it]
79%|███████▉ | 34746/43939 [56:02:27<14:33:19, 5.70s/it]
79%|███████▉ | 34747/43939 [56:02:33<14:33:53, 5.70s/it]
79%|███████▉ | 34748/43939 [56:02:39<14:34:11, 5.71s/it]
79%|███████▉ | 34749/43939 [56:02:44<14:33:25, 5.70s/it]
79%|███████▉ | 34750/43939 [56:02:50<14:34:20, 5.71s/it]
{'loss': 0.651, 'grad_norm': 1.560786247253418, 'learning_rate': 6.37175990083202e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.71, 'epoch': 3.16}
79%|███████▉ | 34750/43939 [56:02:50<14:34:20, 5.71s/it]
79%|███████▉ | 34751/43939 [56:02:56<14:33:27, 5.70s/it]
79%|███████▉ | 34752/43939 [56:03:02<14:35:05, 5.72s/it]
79%|███████▉ | 34753/43939 [56:03:07<14:33:17, 5.70s/it]
79%|███████▉ | 34754/43939 [56:03:13<14:33:25, 5.71s/it]
79%|███████▉ | 34755/43939 [56:03:19<14:34:36, 5.71s/it]
79%|███████▉ | 34756/43939 [56:03:25<14:45:05, 5.78s/it]
79%|███████▉ | 34757/43939 [56:03:30<14:41:11, 5.76s/it]
79%|███████▉ | 34758/43939 [56:03:36<14:39:17, 5.75s/it]
79%|███████▉ | 34759/43939 [56:03:42<14:40:27, 5.75s/it]
79%|███████▉ | 34760/43939 [56:03:48<14:38:45, 5.74s/it]
79%|███████▉ | 34761/43939 [56:03:53<14:36:16, 5.73s/it]
79%|███████▉ | 34762/43939 [56:03:59<14:35:02, 5.72s/it]
79%|███████▉ | 34763/43939 [56:04:05<14:34:12, 5.72s/it]
79%|███████▉ | 34764/43939 [56:04:10<14:33:50, 5.71s/it]
79%|███████▉ | 34765/43939 [56:04:16<14:34:55, 5.72s/it]
79%|███████▉ | 34766/43939 [56:04:22<14:32:07, 5.70s/it]
79%|███████▉ | 34767/43939 [56:04:27<14:31:03, 5.70s/it]
79%|███████▉ | 34768/43939 [56:04:33<14:32:43, 5.71s/it]
79%|███████▉ | 34769/43939 [56:04:39<14:32:05, 5.71s/it]
79%|███████▉ | 34770/43939 [56:04:45<14:32:01, 5.71s/it]
79%|███████▉ | 34771/43939 [56:04:50<14:30:16, 5.70s/it]
79%|███████▉ | 34772/43939 [56:04:56<14:30:03, 5.69s/it]
79%|███████▉ | 34773/43939 [56:05:02<14:31:05, 5.70s/it]
79%|███████▉ | 34774/43939 [56:05:07<14:32:01, 5.71s/it]
79%|███████▉ | 34775/43939 [56:05:13<14:31:11, 5.70s/it]
79%|███████▉ | 34776/43939 [56:05:19<14:29:43, 5.70s/it]
79%|███████▉ | 34777/43939 [56:05:25<14:42:13, 5.78s/it]
79%|███████▉ | 34778/43939 [56:05:30<14:40:15, 5.77s/it]
79%|███████▉ | 34779/43939 [56:05:36<14:37:28, 5.75s/it]
79%|███████▉ | 34780/43939 [56:05:42<14:35:58, 5.74s/it]
79%|███████▉ | 34781/43939 [56:05:48<14:34:40, 5.73s/it]
79%|███████▉ | 34782/43939 [56:05:53<14:32:10, 5.71s/it]
79%|███████▉ | 34783/43939 [56:05:59<14:29:52, 5.70s/it]
79%|███████▉ | 34784/43939 [56:06:05<14:30:25, 5.70s/it]
79%|███████▉ | 34785/43939 [56:06:10<14:30:04, 5.70s/it]
79%|███████▉ | 34786/43939 [56:06:16<14:30:46, 5.71s/it]
79%|███████▉ | 34787/43939 [56:06:22<14:29:39, 5.70s/it]
79%|███████▉ | 34788/43939 [56:06:27<14:28:33, 5.69s/it]
79%|███████▉ | 34789/43939 [56:06:33<14:28:23, 5.69s/it]
79%|███████▉ | 34790/43939 [56:06:39<14:31:44, 5.72s/it]
79%|███████▉ | 34791/43939 [56:06:45<14:31:41, 5.72s/it]
79%|███████▉ | 34792/43939 [56:06:50<14:29:28, 5.70s/it]
79%|███████▉ | 34793/43939 [56:06:56<14:28:20, 5.70s/it]
79%|███████▉ | 34794/43939 [56:07:02<14:27:58, 5.69s/it]
79%|███████▉ | 34795/43939 [56:07:07<14:30:16, 5.71s/it]
79%|███████▉ | 34796/43939 [56:07:13<14:27:38, 5.69s/it]
79%|███████▉ | 34797/43939 [56:07:19<14:29:01, 5.70s/it]
79%|███████▉ | 34798/43939 [56:07:24<14:27:36, 5.69s/it]
79%|███████▉ | 34799/43939 [56:07:30<14:38:35, 5.77s/it]
79%|███████▉ | 34800/43939 [56:07:36<14:36:33, 5.75s/it]
{'loss': 0.6719, 'grad_norm': 1.3014919757843018, 'learning_rate': 6.305680670476436e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.92, 'epoch': 3.17}
79%|███████▉ | 34800/43939 [56:07:36<14:36:33, 5.75s/it]
79%|███████▉ | 34801/43939 [56:07:42<14:36:09, 5.75s/it]
79%|███████▉ | 34802/43939 [56:07:48<14:35:52, 5.75s/it]
79%|███████▉ | 34803/43939 [56:07:53<14:33:17, 5.74s/it]
79%|███████▉ | 34804/43939 [56:07:59<14:31:10, 5.72s/it]
79%|███████▉ | 34805/43939 [56:08:05<14:30:55, 5.72s/it]
79%|███████▉ | 34806/43939 [56:08:10<14:30:32, 5.72s/it]
79%|███████▉ | 34807/43939 [56:08:16<14:27:34, 5.70s/it]
79%|███████▉ | 34808/43939 [56:08:22<14:26:19, 5.69s/it]
79%|███████▉ | 34809/43939 [56:08:27<14:26:06, 5.69s/it]
79%|███████▉ | 34810/43939 [56:08:33<14:26:08, 5.69s/it]
79%|███████▉ | 34811/43939 [56:08:39<14:26:34, 5.70s/it]
79%|███████▉ | 34812/43939 [56:08:45<14:26:25, 5.70s/it]
79%|███████▉ | 34813/43939 [56:08:50<14:24:36, 5.68s/it]
79%|███████▉ | 34814/43939 [56:08:56<14:25:22, 5.69s/it]
79%|███████▉ | 34815/43939 [56:09:02<14:28:12, 5.71s/it]
79%|███████▉ | 34816/43939 [56:09:07<14:27:14, 5.70s/it]
79%|███████▉ | 34817/43939 [56:09:13<14:26:31, 5.70s/it]
79%|███████▉ | 34818/43939 [56:09:19<14:26:08, 5.70s/it]
79%|███████▉ | 34819/43939 [56:09:24<14:25:08, 5.69s/it]
79%|███████▉ | 34820/43939 [56:09:30<14:25:56, 5.70s/it]
79%|███████▉ | 34821/43939 [56:09:36<14:34:59, 5.76s/it]
79%|███████▉ | 34822/43939 [56:09:42<14:33:42, 5.75s/it]
79%|███████▉ | 34823/43939 [56:09:47<14:30:37, 5.73s/it]
79%|███████▉ | 34824/43939 [56:09:53<14:29:11, 5.72s/it]
79%|███████▉ | 34825/43939 [56:09:59<14:27:44, 5.71s/it]
79%|███████▉ | 34826/43939 [56:10:05<14:26:40, 5.71s/it]
79%|███████▉ | 34827/43939 [56:10:10<14:27:26, 5.71s/it]
79%|███████▉ | 34828/43939 [56:10:16<14:26:45, 5.71s/it]
79%|███████▉ | 34829/43939 [56:10:22<14:26:13, 5.71s/it]
79%|███████▉ | 34830/43939 [56:10:27<14:25:14, 5.70s/it]
79%|███████▉ | 34831/43939 [56:10:33<14:25:56, 5.70s/it]
79%|███████▉ | 34832/43939 [56:10:39<14:25:41, 5.70s/it]
79%|███████▉ | 34833/43939 [56:10:44<14:24:57, 5.70s/it]
79%|███████▉ | 34834/43939 [56:10:50<14:26:05, 5.71s/it]
79%|███████▉ | 34835/43939 [56:10:56<14:25:06, 5.70s/it]
79%|███████▉ | 34836/43939 [56:11:02<14:25:52, 5.71s/it]
79%|███████▉ | 34837/43939 [56:11:07<14:23:35, 5.69s/it]
79%|███████▉ | 34838/43939 [56:11:13<14:23:35, 5.69s/it]
79%|███████▉ | 34839/43939 [56:11:19<14:25:48, 5.71s/it]
79%|███████▉ | 34840/43939 [56:11:24<14:24:11, 5.70s/it]
79%|███████▉ | 34841/43939 [56:11:30<14:22:46, 5.69s/it]
79%|███████▉ | 34842/43939 [56:11:36<14:34:58, 5.77s/it]
79%|███████▉ | 34843/43939 [56:11:42<14:32:38, 5.76s/it]
79%|███████▉ | 34844/43939 [56:11:47<14:28:10, 5.73s/it]
79%|███████▉ | 34845/43939 [56:11:53<14:26:50, 5.72s/it]
79%|███████▉ | 34846/43939 [56:11:59<14:25:43, 5.71s/it]
79%|███████▉ | 34847/43939 [56:12:04<14:24:40, 5.71s/it]
79%|███████▉ | 34848/43939 [56:12:10<14:25:35, 5.71s/it]
79%|███████▉ | 34849/43939 [56:12:16<14:24:37, 5.71s/it]
79%|███████▉ | 34850/43939 [56:12:22<14:24:24, 5.71s/it]
{'loss': 0.6454, 'grad_norm': 1.5218391418457031, 'learning_rate': 6.239896386939323e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 451.03, 'epoch': 3.17}
79%|███████▉ | 34850/43939 [56:12:22<14:24:24, 5.71s/it]
79%|███████▉ | 34851/43939 [56:12:27<14:23:03, 5.70s/it]
79%|███████▉ | 34852/43939 [56:12:33<14:23:33, 5.70s/it]
79%|███████▉ | 34853/43939 [56:12:39<14:21:58, 5.69s/it]
79%|███████▉ | 34854/43939 [56:12:44<14:22:35, 5.70s/it]
79%|███████▉ | 34855/43939 [56:12:50<14:23:45, 5.71s/it]
79%|███████▉ | 34856/43939 [56:12:56<14:21:42, 5.69s/it]
79%|███████▉ | 34857/43939 [56:13:02<14:24:21, 5.71s/it]
79%|███████▉ | 34858/43939 [56:13:07<14:22:52, 5.70s/it]
79%|███████▉ | 34859/43939 [56:13:13<14:23:14, 5.70s/it]
79%|███████▉ | 34860/43939 [56:13:19<14:22:28, 5.70s/it]
79%|███████▉ | 34861/43939 [56:13:24<14:24:50, 5.72s/it]
79%|███████▉ | 34862/43939 [56:13:30<14:24:42, 5.72s/it]
79%|███████▉ | 34863/43939 [56:13:36<14:22:55, 5.70s/it]
79%|███████▉ | 34864/43939 [56:13:42<14:33:55, 5.78s/it]
79%|███████▉ | 34865/43939 [56:13:47<14:30:51, 5.76s/it]
79%|███████▉ | 34866/43939 [56:13:53<14:29:55, 5.75s/it]
79%|███████▉ | 34867/43939 [56:13:59<14:27:22, 5.74s/it]
79%|███████▉ | 34868/43939 [56:14:05<14:24:45, 5.72s/it]
79%|███████▉ | 34869/43939 [56:14:10<14:25:26, 5.73s/it]
79%|███████▉ | 34870/43939 [56:14:16<14:24:54, 5.72s/it]
79%|███████▉ | 34871/43939 [56:14:22<14:24:07, 5.72s/it]
79%|███████▉ | 34872/43939 [56:14:27<14:21:49, 5.70s/it]
79%|███████▉ | 34873/43939 [56:14:33<14:22:21, 5.71s/it]
79%|███████▉ | 34874/43939 [56:14:39<14:23:08, 5.71s/it]
79%|███████▉ | 34875/43939 [56:14:45<14:23:09, 5.71s/it]
79%|███████▉ | 34876/43939 [56:14:50<14:23:29, 5.72s/it]
79%|███████▉ | 34877/43939 [56:14:56<14:21:44, 5.71s/it]
79%|███████▉ | 34878/43939 [56:15:02<14:20:55, 5.70s/it]
79%|███████▉ | 34879/43939 [56:15:07<14:20:25, 5.70s/it]
79%|███████▉ | 34880/43939 [56:15:13<14:21:21, 5.70s/it]
79%|███████▉ | 34881/43939 [56:15:19<14:20:32, 5.70s/it]
79%|███████▉ | 34882/43939 [56:15:24<14:19:23, 5.69s/it]
79%|███████▉ | 34883/43939 [56:15:30<14:18:20, 5.69s/it]
79%|███████▉ | 34884/43939 [56:15:36<14:18:59, 5.69s/it]
79%|███████▉ | 34885/43939 [56:15:42<14:31:23, 5.77s/it]
79%|███████▉ | 34886/43939 [56:15:47<14:27:47, 5.75s/it]
79%|███████▉ | 34887/43939 [56:15:53<14:25:10, 5.73s/it]
79%|███████▉ | 34888/43939 [56:15:59<14:25:21, 5.74s/it]
79%|███████▉ | 34889/43939 [56:16:05<14:24:23, 5.73s/it]
79%|███████▉ | 34890/43939 [56:16:10<14:22:38, 5.72s/it]
79%|███████▉ | 34891/43939 [56:16:16<14:22:10, 5.72s/it]
79%|███████▉ | 34892/43939 [56:16:22<14:21:41, 5.71s/it]
79%|███████▉ | 34893/43939 [56:16:27<14:20:23, 5.71s/it]
79%|███████▉ | 34894/43939 [56:16:33<14:19:55, 5.70s/it]
79%|███████▉ | 34895/43939 [56:16:39<14:18:44, 5.70s/it]
79%|███████▉ | 34896/43939 [56:16:44<14:17:18, 5.69s/it]
79%|███████▉ | 34897/43939 [56:16:50<14:19:52, 5.71s/it]
79%|███████▉ | 34898/43939 [56:16:56<14:19:08, 5.70s/it]
79%|███████▉ | 34899/43939 [56:17:02<14:19:21, 5.70s/it]
79%|███████▉ | 34900/43939 [56:17:07<14:18:05, 5.70s/it]
{'loss': 0.6532, 'grad_norm': 1.9518988132476807, 'learning_rate': 6.174408088122344e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.78, 'epoch': 3.18}
79%|███████▉ | 34900/43939 [56:17:07<14:18:05, 5.70s/it]
79%|███████▉ | 34901/43939 [56:17:13<14:18:09, 5.70s/it]
79%|███████▉ | 34902/43939 [56:17:19<14:17:48, 5.70s/it]
79%|███████▉ | 34903/43939 [56:17:24<14:18:23, 5.70s/it]
79%|███████▉ | 34904/43939 [56:17:30<14:18:46, 5.70s/it]
79%|███████▉ | 34905/43939 [56:17:36<14:18:10, 5.70s/it]
79%|███████▉ | 34906/43939 [56:17:41<14:18:01, 5.70s/it]
79%|███████▉ | 34907/43939 [56:17:47<14:27:43, 5.76s/it]
79%|███████▉ | 34908/43939 [56:17:53<14:23:59, 5.74s/it]
79%|███████▉ | 34909/43939 [56:17:59<14:22:24, 5.73s/it]
79%|███████▉ | 34910/43939 [56:18:04<14:19:28, 5.71s/it]
79%|███████▉ | 34911/43939 [56:18:10<14:18:00, 5.70s/it]
79%|███████▉ | 34912/43939 [56:18:16<14:17:26, 5.70s/it]
79%|███████▉ | 34913/43939 [56:18:21<14:15:56, 5.69s/it]
79%|███████▉ | 34914/43939 [56:18:27<14:17:09, 5.70s/it]
79%|███████▉ | 34915/43939 [56:18:33<14:18:20, 5.71s/it]
79%|███████▉ | 34916/43939 [56:18:39<14:20:08, 5.72s/it]
79%|███████▉ | 34917/43939 [56:18:44<14:19:32, 5.72s/it]
79%|███████▉ | 34918/43939 [56:18:50<14:18:43, 5.71s/it]
79%|███████▉ | 34919/43939 [56:18:56<14:18:04, 5.71s/it]
79%|███████▉ | 34920/43939 [56:19:01<14:18:30, 5.71s/it]
79%|███████▉ | 34921/43939 [56:19:07<14:17:42, 5.71s/it]
79%|███████▉ | 34922/43939 [56:19:13<14:15:52, 5.70s/it]
79%|███████▉ | 34923/43939 [56:19:19<14:16:30, 5.70s/it]
79%|███████▉ | 34924/43939 [56:19:24<14:15:56, 5.70s/it]
79%|███████▉ | 34925/43939 [56:19:30<14:16:44, 5.70s/it]
79%|███████▉ | 34926/43939 [56:19:36<14:15:48, 5.70s/it]
79%|███████▉ | 34927/43939 [56:19:41<14:16:35, 5.70s/it]
79%|███████▉ | 34928/43939 [56:19:47<14:28:59, 5.79s/it]
79%|███████▉ | 34929/43939 [56:19:53<14:23:55, 5.75s/it]
79%|███████▉ | 34930/43939 [56:19:59<14:21:52, 5.74s/it]
79%|███████▉ | 34931/43939 [56:20:04<14:19:14, 5.72s/it]
80%|███████▉ | 34932/43939 [56:20:10<14:21:08, 5.74s/it]
80%|███████▉ | 34933/43939 [56:20:16<14:21:23, 5.74s/it]
80%|███████▉ | 34934/43939 [56:20:22<14:18:54, 5.72s/it]
80%|███████▉ | 34935/43939 [56:20:27<14:19:17, 5.73s/it]
80%|███████▉ | 34936/43939 [56:20:33<14:17:40, 5.72s/it]
80%|███████▉ | 34937/43939 [56:20:39<14:16:39, 5.71s/it]
80%|███████▉ | 34938/43939 [56:20:44<14:17:16, 5.71s/it]
80%|███████▉ | 34939/43939 [56:20:50<14:17:17, 5.72s/it]
80%|███████▉ | 34940/43939 [56:20:56<14:17:26, 5.72s/it]
80%|███████▉ | 34941/43939 [56:21:02<14:17:58, 5.72s/it]
80%|███████▉ | 34942/43939 [56:21:07<14:16:18, 5.71s/it]
80%|███████▉ | 34943/43939 [56:21:13<14:16:47, 5.71s/it]
80%|███████▉ | 34944/43939 [56:21:19<14:16:59, 5.72s/it]
80%|███████▉ | 34945/43939 [56:21:24<14:16:47, 5.72s/it]
80%|███████▉ | 34946/43939 [56:21:30<14:17:56, 5.72s/it]
80%|███████▉ | 34947/43939 [56:21:36<14:17:08, 5.72s/it]
80%|███████▉ | 34948/43939 [56:21:42<14:17:03, 5.72s/it]
80%|███████▉ | 34949/43939 [56:21:47<14:15:51, 5.71s/it]
80%|███████▉ | 34950/43939 [56:21:53<14:24:38, 5.77s/it]
{'loss': 0.6719, 'grad_norm': 1.97776198387146, 'learning_rate': 6.1092168072573095e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.38, 'epoch': 3.18}
80%|███████▉ | 34950/43939 [56:21:53<14:24:38, 5.77s/it]
80%|███████▉ | 34951/43939 [56:21:59<14:23:37, 5.77s/it]
80%|███████▉ | 34952/43939 [56:22:05<14:20:20, 5.74s/it]
80%|███████▉ | 34953/43939 [56:22:10<14:19:12, 5.74s/it]
80%|███████▉ | 34954/43939 [56:22:16<14:20:06, 5.74s/it]
80%|███████▉ | 34955/43939 [56:22:22<14:18:54, 5.74s/it]
80%|███████▉ | 34956/43939 [56:22:28<14:16:30, 5.72s/it]
80%|███████▉ | 34957/43939 [56:22:33<14:15:08, 5.71s/it]
80%|███████▉ | 34958/43939 [56:22:39<14:15:10, 5.71s/it]
80%|███████▉ | 34959/43939 [56:22:45<14:16:25, 5.72s/it]
80%|███████▉ | 34960/43939 [56:22:50<14:16:34, 5.72s/it]
80%|███████▉ | 34961/43939 [56:22:56<14:14:55, 5.71s/it]
80%|███████▉ | 34962/43939 [56:23:02<14:15:18, 5.72s/it]
80%|███████▉ | 34963/43939 [56:23:08<14:15:27, 5.72s/it]
80%|███████▉ | 34964/43939 [56:23:13<14:15:08, 5.72s/it]
80%|███████▉ | 34965/43939 [56:23:19<14:13:15, 5.70s/it]
80%|███████▉ | 34966/43939 [56:23:25<14:13:10, 5.70s/it]
80%|███████▉ | 34967/43939 [56:23:30<14:14:23, 5.71s/it]
80%|███████▉ | 34968/43939 [56:23:36<14:14:25, 5.71s/it]
80%|███████▉ | 34969/43939 [56:23:42<14:14:39, 5.72s/it]
80%|███████▉ | 34970/43939 [56:23:48<14:13:26, 5.71s/it]
80%|███████▉ | 34971/43939 [56:23:53<14:23:27, 5.78s/it]
80%|███████▉ | 34972/43939 [56:23:59<14:19:57, 5.75s/it]
80%|███████▉ | 34973/43939 [56:24:05<14:18:14, 5.74s/it]
80%|███████▉ | 34974/43939 [56:24:11<14:15:53, 5.73s/it]
80%|███████▉ | 34975/43939 [56:24:16<14:15:04, 5.72s/it]
80%|███████▉ | 34976/43939 [56:24:22<14:14:17, 5.72s/it]
80%|███████▉ | 34977/43939 [56:24:28<14:13:18, 5.71s/it]
80%|███████▉ | 34978/43939 [56:24:33<14:12:10, 5.71s/it]
80%|███████▉ | 34979/43939 [56:24:39<14:11:20, 5.70s/it]
80%|███████▉ | 34980/43939 [56:24:45<14:12:59, 5.71s/it]
80%|███████▉ | 34981/43939 [56:24:51<14:12:00, 5.71s/it]
80%|███████▉ | 34982/43939 [56:24:56<14:11:45, 5.71s/it]
80%|███████▉ | 34983/43939 [56:25:02<14:09:54, 5.69s/it]
80%|███████▉ | 34984/43939 [56:25:08<14:10:02, 5.70s/it]
80%|███████▉ | 34985/43939 [56:25:13<14:10:10, 5.70s/it]
80%|███████▉ | 34986/43939 [56:25:19<14:10:03, 5.70s/it]
80%|███████▉ | 34987/43939 [56:25:25<14:10:49, 5.70s/it]
80%|███████▉ | 34988/43939 [56:25:30<14:10:18, 5.70s/it]
80%|███████▉ | 34989/43939 [56:25:36<14:08:57, 5.69s/it]
80%|███████▉ | 34990/43939 [56:25:42<14:08:44, 5.69s/it]
80%|███████▉ | 34991/43939 [56:25:47<14:09:27, 5.70s/it]
80%|███████▉ | 34992/43939 [56:25:53<14:09:08, 5.69s/it]
80%|███████▉ | 34993/43939 [56:25:59<14:18:07, 5.76s/it]
80%|███████▉ | 34994/43939 [56:26:05<14:15:56, 5.74s/it]
80%|███████▉ | 34995/43939 [56:26:10<14:14:01, 5.73s/it]
80%|███████▉ | 34996/43939 [56:26:16<14:12:55, 5.72s/it]
80%|███████▉ | 34997/43939 [56:26:22<14:12:04, 5.72s/it]
80%|███████▉ | 34998/43939 [56:26:28<14:09:24, 5.70s/it]
80%|███████▉ | 34999/43939 [56:26:33<14:08:41, 5.70s/it]
80%|███████▉ | 35000/43939 [56:26:39<14:11:01, 5.71s/it]
{'loss': 0.655, 'grad_norm': 1.3716925382614136, 'learning_rate': 6.04432357288987e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.91, 'epoch': 3.19}
80%|███████▉ | 35000/43939 [56:26:39<14:11:01, 5.71s/it]
80%|███████▉ | 35001/43939 [56:26:45<14:10:15, 5.71s/it]
80%|███████▉ | 35002/43939 [56:26:50<14:11:56, 5.72s/it]
80%|███████▉ | 35003/43939 [56:26:56<14:10:25, 5.71s/it]
80%|███████▉ | 35004/43939 [56:27:02<14:09:24, 5.70s/it]
80%|███████▉ | 35005/43939 [56:27:07<14:08:44, 5.70s/it]
80%|███████▉ | 35006/43939 [56:27:13<14:08:15, 5.70s/it]
80%|███████▉ | 35007/43939 [56:27:19<14:08:37, 5.70s/it]
80%|███████▉ | 35008/43939 [56:27:25<14:10:37, 5.71s/it]
80%|███████▉ | 35009/43939 [56:27:30<14:10:51, 5.72s/it]
80%|███████▉ | 35010/43939 [56:27:36<14:09:32, 5.71s/it]
80%|███████▉ | 35011/43939 [56:27:42<14:09:27, 5.71s/it]
80%|███████▉ | 35012/43939 [56:27:47<14:08:41, 5.70s/it]
80%|███████▉ | 35013/43939 [56:27:53<14:20:06, 5.78s/it]
80%|███████▉ | 35014/43939 [56:27:59<14:15:02, 5.75s/it]
80%|███████▉ | 35015/43939 [56:28:05<14:11:39, 5.73s/it]
80%|███████▉ | 35016/43939 [56:28:11<14:13:10, 5.74s/it]
80%|███████▉ | 35017/43939 [56:28:16<14:11:57, 5.73s/it]
80%|███████▉ | 35018/43939 [56:28:22<14:11:01, 5.72s/it]
80%|███████▉ | 35019/43939 [56:28:28<14:10:31, 5.72s/it]
80%|███████▉ | 35020/43939 [56:28:33<14:10:32, 5.72s/it]
80%|███████▉ | 35021/43939 [56:28:39<14:11:13, 5.73s/it]
80%|███████▉ | 35022/43939 [56:28:45<14:10:49, 5.72s/it]
80%|███████▉ | 35023/43939 [56:28:51<14:10:41, 5.72s/it]
80%|███████▉ | 35024/43939 [56:28:56<14:13:03, 5.74s/it]
80%|███████▉ | 35025/43939 [56:29:02<14:10:03, 5.72s/it]
80%|███████▉ | 35026/43939 [56:29:08<14:08:30, 5.71s/it]
80%|███████▉ | 35027/43939 [56:29:13<14:07:45, 5.71s/it]
80%|███████▉ | 35028/43939 [56:29:19<14:08:07, 5.71s/it]
80%|███████▉ | 35029/43939 [56:29:25<14:06:38, 5.70s/it]
80%|███████▉ | 35030/43939 [56:29:31<14:07:37, 5.71s/it]
80%|███████▉ | 35031/43939 [56:29:36<14:08:44, 5.72s/it]
80%|███████▉ | 35032/43939 [56:29:42<14:10:16, 5.73s/it]
80%|███████▉ | 35033/43939 [56:29:48<14:08:02, 5.71s/it]
80%|███████▉ | 35034/43939 [56:29:53<14:07:37, 5.71s/it]
80%|███████▉ | 35035/43939 [56:29:59<14:19:06, 5.79s/it]
80%|███████▉ | 35036/43939 [56:30:05<14:16:32, 5.77s/it]
80%|███████▉ | 35037/43939 [56:30:11<14:13:20, 5.75s/it]
80%|███████▉ | 35038/43939 [56:30:17<14:10:38, 5.73s/it]
80%|███████▉ | 35039/43939 [56:30:22<14:12:12, 5.75s/it]
80%|███████▉ | 35040/43939 [56:30:28<14:10:29, 5.73s/it]
80%|███████▉ | 35041/43939 [56:30:34<14:09:45, 5.73s/it]
80%|███████▉ | 35042/43939 [56:30:39<14:07:33, 5.72s/it]
80%|███████▉ | 35043/43939 [56:30:45<14:06:46, 5.71s/it]
80%|███████▉ | 35044/43939 [56:30:51<14:07:50, 5.72s/it]
80%|███████▉ | 35045/43939 [56:30:57<14:07:25, 5.72s/it]
80%|███████▉ | 35046/43939 [56:31:02<14:08:10, 5.72s/it]
80%|███████▉ | 35047/43939 [56:31:08<14:07:21, 5.72s/it]
80%|███████▉ | 35048/43939 [56:31:14<14:07:05, 5.72s/it]
80%|███████▉ | 35049/43939 [56:31:19<14:05:50, 5.71s/it]
80%|███████▉ | 35050/43939 [56:31:25<14:04:28, 5.70s/it]
{'loss': 0.6699, 'grad_norm': 1.4617153406143188, 'learning_rate': 5.979729408863299e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.89, 'epoch': 3.19}
80%|███████▉ | 35050/43939 [56:31:25<14:04:28, 5.70s/it]
80%|███████▉ | 35051/43939 [56:31:31<14:05:39, 5.71s/it]
80%|███████▉ | 35052/43939 [56:31:36<14:05:01, 5.71s/it]
80%|███████▉ | 35053/43939 [56:31:42<14:04:57, 5.71s/it]
80%|███████▉ | 35054/43939 [56:31:48<14:03:25, 5.70s/it]
80%|███████▉ | 35055/43939 [56:31:54<14:04:09, 5.70s/it]
80%|███████▉ | 35056/43939 [56:31:59<14:07:15, 5.72s/it]
80%|███████▉ | 35057/43939 [56:32:05<14:18:40, 5.80s/it]
80%|███████▉ | 35058/43939 [56:32:11<14:14:45, 5.77s/it]
80%|███████▉ | 35059/43939 [56:32:17<14:10:43, 5.75s/it]
80%|███████▉ | 35060/43939 [56:32:22<14:07:56, 5.73s/it]
80%|███████▉ | 35061/43939 [56:32:28<14:07:06, 5.72s/it]
80%|███████▉ | 35062/43939 [56:32:34<14:05:00, 5.71s/it]
80%|███████▉ | 35063/43939 [56:32:39<14:03:16, 5.70s/it]
80%|███████▉ | 35064/43939 [56:32:45<14:05:10, 5.71s/it]
80%|███████▉ | 35065/43939 [56:32:51<14:08:20, 5.74s/it]
80%|███████▉ | 35066/43939 [56:32:57<14:05:58, 5.72s/it]
80%|███████▉ | 35067/43939 [56:33:02<14:02:37, 5.70s/it]
80%|███████▉ | 35068/43939 [56:33:08<14:01:48, 5.69s/it]
80%|███████▉ | 35069/43939 [56:33:14<14:01:55, 5.70s/it]
80%|███████▉ | 35070/43939 [56:33:19<14:03:29, 5.71s/it]
80%|███████▉ | 35071/43939 [56:33:25<14:01:55, 5.70s/it]
80%|███████▉ | 35072/43939 [56:33:31<14:04:12, 5.71s/it]
80%|███████▉ | 35073/43939 [56:33:37<14:04:05, 5.71s/it]
80%|███████▉ | 35074/43939 [56:33:42<14:02:26, 5.70s/it]
80%|███████▉ | 35075/43939 [56:33:48<14:02:09, 5.70s/it]
80%|███████▉ | 35076/43939 [56:33:54<14:04:19, 5.72s/it]
80%|███████▉ | 35077/43939 [56:33:59<14:03:08, 5.71s/it]
80%|███████▉ | 35078/43939 [56:34:05<14:12:12, 5.77s/it]
80%|███████▉ | 35079/43939 [56:34:11<14:10:18, 5.76s/it]
80%|███████▉ | 35080/43939 [56:34:17<14:08:29, 5.75s/it]
80%|███████▉ | 35081/43939 [56:34:23<14:08:56, 5.75s/it]
80%|███████▉ | 35082/43939 [56:34:28<14:06:54, 5.74s/it]
80%|███████▉ | 35083/43939 [56:34:34<14:06:37, 5.74s/it]
80%|███████▉ | 35084/43939 [56:34:40<14:04:18, 5.72s/it]
80%|███████▉ | 35085/43939 [56:34:45<14:04:23, 5.72s/it]
80%|███████▉ | 35086/43939 [56:34:51<14:02:03, 5.71s/it]
80%|███████▉ | 35087/43939 [56:34:57<14:03:38, 5.72s/it]
80%|███████▉ | 35088/43939 [56:35:03<14:04:18, 5.72s/it]
80%|███████▉ | 35089/43939 [56:35:08<14:03:56, 5.72s/it]
80%|███████▉ | 35090/43939 [56:35:14<14:02:27, 5.71s/it]
80%|███████▉ | 35091/43939 [56:35:20<13:59:40, 5.69s/it]
80%|███████▉ | 35092/43939 [56:35:25<13:58:06, 5.68s/it]
80%|███████▉ | 35093/43939 [56:35:31<13:58:33, 5.69s/it]
80%|███████▉ | 35094/43939 [56:35:37<14:01:57, 5.71s/it]
80%|███████▉ | 35095/43939 [56:35:42<13:59:22, 5.69s/it]
80%|███████▉ | 35096/43939 [56:35:48<13:58:56, 5.69s/it]
80%|███████▉ | 35097/43939 [56:35:54<13:59:23, 5.70s/it]
80%|███████▉ | 35098/43939 [56:35:59<13:59:25, 5.70s/it]
80%|███████▉ | 35099/43939 [56:36:05<14:01:16, 5.71s/it]
80%|███████▉ | 35100/43939 [56:36:11<14:10:07, 5.77s/it]
{'loss': 0.651, 'grad_norm': 1.5557167530059814, 'learning_rate': 5.9154353343023245e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.63, 'epoch': 3.2}
80%|███████▉ | 35100/43939 [56:36:11<14:10:07, 5.77s/it]
80%|███████▉ | 35101/43939 [56:36:17<14:06:41, 5.75s/it]
80%|███████▉ | 35102/43939 [56:36:22<14:02:23, 5.72s/it]
80%|███████▉ | 35103/43939 [56:36:28<14:01:44, 5.72s/it]
80%|███████▉ | 35104/43939 [56:36:34<14:01:32, 5.72s/it]
80%|███████▉ | 35105/43939 [56:36:40<14:02:34, 5.72s/it]
80%|███████▉ | 35106/43939 [56:36:45<14:00:43, 5.71s/it]
80%|███████▉ | 35107/43939 [56:36:51<14:00:17, 5.71s/it]
80%|███████▉ | 35108/43939 [56:36:57<14:00:06, 5.71s/it]
80%|███████▉ | 35109/43939 [56:37:02<14:00:17, 5.71s/it]
80%|███████▉ | 35110/43939 [56:37:08<13:59:06, 5.70s/it]
80%|███████▉ | 35111/43939 [56:37:14<13:58:03, 5.70s/it]
80%|███████▉ | 35112/43939 [56:37:20<13:59:15, 5.70s/it]
80%|███████▉ | 35113/43939 [56:37:25<14:00:02, 5.71s/it]
80%|███████▉ | 35114/43939 [56:37:31<13:59:38, 5.71s/it]
80%|███████▉ | 35115/43939 [56:37:37<13:58:45, 5.70s/it]
80%|███████▉ | 35116/43939 [56:37:42<13:58:25, 5.70s/it]
80%|███████▉ | 35117/43939 [56:37:48<13:58:13, 5.70s/it]
80%|███████▉ | 35118/43939 [56:37:54<13:59:05, 5.71s/it]
80%|███████▉ | 35119/43939 [56:38:00<13:59:22, 5.71s/it]
80%|███████▉ | 35120/43939 [56:38:05<13:59:10, 5.71s/it]
80%|███████▉ | 35121/43939 [56:38:11<14:08:15, 5.77s/it]
80%|███████▉ | 35122/43939 [56:38:17<14:05:56, 5.76s/it]
80%|███████▉ | 35123/43939 [56:38:23<14:03:00, 5.74s/it]
80%|███████▉ | 35124/43939 [56:38:28<14:01:22, 5.73s/it]
80%|███████▉ | 35125/43939 [56:38:34<13:59:43, 5.72s/it]
80%|███████▉ | 35126/43939 [56:38:40<13:58:35, 5.71s/it]
80%|███████▉ | 35127/43939 [56:38:45<13:58:09, 5.71s/it]
80%|███████▉ | 35128/43939 [56:38:51<13:58:45, 5.71s/it]
80%|███████▉ | 35129/43939 [56:38:57<13:58:23, 5.71s/it]
80%|███████▉ | 35130/43939 [56:39:02<13:59:32, 5.72s/it]
80%|███████▉ | 35131/43939 [56:39:08<13:59:30, 5.72s/it]
80%|███████▉ | 35132/43939 [56:39:14<13:57:56, 5.71s/it]
80%|███████▉ | 35133/43939 [56:39:20<13:56:41, 5.70s/it]
80%|███████▉ | 35134/43939 [56:39:25<13:57:13, 5.71s/it]
80%|███████▉ | 35135/43939 [56:39:31<13:58:02, 5.71s/it]
80%|███████▉ | 35136/43939 [56:39:37<13:56:42, 5.70s/it]
80%|███████▉ | 35137/43939 [56:39:42<13:58:37, 5.72s/it]
80%|███████▉ | 35138/43939 [56:39:48<13:59:20, 5.72s/it]
80%|███████▉ | 35139/43939 [56:39:54<13:59:00, 5.72s/it]
80%|███████▉ | 35140/43939 [56:40:00<14:00:29, 5.73s/it]
80%|███████▉ | 35141/43939 [56:40:05<14:02:19, 5.74s/it]
80%|███████▉ | 35142/43939 [56:40:11<14:01:54, 5.74s/it]
80%|███████▉ | 35143/43939 [56:40:17<14:09:57, 5.80s/it]
80%|███████▉ | 35144/43939 [56:40:23<14:07:01, 5.78s/it]
80%|███████▉ | 35145/43939 [56:40:29<14:03:05, 5.75s/it]
80%|███████▉ | 35146/43939 [56:40:34<14:00:20, 5.73s/it]
80%|███████▉ | 35147/43939 [56:40:40<14:00:00, 5.73s/it]
80%|███████▉ | 35148/43939 [56:40:46<13:59:26, 5.73s/it]
80%|███████▉ | 35149/43939 [56:40:51<13:58:36, 5.72s/it]
80%|███████▉ | 35150/43939 [56:40:57<13:57:48, 5.72s/it]
{'loss': 0.6672, 'grad_norm': 1.5268049240112305, 'learning_rate': 5.8514423635970626e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.69, 'epoch': 3.2}
80%|███████▉ | 35150/43939 [56:40:57<13:57:48, 5.72s/it]
80%|███████▉ | 35151/43939 [56:41:03<13:55:47, 5.71s/it]
80%|████████ | 35152/43939 [56:41:08<13:54:49, 5.70s/it][2025-12-01 02:45:22,271] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.38s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.94s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:48, 2.67s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.69s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.70s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:40, 2.72s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:41<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:25, 2.74s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.74s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [01:00<02:14, 2.74s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.74s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.74s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.74s/it][A
38%|███▊ | 27/72 [01:11<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.74s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:47, 2.75s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:33<01:41, 2.74s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:44<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.74s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.74s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.74s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.74s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.74s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:52, 2.74s/it][A
75%|███████▌ | 54/72 [02:25<00:49, 2.74s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.74s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.74s/it][A
81%|████████ | 58/72 [02:36<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.74s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.74s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.74s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:06<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.74s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 1.0137217044830322, 'eval_runtime': 196.5158, 'eval_samples_per_second': 5.084, 'eval_steps_per_second': 0.366, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 3.2}
80%|████████ | 35152/43939 [56:44:25<13:54:49, 5.70s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
80%|████████ | 35153/43939 [56:44:31<157:55:56, 64.71s/it]
80%|████████ | 35154/43939 [56:44:37<114:42:54, 47.01s/it]
80%|████████ | 35155/43939 [56:44:42<84:27:25, 34.61s/it]
80%|████████ | 35156/43939 [56:44:48<63:19:44, 25.96s/it]
80%|████████ | 35157/43939 [56:44:54<48:30:56, 19.89s/it]
80%|████████ | 35158/43939 [56:44:59<38:07:04, 15.63s/it]
80%|████████ | 35159/43939 [56:45:05<30:49:46, 12.64s/it]
80%|████████ | 35160/43939 [56:45:11<25:46:57, 10.57s/it]
80%|████████ | 35161/43939 [56:45:17<22:11:59, 9.10s/it]
80%|████████ | 35162/43939 [56:45:22<19:41:26, 8.08s/it]
80%|████████ | 35163/43939 [56:45:28<17:56:54, 7.36s/it]
80%|████████ | 35164/43939 [56:45:34<16:57:09, 6.95s/it]
80%|████████ | 35165/43939 [56:45:40<16:04:01, 6.59s/it]
80%|████████ | 35166/43939 [56:45:45<15:24:59, 6.33s/it]
80%|████████ | 35167/43939 [56:45:51<14:58:13, 6.14s/it]
80%|████████ | 35168/43939 [56:45:57<14:40:23, 6.02s/it]
80%|████████ | 35169/43939 [56:46:03<14:26:02, 5.92s/it]
80%|████████ | 35170/43939 [56:46:08<14:16:15, 5.86s/it]
80%|████████ | 35171/43939 [56:46:14<14:10:20, 5.82s/it]
80%|████████ | 35172/43939 [56:46:20<14:03:21, 5.77s/it]
80%|████████ | 35173/43939 [56:46:25<13:59:40, 5.75s/it]
80%|████████ | 35174/43939 [56:46:31<13:58:51, 5.74s/it]
80%|████████ | 35175/43939 [56:46:37<13:56:30, 5.73s/it]
80%|████████ | 35176/43939 [56:46:42<13:55:44, 5.72s/it]
80%|████████ | 35177/43939 [56:46:48<13:54:13, 5.71s/it]
80%|████████ | 35178/43939 [56:46:54<13:52:33, 5.70s/it]
80%|████████ | 35179/43939 [56:46:59<13:52:47, 5.70s/it]
80%|████████ | 35180/43939 [56:47:05<13:53:29, 5.71s/it]
80%|████████ | 35181/43939 [56:47:11<13:53:06, 5.71s/it]
80%|████████ | 35182/43939 [56:47:17<13:53:30, 5.71s/it]
80%|████████ | 35183/43939 [56:47:22<13:55:33, 5.73s/it]
80%|████████ | 35184/43939 [56:47:28<13:56:00, 5.73s/it]
80%|████████ | 35185/43939 [56:47:34<13:53:45, 5.71s/it]
80%|████████ | 35186/43939 [56:47:40<14:05:12, 5.79s/it]
80%|████████ | 35187/43939 [56:47:45<14:00:23, 5.76s/it]
80%|████████ | 35188/43939 [56:47:51<13:57:00, 5.74s/it]
80%|████████ | 35189/43939 [56:47:57<13:55:22, 5.73s/it]
80%|████████ | 35190/43939 [56:48:03<13:54:37, 5.72s/it]
80%|████████ | 35191/43939 [56:48:08<13:53:39, 5.72s/it]
80%|████████ | 35192/43939 [56:48:14<13:53:01, 5.71s/it]
80%|████████ | 35193/43939 [56:48:20<13:51:37, 5.71s/it]
80%|████████ | 35194/43939 [56:48:25<13:51:19, 5.70s/it]
80%|████████ | 35195/43939 [56:48:31<13:51:16, 5.70s/it]
80%|████████ | 35196/43939 [56:48:37<13:51:32, 5.71s/it]
80%|████████ | 35197/43939 [56:48:43<13:51:45, 5.71s/it]
80%|████████ | 35198/43939 [56:48:48<13:52:31, 5.71s/it]
80%|████████ | 35199/43939 [56:48:54<13:52:02, 5.71s/it]
80%|████████ | 35200/43939 [56:49:00<13:51:46, 5.71s/it]
{'loss': 0.6643, 'grad_norm': 1.5790226459503174, 'learning_rate': 5.787751506387007e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.58, 'epoch': 3.2}
80%|████████ | 35200/43939 [56:49:00<13:51:46, 5.71s/it]
80%|████████ | 35201/43939 [56:49:05<13:51:00, 5.71s/it]
80%|████████ | 35202/43939 [56:49:11<13:50:47, 5.71s/it]
80%|████████ | 35203/43939 [56:49:17<13:50:48, 5.71s/it]
80%|████████ | 35204/43939 [56:49:23<13:53:31, 5.73s/it]
80%|████████ | 35205/43939 [56:49:28<13:52:02, 5.72s/it]
80%|████████ | 35206/43939 [56:49:34<13:51:29, 5.71s/it]
80%|████████ | 35207/43939 [56:49:40<14:03:45, 5.80s/it]
80%|████████ | 35208/43939 [56:49:46<13:58:18, 5.76s/it]
80%|████████ | 35209/43939 [56:49:51<13:55:35, 5.74s/it]
80%|████████ | 35210/43939 [56:49:57<13:54:03, 5.73s/it]
80%|████████ | 35211/43939 [56:50:03<13:51:35, 5.72s/it]
80%|████████ | 35212/43939 [56:50:08<13:51:25, 5.72s/it]
80%|████████ | 35213/43939 [56:50:14<13:50:25, 5.71s/it]
80%|████████ | 35214/43939 [56:50:20<13:50:21, 5.71s/it]
80%|████████ | 35215/43939 [56:50:26<13:50:04, 5.71s/it]
80%|████████ | 35216/43939 [56:50:31<13:49:32, 5.71s/it]
80%|████████ | 35217/43939 [56:50:37<13:49:55, 5.71s/it]
80%|████████ | 35218/43939 [56:50:43<13:49:42, 5.71s/it]
80%|████████ | 35219/43939 [56:50:48<13:50:46, 5.72s/it]
80%|████████ | 35220/43939 [56:50:54<13:48:42, 5.70s/it]
80%|████████ | 35221/43939 [56:51:00<13:49:37, 5.71s/it]
80%|████████ | 35222/43939 [56:51:06<13:51:22, 5.72s/it]
80%|████████ | 35223/43939 [56:51:11<13:50:36, 5.72s/it]
80%|████████ | 35224/43939 [56:51:17<13:48:27, 5.70s/it]
80%|████████ | 35225/43939 [56:51:23<13:48:38, 5.71s/it]
80%|████████ | 35226/43939 [56:51:28<13:49:27, 5.71s/it]
80%|████████ | 35227/43939 [56:51:34<13:48:59, 5.71s/it]
80%|████████ | 35228/43939 [56:51:40<13:59:55, 5.79s/it]
80%|████████ | 35229/43939 [56:51:46<13:58:20, 5.78s/it]
80%|████████ | 35230/43939 [56:51:51<13:56:14, 5.76s/it]
80%|████████ | 35231/43939 [56:51:57<13:53:11, 5.74s/it]
80%|████████ | 35232/43939 [56:52:03<13:50:21, 5.72s/it]
80%|████████ | 35233/43939 [56:52:09<13:50:03, 5.72s/it]
80%|████████ | 35234/43939 [56:52:14<13:49:18, 5.72s/it]
80%|████████ | 35235/43939 [56:52:20<13:47:55, 5.71s/it]
80%|████████ | 35236/43939 [56:52:26<13:49:56, 5.72s/it]
80%|████████ | 35237/43939 [56:52:31<13:48:35, 5.71s/it]
80%|████████ | 35238/43939 [56:52:37<13:47:10, 5.70s/it]
80%|████████ | 35239/43939 [56:52:43<13:46:59, 5.70s/it]
80%|████████ | 35240/43939 [56:52:48<13:46:44, 5.70s/it]
80%|████████ | 35241/43939 [56:52:54<13:45:55, 5.70s/it]
80%|████████ | 35242/43939 [56:53:00<13:44:36, 5.69s/it]
80%|████████ | 35243/43939 [56:53:06<13:44:20, 5.69s/it]
80%|████████ | 35244/43939 [56:53:11<13:44:26, 5.69s/it]
80%|████████ | 35245/43939 [56:53:17<13:43:30, 5.68s/it]
80%|████████ | 35246/43939 [56:53:23<13:42:44, 5.68s/it]
80%|████████ | 35247/43939 [56:53:28<13:44:39, 5.69s/it]
80%|████████ | 35248/43939 [56:53:34<13:46:00, 5.70s/it]
80%|████████ | 35249/43939 [56:53:40<13:55:21, 5.77s/it]
80%|████████ | 35250/43939 [56:53:46<13:51:27, 5.74s/it]
{'loss': 0.6678, 'grad_norm': 1.5815094709396362, 'learning_rate': 5.7243637675451014e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.89, 'epoch': 3.21}
80%|████████ | 35250/43939 [56:53:46<13:51:27, 5.74s/it]
80%|████████ | 35251/43939 [56:53:51<13:48:33, 5.72s/it]
80%|████████ | 35252/43939 [56:53:57<13:49:57, 5.73s/it]
80%|████████ | 35253/43939 [56:54:03<13:48:32, 5.72s/it]
80%|████████ | 35254/43939 [56:54:08<13:47:54, 5.72s/it]
80%|████████ | 35255/43939 [56:54:14<13:47:40, 5.72s/it]
80%|████████ | 35256/43939 [56:54:20<13:48:27, 5.72s/it]
80%|████████ | 35257/43939 [56:54:26<13:47:10, 5.72s/it]
80%|████████ | 35258/43939 [56:54:31<13:46:47, 5.71s/it]
80%|████████ | 35259/43939 [56:54:37<13:46:14, 5.71s/it]
80%|████████ | 35260/43939 [56:54:43<13:45:45, 5.71s/it]
80%|████████ | 35261/43939 [56:54:48<13:44:03, 5.70s/it]
80%|████████ | 35262/43939 [56:54:54<13:46:32, 5.72s/it]
80%|████████ | 35263/43939 [56:55:00<13:46:05, 5.71s/it]
80%|████████ | 35264/43939 [56:55:06<13:44:27, 5.70s/it]
80%|████████ | 35265/43939 [56:55:11<13:43:01, 5.69s/it]
80%|████████ | 35266/43939 [56:55:17<13:42:20, 5.69s/it]
80%|████████ | 35267/43939 [56:55:23<13:43:19, 5.70s/it]
80%|████████ | 35268/43939 [56:55:28<13:42:11, 5.69s/it]
80%|████████ | 35269/43939 [56:55:34<13:44:30, 5.71s/it]
80%|████████ | 35270/43939 [56:55:40<13:43:21, 5.70s/it]
80%|████████ | 35271/43939 [56:55:46<13:54:09, 5.77s/it]
80%|████████ | 35272/43939 [56:55:51<13:51:20, 5.76s/it]
80%|████████ | 35273/43939 [56:55:57<13:48:56, 5.74s/it]
80%|████████ | 35274/43939 [56:56:03<13:47:01, 5.73s/it]
80%|████████ | 35275/43939 [56:56:08<13:46:32, 5.72s/it]
80%|████████ | 35276/43939 [56:56:14<13:44:40, 5.71s/it]
80%|████████ | 35277/43939 [56:56:20<13:45:34, 5.72s/it]
80%|████████ | 35278/43939 [56:56:26<13:44:37, 5.71s/it]
80%|████████ | 35279/43939 [56:56:31<13:44:23, 5.71s/it]
80%|████████ | 35280/43939 [56:56:37<13:48:17, 5.74s/it]
80%|████████ | 35281/43939 [56:56:43<13:46:30, 5.73s/it]
80%|████████ | 35282/43939 [56:56:49<13:46:31, 5.73s/it]
80%|████████ | 35283/43939 [56:56:54<13:43:57, 5.71s/it]
80%|████████ | 35284/43939 [56:57:00<13:43:52, 5.71s/it]
80%|████████ | 35285/43939 [56:57:06<13:44:24, 5.72s/it]
80%|████████ | 35286/43939 [56:57:11<13:43:57, 5.71s/it]
80%|████████ | 35287/43939 [56:57:17<13:43:56, 5.71s/it]
80%|████████ | 35288/43939 [56:57:23<13:42:32, 5.70s/it]
80%|████████ | 35289/43939 [56:57:28<13:41:55, 5.70s/it]
80%|████████ | 35290/43939 [56:57:34<13:43:50, 5.72s/it]
80%|████████ | 35291/43939 [56:57:40<13:41:58, 5.70s/it]
80%|████████ | 35292/43939 [56:57:46<13:41:26, 5.70s/it]
80%|████████ | 35293/43939 [56:57:52<13:51:48, 5.77s/it]
80%|████████ | 35294/43939 [56:57:57<13:48:49, 5.75s/it]
80%|████████ | 35295/43939 [56:58:03<13:47:24, 5.74s/it]
80%|████████ | 35296/43939 [56:58:09<13:45:10, 5.73s/it]
80%|████████ | 35297/43939 [56:58:14<13:45:12, 5.73s/it]
80%|████████ | 35298/43939 [56:58:20<13:43:43, 5.72s/it]
80%|████████ | 35299/43939 [56:58:26<13:43:33, 5.72s/it]
80%|████████ | 35300/43939 [56:58:32<13:44:27, 5.73s/it]
{'loss': 0.6714, 'grad_norm': 1.5472116470336914, 'learning_rate': 5.661280147161888e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.03, 'epoch': 3.21}
80%|████████ | 35300/43939 [56:58:32<13:44:27, 5.73s/it]
80%|████████ | 35301/43939 [56:58:37<13:43:21, 5.72s/it]
80%|████████ | 35302/43939 [56:58:43<13:40:59, 5.70s/it]
80%|████████ | 35303/43939 [56:58:49<13:42:00, 5.71s/it]
80%|████████ | 35304/43939 [56:58:54<13:41:39, 5.71s/it]
80%|████████ | 35305/43939 [56:59:00<13:42:19, 5.71s/it]
80%|████████ | 35306/43939 [56:59:06<13:40:38, 5.70s/it]
80%|████████ | 35307/43939 [56:59:11<13:40:58, 5.71s/it]
80%|████████ | 35308/43939 [56:59:17<13:40:05, 5.70s/it]
80%|████████ | 35309/43939 [56:59:23<13:41:05, 5.71s/it]
80%|████████ | 35310/43939 [56:59:29<13:39:18, 5.70s/it]
80%|████████ | 35311/43939 [56:59:34<13:40:44, 5.71s/it]
80%|████████ | 35312/43939 [56:59:40<13:43:20, 5.73s/it]
80%|████████ | 35313/43939 [56:59:46<13:41:32, 5.71s/it]
80%|████████ | 35314/43939 [56:59:52<13:51:51, 5.79s/it]
80%|████████ | 35315/43939 [56:59:57<13:46:28, 5.75s/it]
80%|████████ | 35316/43939 [57:00:03<13:45:10, 5.74s/it]
80%|████████ | 35317/43939 [57:00:09<13:42:38, 5.72s/it]
80%|████████ | 35318/43939 [57:00:14<13:41:27, 5.72s/it]
80%|████████ | 35319/43939 [57:00:20<13:39:44, 5.71s/it]
80%|████████ | 35320/43939 [57:00:26<13:39:02, 5.70s/it]
80%|████████ | 35321/43939 [57:00:32<13:41:48, 5.72s/it]
80%|████████ | 35322/43939 [57:00:37<13:40:46, 5.72s/it]
80%|████████ | 35323/43939 [57:00:43<13:40:19, 5.71s/it]
80%|████████ | 35324/43939 [57:00:49<13:39:26, 5.71s/it]
80%|████████ | 35325/43939 [57:00:54<13:38:24, 5.70s/it]
80%|████████ | 35326/43939 [57:01:00<13:38:39, 5.70s/it]
80%|████████ | 35327/43939 [57:01:06<13:39:08, 5.71s/it]
80%|████████ | 35328/43939 [57:01:11<13:37:45, 5.70s/it]
80%|████████ | 35329/43939 [57:01:17<13:38:06, 5.70s/it]
80%|████████ | 35330/43939 [57:01:23<13:38:15, 5.70s/it]
80%|████████ | 35331/43939 [57:01:29<13:39:01, 5.71s/it]
80%|████████ | 35332/43939 [57:01:34<13:38:52, 5.71s/it]
80%|████████ | 35333/43939 [57:01:40<13:37:36, 5.70s/it]
80%|████████ | 35334/43939 [57:01:46<13:35:22, 5.69s/it]
80%|████████ | 35335/43939 [57:01:52<13:48:37, 5.78s/it]
80%|████████ | 35336/43939 [57:01:57<13:45:41, 5.76s/it]
80%|████████ | 35337/43939 [57:02:03<13:42:10, 5.73s/it]
80%|████████ | 35338/43939 [57:02:09<13:40:59, 5.73s/it]
80%|████████ | 35339/43939 [57:02:14<13:38:33, 5.71s/it]
80%|████████ | 35340/43939 [57:02:20<13:41:13, 5.73s/it]
80%|████████ | 35341/43939 [57:02:26<13:39:41, 5.72s/it]
80%|████████ | 35342/43939 [57:02:32<13:41:19, 5.73s/it]
80%|████████ | 35343/43939 [57:02:37<13:39:48, 5.72s/it]
80%|████████ | 35344/43939 [57:02:43<13:38:30, 5.71s/it]
80%|████████ | 35345/43939 [57:02:49<13:36:08, 5.70s/it]
80%|████████ | 35346/43939 [57:02:54<13:35:32, 5.69s/it]
80%|████████ | 35347/43939 [57:03:00<13:35:40, 5.70s/it]
80%|████████ | 35348/43939 [57:03:06<13:34:35, 5.69s/it]
80%|████████ | 35349/43939 [57:03:11<13:35:00, 5.69s/it]
80%|████████ | 35350/43939 [57:03:17<13:35:05, 5.69s/it]
{'loss': 0.6473, 'grad_norm': 1.6719571352005005, 'learning_rate': 5.598501640529713e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.74, 'epoch': 3.22}
80%|████████ | 35350/43939 [57:03:17<13:35:05, 5.69s/it]
80%|████████ | 35351/43939 [57:03:23<13:37:42, 5.71s/it]
80%|████████ | 35352/43939 [57:03:29<13:37:54, 5.72s/it]
80%|████████ | 35353/43939 [57:03:34<13:36:18, 5.70s/it]
80%|████████ | 35354/43939 [57:03:40<13:36:34, 5.71s/it]
80%|████████ | 35355/43939 [57:03:46<13:35:56, 5.70s/it]
80%|████████ | 35356/43939 [57:03:51<13:38:38, 5.72s/it]
80%|████████ | 35357/43939 [57:03:57<13:49:04, 5.80s/it]
80%|████████ | 35358/43939 [57:04:03<13:42:55, 5.75s/it]
80%|████████ | 35359/43939 [57:04:09<13:40:51, 5.74s/it]
80%|████████ | 35360/43939 [57:04:15<13:39:02, 5.73s/it]
80%|████████ | 35361/43939 [57:04:20<13:36:07, 5.71s/it]
80%|████████ | 35362/43939 [57:04:26<13:35:07, 5.70s/it]
80%|████████ | 35363/43939 [57:04:32<13:34:46, 5.70s/it]
80%|████████ | 35364/43939 [57:04:37<13:35:03, 5.70s/it]
80%|████████ | 35365/43939 [57:04:43<13:34:41, 5.70s/it]
80%|████████ | 35366/43939 [57:04:49<13:37:28, 5.72s/it]
80%|████████ | 35367/43939 [57:04:54<13:35:32, 5.71s/it]
80%|████████ | 35368/43939 [57:05:00<13:35:54, 5.71s/it]
80%|████████ | 35369/43939 [57:05:06<13:33:57, 5.70s/it]
80%|████████ | 35370/43939 [57:05:12<13:34:54, 5.71s/it]
81%|████████ | 35371/43939 [57:05:17<13:34:45, 5.71s/it]
81%|████████ | 35372/43939 [57:05:23<13:32:10, 5.69s/it]
81%|████████ | 35373/43939 [57:05:29<13:31:56, 5.69s/it]
81%|████████ | 35374/43939 [57:05:34<13:31:46, 5.69s/it]
81%|████████ | 35375/43939 [57:05:40<13:35:32, 5.71s/it]
81%|████████ | 35376/43939 [57:05:46<13:33:44, 5.70s/it]
81%|████████ | 35377/43939 [57:05:51<13:34:41, 5.71s/it]
81%|████████ | 35378/43939 [57:05:57<13:32:22, 5.69s/it]
81%|████████ | 35379/43939 [57:06:03<13:42:41, 5.77s/it]
81%|████████ | 35380/43939 [57:06:09<13:39:27, 5.74s/it]
81%|████████ | 35381/43939 [57:06:14<13:37:18, 5.73s/it]
81%|████████ | 35382/43939 [57:06:20<13:35:58, 5.72s/it]
81%|████████ | 35383/43939 [57:06:26<13:34:35, 5.71s/it]
81%|████████ | 35384/43939 [57:06:32<13:34:54, 5.72s/it]
81%|████████ | 35385/43939 [57:06:37<13:35:12, 5.72s/it]
81%|████████ | 35386/43939 [57:06:43<13:34:01, 5.71s/it]
81%|████████ | 35387/43939 [57:06:49<13:34:50, 5.72s/it]
81%|████████ | 35388/43939 [57:06:54<13:34:49, 5.72s/it]
81%|████████ | 35389/43939 [57:07:00<13:35:13, 5.72s/it]
81%|████████ | 35390/43939 [57:07:06<13:34:03, 5.71s/it]
81%|████████ | 35391/43939 [57:07:11<13:32:10, 5.70s/it]
81%|████████ | 35392/43939 [57:07:17<13:30:48, 5.69s/it]
81%|████████ | 35393/43939 [57:07:23<13:30:53, 5.69s/it]
81%|████████ | 35394/43939 [57:07:29<13:31:41, 5.70s/it]
81%|████████ | 35395/43939 [57:07:34<13:32:13, 5.70s/it]
81%|████████ | 35396/43939 [57:07:40<13:32:17, 5.71s/it]
81%|████████ | 35397/43939 [57:07:46<13:32:15, 5.71s/it]
81%|████████ | 35398/43939 [57:07:51<13:31:21, 5.70s/it]
81%|████████ | 35399/43939 [57:07:57<13:33:24, 5.71s/it]
81%|████████ | 35400/43939 [57:08:03<13:41:59, 5.78s/it]
{'loss': 0.6653, 'grad_norm': 1.6827168464660645, 'learning_rate': 5.5360292381270604e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.9, 'epoch': 3.22}
81%|████████ | 35400/43939 [57:08:03<13:41:59, 5.78s/it]
81%|████████ | 35401/43939 [57:08:09<13:37:37, 5.75s/it]
81%|████████ | 35402/43939 [57:08:14<13:33:08, 5.71s/it]
81%|████████ | 35403/43939 [57:08:20<13:34:43, 5.73s/it]
81%|████████ | 35404/43939 [57:08:26<13:33:26, 5.72s/it]
81%|████████ | 35405/43939 [57:08:32<13:31:17, 5.70s/it]
81%|████████ | 35406/43939 [57:08:37<13:30:01, 5.70s/it]
81%|████████ | 35407/43939 [57:08:43<13:30:20, 5.70s/it]
81%|████████ | 35408/43939 [57:08:49<13:29:55, 5.70s/it]
81%|████████ | 35409/43939 [57:08:54<13:29:53, 5.70s/it]
81%|████████ | 35410/43939 [57:09:00<13:29:17, 5.69s/it]
81%|████████ | 35411/43939 [57:09:06<13:29:13, 5.69s/it]
81%|████████ | 35412/43939 [57:09:11<13:28:31, 5.69s/it]
81%|████████ | 35413/43939 [57:09:17<13:28:39, 5.69s/it]
81%|████████ | 35414/43939 [57:09:23<13:28:38, 5.69s/it]
81%|████████ | 35415/43939 [57:09:28<13:32:10, 5.72s/it]
81%|████████ | 35416/43939 [57:09:34<13:31:38, 5.71s/it]
81%|████████ | 35417/43939 [57:09:40<13:30:03, 5.70s/it]
81%|████████ | 35418/43939 [57:09:46<13:30:25, 5.71s/it]
81%|████████ | 35419/43939 [57:09:51<13:31:23, 5.71s/it]
81%|████████ | 35420/43939 [57:09:57<13:32:16, 5.72s/it]
81%|████████ | 35421/43939 [57:10:03<13:30:40, 5.71s/it]
81%|████████ | 35422/43939 [57:10:09<13:37:58, 5.76s/it]
81%|████████ | 35423/43939 [57:10:14<13:34:52, 5.74s/it]
81%|████████ | 35424/43939 [57:10:20<13:34:50, 5.74s/it]
81%|████████ | 35425/43939 [57:10:26<13:32:23, 5.73s/it]
81%|████████ | 35426/43939 [57:10:31<13:32:08, 5.72s/it]
81%|████████ | 35427/43939 [57:10:37<13:34:12, 5.74s/it]
81%|████████ | 35428/43939 [57:10:43<13:31:38, 5.72s/it]
81%|████████ | 35429/43939 [57:10:49<13:29:47, 5.71s/it]
81%|████████ | 35430/43939 [57:10:54<13:28:37, 5.70s/it]
81%|████████ | 35431/43939 [57:11:00<13:26:52, 5.69s/it]
81%|████████ | 35432/43939 [57:11:06<13:26:08, 5.69s/it]
81%|████████ | 35433/43939 [57:11:11<13:26:32, 5.69s/it]
81%|████████ | 35434/43939 [57:11:17<13:26:14, 5.69s/it]
81%|████████ | 35435/43939 [57:11:23<13:27:15, 5.70s/it]
81%|████████ | 35436/43939 [57:11:28<13:25:24, 5.68s/it]
81%|████████ | 35437/43939 [57:11:34<13:25:54, 5.69s/it]
81%|████████ | 35438/43939 [57:11:40<13:26:40, 5.69s/it]
81%|████████ | 35439/43939 [57:11:45<13:26:28, 5.69s/it]
81%|████████ | 35440/43939 [57:11:51<13:29:35, 5.72s/it]
81%|████████ | 35441/43939 [57:11:57<13:29:01, 5.71s/it]
81%|████████ | 35442/43939 [57:12:03<13:28:45, 5.71s/it]
81%|████████ | 35443/43939 [57:12:09<13:37:21, 5.77s/it]
81%|████████ | 35444/43939 [57:12:14<13:34:17, 5.75s/it]
81%|████████ | 35445/43939 [57:12:20<13:30:11, 5.72s/it]
81%|████████ | 35446/43939 [57:12:26<13:28:44, 5.71s/it]
81%|████████ | 35447/43939 [57:12:31<13:26:29, 5.70s/it]
81%|████████ | 35448/43939 [57:12:37<13:25:27, 5.69s/it]
81%|████████ | 35449/43939 [57:12:43<13:26:04, 5.70s/it]
81%|████████ | 35450/43939 [57:12:48<13:25:30, 5.69s/it]
{'loss': 0.6473, 'grad_norm': 1.5358210802078247, 'learning_rate': 5.473863925602881e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.42, 'epoch': 3.23}
81%|████████ | 35450/43939 [57:12:48<13:25:30, 5.69s/it]
81%|████████ | 35451/43939 [57:12:54<13:26:23, 5.70s/it]
81%|████████ | 35452/43939 [57:13:00<13:25:44, 5.70s/it]
81%|████████ | 35453/43939 [57:13:06<13:28:05, 5.71s/it]
81%|████████ | 35454/43939 [57:13:11<13:27:50, 5.71s/it]
81%|████████ | 35455/43939 [57:13:17<13:28:35, 5.72s/it]
81%|████████ | 35456/43939 [57:13:23<13:27:15, 5.71s/it]
81%|████████ | 35457/43939 [57:13:28<13:25:47, 5.70s/it]
81%|████████ | 35458/43939 [57:13:34<13:26:29, 5.71s/it]
81%|████████ | 35459/43939 [57:13:40<13:26:58, 5.71s/it]
81%|████████ | 35460/43939 [57:13:46<13:27:56, 5.72s/it]
81%|████████ | 35461/43939 [57:13:51<13:26:00, 5.70s/it]
81%|████████ | 35462/43939 [57:13:57<13:27:25, 5.71s/it]
81%|████████ | 35463/43939 [57:14:03<13:27:42, 5.72s/it]
81%|████████ | 35464/43939 [57:14:08<13:27:04, 5.71s/it]
81%|████████ | 35465/43939 [57:14:14<13:36:24, 5.78s/it]
81%|████████ | 35466/43939 [57:14:20<13:34:38, 5.77s/it]
81%|████████ | 35467/43939 [57:14:26<13:30:40, 5.74s/it]
81%|████████ | 35468/43939 [57:14:31<13:29:29, 5.73s/it]
81%|████████ | 35469/43939 [57:14:37<13:28:22, 5.73s/it]
81%|████████ | 35470/43939 [57:14:43<13:31:24, 5.75s/it]
81%|████████ | 35471/43939 [57:14:49<13:29:01, 5.73s/it]
81%|████████ | 35472/43939 [57:14:54<13:27:44, 5.72s/it]
81%|████████ | 35473/43939 [57:15:00<13:28:59, 5.73s/it]
81%|████████ | 35474/43939 [57:15:06<13:26:44, 5.72s/it]
81%|████████ | 35475/43939 [57:15:11<13:24:49, 5.71s/it]
81%|████████ | 35476/43939 [57:15:17<13:24:56, 5.71s/it]
81%|████████ | 35477/43939 [57:15:23<13:25:04, 5.71s/it]
81%|████████ | 35478/43939 [57:15:29<13:24:50, 5.71s/it]
81%|████████ | 35479/43939 [57:15:34<13:25:02, 5.71s/it]
81%|████████ | 35480/43939 [57:15:40<13:25:06, 5.71s/it]
81%|████████ | 35481/43939 [57:15:46<13:23:21, 5.70s/it]
81%|████████ | 35482/43939 [57:15:51<13:22:11, 5.69s/it]
81%|████████ | 35483/43939 [57:15:57<13:21:31, 5.69s/it]
81%|████████ | 35484/43939 [57:16:03<13:21:15, 5.69s/it]
81%|████████ | 35485/43939 [57:16:08<13:22:34, 5.70s/it]
81%|████████ | 35486/43939 [57:16:14<13:32:16, 5.77s/it]
81%|████████ | 35487/43939 [57:16:20<13:30:16, 5.75s/it]
81%|████████ | 35488/43939 [57:16:26<13:29:37, 5.75s/it]
81%|████████ | 35489/43939 [57:16:32<13:28:00, 5.74s/it]
81%|████████ | 35490/43939 [57:16:37<13:24:54, 5.72s/it]
81%|████████ | 35491/43939 [57:16:43<13:23:23, 5.71s/it]
81%|████████ | 35492/43939 [57:16:49<13:25:33, 5.72s/it]
81%|████████ | 35493/43939 [57:16:54<13:25:13, 5.72s/it]
81%|████████ | 35494/43939 [57:17:00<13:22:56, 5.70s/it]
81%|████████ | 35495/43939 [57:17:06<13:22:44, 5.70s/it]
81%|████████ | 35496/43939 [57:17:11<13:23:04, 5.71s/it]
81%|████████ | 35497/43939 [57:17:17<13:23:15, 5.71s/it]
81%|████████ | 35498/43939 [57:17:23<13:21:46, 5.70s/it]
81%|████████ | 35499/43939 [57:17:29<13:22:58, 5.71s/it]
81%|████████ | 35500/43939 [57:17:34<13:23:04, 5.71s/it]
{'loss': 0.6668, 'grad_norm': 1.635041356086731, 'learning_rate': 5.412006683761065e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 509.06, 'epoch': 3.23}
81%|████████ | 35500/43939 [57:17:34<13:23:04, 5.71s/it]
81%|████████ | 35501/43939 [57:17:40<13:21:33, 5.70s/it]
81%|████████ | 35502/43939 [57:17:46<13:21:24, 5.70s/it]
81%|████████ | 35503/43939 [57:17:51<13:20:12, 5.69s/it]
81%|████████ | 35504/43939 [57:17:57<13:21:00, 5.70s/it]
81%|████████ | 35505/43939 [57:18:03<13:20:35, 5.70s/it]
81%|████████ | 35506/43939 [57:18:08<13:20:53, 5.70s/it]
81%|████████ | 35507/43939 [57:18:14<13:32:08, 5.78s/it]
81%|████████ | 35508/43939 [57:18:20<13:28:44, 5.76s/it]
81%|████████ | 35509/43939 [57:18:26<13:28:58, 5.76s/it]
81%|████████ | 35510/43939 [57:18:32<13:28:02, 5.75s/it]
81%|████████ | 35511/43939 [57:18:37<13:26:27, 5.74s/it]
81%|████████ | 35512/43939 [57:18:43<13:24:46, 5.73s/it]
81%|████████ | 35513/43939 [57:18:49<13:22:11, 5.71s/it]
81%|████████ | 35514/43939 [57:18:54<13:23:53, 5.73s/it]
81%|████████ | 35515/43939 [57:19:00<13:23:18, 5.72s/it]
81%|████████ | 35516/43939 [57:19:06<13:21:21, 5.71s/it]
81%|████████ | 35517/43939 [57:19:12<13:22:07, 5.71s/it]
81%|████████ | 35518/43939 [57:19:17<13:22:47, 5.72s/it]
81%|████████ | 35519/43939 [57:19:23<13:21:22, 5.71s/it]
81%|████████ | 35520/43939 [57:19:29<13:21:29, 5.71s/it]
81%|████████ | 35521/43939 [57:19:34<13:22:34, 5.72s/it]
81%|████████ | 35522/43939 [57:19:40<13:21:43, 5.72s/it]
81%|████████ | 35523/43939 [57:19:46<13:21:51, 5.72s/it]
81%|████████ | 35524/43939 [57:19:52<13:23:30, 5.73s/it]
81%|████████ | 35525/43939 [57:19:57<13:22:23, 5.72s/it]
81%|████████ | 35526/43939 [57:20:03<13:21:30, 5.72s/it]
81%|████████ | 35527/43939 [57:20:09<13:19:45, 5.70s/it]
81%|████████ | 35528/43939 [57:20:14<13:21:20, 5.72s/it]
81%|████████ | 35529/43939 [57:20:20<13:29:13, 5.77s/it]
81%|████████ | 35530/43939 [57:20:26<13:25:06, 5.74s/it]
81%|████████ | 35531/43939 [57:20:32<13:23:23, 5.73s/it]
81%|████████ | 35532/43939 [57:20:37<13:21:57, 5.72s/it]
81%|████████ | 35533/43939 [57:20:43<13:22:49, 5.73s/it]
81%|████████ | 35534/43939 [57:20:49<13:21:56, 5.72s/it]
81%|████████ | 35535/43939 [57:20:55<13:19:45, 5.71s/it]
81%|████████ | 35536/43939 [57:21:00<13:21:47, 5.73s/it]
81%|████████ | 35537/43939 [57:21:06<13:22:15, 5.73s/it]
81%|████████ | 35538/43939 [57:21:12<13:21:19, 5.72s/it]
81%|████████ | 35539/43939 [57:21:17<13:19:51, 5.71s/it]
81%|████████ | 35540/43939 [57:21:23<13:18:36, 5.71s/it]
81%|████████ | 35541/43939 [57:21:29<13:17:28, 5.70s/it]
81%|████████ | 35542/43939 [57:21:35<13:17:15, 5.70s/it]
81%|████████ | 35543/43939 [57:21:40<13:16:58, 5.70s/it]
81%|████████ | 35544/43939 [57:21:46<13:15:32, 5.69s/it]
81%|████████ | 35545/43939 [57:21:52<13:16:46, 5.70s/it]
81%|████████ | 35546/43939 [57:21:57<13:18:08, 5.71s/it]
81%|████████ | 35547/43939 [57:22:03<13:18:35, 5.71s/it]
81%|████████ | 35548/43939 [57:22:09<13:18:06, 5.71s/it]
81%|████████ | 35549/43939 [57:22:14<13:17:39, 5.70s/it]
81%|████████ | 35550/43939 [57:22:20<13:19:23, 5.72s/it]
{'loss': 0.6642, 'grad_norm': 1.571561574935913, 'learning_rate': 5.350458488544966e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.38, 'epoch': 3.24}
81%|████████ | 35550/43939 [57:22:20<13:19:23, 5.72s/it]
81%|████████ | 35551/43939 [57:22:26<13:26:52, 5.77s/it]
81%|████████ | 35552/43939 [57:22:32<13:24:35, 5.76s/it]
81%|████████ | 35553/43939 [57:22:38<13:22:10, 5.74s/it]
81%|████████ | 35554/43939 [57:22:43<13:20:58, 5.73s/it]
81%|████████ | 35555/43939 [57:22:49<13:21:56, 5.74s/it]
81%|████████ | 35556/43939 [57:22:55<13:19:17, 5.72s/it]
81%|████████ | 35557/43939 [57:23:00<13:17:24, 5.71s/it]
81%|████████ | 35558/43939 [57:23:06<13:16:23, 5.70s/it]
81%|████████ | 35559/43939 [57:23:12<13:16:03, 5.70s/it]
81%|████████ | 35560/43939 [57:23:17<13:15:26, 5.70s/it]
81%|████████ | 35561/43939 [57:23:23<13:15:24, 5.70s/it]
81%|████████ | 35562/43939 [57:23:29<13:15:28, 5.70s/it]
81%|████████ | 35563/43939 [57:23:34<13:13:34, 5.68s/it]
81%|████████ | 35564/43939 [57:23:40<13:12:43, 5.68s/it]
81%|████████ | 35565/43939 [57:23:46<13:12:13, 5.68s/it]
81%|████████ | 35566/43939 [57:23:52<13:15:24, 5.70s/it]
81%|████████ | 35567/43939 [57:23:57<13:16:05, 5.71s/it]
81%|████████ | 35568/43939 [57:24:03<13:15:24, 5.70s/it]
81%|████████ | 35569/43939 [57:24:09<13:15:17, 5.70s/it]
81%|████████ | 35570/43939 [57:24:14<13:16:46, 5.71s/it]
81%|████████ | 35571/43939 [57:24:20<13:25:22, 5.77s/it]
81%|████████ | 35572/43939 [57:24:26<13:20:45, 5.74s/it]
81%|████████ | 35573/43939 [57:24:32<13:19:27, 5.73s/it]
81%|████████ | 35574/43939 [57:24:37<13:18:12, 5.73s/it]
81%|████████ | 35575/43939 [57:24:43<13:16:42, 5.72s/it]
81%|████████ | 35576/43939 [57:24:49<13:18:18, 5.73s/it]
81%|████████ | 35577/43939 [57:24:55<13:17:10, 5.72s/it]
81%|████████ | 35578/43939 [57:25:00<13:16:09, 5.71s/it]
81%|████████ | 35579/43939 [57:25:06<13:14:54, 5.71s/it]
81%|████████ | 35580/43939 [57:25:12<13:13:48, 5.70s/it]
81%|████████ | 35581/43939 [57:25:17<13:13:31, 5.70s/it]
81%|████████ | 35582/43939 [57:25:23<13:12:42, 5.69s/it]
81%|████████ | 35583/43939 [57:25:29<13:15:16, 5.71s/it]
81%|████████ | 35584/43939 [57:25:34<13:13:55, 5.70s/it]
81%|████████ | 35585/43939 [57:25:40<13:13:42, 5.70s/it]
81%|████████ | 35586/43939 [57:25:46<13:14:11, 5.70s/it]
81%|████████ | 35587/43939 [57:25:51<13:11:39, 5.69s/it]
81%|████████ | 35588/43939 [57:25:57<13:13:51, 5.70s/it]
81%|████████ | 35589/43939 [57:26:03<13:14:57, 5.71s/it]
81%|████████ | 35590/43939 [57:26:09<13:16:29, 5.72s/it]
81%|████████ | 35591/43939 [57:26:14<13:13:59, 5.71s/it]
81%|████████ | 35592/43939 [57:26:20<13:13:38, 5.70s/it]
81%|████████ | 35593/43939 [57:26:26<13:23:44, 5.78s/it]
81%|████████ | 35594/43939 [57:26:32<13:20:56, 5.76s/it]
81%|████████ | 35595/43939 [57:26:37<13:19:34, 5.75s/it]
81%|████████ | 35596/43939 [57:26:43<13:17:50, 5.74s/it]
81%|████████ | 35597/43939 [57:26:49<13:18:22, 5.74s/it]
81%|████████ | 35598/43939 [57:26:55<13:15:53, 5.73s/it]
81%|████████ | 35599/43939 [57:27:00<13:13:54, 5.71s/it]
81%|████████ | 35600/43939 [57:27:06<13:12:46, 5.70s/it]
{'loss': 0.6645, 'grad_norm': 1.8174020051956177, 'learning_rate': 5.289220311021997e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.42, 'epoch': 3.24}
81%|████████ | 35600/43939 [57:27:06<13:12:46, 5.70s/it]
81%|████████ | 35601/43939 [57:27:12<13:12:39, 5.70s/it]
81%|████████ | 35602/43939 [57:27:17<13:13:51, 5.71s/it]
81%|████████ | 35603/43939 [57:27:23<13:12:58, 5.71s/it]
81%|████████ | 35604/43939 [57:27:29<13:12:12, 5.70s/it]
81%|████████ | 35605/43939 [57:27:35<13:13:32, 5.71s/it]
81%|████████ | 35606/43939 [57:27:40<13:12:59, 5.71s/it]
81%|████████ | 35607/43939 [57:27:46<13:12:22, 5.71s/it]
81%|████████ | 35608/43939 [57:27:52<13:12:25, 5.71s/it]
81%|████████ | 35609/43939 [57:27:57<13:15:03, 5.73s/it]
81%|████████ | 35610/43939 [57:28:03<13:13:33, 5.72s/it]
81%|████████ | 35611/43939 [57:28:09<13:15:31, 5.73s/it]
81%|████████ | 35612/43939 [57:28:15<13:15:29, 5.73s/it]
81%|████████ | 35613/43939 [57:28:20<13:15:52, 5.74s/it]
81%|████████ | 35614/43939 [57:28:26<13:26:03, 5.81s/it]
81%|████████ | 35615/43939 [57:28:32<13:21:46, 5.78s/it]
81%|████████ | 35616/43939 [57:28:38<13:18:33, 5.76s/it]
81%|████████ | 35617/43939 [57:28:44<13:19:03, 5.76s/it]
81%|████████ | 35618/43939 [57:28:49<13:18:46, 5.76s/it]
81%|████████ | 35619/43939 [57:28:55<13:17:27, 5.75s/it]
81%|████████ | 35620/43939 [57:29:01<13:15:16, 5.74s/it]
81%|████████ | 35621/43939 [57:29:06<13:14:38, 5.73s/it]
81%|████████ | 35622/43939 [57:29:12<13:12:48, 5.72s/it]
81%|████████ | 35623/43939 [57:29:18<13:12:47, 5.72s/it]
81%|████████ | 35624/43939 [57:29:24<13:12:18, 5.72s/it]
81%|████████ | 35625/43939 [57:29:29<13:11:56, 5.72s/it]
81%|████████ | 35626/43939 [57:29:35<13:11:07, 5.71s/it]
81%|████████ | 35627/43939 [57:29:41<13:12:29, 5.72s/it]
81%|████████ | 35628/43939 [57:29:46<13:11:38, 5.72s/it]
81%|████████ | 35629/43939 [57:29:52<13:09:29, 5.70s/it]
81%|████████ | 35630/43939 [57:29:58<13:10:51, 5.71s/it]
81%|████████ | 35631/43939 [57:30:03<13:09:13, 5.70s/it]
81%|████████ | 35632/43939 [57:30:09<13:11:31, 5.72s/it]
81%|████████ | 35633/43939 [57:30:15<13:09:29, 5.70s/it]
81%|████████ | 35634/43939 [57:30:21<13:10:44, 5.71s/it]
81%|████████ | 35635/43939 [57:30:27<13:20:29, 5.78s/it]
81%|████████ | 35636/43939 [57:30:32<13:17:13, 5.76s/it]
81%|████████ | 35637/43939 [57:30:38<13:15:00, 5.75s/it]
81%|████████ | 35638/43939 [57:30:44<13:13:54, 5.74s/it]
81%|████████ | 35639/43939 [57:30:49<13:12:34, 5.73s/it]
81%|████████ | 35640/43939 [57:30:55<13:10:13, 5.71s/it]
81%|████████ | 35641/43939 [57:31:01<13:09:47, 5.71s/it]
81%|████████ | 35642/43939 [57:31:07<13:09:37, 5.71s/it]
81%|████████ | 35643/43939 [57:31:12<13:08:20, 5.70s/it]
81%|████████ | 35644/43939 [57:31:18<13:07:56, 5.70s/it]
81%|████████ | 35645/43939 [57:31:24<13:08:13, 5.70s/it]
81%|████████ | 35646/43939 [57:31:29<13:08:20, 5.70s/it]
81%|████████ | 35647/43939 [57:31:35<13:08:01, 5.70s/it]
81%|████████ | 35648/43939 [57:31:41<13:09:26, 5.71s/it]
81%|████████ | 35649/43939 [57:31:46<13:08:10, 5.70s/it]
81%|████████ | 35650/43939 [57:31:52<13:09:04, 5.71s/it]
{'loss': 0.6583, 'grad_norm': 1.6712254285812378, 'learning_rate': 5.228293117368319e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.9, 'epoch': 3.25}
81%|████████ | 35650/43939 [57:31:52<13:09:04, 5.71s/it]
81%|████████ | 35651/43939 [57:31:58<13:07:26, 5.70s/it]
81%|████████ | 35652/43939 [57:32:04<13:07:41, 5.70s/it]
81%|████████ | 35653/43939 [57:32:09<13:05:31, 5.69s/it]
81%|████████ | 35654/43939 [57:32:15<13:05:08, 5.69s/it]
81%|████████ | 35655/43939 [57:32:21<13:05:50, 5.69s/it]
81%|████████ | 35656/43939 [57:32:27<13:16:06, 5.77s/it]
81%|████████ | 35657/43939 [57:32:32<13:13:46, 5.75s/it]
81%|████████ | 35658/43939 [57:32:38<13:10:38, 5.73s/it]
81%|████████ | 35659/43939 [57:32:44<13:11:07, 5.73s/it]
81%|████████ | 35660/43939 [57:32:49<13:10:07, 5.73s/it]
81%|████████ | 35661/43939 [57:32:55<13:08:30, 5.72s/it]
81%|████████ | 35662/43939 [57:33:01<13:09:13, 5.72s/it]
81%|████████ | 35663/43939 [57:33:06<13:06:00, 5.70s/it]
81%|████████ | 35664/43939 [57:33:12<13:04:42, 5.69s/it]
81%|████████ | 35665/43939 [57:33:18<13:04:58, 5.69s/it]
81%|████████ | 35666/43939 [57:33:24<13:05:06, 5.69s/it]
81%|████████ | 35667/43939 [57:33:29<13:04:52, 5.69s/it]
81%|████████ | 35668/43939 [57:33:35<13:03:48, 5.69s/it]
81%|████████ | 35669/43939 [57:33:41<13:02:31, 5.68s/it]
81%|████████ | 35670/43939 [57:33:46<13:03:50, 5.69s/it]
81%|████████ | 35671/43939 [57:33:52<13:04:11, 5.69s/it]
81%|████████ | 35672/43939 [57:33:58<13:03:35, 5.69s/it]
81%|████████ | 35673/43939 [57:34:03<13:04:36, 5.70s/it]
81%|████████ | 35674/43939 [57:34:09<13:06:06, 5.71s/it]
81%|████████ | 35675/43939 [57:34:15<13:04:33, 5.70s/it]
81%|████████ | 35676/43939 [57:34:20<13:04:40, 5.70s/it]
81%|████████ | 35677/43939 [57:34:26<13:04:42, 5.70s/it]
81%|████████ | 35678/43939 [57:34:32<13:12:00, 5.75s/it]
81%|████████ | 35679/43939 [57:34:38<13:10:59, 5.75s/it]
81%|████████ | 35680/43939 [57:34:43<13:09:11, 5.73s/it]
81%|████████ | 35681/43939 [57:34:49<13:08:39, 5.73s/it]
81%|████████ | 35682/43939 [57:34:55<13:08:25, 5.73s/it]
81%|████████ | 35683/43939 [57:35:01<13:07:02, 5.72s/it]
81%|████████ | 35684/43939 [57:35:06<13:06:25, 5.72s/it]
81%|████████ | 35685/43939 [57:35:12<13:05:58, 5.71s/it]
81%|████████ | 35686/43939 [57:35:18<13:05:30, 5.71s/it]
81%|████████ | 35687/43939 [57:35:23<13:04:59, 5.71s/it]
81%|████████ | 35688/43939 [57:35:29<13:05:08, 5.71s/it]
81%|████████ | 35689/43939 [57:35:35<13:03:14, 5.70s/it]
81%|████████ | 35690/43939 [57:35:41<13:04:34, 5.71s/it]
81%|████████ | 35691/43939 [57:35:46<13:04:17, 5.71s/it]
81%|████████ | 35692/43939 [57:35:52<13:04:02, 5.70s/it]
81%|████████ | 35693/43939 [57:35:58<13:05:31, 5.72s/it]
81%|████████ | 35694/43939 [57:36:03<13:05:54, 5.72s/it]
81%|████████ | 35695/43939 [57:36:09<13:04:04, 5.71s/it]
81%|████████ | 35696/43939 [57:36:15<13:04:10, 5.71s/it]
81%|████████ | 35697/43939 [57:36:21<13:03:52, 5.71s/it]
81%|████████ | 35698/43939 [57:36:26<13:03:44, 5.71s/it]
81%|████████ | 35699/43939 [57:36:32<13:05:10, 5.72s/it]
81%|████████ | 35700/43939 [57:36:38<13:13:34, 5.78s/it]
{'loss': 0.6456, 'grad_norm': 1.5725531578063965, 'learning_rate': 5.1676778688535876e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 448.45, 'epoch': 3.25}
81%|████████ | 35700/43939 [57:36:38<13:13:34, 5.78s/it]
81%|████████▏ | 35701/43939 [57:36:44<13:10:32, 5.76s/it]
81%|████████▏ | 35702/43939 [57:36:49<13:06:20, 5.73s/it]
81%|████████▏ | 35703/43939 [57:36:55<13:05:28, 5.72s/it]
81%|████████▏ | 35704/43939 [57:37:01<13:03:32, 5.71s/it]
81%|████████▏ | 35705/43939 [57:37:06<13:02:49, 5.70s/it]
81%|████████▏ | 35706/43939 [57:37:12<13:04:04, 5.71s/it]
81%|████████▏ | 35707/43939 [57:37:18<13:03:44, 5.71s/it]
81%|████████▏ | 35708/43939 [57:37:23<13:03:13, 5.71s/it]
81%|████████▏ | 35709/43939 [57:37:29<13:06:11, 5.73s/it]
81%|████████▏ | 35710/43939 [57:37:35<13:04:36, 5.72s/it]
81%|████████▏ | 35711/43939 [57:37:41<13:04:36, 5.72s/it]
81%|████████▏ | 35712/43939 [57:37:46<13:02:08, 5.70s/it]
81%|████████▏ | 35713/43939 [57:37:52<13:00:56, 5.70s/it]
81%|████████▏ | 35714/43939 [57:37:58<13:00:38, 5.69s/it]
81%|████████▏ | 35715/43939 [57:38:03<13:02:26, 5.71s/it]
81%|████████▏ | 35716/43939 [57:38:09<13:02:43, 5.71s/it]
81%|████████▏ | 35717/43939 [57:38:15<13:02:12, 5.71s/it]
81%|████████▏ | 35718/43939 [57:38:21<13:02:16, 5.71s/it]
81%|████████▏ | 35719/43939 [57:38:26<13:02:06, 5.71s/it]
81%|████████▏ | 35720/43939 [57:38:32<13:02:06, 5.71s/it]
81%|████████▏ | 35721/43939 [57:38:38<13:09:39, 5.77s/it]
81%|████████▏ | 35722/43939 [57:38:44<13:07:03, 5.75s/it]
81%|████████▏ | 35723/43939 [57:38:49<13:04:51, 5.73s/it]
81%|████████▏ | 35724/43939 [57:38:55<13:03:06, 5.72s/it]
81%|████████▏ | 35725/43939 [57:39:01<13:03:45, 5.73s/it]
81%|████████▏ | 35726/43939 [57:39:06<13:01:47, 5.71s/it]
81%|████████▏ | 35727/43939 [57:39:12<13:01:02, 5.71s/it]
81%|████████▏ | 35728/43939 [57:39:18<13:01:16, 5.71s/it]
81%|████████▏ | 35729/43939 [57:39:24<13:02:07, 5.72s/it]
81%|████████▏ | 35730/43939 [57:39:29<13:00:31, 5.70s/it]
81%|████████▏ | 35731/43939 [57:39:35<13:02:14, 5.72s/it]
81%|████████▏ | 35732/43939 [57:39:41<13:01:12, 5.71s/it]
81%|████████▏ | 35733/43939 [57:39:46<13:01:07, 5.71s/it]
81%|████████▏ | 35734/43939 [57:39:52<12:59:11, 5.70s/it]
81%|████████▏ | 35735/43939 [57:39:58<12:59:12, 5.70s/it]
81%|████████▏ | 35736/43939 [57:40:03<13:00:19, 5.71s/it]
81%|████████▏ | 35737/43939 [57:40:09<12:59:31, 5.70s/it]
81%|████████▏ | 35738/43939 [57:40:15<12:58:24, 5.69s/it]
81%|████████▏ | 35739/43939 [57:40:21<12:58:23, 5.70s/it]
81%|████████▏ | 35740/43939 [57:40:26<12:59:49, 5.71s/it]
81%|████████▏ | 35741/43939 [57:40:32<12:58:31, 5.70s/it]
81%|████████▏ | 35742/43939 [57:40:38<12:59:52, 5.71s/it]
81%|████████▏ | 35743/43939 [57:40:44<13:09:10, 5.78s/it]
81%|████████▏ | 35744/43939 [57:40:49<13:07:46, 5.77s/it]
81%|████████▏ | 35745/43939 [57:40:55<13:06:38, 5.76s/it]
81%|████████▏ | 35746/43939 [57:41:01<13:05:11, 5.75s/it]
81%|████████▏ | 35747/43939 [57:41:07<13:03:11, 5.74s/it]
81%|████████▏ | 35748/43939 [57:41:12<13:01:55, 5.73s/it]
81%|████████▏ | 35749/43939 [57:41:18<13:01:48, 5.73s/it]
81%|████████▏ | 35750/43939 [57:41:24<13:01:24, 5.73s/it]
{'loss': 0.6559, 'grad_norm': 1.533808708190918, 'learning_rate': 5.107375521825791e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.25, 'epoch': 3.25}
81%|████████▏ | 35750/43939 [57:41:24<13:01:24, 5.73s/it]
81%|████████▏ | 35751/43939 [57:41:29<13:01:31, 5.73s/it]
81%|████████▏ | 35752/43939 [57:41:35<12:59:38, 5.71s/it]
81%|████████▏ | 35753/43939 [57:41:41<12:59:08, 5.71s/it]
81%|████████▏ | 35754/43939 [57:41:47<13:01:36, 5.73s/it]
81%|████████▏ | 35755/43939 [57:41:52<13:01:07, 5.73s/it]
81%|████████▏ | 35756/43939 [57:41:58<13:00:12, 5.72s/it]
81%|████████▏ | 35757/43939 [57:42:04<13:00:25, 5.72s/it]
81%|████████▏ | 35758/43939 [57:42:09<12:59:19, 5.72s/it]
81%|████████▏ | 35759/43939 [57:42:15<12:58:57, 5.71s/it]
81%|████████▏ | 35760/43939 [57:42:21<12:57:32, 5.70s/it]
81%|████████▏ | 35761/43939 [57:42:27<12:57:32, 5.70s/it]
81%|████████▏ | 35762/43939 [57:42:32<12:57:32, 5.71s/it]
81%|████████▏ | 35763/43939 [57:42:38<12:57:02, 5.70s/it]
81%|████████▏ | 35764/43939 [57:42:44<13:06:29, 5.77s/it]
81%|████████▏ | 35765/43939 [57:42:50<13:03:05, 5.75s/it]
81%|████████▏ | 35766/43939 [57:42:55<13:01:30, 5.74s/it]
81%|████████▏ | 35767/43939 [57:43:01<13:02:10, 5.74s/it]
81%|████████▏ | 35768/43939 [57:43:07<13:01:33, 5.74s/it]
81%|████████▏ | 35769/43939 [57:43:12<13:00:19, 5.73s/it]
81%|████████▏ | 35770/43939 [57:43:18<12:57:41, 5.71s/it]
81%|████████▏ | 35771/43939 [57:43:24<12:56:09, 5.70s/it]
81%|████████▏ | 35772/43939 [57:43:30<12:58:05, 5.72s/it]
81%|████████▏ | 35773/43939 [57:43:35<12:59:29, 5.73s/it]
81%|████████▏ | 35774/43939 [57:43:41<13:00:09, 5.73s/it]
81%|████████▏ | 35775/43939 [57:43:47<12:59:44, 5.73s/it]
81%|████████▏ | 35776/43939 [57:43:52<12:58:02, 5.72s/it]
81%|████████▏ | 35777/43939 [57:43:58<12:57:16, 5.71s/it]
81%|████████▏ | 35778/43939 [57:44:04<12:57:05, 5.71s/it]
81%|████████▏ | 35779/43939 [57:44:10<12:57:19, 5.72s/it]
81%|████████▏ | 35780/43939 [57:44:15<12:57:13, 5.72s/it]
81%|████████▏ | 35781/43939 [57:44:21<12:57:18, 5.72s/it]
81%|████████▏ | 35782/43939 [57:44:27<12:55:19, 5.70s/it]
81%|████████▏ | 35783/43939 [57:44:32<12:56:18, 5.71s/it]
81%|████████▏ | 35784/43939 [57:44:38<12:55:37, 5.71s/it]
81%|████████▏ | 35785/43939 [57:44:44<12:54:12, 5.70s/it]
81%|████████▏ | 35786/43939 [57:44:50<13:04:01, 5.77s/it]
81%|████████▏ | 35787/43939 [57:44:56<13:02:49, 5.76s/it]
81%|████████▏ | 35788/43939 [57:45:01<12:59:17, 5.74s/it]
81%|████████▏ | 35789/43939 [57:45:07<12:57:08, 5.72s/it]
81%|████████▏ | 35790/43939 [57:45:13<12:56:47, 5.72s/it]
81%|████████▏ | 35791/43939 [57:45:18<12:56:26, 5.72s/it]
81%|████████▏ | 35792/43939 [57:45:24<12:55:00, 5.71s/it]
81%|████████▏ | 35793/43939 [57:45:30<12:55:52, 5.71s/it]
81%|████████▏ | 35794/43939 [57:45:35<12:54:26, 5.70s/it]
81%|████████▏ | 35795/43939 [57:45:41<12:53:04, 5.70s/it]
81%|████████▏ | 35796/43939 [57:45:47<12:53:20, 5.70s/it]
81%|████████▏ | 35797/43939 [57:45:53<12:55:27, 5.71s/it]
81%|████████▏ | 35798/43939 [57:45:58<12:54:23, 5.71s/it]
81%|████████▏ | 35799/43939 [57:46:04<12:52:55, 5.70s/it]
81%|████████▏ | 35800/43939 [57:46:10<12:52:36, 5.70s/it]
{'loss': 0.6654, 'grad_norm': 1.455149531364441, 'learning_rate': 5.047387027696165e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.17, 'epoch': 3.26}
81%|████████▏ | 35800/43939 [57:46:10<12:52:36, 5.70s/it]
81%|████████▏ | 35801/43939 [57:46:15<12:52:47, 5.70s/it]
81%|████████▏ | 35802/43939 [57:46:21<12:52:45, 5.70s/it]
81%|████████▏ | 35803/43939 [57:46:27<12:52:49, 5.70s/it]
81%|████████▏ | 35804/43939 [57:46:32<12:54:52, 5.72s/it]
81%|████████▏ | 35805/43939 [57:46:38<12:53:24, 5.70s/it]
81%|████████▏ | 35806/43939 [57:46:44<12:53:41, 5.71s/it]
81%|████████▏ | 35807/43939 [57:46:50<13:02:13, 5.77s/it]
81%|████████▏ | 35808/43939 [57:46:55<12:59:01, 5.75s/it]
81%|████████▏ | 35809/43939 [57:47:01<12:58:54, 5.75s/it]
81%|████████▏ | 35810/43939 [57:47:07<13:00:38, 5.76s/it]
82%|████████▏ | 35811/43939 [57:47:13<12:58:59, 5.75s/it]
82%|████████▏ | 35812/43939 [57:47:18<12:58:09, 5.74s/it]
82%|████████▏ | 35813/43939 [57:47:24<12:57:13, 5.74s/it]
82%|████████▏ | 35814/43939 [57:47:30<12:55:19, 5.73s/it]
82%|████████▏ | 35815/43939 [57:47:36<12:54:54, 5.72s/it]
82%|████████▏ | 35816/43939 [57:47:41<12:54:18, 5.72s/it]
82%|████████▏ | 35817/43939 [57:47:47<12:54:43, 5.72s/it]
82%|████████▏ | 35818/43939 [57:47:53<12:52:21, 5.71s/it]
82%|████████▏ | 35819/43939 [57:47:58<12:50:43, 5.70s/it]
82%|████████▏ | 35820/43939 [57:48:04<12:52:38, 5.71s/it]
82%|████████▏ | 35821/43939 [57:48:10<12:51:15, 5.70s/it]
82%|████████▏ | 35822/43939 [57:48:16<12:51:42, 5.70s/it]
82%|████████▏ | 35823/43939 [57:48:21<12:49:27, 5.69s/it]
82%|████████▏ | 35824/43939 [57:48:27<12:50:48, 5.70s/it]
82%|████████▏ | 35825/43939 [57:48:33<12:50:42, 5.70s/it]
82%|████████▏ | 35826/43939 [57:48:38<12:50:38, 5.70s/it]
82%|████████▏ | 35827/43939 [57:48:44<12:52:18, 5.71s/it]
82%|████████▏ | 35828/43939 [57:48:50<13:00:47, 5.78s/it]
82%|████████▏ | 35829/43939 [57:48:56<12:57:02, 5.75s/it]
82%|████████▏ | 35830/43939 [57:49:01<12:54:06, 5.73s/it]
82%|████████▏ | 35831/43939 [57:49:07<12:51:57, 5.71s/it]
82%|████████▏ | 35832/43939 [57:49:13<12:51:08, 5.71s/it]
82%|████████▏ | 35833/43939 [57:49:18<12:50:33, 5.70s/it]
82%|████████▏ | 35834/43939 [57:49:24<12:50:05, 5.70s/it]
82%|████████▏ | 35835/43939 [57:49:30<12:54:09, 5.73s/it]
82%|████████▏ | 35836/43939 [57:49:36<12:53:18, 5.73s/it]
82%|████████▏ | 35837/43939 [57:49:41<12:50:53, 5.71s/it]
82%|████████▏ | 35838/43939 [57:49:47<12:53:04, 5.73s/it]
82%|████████▏ | 35839/43939 [57:49:53<12:51:16, 5.71s/it]
82%|████████▏ | 35840/43939 [57:49:58<12:50:53, 5.71s/it]
82%|████████▏ | 35841/43939 [57:50:04<12:50:21, 5.71s/it]
82%|████████▏ | 35842/43939 [57:50:10<12:50:49, 5.71s/it]
82%|████████▏ | 35843/43939 [57:50:16<12:49:56, 5.71s/it]
82%|████████▏ | 35844/43939 [57:50:21<12:50:21, 5.71s/it]
82%|████████▏ | 35845/43939 [57:50:27<12:49:02, 5.70s/it]
82%|████████▏ | 35846/43939 [57:50:33<12:50:52, 5.72s/it]
82%|████████▏ | 35847/43939 [57:50:38<12:50:45, 5.71s/it]
82%|████████▏ | 35848/43939 [57:50:44<12:49:18, 5.70s/it]
82%|████████▏ | 35849/43939 [57:50:50<12:49:08, 5.70s/it]
82%|████████▏ | 35850/43939 [57:50:56<12:58:49, 5.78s/it]
{'loss': 0.6602, 'grad_norm': 1.724476933479309, 'learning_rate': 4.987713332924174e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.33, 'epoch': 3.26}
82%|████████▏ | 35850/43939 [57:50:56<12:58:49, 5.78s/it]
82%|████████▏ | 35851/43939 [57:51:01<12:56:44, 5.76s/it]
82%|████████▏ | 35852/43939 [57:51:07<12:55:26, 5.75s/it]
82%|████████▏ | 35853/43939 [57:51:13<12:52:53, 5.74s/it]
82%|████████▏ | 35854/43939 [57:51:19<12:50:32, 5.72s/it]
82%|████████▏ | 35855/43939 [57:51:24<12:49:27, 5.71s/it]
82%|████████▏ | 35856/43939 [57:51:30<12:51:50, 5.73s/it]
82%|████████▏ | 35857/43939 [57:51:36<12:50:46, 5.72s/it]
82%|████████▏ | 35858/43939 [57:51:41<12:48:58, 5.71s/it]
82%|████████▏ | 35859/43939 [57:51:47<12:49:14, 5.71s/it]
82%|████████▏ | 35860/43939 [57:51:53<12:49:31, 5.72s/it]
82%|████████▏ | 35861/43939 [57:51:59<12:52:00, 5.73s/it]
82%|████████▏ | 35862/43939 [57:52:04<12:51:54, 5.73s/it]
82%|████████▏ | 35863/43939 [57:52:10<12:49:59, 5.72s/it]
82%|████████▏ | 35864/43939 [57:52:16<12:50:03, 5.72s/it]
82%|████████▏ | 35865/43939 [57:52:21<12:49:19, 5.72s/it]
82%|████████▏ | 35866/43939 [57:52:27<12:48:44, 5.71s/it]
82%|████████▏ | 35867/43939 [57:52:33<12:48:54, 5.72s/it]
82%|████████▏ | 35868/43939 [57:52:39<12:48:10, 5.71s/it]
82%|████████▏ | 35869/43939 [57:52:44<12:47:45, 5.71s/it]
82%|████████▏ | 35870/43939 [57:52:50<12:48:36, 5.72s/it]
82%|████████▏ | 35871/43939 [57:52:56<12:56:58, 5.78s/it]
82%|████████▏ | 35872/43939 [57:53:02<12:51:46, 5.74s/it]
82%|████████▏ | 35873/43939 [57:53:07<12:51:07, 5.74s/it]
82%|████████▏ | 35874/43939 [57:53:13<12:50:06, 5.73s/it]
82%|████████▏ | 35875/43939 [57:53:19<12:47:50, 5.71s/it]
82%|████████▏ | 35876/43939 [57:53:24<12:49:31, 5.73s/it]
82%|████████▏ | 35877/43939 [57:53:30<12:50:08, 5.73s/it]
82%|████████▏ | 35878/43939 [57:53:36<12:48:44, 5.72s/it]
82%|████████▏ | 35879/43939 [57:53:42<12:47:31, 5.71s/it]
82%|████████▏ | 35880/43939 [57:53:47<12:48:28, 5.72s/it]
82%|████████▏ | 35881/43939 [57:53:53<12:46:56, 5.71s/it]
82%|████████▏ | 35882/43939 [57:53:59<12:48:04, 5.72s/it]
82%|████████▏ | 35883/43939 [57:54:05<12:48:19, 5.72s/it]
82%|████████▏ | 35884/43939 [57:54:10<12:47:24, 5.72s/it]
82%|████████▏ | 35885/43939 [57:54:16<12:46:42, 5.71s/it]
82%|████████▏ | 35886/43939 [57:54:22<12:46:44, 5.71s/it]
82%|████████▏ | 35887/43939 [57:54:27<12:46:46, 5.71s/it]
82%|████████▏ | 35888/43939 [57:54:33<12:46:38, 5.71s/it]
82%|████████▏ | 35889/43939 [57:54:39<12:48:06, 5.73s/it]
82%|████████▏ | 35890/43939 [57:54:45<12:47:13, 5.72s/it]
82%|████████▏ | 35891/43939 [57:54:50<12:46:54, 5.72s/it]
82%|████████▏ | 35892/43939 [57:54:56<12:53:40, 5.77s/it]
82%|████████▏ | 35893/43939 [57:55:02<12:49:49, 5.74s/it]
82%|████████▏ | 35894/43939 [57:55:08<12:48:38, 5.73s/it]
82%|████████▏ | 35895/43939 [57:55:13<12:49:14, 5.74s/it]
82%|████████▏ | 35896/43939 [57:55:19<12:48:17, 5.73s/it]
82%|████████▏ | 35897/43939 [57:55:25<12:46:34, 5.72s/it]
82%|████████▏ | 35898/43939 [57:55:30<12:45:48, 5.71s/it]
82%|████████▏ | 35899/43939 [57:55:36<12:44:03, 5.70s/it]
82%|████████▏ | 35900/43939 [57:55:42<12:42:43, 5.69s/it]
{'loss': 0.6563, 'grad_norm': 1.6139812469482422, 'learning_rate': 4.928355379002584e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.57, 'epoch': 3.27}
82%|████████▏ | 35900/43939 [57:55:42<12:42:43, 5.69s/it]
82%|████████▏ | 35901/43939 [57:55:47<12:42:55, 5.69s/it]
82%|████████▏ | 35902/43939 [57:55:53<12:43:30, 5.70s/it]
82%|████████▏ | 35903/43939 [57:55:59<12:43:35, 5.70s/it]
82%|████████▏ | 35904/43939 [57:56:05<12:43:13, 5.70s/it]
82%|████████▏ | 35905/43939 [57:56:10<12:44:49, 5.71s/it]
82%|████████▏ | 35906/43939 [57:56:16<12:43:33, 5.70s/it]
82%|████████▏ | 35907/43939 [57:56:22<12:41:12, 5.69s/it]
82%|████████▏ | 35908/43939 [57:56:27<12:41:27, 5.69s/it]
82%|████████▏ | 35909/43939 [57:56:33<12:41:29, 5.69s/it]
82%|████████▏ | 35910/43939 [57:56:39<12:43:05, 5.70s/it]
82%|████████▏ | 35911/43939 [57:56:44<12:43:01, 5.70s/it]
82%|████████▏ | 35912/43939 [57:56:50<12:42:39, 5.70s/it]
82%|████████▏ | 35913/43939 [57:56:56<12:43:33, 5.71s/it]
82%|████████▏ | 35914/43939 [57:57:02<12:52:34, 5.78s/it]
82%|████████▏ | 35915/43939 [57:57:07<12:49:52, 5.76s/it]
82%|████████▏ | 35916/43939 [57:57:13<12:47:52, 5.74s/it]
82%|████████▏ | 35917/43939 [57:57:19<12:45:11, 5.72s/it]
82%|████████▏ | 35918/43939 [57:57:25<12:42:13, 5.70s/it]
82%|████████▏ | 35919/43939 [57:57:30<12:40:25, 5.69s/it]
82%|████████▏ | 35920/43939 [57:57:36<12:40:38, 5.69s/it]
82%|████████▏ | 35921/43939 [57:57:42<12:42:36, 5.71s/it]
82%|████████▏ | 35922/43939 [57:57:47<12:42:26, 5.71s/it]
82%|████████▏ | 35923/43939 [57:57:53<12:41:45, 5.70s/it]
82%|████████▏ | 35924/43939 [57:57:59<12:40:33, 5.69s/it]
82%|████████▏ | 35925/43939 [57:58:04<12:42:10, 5.71s/it]
82%|████████▏ | 35926/43939 [57:58:10<12:42:13, 5.71s/it]
82%|████████▏ | 35927/43939 [57:58:16<12:41:36, 5.70s/it]
82%|████████▏ | 35928/43939 [57:58:22<12:40:43, 5.70s/it]
82%|████████▏ | 35929/43939 [57:58:27<12:41:12, 5.70s/it]
82%|████████▏ | 35930/43939 [57:58:33<12:40:00, 5.69s/it]
82%|████████▏ | 35931/43939 [57:58:39<12:40:48, 5.70s/it]
82%|████████▏ | 35932/43939 [57:58:44<12:42:04, 5.71s/it]
82%|████████▏ | 35933/43939 [57:58:50<12:42:54, 5.72s/it]
82%|████████▏ | 35934/43939 [57:58:56<12:42:00, 5.71s/it]
82%|████████▏ | 35935/43939 [57:59:02<12:43:41, 5.72s/it]
82%|████████▏ | 35936/43939 [57:59:07<12:53:00, 5.80s/it]
82%|████████▏ | 35937/43939 [57:59:13<12:49:35, 5.77s/it]
82%|████████▏ | 35938/43939 [57:59:19<12:45:50, 5.74s/it]
82%|████████▏ | 35939/43939 [57:59:25<12:42:50, 5.72s/it]
82%|████████▏ | 35940/43939 [57:59:30<12:42:27, 5.72s/it]
82%|████████▏ | 35941/43939 [57:59:36<12:41:30, 5.71s/it]
82%|████████▏ | 35942/43939 [57:59:42<12:42:28, 5.72s/it]
82%|████████▏ | 35943/43939 [57:59:47<12:41:27, 5.71s/it]
82%|████████▏ | 35944/43939 [57:59:53<12:40:51, 5.71s/it]
82%|████████▏ | 35945/43939 [57:59:59<12:41:29, 5.72s/it]
82%|████████▏ | 35946/43939 [58:00:05<12:41:02, 5.71s/it]
82%|████████▏ | 35947/43939 [58:00:10<12:44:15, 5.74s/it]
82%|████████▏ | 35948/43939 [58:00:16<12:41:56, 5.72s/it]
82%|████████▏ | 35949/43939 [58:00:22<12:40:45, 5.71s/it]
82%|████████▏ | 35950/43939 [58:00:27<12:40:19, 5.71s/it]
{'loss': 0.6564, 'grad_norm': 1.7342591285705566, 'learning_rate': 4.869314102442604e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.59, 'epoch': 3.27}
82%|████████▏ | 35950/43939 [58:00:27<12:40:19, 5.71s/it]
82%|████████▏ | 35951/43939 [58:00:33<12:40:38, 5.71s/it]
82%|████████▏ | 35952/43939 [58:00:39<12:40:05, 5.71s/it]
82%|████████▏ | 35953/43939 [58:00:45<12:40:58, 5.72s/it]
82%|████████▏ | 35954/43939 [58:00:50<12:39:56, 5.71s/it]
82%|████████▏ | 35955/43939 [58:00:56<12:39:53, 5.71s/it]
82%|████████▏ | 35956/43939 [58:01:02<12:39:41, 5.71s/it]
82%|████████▏ | 35957/43939 [58:01:08<12:48:44, 5.78s/it]
82%|████████▏ | 35958/43939 [58:01:13<12:46:08, 5.76s/it]
82%|████████▏ | 35959/43939 [58:01:19<12:43:01, 5.74s/it]
82%|████████▏ | 35960/43939 [58:01:25<12:40:30, 5.72s/it]
82%|████████▏ | 35961/43939 [58:01:30<12:40:05, 5.72s/it]
82%|████████▏ | 35962/43939 [58:01:36<12:39:18, 5.71s/it]
82%|████████▏ | 35963/43939 [58:01:42<12:39:42, 5.72s/it]
82%|████████▏ | 35964/43939 [58:01:48<12:39:31, 5.71s/it]
82%|████████▏ | 35965/43939 [58:01:53<12:40:34, 5.72s/it]
82%|████████▏ | 35966/43939 [58:01:59<12:39:21, 5.71s/it]
82%|████████▏ | 35967/43939 [58:02:05<12:40:40, 5.73s/it]
82%|████████▏ | 35968/43939 [58:02:10<12:39:27, 5.72s/it]
82%|████████▏ | 35969/43939 [58:02:16<12:38:16, 5.71s/it]
82%|████████▏ | 35970/43939 [58:02:22<12:36:43, 5.70s/it]
82%|████████▏ | 35971/43939 [58:02:27<12:36:39, 5.70s/it]
82%|████████▏ | 35972/43939 [58:02:33<12:35:58, 5.69s/it]
82%|████████▏ | 35973/43939 [58:02:39<12:37:43, 5.71s/it]
82%|████████▏ | 35974/43939 [58:02:45<12:37:09, 5.70s/it]
82%|████████▏ | 35975/43939 [58:02:50<12:37:16, 5.71s/it]
82%|████████▏ | 35976/43939 [58:02:56<12:37:03, 5.70s/it]
82%|████████▏ | 35977/43939 [58:03:02<12:36:46, 5.70s/it]
82%|████████▏ | 35978/43939 [58:03:07<12:35:37, 5.69s/it]
82%|████████▏ | 35979/43939 [58:03:13<12:46:42, 5.78s/it]
82%|████████▏ | 35980/43939 [58:03:19<12:45:36, 5.77s/it]
82%|████████▏ | 35981/43939 [58:03:25<12:41:42, 5.74s/it]
82%|████████▏ | 35982/43939 [58:03:31<12:39:47, 5.73s/it]
82%|████████▏ | 35983/43939 [58:03:36<12:38:45, 5.72s/it]
82%|████████▏ | 35984/43939 [58:03:42<12:36:49, 5.71s/it]
82%|████████▏ | 35985/43939 [58:03:48<12:35:26, 5.70s/it]
82%|████████▏ | 35986/43939 [58:03:53<12:34:37, 5.69s/it]
82%|████████▏ | 35987/43939 [58:03:59<12:34:53, 5.70s/it]
82%|████████▏ | 35988/43939 [58:04:05<12:33:07, 5.68s/it]
82%|████████▏ | 35989/43939 [58:04:10<12:34:26, 5.69s/it]
82%|████████▏ | 35990/43939 [58:04:16<12:34:19, 5.69s/it]
82%|████████▏ | 35991/43939 [58:04:22<12:34:30, 5.70s/it]
82%|████████▏ | 35992/43939 [58:04:27<12:35:18, 5.70s/it]
82%|████████▏ | 35993/43939 [58:04:33<12:33:15, 5.69s/it]
82%|████████▏ | 35994/43939 [58:04:39<12:36:22, 5.71s/it]
82%|████████▏ | 35995/43939 [58:04:45<12:37:17, 5.72s/it]
82%|████████▏ | 35996/43939 [58:04:50<12:37:16, 5.72s/it]
82%|████████▏ | 35997/43939 [58:04:56<12:36:37, 5.72s/it]
82%|████████▏ | 35998/43939 [58:05:02<12:38:21, 5.73s/it]
82%|████████▏ | 35999/43939 [58:05:07<12:36:15, 5.71s/it]
82%|████████▏ | 36000/43939 [58:05:13<12:45:37, 5.79s/it]
{'loss': 0.6659, 'grad_norm': 1.4513487815856934, 'learning_rate': 4.8105904347591245e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.9, 'epoch': 3.28}
82%|████████▏ | 36000/43939 [58:05:13<12:45:37, 5.79s/it]
82%|████████▏ | 36001/43939 [58:05:19<12:41:36, 5.76s/it]
82%|████████▏ | 36002/43939 [58:05:25<12:41:33, 5.76s/it]
82%|████████▏ | 36003/43939 [58:05:31<12:38:40, 5.74s/it]
82%|████████▏ | 36004/43939 [58:05:36<12:36:50, 5.72s/it]
82%|████████▏ | 36005/43939 [58:05:42<12:35:01, 5.71s/it]
82%|████████▏ | 36006/43939 [58:05:48<12:35:43, 5.72s/it]
82%|████████▏ | 36007/43939 [58:05:53<12:36:02, 5.72s/it]
82%|████████▏ | 36008/43939 [58:05:59<12:36:35, 5.72s/it]
82%|████████▏ | 36009/43939 [58:06:05<12:36:18, 5.72s/it]
82%|████████▏ | 36010/43939 [58:06:11<12:34:29, 5.71s/it]
82%|████████▏ | 36011/43939 [58:06:16<12:34:21, 5.71s/it]
82%|████████▏ | 36012/43939 [58:06:22<12:33:08, 5.70s/it]
82%|████████▏ | 36013/43939 [58:06:28<12:31:50, 5.69s/it]
82%|████████▏ | 36014/43939 [58:06:33<12:30:55, 5.69s/it]
82%|████████▏ | 36015/43939 [58:06:39<12:31:25, 5.69s/it]
82%|████████▏ | 36016/43939 [58:06:45<12:31:46, 5.69s/it]
82%|████████▏ | 36017/43939 [58:06:50<12:31:41, 5.69s/it]
82%|████████▏ | 36018/43939 [58:06:56<12:31:03, 5.69s/it]
82%|████████▏ | 36019/43939 [58:07:02<12:31:28, 5.69s/it]
82%|████████▏ | 36020/43939 [58:07:07<12:32:10, 5.70s/it]
82%|████████▏ | 36021/43939 [58:07:13<12:42:47, 5.78s/it]
82%|████████▏ | 36022/43939 [58:07:19<12:39:26, 5.76s/it]
82%|████████▏ | 36023/43939 [58:07:25<12:37:08, 5.74s/it]
82%|████████▏ | 36024/43939 [58:07:31<12:36:34, 5.74s/it]
82%|████████▏ | 36025/43939 [58:07:36<12:33:40, 5.71s/it]
82%|████████▏ | 36026/43939 [58:07:42<12:33:40, 5.71s/it]
82%|████████▏ | 36027/43939 [58:07:48<12:31:52, 5.70s/it]
82%|████████▏ | 36028/43939 [58:07:53<12:32:32, 5.71s/it]
82%|████████▏ | 36029/43939 [58:07:59<12:32:04, 5.70s/it]
82%|████████▏ | 36030/43939 [58:08:05<12:33:13, 5.71s/it]
82%|████████▏ | 36031/43939 [58:08:10<12:32:14, 5.71s/it]
82%|████████▏ | 36032/43939 [58:08:16<12:31:17, 5.70s/it]
82%|████████▏ | 36033/43939 [58:08:22<12:32:35, 5.71s/it]
82%|████████▏ | 36034/43939 [58:08:28<12:31:56, 5.71s/it]
82%|████████▏ | 36035/43939 [58:08:33<12:31:01, 5.70s/it]
82%|████████▏ | 36036/43939 [58:08:39<12:31:32, 5.71s/it]
82%|████████▏ | 36037/43939 [58:08:45<12:30:27, 5.70s/it]
82%|████████▏ | 36038/43939 [58:08:50<12:30:54, 5.70s/it]
82%|████████▏ | 36039/43939 [58:08:56<12:29:51, 5.70s/it]
82%|████████▏ | 36040/43939 [58:09:02<12:30:45, 5.70s/it]
82%|████████▏ | 36041/43939 [58:09:07<12:30:08, 5.70s/it]
82%|████████▏ | 36042/43939 [58:09:13<12:38:51, 5.77s/it]
82%|████████▏ | 36043/43939 [58:09:19<12:36:52, 5.75s/it]
82%|████████▏ | 36044/43939 [58:09:25<12:34:37, 5.73s/it]
82%|████████▏ | 36045/43939 [58:09:30<12:32:12, 5.72s/it]
82%|████████▏ | 36046/43939 [58:09:36<12:30:43, 5.71s/it]
82%|████████▏ | 36047/43939 [58:09:42<12:29:46, 5.70s/it]
82%|████████▏ | 36048/43939 [58:09:48<12:30:22, 5.71s/it]
82%|████████▏ | 36049/43939 [58:09:53<12:29:42, 5.70s/it]
82%|████████▏ | 36050/43939 [58:09:59<12:30:53, 5.71s/it]
{'loss': 0.6458, 'grad_norm': 1.3770030736923218, 'learning_rate': 4.752185302455997e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.99, 'epoch': 3.28}
82%|████████▏ | 36050/43939 [58:09:59<12:30:53, 5.71s/it]
82%|████████▏ | 36051/43939 [58:10:05<12:31:16, 5.71s/it]
82%|████████▏ | 36052/43939 [58:10:10<12:30:38, 5.71s/it]
82%|████████▏ | 36053/43939 [58:10:16<12:28:52, 5.70s/it]
82%|████████▏ | 36054/43939 [58:10:22<12:29:03, 5.70s/it]
82%|████████▏ | 36055/43939 [58:10:27<12:28:41, 5.70s/it]
82%|████████▏ | 36056/43939 [58:10:33<12:32:05, 5.72s/it]
82%|████████▏ | 36057/43939 [58:10:39<12:34:21, 5.74s/it]
82%|████████▏ | 36058/43939 [58:10:45<12:33:16, 5.73s/it]
82%|████████▏ | 36059/43939 [58:10:50<12:31:09, 5.72s/it]
82%|████████▏ | 36060/43939 [58:10:56<12:32:20, 5.73s/it]
82%|████████▏ | 36061/43939 [58:11:02<12:30:27, 5.72s/it]
82%|████████▏ | 36062/43939 [58:11:08<12:29:12, 5.71s/it]
82%|████████▏ | 36063/43939 [58:11:13<12:30:10, 5.71s/it]
82%|████████▏ | 36064/43939 [58:11:19<12:38:32, 5.78s/it]
82%|████████▏ | 36065/43939 [58:11:25<12:35:51, 5.76s/it]
82%|████████▏ | 36066/43939 [58:11:31<12:32:54, 5.74s/it]
82%|████████▏ | 36067/43939 [58:11:36<12:31:30, 5.73s/it]
82%|████████▏ | 36068/43939 [58:11:42<12:32:28, 5.74s/it]
82%|████████▏ | 36069/43939 [58:11:48<12:30:28, 5.72s/it]
82%|████████▏ | 36070/43939 [58:11:54<12:31:57, 5.73s/it]
82%|████████▏ | 36071/43939 [58:11:59<12:29:40, 5.72s/it]
82%|████████▏ | 36072/43939 [58:12:05<12:31:27, 5.73s/it]
82%|████████▏ | 36073/43939 [58:12:11<12:29:02, 5.71s/it]
82%|████████▏ | 36074/43939 [58:12:16<12:30:16, 5.72s/it]
82%|████████▏ | 36075/43939 [58:12:22<12:27:25, 5.70s/it]
82%|████████▏ | 36076/43939 [58:12:28<12:25:47, 5.69s/it]
82%|████████▏ | 36077/43939 [58:12:33<12:26:04, 5.69s/it]
82%|████████▏ | 36078/43939 [58:12:39<12:24:48, 5.68s/it]
82%|████████▏ | 36079/43939 [58:12:45<12:25:02, 5.69s/it]
82%|████████▏ | 36080/43939 [58:12:50<12:24:35, 5.68s/it]
82%|████████▏ | 36081/43939 [58:12:56<12:23:31, 5.68s/it]
82%|████████▏ | 36082/43939 [58:13:02<12:24:07, 5.68s/it]
82%|████████▏ | 36083/43939 [58:13:07<12:24:18, 5.68s/it]
82%|████████▏ | 36084/43939 [58:13:13<12:25:00, 5.69s/it]
82%|████████▏ | 36085/43939 [58:13:19<12:26:30, 5.70s/it]
82%|████████▏ | 36086/43939 [58:13:25<12:36:12, 5.78s/it]
82%|████████▏ | 36087/43939 [58:13:31<12:32:29, 5.75s/it]
82%|████████▏ | 36088/43939 [58:13:36<12:33:23, 5.76s/it]
82%|████████▏ | 36089/43939 [58:13:42<12:31:59, 5.75s/it]
82%|████████▏ | 36090/43939 [58:13:48<12:30:25, 5.74s/it]
82%|████████▏ | 36091/43939 [58:13:53<12:27:57, 5.72s/it]
82%|████████▏ | 36092/43939 [58:13:59<12:25:54, 5.70s/it]
82%|████████▏ | 36093/43939 [58:14:05<12:27:02, 5.71s/it]
82%|████████▏ | 36094/43939 [58:14:11<12:26:20, 5.71s/it]
82%|████████▏ | 36095/43939 [58:14:16<12:23:55, 5.69s/it]
82%|████████▏ | 36096/43939 [58:14:22<12:23:36, 5.69s/it]
82%|████████▏ | 36097/43939 [58:14:28<12:25:49, 5.71s/it]
82%|████████▏ | 36098/43939 [58:14:33<12:26:15, 5.71s/it]
82%|████████▏ | 36099/43939 [58:14:39<12:26:30, 5.71s/it]
82%|████████▏ | 36100/43939 [58:14:45<12:26:18, 5.71s/it]
{'loss': 0.661, 'grad_norm': 2.054734468460083, 'learning_rate': 4.694099627011433e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.42, 'epoch': 3.29}
82%|████████▏ | 36100/43939 [58:14:45<12:26:18, 5.71s/it]
82%|████████▏ | 36101/43939 [58:14:50<12:27:08, 5.72s/it]
82%|████████▏ | 36102/43939 [58:14:56<12:27:48, 5.73s/it]
82%|████████▏ | 36103/43939 [58:15:02<12:25:48, 5.71s/it]
82%|████████▏ | 36104/43939 [58:15:08<12:25:38, 5.71s/it]
82%|████████▏ | 36105/43939 [58:15:13<12:24:59, 5.71s/it]
82%|████████▏ | 36106/43939 [58:15:19<12:24:10, 5.70s/it]
82%|████████▏ | 36107/43939 [58:15:25<12:23:41, 5.70s/it]
82%|████████▏ | 36108/43939 [58:15:31<12:31:12, 5.76s/it]
82%|████████▏ | 36109/43939 [58:15:36<12:29:57, 5.75s/it]
82%|████████▏ | 36110/43939 [58:15:42<12:30:09, 5.75s/it]
82%|████████▏ | 36111/43939 [58:15:48<12:29:26, 5.74s/it]
82%|████████▏ | 36112/43939 [58:15:54<12:28:01, 5.73s/it]
82%|████████▏ | 36113/43939 [58:15:59<12:27:47, 5.73s/it]
82%|████████▏ | 36114/43939 [58:16:05<12:27:01, 5.73s/it]
82%|████████▏ | 36115/43939 [58:16:11<12:25:45, 5.72s/it]
82%|████████▏ | 36116/43939 [58:16:16<12:25:02, 5.71s/it]
82%|████████▏ | 36117/43939 [58:16:22<12:22:33, 5.70s/it]
82%|████████▏ | 36118/43939 [58:16:28<12:21:26, 5.69s/it]
82%|████████▏ | 36119/43939 [58:16:33<12:21:00, 5.69s/it]
82%|████████▏ | 36120/43939 [58:16:39<12:20:13, 5.68s/it]
82%|████████▏ | 36121/43939 [58:16:45<12:21:32, 5.69s/it]
82%|████████▏ | 36122/43939 [58:16:50<12:22:30, 5.70s/it]
82%|████████▏ | 36123/43939 [58:16:56<12:21:56, 5.70s/it]
82%|████████▏ | 36124/43939 [58:17:02<12:22:37, 5.70s/it]
82%|████████▏ | 36125/43939 [58:17:08<12:22:36, 5.70s/it]
82%|████████▏ | 36126/43939 [58:17:13<12:25:16, 5.72s/it]
82%|████████▏ | 36127/43939 [58:17:19<12:24:19, 5.72s/it]
82%|████████▏ | 36128/43939 [58:17:25<12:24:00, 5.72s/it]
82%|████████▏ | 36129/43939 [58:17:31<12:31:57, 5.78s/it]
82%|████████▏ | 36130/43939 [58:17:36<12:27:57, 5.75s/it]
82%|████████▏ | 36131/43939 [58:17:42<12:26:28, 5.74s/it]
82%|████████▏ | 36132/43939 [58:17:48<12:27:07, 5.74s/it]
82%|████████▏ | 36133/43939 [58:17:54<12:26:19, 5.74s/it]
82%|████████▏ | 36134/43939 [58:17:59<12:24:34, 5.72s/it]
82%|████████▏ | 36135/43939 [58:18:05<12:22:12, 5.71s/it]
82%|████████▏ | 36136/43939 [58:18:11<12:22:59, 5.71s/it]
82%|████████▏ | 36137/43939 [58:18:16<12:21:24, 5.70s/it]
82%|████████▏ | 36138/43939 [58:18:22<12:21:39, 5.70s/it]
82%|████████▏ | 36139/43939 [58:18:28<12:21:35, 5.70s/it]
82%|████████▏ | 36140/43939 [58:18:33<12:22:55, 5.72s/it]
82%|████████▏ | 36141/43939 [58:18:39<12:22:28, 5.71s/it]
82%|████████▏ | 36142/43939 [58:18:45<12:21:53, 5.71s/it]
82%|████████▏ | 36143/43939 [58:18:51<12:23:52, 5.73s/it]
82%|████████▏ | 36144/43939 [58:18:56<12:22:54, 5.72s/it]
82%|████████▏ | 36145/43939 [58:19:02<12:24:39, 5.73s/it]
82%|████████▏ | 36146/43939 [58:19:08<12:24:32, 5.73s/it]
82%|████████▏ | 36147/43939 [58:19:14<12:23:17, 5.72s/it]
82%|████████▏ | 36148/43939 [58:19:19<12:22:52, 5.72s/it]
82%|████████▏ | 36149/43939 [58:19:25<12:21:37, 5.71s/it]
82%|████████▏ | 36150/43939 [58:19:31<12:20:55, 5.71s/it]
{'loss': 0.6722, 'grad_norm': 1.678490161895752, 'learning_rate': 4.6363343248634675e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.69, 'epoch': 3.29}
82%|████████▏ | 36150/43939 [58:19:31<12:20:55, 5.71s/it]
82%|████████▏ | 36151/43939 [58:19:37<12:29:57, 5.78s/it]
82%|████████▏ | 36152/43939 [58:19:42<12:26:16, 5.75s/it]
82%|████████▏ | 36153/43939 [58:19:48<12:24:37, 5.74s/it]
82%|████████▏ | 36154/43939 [58:19:54<12:22:22, 5.72s/it]
82%|████████▏ | 36155/43939 [58:19:59<12:22:42, 5.72s/it]
82%|████████▏ | 36156/43939 [58:20:05<12:21:27, 5.72s/it]
82%|████████▏ | 36157/43939 [58:20:11<12:21:35, 5.72s/it]
82%|████████▏ | 36158/43939 [58:20:17<12:24:43, 5.74s/it]
82%|████████▏ | 36159/43939 [58:20:22<12:23:48, 5.74s/it]
82%|████████▏ | 36160/43939 [58:20:28<12:22:21, 5.73s/it]
82%|████████▏ | 36161/43939 [58:20:34<12:21:40, 5.72s/it]
82%|████████▏ | 36162/43939 [58:20:39<12:20:43, 5.71s/it]
82%|████████▏ | 36163/43939 [58:20:45<12:20:10, 5.71s/it]
82%|████████▏ | 36164/43939 [58:20:51<12:20:28, 5.71s/it]
82%|████████▏ | 36165/43939 [58:20:57<12:19:12, 5.71s/it]
82%|████████▏ | 36166/43939 [58:21:02<12:18:56, 5.70s/it]
82%|████████▏ | 36167/43939 [58:21:08<12:18:35, 5.70s/it]
82%|████████▏ | 36168/43939 [58:21:14<12:18:41, 5.70s/it]
82%|████████▏ | 36169/43939 [58:21:19<12:20:50, 5.72s/it]
82%|████████▏ | 36170/43939 [58:21:25<12:20:56, 5.72s/it]
82%|████████▏ | 36171/43939 [58:21:31<12:21:52, 5.73s/it]
82%|████████▏ | 36172/43939 [58:21:37<12:33:27, 5.82s/it]
82%|████████▏ | 36173/43939 [58:21:43<12:29:51, 5.79s/it]
82%|████████▏ | 36174/43939 [58:21:48<12:25:49, 5.76s/it]
82%|████████▏ | 36175/43939 [58:21:54<12:24:14, 5.75s/it]
82%|████████▏ | 36176/43939 [58:22:00<12:22:31, 5.74s/it]
82%|████████▏ | 36177/43939 [58:22:05<12:20:29, 5.72s/it]
82%|████████▏ | 36178/43939 [58:22:11<12:20:35, 5.73s/it]
82%|████████▏ | 36179/43939 [58:22:17<12:20:45, 5.73s/it]
82%|████████▏ | 36180/43939 [58:22:23<12:19:32, 5.72s/it]
82%|████████▏ | 36181/43939 [58:22:28<12:17:15, 5.70s/it]
82%|████████▏ | 36182/43939 [58:22:34<12:19:12, 5.72s/it]
82%|████████▏ | 36183/43939 [58:22:40<12:18:25, 5.71s/it]
82%|████████▏ | 36184/43939 [58:22:45<12:18:26, 5.71s/it]
82%|████████▏ | 36185/43939 [58:22:51<12:19:23, 5.72s/it]
82%|████████▏ | 36186/43939 [58:22:57<12:18:28, 5.72s/it]
82%|████████▏ | 36187/43939 [58:23:03<12:17:59, 5.71s/it]
82%|████████▏ | 36188/43939 [58:23:08<12:19:21, 5.72s/it]
82%|████████▏ | 36189/43939 [58:23:14<12:18:57, 5.72s/it]
82%|████████▏ | 36190/43939 [58:23:20<12:19:02, 5.72s/it]
82%|████████▏ | 36191/43939 [58:23:26<12:19:17, 5.72s/it]
82%|████████▏ | 36192/43939 [58:23:31<12:20:48, 5.74s/it]
82%|████████▏ | 36193/43939 [58:23:37<12:20:13, 5.73s/it]
82%|████████▏ | 36194/43939 [58:23:43<12:30:48, 5.82s/it]
82%|████████▏ | 36195/43939 [58:23:49<12:26:10, 5.78s/it]
82%|████████▏ | 36196/43939 [58:23:54<12:23:36, 5.76s/it]
82%|████████▏ | 36197/43939 [58:24:00<12:20:08, 5.74s/it]
82%|████████▏ | 36198/43939 [58:24:06<12:18:46, 5.73s/it]
82%|████████▏ | 36199/43939 [58:24:12<12:17:45, 5.72s/it]
82%|████████▏ | 36200/43939 [58:24:17<12:16:48, 5.71s/it]
{'loss': 0.6529, 'grad_norm': 1.412787914276123, 'learning_rate': 4.5788903073954805e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.59, 'epoch': 3.3}
82%|████████▏ | 36200/43939 [58:24:17<12:16:48, 5.71s/it]
82%|████████▏ | 36201/43939 [58:24:23<12:17:19, 5.72s/it]
82%|████████▏ | 36202/43939 [58:24:29<12:16:31, 5.71s/it]
82%|████████▏ | 36203/43939 [58:24:34<12:17:19, 5.72s/it]
82%|████████▏ | 36204/43939 [58:24:40<12:17:13, 5.72s/it]
82%|████████▏ | 36205/43939 [58:24:46<12:17:15, 5.72s/it]
82%|████████▏ | 36206/43939 [58:24:52<12:16:20, 5.71s/it]
82%|████████▏ | 36207/43939 [58:24:57<12:17:50, 5.73s/it]
82%|████████▏ | 36208/43939 [58:25:03<12:16:23, 5.72s/it]
82%|████████▏ | 36209/43939 [58:25:09<12:18:16, 5.73s/it]
82%|████████▏ | 36210/43939 [58:25:14<12:17:54, 5.73s/it]
82%|████████▏ | 36211/43939 [58:25:20<12:17:34, 5.73s/it]
82%|████████▏ | 36212/43939 [58:25:26<12:18:34, 5.74s/it]
82%|████████▏ | 36213/43939 [58:25:32<12:17:42, 5.73s/it]
82%|████████▏ | 36214/43939 [58:25:37<12:16:35, 5.72s/it]
82%|████████▏ | 36215/43939 [58:25:43<12:25:22, 5.79s/it]
82%|████████▏ | 36216/43939 [58:25:49<12:22:24, 5.77s/it]
82%|████████▏ | 36217/43939 [58:25:55<12:19:59, 5.75s/it]
82%|████████▏ | 36218/43939 [58:26:00<12:19:41, 5.75s/it]
82%|████████▏ | 36219/43939 [58:26:06<12:18:07, 5.74s/it]
82%|████████▏ | 36220/43939 [58:26:12<12:17:25, 5.73s/it]
82%|████████▏ | 36221/43939 [58:26:18<12:15:02, 5.71s/it]
82%|████████▏ | 36222/43939 [58:26:23<12:15:38, 5.72s/it]
82%|████████▏ | 36223/43939 [58:26:29<12:14:53, 5.71s/it]
82%|████████▏ | 36224/43939 [58:26:35<12:13:46, 5.71s/it]
82%|████████▏ | 36225/43939 [58:26:40<12:14:08, 5.71s/it]
82%|████████▏ | 36226/43939 [58:26:46<12:11:52, 5.69s/it]
82%|████████▏ | 36227/43939 [58:26:52<12:13:38, 5.71s/it]
82%|████████▏ | 36228/43939 [58:26:57<12:12:09, 5.70s/it]
82%|████████▏ | 36229/43939 [58:27:03<12:12:06, 5.70s/it]
82%|████████▏ | 36230/43939 [58:27:09<12:11:33, 5.69s/it]
82%|████████▏ | 36231/43939 [58:27:15<12:12:51, 5.70s/it]
82%|████████▏ | 36232/43939 [58:27:20<12:12:38, 5.70s/it]
82%|████████▏ | 36233/43939 [58:27:26<12:12:35, 5.70s/it]
82%|████████▏ | 36234/43939 [58:27:32<12:14:50, 5.72s/it]
82%|████████▏ | 36235/43939 [58:27:37<12:13:14, 5.71s/it]
82%|████████▏ | 36236/43939 [58:27:43<12:21:44, 5.78s/it]
82%|████████▏ | 36237/43939 [58:27:49<12:18:47, 5.76s/it]
82%|████████▏ | 36238/43939 [58:27:55<12:16:03, 5.73s/it]
82%|████████▏ | 36239/43939 [58:28:00<12:14:19, 5.72s/it]
82%|████████▏ | 36240/43939 [58:28:06<12:13:16, 5.71s/it]
82%|████████▏ | 36241/43939 [58:28:12<12:12:12, 5.71s/it]
82%|████████▏ | 36242/43939 [58:28:18<12:12:46, 5.71s/it]
82%|████████▏ | 36243/43939 [58:28:23<12:12:40, 5.71s/it]
82%|████████▏ | 36244/43939 [58:28:29<12:11:59, 5.71s/it]
82%|████████▏ | 36245/43939 [58:28:35<12:10:12, 5.69s/it]
82%|████████▏ | 36246/43939 [58:28:40<12:10:16, 5.70s/it]
82%|████████▏ | 36247/43939 [58:28:46<12:10:57, 5.70s/it]
82%|████████▏ | 36248/43939 [58:28:52<12:11:48, 5.71s/it]
82%|████████▏ | 36249/43939 [58:28:57<12:11:18, 5.71s/it]
83%|████████▎ | 36250/43939 [58:29:03<12:11:22, 5.71s/it]
{'loss': 0.6504, 'grad_norm': 1.6679067611694336, 'learning_rate': 4.521768480921845e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.94, 'epoch': 3.3}
83%|████████▎ | 36250/43939 [58:29:03<12:11:22, 5.71s/it]
83%|████████▎ | 36251/43939 [58:29:09<12:11:00, 5.71s/it]
83%|████████▎ | 36252/43939 [58:29:15<12:11:39, 5.71s/it]
83%|████████▎ | 36253/43939 [58:29:20<12:10:30, 5.70s/it]
83%|████████▎ | 36254/43939 [58:29:26<12:09:40, 5.70s/it]
83%|████████▎ | 36255/43939 [58:29:32<12:12:05, 5.72s/it]
83%|████████▎ | 36256/43939 [58:29:37<12:10:26, 5.70s/it]
83%|████████▎ | 36257/43939 [58:29:43<12:10:40, 5.71s/it]
83%|████████▎ | 36258/43939 [58:29:49<12:18:45, 5.77s/it]
83%|████████▎ | 36259/43939 [58:29:55<12:15:58, 5.75s/it]
83%|████████▎ | 36260/43939 [58:30:00<12:13:58, 5.73s/it]
83%|████████▎ | 36261/43939 [58:30:06<12:11:51, 5.72s/it]
83%|████████▎ | 36262/43939 [58:30:12<12:10:31, 5.71s/it]
83%|████████▎ | 36263/43939 [58:30:18<12:09:57, 5.71s/it]
83%|████████▎ | 36264/43939 [58:30:23<12:11:12, 5.72s/it]
83%|████████▎ | 36265/43939 [58:30:29<12:10:10, 5.71s/it]
83%|████████▎ | 36266/43939 [58:30:35<12:09:48, 5.71s/it]
83%|████████▎ | 36267/43939 [58:30:40<12:10:28, 5.71s/it]
83%|████████▎ | 36268/43939 [58:30:46<12:11:06, 5.72s/it]
83%|████████▎ | 36269/43939 [58:30:52<12:10:41, 5.72s/it]
83%|████████▎ | 36270/43939 [58:30:58<12:09:47, 5.71s/it]
83%|████████▎ | 36271/43939 [58:31:03<12:09:12, 5.71s/it]
83%|████████▎ | 36272/43939 [58:31:09<12:08:29, 5.70s/it]
83%|████████▎ | 36273/43939 [58:31:15<12:09:04, 5.71s/it]
83%|████████▎ | 36274/43939 [58:31:20<12:08:30, 5.70s/it]
83%|████████▎ | 36275/43939 [58:31:26<12:08:53, 5.71s/it]
83%|████████▎ | 36276/43939 [58:31:32<12:10:03, 5.72s/it]
83%|████████▎ | 36277/43939 [58:31:38<12:10:46, 5.72s/it]
83%|████████▎ | 36278/43939 [58:31:43<12:11:51, 5.73s/it]
83%|████████▎ | 36279/43939 [58:31:49<12:10:41, 5.72s/it]
83%|████████▎ | 36280/43939 [58:31:55<12:18:58, 5.79s/it]
83%|████████▎ | 36281/43939 [58:32:01<12:15:20, 5.76s/it]
83%|████████▎ | 36282/43939 [58:32:06<12:13:34, 5.75s/it]
83%|████████▎ | 36283/43939 [58:32:12<12:10:29, 5.72s/it]
83%|████████▎ | 36284/43939 [58:32:18<12:11:03, 5.73s/it]
83%|████████▎ | 36285/43939 [58:32:23<12:11:00, 5.73s/it]
83%|████████▎ | 36286/43939 [58:32:29<12:09:43, 5.72s/it]
83%|████████▎ | 36287/43939 [58:32:35<12:11:07, 5.73s/it]
83%|████████▎ | 36288/43939 [58:32:41<12:11:26, 5.74s/it]
83%|████████▎ | 36289/43939 [58:32:46<12:10:57, 5.73s/it]
83%|████████▎ | 36290/43939 [58:32:52<12:09:31, 5.72s/it]
83%|████████▎ | 36291/43939 [58:32:58<12:09:20, 5.72s/it]
83%|████████▎ | 36292/43939 [58:33:04<12:10:58, 5.74s/it]
83%|████████▎ | 36293/43939 [58:33:09<12:13:25, 5.76s/it]
83%|████████▎ | 36294/43939 [58:33:15<12:10:45, 5.74s/it]
83%|████████▎ | 36295/43939 [58:33:21<12:09:24, 5.73s/it]
83%|████████▎ | 36296/43939 [58:33:27<12:11:09, 5.74s/it]
83%|████████▎ | 36297/43939 [58:33:32<12:10:47, 5.74s/it]
83%|████████▎ | 36298/43939 [58:33:38<12:08:58, 5.72s/it]
83%|████████▎ | 36299/43939 [58:33:44<12:07:40, 5.71s/it]
83%|████████▎ | 36300/43939 [58:33:50<12:14:14, 5.77s/it]
{'loss': 0.673, 'grad_norm': 1.6302002668380737, 'learning_rate': 4.464969746673609e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.5, 'epoch': 3.3}
83%|████████▎ | 36300/43939 [58:33:50<12:14:14, 5.77s/it]
83%|████████▎ | 36301/43939 [58:33:55<12:12:05, 5.75s/it]
83%|████████▎ | 36302/43939 [58:34:01<12:11:49, 5.75s/it]
83%|████████▎ | 36303/43939 [58:34:07<12:10:03, 5.74s/it]
83%|████████▎ | 36304/43939 [58:34:12<12:09:25, 5.73s/it]
83%|████████▎ | 36305/43939 [58:34:18<12:07:22, 5.72s/it]
83%|████████▎ | 36306/43939 [58:34:24<12:07:16, 5.72s/it]
83%|████████▎ | 36307/43939 [58:34:30<12:07:34, 5.72s/it]
83%|████████▎ | 36308/43939 [58:34:35<12:07:00, 5.72s/it]
83%|████████▎ | 36309/43939 [58:34:41<12:05:40, 5.71s/it]
83%|████████▎ | 36310/43939 [58:34:47<12:05:58, 5.71s/it]
83%|████████▎ | 36311/43939 [58:34:52<12:06:52, 5.72s/it]
83%|████████▎ | 36312/43939 [58:34:58<12:07:14, 5.72s/it]
83%|████████▎ | 36313/43939 [58:35:04<12:06:37, 5.72s/it]
83%|████████▎ | 36314/43939 [58:35:10<12:04:55, 5.70s/it]
83%|████████▎ | 36315/43939 [58:35:15<12:04:16, 5.70s/it]
83%|████████▎ | 36316/43939 [58:35:21<12:03:47, 5.70s/it]
83%|████████▎ | 36317/43939 [58:35:27<12:04:33, 5.70s/it]
83%|████████▎ | 36318/43939 [58:35:32<12:03:29, 5.70s/it]
83%|████████▎ | 36319/43939 [58:35:38<12:03:06, 5.69s/it]
83%|████████▎ | 36320/43939 [58:35:44<12:03:22, 5.70s/it]
83%|████████▎ | 36321/43939 [58:35:49<12:04:31, 5.71s/it]
83%|████████▎ | 36322/43939 [58:35:55<12:11:50, 5.76s/it]
83%|████████▎ | 36323/43939 [58:36:01<12:11:03, 5.76s/it]
83%|████████▎ | 36324/43939 [58:36:07<12:09:27, 5.75s/it]
83%|████████▎ | 36325/43939 [58:36:12<12:06:54, 5.73s/it]
83%|████████▎ | 36326/43939 [58:36:18<12:08:48, 5.74s/it]
83%|████████▎ | 36327/43939 [58:36:24<12:07:10, 5.73s/it]
83%|████████▎ | 36328/43939 [58:36:30<12:07:09, 5.73s/it]
83%|████████▎ | 36329/43939 [58:36:35<12:04:42, 5.71s/it]
83%|████████▎ | 36330/43939 [58:36:41<12:04:17, 5.71s/it]
83%|████████▎ | 36331/43939 [58:36:47<12:04:12, 5.71s/it]
83%|████████▎ | 36332/43939 [58:36:52<12:03:08, 5.70s/it]
83%|████████▎ | 36333/43939 [58:36:58<12:03:01, 5.70s/it]
83%|████████▎ | 36334/43939 [58:37:04<12:03:08, 5.71s/it]
83%|████████▎ | 36335/43939 [58:37:10<12:02:45, 5.70s/it]
83%|████████▎ | 36336/43939 [58:37:15<12:03:36, 5.71s/it]
83%|████████▎ | 36337/43939 [58:37:21<12:02:26, 5.70s/it]
83%|████████▎ | 36338/43939 [58:37:27<12:01:58, 5.70s/it]
83%|████████▎ | 36339/43939 [58:37:32<12:04:27, 5.72s/it]
83%|████████▎ | 36340/43939 [58:37:38<12:03:23, 5.71s/it]
83%|████████▎ | 36341/43939 [58:37:44<12:02:48, 5.71s/it]
83%|████████▎ | 36342/43939 [58:37:50<12:04:31, 5.72s/it]
83%|████████▎ | 36343/43939 [58:37:55<12:04:01, 5.72s/it]
83%|████████▎ | 36344/43939 [58:38:01<12:12:47, 5.79s/it]
83%|████████▎ | 36345/43939 [58:38:07<12:10:39, 5.77s/it]
83%|████████▎ | 36346/43939 [58:38:13<12:08:22, 5.76s/it]
83%|████████▎ | 36347/43939 [58:38:18<12:07:52, 5.75s/it]
83%|████████▎ | 36348/43939 [58:38:24<12:05:43, 5.74s/it]
83%|████████▎ | 36349/43939 [58:38:30<12:04:33, 5.73s/it]
83%|████████▎ | 36350/43939 [58:38:36<12:03:21, 5.72s/it]
{'loss': 0.6477, 'grad_norm': 1.4526721239089966, 'learning_rate': 4.408495000784274e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.37, 'epoch': 3.31}
83%|████████▎ | 36350/43939 [58:38:36<12:03:21, 5.72s/it]
83%|████████▎ | 36351/43939 [58:38:41<12:03:06, 5.72s/it]
83%|████████▎ | 36352/43939 [58:38:47<12:03:51, 5.72s/it]
83%|████████▎ | 36353/43939 [58:38:53<12:03:50, 5.73s/it]
83%|████████▎ | 36354/43939 [58:38:58<12:04:23, 5.73s/it]
83%|████████▎ | 36355/43939 [58:39:04<12:05:48, 5.74s/it]
83%|████████▎ | 36356/43939 [58:39:10<12:04:22, 5.73s/it]
83%|████████▎ | 36357/43939 [58:39:16<12:03:52, 5.73s/it]
83%|████████▎ | 36358/43939 [58:39:21<12:02:35, 5.72s/it]
83%|████████▎ | 36359/43939 [58:39:27<12:01:20, 5.71s/it]
83%|████████▎ | 36360/43939 [58:39:33<11:59:38, 5.70s/it]
83%|████████▎ | 36361/43939 [58:39:38<11:59:45, 5.70s/it]
83%|████████▎ | 36362/43939 [58:39:44<12:01:25, 5.71s/it]
83%|████████▎ | 36363/43939 [58:39:50<12:01:35, 5.71s/it]
83%|████████▎ | 36364/43939 [58:39:56<12:01:48, 5.72s/it]
83%|████████▎ | 36365/43939 [58:40:02<12:10:02, 5.78s/it]
83%|████████▎ | 36366/43939 [58:40:07<12:06:56, 5.76s/it]
83%|████████▎ | 36367/43939 [58:40:13<12:04:25, 5.74s/it]
83%|████████▎ | 36368/43939 [58:40:19<12:02:39, 5.73s/it]
83%|████████▎ | 36369/43939 [58:40:24<12:01:24, 5.72s/it]
83%|████████▎ | 36370/43939 [58:40:30<12:00:02, 5.71s/it]
83%|████████▎ | 36371/43939 [58:40:36<11:59:46, 5.71s/it]
83%|████████▎ | 36372/43939 [58:40:41<12:00:00, 5.71s/it]
83%|████████▎ | 36373/43939 [58:40:47<12:01:12, 5.72s/it]
83%|████████▎ | 36374/43939 [58:40:53<12:01:11, 5.72s/it]
83%|████████▎ | 36375/43939 [58:40:59<11:59:44, 5.71s/it]
83%|████████▎ | 36376/43939 [58:41:04<11:59:59, 5.71s/it]
83%|████████▎ | 36377/43939 [58:41:10<12:00:42, 5.72s/it]
83%|████████▎ | 36378/43939 [58:41:16<11:57:54, 5.70s/it]
83%|████████▎ | 36379/43939 [58:41:21<11:58:14, 5.70s/it]
83%|████████▎ | 36380/43939 [58:41:27<11:58:42, 5.70s/it]
83%|████████▎ | 36381/43939 [58:41:33<11:58:22, 5.70s/it]
83%|████████▎ | 36382/43939 [58:41:39<11:58:13, 5.70s/it]
83%|████████▎ | 36383/43939 [58:41:44<11:59:17, 5.71s/it]
83%|████████▎ | 36384/43939 [58:41:50<11:58:50, 5.71s/it]
83%|████████▎ | 36385/43939 [58:41:56<11:58:19, 5.71s/it]
83%|████████▎ | 36386/43939 [58:42:01<11:58:07, 5.70s/it]
83%|████████▎ | 36387/43939 [58:42:07<12:04:58, 5.76s/it]
83%|████████▎ | 36388/43939 [58:42:13<12:03:09, 5.75s/it]
83%|████████▎ | 36389/43939 [58:42:19<12:02:07, 5.74s/it]
83%|████████▎ | 36390/43939 [58:42:24<12:00:32, 5.73s/it]
83%|████████▎ | 36391/43939 [58:42:30<12:01:39, 5.74s/it]
83%|████████▎ | 36392/43939 [58:42:36<12:00:26, 5.73s/it]
83%|████████▎ | 36393/43939 [58:42:42<11:59:40, 5.72s/it]
83%|████████▎ | 36394/43939 [58:42:47<11:59:01, 5.72s/it]
83%|████████▎ | 36395/43939 [58:42:53<11:57:43, 5.71s/it]
83%|████████▎ | 36396/43939 [58:42:59<11:57:24, 5.71s/it]
83%|████████▎ | 36397/43939 [58:43:04<11:55:53, 5.70s/it]
83%|████████▎ | 36398/43939 [58:43:10<11:56:39, 5.70s/it]
83%|████████▎ | 36399/43939 [58:43:16<11:55:28, 5.69s/it]
83%|████████▎ | 36400/43939 [58:43:21<11:57:46, 5.71s/it]
{'loss': 0.6436, 'grad_norm': 1.7062057256698608, 'learning_rate': 4.3523451342756705e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 442.05, 'epoch': 3.31}
83%|████████▎ | 36400/43939 [58:43:21<11:57:46, 5.71s/it]
83%|████████▎ | 36401/43939 [58:43:27<11:57:34, 5.71s/it]
83%|████████▎ | 36402/43939 [58:43:33<11:56:24, 5.70s/it]
83%|████████▎ | 36403/43939 [58:43:39<11:56:07, 5.70s/it]
83%|████████▎ | 36404/43939 [58:43:44<11:56:28, 5.71s/it]
83%|████████▎ | 36405/43939 [58:43:50<11:56:38, 5.71s/it]
83%|████████▎ | 36406/43939 [58:43:56<11:56:29, 5.71s/it]
83%|████████▎ | 36407/43939 [58:44:01<11:57:14, 5.71s/it]
83%|████████▎ | 36408/43939 [58:44:07<12:05:43, 5.78s/it]
83%|████████▎ | 36409/43939 [58:44:13<12:04:18, 5.77s/it]
83%|████████▎ | 36410/43939 [58:44:19<12:01:17, 5.75s/it]
83%|████████▎ | 36411/43939 [58:44:24<11:58:29, 5.73s/it]
83%|████████▎ | 36412/43939 [58:44:30<11:57:47, 5.72s/it]
83%|████████▎ | 36413/43939 [58:44:36<11:56:14, 5.71s/it]
83%|████████▎ | 36414/43939 [58:44:42<11:56:23, 5.71s/it]
83%|████████▎ | 36415/43939 [58:44:47<11:55:21, 5.70s/it]
83%|████████▎ | 36416/43939 [58:44:53<11:54:55, 5.70s/it]
83%|████████▎ | 36417/43939 [58:44:59<11:55:44, 5.71s/it]
83%|████████▎ | 36418/43939 [58:45:04<11:54:52, 5.70s/it]
83%|████████▎ | 36419/43939 [58:45:10<11:54:12, 5.70s/it]
83%|████████▎ | 36420/43939 [58:45:16<11:53:28, 5.69s/it]
83%|████████▎ | 36421/43939 [58:45:21<11:52:41, 5.69s/it]
83%|████████▎ | 36422/43939 [58:45:27<11:53:09, 5.69s/it]
83%|████████▎ | 36423/43939 [58:45:33<11:52:04, 5.68s/it]
83%|████████▎ | 36424/43939 [58:45:38<11:51:01, 5.68s/it]
83%|████████▎ | 36425/43939 [58:45:44<11:52:21, 5.69s/it]
83%|████████▎ | 36426/43939 [58:45:50<11:51:17, 5.68s/it]
83%|████████▎ | 36427/43939 [58:45:56<11:51:07, 5.68s/it]
83%|████████▎ | 36428/43939 [58:46:01<11:51:31, 5.68s/it]
83%|████████▎ | 36429/43939 [58:46:07<11:50:49, 5.68s/it]
83%|████████▎ | 36430/43939 [58:46:13<12:02:25, 5.77s/it]
83%|████████▎ | 36431/43939 [58:46:19<11:58:14, 5.74s/it]
83%|████████▎ | 36432/43939 [58:46:24<11:58:38, 5.74s/it]
83%|████████▎ | 36433/43939 [58:46:30<11:57:39, 5.74s/it]
83%|████████▎ | 36434/43939 [58:46:36<11:55:16, 5.72s/it]
83%|████████▎ | 36435/43939 [58:46:41<11:52:58, 5.70s/it]
83%|████████▎ | 36436/43939 [58:46:47<11:54:28, 5.71s/it]
83%|████████▎ | 36437/43939 [58:46:53<11:54:52, 5.72s/it]
83%|████████▎ | 36438/43939 [58:46:58<11:52:31, 5.70s/it]
83%|████████▎ | 36439/43939 [58:47:04<11:51:25, 5.69s/it]
83%|████████▎ | 36440/43939 [58:47:10<11:51:58, 5.70s/it]
83%|████████▎ | 36441/43939 [58:47:16<11:54:47, 5.72s/it]
83%|████████▎ | 36442/43939 [58:47:21<11:54:15, 5.72s/it]
83%|████████▎ | 36443/43939 [58:47:27<11:52:41, 5.70s/it]
83%|████████▎ | 36444/43939 [58:47:33<11:53:36, 5.71s/it]
83%|████████▎ | 36445/43939 [58:47:38<11:52:22, 5.70s/it]
83%|████████▎ | 36446/43939 [58:47:44<11:52:13, 5.70s/it]
83%|████████▎ | 36447/43939 [58:47:50<11:54:52, 5.73s/it]
83%|████████▎ | 36448/43939 [58:47:56<11:54:27, 5.72s/it]
83%|████████▎ | 36449/43939 [58:48:01<11:53:45, 5.72s/it]
83%|████████▎ | 36450/43939 [58:48:07<11:53:14, 5.71s/it]
{'loss': 0.6473, 'grad_norm': 1.7726435661315918, 'learning_rate': 4.2965210330439015e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.83, 'epoch': 3.32}
83%|████████▎ | 36450/43939 [58:48:07<11:53:14, 5.71s/it]
83%|████████▎ | 36451/43939 [58:48:13<12:02:34, 5.79s/it]
83%|████████▎ | 36452/43939 [58:48:19<12:00:11, 5.77s/it]
83%|████████▎ | 36453/43939 [58:48:24<11:58:23, 5.76s/it]
83%|████████▎ | 36454/43939 [58:48:30<11:55:50, 5.74s/it]
83%|████████▎ | 36455/43939 [58:48:36<11:53:38, 5.72s/it]
83%|████████▎ | 36456/43939 [58:48:42<11:52:03, 5.71s/it]
83%|████████▎ | 36457/43939 [58:48:47<11:51:35, 5.71s/it]
83%|████████▎ | 36458/43939 [58:48:53<11:50:55, 5.70s/it]
83%|████████▎ | 36459/43939 [58:48:59<11:51:12, 5.70s/it]
83%|████████▎ | 36460/43939 [58:49:04<11:50:59, 5.70s/it]
83%|████████▎ | 36461/43939 [58:49:10<11:51:31, 5.71s/it]
83%|████████▎ | 36462/43939 [58:49:16<11:51:20, 5.71s/it]
83%|████████▎ | 36463/43939 [58:49:21<11:52:24, 5.72s/it]
83%|████████▎ | 36464/43939 [58:49:27<11:52:22, 5.72s/it]
83%|████████▎ | 36465/43939 [58:49:33<11:51:21, 5.71s/it]
83%|████████▎ | 36466/43939 [58:49:39<11:50:07, 5.70s/it]
83%|████████▎ | 36467/43939 [58:49:44<11:49:28, 5.70s/it]
83%|████████▎ | 36468/43939 [58:49:50<11:48:42, 5.69s/it]
83%|████████▎ | 36469/43939 [58:49:56<11:51:02, 5.71s/it]
83%|████████▎ | 36470/43939 [58:50:01<11:50:29, 5.71s/it]
83%|████████▎ | 36471/43939 [58:50:07<11:49:50, 5.70s/it]
83%|████████▎ | 36472/43939 [58:50:13<11:50:20, 5.71s/it]
83%|████████▎ | 36473/43939 [58:50:19<11:59:15, 5.78s/it]
83%|████████▎ | 36474/43939 [58:50:24<11:54:46, 5.75s/it]
83%|████████▎ | 36475/43939 [58:50:30<11:53:09, 5.73s/it]
83%|████████▎ | 36476/43939 [58:50:36<11:51:58, 5.72s/it]
83%|████████▎ | 36477/43939 [58:50:42<11:49:50, 5.71s/it]
83%|████████▎ | 36478/43939 [58:50:47<11:49:06, 5.70s/it]
83%|████████▎ | 36479/43939 [58:50:53<11:49:27, 5.71s/it]
83%|████████▎ | 36480/43939 [58:50:59<11:49:40, 5.71s/it]
83%|████████▎ | 36481/43939 [58:51:04<11:49:58, 5.71s/it]
83%|████████▎ | 36482/43939 [58:51:10<11:50:25, 5.72s/it]
83%|████████▎ | 36483/43939 [58:51:16<11:48:38, 5.70s/it]
83%|████████▎ | 36484/43939 [58:51:21<11:48:53, 5.71s/it]
83%|████████▎ | 36485/43939 [58:51:27<11:49:49, 5.71s/it]
83%|████████▎ | 36486/43939 [58:51:33<11:48:43, 5.71s/it]
83%|████████▎ | 36487/43939 [58:51:39<11:48:59, 5.71s/it]
83%|████████▎ | 36488/43939 [58:51:44<11:49:47, 5.72s/it]
83%|████████▎ | 36489/43939 [58:51:50<11:50:03, 5.72s/it]
83%|████████▎ | 36490/43939 [58:51:56<11:49:17, 5.71s/it]
83%|████████▎ | 36491/43939 [58:52:01<11:48:57, 5.71s/it]
83%|████████▎ | 36492/43939 [58:52:07<11:48:20, 5.71s/it]
83%|████████▎ | 36493/43939 [58:52:13<11:48:52, 5.71s/it]
83%|████████▎ | 36494/43939 [58:52:19<11:59:47, 5.80s/it]
83%|████████▎ | 36495/43939 [58:52:25<11:56:02, 5.77s/it]
83%|████████▎ | 36496/43939 [58:52:30<11:53:11, 5.75s/it]
83%|████████▎ | 36497/43939 [58:52:36<11:51:21, 5.74s/it]
83%|████████▎ | 36498/43939 [58:52:42<11:50:01, 5.73s/it]
83%|████████▎ | 36499/43939 [58:52:47<11:49:48, 5.72s/it]
83%|████████▎ | 36500/43939 [58:52:53<11:47:23, 5.71s/it]
{'loss': 0.6618, 'grad_norm': 1.626259684562683, 'learning_rate': 4.241023577845343e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.19, 'epoch': 3.32}
83%|████████▎ | 36500/43939 [58:52:53<11:47:23, 5.71s/it]
83%|████████▎ | 36501/43939 [58:52:59<11:47:43, 5.71s/it]
83%|████████▎ | 36502/43939 [58:53:05<11:50:03, 5.73s/it]
83%|████████▎ | 36503/43939 [58:53:10<11:48:03, 5.71s/it]
83%|████████▎ | 36504/43939 [58:53:16<11:49:33, 5.73s/it]
83%|████████▎ | 36505/43939 [58:53:22<11:49:21, 5.73s/it]
83%|████████▎ | 36506/43939 [58:53:27<11:48:01, 5.72s/it]
83%|████████▎ | 36507/43939 [58:53:33<11:47:36, 5.71s/it]
83%|████████▎ | 36508/43939 [58:53:39<11:48:34, 5.72s/it]
83%|████████▎ | 36509/43939 [58:53:45<11:49:40, 5.73s/it]
83%|████████▎ | 36510/43939 [58:53:50<11:48:38, 5.72s/it]
83%|████████▎ | 36511/43939 [58:53:56<11:46:00, 5.70s/it]
83%|████████▎ | 36512/43939 [58:54:02<11:46:35, 5.71s/it]
83%|████████▎ | 36513/43939 [58:54:07<11:46:08, 5.71s/it]
83%|████████▎ | 36514/43939 [58:54:13<11:45:34, 5.70s/it]
83%|████████▎ | 36515/43939 [58:54:19<11:55:52, 5.79s/it]
83%|████████▎ | 36516/43939 [58:54:25<11:51:25, 5.75s/it]
83%|████████▎ | 36517/43939 [58:54:30<11:50:28, 5.74s/it]
83%|████████▎ | 36518/43939 [58:54:36<11:49:21, 5.74s/it]
83%|████████▎ | 36519/43939 [58:54:42<11:46:30, 5.71s/it]
83%|████████▎ | 36520/43939 [58:54:48<11:45:50, 5.71s/it]
83%|████████▎ | 36521/43939 [58:54:53<11:45:40, 5.71s/it]
83%|████████▎ | 36522/43939 [58:54:59<11:45:01, 5.70s/it]
83%|████████▎ | 36523/43939 [58:55:05<11:44:40, 5.70s/it]
83%|████████▎ | 36524/43939 [58:55:10<11:42:47, 5.69s/it]
83%|████████▎ | 36525/43939 [58:55:16<11:43:06, 5.69s/it]
83%|████████▎ | 36526/43939 [58:55:22<11:42:09, 5.68s/it]
83%|████████▎ | 36527/43939 [58:55:27<11:43:00, 5.69s/it]
83%|████████▎ | 36528/43939 [58:55:33<11:43:57, 5.70s/it]
83%|████████▎ | 36529/43939 [58:55:39<11:44:31, 5.70s/it]
83%|████████▎ | 36530/43939 [58:55:45<11:44:35, 5.71s/it]
83%|████████▎ | 36531/43939 [58:55:50<11:43:40, 5.70s/it]
83%|████████▎ | 36532/43939 [58:55:56<11:43:57, 5.70s/it]
83%|████████▎ | 36533/43939 [58:56:02<11:44:19, 5.71s/it]
83%|████████▎ | 36534/43939 [58:56:07<11:45:07, 5.71s/it]
83%|████████▎ | 36535/43939 [58:56:13<11:43:47, 5.70s/it]
83%|████████▎ | 36536/43939 [58:56:19<11:52:44, 5.78s/it]
83%|████████▎ | 36537/43939 [58:56:25<11:51:42, 5.77s/it]
83%|████████▎ | 36538/43939 [58:56:30<11:48:53, 5.75s/it]
83%|████████▎ | 36539/43939 [58:56:36<11:48:33, 5.75s/it]
83%|████████▎ | 36540/43939 [58:56:42<11:47:37, 5.74s/it]
83%|████████▎ | 36541/43939 [58:56:48<11:45:34, 5.72s/it]
83%|████████▎ | 36542/43939 [58:56:53<11:45:04, 5.72s/it]
83%|████████▎ | 36543/43939 [58:56:59<11:44:31, 5.72s/it]
83%|████████▎ | 36544/43939 [58:57:05<11:43:26, 5.71s/it]
83%|████████▎ | 36545/43939 [58:57:10<11:43:16, 5.71s/it]
83%|████████▎ | 36546/43939 [58:57:16<11:44:58, 5.72s/it]
83%|████████▎ | 36547/43939 [58:57:22<11:43:58, 5.71s/it]
83%|████████▎ | 36548/43939 [58:57:28<11:43:11, 5.71s/it]
83%|████████▎ | 36549/43939 [58:57:33<11:42:58, 5.71s/it]
83%|████████▎ | 36550/43939 [58:57:39<11:43:51, 5.72s/it]
{'loss': 0.6607, 'grad_norm': 1.914724349975586, 'learning_rate': 4.185853644282769e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.76, 'epoch': 3.33}
83%|████████▎ | 36550/43939 [58:57:39<11:43:51, 5.72s/it]
83%|████████▎ | 36551/43939 [58:57:45<11:43:10, 5.71s/it]
83%|████████▎ | 36552/43939 [58:57:50<11:42:11, 5.70s/it]
83%|████████▎ | 36553/43939 [58:57:56<11:42:02, 5.70s/it]
83%|████████▎ | 36554/43939 [58:58:02<11:45:27, 5.73s/it]
83%|████████▎ | 36555/43939 [58:58:08<11:44:34, 5.73s/it]
83%|████████▎ | 36556/43939 [58:58:13<11:43:27, 5.72s/it]
83%|████████▎ | 36557/43939 [58:58:19<11:42:47, 5.71s/it]
83%|████████▎ | 36558/43939 [58:58:25<11:51:30, 5.78s/it]
83%|████████▎ | 36559/43939 [58:58:31<11:49:37, 5.77s/it]
83%|████████▎ | 36560/43939 [58:58:36<11:46:15, 5.74s/it]
83%|████████▎ | 36561/43939 [58:58:42<11:44:42, 5.73s/it]
83%|████████▎ | 36562/43939 [58:58:48<11:42:44, 5.72s/it]
83%|████████▎ | 36563/43939 [58:58:53<11:45:13, 5.74s/it]
83%|████████▎ | 36564/43939 [58:58:59<11:44:38, 5.73s/it]
83%|████████▎ | 36565/43939 [58:59:05<11:43:12, 5.72s/it]
83%|████████▎ | 36566/43939 [58:59:11<11:42:08, 5.71s/it]
83%|████████▎ | 36567/43939 [58:59:16<11:42:46, 5.72s/it]
83%|████████▎ | 36568/43939 [58:59:22<11:42:46, 5.72s/it]
83%|████████▎ | 36569/43939 [58:59:28<11:40:36, 5.70s/it]
83%|████████▎ | 36570/43939 [58:59:33<11:40:21, 5.70s/it]
83%|████████▎ | 36571/43939 [58:59:39<11:38:21, 5.69s/it]
83%|████████▎ | 36572/43939 [58:59:45<11:40:05, 5.70s/it]
83%|████████▎ | 36573/43939 [58:59:51<11:40:27, 5.71s/it]
83%|████████▎ | 36574/43939 [58:59:56<11:41:32, 5.72s/it]
83%|████████▎ | 36575/43939 [59:00:02<11:40:27, 5.71s/it]
83%|████████▎ | 36576/43939 [59:00:08<11:41:35, 5.72s/it]
83%|████████▎ | 36577/43939 [59:00:13<11:40:34, 5.71s/it]
83%|████████▎ | 36578/43939 [59:00:19<11:39:42, 5.70s/it]
83%|████████▎ | 36579/43939 [59:00:25<11:50:10, 5.79s/it]
83%|████████▎ | 36580/43939 [59:00:31<11:46:40, 5.76s/it]
83%|████████▎ | 36581/43939 [59:00:36<11:44:39, 5.75s/it]
83%|████████▎ | 36582/43939 [59:00:42<11:43:09, 5.73s/it]
83%|████████▎ | 36583/43939 [59:00:48<11:43:24, 5.74s/it]
83%|████████▎ | 36584/43939 [59:00:54<11:40:58, 5.72s/it]
83%|████████▎ | 36585/43939 [59:00:59<11:39:55, 5.71s/it]
83%|████████▎ | 36586/43939 [59:01:05<11:39:04, 5.70s/it]
83%|████████▎ | 36587/43939 [59:01:11<11:39:15, 5.71s/it]
83%|████████▎ | 36588/43939 [59:01:16<11:37:49, 5.70s/it]
83%|████████▎ | 36589/43939 [59:01:22<11:39:19, 5.71s/it]
83%|████████▎ | 36590/43939 [59:01:28<11:40:07, 5.72s/it]
83%|████████▎ | 36591/43939 [59:01:34<11:40:13, 5.72s/it]
83%|████████▎ | 36592/43939 [59:01:39<11:39:06, 5.71s/it]
83%|████████▎ | 36593/43939 [59:01:45<11:39:46, 5.72s/it]
83%|████████▎ | 36594/43939 [59:01:51<11:39:21, 5.71s/it]
83%|████████▎ | 36595/43939 [59:01:56<11:39:02, 5.71s/it]
83%|████████▎ | 36596/43939 [59:02:02<11:39:22, 5.71s/it]
83%|████████▎ | 36597/43939 [59:02:08<11:38:04, 5.70s/it]
83%|████████▎ | 36598/43939 [59:02:14<11:38:19, 5.71s/it]
83%|████████▎ | 36599/43939 [59:02:19<11:38:35, 5.71s/it]
83%|████████▎ | 36600/43939 [59:02:25<11:47:47, 5.79s/it]
{'loss': 0.6544, 'grad_norm': 1.4026740789413452, 'learning_rate': 4.131012102791532e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.25, 'epoch': 3.33}
83%|████████▎ | 36600/43939 [59:02:25<11:47:47, 5.79s/it]
83%|████████▎ | 36601/43939 [59:02:31<11:44:38, 5.76s/it]
83%|████████▎ | 36602/43939 [59:02:37<11:41:47, 5.74s/it]
83%|████████▎ | 36603/43939 [59:02:42<11:41:34, 5.74s/it]
83%|████████▎ | 36604/43939 [59:02:48<11:39:22, 5.72s/it]
83%|████████▎ | 36605/43939 [59:02:54<11:40:15, 5.73s/it]
83%|████████▎ | 36606/43939 [59:02:59<11:37:39, 5.71s/it]
83%|████████▎ | 36607/43939 [59:03:05<11:38:17, 5.71s/it]
83%|████████▎ | 36608/43939 [59:03:11<11:38:51, 5.72s/it]
83%|████████▎ | 36609/43939 [59:03:17<11:37:57, 5.71s/it]
83%|████████▎ | 36610/43939 [59:03:22<11:37:45, 5.71s/it]
83%|████████▎ | 36611/43939 [59:03:28<11:35:41, 5.70s/it]
83%|████████▎ | 36612/43939 [59:03:34<11:36:14, 5.70s/it]
83%|████████▎ | 36613/43939 [59:03:39<11:35:51, 5.70s/it]
83%|████████▎ | 36614/43939 [59:03:45<11:36:33, 5.71s/it]
83%|████████▎ | 36615/43939 [59:03:51<11:35:30, 5.70s/it]
83%|████████▎ | 36616/43939 [59:03:56<11:35:55, 5.70s/it]
83%|████████▎ | 36617/43939 [59:04:02<11:35:11, 5.70s/it]
83%|████████▎ | 36618/43939 [59:04:08<11:35:44, 5.70s/it]
83%|████████▎ | 36619/43939 [59:04:14<11:36:31, 5.71s/it]
83%|████████▎ | 36620/43939 [59:04:19<11:37:03, 5.71s/it]
83%|████████▎ | 36621/43939 [59:04:25<11:44:15, 5.77s/it]
83%|████████▎ | 36622/43939 [59:04:31<11:41:09, 5.75s/it]
83%|████████▎ | 36623/43939 [59:04:37<11:40:14, 5.74s/it]
83%|████████▎ | 36624/43939 [59:04:42<11:39:33, 5.74s/it]
83%|████████▎ | 36625/43939 [59:04:48<11:39:00, 5.73s/it]
83%|████████▎ | 36626/43939 [59:04:54<11:38:28, 5.73s/it]
83%|████████▎ | 36627/43939 [59:05:00<11:37:09, 5.72s/it]
83%|████████▎ | 36628/43939 [59:05:05<11:36:21, 5.71s/it]
83%|████████▎ | 36629/43939 [59:05:11<11:36:29, 5.72s/it]
83%|████████▎ | 36630/43939 [59:05:17<11:37:33, 5.73s/it]
83%|████████▎ | 36631/43939 [59:05:22<11:36:25, 5.72s/it]
83%|████████▎ | 36632/43939 [59:05:28<11:36:18, 5.72s/it]
83%|████████▎ | 36633/43939 [59:05:34<11:35:22, 5.71s/it]
83%|████████▎ | 36634/43939 [59:05:39<11:33:49, 5.70s/it]
83%|████████▎ | 36635/43939 [59:05:45<11:33:51, 5.70s/it]
83%|████████▎ | 36636/43939 [59:05:51<11:34:15, 5.70s/it]
83%|████████▎ | 36637/43939 [59:05:57<11:33:23, 5.70s/it]
83%|████████▎ | 36638/43939 [59:06:02<11:35:57, 5.72s/it]
83%|████████▎ | 36639/43939 [59:06:08<11:35:15, 5.71s/it]
83%|████████▎ | 36640/43939 [59:06:14<11:35:19, 5.72s/it]
83%|████████▎ | 36641/43939 [59:06:19<11:34:25, 5.71s/it]
83%|████████▎ | 36642/43939 [59:06:25<11:33:23, 5.70s/it]
83%|████████▎ | 36643/43939 [59:06:31<11:41:11, 5.77s/it]
83%|████████▎ | 36644/43939 [59:06:37<11:38:31, 5.75s/it]
83%|████████▎ | 36645/43939 [59:06:42<11:36:46, 5.73s/it]
83%|████████▎ | 36646/43939 [59:06:48<11:36:23, 5.73s/it]
83%|████████▎ | 36647/43939 [59:06:54<11:34:35, 5.72s/it]
83%|████████▎ | 36648/43939 [59:07:00<11:33:44, 5.71s/it]
83%|████████▎ | 36649/43939 [59:07:05<11:34:33, 5.72s/it]
83%|████████▎ | 36650/43939 [59:07:11<11:33:09, 5.71s/it]
{'loss': 0.6583, 'grad_norm': 1.5948106050491333, 'learning_rate': 4.07649981862582e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.42, 'epoch': 3.34}
83%|████████▎ | 36650/43939 [59:07:11<11:33:09, 5.71s/it]
83%|████████▎ | 36651/43939 [59:07:17<11:33:04, 5.71s/it]
83%|████████▎ | 36652/43939 [59:07:22<11:31:59, 5.70s/it]
83%|████████▎ | 36653/43939 [59:07:28<11:33:42, 5.71s/it]
83%|████████▎ | 36654/43939 [59:07:34<11:31:49, 5.70s/it]
83%|████████▎ | 36655/43939 [59:07:39<11:30:04, 5.68s/it]
83%|████████▎ | 36656/43939 [59:07:45<11:31:00, 5.69s/it]
83%|████████▎ | 36657/43939 [59:07:51<11:30:23, 5.69s/it]
83%|████████▎ | 36658/43939 [59:07:57<11:32:27, 5.71s/it]
83%|████████▎ | 36659/43939 [59:08:02<11:34:03, 5.72s/it]
83%|████████▎ | 36660/43939 [59:08:08<11:35:05, 5.73s/it]
83%|████████▎ | 36661/43939 [59:08:14<11:34:25, 5.72s/it]
83%|████████▎ | 36662/43939 [59:08:19<11:32:42, 5.71s/it]
83%|████████▎ | 36663/43939 [59:08:25<11:32:57, 5.71s/it]
83%|████████▎ | 36664/43939 [59:08:31<11:32:02, 5.71s/it]
83%|████████▎ | 36665/43939 [59:08:37<11:39:57, 5.77s/it]
83%|████████▎ | 36666/43939 [59:08:43<11:38:33, 5.76s/it]
83%|████████▎ | 36667/43939 [59:08:48<11:36:14, 5.74s/it]
83%|████████▎ | 36668/43939 [59:08:54<11:33:59, 5.73s/it]
83%|████████▎ | 36669/43939 [59:09:00<11:33:15, 5.72s/it]
83%|████████▎ | 36670/43939 [59:09:05<11:31:51, 5.71s/it]
83%|████████▎ | 36671/43939 [59:09:11<11:30:32, 5.70s/it]
83%|████████▎ | 36672/43939 [59:09:17<11:30:09, 5.70s/it]
83%|████████▎ | 36673/43939 [59:09:22<11:33:00, 5.72s/it]
83%|████████▎ | 36674/43939 [59:09:28<11:30:55, 5.71s/it]
83%|████████▎ | 36675/43939 [59:09:34<11:30:19, 5.70s/it]
83%|████████▎ | 36676/43939 [59:09:40<11:30:01, 5.70s/it]
83%|████████▎ | 36677/43939 [59:09:45<11:32:20, 5.72s/it]
83%|████████▎ | 36678/43939 [59:09:51<11:31:16, 5.71s/it]
83%|████████▎ | 36679/43939 [59:09:57<11:30:50, 5.71s/it]
83%|████████▎ | 36680/43939 [59:10:02<11:30:09, 5.70s/it]
83%|████████▎ | 36681/43939 [59:10:08<11:32:07, 5.72s/it]
83%|████████▎ | 36682/43939 [59:10:14<11:32:02, 5.72s/it]
83%|████████▎ | 36683/43939 [59:10:20<11:33:02, 5.73s/it]
83%|████████▎ | 36684/43939 [59:10:25<11:31:39, 5.72s/it]
83%|████████▎ | 36685/43939 [59:10:31<11:30:01, 5.71s/it]
83%|████████▎ | 36686/43939 [59:10:37<11:37:20, 5.77s/it]
83%|████████▎ | 36687/43939 [59:10:43<11:33:03, 5.73s/it]
83%|████████▎ | 36688/43939 [59:10:48<11:31:36, 5.72s/it]
83%|████████▎ | 36689/43939 [59:10:54<11:30:09, 5.71s/it]
84%|████████▎ | 36690/43939 [59:11:00<11:30:02, 5.71s/it]
84%|████████▎ | 36691/43939 [59:11:05<11:31:19, 5.72s/it]
84%|████████▎ | 36692/43939 [59:11:11<11:32:14, 5.73s/it]
84%|████████▎ | 36693/43939 [59:11:17<11:31:30, 5.73s/it]
84%|████████▎ | 36694/43939 [59:11:23<11:29:52, 5.71s/it]
84%|████████▎ | 36695/43939 [59:11:28<11:29:34, 5.71s/it]
84%|████████▎ | 36696/43939 [59:11:34<11:29:15, 5.71s/it]
84%|████████▎ | 36697/43939 [59:11:40<11:29:19, 5.71s/it]
84%|████████▎ | 36698/43939 [59:11:45<11:30:56, 5.73s/it]
84%|████████▎ | 36699/43939 [59:11:51<11:31:51, 5.73s/it]
84%|████████▎ | 36700/43939 [59:11:57<11:30:08, 5.72s/it]
{'loss': 0.6687, 'grad_norm': 1.573398232460022, 'learning_rate': 4.02231765184502e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 513.38, 'epoch': 3.34}
84%|████████▎ | 36700/43939 [59:11:57<11:30:08, 5.72s/it]
84%|████████▎ | 36701/43939 [59:12:03<11:29:25, 5.71s/it]
84%|████████▎ | 36702/43939 [59:12:08<11:27:01, 5.70s/it]
84%|████████▎ | 36703/43939 [59:12:14<11:27:03, 5.70s/it]
84%|████████▎ | 36704/43939 [59:12:20<11:27:33, 5.70s/it]
84%|████████▎ | 36705/43939 [59:12:25<11:28:23, 5.71s/it]
84%|████████▎ | 36706/43939 [59:12:31<11:26:42, 5.70s/it]
84%|████████▎ | 36707/43939 [59:12:37<11:26:59, 5.70s/it]
84%|████████▎ | 36708/43939 [59:12:43<11:33:19, 5.75s/it]
84%|████████▎ | 36709/43939 [59:12:48<11:30:28, 5.73s/it]
84%|████████▎ | 36710/43939 [59:12:54<11:28:59, 5.72s/it]
84%|████████▎ | 36711/43939 [59:13:00<11:27:31, 5.71s/it]
84%|████████▎ | 36712/43939 [59:13:05<11:26:12, 5.70s/it]
84%|████████▎ | 36713/43939 [59:13:11<11:27:17, 5.71s/it]
84%|████████▎ | 36714/43939 [59:13:17<11:26:19, 5.70s/it]
84%|████████▎ | 36715/43939 [59:13:22<11:26:19, 5.70s/it]
84%|████████▎ | 36716/43939 [59:13:28<11:25:41, 5.70s/it]
84%|████████▎ | 36717/43939 [59:13:34<11:25:13, 5.69s/it]
84%|████████▎ | 36718/43939 [59:13:40<11:25:58, 5.70s/it]
84%|████████▎ | 36719/43939 [59:13:45<11:25:27, 5.70s/it]
84%|████████▎ | 36720/43939 [59:13:51<11:25:49, 5.70s/it]
84%|████████▎ | 36721/43939 [59:13:57<11:26:02, 5.70s/it]
84%|████████▎ | 36722/43939 [59:14:02<11:24:33, 5.69s/it]
84%|████████▎ | 36723/43939 [59:14:08<11:24:17, 5.69s/it]
84%|████████▎ | 36724/43939 [59:14:14<11:24:37, 5.69s/it]
84%|████████▎ | 36725/43939 [59:14:19<11:24:29, 5.69s/it]
84%|████████▎ | 36726/43939 [59:14:25<11:25:14, 5.70s/it]
84%|████████▎ | 36727/43939 [59:14:31<11:24:48, 5.70s/it]
84%|████████▎ | 36728/43939 [59:14:37<11:26:27, 5.71s/it]
84%|████████▎ | 36729/43939 [59:14:42<11:35:15, 5.79s/it]
84%|████████▎ | 36730/43939 [59:14:48<11:32:44, 5.77s/it]
84%|████████▎ | 36731/43939 [59:14:54<11:30:31, 5.75s/it]
84%|████████▎ | 36732/43939 [59:15:00<11:27:32, 5.72s/it]
84%|████████▎ | 36733/43939 [59:15:05<11:26:54, 5.72s/it]
84%|████████▎ | 36734/43939 [59:15:11<11:25:32, 5.71s/it]
84%|████████▎ | 36735/43939 [59:15:17<11:23:53, 5.70s/it]
84%|████████▎ | 36736/43939 [59:15:22<11:24:18, 5.70s/it]
84%|████████▎ | 36737/43939 [59:15:28<11:22:33, 5.69s/it]
84%|████████▎ | 36738/43939 [59:15:34<11:22:40, 5.69s/it]
84%|████████▎ | 36739/43939 [59:15:39<11:23:13, 5.69s/it]
84%|████████▎ | 36740/43939 [59:15:45<11:25:02, 5.71s/it]
84%|████████▎ | 36741/43939 [59:15:51<11:24:57, 5.71s/it]
84%|████████▎ | 36742/43939 [59:15:57<11:24:42, 5.71s/it]
84%|████████▎ | 36743/43939 [59:16:02<11:23:53, 5.70s/it]
84%|████████▎ | 36744/43939 [59:16:08<11:23:52, 5.70s/it]
84%|████████▎ | 36745/43939 [59:16:14<11:23:08, 5.70s/it]
84%|████████▎ | 36746/43939 [59:16:19<11:23:42, 5.70s/it]
84%|████████▎ | 36747/43939 [59:16:25<11:23:17, 5.70s/it]
84%|████████▎ | 36748/43939 [59:16:31<11:22:32, 5.70s/it]
84%|████████▎ | 36749/43939 [59:16:36<11:22:27, 5.70s/it]
84%|████████▎ | 36750/43939 [59:16:42<11:22:36, 5.70s/it]
{'loss': 0.6481, 'grad_norm': 1.5159857273101807, 'learning_rate': 3.968466457300138e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 459.02, 'epoch': 3.35}
84%|████████▎ | 36750/43939 [59:16:42<11:22:36, 5.70s/it]
84%|████████▎ | 36751/43939 [59:16:48<11:30:34, 5.76s/it]
84%|████████▎ | 36752/43939 [59:16:54<11:28:07, 5.74s/it]
84%|████████▎ | 36753/43939 [59:16:59<11:26:37, 5.73s/it]
84%|████████▎ | 36754/43939 [59:17:05<11:24:04, 5.71s/it]
84%|████████▎ | 36755/43939 [59:17:11<11:22:12, 5.70s/it]
84%|████████▎ | 36756/43939 [59:17:17<11:24:19, 5.72s/it]
84%|████████▎ | 36757/43939 [59:17:22<11:23:40, 5.71s/it]
84%|████████▎ | 36758/43939 [59:17:28<11:23:39, 5.71s/it]
84%|████████▎ | 36759/43939 [59:17:34<11:23:50, 5.71s/it]
84%|████████▎ | 36760/43939 [59:17:39<11:22:20, 5.70s/it]
84%|████████▎ | 36761/43939 [59:17:45<11:23:02, 5.71s/it]
84%|████████▎ | 36762/43939 [59:17:51<11:23:20, 5.71s/it]
84%|████████▎ | 36763/43939 [59:17:56<11:21:18, 5.70s/it]
84%|████████▎ | 36764/43939 [59:18:02<11:23:05, 5.71s/it]
84%|████████▎ | 36765/43939 [59:18:08<11:21:53, 5.70s/it]
84%|████████▎ | 36766/43939 [59:18:14<11:21:32, 5.70s/it]
84%|████████▎ | 36767/43939 [59:18:19<11:20:43, 5.69s/it]
84%|████████▎ | 36768/43939 [59:18:25<11:21:53, 5.71s/it]
84%|████████▎ | 36769/43939 [59:18:31<11:21:12, 5.70s/it]
84%|████████▎ | 36770/43939 [59:18:36<11:20:43, 5.70s/it]
84%|████████▎ | 36771/43939 [59:18:42<11:29:50, 5.77s/it]
84%|████████▎ | 36772/43939 [59:18:48<11:26:54, 5.75s/it]
84%|████████▎ | 36773/43939 [59:18:54<11:26:04, 5.74s/it]
84%|████████▎ | 36774/43939 [59:18:59<11:23:07, 5.72s/it]
84%|████████▎ | 36775/43939 [59:19:05<11:22:16, 5.71s/it]
84%|████████▎ | 36776/43939 [59:19:11<11:23:14, 5.72s/it]
84%|████████▎ | 36777/43939 [59:19:17<11:21:41, 5.71s/it]
84%|████████▎ | 36778/43939 [59:19:22<11:23:04, 5.72s/it]
84%|████████▎ | 36779/43939 [59:19:28<11:21:13, 5.71s/it]
84%|████████▎ | 36780/43939 [59:19:34<11:20:04, 5.70s/it]
84%|████████▎ | 36781/43939 [59:19:39<11:19:49, 5.70s/it]
84%|████████▎ | 36782/43939 [59:19:45<11:20:49, 5.71s/it]
84%|████████▎ | 36783/43939 [59:19:51<11:21:52, 5.72s/it]
84%|████████▎ | 36784/43939 [59:19:57<11:21:36, 5.72s/it]
84%|████████▎ | 36785/43939 [59:20:02<11:23:18, 5.73s/it]
84%|████████▎ | 36786/43939 [59:20:08<11:22:07, 5.72s/it]
84%|████████▎ | 36787/43939 [59:20:14<11:20:59, 5.71s/it]
84%|████████▎ | 36788/43939 [59:20:19<11:21:24, 5.72s/it]
84%|████████▎ | 36789/43939 [59:20:25<11:20:16, 5.71s/it]
84%|████████▎ | 36790/43939 [59:20:31<11:19:13, 5.70s/it]
84%|████████▎ | 36791/43939 [59:20:36<11:18:47, 5.70s/it]
84%|████████▎ | 36792/43939 [59:20:42<11:19:39, 5.71s/it]
84%|████████▎ | 36793/43939 [59:20:48<11:29:25, 5.79s/it]
84%|████████▎ | 36794/43939 [59:20:54<11:27:12, 5.77s/it]
84%|████████▎ | 36795/43939 [59:21:00<11:25:11, 5.75s/it]
84%|████████▎ | 36796/43939 [59:21:05<11:22:22, 5.73s/it]
84%|████████▎ | 36797/43939 [59:21:11<11:21:48, 5.73s/it]
84%|████████▎ | 36798/43939 [59:21:17<11:19:34, 5.71s/it]
84%|████████▍ | 36799/43939 [59:21:22<11:19:06, 5.71s/it]
84%|████████▍ | 36800/43939 [59:21:28<11:17:53, 5.70s/it]
{'loss': 0.6561, 'grad_norm': 1.6860300302505493, 'learning_rate': 3.914947084620316e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.08, 'epoch': 3.35}
84%|████████▍ | 36800/43939 [59:21:28<11:17:53, 5.70s/it]
84%|████████▍ | 36801/43939 [59:21:34<11:16:40, 5.69s/it]
84%|████████▍ | 36802/43939 [59:21:39<11:17:28, 5.70s/it]
84%|████████▍ | 36803/43939 [59:21:45<11:17:56, 5.70s/it]
84%|████████▍ | 36804/43939 [59:21:51<11:17:41, 5.70s/it]
84%|████████▍ | 36805/43939 [59:21:57<11:18:20, 5.71s/it]
84%|████████▍ | 36806/43939 [59:22:02<11:17:46, 5.70s/it]
84%|████████▍ | 36807/43939 [59:22:08<11:19:07, 5.71s/it]
84%|████████▍ | 36808/43939 [59:22:14<11:17:56, 5.70s/it]
84%|████████▍ | 36809/43939 [59:22:19<11:19:13, 5.72s/it]
84%|████████▍ | 36810/43939 [59:22:25<11:18:37, 5.71s/it]
84%|████████▍ | 36811/43939 [59:22:31<11:18:06, 5.71s/it]
84%|████████▍ | 36812/43939 [59:22:37<11:16:40, 5.70s/it]
84%|████████▍ | 36813/43939 [59:22:42<11:16:57, 5.70s/it]
84%|████████▍ | 36814/43939 [59:22:48<11:24:42, 5.77s/it]
84%|████████▍ | 36815/43939 [59:22:54<11:24:23, 5.76s/it]
84%|████████▍ | 36816/43939 [59:23:00<11:22:23, 5.75s/it]
84%|████████▍ | 36817/43939 [59:23:05<11:20:10, 5.73s/it]
84%|████████▍ | 36818/43939 [59:23:11<11:19:23, 5.72s/it]
84%|████████▍ | 36819/43939 [59:23:17<11:17:58, 5.71s/it]
84%|████████▍ | 36820/43939 [59:23:22<11:17:43, 5.71s/it]
84%|████████▍ | 36821/43939 [59:23:28<11:16:46, 5.70s/it]
84%|████████▍ | 36822/43939 [59:23:34<11:16:05, 5.70s/it]
84%|████████▍ | 36823/43939 [59:23:39<11:16:49, 5.71s/it]
84%|████████▍ | 36824/43939 [59:23:45<11:17:06, 5.71s/it]
84%|████████▍ | 36825/43939 [59:23:51<11:18:32, 5.72s/it]
84%|████████▍ | 36826/43939 [59:23:57<11:18:24, 5.72s/it]
84%|████████▍ | 36827/43939 [59:24:02<11:19:50, 5.74s/it]
84%|████████▍ | 36828/43939 [59:24:08<11:18:10, 5.72s/it]
84%|████████▍ | 36829/43939 [59:24:14<11:17:26, 5.72s/it]
84%|████████▍ | 36830/43939 [59:24:20<11:16:01, 5.71s/it]
84%|████████▍ | 36831/43939 [59:24:25<11:17:08, 5.72s/it]
84%|████████▍ | 36832/43939 [59:24:31<11:17:20, 5.72s/it]
84%|████████▍ | 36833/43939 [59:24:37<11:18:23, 5.73s/it]
84%|████████▍ | 36834/43939 [59:24:42<11:17:43, 5.72s/it]
84%|████████▍ | 36835/43939 [59:24:48<11:24:16, 5.78s/it]
84%|████████▍ | 36836/43939 [59:24:54<11:22:27, 5.76s/it]
84%|████████▍ | 36837/43939 [59:25:00<11:20:28, 5.75s/it]
84%|████████▍ | 36838/43939 [59:25:05<11:17:20, 5.72s/it]
84%|████████▍ | 36839/43939 [59:25:11<11:16:49, 5.72s/it]
84%|████████▍ | 36840/43939 [59:25:17<11:15:25, 5.71s/it]
84%|████████▍ | 36841/43939 [59:25:23<11:16:30, 5.72s/it]
84%|████████▍ | 36842/43939 [59:25:28<11:14:47, 5.70s/it]
84%|████████▍ | 36843/43939 [59:25:34<11:14:15, 5.70s/it]
84%|████████▍ | 36844/43939 [59:25:40<11:13:23, 5.69s/it]
84%|████████▍ | 36845/43939 [59:25:45<11:13:18, 5.69s/it]
84%|████████▍ | 36846/43939 [59:25:51<11:13:05, 5.69s/it]
84%|████████▍ | 36847/43939 [59:25:57<11:12:52, 5.69s/it]
84%|████████▍ | 36848/43939 [59:26:02<11:15:12, 5.71s/it]
84%|████████▍ | 36849/43939 [59:26:08<11:14:21, 5.71s/it]
84%|████████▍ | 36850/43939 [59:26:14<11:15:08, 5.71s/it]
{'loss': 0.6572, 'grad_norm': 1.3102531433105469, 'learning_rate': 3.861760378199431e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.49, 'epoch': 3.35}
84%|████████▍ | 36850/43939 [59:26:14<11:15:08, 5.71s/it]
84%|████████▍ | 36851/43939 [59:26:20<11:14:45, 5.71s/it]
84%|████████▍ | 36852/43939 [59:26:25<11:14:41, 5.71s/it]
84%|████████▍ | 36853/43939 [59:26:31<11:14:20, 5.71s/it]
84%|████████▍ | 36854/43939 [59:26:37<11:13:59, 5.71s/it]
84%|████████▍ | 36855/43939 [59:26:42<11:13:38, 5.71s/it]
84%|████████▍ | 36856/43939 [59:26:48<11:14:53, 5.72s/it]
84%|████████▍ | 36857/43939 [59:26:54<11:22:18, 5.78s/it]
84%|████████▍ | 36858/43939 [59:27:00<11:20:20, 5.76s/it]
84%|████████▍ | 36859/43939 [59:27:06<11:18:24, 5.75s/it]
84%|████████▍ | 36860/43939 [59:27:11<11:16:07, 5.73s/it]
84%|████████▍ | 36861/43939 [59:27:17<11:17:23, 5.74s/it]
84%|████████▍ | 36862/43939 [59:27:23<11:14:59, 5.72s/it]
84%|████████▍ | 36863/43939 [59:27:28<11:13:49, 5.71s/it]
84%|████████▍ | 36864/43939 [59:27:34<11:13:36, 5.71s/it]
84%|████████▍ | 36865/43939 [59:27:40<11:10:56, 5.69s/it]
84%|████████▍ | 36866/43939 [59:27:45<11:11:21, 5.70s/it]
84%|████████▍ | 36867/43939 [59:27:51<11:11:36, 5.70s/it]
84%|████████▍ | 36868/43939 [59:27:57<11:12:43, 5.71s/it]
84%|████████▍ | 36869/43939 [59:28:03<11:12:44, 5.71s/it]
84%|████████▍ | 36870/43939 [59:28:08<11:13:12, 5.71s/it]
84%|████████▍ | 36871/43939 [59:28:14<11:13:56, 5.72s/it]
84%|████████▍ | 36872/43939 [59:28:20<11:14:27, 5.73s/it]
84%|████████▍ | 36873/43939 [59:28:25<11:12:54, 5.71s/it]
84%|████████▍ | 36874/43939 [59:28:31<11:11:44, 5.70s/it]
84%|████████▍ | 36875/43939 [59:28:37<11:11:45, 5.71s/it]
84%|████████▍ | 36876/43939 [59:28:43<11:10:41, 5.70s/it]
84%|████████▍ | 36877/43939 [59:28:48<11:10:57, 5.70s/it]
84%|████████▍ | 36878/43939 [59:28:54<11:19:53, 5.78s/it]
84%|████████▍ | 36879/43939 [59:29:00<11:16:52, 5.75s/it]
84%|████████▍ | 36880/43939 [59:29:06<11:16:19, 5.75s/it]
84%|████████▍ | 36881/43939 [59:29:11<11:14:03, 5.73s/it]
84%|████████▍ | 36882/43939 [59:29:17<11:12:51, 5.72s/it]
84%|████████▍ | 36883/43939 [59:29:23<11:12:09, 5.72s/it]
84%|████████▍ | 36884/43939 [59:29:28<11:12:11, 5.72s/it]
84%|████████▍ | 36885/43939 [59:29:34<11:13:02, 5.72s/it]
84%|████████▍ | 36886/43939 [59:29:40<11:10:58, 5.71s/it]
84%|████████▍ | 36887/43939 [59:29:46<11:09:46, 5.70s/it]
84%|████████▍ | 36888/43939 [59:29:51<11:10:57, 5.71s/it]
84%|████████▍ | 36889/43939 [59:29:57<11:10:15, 5.70s/it]
84%|████████▍ | 36890/43939 [59:30:03<11:09:39, 5.70s/it]
84%|████████▍ | 36891/43939 [59:30:08<11:09:52, 5.70s/it]
84%|████████▍ | 36892/43939 [59:30:14<11:09:35, 5.70s/it]
84%|████████▍ | 36893/43939 [59:30:20<11:10:15, 5.71s/it]
84%|████████▍ | 36894/43939 [59:30:25<11:08:41, 5.70s/it]
84%|████████▍ | 36895/43939 [59:30:31<11:09:32, 5.70s/it]
84%|████████▍ | 36896/43939 [59:30:37<11:09:45, 5.71s/it]
84%|████████▍ | 36897/43939 [59:30:43<11:11:42, 5.72s/it]
84%|████████▍ | 36898/43939 [59:30:48<11:12:11, 5.73s/it]
84%|████████▍ | 36899/43939 [59:30:54<11:11:35, 5.72s/it]
84%|████████▍ | 36900/43939 [59:31:00<11:19:26, 5.79s/it]
{'loss': 0.6635, 'grad_norm': 1.665374994277954, 'learning_rate': 3.8089071771827607e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.35, 'epoch': 3.36}
84%|████████▍ | 36900/43939 [59:31:00<11:19:26, 5.79s/it]
84%|████████▍ | 36901/43939 [59:31:06<11:16:15, 5.77s/it]
84%|████████▍ | 36902/43939 [59:31:11<11:14:00, 5.75s/it]
84%|████████▍ | 36903/43939 [59:31:17<11:11:53, 5.73s/it]
84%|████████▍ | 36904/43939 [59:31:23<11:12:09, 5.73s/it]
84%|████████▍ | 36905/43939 [59:31:29<11:11:56, 5.73s/it]
84%|████████▍ | 36906/43939 [59:31:34<11:10:28, 5.72s/it]
84%|████████▍ | 36907/43939 [59:31:40<11:09:59, 5.72s/it]
84%|████████▍ | 36908/43939 [59:31:46<11:10:39, 5.72s/it]
84%|████████▍ | 36909/43939 [59:31:51<11:11:10, 5.73s/it]
84%|████████▍ | 36910/43939 [59:31:57<11:11:36, 5.73s/it]
84%|████████▍ | 36911/43939 [59:32:03<11:10:42, 5.73s/it]
84%|████████▍ | 36912/43939 [59:32:09<11:10:57, 5.73s/it]
84%|████████▍ | 36913/43939 [59:32:14<11:09:01, 5.71s/it]
84%|████████▍ | 36914/43939 [59:32:20<11:09:26, 5.72s/it]
84%|████████▍ | 36915/43939 [59:32:26<11:08:38, 5.71s/it]
84%|████████▍ | 36916/43939 [59:32:31<11:08:15, 5.71s/it]
84%|████████▍ | 36917/43939 [59:32:37<11:07:49, 5.71s/it]
84%|████████▍ | 36918/43939 [59:32:43<11:06:09, 5.69s/it]
84%|████████▍ | 36919/43939 [59:32:49<11:05:53, 5.69s/it]
84%|████████▍ | 36920/43939 [59:32:54<11:05:41, 5.69s/it]
84%|████████▍ | 36921/43939 [59:33:00<11:13:58, 5.76s/it]
84%|████████▍ | 36922/43939 [59:33:06<11:11:21, 5.74s/it]
84%|████████▍ | 36923/43939 [59:33:12<11:11:23, 5.74s/it]
84%|████████▍ | 36924/43939 [59:33:17<11:11:06, 5.74s/it]
84%|████████▍ | 36925/43939 [59:33:23<11:10:12, 5.73s/it]
84%|████████▍ | 36926/43939 [59:33:29<11:09:39, 5.73s/it]
84%|████████▍ | 36927/43939 [59:33:34<11:08:11, 5.72s/it]
84%|████████▍ | 36928/43939 [59:33:40<11:06:59, 5.71s/it]
84%|████████▍ | 36929/43939 [59:33:46<11:07:13, 5.71s/it]
84%|████████▍ | 36930/43939 [59:33:52<11:06:51, 5.71s/it]
84%|████████▍ | 36931/43939 [59:33:57<11:06:17, 5.70s/it]
84%|████████▍ | 36932/43939 [59:34:03<11:06:02, 5.70s/it]
84%|████████▍ | 36933/43939 [59:34:09<11:04:42, 5.69s/it]
84%|████████▍ | 36934/43939 [59:34:14<11:03:54, 5.69s/it]
84%|████████▍ | 36935/43939 [59:34:20<11:04:38, 5.69s/it]
84%|████████▍ | 36936/43939 [59:34:26<11:04:07, 5.69s/it]
84%|████████▍ | 36937/43939 [59:34:31<11:03:18, 5.68s/it]
84%|████████▍ | 36938/43939 [59:34:37<11:06:25, 5.71s/it]
84%|████████▍ | 36939/43939 [59:34:43<11:06:55, 5.72s/it]
84%|████████▍ | 36940/43939 [59:34:49<11:05:10, 5.70s/it]
84%|████████▍ | 36941/43939 [59:34:54<11:07:34, 5.72s/it]
84%|████████▍ | 36942/43939 [59:35:00<11:12:50, 5.77s/it]
84%|████████▍ | 36943/43939 [59:35:06<11:11:17, 5.76s/it]
84%|████████▍ | 36944/43939 [59:35:12<11:08:05, 5.73s/it]
84%|████████▍ | 36945/43939 [59:35:17<11:06:51, 5.72s/it]
84%|████████▍ | 36946/43939 [59:35:23<11:07:07, 5.72s/it]
84%|████████▍ | 36947/43939 [59:35:29<11:07:50, 5.73s/it]
84%|████████▍ | 36948/43939 [59:35:34<11:05:44, 5.71s/it]
84%|████████▍ | 36949/43939 [59:35:40<11:04:28, 5.70s/it]
84%|████████▍ | 36950/43939 [59:35:46<11:02:53, 5.69s/it]
{'loss': 0.6745, 'grad_norm': 1.8100502490997314, 'learning_rate': 3.7563883154537545e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.77, 'epoch': 3.36}
84%|████████▍ | 36950/43939 [59:35:46<11:02:53, 5.69s/it]
84%|████████▍ | 36951/43939 [59:35:51<11:03:08, 5.69s/it]
84%|████████▍ | 36952/43939 [59:35:57<11:03:40, 5.70s/it]
84%|████████▍ | 36953/43939 [59:36:03<11:03:07, 5.70s/it]
84%|████████▍ | 36954/43939 [59:36:09<11:03:44, 5.70s/it]
84%|████████▍ | 36955/43939 [59:36:14<11:03:17, 5.70s/it]
84%|████████▍ | 36956/43939 [59:36:20<11:04:12, 5.71s/it]
84%|████████▍ | 36957/43939 [59:36:26<11:04:28, 5.71s/it]
84%|████████▍ | 36958/43939 [59:36:31<11:05:24, 5.72s/it]
84%|████████▍ | 36959/43939 [59:36:37<11:03:36, 5.70s/it]
84%|████████▍ | 36960/43939 [59:36:43<11:03:06, 5.70s/it]
84%|████████▍ | 36961/43939 [59:36:49<11:03:58, 5.71s/it]
84%|████████▍ | 36962/43939 [59:36:54<11:03:31, 5.71s/it]
84%|████████▍ | 36963/43939 [59:37:00<11:04:33, 5.72s/it]
84%|████████▍ | 36964/43939 [59:37:06<11:12:34, 5.79s/it]
84%|████████▍ | 36965/43939 [59:37:12<11:09:44, 5.76s/it]
84%|████████▍ | 36966/43939 [59:37:17<11:09:44, 5.76s/it]
84%|████████▍ | 36967/43939 [59:37:23<11:06:55, 5.74s/it]
84%|████████▍ | 36968/43939 [59:37:29<11:04:40, 5.72s/it]
84%|████████▍ | 36969/43939 [59:37:34<11:03:05, 5.71s/it]
84%|████████▍ | 36970/43939 [59:37:40<11:01:54, 5.70s/it]
84%|████████▍ | 36971/43939 [59:37:46<11:02:45, 5.71s/it]
84%|████████▍ | 36972/43939 [59:37:52<11:03:00, 5.71s/it]
84%|████████▍ | 36973/43939 [59:37:57<11:01:07, 5.69s/it]
84%|████████▍ | 36974/43939 [59:38:03<11:04:39, 5.73s/it]
84%|████████▍ | 36975/43939 [59:38:09<11:02:21, 5.71s/it]
84%|████████▍ | 36976/43939 [59:38:14<11:01:15, 5.70s/it]
84%|████████▍ | 36977/43939 [59:38:20<11:01:31, 5.70s/it]
84%|████████▍ | 36978/43939 [59:38:26<11:01:43, 5.70s/it]
84%|████████▍ | 36979/43939 [59:38:31<11:01:28, 5.70s/it]
84%|████████▍ | 36980/43939 [59:38:37<11:01:06, 5.70s/it]
84%|████████▍ | 36981/43939 [59:38:43<11:01:49, 5.71s/it]
84%|████████▍ | 36982/43939 [59:38:49<11:00:38, 5.70s/it]
84%|████████▍ | 36983/43939 [59:38:54<11:01:48, 5.71s/it]
84%|████████▍ | 36984/43939 [59:39:00<11:01:01, 5.70s/it]
84%|████████▍ | 36985/43939 [59:39:06<11:01:11, 5.70s/it]
84%|████████▍ | 36986/43939 [59:39:12<11:09:45, 5.78s/it]
84%|████████▍ | 36987/43939 [59:39:17<11:06:22, 5.75s/it]
84%|████████▍ | 36988/43939 [59:39:23<11:04:36, 5.74s/it]
84%|████████▍ | 36989/43939 [59:39:29<11:03:20, 5.73s/it]
84%|████████▍ | 36990/43939 [59:39:34<11:02:08, 5.72s/it]
84%|████████▍ | 36991/43939 [59:39:40<11:01:25, 5.71s/it]
84%|████████▍ | 36992/43939 [59:39:46<11:02:06, 5.72s/it]
84%|████████▍ | 36993/43939 [59:39:52<11:01:41, 5.72s/it]
84%|████████▍ | 36994/43939 [59:39:57<11:02:50, 5.73s/it]
84%|████████▍ | 36995/43939 [59:40:03<11:02:26, 5.72s/it]
84%|████████▍ | 36996/43939 [59:40:09<11:02:02, 5.72s/it]
84%|████████▍ | 36997/43939 [59:40:15<11:02:43, 5.73s/it]
84%|████████▍ | 36998/43939 [59:40:20<11:03:10, 5.73s/it]
84%|████████▍ | 36999/43939 [59:40:26<11:02:40, 5.73s/it]
84%|████████▍ | 37000/43939 [59:40:32<11:01:06, 5.72s/it]
{'loss': 0.6504, 'grad_norm': 2.081054449081421, 'learning_rate': 3.7042046216208686e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 450.25, 'epoch': 3.37}
84%|████████▍ | 37000/43939 [59:40:32<11:01:06, 5.72s/it]
84%|████████▍ | 37001/43939 [59:40:37<11:01:44, 5.72s/it]
84%|████████▍ | 37002/43939 [59:40:43<11:00:20, 5.71s/it]
84%|████████▍ | 37003/43939 [59:40:49<10:59:53, 5.71s/it]
84%|████████▍ | 37004/43939 [59:40:54<10:59:02, 5.70s/it]
84%|████████▍ | 37005/43939 [59:41:00<10:59:08, 5.70s/it]
84%|████████▍ | 37006/43939 [59:41:06<10:59:24, 5.71s/it]
84%|████████▍ | 37007/43939 [59:41:12<11:08:01, 5.78s/it]
84%|████████▍ | 37008/43939 [59:41:18<11:06:23, 5.77s/it]
84%|████████▍ | 37009/43939 [59:41:23<11:05:57, 5.77s/it]
84%|████████▍ | 37010/43939 [59:41:29<11:02:33, 5.74s/it]
84%|████████▍ | 37011/43939 [59:41:35<11:00:31, 5.72s/it]
84%|████████▍ | 37012/43939 [59:41:40<10:59:55, 5.72s/it]
84%|████████▍ | 37013/43939 [59:41:46<10:58:15, 5.70s/it]
84%|████████▍ | 37014/43939 [59:41:52<10:58:11, 5.70s/it]
84%|████████▍ | 37015/43939 [59:41:58<10:58:16, 5.70s/it]
84%|████████▍ | 37016/43939 [59:42:03<10:58:55, 5.71s/it]
84%|████████▍ | 37017/43939 [59:42:09<11:00:30, 5.73s/it]
84%|████████▍ | 37018/43939 [59:42:15<11:01:26, 5.73s/it]
84%|████████▍ | 37019/43939 [59:42:20<11:00:54, 5.73s/it]
84%|████████▍ | 37020/43939 [59:42:26<10:59:36, 5.72s/it]
84%|████████▍ | 37021/43939 [59:42:32<10:58:09, 5.71s/it]
84%|████████▍ | 37022/43939 [59:42:38<10:57:50, 5.71s/it]
84%|████████▍ | 37023/43939 [59:42:43<10:58:54, 5.72s/it]
84%|████████▍ | 37024/43939 [59:42:49<10:59:03, 5.72s/it]
84%|████████▍ | 37025/43939 [59:42:55<10:59:44, 5.73s/it]
84%|████████▍ | 37026/43939 [59:43:00<10:59:31, 5.72s/it]
84%|████████▍ | 37027/43939 [59:43:06<10:57:05, 5.70s/it]
84%|████████▍ | 37028/43939 [59:43:12<10:58:39, 5.72s/it]
84%|████████▍ | 37029/43939 [59:43:18<11:05:37, 5.78s/it]
84%|████████▍ | 37030/43939 [59:43:23<11:02:13, 5.75s/it]
84%|████████▍ | 37031/43939 [59:43:29<10:59:04, 5.72s/it]
84%|████████▍ | 37032/43939 [59:43:35<10:58:38, 5.72s/it]
84%|████████▍ | 37033/43939 [59:43:41<10:58:25, 5.72s/it]
84%|████████▍ | 37034/43939 [59:43:46<10:57:40, 5.71s/it]
84%|████████▍ | 37035/43939 [59:43:52<10:57:45, 5.72s/it]
84%|████████▍ | 37036/43939 [59:43:58<10:57:06, 5.71s/it]
84%|████████▍ | 37037/43939 [59:44:03<10:56:49, 5.71s/it]
84%|████████▍ | 37038/43939 [59:44:09<10:57:51, 5.72s/it]
84%|████████▍ | 37039/43939 [59:44:15<10:58:23, 5.73s/it]
84%|████████▍ | 37040/43939 [59:44:21<11:01:11, 5.75s/it]
84%|████████▍ | 37041/43939 [59:44:26<10:58:50, 5.73s/it]
84%|████████▍ | 37042/43939 [59:44:32<10:57:47, 5.72s/it]
84%|████████▍ | 37043/43939 [59:44:38<10:59:32, 5.74s/it]
84%|████████▍ | 37044/43939 [59:44:44<10:59:03, 5.74s/it]
84%|████████▍ | 37045/43939 [59:44:49<10:57:49, 5.73s/it]
84%|████████▍ | 37046/43939 [59:44:55<10:56:48, 5.72s/it]
84%|████████▍ | 37047/43939 [59:45:01<10:55:55, 5.71s/it]
84%|████████▍ | 37048/43939 [59:45:06<10:54:29, 5.70s/it]
84%|████████▍ | 37049/43939 [59:45:12<10:54:59, 5.70s/it]
84%|████████▍ | 37050/43939 [59:45:18<11:03:57, 5.78s/it]
{'loss': 0.6571, 'grad_norm': 1.6421407461166382, 'learning_rate': 3.6523569190045126e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.34, 'epoch': 3.37}
84%|████████▍ | 37050/43939 [59:45:18<11:03:57, 5.78s/it]
84%|████████▍ | 37051/43939 [59:45:24<11:01:40, 5.76s/it]
84%|████████▍ | 37052/43939 [59:45:29<10:59:51, 5.75s/it]
84%|████████▍ | 37053/43939 [59:45:35<10:58:19, 5.74s/it]
84%|████████▍ | 37054/43939 [59:45:41<10:56:58, 5.73s/it]
84%|████████▍ | 37055/43939 [59:45:47<10:55:32, 5.71s/it]
84%|████████▍ | 37056/43939 [59:45:52<10:55:56, 5.72s/it]
84%|████████▍ | 37057/43939 [59:45:58<10:55:08, 5.71s/it]
84%|████████▍ | 37058/43939 [59:46:04<10:54:19, 5.71s/it]
84%|████████▍ | 37059/43939 [59:46:09<10:52:24, 5.69s/it]
84%|████████▍ | 37060/43939 [59:46:15<10:52:32, 5.69s/it]
84%|████████▍ | 37061/43939 [59:46:21<10:53:09, 5.70s/it]
84%|████████▍ | 37062/43939 [59:46:26<10:53:18, 5.70s/it]
84%|████████▍ | 37063/43939 [59:46:32<10:52:49, 5.70s/it]
84%|████████▍ | 37064/43939 [59:46:38<10:52:36, 5.70s/it]
84%|████████▍ | 37065/43939 [59:46:44<10:54:18, 5.71s/it]
84%|████████▍ | 37066/43939 [59:46:49<10:54:51, 5.72s/it]
84%|████████▍ | 37067/43939 [59:46:55<10:57:10, 5.74s/it]
84%|████████▍ | 37068/43939 [59:47:01<10:56:53, 5.74s/it]
84%|████████▍ | 37069/43939 [59:47:07<10:56:39, 5.74s/it]
84%|████████▍ | 37070/43939 [59:47:12<10:55:02, 5.72s/it]
84%|████████▍ | 37071/43939 [59:47:18<11:03:36, 5.80s/it]
84%|████████▍ | 37072/43939 [59:47:24<10:59:01, 5.76s/it]
84%|████████▍ | 37073/43939 [59:47:30<10:55:57, 5.73s/it]
84%|████████▍ | 37074/43939 [59:47:35<10:55:18, 5.73s/it]
84%|████████▍ | 37075/43939 [59:47:41<10:54:42, 5.72s/it]
84%|████████▍ | 37076/43939 [59:47:47<10:56:12, 5.74s/it]
84%|████████▍ | 37077/43939 [59:47:52<10:54:34, 5.72s/it]
84%|████████▍ | 37078/43939 [59:47:58<10:54:21, 5.72s/it]
84%|████████▍ | 37079/43939 [59:48:04<10:54:43, 5.73s/it]
84%|████████▍ | 37080/43939 [59:48:10<10:53:11, 5.71s/it]
84%|████████▍ | 37081/43939 [59:48:15<10:54:58, 5.73s/it]
84%|████████▍ | 37082/43939 [59:48:21<10:55:19, 5.73s/it]
84%|████████▍ | 37083/43939 [59:48:27<10:55:57, 5.74s/it]
84%|████████▍ | 37084/43939 [59:48:33<10:56:51, 5.75s/it]
84%|████████▍ | 37085/43939 [59:48:38<10:56:28, 5.75s/it]
84%|████████▍ | 37086/43939 [59:48:44<10:55:10, 5.74s/it]
84%|████████▍ | 37087/43939 [59:48:50<10:54:15, 5.73s/it]
84%|████████▍ | 37088/43939 [59:48:55<10:52:15, 5.71s/it]
84%|████████▍ | 37089/43939 [59:49:01<10:51:15, 5.70s/it]
84%|████████▍ | 37090/43939 [59:49:07<10:52:08, 5.71s/it]
84%|████████▍ | 37091/43939 [59:49:13<10:51:33, 5.71s/it]
84%|████████▍ | 37092/43939 [59:49:18<10:58:12, 5.77s/it]
84%|████████▍ | 37093/43939 [59:49:24<10:54:15, 5.73s/it]
84%|████████▍ | 37094/43939 [59:49:30<10:52:32, 5.72s/it]
84%|████████▍ | 37095/43939 [59:49:36<10:52:52, 5.72s/it]
84%|████████▍ | 37096/43939 [59:49:41<10:51:26, 5.71s/it]
84%|████████▍ | 37097/43939 [59:49:47<10:49:15, 5.69s/it]
84%|████████▍ | 37098/43939 [59:49:53<10:50:12, 5.70s/it]
84%|████████▍ | 37099/43939 [59:49:58<10:50:58, 5.71s/it]
84%|████████▍ | 37100/43939 [59:50:04<10:50:42, 5.71s/it]
{'loss': 0.6757, 'grad_norm': 1.7119728326797485, 'learning_rate': 3.6008460256240328e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.88, 'epoch': 3.38}
84%|████████▍ | 37100/43939 [59:50:04<10:50:42, 5.71s/it]
84%|████████▍ | 37101/43939 [59:50:10<10:50:47, 5.71s/it]
84%|████████▍ | 37102/43939 [59:50:15<10:51:07, 5.71s/it]
84%|████████▍ | 37103/43939 [59:50:21<10:49:37, 5.70s/it]
84%|████████▍ | 37104/43939 [59:50:27<10:49:12, 5.70s/it]
84%|████████▍ | 37105/43939 [59:50:33<10:50:44, 5.71s/it]
84%|████████▍ | 37106/43939 [59:50:38<10:51:15, 5.72s/it]
84%|████████▍ | 37107/43939 [59:50:44<10:50:16, 5.71s/it]
84%|████████▍ | 37108/43939 [59:50:50<10:50:34, 5.71s/it]
84%|████████▍ | 37109/43939 [59:50:55<10:49:47, 5.71s/it]
84%|████████▍ | 37110/43939 [59:51:01<10:48:42, 5.70s/it]
84%|████████▍ | 37111/43939 [59:51:07<10:48:46, 5.70s/it]
84%|████████▍ | 37112/43939 [59:51:13<10:47:55, 5.69s/it]
84%|████████▍ | 37113/43939 [59:51:18<10:47:57, 5.70s/it]
84%|████████▍ | 37114/43939 [59:51:24<10:56:54, 5.77s/it]
84%|████████▍ | 37115/43939 [59:51:30<10:53:50, 5.75s/it]
84%|████████▍ | 37116/43939 [59:51:36<10:51:55, 5.73s/it]
84%|████████▍ | 37117/43939 [59:51:41<10:50:49, 5.72s/it]
84%|████████▍ | 37118/43939 [59:51:47<10:50:14, 5.72s/it]
84%|████████▍ | 37119/43939 [59:51:53<10:50:25, 5.72s/it]
84%|████████▍ | 37120/43939 [59:51:58<10:49:33, 5.72s/it]
84%|████████▍ | 37121/43939 [59:52:04<10:49:11, 5.71s/it]
84%|████████▍ | 37122/43939 [59:52:10<10:50:28, 5.73s/it]
84%|████████▍ | 37123/43939 [59:52:16<10:49:09, 5.71s/it]
84%|████████▍ | 37124/43939 [59:52:21<10:50:46, 5.73s/it]
84%|████████▍ | 37125/43939 [59:52:27<10:49:20, 5.72s/it]
84%|████████▍ | 37126/43939 [59:52:33<10:48:46, 5.71s/it]
84%|████████▍ | 37127/43939 [59:52:38<10:48:07, 5.71s/it]
84%|████████▍ | 37128/43939 [59:52:44<10:46:52, 5.70s/it]
85%|████████▍ | 37129/43939 [59:52:50<10:46:04, 5.69s/it]
85%|████████▍ | 37130/43939 [59:52:55<10:45:34, 5.69s/it]
85%|████████▍ | 37131/43939 [59:53:01<10:45:44, 5.69s/it]
85%|████████▍ | 37132/43939 [59:53:07<10:45:40, 5.69s/it]
85%|████████▍ | 37133/43939 [59:53:13<10:45:49, 5.69s/it]
85%|████████▍ | 37134/43939 [59:53:18<10:47:10, 5.71s/it]
85%|████████▍ | 37135/43939 [59:53:24<10:47:32, 5.71s/it]
85%|████████▍ | 37136/43939 [59:53:30<10:54:50, 5.78s/it]
85%|████████▍ | 37137/43939 [59:53:36<10:51:25, 5.75s/it]
85%|████████▍ | 37138/43939 [59:53:41<10:51:00, 5.74s/it]
85%|████████▍ | 37139/43939 [59:53:47<10:48:56, 5.73s/it]
85%|████████▍ | 37140/43939 [59:53:53<10:48:55, 5.73s/it]
85%|████████▍ | 37141/43939 [59:53:58<10:48:38, 5.73s/it]
85%|████████▍ | 37142/43939 [59:54:04<10:46:38, 5.71s/it]
85%|████████▍ | 37143/43939 [59:54:10<10:45:25, 5.70s/it]
85%|████████▍ | 37144/43939 [59:54:15<10:44:45, 5.69s/it]
85%|████████▍ | 37145/43939 [59:54:21<10:44:15, 5.69s/it]
85%|████████▍ | 37146/43939 [59:54:27<10:44:10, 5.69s/it]
85%|████████▍ | 37147/43939 [59:54:33<10:43:42, 5.69s/it]
85%|████████▍ | 37148/43939 [59:54:38<10:45:12, 5.70s/it]
85%|████████▍ | 37149/43939 [59:54:44<10:45:49, 5.71s/it]
85%|████████▍ | 37150/43939 [59:54:50<10:45:08, 5.70s/it]
{'loss': 0.6446, 'grad_norm': 1.333788514137268, 'learning_rate': 3.5496727541848215e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.93, 'epoch': 3.38}
85%|████████▍ | 37150/43939 [59:54:50<10:45:08, 5.70s/it]
85%|████████▍ | 37151/43939 [59:54:55<10:47:57, 5.73s/it]
85%|████████▍ | 37152/43939 [59:55:01<10:46:00, 5.71s/it]
85%|████████▍ | 37153/43939 [59:55:07<10:48:11, 5.73s/it]
85%|████████▍ | 37154/43939 [59:55:13<10:48:30, 5.73s/it]
85%|████████▍ | 37155/43939 [59:55:18<10:48:25, 5.73s/it]
85%|████████▍ | 37156/43939 [59:55:24<10:47:44, 5.73s/it]
85%|████████▍ | 37157/43939 [59:55:30<10:47:07, 5.73s/it]
85%|████████▍ | 37158/43939 [59:55:36<10:53:06, 5.78s/it]
85%|████████▍ | 37159/43939 [59:55:41<10:51:20, 5.76s/it]
85%|████████▍ | 37160/43939 [59:55:47<10:48:36, 5.74s/it]
85%|████████▍ | 37161/43939 [59:55:53<10:48:52, 5.74s/it]
85%|████████▍ | 37162/43939 [59:55:59<10:48:26, 5.74s/it]
85%|████████▍ | 37163/43939 [59:56:04<10:47:34, 5.73s/it]
85%|████████▍ | 37164/43939 [59:56:10<10:49:30, 5.75s/it]
85%|████████▍ | 37165/43939 [59:56:16<10:49:02, 5.75s/it]
85%|████████▍ | 37166/43939 [59:56:22<10:46:17, 5.73s/it]
85%|████████▍ | 37167/43939 [59:56:27<10:45:47, 5.72s/it]
85%|████████▍ | 37168/43939 [59:56:33<10:44:34, 5.71s/it]
85%|████████▍ | 37169/43939 [59:56:39<10:44:05, 5.71s/it]
85%|████████▍ | 37170/43939 [59:56:44<10:44:36, 5.71s/it]
85%|████████▍ | 37171/43939 [59:56:50<10:44:23, 5.71s/it]
85%|████████▍ | 37172/43939 [59:56:56<10:43:04, 5.70s/it]
85%|████████▍ | 37173/43939 [59:57:01<10:43:39, 5.71s/it]
85%|████████▍ | 37174/43939 [59:57:07<10:44:00, 5.71s/it]
85%|████████▍ | 37175/43939 [59:57:13<10:43:29, 5.71s/it]
85%|████████▍ | 37176/43939 [59:57:19<10:43:02, 5.70s/it]
85%|████████▍ | 37177/43939 [59:57:24<10:41:44, 5.69s/it]
85%|████████▍ | 37178/43939 [59:57:30<10:42:56, 5.71s/it]
85%|████████▍ | 37179/43939 [59:57:36<10:49:37, 5.77s/it]
85%|████████▍ | 37180/43939 [59:57:42<10:47:20, 5.75s/it]
85%|████████▍ | 37181/43939 [59:57:47<10:45:54, 5.73s/it]
85%|████████▍ | 37182/43939 [59:57:53<10:45:14, 5.73s/it]
85%|████████▍ | 37183/43939 [59:57:59<10:45:14, 5.73s/it]
85%|████████▍ | 37184/43939 [59:58:04<10:42:24, 5.71s/it]
85%|████████▍ | 37185/43939 [59:58:10<10:42:51, 5.71s/it]
85%|████████▍ | 37186/43939 [59:58:16<10:44:23, 5.73s/it]
85%|████████▍ | 37187/43939 [59:58:22<10:44:46, 5.73s/it]
85%|████████▍ | 37188/43939 [59:58:27<10:42:43, 5.71s/it]
85%|████████▍ | 37189/43939 [59:58:33<10:42:13, 5.71s/it]
85%|████████▍ | 37190/43939 [59:58:39<10:42:06, 5.71s/it]
85%|████████▍ | 37191/43939 [59:58:44<10:40:17, 5.69s/it]
85%|████████▍ | 37192/43939 [59:58:50<10:39:55, 5.69s/it]
85%|████████▍ | 37193/43939 [59:58:56<10:41:08, 5.70s/it]
85%|████████▍ | 37194/43939 [59:59:01<10:40:35, 5.70s/it]
85%|████████▍ | 37195/43939 [59:59:07<10:41:15, 5.71s/it]
85%|████████▍ | 37196/43939 [59:59:13<10:40:41, 5.70s/it]
85%|████████▍ | 37197/43939 [59:59:19<10:40:30, 5.70s/it]
85%|████████▍ | 37198/43939 [59:59:24<10:40:45, 5.70s/it]
85%|████████▍ | 37199/43939 [59:59:30<10:42:05, 5.72s/it]
85%|████████▍ | 37200/43939 [59:59:36<10:40:41, 5.70s/it]
{'loss': 0.6681, 'grad_norm': 1.793472170829773, 'learning_rate': 3.498837912065489e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.56, 'epoch': 3.39}
85%|████████▍ | 37200/43939 [59:59:36<10:40:41, 5.70s/it]
85%|████████▍ | 37201/43939 [59:59:42<10:47:29, 5.77s/it]
85%|████████▍ | 37202/43939 [59:59:47<10:45:56, 5.75s/it]
85%|████████▍ | 37203/43939 [59:59:53<10:44:13, 5.74s/it]
85%|████████▍ | 37204/43939 [59:59:59<10:42:45, 5.73s/it]
85%|████████▍ | 37205/43939 [60:00:04<10:43:03, 5.73s/it]
85%|████████▍ | 37206/43939 [60:00:10<10:40:48, 5.71s/it]
85%|████████▍ | 37207/43939 [60:00:16<10:39:30, 5.70s/it]
85%|████████▍ | 37208/43939 [60:00:21<10:38:03, 5.69s/it]
85%|████████▍ | 37209/43939 [60:00:27<10:38:13, 5.69s/it]
85%|████████▍ | 37210/43939 [60:00:33<10:37:30, 5.68s/it]
85%|████████▍ | 37211/43939 [60:00:39<10:38:13, 5.69s/it]
85%|████████▍ | 37212/43939 [60:00:44<10:38:57, 5.70s/it]
85%|████████▍ | 37213/43939 [60:00:50<10:39:02, 5.70s/it]
85%|████████▍ | 37214/43939 [60:00:56<10:40:47, 5.72s/it]
85%|████████▍ | 37215/43939 [60:01:01<10:40:02, 5.71s/it]
85%|████████▍ | 37216/43939 [60:01:07<10:39:56, 5.71s/it]
85%|████████▍ | 37217/43939 [60:01:13<10:39:31, 5.71s/it]
85%|████████▍ | 37218/43939 [60:01:19<10:40:10, 5.72s/it]
85%|████████▍ | 37219/43939 [60:01:24<10:39:51, 5.71s/it]
85%|████████▍ | 37220/43939 [60:01:30<10:41:50, 5.73s/it]
85%|████████▍ | 37221/43939 [60:01:36<10:40:58, 5.72s/it]
85%|████████▍ | 37222/43939 [60:01:42<10:48:37, 5.79s/it]
85%|████████▍ | 37223/43939 [60:01:47<10:45:07, 5.76s/it]
85%|████████▍ | 37224/43939 [60:01:53<10:41:55, 5.74s/it]
85%|████████▍ | 37225/43939 [60:01:59<10:41:31, 5.73s/it]
85%|████████▍ | 37226/43939 [60:02:05<10:40:20, 5.72s/it]
85%|████████▍ | 37227/43939 [60:02:10<10:40:31, 5.73s/it]
85%|████████▍ | 37228/43939 [60:02:16<10:39:29, 5.72s/it]
85%|████████▍ | 37229/43939 [60:02:22<10:38:57, 5.71s/it]
85%|████████▍ | 37230/43939 [60:02:27<10:39:58, 5.72s/it]
85%|████████▍ | 37231/43939 [60:02:33<10:39:33, 5.72s/it]
85%|████████▍ | 37232/43939 [60:02:39<10:37:29, 5.70s/it]
85%|████████▍ | 37233/43939 [60:02:45<10:37:51, 5.71s/it]
85%|████████▍ | 37234/43939 [60:02:50<10:37:08, 5.70s/it]
85%|████████▍ | 37235/43939 [60:02:56<10:36:28, 5.70s/it]
85%|████████▍ | 37236/43939 [60:03:02<10:38:53, 5.72s/it]
85%|████████▍ | 37237/43939 [60:03:07<10:38:40, 5.72s/it]
85%|████████▍ | 37238/43939 [60:03:13<10:37:41, 5.71s/it]
85%|████████▍ | 37239/43939 [60:03:19<10:38:05, 5.71s/it]
85%|████████▍ | 37240/43939 [60:03:24<10:37:44, 5.71s/it]
85%|████████▍ | 37241/43939 [60:03:30<10:37:31, 5.71s/it]
85%|████████▍ | 37242/43939 [60:03:36<10:37:40, 5.71s/it]
85%|████████▍ | 37243/43939 [60:03:42<10:36:51, 5.71s/it]
85%|████████▍ | 37244/43939 [60:03:48<10:45:23, 5.78s/it]
85%|████████▍ | 37245/43939 [60:03:53<10:42:52, 5.76s/it]
85%|████████▍ | 37246/43939 [60:03:59<10:40:54, 5.75s/it]
85%|████████▍ | 37247/43939 [60:04:05<10:39:16, 5.73s/it]
85%|████████▍ | 37248/43939 [60:04:10<10:39:20, 5.73s/it]
85%|████████▍ | 37249/43939 [60:04:16<10:40:08, 5.74s/it]
85%|████████▍ | 37250/43939 [60:04:22<10:38:32, 5.73s/it]
{'loss': 0.6414, 'grad_norm': 1.61867356300354, 'learning_rate': 3.4483423013051323e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.26, 'epoch': 3.39}
85%|████████▍ | 37250/43939 [60:04:22<10:38:32, 5.73s/it]
85%|████████▍ | 37251/43939 [60:04:28<10:37:47, 5.72s/it]
85%|████████▍ | 37252/43939 [60:04:33<10:36:54, 5.71s/it]
85%|████████▍ | 37253/43939 [60:04:39<10:36:56, 5.72s/it]
85%|████████▍ | 37254/43939 [60:04:45<10:35:49, 5.71s/it]
85%|████████▍ | 37255/43939 [60:04:50<10:36:20, 5.71s/it]
85%|████████▍ | 37256/43939 [60:04:56<10:38:32, 5.73s/it]
85%|████████▍ | 37257/43939 [60:05:02<10:36:46, 5.72s/it]
85%|████████▍ | 37258/43939 [60:05:08<10:35:55, 5.71s/it]
85%|████████▍ | 37259/43939 [60:05:13<10:35:10, 5.71s/it]
85%|████████▍ | 37260/43939 [60:05:19<10:35:14, 5.71s/it]
85%|████████▍ | 37261/43939 [60:05:25<10:35:59, 5.71s/it]
85%|████████▍ | 37262/43939 [60:05:30<10:37:58, 5.73s/it]
85%|████████▍ | 37263/43939 [60:05:36<10:36:39, 5.72s/it]
85%|████████▍ | 37264/43939 [60:05:42<10:43:54, 5.79s/it]
85%|████████▍ | 37265/43939 [60:05:48<10:41:44, 5.77s/it]
85%|████████▍ | 37266/43939 [60:05:54<10:40:46, 5.76s/it]
85%|████████▍ | 37267/43939 [60:05:59<10:38:23, 5.74s/it]
85%|████████▍ | 37268/43939 [60:06:05<10:37:43, 5.74s/it]
85%|████████▍ | 37269/43939 [60:06:11<10:36:26, 5.73s/it]
85%|████████▍ | 37270/43939 [60:06:16<10:34:35, 5.71s/it]
85%|████████▍ | 37271/43939 [60:06:22<10:33:48, 5.70s/it]
85%|████████▍ | 37272/43939 [60:06:28<10:34:21, 5.71s/it]
85%|████████▍ | 37273/43939 [60:06:34<10:35:00, 5.72s/it]
85%|████████▍ | 37274/43939 [60:06:39<10:34:40, 5.71s/it]
85%|████████▍ | 37275/43939 [60:06:45<10:34:27, 5.71s/it]
85%|████████▍ | 37276/43939 [60:06:51<10:34:40, 5.72s/it]
85%|████████▍ | 37277/43939 [60:06:56<10:34:24, 5.71s/it]
85%|████████▍ | 37278/43939 [60:07:02<10:32:28, 5.70s/it]
85%|████████▍ | 37279/43939 [60:07:08<10:32:34, 5.70s/it]
85%|████████▍ | 37280/43939 [60:07:13<10:32:54, 5.70s/it]
85%|████████▍ | 37281/43939 [60:07:19<10:32:30, 5.70s/it]
85%|████████▍ | 37282/43939 [60:07:25<10:33:02, 5.71s/it]
85%|████████▍ | 37283/43939 [60:07:31<10:34:26, 5.72s/it]
85%|████████▍ | 37284/43939 [60:07:36<10:32:52, 5.71s/it]
85%|████████▍ | 37285/43939 [60:07:42<10:33:16, 5.71s/it]
85%|████████▍ | 37286/43939 [60:07:48<10:40:27, 5.78s/it]
85%|████████▍ | 37287/43939 [60:07:54<10:39:53, 5.77s/it]
85%|████████▍ | 37288/43939 [60:07:59<10:37:06, 5.75s/it]
85%|████████▍ | 37289/43939 [60:08:05<10:33:08, 5.71s/it]
85%|████████▍ | 37290/43939 [60:08:11<10:34:56, 5.73s/it]
85%|████████▍ | 37291/43939 [60:08:16<10:33:37, 5.72s/it]
85%|████████▍ | 37292/43939 [60:08:22<10:32:03, 5.71s/it]
85%|████████▍ | 37293/43939 [60:08:28<10:33:07, 5.72s/it]
85%|████████▍ | 37294/43939 [60:08:34<10:32:04, 5.71s/it]
85%|████████▍ | 37295/43939 [60:08:39<10:31:26, 5.70s/it]
85%|████████▍ | 37296/43939 [60:08:45<10:33:06, 5.72s/it]
85%|████████▍ | 37297/43939 [60:08:51<10:32:04, 5.71s/it]
85%|████████▍ | 37298/43939 [60:08:56<10:30:34, 5.70s/it]
85%|████████▍ | 37299/43939 [60:09:02<10:30:31, 5.70s/it]
85%|████████▍ | 37300/43939 [60:09:08<10:30:06, 5.69s/it]
{'loss': 0.6657, 'grad_norm': 1.9037747383117676, 'learning_rate': 3.3981867185906706e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.37, 'epoch': 3.4}
85%|████████▍ | 37300/43939 [60:09:08<10:30:06, 5.69s/it]
85%|████████▍ | 37301/43939 [60:09:14<10:32:21, 5.72s/it]
85%|████████▍ | 37302/43939 [60:09:19<10:33:20, 5.73s/it]
85%|████████▍ | 37303/43939 [60:09:25<10:32:39, 5.72s/it]
85%|████████▍ | 37304/43939 [60:09:31<10:31:14, 5.71s/it]
85%|████████▍ | 37305/43939 [60:09:36<10:31:22, 5.71s/it]
85%|████████▍ | 37306/43939 [60:09:42<10:32:10, 5.72s/it]
85%|████████▍ | 37307/43939 [60:09:48<10:41:00, 5.80s/it]
85%|████████▍ | 37308/43939 [60:09:54<10:36:52, 5.76s/it]
85%|████████▍ | 37309/43939 [60:09:59<10:34:47, 5.74s/it]
85%|████████▍ | 37310/43939 [60:10:05<10:33:50, 5.74s/it]
85%|████████▍ | 37311/43939 [60:10:11<10:31:39, 5.72s/it]
85%|████████▍ | 37312/43939 [60:10:17<10:30:18, 5.71s/it]
85%|████████▍ | 37313/43939 [60:10:22<10:31:34, 5.72s/it]
85%|████████▍ | 37314/43939 [60:10:28<10:29:06, 5.70s/it]
85%|████████▍ | 37315/43939 [60:10:34<10:27:35, 5.68s/it]
85%|████████▍ | 37316/43939 [60:10:39<10:27:17, 5.68s/it]
85%|████████▍ | 37317/43939 [60:10:45<10:28:18, 5.69s/it]
85%|████████▍ | 37318/43939 [60:10:51<10:29:09, 5.70s/it]
85%|████████▍ | 37319/43939 [60:10:56<10:29:17, 5.70s/it]
85%|████████▍ | 37320/43939 [60:11:02<10:28:56, 5.70s/it]
85%|████████▍ | 37321/43939 [60:11:08<10:28:54, 5.70s/it]
85%|████████▍ | 37322/43939 [60:11:14<10:32:19, 5.73s/it]
85%|████████▍ | 37323/43939 [60:11:19<10:30:21, 5.72s/it]
85%|████████▍ | 37324/43939 [60:11:25<10:28:57, 5.70s/it]
85%|████████▍ | 37325/43939 [60:11:31<10:28:18, 5.70s/it]
85%|████████▍ | 37326/43939 [60:11:36<10:28:34, 5.70s/it]
85%|████████▍ | 37327/43939 [60:11:42<10:29:43, 5.71s/it]
85%|████████▍ | 37328/43939 [60:11:48<10:36:18, 5.77s/it]
85%|████████▍ | 37329/43939 [60:11:54<10:32:30, 5.74s/it]
85%|████████▍ | 37330/43939 [60:11:59<10:30:30, 5.72s/it]
85%|████████▍ | 37331/43939 [60:12:05<10:29:26, 5.72s/it]
85%|████████▍ | 37332/43939 [60:12:11<10:28:08, 5.70s/it]
85%|████████▍ | 37333/43939 [60:12:16<10:27:44, 5.70s/it]
85%|████████▍ | 37334/43939 [60:12:22<10:27:51, 5.70s/it]
85%|████████▍ | 37335/43939 [60:12:28<10:28:20, 5.71s/it]
85%|████████▍ | 37336/43939 [60:12:34<10:27:50, 5.71s/it]
85%|████████▍ | 37337/43939 [60:12:39<10:28:32, 5.71s/it]
85%|████████▍ | 37338/43939 [60:12:45<10:30:44, 5.73s/it]
85%|████████▍ | 37339/43939 [60:12:51<10:30:03, 5.73s/it]
85%|████████▍ | 37340/43939 [60:12:56<10:27:38, 5.71s/it]
85%|████████▍ | 37341/43939 [60:13:02<10:28:10, 5.71s/it]
85%|████████▍ | 37342/43939 [60:13:08<10:27:09, 5.70s/it]
85%|████████▍ | 37343/43939 [60:13:14<10:26:59, 5.70s/it]
85%|████████▍ | 37344/43939 [60:13:19<10:26:58, 5.70s/it]
85%|████████▍ | 37345/43939 [60:13:25<10:26:53, 5.70s/it]
85%|████████▍ | 37346/43939 [60:13:31<10:26:40, 5.70s/it]
85%|████████▍ | 37347/43939 [60:13:36<10:26:58, 5.71s/it]
85%|████████▍ | 37348/43939 [60:13:42<10:26:58, 5.71s/it]
85%|████████▌ | 37349/43939 [60:13:48<10:33:04, 5.76s/it][2025-12-01 06:18:01,842] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:41, 2.41s/it][A
8%|▊ | 6/72 [00:13<02:46, 2.52s/it][A
10%|▉ | 7/72 [00:16<02:48, 2.59s/it][A
11%|█ | 8/72 [00:19<02:48, 2.64s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:46, 2.68s/it][A
15%|█▌ | 11/72 [00:27<02:44, 2.69s/it][A
17%|█▋ | 12/72 [00:30<02:42, 2.71s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.72s/it][A
21%|██ | 15/72 [00:38<02:35, 2.72s/it][A
22%|██▏ | 16/72 [00:40<02:32, 2.72s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.73s/it][A
25%|██▌ | 18/72 [00:46<02:27, 2.73s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.73s/it][A
28%|██▊ | 20/72 [00:51<02:22, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.74s/it][A
31%|███ | 22/72 [00:57<02:16, 2.74s/it][A
32%|███▏ | 23/72 [01:00<02:13, 2.73s/it][A
33%|███▎ | 24/72 [01:02<02:11, 2.73s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:11<02:03, 2.74s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.74s/it][A
40%|████ | 29/72 [01:16<01:57, 2.74s/it][A
42%|████▏ | 30/72 [01:19<01:55, 2.74s/it][A
43%|████▎ | 31/72 [01:22<01:52, 2.74s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.74s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.74s/it][A
47%|████▋ | 34/72 [01:30<01:44, 2.74s/it][A
49%|████▊ | 35/72 [01:32<01:41, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.74s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.74s/it][A
53%|█████▎ | 38/72 [01:41<01:33, 2.74s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.74s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.74s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:52<01:22, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.74s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.74s/it][A
62%|██████▎ | 45/72 [02:00<01:13, 2.74s/it][A
64%|██████▍ | 46/72 [02:03<01:11, 2.74s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.73s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.74s/it][A
68%|██████▊ | 49/72 [02:11<01:02, 2.74s/it][A
69%|██████▉ | 50/72 [02:14<01:00, 2.74s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.74s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.74s/it][A
74%|███████▎ | 53/72 [02:22<00:52, 2.74s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.74s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:30<00:43, 2.74s/it][A
79%|███████▉ | 57/72 [02:33<00:41, 2.74s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:41<00:32, 2.74s/it][A
85%|████████▍ | 61/72 [02:44<00:30, 2.74s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:52<00:21, 2.73s/it][A
90%|█████████ | 65/72 [02:55<00:19, 2.73s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.73s/it][A
93%|█████████▎| 67/72 [03:00<00:13, 2.73s/it][A
94%|█████████▍| 68/72 [03:03<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.73s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.73s/it][A
99%|█████████▊| 71/72 [03:11<00:02, 2.73s/it][A
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A{'eval_loss': 1.015528678894043, 'eval_runtime': 195.8565, 'eval_samples_per_second': 5.101, 'eval_steps_per_second': 0.368, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 3.4}
85%|████████▌ | 37349/43939 [60:17:04<10:33:04, 5.76s/it]
100%|██████████| 72/72 [03:12<00:00, 2.26s/it][A
[A
85%|████████▌ | 37350/43939 [60:17:10<118:07:50, 64.54s/it]
{'loss': 0.6504, 'grad_norm': 1.6967730522155762, 'learning_rate': 3.3483719552442894e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 76.01, 'tokens_per_second_per_gpu': 8.22, 'epoch': 3.4}
85%|████████▌ | 37350/43939 [60:17:10<118:07:50, 64.54s/it]
85%|████████▌ | 37351/43939 [60:17:15<85:48:16, 46.89s/it]
85%|████████▌ | 37352/43939 [60:17:21<63:10:39, 34.53s/it]
85%|████████▌ | 37353/43939 [60:17:27<47:18:45, 25.86s/it]
85%|████████▌ | 37354/43939 [60:17:32<36:14:59, 19.82s/it]
85%|████████▌ | 37355/43939 [60:17:38<28:28:39, 15.57s/it]
85%|████████▌ | 37356/43939 [60:17:44<23:04:12, 12.62s/it]
85%|████████▌ | 37357/43939 [60:17:50<19:16:35, 10.54s/it]
85%|████████▌ | 37358/43939 [60:17:55<16:36:55, 9.09s/it]
85%|████████▌ | 37359/43939 [60:18:01<14:45:25, 8.07s/it]
85%|████████▌ | 37360/43939 [60:18:07<13:28:31, 7.37s/it]
85%|████████▌ | 37361/43939 [60:18:12<12:33:13, 6.87s/it]
85%|████████▌ | 37362/43939 [60:18:18<11:53:42, 6.51s/it]
85%|████████▌ | 37363/43939 [60:18:24<11:26:40, 6.27s/it]
85%|████████▌ | 37364/43939 [60:18:29<11:07:46, 6.09s/it]
85%|████████▌ | 37365/43939 [60:18:35<10:55:00, 5.98s/it]
85%|████████▌ | 37366/43939 [60:18:41<10:44:50, 5.89s/it]
85%|████████▌ | 37367/43939 [60:18:47<10:40:04, 5.84s/it]
85%|████████▌ | 37368/43939 [60:18:52<10:35:51, 5.81s/it]
85%|████████▌ | 37369/43939 [60:18:58<10:31:34, 5.77s/it]
85%|████████▌ | 37370/43939 [60:19:04<10:29:45, 5.75s/it]
85%|████████▌ | 37371/43939 [60:19:10<10:37:05, 5.82s/it]
85%|████████▌ | 37372/43939 [60:19:15<10:32:34, 5.78s/it]
85%|████████▌ | 37373/43939 [60:19:21<10:29:27, 5.75s/it]
85%|████████▌ | 37374/43939 [60:19:27<10:28:17, 5.74s/it]
85%|████████▌ | 37375/43939 [60:19:32<10:27:21, 5.73s/it]
85%|████████▌ | 37376/43939 [60:19:38<10:26:41, 5.73s/it]
85%|████████▌ | 37377/43939 [60:19:44<10:26:24, 5.73s/it]
85%|████████▌ | 37378/43939 [60:19:50<10:27:16, 5.74s/it]
85%|████████▌ | 37379/43939 [60:19:55<10:27:47, 5.74s/it]
85%|████████▌ | 37380/43939 [60:20:01<10:26:19, 5.73s/it]
85%|████████▌ | 37381/43939 [60:20:07<10:25:17, 5.72s/it]
85%|████████▌ | 37382/43939 [60:20:13<10:25:08, 5.72s/it]
85%|████████▌ | 37383/43939 [60:20:18<10:23:57, 5.71s/it]
85%|████████▌ | 37384/43939 [60:20:24<10:23:13, 5.70s/it]
85%|████████▌ | 37385/43939 [60:20:30<10:21:56, 5.69s/it]
85%|████████▌ | 37386/43939 [60:20:35<10:21:46, 5.69s/it]
85%|████████▌ | 37387/43939 [60:20:41<10:22:15, 5.70s/it]
85%|████████▌ | 37388/43939 [60:20:47<10:23:12, 5.71s/it]
85%|████████▌ | 37389/43939 [60:20:52<10:21:04, 5.69s/it]
85%|████████▌ | 37390/43939 [60:20:58<10:21:19, 5.69s/it]
85%|████████▌ | 37391/43939 [60:21:04<10:22:34, 5.70s/it]
85%|████████▌ | 37392/43939 [60:21:09<10:21:43, 5.70s/it]
85%|████████▌ | 37393/43939 [60:21:15<10:29:06, 5.77s/it]
85%|████████▌ | 37394/43939 [60:21:21<10:26:13, 5.74s/it]
85%|████████▌ | 37395/43939 [60:21:27<10:25:53, 5.74s/it]
85%|████████▌ | 37396/43939 [60:21:33<10:24:13, 5.72s/it]
85%|████████▌ | 37397/43939 [60:21:38<10:25:55, 5.74s/it]
85%|████████▌ | 37398/43939 [60:21:44<10:26:32, 5.75s/it]
85%|████████▌ | 37399/43939 [60:21:50<10:25:06, 5.73s/it]
85%|████████▌ | 37400/43939 [60:21:55<10:25:15, 5.74s/it]
{'loss': 0.6673, 'grad_norm': 1.3011078834533691, 'learning_rate': 3.2988987972109418e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.7, 'epoch': 3.4}
85%|████████▌ | 37400/43939 [60:21:56<10:25:15, 5.74s/it]
85%|████████▌ | 37401/43939 [60:22:01<10:24:38, 5.73s/it]
85%|████████▌ | 37402/43939 [60:22:07<10:22:40, 5.72s/it]
85%|████████▌ | 37403/43939 [60:22:13<10:22:18, 5.71s/it]
85%|████████▌ | 37404/43939 [60:22:18<10:22:18, 5.71s/it]
85%|████████▌ | 37405/43939 [60:22:24<10:21:09, 5.70s/it]
85%|████████▌ | 37406/43939 [60:22:30<10:21:24, 5.71s/it]
85%|████████▌ | 37407/43939 [60:22:35<10:20:42, 5.70s/it]
85%|████████▌ | 37408/43939 [60:22:41<10:21:41, 5.71s/it]
85%|████████▌ | 37409/43939 [60:22:47<10:21:09, 5.71s/it]
85%|████████▌ | 37410/43939 [60:22:53<10:19:52, 5.70s/it]
85%|████████▌ | 37411/43939 [60:22:58<10:19:15, 5.69s/it]
85%|████████▌ | 37412/43939 [60:23:04<10:19:03, 5.69s/it]
85%|████████▌ | 37413/43939 [60:23:10<10:20:06, 5.70s/it]
85%|████████▌ | 37414/43939 [60:23:16<10:28:28, 5.78s/it]
85%|████████▌ | 37415/43939 [60:23:21<10:25:04, 5.75s/it]
85%|████████▌ | 37416/43939 [60:23:27<10:24:38, 5.75s/it]
85%|████████▌ | 37417/43939 [60:23:33<10:23:02, 5.73s/it]
85%|████████▌ | 37418/43939 [60:23:38<10:21:31, 5.72s/it]
85%|████████▌ | 37419/43939 [60:23:44<10:21:16, 5.72s/it]
85%|████████▌ | 37420/43939 [60:23:50<10:19:47, 5.70s/it]
85%|████████▌ | 37421/43939 [60:23:55<10:20:10, 5.71s/it]
85%|████████▌ | 37422/43939 [60:24:01<10:19:18, 5.70s/it]
85%|████████▌ | 37423/43939 [60:24:07<10:17:50, 5.69s/it]
85%|████████▌ | 37424/43939 [60:24:13<10:17:38, 5.69s/it]
85%|████████▌ | 37425/43939 [60:24:18<10:18:38, 5.70s/it]
85%|████████▌ | 37426/43939 [60:24:24<10:18:57, 5.70s/it]
85%|████████▌ | 37427/43939 [60:24:30<10:18:12, 5.70s/it]
85%|████████▌ | 37428/43939 [60:24:35<10:17:41, 5.69s/it]
85%|████████▌ | 37429/43939 [60:24:41<10:17:23, 5.69s/it]
85%|████████▌ | 37430/43939 [60:24:47<10:17:04, 5.69s/it]
85%|████████▌ | 37431/43939 [60:24:52<10:18:09, 5.70s/it]
85%|████████▌ | 37432/43939 [60:24:58<10:19:15, 5.71s/it]
85%|████████▌ | 37433/43939 [60:25:04<10:19:46, 5.72s/it]
85%|████████▌ | 37434/43939 [60:25:10<10:19:09, 5.71s/it]
85%|████████▌ | 37435/43939 [60:25:15<10:25:48, 5.77s/it]
85%|████████▌ | 37436/43939 [60:25:21<10:23:52, 5.76s/it]
85%|████████▌ | 37437/43939 [60:25:27<10:22:04, 5.74s/it]
85%|████████▌ | 37438/43939 [60:25:33<10:21:32, 5.74s/it]
85%|████████▌ | 37439/43939 [60:25:38<10:20:08, 5.72s/it]
85%|████████▌ | 37440/43939 [60:25:44<10:19:24, 5.72s/it]
85%|████████▌ | 37441/43939 [60:25:50<10:19:34, 5.72s/it]
85%|████████▌ | 37442/43939 [60:25:55<10:19:35, 5.72s/it]
85%|████████▌ | 37443/43939 [60:26:01<10:18:58, 5.72s/it]
85%|████████▌ | 37444/43939 [60:26:07<10:17:32, 5.70s/it]
85%|████████▌ | 37445/43939 [60:26:13<10:17:27, 5.70s/it]
85%|████████▌ | 37446/43939 [60:26:18<10:16:17, 5.69s/it]
85%|████████▌ | 37447/43939 [60:26:24<10:17:19, 5.71s/it]
85%|████████▌ | 37448/43939 [60:26:30<10:16:41, 5.70s/it]
85%|████████▌ | 37449/43939 [60:26:35<10:16:57, 5.70s/it]
85%|████████▌ | 37450/43939 [60:26:41<10:17:38, 5.71s/it]
{'loss': 0.6342, 'grad_norm': 1.3419616222381592, 'learning_rate': 3.2497680250459547e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 451.1, 'epoch': 3.41}
85%|████████▌ | 37450/43939 [60:26:41<10:17:38, 5.71s/it]
85%|████████▌ | 37451/43939 [60:26:47<10:18:01, 5.72s/it]
85%|████████▌ | 37452/43939 [60:26:53<10:18:41, 5.72s/it]
85%|████████▌ | 37453/43939 [60:26:58<10:19:40, 5.73s/it]
85%|████████▌ | 37454/43939 [60:27:04<10:19:16, 5.73s/it]
85%|████████▌ | 37455/43939 [60:27:10<10:17:47, 5.72s/it]
85%|████████▌ | 37456/43939 [60:27:15<10:16:29, 5.71s/it]
85%|████████▌ | 37457/43939 [60:27:21<10:23:59, 5.78s/it]
85%|████████▌ | 37458/43939 [60:27:27<10:20:44, 5.75s/it]
85%|████████▌ | 37459/43939 [60:27:33<10:18:52, 5.73s/it]
85%|████████▌ | 37460/43939 [60:27:38<10:17:45, 5.72s/it]
85%|████████▌ | 37461/43939 [60:27:44<10:16:37, 5.71s/it]
85%|████████▌ | 37462/43939 [60:27:50<10:15:37, 5.70s/it]
85%|████████▌ | 37463/43939 [60:27:55<10:15:56, 5.71s/it]
85%|████████▌ | 37464/43939 [60:28:01<10:16:42, 5.71s/it]
85%|████████▌ | 37465/43939 [60:28:07<10:16:23, 5.71s/it]
85%|████████▌ | 37466/43939 [60:28:13<10:17:44, 5.73s/it]
85%|████████▌ | 37467/43939 [60:28:18<10:18:00, 5.73s/it]
85%|████████▌ | 37468/43939 [60:28:24<10:18:14, 5.73s/it]
85%|████████▌ | 37469/43939 [60:28:30<10:16:58, 5.72s/it]
85%|████████▌ | 37470/43939 [60:28:36<10:16:05, 5.71s/it]
85%|████████▌ | 37471/43939 [60:28:41<10:14:32, 5.70s/it]
85%|████████▌ | 37472/43939 [60:28:47<10:14:59, 5.71s/it]
85%|████████▌ | 37473/43939 [60:28:53<10:15:04, 5.71s/it]
85%|████████▌ | 37474/43939 [60:28:58<10:14:19, 5.70s/it]
85%|████████▌ | 37475/43939 [60:29:04<10:15:15, 5.71s/it]
85%|████████▌ | 37476/43939 [60:29:10<10:14:42, 5.71s/it]
85%|████████▌ | 37477/43939 [60:29:15<10:14:11, 5.70s/it]
85%|████████▌ | 37478/43939 [60:29:21<10:14:07, 5.70s/it]
85%|████████▌ | 37479/43939 [60:29:27<10:23:04, 5.79s/it]
85%|████████▌ | 37480/43939 [60:29:33<10:19:22, 5.75s/it]
85%|████████▌ | 37481/43939 [60:29:39<10:17:39, 5.74s/it]
85%|████████▌ | 37482/43939 [60:29:44<10:18:31, 5.75s/it]
85%|████████▌ | 37483/43939 [60:29:50<10:18:17, 5.75s/it]
85%|████████▌ | 37484/43939 [60:29:56<10:17:26, 5.74s/it]
85%|████████▌ | 37485/43939 [60:30:01<10:16:42, 5.73s/it]
85%|████████▌ | 37486/43939 [60:30:07<10:15:35, 5.72s/it]
85%|████████▌ | 37487/43939 [60:30:13<10:15:42, 5.73s/it]
85%|████████▌ | 37488/43939 [60:30:19<10:16:39, 5.74s/it]
85%|████████▌ | 37489/43939 [60:30:24<10:14:36, 5.72s/it]
85%|████████▌ | 37490/43939 [60:30:30<10:14:44, 5.72s/it]
85%|████████▌ | 37491/43939 [60:30:36<10:14:57, 5.72s/it]
85%|████████▌ | 37492/43939 [60:30:42<10:13:51, 5.71s/it]
85%|████████▌ | 37493/43939 [60:30:47<10:12:35, 5.70s/it]
85%|████████▌ | 37494/43939 [60:30:53<10:11:43, 5.69s/it]
85%|████████▌ | 37495/43939 [60:30:59<10:12:10, 5.70s/it]
85%|████████▌ | 37496/43939 [60:31:04<10:12:59, 5.71s/it]
85%|████████▌ | 37497/43939 [60:31:10<10:13:59, 5.72s/it]
85%|████████▌ | 37498/43939 [60:31:16<10:13:42, 5.72s/it]
85%|████████▌ | 37499/43939 [60:31:22<10:14:58, 5.73s/it]
85%|████████▌ | 37500/43939 [60:31:27<10:20:55, 5.79s/it]
{'loss': 0.6598, 'grad_norm': 1.4529420137405396, 'learning_rate': 3.2009804139027155e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.52, 'epoch': 3.41}
85%|████████▌ | 37500/43939 [60:31:27<10:20:55, 5.79s/it]
85%|████████▌ | 37501/43939 [60:31:33<10:17:50, 5.76s/it]
85%|████████▌ | 37502/43939 [60:31:39<10:15:36, 5.74s/it]
85%|████████▌ | 37503/43939 [60:31:45<10:14:17, 5.73s/it]
85%|████████▌ | 37504/43939 [60:31:50<10:14:20, 5.73s/it]
85%|████████▌ | 37505/43939 [60:31:56<10:13:53, 5.72s/it]
85%|████████▌ | 37506/43939 [60:32:02<10:14:10, 5.73s/it]
85%|████████▌ | 37507/43939 [60:32:07<10:12:45, 5.72s/it]
85%|████████▌ | 37508/43939 [60:32:13<10:11:02, 5.70s/it]
85%|████████▌ | 37509/43939 [60:32:19<10:11:10, 5.70s/it]
85%|████████▌ | 37510/43939 [60:32:24<10:10:19, 5.70s/it]
85%|████████▌ | 37511/43939 [60:32:30<10:09:14, 5.69s/it]
85%|████████▌ | 37512/43939 [60:32:36<10:09:49, 5.69s/it]
85%|████████▌ | 37513/43939 [60:32:42<10:10:17, 5.70s/it]
85%|████████▌ | 37514/43939 [60:32:47<10:09:39, 5.69s/it]
85%|████████▌ | 37515/43939 [60:32:53<10:11:32, 5.71s/it]
85%|████████▌ | 37516/43939 [60:32:59<10:12:21, 5.72s/it]
85%|████████▌ | 37517/43939 [60:33:04<10:10:06, 5.70s/it]
85%|████████▌ | 37518/43939 [60:33:10<10:09:25, 5.69s/it]
85%|████████▌ | 37519/43939 [60:33:16<10:09:44, 5.70s/it]
85%|████████▌ | 37520/43939 [60:33:22<10:16:16, 5.76s/it]
85%|████████▌ | 37521/43939 [60:33:27<10:13:35, 5.74s/it]
85%|████████▌ | 37522/43939 [60:33:33<10:10:52, 5.71s/it]
85%|████████▌ | 37523/43939 [60:33:39<10:09:57, 5.70s/it]
85%|████████▌ | 37524/43939 [60:33:44<10:08:59, 5.70s/it]
85%|████████▌ | 37525/43939 [60:33:50<10:08:36, 5.69s/it]
85%|████████▌ | 37526/43939 [60:33:56<10:07:49, 5.69s/it]
85%|████████▌ | 37527/43939 [60:34:01<10:08:07, 5.69s/it]
85%|████████▌ | 37528/43939 [60:34:07<10:07:54, 5.69s/it]
85%|████████▌ | 37529/43939 [60:34:13<10:09:31, 5.71s/it]
85%|████████▌ | 37530/43939 [60:34:19<10:09:49, 5.71s/it]
85%|████████▌ | 37531/43939 [60:34:24<10:08:39, 5.70s/it]
85%|████████▌ | 37532/43939 [60:34:30<10:07:18, 5.69s/it]
85%|████████▌ | 37533/43939 [60:34:36<10:06:57, 5.68s/it]
85%|████████▌ | 37534/43939 [60:34:41<10:07:48, 5.69s/it]
85%|████████▌ | 37535/43939 [60:34:47<10:07:03, 5.69s/it]
85%|████████▌ | 37536/43939 [60:34:53<10:08:08, 5.70s/it]
85%|████████▌ | 37537/43939 [60:34:58<10:07:48, 5.70s/it]
85%|████████▌ | 37538/43939 [60:35:04<10:07:26, 5.69s/it]
85%|████████▌ | 37539/43939 [60:35:10<10:06:57, 5.69s/it]
85%|████████▌ | 37540/43939 [60:35:15<10:06:25, 5.69s/it]
85%|████████▌ | 37541/43939 [60:35:21<10:08:07, 5.70s/it]
85%|████████▌ | 37542/43939 [60:35:27<10:15:40, 5.77s/it]
85%|████████▌ | 37543/43939 [60:35:33<10:13:14, 5.75s/it]
85%|████████▌ | 37544/43939 [60:35:39<10:11:41, 5.74s/it]
85%|████████▌ | 37545/43939 [60:35:44<10:09:44, 5.72s/it]
85%|████████▌ | 37546/43939 [60:35:50<10:09:10, 5.72s/it]
85%|████████▌ | 37547/43939 [60:35:56<10:07:21, 5.70s/it]
85%|████████▌ | 37548/43939 [60:36:01<10:07:18, 5.70s/it]
85%|████████▌ | 37549/43939 [60:36:07<10:07:02, 5.70s/it]
85%|████████▌ | 37550/43939 [60:36:13<10:08:03, 5.71s/it]
{'loss': 0.6423, 'grad_norm': 1.46367347240448, 'learning_rate': 3.152536733520442e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 463.7, 'epoch': 3.42}
85%|████████▌ | 37550/43939 [60:36:13<10:08:03, 5.71s/it]
85%|████████▌ | 37551/43939 [60:36:18<10:07:28, 5.71s/it]
85%|████████▌ | 37552/43939 [60:36:24<10:07:39, 5.71s/it]
85%|████████▌ | 37553/43939 [60:36:30<10:07:05, 5.70s/it]
85%|████████▌ | 37554/43939 [60:36:36<10:07:14, 5.71s/it]
85%|████████▌ | 37555/43939 [60:36:41<10:06:43, 5.70s/it]
85%|████████▌ | 37556/43939 [60:36:47<10:06:00, 5.70s/it]
85%|████████▌ | 37557/43939 [60:36:53<10:06:10, 5.70s/it]
85%|████████▌ | 37558/43939 [60:36:58<10:06:27, 5.70s/it]
85%|████████▌ | 37559/43939 [60:37:04<10:06:47, 5.71s/it]
85%|████████▌ | 37560/43939 [60:37:10<10:06:05, 5.70s/it]
85%|████████▌ | 37561/43939 [60:37:15<10:06:25, 5.70s/it]
85%|████████▌ | 37562/43939 [60:37:21<10:06:38, 5.71s/it]
85%|████████▌ | 37563/43939 [60:37:27<10:05:08, 5.69s/it]
85%|████████▌ | 37564/43939 [60:37:33<10:13:42, 5.78s/it]
85%|████████▌ | 37565/43939 [60:37:38<10:10:43, 5.75s/it]
85%|████████▌ | 37566/43939 [60:37:44<10:08:53, 5.73s/it]
85%|████████▌ | 37567/43939 [60:37:50<10:07:29, 5.72s/it]
86%|████████▌ | 37568/43939 [60:37:56<10:06:23, 5.71s/it]
86%|████████▌ | 37569/43939 [60:38:01<10:04:40, 5.70s/it]
86%|████████▌ | 37570/43939 [60:38:07<10:04:06, 5.69s/it]
86%|████████▌ | 37571/43939 [60:38:13<10:04:36, 5.70s/it]
86%|████████▌ | 37572/43939 [60:38:18<10:04:33, 5.70s/it]
86%|████████▌ | 37573/43939 [60:38:24<10:05:23, 5.71s/it]
86%|████████▌ | 37574/43939 [60:38:30<10:06:01, 5.71s/it]
86%|████████▌ | 37575/43939 [60:38:35<10:06:04, 5.71s/it]
86%|████████▌ | 37576/43939 [60:38:41<10:08:18, 5.74s/it]
86%|████████▌ | 37577/43939 [60:38:47<10:07:17, 5.73s/it]
86%|████████▌ | 37578/43939 [60:38:53<10:07:20, 5.73s/it]
86%|████████▌ | 37579/43939 [60:38:58<10:07:34, 5.73s/it]
86%|████████▌ | 37580/43939 [60:39:04<10:07:13, 5.73s/it]
86%|████████▌ | 37581/43939 [60:39:10<10:05:19, 5.71s/it]
86%|████████▌ | 37582/43939 [60:39:16<10:06:07, 5.72s/it]
86%|████████▌ | 37583/43939 [60:39:21<10:06:24, 5.72s/it]
86%|████████▌ | 37584/43939 [60:39:27<10:04:52, 5.71s/it]
86%|████████▌ | 37585/43939 [60:39:33<10:12:07, 5.78s/it]
86%|████████▌ | 37586/43939 [60:39:39<10:10:00, 5.76s/it]
86%|████████▌ | 37587/43939 [60:39:44<10:07:34, 5.74s/it]
86%|████████▌ | 37588/43939 [60:39:50<10:06:45, 5.73s/it]
86%|████████▌ | 37589/43939 [60:39:56<10:06:15, 5.73s/it]
86%|████████▌ | 37590/43939 [60:40:01<10:03:43, 5.71s/it]
86%|████████▌ | 37591/43939 [60:40:07<10:02:23, 5.69s/it]
86%|████████▌ | 37592/43939 [60:40:13<10:04:33, 5.72s/it]
86%|████████▌ | 37593/43939 [60:40:19<10:04:45, 5.72s/it]
86%|████████▌ | 37594/43939 [60:40:24<10:04:28, 5.72s/it]
86%|████████▌ | 37595/43939 [60:40:30<10:05:16, 5.72s/it]
86%|████████▌ | 37596/43939 [60:40:36<10:04:27, 5.72s/it]
86%|████████▌ | 37597/43939 [60:40:41<10:03:31, 5.71s/it]
86%|████████▌ | 37598/43939 [60:40:47<10:05:12, 5.73s/it]
86%|████████▌ | 37599/43939 [60:40:53<10:04:40, 5.72s/it]
86%|████████▌ | 37600/43939 [60:40:59<10:04:15, 5.72s/it]
{'loss': 0.663, 'grad_norm': 1.658532977104187, 'learning_rate': 3.1044377482120246e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 492.04, 'epoch': 3.42}
86%|████████▌ | 37600/43939 [60:40:59<10:04:15, 5.72s/it]
86%|████████▌ | 37601/43939 [60:41:04<10:04:28, 5.72s/it]
86%|████████▌ | 37602/43939 [60:41:10<10:03:30, 5.71s/it]
86%|████████▌ | 37603/43939 [60:41:16<10:01:51, 5.70s/it]
86%|████████▌ | 37604/43939 [60:41:21<10:01:21, 5.70s/it]
86%|████████▌ | 37605/43939 [60:41:27<10:01:00, 5.69s/it]
86%|████████▌ | 37606/43939 [60:41:33<10:01:13, 5.70s/it]
86%|████████▌ | 37607/43939 [60:41:39<10:09:48, 5.78s/it]
86%|████████▌ | 37608/43939 [60:41:44<10:07:51, 5.76s/it]
86%|████████▌ | 37609/43939 [60:41:50<10:05:33, 5.74s/it]
86%|████████▌ | 37610/43939 [60:41:56<10:03:22, 5.72s/it]
86%|████████▌ | 37611/43939 [60:42:01<10:01:43, 5.71s/it]
86%|████████▌ | 37612/43939 [60:42:07<10:01:26, 5.70s/it]
86%|████████▌ | 37613/43939 [60:42:13<10:01:23, 5.70s/it]
86%|████████▌ | 37614/43939 [60:42:19<10:01:53, 5.71s/it]
86%|████████▌ | 37615/43939 [60:42:24<10:02:08, 5.71s/it]
86%|████████▌ | 37616/43939 [60:42:30<10:02:14, 5.71s/it]
86%|████████▌ | 37617/43939 [60:42:36<10:01:47, 5.71s/it]
86%|████████▌ | 37618/43939 [60:42:41<10:01:43, 5.71s/it]
86%|████████▌ | 37619/43939 [60:42:47<10:00:49, 5.70s/it]
86%|████████▌ | 37620/43939 [60:42:53<10:02:15, 5.72s/it]
86%|████████▌ | 37621/43939 [60:42:59<10:01:13, 5.71s/it]
86%|████████▌ | 37622/43939 [60:43:04<10:01:14, 5.71s/it]
86%|████████▌ | 37623/43939 [60:43:10<10:02:02, 5.72s/it]
86%|████████▌ | 37624/43939 [60:43:16<10:02:13, 5.72s/it]
86%|████████▌ | 37625/43939 [60:43:22<10:02:29, 5.73s/it]
86%|████████▌ | 37626/43939 [60:43:27<10:01:09, 5.71s/it]
86%|████████▌ | 37627/43939 [60:43:33<10:01:36, 5.72s/it]
86%|████████▌ | 37628/43939 [60:43:39<10:00:25, 5.71s/it]
86%|████████▌ | 37629/43939 [60:43:45<10:08:34, 5.79s/it]
86%|████████▌ | 37630/43939 [60:43:50<10:06:59, 5.77s/it]
86%|████████▌ | 37631/43939 [60:43:56<10:04:49, 5.75s/it]
86%|████████▌ | 37632/43939 [60:44:02<10:03:16, 5.74s/it]
86%|████████▌ | 37633/43939 [60:44:07<10:00:57, 5.72s/it]
86%|████████▌ | 37634/43939 [60:44:13<10:01:40, 5.73s/it]
86%|████████▌ | 37635/43939 [60:44:19<10:00:55, 5.72s/it]
86%|████████▌ | 37636/43939 [60:44:25<10:00:17, 5.71s/it]
86%|████████▌ | 37637/43939 [60:44:30<9:59:02, 5.70s/it]
86%|████████▌ | 37638/43939 [60:44:36<9:58:18, 5.70s/it]
86%|████████▌ | 37639/43939 [60:44:42<9:57:50, 5.69s/it]
86%|████████▌ | 37640/43939 [60:44:47<9:57:49, 5.69s/it]
86%|████████▌ | 37641/43939 [60:44:53<9:58:12, 5.70s/it]
86%|████████▌ | 37642/43939 [60:44:59<9:56:55, 5.69s/it]
86%|████████▌ | 37643/43939 [60:45:04<9:57:14, 5.69s/it]
86%|████████▌ | 37644/43939 [60:45:10<9:58:01, 5.70s/it]
86%|████████▌ | 37645/43939 [60:45:16<9:58:43, 5.71s/it]
86%|████████▌ | 37646/43939 [60:45:22<9:58:49, 5.71s/it]
86%|████████▌ | 37647/43939 [60:45:27<10:01:35, 5.74s/it]
86%|████████▌ | 37648/43939 [60:45:33<9:59:40, 5.72s/it]
86%|████████▌ | 37649/43939 [60:45:39<9:59:56, 5.72s/it]
86%|████████▌ | 37650/43939 [60:45:45<10:06:47, 5.79s/it]
{'loss': 0.6693, 'grad_norm': 1.5503042936325073, 'learning_rate': 3.056684216851999e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.13, 'epoch': 3.43}
86%|████████▌ | 37650/43939 [60:45:45<10:06:47, 5.79s/it]
86%|████████▌ | 37651/43939 [60:45:50<10:04:13, 5.77s/it]
86%|████████▌ | 37652/43939 [60:45:56<10:01:43, 5.74s/it]
86%|████████▌ | 37653/43939 [60:46:02<10:00:27, 5.73s/it]
86%|████████▌ | 37654/43939 [60:46:07<9:59:54, 5.73s/it]
86%|████████▌ | 37655/43939 [60:46:13<9:58:57, 5.72s/it]
86%|████████▌ | 37656/43939 [60:46:19<9:57:11, 5.70s/it]
86%|████████▌ | 37657/43939 [60:46:25<9:56:27, 5.70s/it]
86%|████████▌ | 37658/43939 [60:46:30<9:56:05, 5.69s/it]
86%|████████▌ | 37659/43939 [60:46:36<9:56:25, 5.70s/it]
86%|████████▌ | 37660/43939 [60:46:42<9:55:43, 5.69s/it]
86%|████████▌ | 37661/43939 [60:46:47<9:55:12, 5.69s/it]
86%|████████▌ | 37662/43939 [60:46:53<9:56:38, 5.70s/it]
86%|████████▌ | 37663/43939 [60:46:59<9:55:51, 5.70s/it]
86%|████████▌ | 37664/43939 [60:47:04<9:57:49, 5.72s/it]
86%|████████▌ | 37665/43939 [60:47:10<9:57:20, 5.71s/it]
86%|████████▌ | 37666/43939 [60:47:16<9:56:30, 5.71s/it]
86%|████████▌ | 37667/43939 [60:47:22<9:56:04, 5.70s/it]
86%|████████▌ | 37668/43939 [60:47:27<9:55:13, 5.69s/it]
86%|████████▌ | 37669/43939 [60:47:33<9:55:31, 5.70s/it]
86%|████████▌ | 37670/43939 [60:47:39<10:02:27, 5.77s/it]
86%|████████▌ | 37671/43939 [60:47:45<10:00:20, 5.75s/it]
86%|████████▌ | 37672/43939 [60:47:50<9:58:55, 5.73s/it]
86%|████████▌ | 37673/43939 [60:47:56<9:59:15, 5.74s/it]
86%|████████▌ | 37674/43939 [60:48:02<9:57:56, 5.73s/it]
86%|████████▌ | 37675/43939 [60:48:07<9:56:12, 5.71s/it]
86%|████████▌ | 37676/43939 [60:48:13<9:55:26, 5.70s/it]
86%|████████▌ | 37677/43939 [60:48:19<9:55:40, 5.71s/it]
86%|████████▌ | 37678/43939 [60:48:24<9:54:06, 5.69s/it]
86%|████████▌ | 37679/43939 [60:48:30<9:53:23, 5.69s/it]
86%|████████▌ | 37680/43939 [60:48:36<9:53:28, 5.69s/it]
86%|████████▌ | 37681/43939 [60:48:42<9:53:52, 5.69s/it]
86%|████████▌ | 37682/43939 [60:48:47<9:53:44, 5.69s/it]
86%|████████▌ | 37683/43939 [60:48:53<9:54:25, 5.70s/it]
86%|████████▌ | 37684/43939 [60:48:59<9:54:05, 5.70s/it]
86%|████████▌ | 37685/43939 [60:49:04<9:54:50, 5.71s/it]
86%|████████▌ | 37686/43939 [60:49:10<9:55:22, 5.71s/it]
86%|████████▌ | 37687/43939 [60:49:16<9:54:42, 5.71s/it]
86%|████████▌ | 37688/43939 [60:49:22<9:56:11, 5.72s/it]
86%|████████▌ | 37689/43939 [60:49:27<9:57:43, 5.74s/it]
86%|████████▌ | 37690/43939 [60:49:33<9:56:09, 5.72s/it]
86%|████████▌ | 37691/43939 [60:49:39<9:55:05, 5.71s/it]
86%|████████▌ | 37692/43939 [60:49:45<10:01:12, 5.77s/it]
86%|████████▌ | 37693/43939 [60:49:50<9:59:38, 5.76s/it]
86%|████████▌ | 37694/43939 [60:49:56<9:57:38, 5.74s/it]
86%|████████▌ | 37695/43939 [60:50:02<9:57:43, 5.74s/it]
86%|████████▌ | 37696/43939 [60:50:08<9:57:44, 5.74s/it]
86%|████████▌ | 37697/43939 [60:50:13<9:56:07, 5.73s/it]
86%|████████▌ | 37698/43939 [60:50:19<9:55:51, 5.73s/it]
86%|████████▌ | 37699/43939 [60:50:25<9:55:11, 5.72s/it]
86%|████████▌ | 37700/43939 [60:50:30<9:54:14, 5.71s/it]
{'loss': 0.6674, 'grad_norm': 1.3065074682235718, 'learning_rate': 3.0092768928645375e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 498.52, 'epoch': 3.43}
86%|████████▌ | 37700/43939 [60:50:30<9:54:14, 5.71s/it]
86%|████████▌ | 37701/43939 [60:50:36<9:54:52, 5.72s/it]
86%|████████▌ | 37702/43939 [60:50:42<9:55:02, 5.72s/it]
86%|████████▌ | 37703/43939 [60:50:48<9:54:07, 5.72s/it]
86%|████████▌ | 37704/43939 [60:50:53<9:52:44, 5.70s/it]
86%|████████▌ | 37705/43939 [60:50:59<9:53:43, 5.71s/it]
86%|████████▌ | 37706/43939 [60:51:05<9:53:16, 5.71s/it]
86%|████████▌ | 37707/43939 [60:51:10<9:53:09, 5.71s/it]
86%|████████▌ | 37708/43939 [60:51:16<9:53:51, 5.72s/it]
86%|████████▌ | 37709/43939 [60:51:22<9:52:10, 5.70s/it]
86%|████████▌ | 37710/43939 [60:51:27<9:52:03, 5.70s/it]
86%|████████▌ | 37711/43939 [60:51:33<9:52:46, 5.71s/it]
86%|████████▌ | 37712/43939 [60:51:39<9:51:18, 5.70s/it]
86%|████████▌ | 37713/43939 [60:51:45<9:50:33, 5.69s/it]
86%|████████▌ | 37714/43939 [60:51:50<9:58:05, 5.76s/it]
86%|████████▌ | 37715/43939 [60:51:56<9:56:07, 5.75s/it]
86%|████████▌ | 37716/43939 [60:52:02<9:55:55, 5.75s/it]
86%|████████▌ | 37717/43939 [60:52:08<9:55:49, 5.75s/it]
86%|████████▌ | 37718/43939 [60:52:13<9:54:13, 5.73s/it]
86%|████████▌ | 37719/43939 [60:52:19<9:53:38, 5.73s/it]
86%|████████▌ | 37720/43939 [60:52:25<9:53:37, 5.73s/it]
86%|████████▌ | 37721/43939 [60:52:31<9:52:33, 5.72s/it]
86%|████████▌ | 37722/43939 [60:52:36<9:51:02, 5.70s/it]
86%|████████▌ | 37723/43939 [60:52:42<9:50:03, 5.70s/it]
86%|████████▌ | 37724/43939 [60:52:48<9:49:58, 5.70s/it]
86%|████████▌ | 37725/43939 [60:52:53<9:50:48, 5.70s/it]
86%|████████▌ | 37726/43939 [60:52:59<9:49:32, 5.69s/it]
86%|████████▌ | 37727/43939 [60:53:05<9:49:16, 5.69s/it]
86%|████████▌ | 37728/43939 [60:53:10<9:51:11, 5.71s/it]
86%|████████▌ | 37729/43939 [60:53:16<9:50:13, 5.70s/it]
86%|████████▌ | 37730/43939 [60:53:22<9:50:15, 5.70s/it]
86%|████████▌ | 37731/43939 [60:53:27<9:49:20, 5.70s/it]
86%|████████▌ | 37732/43939 [60:53:33<9:50:39, 5.71s/it]
86%|████████▌ | 37733/43939 [60:53:39<9:51:19, 5.72s/it]
86%|████████▌ | 37734/43939 [60:53:45<9:52:03, 5.73s/it]
86%|████████▌ | 37735/43939 [60:53:51<9:57:41, 5.78s/it]
86%|████████▌ | 37736/43939 [60:53:56<9:54:58, 5.76s/it]
86%|████████▌ | 37737/43939 [60:54:02<9:52:38, 5.73s/it]
86%|████████▌ | 37738/43939 [60:54:08<9:50:52, 5.72s/it]
86%|████████▌ | 37739/43939 [60:54:13<9:50:21, 5.71s/it]
86%|████████▌ | 37740/43939 [60:54:19<9:50:21, 5.71s/it]
86%|████████▌ | 37741/43939 [60:54:25<9:49:31, 5.71s/it]
86%|████████▌ | 37742/43939 [60:54:30<9:48:12, 5.70s/it]
86%|████████▌ | 37743/43939 [60:54:36<9:48:20, 5.70s/it]
86%|████████▌ | 37744/43939 [60:54:42<9:48:31, 5.70s/it]
86%|████████▌ | 37745/43939 [60:54:48<9:47:50, 5.69s/it]
86%|████████▌ | 37746/43939 [60:54:53<9:47:54, 5.70s/it]
86%|████████▌ | 37747/43939 [60:54:59<9:47:16, 5.69s/it]
86%|████████▌ | 37748/43939 [60:55:05<9:47:19, 5.69s/it]
86%|████████▌ | 37749/43939 [60:55:10<9:48:02, 5.70s/it]
86%|████████▌ | 37750/43939 [60:55:16<9:47:15, 5.69s/it]
{'loss': 0.6659, 'grad_norm': 1.7457201480865479, 'learning_rate': 2.9622165242115808e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.49, 'epoch': 3.44}
86%|████████▌ | 37750/43939 [60:55:16<9:47:15, 5.69s/it]
86%|████████▌ | 37751/43939 [60:55:22<9:47:26, 5.70s/it]
86%|████████▌ | 37752/43939 [60:55:27<9:47:33, 5.70s/it]
86%|████████▌ | 37753/43939 [60:55:33<9:47:38, 5.70s/it]
86%|████████▌ | 37754/43939 [60:55:39<9:47:39, 5.70s/it]
86%|████████▌ | 37755/43939 [60:55:44<9:47:00, 5.70s/it]
86%|████████▌ | 37756/43939 [60:55:50<9:53:04, 5.76s/it]
86%|████████▌ | 37757/43939 [60:55:56<9:51:43, 5.74s/it]
86%|████████▌ | 37758/43939 [60:56:02<9:51:27, 5.74s/it]
86%|████████▌ | 37759/43939 [60:56:08<9:50:15, 5.73s/it]
86%|████████▌ | 37760/43939 [60:56:13<9:51:17, 5.74s/it]
86%|████████▌ | 37761/43939 [60:56:19<9:51:52, 5.75s/it]
86%|████████▌ | 37762/43939 [60:56:25<9:49:51, 5.73s/it]
86%|████████▌ | 37763/43939 [60:56:31<9:50:51, 5.74s/it]
86%|████████▌ | 37764/43939 [60:56:36<9:49:57, 5.73s/it]
86%|████████▌ | 37765/43939 [60:56:42<9:48:34, 5.72s/it]
86%|████████▌ | 37766/43939 [60:56:48<9:49:28, 5.73s/it]
86%|████████▌ | 37767/43939 [60:56:53<9:48:23, 5.72s/it]
86%|████████▌ | 37768/43939 [60:56:59<9:48:59, 5.73s/it]
86%|████████▌ | 37769/43939 [60:57:05<9:47:18, 5.71s/it]
86%|████████▌ | 37770/43939 [60:57:10<9:46:57, 5.71s/it]
86%|████████▌ | 37771/43939 [60:57:16<9:48:37, 5.73s/it]
86%|████████▌ | 37772/43939 [60:57:22<9:48:06, 5.72s/it]
86%|████████▌ | 37773/43939 [60:57:28<9:49:43, 5.74s/it]
86%|████████▌ | 37774/43939 [60:57:33<9:47:33, 5.72s/it]
86%|████████▌ | 37775/43939 [60:57:39<9:48:09, 5.73s/it]
86%|████████▌ | 37776/43939 [60:57:45<9:48:34, 5.73s/it]
86%|████████▌ | 37777/43939 [60:57:51<9:46:15, 5.71s/it]
86%|████████▌ | 37778/43939 [60:57:56<9:51:32, 5.76s/it]
86%|████████▌ | 37779/43939 [60:58:02<9:49:20, 5.74s/it]
86%|████████▌ | 37780/43939 [60:58:08<9:46:38, 5.72s/it]
86%|████████▌ | 37781/43939 [60:58:13<9:45:32, 5.71s/it]
86%|████████▌ | 37782/43939 [60:58:19<9:46:12, 5.71s/it]
86%|████████▌ | 37783/43939 [60:58:25<9:46:17, 5.71s/it]
86%|████████▌ | 37784/43939 [60:58:31<9:47:42, 5.73s/it]
86%|████████▌ | 37785/43939 [60:58:36<9:46:58, 5.72s/it]
86%|████████▌ | 37786/43939 [60:58:42<9:46:07, 5.72s/it]
86%|████████▌ | 37787/43939 [60:58:48<9:44:53, 5.70s/it]
86%|████████▌ | 37788/43939 [60:58:53<9:44:13, 5.70s/it]
86%|████████▌ | 37789/43939 [60:58:59<9:44:40, 5.70s/it]
86%|████████▌ | 37790/43939 [60:59:05<9:46:15, 5.72s/it]
86%|████████▌ | 37791/43939 [60:59:11<9:45:20, 5.71s/it]
86%|████████▌ | 37792/43939 [60:59:16<9:45:02, 5.71s/it]
86%|████████▌ | 37793/43939 [60:59:22<9:44:25, 5.71s/it]
86%|████████▌ | 37794/43939 [60:59:28<9:44:54, 5.71s/it]
86%|████████▌ | 37795/43939 [60:59:33<9:45:14, 5.72s/it]
86%|████████▌ | 37796/43939 [60:59:39<9:44:49, 5.71s/it]
86%|████████▌ | 37797/43939 [60:59:45<9:44:33, 5.71s/it]
86%|████████▌ | 37798/43939 [60:59:51<9:44:22, 5.71s/it]
86%|████████▌ | 37799/43939 [60:59:56<9:45:03, 5.72s/it]
86%|████████▌ | 37800/43939 [61:00:02<9:52:38, 5.79s/it]
{'loss': 0.6687, 'grad_norm': 1.7811120748519897, 'learning_rate': 2.9155038533810325e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.01, 'epoch': 3.44}
86%|████████▌ | 37800/43939 [61:00:02<9:52:38, 5.79s/it]
86%|████████▌ | 37801/43939 [61:00:08<9:50:22, 5.77s/it]
86%|████████▌ | 37802/43939 [61:00:14<9:49:03, 5.76s/it]
86%|████████▌ | 37803/43939 [61:00:19<9:48:40, 5.76s/it]
86%|████████▌ | 37804/43939 [61:00:25<9:46:55, 5.74s/it]
86%|████████▌ | 37805/43939 [61:00:31<9:45:22, 5.73s/it]
86%|████████▌ | 37806/43939 [61:00:37<9:43:54, 5.71s/it]
86%|████████▌ | 37807/43939 [61:00:42<9:44:24, 5.72s/it]
86%|████████▌ | 37808/43939 [61:00:48<9:43:31, 5.71s/it]
86%|████████▌ | 37809/43939 [61:00:54<9:42:17, 5.70s/it]
86%|████████▌ | 37810/43939 [61:00:59<9:42:12, 5.70s/it]
86%|████████▌ | 37811/43939 [61:01:05<9:43:00, 5.71s/it]
86%|████████▌ | 37812/43939 [61:01:11<9:43:14, 5.71s/it]
86%|████████▌ | 37813/43939 [61:01:17<9:44:29, 5.72s/it]
86%|████████▌ | 37814/43939 [61:01:22<9:43:05, 5.71s/it]
86%|████████▌ | 37815/43939 [61:01:28<9:41:05, 5.69s/it]
86%|████████▌ | 37816/43939 [61:01:34<9:41:03, 5.69s/it]
86%|████████▌ | 37817/43939 [61:01:39<9:43:18, 5.72s/it]
86%|████████▌ | 37818/43939 [61:01:45<9:43:48, 5.72s/it]
86%|████████▌ | 37819/43939 [61:01:51<9:43:34, 5.72s/it]
86%|████████▌ | 37820/43939 [61:01:57<9:43:06, 5.72s/it]
86%|████████▌ | 37821/43939 [61:02:03<9:50:49, 5.79s/it]
86%|████████▌ | 37822/43939 [61:02:08<9:48:28, 5.77s/it]
86%|████████▌ | 37823/43939 [61:02:14<9:47:13, 5.76s/it]
86%|████████▌ | 37824/43939 [61:02:20<9:45:46, 5.75s/it]
86%|████████▌ | 37825/43939 [61:02:25<9:44:21, 5.73s/it]
86%|████████▌ | 37826/43939 [61:02:31<9:42:35, 5.72s/it]
86%|████████▌ | 37827/43939 [61:02:37<9:41:08, 5.70s/it]
86%|████████▌ | 37828/43939 [61:02:42<9:40:58, 5.70s/it]
86%|████████▌ | 37829/43939 [61:02:48<9:41:15, 5.71s/it]
86%|████████▌ | 37830/43939 [61:02:54<9:41:40, 5.71s/it]
86%|████████▌ | 37831/43939 [61:03:00<9:42:32, 5.72s/it]
86%|████████▌ | 37832/43939 [61:03:05<9:43:55, 5.74s/it]
86%|████████▌ | 37833/43939 [61:03:11<9:42:45, 5.73s/it]
86%|████████▌ | 37834/43939 [61:03:17<9:41:02, 5.71s/it]
86%|████████▌ | 37835/43939 [61:03:22<9:40:14, 5.70s/it]
86%|████████▌ | 37836/43939 [61:03:28<9:40:08, 5.70s/it]
86%|████████▌ | 37837/43939 [61:03:34<9:40:01, 5.70s/it]
86%|████████▌ | 37838/43939 [61:03:40<9:38:45, 5.69s/it]
86%|████████▌ | 37839/43939 [61:03:45<9:38:42, 5.69s/it]
86%|████████▌ | 37840/43939 [61:03:51<9:38:54, 5.70s/it]
86%|████████▌ | 37841/43939 [61:03:57<9:38:33, 5.69s/it]
86%|████████▌ | 37842/43939 [61:04:02<9:38:29, 5.69s/it]
86%|████████▌ | 37843/43939 [61:04:08<9:47:55, 5.79s/it]
86%|████████▌ | 37844/43939 [61:04:14<9:44:18, 5.75s/it]
86%|████████▌ | 37845/43939 [61:04:20<9:42:33, 5.74s/it]
86%|████████▌ | 37846/43939 [61:04:25<9:41:55, 5.73s/it]
86%|████████▌ | 37847/43939 [61:04:31<9:42:27, 5.74s/it]
86%|████████▌ | 37848/43939 [61:04:37<9:42:02, 5.73s/it]
86%|████████▌ | 37849/43939 [61:04:43<9:41:45, 5.73s/it]
86%|████████▌ | 37850/43939 [61:04:48<9:39:20, 5.71s/it]
{'loss': 0.669, 'grad_norm': 2.163778781890869, 'learning_rate': 2.869139617375055e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 537.78, 'epoch': 3.45}
86%|████████▌ | 37850/43939 [61:04:48<9:39:20, 5.71s/it]
86%|████████▌ | 37851/43939 [61:04:54<9:38:04, 5.70s/it]
86%|████████▌ | 37852/43939 [61:05:00<9:36:26, 5.68s/it]
86%|████████▌ | 37853/43939 [61:05:05<9:37:03, 5.69s/it]
86%|████████▌ | 37854/43939 [61:05:11<9:38:20, 5.70s/it]
86%|████████▌ | 37855/43939 [61:05:17<9:38:33, 5.71s/it]
86%|████████▌ | 37856/43939 [61:05:22<9:38:53, 5.71s/it]
86%|████████▌ | 37857/43939 [61:05:28<9:38:40, 5.71s/it]
86%|████████▌ | 37858/43939 [61:05:34<9:37:16, 5.70s/it]
86%|████████▌ | 37859/43939 [61:05:40<9:37:07, 5.70s/it]
86%|████████▌ | 37860/43939 [61:05:45<9:36:56, 5.69s/it]
86%|████████▌ | 37861/43939 [61:05:51<9:36:51, 5.69s/it]
86%|████████▌ | 37862/43939 [61:05:57<9:37:03, 5.70s/it]
86%|████████▌ | 37863/43939 [61:06:02<9:37:06, 5.70s/it]
86%|████████▌ | 37864/43939 [61:06:08<9:39:06, 5.72s/it]
86%|████████▌ | 37865/43939 [61:06:14<9:45:42, 5.79s/it]
86%|████████▌ | 37866/43939 [61:06:20<9:42:57, 5.76s/it]
86%|████████▌ | 37867/43939 [61:06:25<9:42:19, 5.75s/it]
86%|████████▌ | 37868/43939 [61:06:31<9:41:09, 5.74s/it]
86%|████████▌ | 37869/43939 [61:06:37<9:40:17, 5.74s/it]
86%|████████▌ | 37870/43939 [61:06:43<9:40:03, 5.73s/it]
86%|████████▌ | 37871/43939 [61:06:48<9:38:45, 5.72s/it]
86%|████████▌ | 37872/43939 [61:06:54<9:38:27, 5.72s/it]
86%|████████▌ | 37873/43939 [61:07:00<9:37:13, 5.71s/it]
86%|████████▌ | 37874/43939 [61:07:05<9:38:07, 5.72s/it]
86%|████████▌ | 37875/43939 [61:07:11<9:36:40, 5.71s/it]
86%|████████▌ | 37876/43939 [61:07:17<9:37:01, 5.71s/it]
86%|████████▌ | 37877/43939 [61:07:23<9:38:16, 5.72s/it]
86%|████████▌ | 37878/43939 [61:07:28<9:36:56, 5.71s/it]
86%|████████▌ | 37879/43939 [61:07:34<9:38:48, 5.73s/it]
86%|████████▌ | 37880/43939 [61:07:40<9:38:04, 5.72s/it]
86%|████████▌ | 37881/43939 [61:07:46<9:39:35, 5.74s/it]
86%|████████▌ | 37882/43939 [61:07:51<9:39:49, 5.74s/it]
86%|████████▌ | 37883/43939 [61:07:57<9:38:42, 5.73s/it]
86%|████████▌ | 37884/43939 [61:08:03<9:37:57, 5.73s/it]
86%|████████▌ | 37885/43939 [61:08:08<9:37:12, 5.72s/it]
86%|████████▌ | 37886/43939 [61:08:14<9:43:20, 5.78s/it]
86%|████████▌ | 37887/43939 [61:08:20<9:40:05, 5.75s/it]
86%|████████▌ | 37888/43939 [61:08:26<9:38:58, 5.74s/it]
86%|████████▌ | 37889/43939 [61:08:31<9:37:34, 5.73s/it]
86%|████████▌ | 37890/43939 [61:08:37<9:37:00, 5.72s/it]
86%|████████▌ | 37891/43939 [61:08:43<9:36:27, 5.72s/it]
86%|████████▌ | 37892/43939 [61:08:49<9:37:38, 5.73s/it]
86%|████████▌ | 37893/43939 [61:08:54<9:36:42, 5.72s/it]
86%|████████▌ | 37894/43939 [61:09:00<9:35:27, 5.71s/it]
86%|████████▌ | 37895/43939 [61:09:06<9:34:57, 5.71s/it]
86%|████████▌ | 37896/43939 [61:09:11<9:34:46, 5.71s/it]
86%|████████▌ | 37897/43939 [61:09:17<9:33:41, 5.70s/it]
86%|████████▋ | 37898/43939 [61:09:23<9:35:14, 5.71s/it]
86%|████████▋ | 37899/43939 [61:09:29<9:34:43, 5.71s/it]
86%|████████▋ | 37900/43939 [61:09:34<9:34:42, 5.71s/it]
{'loss': 0.6516, 'grad_norm': 1.4239171743392944, 'learning_rate': 2.8231245476984192e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.68, 'epoch': 3.45}
86%|████████▋ | 37900/43939 [61:09:34<9:34:42, 5.71s/it]
86%|████████▋ | 37901/43939 [61:09:40<9:33:48, 5.70s/it]
86%|████████▋ | 37902/43939 [61:09:46<9:33:02, 5.70s/it]
86%|████████▋ | 37903/43939 [61:09:51<9:35:14, 5.72s/it]
86%|████████▋ | 37904/43939 [61:09:57<9:33:42, 5.70s/it]
86%|████████▋ | 37905/43939 [61:10:03<9:34:02, 5.71s/it]
86%|████████▋ | 37906/43939 [61:10:09<9:39:52, 5.77s/it]
86%|████████▋ | 37907/43939 [61:10:14<9:37:27, 5.74s/it]
86%|████████▋ | 37908/43939 [61:10:20<9:35:43, 5.73s/it]
86%|████████▋ | 37909/43939 [61:10:26<9:34:37, 5.72s/it]
86%|████████▋ | 37910/43939 [61:10:31<9:34:25, 5.72s/it]
86%|████████▋ | 37911/43939 [61:10:37<9:33:48, 5.71s/it]
86%|████████▋ | 37912/43939 [61:10:43<9:33:49, 5.71s/it]
86%|████████▋ | 37913/43939 [61:10:49<9:33:05, 5.71s/it]
86%|████████▋ | 37914/43939 [61:10:54<9:31:34, 5.69s/it]
86%|████████▋ | 37915/43939 [61:11:00<9:32:00, 5.70s/it]
86%|████████▋ | 37916/43939 [61:11:06<9:32:20, 5.70s/it]
86%|████████▋ | 37917/43939 [61:11:11<9:32:47, 5.71s/it]
86%|████████▋ | 37918/43939 [61:11:17<9:32:15, 5.70s/it]
86%|████████▋ | 37919/43939 [61:11:23<9:32:03, 5.70s/it]
86%|████████▋ | 37920/43939 [61:11:28<9:31:57, 5.70s/it]
86%|████████▋ | 37921/43939 [61:11:34<9:31:05, 5.69s/it]
86%|████████▋ | 37922/43939 [61:11:40<9:32:10, 5.71s/it]
86%|████████▋ | 37923/43939 [61:11:46<9:31:42, 5.70s/it]
86%|████████▋ | 37924/43939 [61:11:51<9:31:11, 5.70s/it]
86%|████████▋ | 37925/43939 [61:11:57<9:31:50, 5.71s/it]
86%|████████▋ | 37926/43939 [61:12:03<9:31:05, 5.70s/it]
86%|████████▋ | 37927/43939 [61:12:08<9:30:28, 5.69s/it]
86%|████████▋ | 37928/43939 [61:12:14<9:37:09, 5.76s/it]
86%|████████▋ | 37929/43939 [61:12:20<9:34:01, 5.73s/it]
86%|████████▋ | 37930/43939 [61:12:26<9:33:24, 5.73s/it]
86%|████████▋ | 37931/43939 [61:12:31<9:31:52, 5.71s/it]
86%|████████▋ | 37932/43939 [61:12:37<9:31:43, 5.71s/it]
86%|████████▋ | 37933/43939 [61:12:43<9:30:38, 5.70s/it]
86%|████████▋ | 37934/43939 [61:12:48<9:29:47, 5.69s/it]
86%|████████▋ | 37935/43939 [61:12:54<9:29:30, 5.69s/it]
86%|████████▋ | 37936/43939 [61:13:00<9:29:13, 5.69s/it]
86%|████████▋ | 37937/43939 [61:13:05<9:28:04, 5.68s/it]
86%|████████▋ | 37938/43939 [61:13:11<9:28:56, 5.69s/it]
86%|████████▋ | 37939/43939 [61:13:17<9:29:27, 5.69s/it]
86%|████████▋ | 37940/43939 [61:13:22<9:28:36, 5.69s/it]
86%|████████▋ | 37941/43939 [61:13:28<9:28:17, 5.68s/it]
86%|████████▋ | 37942/43939 [61:13:34<9:28:30, 5.69s/it]
86%|████████▋ | 37943/43939 [61:13:40<9:29:15, 5.70s/it]
86%|████████▋ | 37944/43939 [61:13:45<9:29:26, 5.70s/it]
86%|████████▋ | 37945/43939 [61:13:51<9:28:40, 5.69s/it]
86%|████████▋ | 37946/43939 [61:13:57<9:28:16, 5.69s/it]
86%|████████▋ | 37947/43939 [61:14:02<9:29:20, 5.70s/it]
86%|████████▋ | 37948/43939 [61:14:08<9:29:53, 5.71s/it]
86%|████████▋ | 37949/43939 [61:14:14<9:30:26, 5.71s/it]
86%|████████▋ | 37950/43939 [61:14:20<9:35:48, 5.77s/it]
{'loss': 0.6496, 'grad_norm': 1.7351467609405518, 'learning_rate': 2.7774593703469852e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.95, 'epoch': 3.45}
86%|████████▋ | 37950/43939 [61:14:20<9:35:48, 5.77s/it]
86%|████████▋ | 37951/43939 [61:14:25<9:33:48, 5.75s/it]
86%|████████▋ | 37952/43939 [61:14:31<9:31:53, 5.73s/it]
86%|████████▋ | 37953/43939 [61:14:37<9:30:24, 5.72s/it]
86%|████████▋ | 37954/43939 [61:14:43<9:30:27, 5.72s/it]
86%|████████▋ | 37955/43939 [61:14:48<9:29:00, 5.71s/it]
86%|████████▋ | 37956/43939 [61:14:54<9:28:05, 5.70s/it]
86%|████████▋ | 37957/43939 [61:15:00<9:28:03, 5.70s/it]
86%|████████▋ | 37958/43939 [61:15:05<9:27:21, 5.69s/it]
86%|████████▋ | 37959/43939 [61:15:11<9:27:13, 5.69s/it]
86%|████████▋ | 37960/43939 [61:15:17<9:27:36, 5.70s/it]
86%|████████▋ | 37961/43939 [61:15:22<9:28:21, 5.70s/it]
86%|████████▋ | 37962/43939 [61:15:28<9:27:43, 5.70s/it]
86%|████████▋ | 37963/43939 [61:15:34<9:28:07, 5.70s/it]
86%|████████▋ | 37964/43939 [61:15:40<9:28:49, 5.71s/it]
86%|████████▋ | 37965/43939 [61:15:45<9:29:46, 5.72s/it]
86%|████████▋ | 37966/43939 [61:15:51<9:29:03, 5.72s/it]
86%|████████▋ | 37967/43939 [61:15:57<9:28:45, 5.71s/it]
86%|████████▋ | 37968/43939 [61:16:02<9:28:36, 5.71s/it]
86%|████████▋ | 37969/43939 [61:16:08<9:28:05, 5.71s/it]
86%|████████▋ | 37970/43939 [61:16:14<9:28:26, 5.71s/it]
86%|████████▋ | 37971/43939 [61:16:20<9:34:45, 5.78s/it]
86%|████████▋ | 37972/43939 [61:16:25<9:33:34, 5.77s/it]
86%|████████▋ | 37973/43939 [61:16:31<9:30:44, 5.74s/it]
86%|████████▋ | 37974/43939 [61:16:37<9:29:19, 5.73s/it]
86%|████████▋ | 37975/43939 [61:16:43<9:28:58, 5.72s/it]
86%|████████▋ | 37976/43939 [61:16:48<9:27:53, 5.71s/it]
86%|████████▋ | 37977/43939 [61:16:54<9:27:05, 5.71s/it]
86%|████████▋ | 37978/43939 [61:17:00<9:26:32, 5.70s/it]
86%|████████▋ | 37979/43939 [61:17:05<9:26:26, 5.70s/it]
86%|████████▋ | 37980/43939 [61:17:11<9:24:47, 5.69s/it]
86%|████████▋ | 37981/43939 [61:17:17<9:24:04, 5.68s/it]
86%|████████▋ | 37982/43939 [61:17:22<9:24:58, 5.69s/it]
86%|████████▋ | 37983/43939 [61:17:28<9:25:02, 5.69s/it]
86%|████████▋ | 37984/43939 [61:17:34<9:26:24, 5.71s/it]
86%|████████▋ | 37985/43939 [61:17:40<9:25:56, 5.70s/it]
86%|████████▋ | 37986/43939 [61:17:45<9:25:17, 5.70s/it]
86%|████████▋ | 37987/43939 [61:17:51<9:25:01, 5.70s/it]
86%|████████▋ | 37988/43939 [61:17:57<9:24:11, 5.69s/it]
86%|████████▋ | 37989/43939 [61:18:02<9:27:13, 5.72s/it]
86%|████████▋ | 37990/43939 [61:18:08<9:26:52, 5.72s/it]
86%|████████▋ | 37991/43939 [61:18:14<9:28:36, 5.74s/it]
86%|████████▋ | 37992/43939 [61:18:20<9:27:34, 5.73s/it]
86%|████████▋ | 37993/43939 [61:18:25<9:34:10, 5.79s/it]
86%|████████▋ | 37994/43939 [61:18:31<9:31:48, 5.77s/it]
86%|████████▋ | 37995/43939 [61:18:37<9:29:18, 5.75s/it]
86%|████████▋ | 37996/43939 [61:18:43<9:27:29, 5.73s/it]
86%|████████▋ | 37997/43939 [61:18:48<9:26:31, 5.72s/it]
86%|████████▋ | 37998/43939 [61:18:54<9:25:39, 5.71s/it]
86%|████████▋ | 37999/43939 [61:19:00<9:26:02, 5.72s/it]
86%|████████▋ | 38000/43939 [61:19:05<9:25:23, 5.71s/it]
{'loss': 0.6482, 'grad_norm': 1.7165180444717407, 'learning_rate': 2.732144805796236e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.65, 'epoch': 3.46}
86%|████████▋ | 38000/43939 [61:19:05<9:25:23, 5.71s/it]
86%|████████▋ | 38001/43939 [61:19:11<9:25:49, 5.72s/it]
86%|████████▋ | 38002/43939 [61:19:17<9:24:55, 5.71s/it]
86%|████████▋ | 38003/43939 [61:19:23<9:27:30, 5.74s/it]
86%|████████▋ | 38004/43939 [61:19:28<9:27:25, 5.74s/it]
86%|████████▋ | 38005/43939 [61:19:34<9:26:06, 5.72s/it]
86%|████████▋ | 38006/43939 [61:19:40<9:24:23, 5.71s/it]
86%|████████▋ | 38007/43939 [61:19:45<9:24:08, 5.71s/it]
87%|████████▋ | 38008/43939 [61:19:51<9:24:33, 5.71s/it]
87%|████████▋ | 38009/43939 [61:19:57<9:24:19, 5.71s/it]
87%|████████▋ | 38010/43939 [61:20:03<9:25:02, 5.72s/it]
87%|████████▋ | 38011/43939 [61:20:08<9:24:16, 5.71s/it]
87%|████████▋ | 38012/43939 [61:20:14<9:23:52, 5.71s/it]
87%|████████▋ | 38013/43939 [61:20:20<9:23:21, 5.70s/it]
87%|████████▋ | 38014/43939 [61:20:26<9:30:37, 5.78s/it]
87%|████████▋ | 38015/43939 [61:20:31<9:29:18, 5.77s/it]
87%|████████▋ | 38016/43939 [61:20:37<9:29:57, 5.77s/it]
87%|████████▋ | 38017/43939 [61:20:43<9:26:40, 5.74s/it]
87%|████████▋ | 38018/43939 [61:20:49<9:27:14, 5.75s/it]
87%|████████▋ | 38019/43939 [61:20:54<9:24:33, 5.72s/it]
87%|████████▋ | 38020/43939 [61:21:00<9:23:08, 5.71s/it]
87%|████████▋ | 38021/43939 [61:21:06<9:22:44, 5.71s/it]
87%|████████▋ | 38022/43939 [61:21:11<9:23:05, 5.71s/it]
87%|████████▋ | 38023/43939 [61:21:17<9:23:01, 5.71s/it]
87%|████████▋ | 38024/43939 [61:21:23<9:22:54, 5.71s/it]
87%|████████▋ | 38025/43939 [61:21:28<9:22:02, 5.70s/it]
87%|████████▋ | 38026/43939 [61:21:34<9:22:44, 5.71s/it]
87%|████████▋ | 38027/43939 [61:21:40<9:22:27, 5.71s/it]
87%|████████▋ | 38028/43939 [61:21:46<9:21:13, 5.70s/it]
87%|████████▋ | 38029/43939 [61:21:51<9:21:31, 5.70s/it]
87%|████████▋ | 38030/43939 [61:21:57<9:22:19, 5.71s/it]
87%|████████▋ | 38031/43939 [61:22:03<9:21:52, 5.71s/it]
87%|████████▋ | 38032/43939 [61:22:08<9:21:56, 5.71s/it]
87%|████████▋ | 38033/43939 [61:22:14<9:20:52, 5.70s/it]
87%|████████▋ | 38034/43939 [61:22:20<9:23:12, 5.72s/it]
87%|████████▋ | 38035/43939 [61:22:26<9:22:40, 5.72s/it]
87%|████████▋ | 38036/43939 [61:22:32<9:29:09, 5.79s/it]
87%|████████▋ | 38037/43939 [61:22:37<9:25:19, 5.75s/it]
87%|████████▋ | 38038/43939 [61:22:43<9:23:59, 5.73s/it]
87%|████████▋ | 38039/43939 [61:22:49<9:22:54, 5.72s/it]
87%|████████▋ | 38040/43939 [61:22:54<9:21:43, 5.71s/it]
87%|████████▋ | 38041/43939 [61:23:00<9:20:33, 5.70s/it]
87%|████████▋ | 38042/43939 [61:23:06<9:20:50, 5.71s/it]
87%|████████▋ | 38043/43939 [61:23:11<9:20:25, 5.70s/it]
87%|████████▋ | 38044/43939 [61:23:17<9:21:31, 5.72s/it]
87%|████████▋ | 38045/43939 [61:23:23<9:19:24, 5.69s/it]
87%|████████▋ | 38046/43939 [61:23:28<9:19:18, 5.69s/it]
87%|████████▋ | 38047/43939 [61:23:34<9:19:17, 5.70s/it]
87%|████████▋ | 38048/43939 [61:23:40<9:18:46, 5.69s/it]
87%|████████▋ | 38049/43939 [61:23:46<9:18:32, 5.69s/it]
87%|████████▋ | 38050/43939 [61:23:51<9:19:51, 5.70s/it]
{'loss': 0.6565, 'grad_norm': 1.383346438407898, 'learning_rate': 2.687181568989919e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.46, 'epoch': 3.46}
87%|████████▋ | 38050/43939 [61:23:51<9:19:51, 5.70s/it]
87%|████████▋ | 38051/43939 [61:23:57<9:20:36, 5.71s/it]
87%|████████▋ | 38052/43939 [61:24:03<9:19:34, 5.70s/it]
87%|████████▋ | 38053/43939 [61:24:08<9:18:56, 5.70s/it]
87%|████████▋ | 38054/43939 [61:24:14<9:18:11, 5.69s/it]
87%|████████▋ | 38055/43939 [61:24:20<9:19:04, 5.70s/it]
87%|████████▋ | 38056/43939 [61:24:25<9:20:10, 5.71s/it]
87%|████████▋ | 38057/43939 [61:24:31<9:27:20, 5.79s/it]
87%|████████▋ | 38058/43939 [61:24:37<9:25:23, 5.77s/it]
87%|████████▋ | 38059/43939 [61:24:43<9:22:36, 5.74s/it]
87%|████████▋ | 38060/43939 [61:24:49<9:21:48, 5.73s/it]
87%|████████▋ | 38061/43939 [61:24:54<9:20:27, 5.72s/it]
87%|████████▋ | 38062/43939 [61:25:00<9:20:14, 5.72s/it]
87%|████████▋ | 38063/43939 [61:25:06<9:21:19, 5.73s/it]
87%|████████▋ | 38064/43939 [61:25:11<9:20:29, 5.72s/it]
87%|████████▋ | 38065/43939 [61:25:17<9:20:03, 5.72s/it]
87%|████████▋ | 38066/43939 [61:25:23<9:18:43, 5.71s/it]
87%|████████▋ | 38067/43939 [61:25:29<9:18:49, 5.71s/it]
87%|████████▋ | 38068/43939 [61:25:34<9:18:41, 5.71s/it]
87%|████████▋ | 38069/43939 [61:25:40<9:19:04, 5.71s/it]
87%|████████▋ | 38070/43939 [61:25:46<9:18:29, 5.71s/it]
87%|████████▋ | 38071/43939 [61:25:51<9:18:55, 5.72s/it]
87%|████████▋ | 38072/43939 [61:25:57<9:17:44, 5.70s/it]
87%|████████▋ | 38073/43939 [61:26:03<9:18:36, 5.71s/it]
87%|████████▋ | 38074/43939 [61:26:08<9:17:25, 5.70s/it]
87%|████████▋ | 38075/43939 [61:26:14<9:17:09, 5.70s/it]
87%|████████▋ | 38076/43939 [61:26:20<9:16:48, 5.70s/it]
87%|████████▋ | 38077/43939 [61:26:26<9:17:07, 5.70s/it]
87%|████████▋ | 38078/43939 [61:26:32<9:25:25, 5.79s/it]
87%|████████▋ | 38079/43939 [61:26:37<9:24:26, 5.78s/it]
87%|████████▋ | 38080/43939 [61:26:43<9:22:02, 5.76s/it]
87%|████████▋ | 38081/43939 [61:26:49<9:20:37, 5.74s/it]
87%|████████▋ | 38082/43939 [61:26:55<9:20:54, 5.75s/it]
87%|████████▋ | 38083/43939 [61:27:00<9:18:40, 5.72s/it]
87%|████████▋ | 38084/43939 [61:27:06<9:18:19, 5.72s/it]
87%|████████▋ | 38085/43939 [61:27:12<9:17:23, 5.71s/it]
87%|████████▋ | 38086/43939 [61:27:17<9:16:02, 5.70s/it]
87%|████████▋ | 38087/43939 [61:27:23<9:17:08, 5.71s/it]
87%|████████▋ | 38088/43939 [61:27:29<9:17:21, 5.72s/it]
87%|████████▋ | 38089/43939 [61:27:34<9:18:27, 5.73s/it]
87%|████████▋ | 38090/43939 [61:27:40<9:17:22, 5.72s/it]
87%|████████▋ | 38091/43939 [61:27:46<9:17:39, 5.72s/it]
87%|████████▋ | 38092/43939 [61:27:52<9:16:51, 5.71s/it]
87%|████████▋ | 38093/43939 [61:27:57<9:15:46, 5.70s/it]
87%|████████▋ | 38094/43939 [61:28:03<9:15:22, 5.70s/it]
87%|████████▋ | 38095/43939 [61:28:09<9:14:39, 5.69s/it]
87%|████████▋ | 38096/43939 [61:28:14<9:14:48, 5.70s/it]
87%|████████▋ | 38097/43939 [61:28:20<9:13:41, 5.69s/it]
87%|████████▋ | 38098/43939 [61:28:26<9:13:50, 5.69s/it]
87%|████████▋ | 38099/43939 [61:28:31<9:13:28, 5.69s/it]
87%|████████▋ | 38100/43939 [61:28:37<9:21:31, 5.77s/it]
{'loss': 0.6534, 'grad_norm': 1.528689980506897, 'learning_rate': 2.6425703693287533e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 452.96, 'epoch': 3.47}
87%|████████▋ | 38100/43939 [61:28:37<9:21:31, 5.77s/it]
87%|████████▋ | 38101/43939 [61:28:43<9:18:58, 5.74s/it]
87%|████████▋ | 38102/43939 [61:28:49<9:18:57, 5.75s/it]
87%|████████▋ | 38103/43939 [61:28:55<9:18:47, 5.74s/it]
87%|████████▋ | 38104/43939 [61:29:00<9:17:07, 5.73s/it]
87%|████████▋ | 38105/43939 [61:29:06<9:16:48, 5.73s/it]
87%|████████▋ | 38106/43939 [61:29:12<9:15:53, 5.72s/it]
87%|████████▋ | 38107/43939 [61:29:17<9:16:12, 5.72s/it]
87%|████████▋ | 38108/43939 [61:29:23<9:16:12, 5.72s/it]
87%|████████▋ | 38109/43939 [61:29:29<9:16:12, 5.72s/it]
87%|████████▋ | 38110/43939 [61:29:35<9:15:30, 5.72s/it]
87%|████████▋ | 38111/43939 [61:29:40<9:14:42, 5.71s/it]
87%|████████▋ | 38112/43939 [61:29:46<9:14:36, 5.71s/it]
87%|████████▋ | 38113/43939 [61:29:52<9:16:08, 5.73s/it]
87%|████████▋ | 38114/43939 [61:29:57<9:17:23, 5.74s/it]
87%|████████▋ | 38115/43939 [61:30:03<9:17:01, 5.74s/it]
87%|████████▋ | 38116/43939 [61:30:09<9:15:41, 5.73s/it]
87%|████████▋ | 38117/43939 [61:30:15<9:15:25, 5.72s/it]
87%|████████▋ | 38118/43939 [61:30:20<9:14:39, 5.72s/it]
87%|████████▋ | 38119/43939 [61:30:26<9:13:19, 5.70s/it]
87%|████████▋ | 38120/43939 [61:30:32<9:12:56, 5.70s/it]
87%|████████▋ | 38121/43939 [61:30:38<9:19:53, 5.77s/it]
87%|████████▋ | 38122/43939 [61:30:43<9:19:03, 5.77s/it]
87%|████████▋ | 38123/43939 [61:30:49<9:16:57, 5.75s/it]
87%|████████▋ | 38124/43939 [61:30:55<9:17:23, 5.75s/it]
87%|████████▋ | 38125/43939 [61:31:01<9:17:09, 5.75s/it]
87%|████████▋ | 38126/43939 [61:31:06<9:15:21, 5.73s/it]
87%|████████▋ | 38127/43939 [61:31:12<9:13:52, 5.72s/it]
87%|████████▋ | 38128/43939 [61:31:18<9:12:04, 5.70s/it]
87%|████████▋ | 38129/43939 [61:31:23<9:11:59, 5.70s/it]
87%|████████▋ | 38130/43939 [61:31:29<9:12:34, 5.71s/it]
87%|████████▋ | 38131/43939 [61:31:35<9:12:38, 5.71s/it]
87%|████████▋ | 38132/43939 [61:31:40<9:12:33, 5.71s/it]
87%|████████▋ | 38133/43939 [61:31:46<9:11:41, 5.70s/it]
87%|████████▋ | 38134/43939 [61:31:52<9:11:38, 5.70s/it]
87%|████████▋ | 38135/43939 [61:31:58<9:11:36, 5.70s/it]
87%|████████▋ | 38136/43939 [61:32:03<9:12:52, 5.72s/it]
87%|████████▋ | 38137/43939 [61:32:09<9:10:58, 5.70s/it]
87%|████████▋ | 38138/43939 [61:32:15<9:12:16, 5.71s/it]
87%|████████▋ | 38139/43939 [61:32:20<9:12:31, 5.72s/it]
87%|████████▋ | 38140/43939 [61:32:26<9:11:06, 5.70s/it]
87%|████████▋ | 38141/43939 [61:32:32<9:11:14, 5.70s/it]
87%|████████▋ | 38142/43939 [61:32:38<9:17:09, 5.77s/it]
87%|████████▋ | 38143/43939 [61:32:43<9:14:32, 5.74s/it]
87%|████████▋ | 38144/43939 [61:32:49<9:13:19, 5.73s/it]
87%|████████▋ | 38145/43939 [61:32:55<9:14:20, 5.74s/it]
87%|████████▋ | 38146/43939 [61:33:01<9:14:52, 5.75s/it]
87%|████████▋ | 38147/43939 [61:33:06<9:14:20, 5.74s/it]
87%|████████▋ | 38148/43939 [61:33:12<9:12:18, 5.72s/it]
87%|████████▋ | 38149/43939 [61:33:18<9:13:56, 5.74s/it]
87%|████████▋ | 38150/43939 [61:33:24<9:13:12, 5.73s/it]
{'loss': 0.6784, 'grad_norm': 1.5609986782073975, 'learning_rate': 2.5983119106592512e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 530.79, 'epoch': 3.47}
87%|████████▋ | 38150/43939 [61:33:24<9:13:12, 5.73s/it]
87%|████████▋ | 38151/43939 [61:33:29<9:12:47, 5.73s/it]
87%|████████▋ | 38152/43939 [61:33:35<9:11:27, 5.72s/it]
87%|████████▋ | 38153/43939 [61:33:41<9:10:34, 5.71s/it]
87%|████████▋ | 38154/43939 [61:33:46<9:10:36, 5.71s/it]
87%|████████▋ | 38155/43939 [61:33:52<9:10:10, 5.71s/it]
87%|████████▋ | 38156/43939 [61:33:58<9:09:57, 5.71s/it]
87%|████████▋ | 38157/43939 [61:34:03<9:10:00, 5.71s/it]
87%|████████▋ | 38158/43939 [61:34:09<9:10:23, 5.71s/it]
87%|████████▋ | 38159/43939 [61:34:15<9:09:34, 5.70s/it]
87%|████████▋ | 38160/43939 [61:34:21<9:09:14, 5.70s/it]
87%|████████▋ | 38161/43939 [61:34:26<9:09:39, 5.71s/it]
87%|████████▋ | 38162/43939 [61:34:32<9:09:21, 5.71s/it]
87%|████████▋ | 38163/43939 [61:34:38<9:10:24, 5.72s/it]
87%|████████▋ | 38164/43939 [61:34:44<9:16:46, 5.78s/it]
87%|████████▋ | 38165/43939 [61:34:49<9:14:44, 5.76s/it]
87%|████████▋ | 38166/43939 [61:34:55<9:12:00, 5.74s/it]
87%|████████▋ | 38167/43939 [61:35:01<9:10:28, 5.72s/it]
87%|████████▋ | 38168/43939 [61:35:06<9:09:30, 5.71s/it]
87%|████████▋ | 38169/43939 [61:35:12<9:09:43, 5.72s/it]
87%|████████▋ | 38170/43939 [61:35:18<9:09:33, 5.72s/it]
87%|████████▋ | 38171/43939 [61:35:24<9:08:30, 5.71s/it]
87%|████████▋ | 38172/43939 [61:35:29<9:08:26, 5.71s/it]
87%|████████▋ | 38173/43939 [61:35:35<9:07:29, 5.70s/it]
87%|████████▋ | 38174/43939 [61:35:41<9:08:05, 5.70s/it]
87%|████████▋ | 38175/43939 [61:35:46<9:07:56, 5.70s/it]
87%|████████▋ | 38176/43939 [61:35:52<9:08:43, 5.71s/it]
87%|████████▋ | 38177/43939 [61:35:58<9:09:19, 5.72s/it]
87%|████████▋ | 38178/43939 [61:36:04<9:08:45, 5.72s/it]
87%|████████▋ | 38179/43939 [61:36:09<9:08:08, 5.71s/it]
87%|████████▋ | 38180/43939 [61:36:15<9:08:15, 5.71s/it]
87%|████████▋ | 38181/43939 [61:36:21<9:08:01, 5.71s/it]
87%|████████▋ | 38182/43939 [61:36:26<9:07:01, 5.70s/it]
87%|████████▋ | 38183/43939 [61:36:32<9:07:19, 5.71s/it]
87%|████████▋ | 38184/43939 [61:36:38<9:07:22, 5.71s/it]
87%|████████▋ | 38185/43939 [61:36:43<9:06:22, 5.70s/it]
87%|████████▋ | 38186/43939 [61:36:49<9:14:02, 5.78s/it]
87%|████████▋ | 38187/43939 [61:36:55<9:10:30, 5.74s/it]
87%|████████▋ | 38188/43939 [61:37:01<9:10:27, 5.74s/it]
87%|████████▋ | 38189/43939 [61:37:07<9:10:58, 5.75s/it]
87%|████████▋ | 38190/43939 [61:37:12<9:10:21, 5.74s/it]
87%|████████▋ | 38191/43939 [61:37:18<9:08:48, 5.73s/it]
87%|████████▋ | 38192/43939 [61:37:24<9:07:14, 5.71s/it]
87%|████████▋ | 38193/43939 [61:37:29<9:06:26, 5.71s/it]
87%|████████▋ | 38194/43939 [61:37:35<9:05:53, 5.70s/it]
87%|████████▋ | 38195/43939 [61:37:41<9:07:33, 5.72s/it]
87%|████████▋ | 38196/43939 [61:37:47<9:06:33, 5.71s/it]
87%|████████▋ | 38197/43939 [61:37:52<9:05:20, 5.70s/it]
87%|████████▋ | 38198/43939 [61:37:58<9:06:26, 5.71s/it]
87%|████████▋ | 38199/43939 [61:38:04<9:06:01, 5.71s/it]
87%|████████▋ | 38200/43939 [61:38:09<9:05:17, 5.70s/it]
{'loss': 0.6474, 'grad_norm': 1.7582324743270874, 'learning_rate': 2.5544068912625977e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.09, 'epoch': 3.48}
87%|████████▋ | 38200/43939 [61:38:09<9:05:17, 5.70s/it]
87%|████████▋ | 38201/43939 [61:38:15<9:06:51, 5.72s/it]
87%|████████▋ | 38202/43939 [61:38:21<9:05:17, 5.70s/it]
87%|████████▋ | 38203/43939 [61:38:26<9:05:03, 5.70s/it]
87%|████████▋ | 38204/43939 [61:38:32<9:05:03, 5.70s/it]
87%|████████▋ | 38205/43939 [61:38:38<9:05:50, 5.71s/it]
87%|████████▋ | 38206/43939 [61:38:44<9:05:15, 5.71s/it]
87%|████████▋ | 38207/43939 [61:38:49<9:05:39, 5.71s/it]
87%|████████▋ | 38208/43939 [61:38:55<9:11:04, 5.77s/it]
87%|████████▋ | 38209/43939 [61:39:01<9:09:23, 5.75s/it]
87%|████████▋ | 38210/43939 [61:39:07<9:07:23, 5.73s/it]
87%|████████▋ | 38211/43939 [61:39:12<9:06:23, 5.72s/it]
87%|████████▋ | 38212/43939 [61:39:18<9:05:09, 5.71s/it]
87%|████████▋ | 38213/43939 [61:39:24<9:04:43, 5.71s/it]
87%|████████▋ | 38214/43939 [61:39:29<9:04:38, 5.71s/it]
87%|████████▋ | 38215/43939 [61:39:35<9:04:43, 5.71s/it]
87%|████████▋ | 38216/43939 [61:39:41<9:06:20, 5.73s/it]
87%|████████▋ | 38217/43939 [61:39:47<9:05:15, 5.72s/it]
87%|████████▋ | 38218/43939 [61:39:52<9:04:59, 5.72s/it]
87%|████████▋ | 38219/43939 [61:39:58<9:04:46, 5.71s/it]
87%|████████▋ | 38220/43939 [61:40:04<9:05:29, 5.72s/it]
87%|████████▋ | 38221/43939 [61:40:09<9:03:34, 5.70s/it]
87%|████████▋ | 38222/43939 [61:40:15<9:03:46, 5.71s/it]
87%|████████▋ | 38223/43939 [61:40:21<9:04:00, 5.71s/it]
87%|████████▋ | 38224/43939 [61:40:27<9:04:50, 5.72s/it]
87%|████████▋ | 38225/43939 [61:40:32<9:03:52, 5.71s/it]
87%|████████▋ | 38226/43939 [61:40:38<9:03:26, 5.71s/it]
87%|████████▋ | 38227/43939 [61:40:44<9:04:27, 5.72s/it]
87%|████████▋ | 38228/43939 [61:40:49<9:03:13, 5.71s/it]
87%|████████▋ | 38229/43939 [61:40:55<9:09:04, 5.77s/it]
87%|████████▋ | 38230/43939 [61:41:01<9:06:58, 5.75s/it]
87%|████████▋ | 38231/43939 [61:41:07<9:05:38, 5.74s/it]
87%|████████▋ | 38232/43939 [61:41:12<9:05:35, 5.74s/it]
87%|████████▋ | 38233/43939 [61:41:18<9:06:16, 5.74s/it]
87%|████████▋ | 38234/43939 [61:41:24<9:05:11, 5.73s/it]
87%|████████▋ | 38235/43939 [61:41:30<9:04:14, 5.72s/it]
87%|████████▋ | 38236/43939 [61:41:35<9:02:40, 5.71s/it]
87%|████████▋ | 38237/43939 [61:41:41<9:04:35, 5.73s/it]
87%|████████▋ | 38238/43939 [61:41:47<9:03:02, 5.72s/it]
87%|████████▋ | 38239/43939 [61:41:52<9:02:17, 5.71s/it]
87%|████████▋ | 38240/43939 [61:41:58<9:02:59, 5.72s/it]
87%|████████▋ | 38241/43939 [61:42:04<9:02:33, 5.71s/it]
87%|████████▋ | 38242/43939 [61:42:10<9:02:14, 5.71s/it]
87%|████████▋ | 38243/43939 [61:42:15<9:01:32, 5.70s/it]
87%|████████▋ | 38244/43939 [61:42:21<9:01:06, 5.70s/it]
87%|████████▋ | 38245/43939 [61:42:27<9:00:42, 5.70s/it]
87%|████████▋ | 38246/43939 [61:42:32<9:00:46, 5.70s/it]
87%|████████▋ | 38247/43939 [61:42:38<9:00:16, 5.70s/it]
87%|████████▋ | 38248/43939 [61:42:44<9:00:22, 5.70s/it]
87%|████████▋ | 38249/43939 [61:42:49<8:59:47, 5.69s/it]
87%|████████▋ | 38250/43939 [61:42:55<9:01:51, 5.71s/it]
{'loss': 0.6492, 'grad_norm': 1.611234188079834, 'learning_rate': 2.510856003843659e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.4, 'epoch': 3.48}
87%|████████▋ | 38250/43939 [61:42:55<9:01:51, 5.71s/it]
87%|████████▋ | 38251/43939 [61:43:01<9:08:23, 5.78s/it]
87%|████████▋ | 38252/43939 [61:43:07<9:05:41, 5.76s/it]
87%|████████▋ | 38253/43939 [61:43:13<9:06:07, 5.76s/it]
87%|████████▋ | 38254/43939 [61:43:18<9:05:11, 5.75s/it]
87%|████████▋ | 38255/43939 [61:43:24<9:05:05, 5.75s/it]
87%|████████▋ | 38256/43939 [61:43:30<9:03:47, 5.74s/it]
87%|████████▋ | 38257/43939 [61:43:36<9:02:40, 5.73s/it]
87%|████████▋ | 38258/43939 [61:43:41<9:01:22, 5.72s/it]
87%|████████▋ | 38259/43939 [61:43:47<9:00:50, 5.71s/it]
87%|████████▋ | 38260/43939 [61:43:53<9:01:08, 5.72s/it]
87%|████████▋ | 38261/43939 [61:43:58<9:00:35, 5.71s/it]
87%|████████▋ | 38262/43939 [61:44:04<9:00:31, 5.71s/it]
87%|████████▋ | 38263/43939 [61:44:10<8:59:48, 5.71s/it]
87%|████████▋ | 38264/43939 [61:44:15<8:59:24, 5.70s/it]
87%|████████▋ | 38265/43939 [61:44:21<8:59:07, 5.70s/it]
87%|████████▋ | 38266/43939 [61:44:27<8:58:55, 5.70s/it]
87%|████████▋ | 38267/43939 [61:44:33<8:59:31, 5.71s/it]
87%|████████▋ | 38268/43939 [61:44:38<8:59:52, 5.71s/it]
87%|████████▋ | 38269/43939 [61:44:44<8:59:18, 5.71s/it]
87%|████████▋ | 38270/43939 [61:44:50<9:00:21, 5.72s/it]
87%|████████▋ | 38271/43939 [61:44:55<9:00:25, 5.72s/it]
87%|████████▋ | 38272/43939 [61:45:01<9:06:19, 5.78s/it]
87%|████████▋ | 38273/43939 [61:45:07<9:03:35, 5.76s/it]
87%|████████▋ | 38274/43939 [61:45:13<9:03:18, 5.75s/it]
87%|████████▋ | 38275/43939 [61:45:19<9:01:52, 5.74s/it]
87%|████████▋ | 38276/43939 [61:45:24<9:00:02, 5.72s/it]
87%|████████▋ | 38277/43939 [61:45:30<8:59:51, 5.72s/it]
87%|████████▋ | 38278/43939 [61:45:36<8:59:16, 5.72s/it]
87%|████████▋ | 38279/43939 [61:45:41<8:59:24, 5.72s/it]
87%|████████▋ | 38280/43939 [61:45:47<9:00:51, 5.73s/it]
87%|████████▋ | 38281/43939 [61:45:53<8:59:41, 5.72s/it]
87%|████████▋ | 38282/43939 [61:45:59<8:58:17, 5.71s/it]
87%|████████▋ | 38283/43939 [61:46:04<8:57:50, 5.71s/it]
87%|████████▋ | 38284/43939 [61:46:10<8:58:18, 5.71s/it]
87%|████████▋ | 38285/43939 [61:46:16<8:59:22, 5.72s/it]
87%|████████▋ | 38286/43939 [61:46:21<8:59:04, 5.72s/it]
87%|████████▋ | 38287/43939 [61:46:27<8:59:30, 5.73s/it]
87%|████████▋ | 38288/43939 [61:46:33<8:59:06, 5.72s/it]
87%|████████▋ | 38289/43939 [61:46:39<8:59:09, 5.73s/it]
87%|████████▋ | 38290/43939 [61:46:44<8:58:52, 5.72s/it]
87%|████████▋ | 38291/43939 [61:46:50<9:00:01, 5.74s/it]
87%|████████▋ | 38292/43939 [61:46:56<8:58:48, 5.72s/it]
87%|████████▋ | 38293/43939 [61:47:01<8:58:18, 5.72s/it]
87%|████████▋ | 38294/43939 [61:47:07<9:03:51, 5.78s/it]
87%|████████▋ | 38295/43939 [61:47:13<9:01:45, 5.76s/it]
87%|████████▋ | 38296/43939 [61:47:19<8:59:49, 5.74s/it]
87%|████████▋ | 38297/43939 [61:47:25<8:59:42, 5.74s/it]
87%|████████▋ | 38298/43939 [61:47:30<8:59:52, 5.74s/it]
87%|████████▋ | 38299/43939 [61:47:36<8:59:05, 5.73s/it]
87%|████████▋ | 38300/43939 [61:47:42<8:58:37, 5.73s/it]
{'loss': 0.6517, 'grad_norm': 1.5946812629699707, 'learning_rate': 2.4676599355200225e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 479.63, 'epoch': 3.49}
87%|████████▋ | 38300/43939 [61:47:42<8:58:37, 5.73s/it]
87%|████████▋ | 38301/43939 [61:47:47<8:58:03, 5.73s/it]
87%|████████▋ | 38302/43939 [61:47:53<8:58:30, 5.73s/it]
87%|████████▋ | 38303/43939 [61:47:59<8:57:51, 5.73s/it]
87%|████████▋ | 38304/43939 [61:48:05<8:55:59, 5.71s/it]
87%|████████▋ | 38305/43939 [61:48:10<8:55:49, 5.71s/it]
87%|████████▋ | 38306/43939 [61:48:16<8:56:02, 5.71s/it]
87%|████████▋ | 38307/43939 [61:48:22<8:55:41, 5.71s/it]
87%|████████▋ | 38308/43939 [61:48:27<8:55:48, 5.71s/it]
87%|████████▋ | 38309/43939 [61:48:33<8:55:08, 5.70s/it]
87%|████████▋ | 38310/43939 [61:48:39<8:54:52, 5.70s/it]
87%|████████▋ | 38311/43939 [61:48:45<8:55:41, 5.71s/it]
87%|████████▋ | 38312/43939 [61:48:50<8:54:51, 5.70s/it]
87%|████████▋ | 38313/43939 [61:48:56<8:53:49, 5.69s/it]
87%|████████▋ | 38314/43939 [61:49:02<8:54:15, 5.70s/it]
87%|████████▋ | 38315/43939 [61:49:08<8:59:40, 5.76s/it]
87%|████████▋ | 38316/43939 [61:49:13<8:58:02, 5.74s/it]
87%|████████▋ | 38317/43939 [61:49:19<8:58:02, 5.74s/it]
87%|████████▋ | 38318/43939 [61:49:25<8:57:24, 5.74s/it]
87%|████████▋ | 38319/43939 [61:49:30<8:57:05, 5.73s/it]
87%|████████▋ | 38320/43939 [61:49:36<8:56:24, 5.73s/it]
87%|████████▋ | 38321/43939 [61:49:42<8:55:06, 5.71s/it]
87%|████████▋ | 38322/43939 [61:49:47<8:54:25, 5.71s/it]
87%|████████▋ | 38323/43939 [61:49:53<8:53:05, 5.70s/it]
87%|████████▋ | 38324/43939 [61:49:59<8:53:29, 5.70s/it]
87%|████████▋ | 38325/43939 [61:50:05<8:53:26, 5.70s/it]
87%|████████▋ | 38326/43939 [61:50:10<8:55:05, 5.72s/it]
87%|████████▋ | 38327/43939 [61:50:16<8:54:23, 5.71s/it]
87%|████████▋ | 38328/43939 [61:50:22<8:53:01, 5.70s/it]
87%|████████▋ | 38329/43939 [61:50:27<8:52:45, 5.70s/it]
87%|████████▋ | 38330/43939 [61:50:33<8:53:50, 5.71s/it]
87%|████████▋ | 38331/43939 [61:50:39<8:56:23, 5.74s/it]
87%|████████▋ | 38332/43939 [61:50:45<8:55:08, 5.73s/it]
87%|████████▋ | 38333/43939 [61:50:50<8:54:13, 5.72s/it]
87%|████████▋ | 38334/43939 [61:50:56<8:53:48, 5.71s/it]
87%|████████▋ | 38335/43939 [61:51:02<8:53:16, 5.71s/it]
87%|████████▋ | 38336/43939 [61:51:08<9:00:38, 5.79s/it]
87%|████████▋ | 38337/43939 [61:51:13<8:58:34, 5.77s/it]
87%|████████▋ | 38338/43939 [61:51:19<8:56:42, 5.75s/it]
87%|████████▋ | 38339/43939 [61:51:25<8:55:07, 5.73s/it]
87%|████████▋ | 38340/43939 [61:51:31<8:54:01, 5.72s/it]
87%|████████▋ | 38341/43939 [61:51:36<8:54:06, 5.72s/it]
87%|████████▋ | 38342/43939 [61:51:42<8:54:10, 5.73s/it]
87%|████████▋ | 38343/43939 [61:51:48<8:53:05, 5.72s/it]
87%|████████▋ | 38344/43939 [61:51:53<8:53:00, 5.72s/it]
87%|████████▋ | 38345/43939 [61:51:59<8:52:40, 5.71s/it]
87%|████████▋ | 38346/43939 [61:52:05<8:55:22, 5.74s/it]
87%|████████▋ | 38347/43939 [61:52:11<8:54:27, 5.73s/it]
87%|████████▋ | 38348/43939 [61:52:16<8:52:38, 5.72s/it]
87%|████████▋ | 38349/43939 [61:52:22<8:52:06, 5.71s/it]
87%|████████▋ | 38350/43939 [61:52:28<8:52:36, 5.72s/it]
{'loss': 0.6606, 'grad_norm': 1.3531148433685303, 'learning_rate': 2.4248193678111735e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.16, 'epoch': 3.49}
87%|████████▋ | 38350/43939 [61:52:28<8:52:36, 5.72s/it]
87%|████████▋ | 38351/43939 [61:52:33<8:53:09, 5.72s/it]
87%|████████▋ | 38352/43939 [61:52:39<8:52:04, 5.71s/it]
87%|████████▋ | 38353/43939 [61:52:45<8:51:26, 5.71s/it]
87%|████████▋ | 38354/43939 [61:52:51<8:50:11, 5.70s/it]
87%|████████▋ | 38355/43939 [61:52:56<8:49:25, 5.69s/it]
87%|████████▋ | 38356/43939 [61:53:02<8:50:09, 5.70s/it]
87%|████████▋ | 38357/43939 [61:53:08<8:50:01, 5.70s/it]
87%|████████▋ | 38358/43939 [61:53:14<8:55:59, 5.76s/it]
87%|████████▋ | 38359/43939 [61:53:19<8:54:45, 5.75s/it]
87%|████████▋ | 38360/43939 [61:53:25<8:54:33, 5.75s/it]
87%|████████▋ | 38361/43939 [61:53:31<8:53:14, 5.74s/it]
87%|████████▋ | 38362/43939 [61:53:36<8:54:38, 5.75s/it]
87%|████████▋ | 38363/43939 [61:53:42<8:53:14, 5.74s/it]
87%|████████▋ | 38364/43939 [61:53:48<8:52:03, 5.73s/it]
87%|████████▋ | 38365/43939 [61:53:54<8:51:15, 5.72s/it]
87%|████████▋ | 38366/43939 [61:53:59<8:50:04, 5.71s/it]
87%|████████▋ | 38367/43939 [61:54:05<8:50:35, 5.71s/it]
87%|████████▋ | 38368/43939 [61:54:11<8:51:18, 5.72s/it]
87%|████████▋ | 38369/43939 [61:54:16<8:50:53, 5.72s/it]
87%|████████▋ | 38370/43939 [61:54:22<8:50:23, 5.71s/it]
87%|████████▋ | 38371/43939 [61:54:28<8:49:40, 5.71s/it]
87%|████████▋ | 38372/43939 [61:54:34<8:49:15, 5.70s/it]
87%|████████▋ | 38373/43939 [61:54:39<8:49:26, 5.71s/it]
87%|████████▋ | 38374/43939 [61:54:45<8:50:17, 5.72s/it]
87%|████████▋ | 38375/43939 [61:54:51<8:49:48, 5.71s/it]
87%|████████▋ | 38376/43939 [61:54:56<8:48:44, 5.70s/it]
87%|████████▋ | 38377/43939 [61:55:02<8:48:38, 5.70s/it]
87%|████████▋ | 38378/43939 [61:55:08<8:48:23, 5.70s/it]
87%|████████▋ | 38379/43939 [61:55:14<8:54:58, 5.77s/it]
87%|████████▋ | 38380/43939 [61:55:19<8:52:36, 5.75s/it]
87%|████████▋ | 38381/43939 [61:55:25<8:51:37, 5.74s/it]
87%|████████▋ | 38382/43939 [61:55:31<8:50:59, 5.73s/it]
87%|████████▋ | 38383/43939 [61:55:37<8:50:08, 5.73s/it]
87%|████████▋ | 38384/43939 [61:55:42<8:48:52, 5.71s/it]
87%|████████▋ | 38385/43939 [61:55:48<8:48:33, 5.71s/it]
87%|████████▋ | 38386/43939 [61:55:54<8:48:19, 5.71s/it]
87%|████████▋ | 38387/43939 [61:55:59<8:47:04, 5.70s/it]
87%|████████▋ | 38388/43939 [61:56:05<8:47:39, 5.70s/it]
87%|████████▋ | 38389/43939 [61:56:11<8:48:37, 5.71s/it]
87%|████████▋ | 38390/43939 [61:56:16<8:47:57, 5.71s/it]
87%|████████▋ | 38391/43939 [61:56:22<8:47:52, 5.71s/it]
87%|████████▋ | 38392/43939 [61:56:28<8:48:18, 5.71s/it]
87%|████████▋ | 38393/43939 [61:56:34<8:48:41, 5.72s/it]
87%|████████▋ | 38394/43939 [61:56:39<8:49:57, 5.73s/it]
87%|████████▋ | 38395/43939 [61:56:45<8:49:09, 5.73s/it]
87%|████████▋ | 38396/43939 [61:56:51<8:48:59, 5.73s/it]
87%|████████▋ | 38397/43939 [61:56:57<8:48:54, 5.73s/it]
87%|████████▋ | 38398/43939 [61:57:02<8:49:23, 5.73s/it]
87%|████████▋ | 38399/43939 [61:57:08<8:48:11, 5.72s/it]
87%|████████▋ | 38400/43939 [61:57:14<8:54:51, 5.79s/it]
{'loss': 0.6763, 'grad_norm': 1.433319330215454, 'learning_rate': 2.3823349766277453e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.35, 'epoch': 3.5}
87%|████████▋ | 38400/43939 [61:57:14<8:54:51, 5.79s/it]
87%|████████▋ | 38401/43939 [61:57:20<8:52:49, 5.77s/it]
87%|████████▋ | 38402/43939 [61:57:25<8:52:55, 5.77s/it]
87%|████████▋ | 38403/43939 [61:57:31<8:50:05, 5.75s/it]
87%|████████▋ | 38404/43939 [61:57:37<8:48:30, 5.73s/it]
87%|████████▋ | 38405/43939 [61:57:43<8:48:29, 5.73s/it]
87%|████████▋ | 38406/43939 [61:57:48<8:47:56, 5.72s/it]
87%|████████▋ | 38407/43939 [61:57:54<8:47:10, 5.72s/it]
87%|████████▋ | 38408/43939 [61:58:00<8:47:42, 5.72s/it]
87%|████████▋ | 38409/43939 [61:58:05<8:47:10, 5.72s/it]
87%|████████▋ | 38410/43939 [61:58:11<8:47:36, 5.73s/it]
87%|████████▋ | 38411/43939 [61:58:17<8:47:06, 5.72s/it]
87%|████████▋ | 38412/43939 [61:58:23<8:46:27, 5.72s/it]
87%|████████▋ | 38413/43939 [61:58:28<8:45:16, 5.70s/it]
87%|████████▋ | 38414/43939 [61:58:34<8:45:00, 5.70s/it]
87%|████████▋ | 38415/43939 [61:58:40<8:44:58, 5.70s/it]
87%|████████▋ | 38416/43939 [61:58:45<8:44:02, 5.69s/it]
87%|████████▋ | 38417/43939 [61:58:51<8:43:55, 5.69s/it]
87%|████████▋ | 38418/43939 [61:58:57<8:44:44, 5.70s/it]
87%|████████▋ | 38419/43939 [61:59:03<8:45:45, 5.71s/it]
87%|████████▋ | 38420/43939 [61:59:08<8:45:07, 5.71s/it]
87%|████████▋ | 38421/43939 [61:59:14<8:44:54, 5.71s/it]
87%|████████▋ | 38422/43939 [61:59:20<8:49:49, 5.76s/it]
87%|████████▋ | 38423/43939 [61:59:26<8:48:17, 5.75s/it]
87%|████████▋ | 38424/43939 [61:59:31<8:48:14, 5.75s/it]
87%|████████▋ | 38425/43939 [61:59:37<8:46:23, 5.73s/it]
87%|████████▋ | 38426/43939 [61:59:43<8:45:42, 5.72s/it]
87%|████████▋ | 38427/43939 [61:59:48<8:45:53, 5.72s/it]
87%|████████▋ | 38428/43939 [61:59:54<8:44:49, 5.71s/it]
87%|████████▋ | 38429/43939 [62:00:00<8:44:17, 5.71s/it]
87%|████████▋ | 38430/43939 [62:00:05<8:44:11, 5.71s/it]
87%|████████▋ | 38431/43939 [62:00:11<8:43:41, 5.70s/it]
87%|████████▋ | 38432/43939 [62:00:17<8:44:15, 5.71s/it]
87%|████████▋ | 38433/43939 [62:00:23<8:43:29, 5.70s/it]
87%|████████▋ | 38434/43939 [62:00:28<8:43:44, 5.71s/it]
87%|████████▋ | 38435/43939 [62:00:34<8:43:08, 5.70s/it]
87%|████████▋ | 38436/43939 [62:00:40<8:43:52, 5.71s/it]
87%|████████▋ | 38437/43939 [62:00:45<8:43:46, 5.71s/it]
87%|████████▋ | 38438/43939 [62:00:51<8:43:35, 5.71s/it]
87%|████████▋ | 38439/43939 [62:00:57<8:42:31, 5.70s/it]
87%|████████▋ | 38440/43939 [62:01:03<8:42:33, 5.70s/it]
87%|████████▋ | 38441/43939 [62:01:08<8:42:15, 5.70s/it]
87%|████████▋ | 38442/43939 [62:01:14<8:41:33, 5.69s/it]
87%|████████▋ | 38443/43939 [62:01:20<8:42:01, 5.70s/it]
87%|████████▋ | 38444/43939 [62:01:26<8:47:30, 5.76s/it]
87%|████████▋ | 38445/43939 [62:01:31<8:45:15, 5.74s/it]
87%|████████▋ | 38446/43939 [62:01:37<8:45:16, 5.74s/it]
88%|████████▊ | 38447/43939 [62:01:43<8:43:11, 5.72s/it]
88%|████████▊ | 38448/43939 [62:01:48<8:42:40, 5.71s/it]
88%|████████▊ | 38449/43939 [62:01:54<8:42:54, 5.71s/it]
88%|████████▊ | 38450/43939 [62:02:00<8:43:30, 5.72s/it]
{'loss': 0.6548, 'grad_norm': 1.3814903497695923, 'learning_rate': 2.3402074322608437e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.8, 'epoch': 3.5}
88%|████████▊ | 38450/43939 [62:02:00<8:43:30, 5.72s/it]
88%|████████▊ | 38451/43939 [62:02:06<8:44:38, 5.74s/it][2025-12-01 08:06:19,355] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-38451
88%|████████▊ | 38452/43939 [62:02:13<9:45:31, 6.40s/it]
88%|████████▊ | 38453/43939 [62:02:19<9:26:02, 6.19s/it]
88%|████████▊ | 38454/43939 [62:02:25<9:11:50, 6.04s/it]
88%|████████▊ | 38455/43939 [62:02:31<9:02:29, 5.94s/it]
88%|████████▊ | 38456/43939 [62:02:36<8:55:03, 5.86s/it]
88%|████████▊ | 38457/43939 [62:02:42<8:50:30, 5.81s/it]
88%|████████▊ | 38458/43939 [62:02:48<8:47:28, 5.77s/it]
88%|████████▊ | 38459/43939 [62:02:53<8:44:29, 5.74s/it]
88%|████████▊ | 38460/43939 [62:02:59<8:44:17, 5.74s/it]
88%|████████▊ | 38461/43939 [62:03:05<8:45:29, 5.76s/it]
88%|████████▊ | 38462/43939 [62:03:11<8:43:19, 5.73s/it]
88%|████████▊ | 38463/43939 [62:03:16<8:42:20, 5.72s/it]
88%|████████▊ | 38464/43939 [62:03:22<8:47:36, 5.78s/it]
88%|████████▊ | 38465/43939 [62:03:28<8:47:06, 5.78s/it]
88%|████████▊ | 38466/43939 [62:03:34<8:45:55, 5.77s/it]
88%|████████▊ | 38467/43939 [62:03:39<8:43:43, 5.74s/it]
88%|████████▊ | 38468/43939 [62:03:45<8:43:04, 5.74s/it]
88%|████████▊ | 38469/43939 [62:03:51<8:42:13, 5.73s/it]
88%|████████▊ | 38470/43939 [62:03:56<8:41:18, 5.72s/it]
88%|████████▊ | 38471/43939 [62:04:02<8:40:27, 5.71s/it]
88%|████████▊ | 38472/43939 [62:04:08<8:40:30, 5.71s/it]
88%|████████▊ | 38473/43939 [62:04:14<8:41:40, 5.73s/it]
88%|████████▊ | 38474/43939 [62:04:19<8:41:19, 5.72s/it]
88%|████████▊ | 38475/43939 [62:04:25<8:40:05, 5.71s/it]
88%|████████▊ | 38476/43939 [62:04:31<8:39:38, 5.71s/it]
88%|████████▊ | 38477/43939 [62:04:36<8:38:47, 5.70s/it]
88%|████████▊ | 38478/43939 [62:04:42<8:39:43, 5.71s/it]
88%|████████▊ | 38479/43939 [62:04:48<8:39:09, 5.70s/it]
88%|████████▊ | 38480/43939 [62:04:53<8:38:05, 5.69s/it]
88%|████████▊ | 38481/43939 [62:04:59<8:39:09, 5.71s/it]
88%|████████▊ | 38482/43939 [62:05:05<8:38:18, 5.70s/it]
88%|████████▊ | 38483/43939 [62:05:11<8:36:55, 5.68s/it]
88%|████████▊ | 38484/43939 [62:05:16<8:36:25, 5.68s/it]
88%|████████▊ | 38485/43939 [62:05:22<8:37:20, 5.69s/it]
88%|████████▊ | 38486/43939 [62:05:28<8:43:15, 5.76s/it]
88%|████████▊ | 38487/43939 [62:05:34<8:41:53, 5.74s/it]
88%|████████▊ | 38488/43939 [62:05:39<8:41:26, 5.74s/it]
88%|████████▊ | 38489/43939 [62:05:45<8:39:46, 5.72s/it]
88%|████████▊ | 38490/43939 [62:05:51<8:38:46, 5.71s/it]
88%|████████▊ | 38491/43939 [62:05:56<8:38:28, 5.71s/it]
88%|████████▊ | 38492/43939 [62:06:02<8:38:33, 5.71s/it]
88%|████████▊ | 38493/43939 [62:06:08<8:37:48, 5.70s/it]
88%|████████▊ | 38494/43939 [62:06:13<8:37:40, 5.70s/it]
88%|████████▊ | 38495/43939 [62:06:19<8:37:50, 5.71s/it]
88%|████████▊ | 38496/43939 [62:06:25<8:37:56, 5.71s/it]
88%|████████▊ | 38497/43939 [62:06:31<8:37:32, 5.71s/it]
88%|████████▊ | 38498/43939 [62:06:36<8:37:40, 5.71s/it]
88%|████████▊ | 38499/43939 [62:06:42<8:37:32, 5.71s/it]
88%|████████▊ | 38500/43939 [62:06:48<8:37:45, 5.71s/it]
{'loss': 0.6371, 'grad_norm': 1.639966368675232, 'learning_rate': 2.298437399371481e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 455.42, 'epoch': 3.5}
88%|████████▊ | 38500/43939 [62:06:48<8:37:45, 5.71s/it]
88%|████████▊ | 38501/43939 [62:06:54<8:39:01, 5.73s/it]
88%|████████▊ | 38502/43939 [62:06:59<8:38:20, 5.72s/it]
88%|████████▊ | 38503/43939 [62:07:05<8:37:31, 5.71s/it]
88%|████████▊ | 38504/43939 [62:07:11<8:36:41, 5.70s/it]
88%|████████▊ | 38505/43939 [62:07:16<8:35:35, 5.69s/it]
88%|████████▊ | 38506/43939 [62:07:22<8:35:37, 5.69s/it]
88%|████████▊ | 38507/43939 [62:07:28<8:40:34, 5.75s/it]
88%|████████▊ | 38508/43939 [62:07:34<8:40:10, 5.75s/it]
88%|████████▊ | 38509/43939 [62:07:39<8:38:52, 5.73s/it]
88%|████████▊ | 38510/43939 [62:07:45<8:38:58, 5.74s/it]
88%|████████▊ | 38511/43939 [62:07:51<8:38:00, 5.73s/it]
88%|████████▊ | 38512/43939 [62:07:56<8:37:44, 5.72s/it]
88%|████████▊ | 38513/43939 [62:08:02<8:36:33, 5.71s/it]
88%|████████▊ | 38514/43939 [62:08:08<8:36:13, 5.71s/it]
88%|████████▊ | 38515/43939 [62:08:14<8:36:43, 5.72s/it]
88%|████████▊ | 38516/43939 [62:08:19<8:36:10, 5.71s/it]
88%|████████▊ | 38517/43939 [62:08:25<8:36:29, 5.72s/it]
88%|████████▊ | 38518/43939 [62:08:31<8:35:40, 5.71s/it]
88%|████████▊ | 38519/43939 [62:08:36<8:35:21, 5.71s/it]
88%|████████▊ | 38520/43939 [62:08:42<8:35:39, 5.71s/it]
88%|████████▊ | 38521/43939 [62:08:48<8:35:54, 5.71s/it]
88%|████████▊ | 38522/43939 [62:08:54<8:35:14, 5.71s/it]
88%|████████▊ | 38523/43939 [62:08:59<8:35:53, 5.72s/it]
88%|████████▊ | 38524/43939 [62:09:05<8:35:53, 5.72s/it]
88%|████████▊ | 38525/43939 [62:09:11<8:36:02, 5.72s/it]
88%|████████▊ | 38526/43939 [62:09:16<8:35:35, 5.72s/it]
88%|████████▊ | 38527/43939 [62:09:22<8:35:17, 5.71s/it]
88%|████████▊ | 38528/43939 [62:09:28<8:35:15, 5.71s/it]
88%|████████▊ | 38529/43939 [62:09:34<8:40:56, 5.78s/it]
88%|████████▊ | 38530/43939 [62:09:39<8:38:44, 5.75s/it]
88%|████████▊ | 38531/43939 [62:09:45<8:38:34, 5.75s/it]
88%|████████▊ | 38532/43939 [62:09:51<8:37:00, 5.74s/it]
88%|████████▊ | 38533/43939 [62:09:57<8:37:00, 5.74s/it]
88%|████████▊ | 38534/43939 [62:10:02<8:36:02, 5.73s/it]
88%|████████▊ | 38535/43939 [62:10:08<8:34:57, 5.72s/it]
88%|████████▊ | 38536/43939 [62:10:14<8:35:24, 5.72s/it]
88%|████████▊ | 38537/43939 [62:10:19<8:34:48, 5.72s/it]
88%|████████▊ | 38538/43939 [62:10:25<8:34:37, 5.72s/it]
88%|████████▊ | 38539/43939 [62:10:31<8:34:11, 5.71s/it]
88%|████████▊ | 38540/43939 [62:10:37<8:34:38, 5.72s/it]
88%|████████▊ | 38541/43939 [62:10:42<8:34:01, 5.71s/it]
88%|████████▊ | 38542/43939 [62:10:48<8:35:10, 5.73s/it]
88%|████████▊ | 38543/43939 [62:10:54<8:33:56, 5.71s/it]
88%|████████▊ | 38544/43939 [62:10:59<8:32:44, 5.70s/it]
88%|████████▊ | 38545/43939 [62:11:05<8:33:16, 5.71s/it]
88%|████████▊ | 38546/43939 [62:11:11<8:32:14, 5.70s/it]
88%|████████▊ | 38547/43939 [62:11:17<8:32:49, 5.71s/it]
88%|████████▊ | 38548/43939 [62:11:22<8:32:49, 5.71s/it]
88%|████████▊ | 38549/43939 [62:11:28<8:32:55, 5.71s/it]
88%|████████▊ | 38550/43939 [62:11:34<8:40:22, 5.79s/it]
{'loss': 0.6771, 'grad_norm': 1.2393807172775269, 'learning_rate': 2.2570255369800815e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.62, 'epoch': 3.51}
88%|████████▊ | 38550/43939 [62:11:34<8:40:22, 5.79s/it]
88%|████████▊ | 38551/43939 [62:11:40<8:38:31, 5.77s/it]
88%|████████▊ | 38552/43939 [62:11:45<8:37:46, 5.77s/it]
88%|████████▊ | 38553/43939 [62:11:51<8:36:07, 5.75s/it]
88%|████████▊ | 38554/43939 [62:11:57<8:34:09, 5.73s/it]
88%|████████▊ | 38555/43939 [62:12:03<8:33:33, 5.72s/it]
88%|████████▊ | 38556/43939 [62:12:08<8:32:09, 5.71s/it]
88%|████████▊ | 38557/43939 [62:12:14<8:30:47, 5.69s/it]
88%|████████▊ | 38558/43939 [62:12:20<8:31:40, 5.71s/it]
88%|████████▊ | 38559/43939 [62:12:25<8:30:48, 5.70s/it]
88%|████████▊ | 38560/43939 [62:12:31<8:31:08, 5.70s/it]
88%|████████▊ | 38561/43939 [62:12:37<8:30:39, 5.70s/it]
88%|████████▊ | 38562/43939 [62:12:42<8:30:37, 5.70s/it]
88%|████████▊ | 38563/43939 [62:12:48<8:30:23, 5.70s/it]
88%|████████▊ | 38564/43939 [62:12:54<8:30:48, 5.70s/it]
88%|████████▊ | 38565/43939 [62:12:59<8:29:50, 5.69s/it]
88%|████████▊ | 38566/43939 [62:13:05<8:29:53, 5.69s/it]
88%|████████▊ | 38567/43939 [62:13:11<8:30:41, 5.70s/it]
88%|████████▊ | 38568/43939 [62:13:17<8:30:16, 5.70s/it]
88%|████████▊ | 38569/43939 [62:13:22<8:29:32, 5.69s/it]
88%|████████▊ | 38570/43939 [62:13:28<8:30:04, 5.70s/it]
88%|████████▊ | 38571/43939 [62:13:34<8:30:19, 5.70s/it]
88%|████████▊ | 38572/43939 [62:13:40<8:35:00, 5.76s/it]
88%|████████▊ | 38573/43939 [62:13:45<8:33:29, 5.74s/it]
88%|████████▊ | 38574/43939 [62:13:51<8:32:06, 5.73s/it]
88%|████████▊ | 38575/43939 [62:13:57<8:31:42, 5.72s/it]
88%|████████▊ | 38576/43939 [62:14:02<8:30:56, 5.72s/it]
88%|████████▊ | 38577/43939 [62:14:08<8:32:16, 5.73s/it]
88%|████████▊ | 38578/43939 [62:14:14<8:31:13, 5.72s/it]
88%|████████▊ | 38579/43939 [62:14:20<8:31:19, 5.72s/it]
88%|████████▊ | 38580/43939 [62:14:25<8:30:24, 5.71s/it]
88%|████████▊ | 38581/43939 [62:14:31<8:29:22, 5.70s/it]
88%|████████▊ | 38582/43939 [62:14:37<8:30:10, 5.71s/it]
88%|████████▊ | 38583/43939 [62:14:42<8:30:06, 5.71s/it]
88%|████████▊ | 38584/43939 [62:14:48<8:29:24, 5.71s/it]
88%|████████▊ | 38585/43939 [62:14:54<8:28:45, 5.70s/it]
88%|████████▊ | 38586/43939 [62:15:00<8:29:03, 5.71s/it]
88%|████████▊ | 38587/43939 [62:15:05<8:29:05, 5.71s/it]
88%|████████▊ | 38588/43939 [62:15:11<8:27:44, 5.69s/it]
88%|████████▊ | 38589/43939 [62:15:17<8:28:15, 5.70s/it]
88%|████████▊ | 38590/43939 [62:15:22<8:28:08, 5.70s/it]
88%|████████▊ | 38591/43939 [62:15:28<8:28:18, 5.70s/it]
88%|████████▊ | 38592/43939 [62:15:34<8:28:29, 5.71s/it]
88%|████████▊ | 38593/43939 [62:15:40<8:34:32, 5.77s/it]
88%|████████▊ | 38594/43939 [62:15:45<8:32:37, 5.75s/it]
88%|████████▊ | 38595/43939 [62:15:51<8:31:51, 5.75s/it]
88%|████████▊ | 38596/43939 [62:15:57<8:30:33, 5.73s/it]
88%|████████▊ | 38597/43939 [62:16:02<8:29:15, 5.72s/it]
88%|████████▊ | 38598/43939 [62:16:08<8:28:54, 5.72s/it]
88%|████████▊ | 38599/43939 [62:16:14<8:27:34, 5.70s/it]
88%|████████▊ | 38600/43939 [62:16:20<8:27:17, 5.70s/it]
{'loss': 0.6452, 'grad_norm': 1.503077507019043, 'learning_rate': 2.2159724984560925e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 476.85, 'epoch': 3.51}
88%|████████▊ | 38600/43939 [62:16:20<8:27:17, 5.70s/it]
88%|████████▊ | 38601/43939 [62:16:25<8:27:22, 5.70s/it]
88%|████████▊ | 38602/43939 [62:16:31<8:26:59, 5.70s/it]
88%|████████▊ | 38603/43939 [62:16:37<8:28:24, 5.72s/it]
88%|████████▊ | 38604/43939 [62:16:42<8:27:12, 5.70s/it]
88%|████████▊ | 38605/43939 [62:16:48<8:26:46, 5.70s/it]
88%|████████▊ | 38606/43939 [62:16:54<8:27:29, 5.71s/it]
88%|████████▊ | 38607/43939 [62:17:00<8:27:35, 5.71s/it]
88%|████████▊ | 38608/43939 [62:17:05<8:26:42, 5.70s/it]
88%|████████▊ | 38609/43939 [62:17:11<8:25:33, 5.69s/it]
88%|████████▊ | 38610/43939 [62:17:17<8:25:29, 5.69s/it]
88%|████████▊ | 38611/43939 [62:17:22<8:25:24, 5.69s/it]
88%|████████▊ | 38612/43939 [62:17:28<8:25:57, 5.70s/it]
88%|████████▊ | 38613/43939 [62:17:34<8:25:08, 5.69s/it]
88%|████████▊ | 38614/43939 [62:17:39<8:25:31, 5.70s/it]
88%|████████▊ | 38615/43939 [62:17:45<8:33:10, 5.78s/it]
88%|████████▊ | 38616/43939 [62:17:51<8:31:37, 5.77s/it]
88%|████████▊ | 38617/43939 [62:17:57<8:30:11, 5.75s/it]
88%|████████▊ | 38618/43939 [62:18:03<8:29:04, 5.74s/it]
88%|████████▊ | 38619/43939 [62:18:08<8:29:16, 5.74s/it]
88%|████████▊ | 38620/43939 [62:18:14<8:28:07, 5.73s/it]
88%|████████▊ | 38621/43939 [62:18:20<8:27:39, 5.73s/it]
88%|████████▊ | 38622/43939 [62:18:25<8:27:06, 5.72s/it]
88%|████████▊ | 38623/43939 [62:18:31<8:26:23, 5.72s/it]
88%|████████▊ | 38624/43939 [62:18:37<8:25:38, 5.71s/it]
88%|████████▊ | 38625/43939 [62:18:42<8:24:27, 5.70s/it]
88%|████████▊ | 38626/43939 [62:18:48<8:24:18, 5.70s/it]
88%|████████▊ | 38627/43939 [62:18:54<8:24:31, 5.70s/it]
88%|████████▊ | 38628/43939 [62:19:00<8:24:39, 5.70s/it]
88%|████████▊ | 38629/43939 [62:19:05<8:24:53, 5.70s/it]
88%|████████▊ | 38630/43939 [62:19:11<8:24:55, 5.71s/it]
88%|████████▊ | 38631/43939 [62:19:17<8:25:32, 5.71s/it]
88%|████████▊ | 38632/43939 [62:19:22<8:25:00, 5.71s/it]
88%|████████▊ | 38633/43939 [62:19:28<8:25:42, 5.72s/it]
88%|████████▊ | 38634/43939 [62:19:34<8:25:17, 5.71s/it]
88%|████████▊ | 38635/43939 [62:19:40<8:25:28, 5.72s/it]
88%|████████▊ | 38636/43939 [62:19:45<8:24:46, 5.71s/it]
88%|████████▊ | 38637/43939 [62:19:51<8:30:40, 5.78s/it]
88%|████████▊ | 38638/43939 [62:19:57<8:27:50, 5.75s/it]
88%|████████▊ | 38639/43939 [62:20:03<8:27:13, 5.74s/it]
88%|████████▊ | 38640/43939 [62:20:08<8:27:34, 5.75s/it]
88%|████████▊ | 38641/43939 [62:20:14<8:25:55, 5.73s/it]
88%|████████▊ | 38642/43939 [62:20:20<8:25:22, 5.72s/it]
88%|████████▊ | 38643/43939 [62:20:25<8:23:46, 5.71s/it]
88%|████████▊ | 38644/43939 [62:20:31<8:22:55, 5.70s/it]
88%|████████▊ | 38645/43939 [62:20:37<8:22:28, 5.69s/it]
88%|████████▊ | 38646/43939 [62:20:42<8:22:11, 5.69s/it]
88%|████████▊ | 38647/43939 [62:20:48<8:20:36, 5.68s/it]
88%|████████▊ | 38648/43939 [62:20:54<8:22:11, 5.69s/it]
88%|████████▊ | 38649/43939 [62:21:00<8:23:36, 5.71s/it]
88%|████████▊ | 38650/43939 [62:21:05<8:23:45, 5.71s/it]
{'loss': 0.6362, 'grad_norm': 1.5892212390899658, 'learning_rate': 2.175278931507671e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.2, 'epoch': 3.52}
88%|████████▊ | 38650/43939 [62:21:05<8:23:45, 5.71s/it]
88%|████████▊ | 38651/43939 [62:21:11<8:23:30, 5.71s/it]
88%|████████▊ | 38652/43939 [62:21:17<8:23:53, 5.72s/it]
88%|████████▊ | 38653/43939 [62:21:23<8:23:43, 5.72s/it]
88%|████████▊ | 38654/43939 [62:21:28<8:21:47, 5.70s/it]
88%|████████▊ | 38655/43939 [62:21:34<8:21:29, 5.69s/it]
88%|████████▊ | 38656/43939 [62:21:40<8:22:21, 5.71s/it]
88%|████████▊ | 38657/43939 [62:21:45<8:22:37, 5.71s/it]
88%|████████▊ | 38658/43939 [62:21:51<8:31:06, 5.81s/it]
88%|████████▊ | 38659/43939 [62:21:57<8:28:46, 5.78s/it]
88%|████████▊ | 38660/43939 [62:22:03<8:26:47, 5.76s/it]
88%|████████▊ | 38661/43939 [62:22:08<8:24:51, 5.74s/it]
88%|████████▊ | 38662/43939 [62:22:14<8:24:10, 5.73s/it]
88%|████████▊ | 38663/43939 [62:22:20<8:23:07, 5.72s/it]
88%|████████▊ | 38664/43939 [62:22:26<8:23:10, 5.72s/it]
88%|████████▊ | 38665/43939 [62:22:31<8:22:13, 5.71s/it]
88%|████████▊ | 38666/43939 [62:22:37<8:22:04, 5.71s/it]
88%|████████▊ | 38667/43939 [62:22:43<8:21:34, 5.71s/it]
88%|████████▊ | 38668/43939 [62:22:48<8:21:48, 5.71s/it]
88%|████████▊ | 38669/43939 [62:22:54<8:20:46, 5.70s/it]
88%|████████▊ | 38670/43939 [62:23:00<8:21:37, 5.71s/it]
88%|████████▊ | 38671/43939 [62:23:06<8:21:59, 5.72s/it]
88%|████████▊ | 38672/43939 [62:23:11<8:20:50, 5.71s/it]
88%|████████▊ | 38673/43939 [62:23:17<8:20:06, 5.70s/it]
88%|████████▊ | 38674/43939 [62:23:23<8:19:49, 5.70s/it]
88%|████████▊ | 38675/43939 [62:23:28<8:19:22, 5.69s/it]
88%|████████▊ | 38676/43939 [62:23:34<8:19:59, 5.70s/it]
88%|████████▊ | 38677/43939 [62:23:40<8:19:39, 5.70s/it]
88%|████████▊ | 38678/43939 [62:23:45<8:20:19, 5.71s/it]
88%|████████▊ | 38679/43939 [62:23:51<8:25:08, 5.76s/it]
88%|████████▊ | 38680/43939 [62:23:57<8:23:31, 5.74s/it]
88%|████████▊ | 38681/43939 [62:24:03<8:22:31, 5.73s/it]
88%|████████▊ | 38682/43939 [62:24:08<8:21:35, 5.72s/it]
88%|████████▊ | 38683/43939 [62:24:14<8:20:10, 5.71s/it]
88%|████████▊ | 38684/43939 [62:24:20<8:21:35, 5.73s/it]
88%|████████▊ | 38685/43939 [62:24:26<8:21:25, 5.73s/it]
88%|████████▊ | 38686/43939 [62:24:31<8:20:04, 5.71s/it]
88%|████████▊ | 38687/43939 [62:24:37<8:21:21, 5.73s/it]
88%|████████▊ | 38688/43939 [62:24:43<8:20:26, 5.72s/it]
88%|████████▊ | 38689/43939 [62:24:48<8:20:48, 5.72s/it]
88%|████████▊ | 38690/43939 [62:24:54<8:20:03, 5.72s/it]
88%|████████▊ | 38691/43939 [62:25:00<8:19:23, 5.71s/it]
88%|████████▊ | 38692/43939 [62:25:06<8:19:31, 5.71s/it]
88%|████████▊ | 38693/43939 [62:25:11<8:19:10, 5.71s/it]
88%|████████▊ | 38694/43939 [62:25:17<8:18:46, 5.71s/it]
88%|████████▊ | 38695/43939 [62:25:23<8:17:55, 5.70s/it]
88%|████████▊ | 38696/43939 [62:25:28<8:19:08, 5.71s/it]
88%|████████▊ | 38697/43939 [62:25:34<8:19:26, 5.72s/it]
88%|████████▊ | 38698/43939 [62:25:40<8:20:57, 5.74s/it]
88%|████████▊ | 38699/43939 [62:25:46<8:21:34, 5.74s/it]
88%|████████▊ | 38700/43939 [62:25:52<8:26:13, 5.80s/it]
{'loss': 0.6657, 'grad_norm': 1.6182997226715088, 'learning_rate': 2.134945478171463e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.72, 'epoch': 3.52}
88%|████████▊ | 38700/43939 [62:25:52<8:26:13, 5.80s/it]
88%|████████▊ | 38701/43939 [62:25:57<8:23:54, 5.77s/it]
88%|████████▊ | 38702/43939 [62:26:03<8:22:58, 5.76s/it]
88%|████████▊ | 38703/43939 [62:26:09<8:22:11, 5.75s/it]
88%|████████▊ | 38704/43939 [62:26:15<8:22:10, 5.76s/it]
88%|████████▊ | 38705/43939 [62:26:20<8:22:00, 5.75s/it]
88%|████████▊ | 38706/43939 [62:26:26<8:20:39, 5.74s/it]
88%|████████▊ | 38707/43939 [62:26:32<8:20:25, 5.74s/it]
88%|████████▊ | 38708/43939 [62:26:37<8:19:21, 5.73s/it]
88%|████████▊ | 38709/43939 [62:26:43<8:18:47, 5.72s/it]
88%|████████▊ | 38710/43939 [62:26:49<8:18:38, 5.72s/it]
88%|████████▊ | 38711/43939 [62:26:55<8:19:03, 5.73s/it]
88%|████████▊ | 38712/43939 [62:27:00<8:17:38, 5.71s/it]
88%|████████▊ | 38713/43939 [62:27:06<8:17:21, 5.71s/it]
88%|████████▊ | 38714/43939 [62:27:12<8:17:08, 5.71s/it]
88%|████████▊ | 38715/43939 [62:27:17<8:16:46, 5.71s/it]
88%|████████▊ | 38716/43939 [62:27:23<8:17:34, 5.72s/it]
88%|████████▊ | 38717/43939 [62:27:29<8:17:50, 5.72s/it]
88%|████████▊ | 38718/43939 [62:27:35<8:17:40, 5.72s/it]
88%|████████▊ | 38719/43939 [62:27:40<8:17:52, 5.72s/it]
88%|████████▊ | 38720/43939 [62:27:46<8:18:10, 5.73s/it]
88%|████████▊ | 38721/43939 [62:27:52<8:17:15, 5.72s/it]
88%|████████▊ | 38722/43939 [62:27:58<8:22:21, 5.78s/it]
88%|████████▊ | 38723/43939 [62:28:03<8:23:37, 5.79s/it]
88%|████████▊ | 38724/43939 [62:28:09<8:20:45, 5.76s/it]
88%|████████▊ | 38725/43939 [62:28:15<8:18:15, 5.73s/it]
88%|████████▊ | 38726/43939 [62:28:21<8:18:37, 5.74s/it]
88%|████████▊ | 38727/43939 [62:28:26<8:17:40, 5.73s/it]
88%|████████▊ | 38728/43939 [62:28:32<8:17:10, 5.72s/it]
88%|████████▊ | 38729/43939 [62:28:38<8:16:04, 5.71s/it]
88%|████████▊ | 38730/43939 [62:28:43<8:15:14, 5.70s/it]
88%|████████▊ | 38731/43939 [62:28:49<8:15:39, 5.71s/it]
88%|████████▊ | 38732/43939 [62:28:55<8:14:33, 5.70s/it]
88%|████████▊ | 38733/43939 [62:29:01<8:15:08, 5.71s/it]
88%|████████▊ | 38734/43939 [62:29:06<8:15:07, 5.71s/it]
88%|████████▊ | 38735/43939 [62:29:12<8:15:27, 5.71s/it]
88%|████████▊ | 38736/43939 [62:29:18<8:14:04, 5.70s/it]
88%|████████▊ | 38737/43939 [62:29:23<8:15:28, 5.71s/it]
88%|████████▊ | 38738/43939 [62:29:29<8:15:06, 5.71s/it]
88%|████████▊ | 38739/43939 [62:29:35<8:16:05, 5.72s/it]
88%|████████▊ | 38740/43939 [62:29:41<8:15:22, 5.72s/it]
88%|████████▊ | 38741/43939 [62:29:46<8:17:42, 5.74s/it]
88%|████████▊ | 38742/43939 [62:29:52<8:15:59, 5.73s/it]
88%|████████▊ | 38743/43939 [62:29:58<8:20:51, 5.78s/it]
88%|████████▊ | 38744/43939 [62:30:04<8:19:34, 5.77s/it]
88%|████████▊ | 38745/43939 [62:30:09<8:17:02, 5.74s/it]
88%|████████▊ | 38746/43939 [62:30:15<8:15:44, 5.73s/it]
88%|████████▊ | 38747/43939 [62:30:21<8:15:17, 5.72s/it]
88%|████████▊ | 38748/43939 [62:30:26<8:14:06, 5.71s/it]
88%|████████▊ | 38749/43939 [62:30:32<8:13:53, 5.71s/it]
88%|████████▊ | 38750/43939 [62:30:38<8:13:42, 5.71s/it]
{'loss': 0.6685, 'grad_norm': 1.5786893367767334, 'learning_rate': 2.0949727748024756e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.61, 'epoch': 3.53}
88%|████████▊ | 38750/43939 [62:30:38<8:13:42, 5.71s/it]
88%|████████▊ | 38751/43939 [62:30:44<8:14:26, 5.72s/it]
88%|████████▊ | 38752/43939 [62:30:49<8:14:02, 5.71s/it]
88%|████████▊ | 38753/43939 [62:30:55<8:13:28, 5.71s/it]
88%|████████▊ | 38754/43939 [62:31:01<8:13:08, 5.71s/it]
88%|████████▊ | 38755/43939 [62:31:06<8:12:52, 5.70s/it]
88%|████████▊ | 38756/43939 [62:31:12<8:12:26, 5.70s/it]
88%|████████▊ | 38757/43939 [62:31:18<8:12:20, 5.70s/it]
88%|████████▊ | 38758/43939 [62:31:23<8:12:24, 5.70s/it]
88%|████████▊ | 38759/43939 [62:31:29<8:12:47, 5.71s/it]
88%|████████▊ | 38760/43939 [62:31:35<8:11:51, 5.70s/it]
88%|████████▊ | 38761/43939 [62:31:41<8:11:53, 5.70s/it]
88%|████████▊ | 38762/43939 [62:31:46<8:11:58, 5.70s/it]
88%|████████▊ | 38763/43939 [62:31:52<8:12:46, 5.71s/it]
88%|████████▊ | 38764/43939 [62:31:58<8:12:20, 5.71s/it]
88%|████████▊ | 38765/43939 [62:32:04<8:17:11, 5.77s/it]
88%|████████▊ | 38766/43939 [62:32:09<8:15:35, 5.75s/it]
88%|████████▊ | 38767/43939 [62:32:15<8:14:23, 5.74s/it]
88%|████████▊ | 38768/43939 [62:32:21<8:12:32, 5.71s/it]
88%|████████▊ | 38769/43939 [62:32:26<8:11:48, 5.71s/it]
88%|████████▊ | 38770/43939 [62:32:32<8:13:06, 5.72s/it]
88%|████████▊ | 38771/43939 [62:32:38<8:12:25, 5.72s/it]
88%|████████▊ | 38772/43939 [62:32:44<8:12:12, 5.72s/it]
88%|████████▊ | 38773/43939 [62:32:49<8:11:30, 5.71s/it]
88%|████████▊ | 38774/43939 [62:32:55<8:12:39, 5.72s/it]
88%|████████▊ | 38775/43939 [62:33:01<8:12:24, 5.72s/it]
88%|████████▊ | 38776/43939 [62:33:06<8:10:45, 5.70s/it]
88%|████████▊ | 38777/43939 [62:33:12<8:12:25, 5.72s/it]
88%|████████▊ | 38778/43939 [62:33:18<8:13:09, 5.73s/it]
88%|████████▊ | 38779/43939 [62:33:24<8:11:25, 5.71s/it]
88%|████████▊ | 38780/43939 [62:33:29<8:11:28, 5.72s/it]
88%|████████▊ | 38781/43939 [62:33:35<8:11:06, 5.71s/it]
88%|████████▊ | 38782/43939 [62:33:41<8:09:56, 5.70s/it]
88%|████████▊ | 38783/43939 [62:33:46<8:10:03, 5.70s/it]
88%|████████▊ | 38784/43939 [62:33:52<8:09:40, 5.70s/it]
88%|████████▊ | 38785/43939 [62:33:58<8:10:02, 5.70s/it]
88%|████████▊ | 38786/43939 [62:34:04<8:16:14, 5.78s/it]
88%|████████▊ | 38787/43939 [62:34:09<8:14:20, 5.76s/it]
88%|████████▊ | 38788/43939 [62:34:15<8:13:45, 5.75s/it]
88%|████████▊ | 38789/43939 [62:34:21<8:13:04, 5.74s/it]
88%|████████▊ | 38790/43939 [62:34:27<8:13:30, 5.75s/it]
88%|████████▊ | 38791/43939 [62:34:32<8:13:25, 5.75s/it]
88%|████████▊ | 38792/43939 [62:34:38<8:11:56, 5.73s/it]
88%|████████▊ | 38793/43939 [62:34:44<8:10:53, 5.72s/it]
88%|████████▊ | 38794/43939 [62:34:50<8:11:05, 5.73s/it]
88%|████████▊ | 38795/43939 [62:34:55<8:11:05, 5.73s/it]
88%|████████▊ | 38796/43939 [62:35:01<8:11:33, 5.73s/it]
88%|████████▊ | 38797/43939 [62:35:07<8:10:25, 5.72s/it]
88%|████████▊ | 38798/43939 [62:35:12<8:09:13, 5.71s/it]
88%|████████▊ | 38799/43939 [62:35:18<8:08:11, 5.70s/it]
88%|████████▊ | 38800/43939 [62:35:24<8:08:39, 5.71s/it]
{'loss': 0.6331, 'grad_norm': 1.6226705312728882, 'learning_rate': 2.0553614520640363e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 464.41, 'epoch': 3.53}
88%|████████▊ | 38800/43939 [62:35:24<8:08:39, 5.71s/it]
88%|████████▊ | 38801/43939 [62:35:30<8:08:37, 5.71s/it]
88%|████████▊ | 38802/43939 [62:35:35<8:08:22, 5.70s/it]
88%|████████▊ | 38803/43939 [62:35:41<8:09:03, 5.71s/it]
88%|████████▊ | 38804/43939 [62:35:47<8:08:53, 5.71s/it]
88%|████████▊ | 38805/43939 [62:35:52<8:08:48, 5.71s/it]
88%|████████▊ | 38806/43939 [62:35:58<8:07:34, 5.70s/it]
88%|████████▊ | 38807/43939 [62:36:04<8:07:38, 5.70s/it]
88%|████████▊ | 38808/43939 [62:36:10<8:13:43, 5.77s/it]
88%|████████▊ | 38809/43939 [62:36:15<8:12:33, 5.76s/it]
88%|████████▊ | 38810/43939 [62:36:21<8:10:51, 5.74s/it]
88%|████████▊ | 38811/43939 [62:36:27<8:09:56, 5.73s/it]
88%|████████▊ | 38812/43939 [62:36:33<8:09:57, 5.73s/it]
88%|████████▊ | 38813/43939 [62:36:38<8:09:31, 5.73s/it]
88%|████████▊ | 38814/43939 [62:36:44<8:08:30, 5.72s/it]
88%|████████▊ | 38815/43939 [62:36:50<8:07:38, 5.71s/it]
88%|████████▊ | 38816/43939 [62:36:55<8:06:16, 5.70s/it]
88%|████████▊ | 38817/43939 [62:37:01<8:07:45, 5.71s/it]
88%|████████▊ | 38818/43939 [62:37:07<8:06:53, 5.70s/it]
88%|████████▊ | 38819/43939 [62:37:13<8:08:08, 5.72s/it]
88%|████████▊ | 38820/43939 [62:37:18<8:07:21, 5.71s/it]
88%|████████▊ | 38821/43939 [62:37:24<8:06:59, 5.71s/it]
88%|████████▊ | 38822/43939 [62:37:30<8:05:59, 5.70s/it]
88%|████████▊ | 38823/43939 [62:37:35<8:06:08, 5.70s/it]
88%|████████▊ | 38824/43939 [62:37:41<8:06:35, 5.71s/it]
88%|████████▊ | 38825/43939 [62:37:47<8:06:31, 5.71s/it]
88%|████████▊ | 38826/43939 [62:37:52<8:06:50, 5.71s/it]
88%|████████▊ | 38827/43939 [62:37:58<8:07:10, 5.72s/it]
88%|████████▊ | 38828/43939 [62:38:04<8:07:27, 5.72s/it]
88%|████████▊ | 38829/43939 [62:38:10<8:12:50, 5.79s/it]
88%|████████▊ | 38830/43939 [62:38:16<8:11:52, 5.78s/it]
88%|████████▊ | 38831/43939 [62:38:21<8:09:57, 5.76s/it]
88%|████████▊ | 38832/43939 [62:38:27<8:08:30, 5.74s/it]
88%|████████▊ | 38833/43939 [62:38:33<8:07:03, 5.72s/it]
88%|████████▊ | 38834/43939 [62:38:38<8:07:11, 5.73s/it]
88%|████████▊ | 38835/43939 [62:38:44<8:06:17, 5.72s/it]
88%|████████▊ | 38836/43939 [62:38:50<8:05:37, 5.71s/it]
88%|████████▊ | 38837/43939 [62:38:56<8:04:34, 5.70s/it]
88%|████████▊ | 38838/43939 [62:39:01<8:04:50, 5.70s/it]
88%|████████▊ | 38839/43939 [62:39:07<8:05:03, 5.71s/it]
88%|████████▊ | 38840/43939 [62:39:13<8:03:48, 5.69s/it]
88%|████████▊ | 38841/43939 [62:39:18<8:04:23, 5.70s/it]
88%|████████▊ | 38842/43939 [62:39:24<8:04:36, 5.70s/it]
88%|████████▊ | 38843/43939 [62:39:30<8:03:51, 5.70s/it]
88%|████████▊ | 38844/43939 [62:39:35<8:03:58, 5.70s/it]
88%|████████▊ | 38845/43939 [62:39:41<8:02:58, 5.69s/it]
88%|████████▊ | 38846/43939 [62:39:47<8:04:34, 5.71s/it]
88%|████████▊ | 38847/43939 [62:39:53<8:04:31, 5.71s/it]
88%|████████▊ | 38848/43939 [62:39:58<8:05:12, 5.72s/it]
88%|████████▊ | 38849/43939 [62:40:04<8:05:35, 5.72s/it]
88%|████████▊ | 38850/43939 [62:40:10<8:05:26, 5.72s/it]
{'loss': 0.6579, 'grad_norm': 1.609815239906311, 'learning_rate': 2.016112134917858e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.45, 'epoch': 3.54}
88%|████████▊ | 38850/43939 [62:40:10<8:05:26, 5.72s/it]
88%|████████▊ | 38851/43939 [62:40:16<8:10:57, 5.79s/it]
88%|████████▊ | 38852/43939 [62:40:21<8:08:59, 5.77s/it]
88%|████████▊ | 38853/43939 [62:40:27<8:07:12, 5.75s/it]
88%|████████▊ | 38854/43939 [62:40:33<8:05:37, 5.73s/it]
88%|████████▊ | 38855/43939 [62:40:39<8:04:38, 5.72s/it]
88%|████████▊ | 38856/43939 [62:40:44<8:04:07, 5.71s/it]
88%|████████▊ | 38857/43939 [62:40:50<8:04:54, 5.72s/it]
88%|████████▊ | 38858/43939 [62:40:56<8:03:43, 5.71s/it]
88%|████████▊ | 38859/43939 [62:41:01<8:03:18, 5.71s/it]
88%|████████▊ | 38860/43939 [62:41:07<8:02:13, 5.70s/it]
88%|████████▊ | 38861/43939 [62:41:13<8:01:54, 5.69s/it]
88%|████████▊ | 38862/43939 [62:41:18<8:02:28, 5.70s/it]
88%|████████▊ | 38863/43939 [62:41:24<8:03:57, 5.72s/it]
88%|████████▊ | 38864/43939 [62:41:30<8:02:46, 5.71s/it]
88%|████████▊ | 38865/43939 [62:41:36<8:02:17, 5.70s/it]
88%|████████▊ | 38866/43939 [62:41:41<8:01:56, 5.70s/it]
88%|████████▊ | 38867/43939 [62:41:47<8:03:01, 5.71s/it]
88%|████████▊ | 38868/43939 [62:41:53<8:02:03, 5.70s/it]
88%|████████▊ | 38869/43939 [62:41:58<8:01:54, 5.70s/it]
88%|████████▊ | 38870/43939 [62:42:04<8:00:35, 5.69s/it]
88%|████████▊ | 38871/43939 [62:42:10<8:01:12, 5.70s/it]
88%|████████▊ | 38872/43939 [62:42:16<8:07:17, 5.77s/it]
88%|████████▊ | 38873/43939 [62:42:21<8:07:02, 5.77s/it]
88%|████████▊ | 38874/43939 [62:42:27<8:04:55, 5.74s/it]
88%|████████▊ | 38875/43939 [62:42:33<8:03:25, 5.73s/it]
88%|████████▊ | 38876/43939 [62:42:39<8:02:52, 5.72s/it]
88%|████████▊ | 38877/43939 [62:42:44<8:03:04, 5.73s/it]
88%|████████▊ | 38878/43939 [62:42:50<8:02:30, 5.72s/it]
88%|████████▊ | 38879/43939 [62:42:56<8:01:45, 5.71s/it]
88%|████████▊ | 38880/43939 [62:43:01<8:01:22, 5.71s/it]
88%|████████▊ | 38881/43939 [62:43:07<8:01:56, 5.72s/it]
88%|████████▊ | 38882/43939 [62:43:13<8:01:07, 5.71s/it]
88%|████████▊ | 38883/43939 [62:43:19<8:01:18, 5.71s/it]
88%|████████▊ | 38884/43939 [62:43:24<8:00:34, 5.70s/it]
88%|████████▊ | 38885/43939 [62:43:30<8:00:31, 5.70s/it]
88%|████████▊ | 38886/43939 [62:43:36<7:59:39, 5.70s/it]
89%|████████▊ | 38887/43939 [62:43:41<8:00:20, 5.70s/it]
89%|████████▊ | 38888/43939 [62:43:47<8:00:33, 5.71s/it]
89%|████████▊ | 38889/43939 [62:43:53<8:01:14, 5.72s/it]
89%|████████▊ | 38890/43939 [62:43:58<8:00:23, 5.71s/it]
89%|████████▊ | 38891/43939 [62:44:04<8:00:18, 5.71s/it]
89%|████████▊ | 38892/43939 [62:44:10<8:00:14, 5.71s/it]
89%|████████▊ | 38893/43939 [62:44:16<8:00:02, 5.71s/it]
89%|████████▊ | 38894/43939 [62:44:22<8:05:23, 5.77s/it]
89%|████████▊ | 38895/43939 [62:44:27<8:03:07, 5.75s/it]
89%|████████▊ | 38896/43939 [62:44:33<8:02:26, 5.74s/it]
89%|████████▊ | 38897/43939 [62:44:39<8:01:23, 5.73s/it]
89%|████████▊ | 38898/43939 [62:44:44<8:01:39, 5.73s/it]
89%|████████▊ | 38899/43939 [62:44:50<8:00:45, 5.72s/it]
89%|████████▊ | 38900/43939 [62:44:56<7:59:57, 5.71s/it]
{'loss': 0.6504, 'grad_norm': 1.6627197265625, 'learning_rate': 1.9772254426141447e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 466.41, 'epoch': 3.54}
89%|████████▊ | 38900/43939 [62:44:56<7:59:57, 5.71s/it]
89%|████████▊ | 38901/43939 [62:45:01<7:59:05, 5.71s/it]
89%|████████▊ | 38902/43939 [62:45:07<7:59:26, 5.71s/it]
89%|████████▊ | 38903/43939 [62:45:13<7:59:03, 5.71s/it]
89%|████████▊ | 38904/43939 [62:45:19<7:58:41, 5.70s/it]
89%|████████▊ | 38905/43939 [62:45:24<7:58:48, 5.71s/it]
89%|████████▊ | 38906/43939 [62:45:30<7:57:00, 5.69s/it]
89%|████████▊ | 38907/43939 [62:45:36<7:57:16, 5.69s/it]
89%|████████▊ | 38908/43939 [62:45:41<7:58:19, 5.70s/it]
89%|████████▊ | 38909/43939 [62:45:47<7:57:57, 5.70s/it]
89%|████████▊ | 38910/43939 [62:45:53<7:57:39, 5.70s/it]
89%|████████▊ | 38911/43939 [62:45:58<7:58:13, 5.71s/it]
89%|████████▊ | 38912/43939 [62:46:04<7:58:45, 5.71s/it]
89%|████████▊ | 38913/43939 [62:46:10<7:59:28, 5.72s/it]
89%|████████▊ | 38914/43939 [62:46:16<7:58:53, 5.72s/it]
89%|████████▊ | 38915/43939 [62:46:22<8:03:55, 5.78s/it]
89%|████████▊ | 38916/43939 [62:46:27<8:01:42, 5.75s/it]
89%|████████▊ | 38917/43939 [62:46:33<8:01:02, 5.75s/it]
89%|████████▊ | 38918/43939 [62:46:39<7:59:13, 5.73s/it]
89%|████████▊ | 38919/43939 [62:46:44<7:58:23, 5.72s/it]
89%|████████▊ | 38920/43939 [62:46:50<7:58:02, 5.71s/it]
89%|████████▊ | 38921/43939 [62:46:56<7:57:57, 5.71s/it]
89%|████████▊ | 38922/43939 [62:47:01<7:56:43, 5.70s/it]
89%|████████▊ | 38923/43939 [62:47:07<7:56:22, 5.70s/it]
89%|████████▊ | 38924/43939 [62:47:13<7:55:37, 5.69s/it]
89%|████████▊ | 38925/43939 [62:47:19<7:56:10, 5.70s/it]
89%|████████▊ | 38926/43939 [62:47:24<7:56:58, 5.71s/it]
89%|████████▊ | 38927/43939 [62:47:30<7:56:42, 5.71s/it]
89%|████████▊ | 38928/43939 [62:47:36<7:56:47, 5.71s/it]
89%|████████▊ | 38929/43939 [62:47:41<7:56:30, 5.71s/it]
89%|████████▊ | 38930/43939 [62:47:47<7:56:32, 5.71s/it]
89%|████████▊ | 38931/43939 [62:47:53<7:55:52, 5.70s/it]
89%|████████▊ | 38932/43939 [62:47:59<7:56:10, 5.71s/it]
89%|████████▊ | 38933/43939 [62:48:04<7:55:30, 5.70s/it]
89%|████████▊ | 38934/43939 [62:48:10<7:54:42, 5.69s/it]
89%|████████▊ | 38935/43939 [62:48:16<7:56:15, 5.71s/it]
89%|████████▊ | 38936/43939 [62:48:22<8:01:24, 5.77s/it]
89%|████████▊ | 38937/43939 [62:48:27<7:59:39, 5.75s/it]
89%|████████▊ | 38938/43939 [62:48:33<7:57:44, 5.73s/it]
89%|████████▊ | 38939/43939 [62:48:39<7:55:35, 5.71s/it]
89%|████████▊ | 38940/43939 [62:48:44<7:55:55, 5.71s/it]
89%|████████▊ | 38941/43939 [62:48:50<7:55:40, 5.71s/it]
89%|████████▊ | 38942/43939 [62:48:56<7:55:30, 5.71s/it]
89%|████████▊ | 38943/43939 [62:49:01<7:55:33, 5.71s/it]
89%|████████▊ | 38944/43939 [62:49:07<7:55:09, 5.71s/it]
89%|████████▊ | 38945/43939 [62:49:13<7:55:15, 5.71s/it]
89%|████████▊ | 38946/43939 [62:49:19<7:55:30, 5.71s/it]
89%|████████▊ | 38947/43939 [62:49:24<7:55:20, 5.71s/it]
89%|████████▊ | 38948/43939 [62:49:30<7:54:57, 5.71s/it]
89%|████████▊ | 38949/43939 [62:49:36<7:54:46, 5.71s/it]
89%|████████▊ | 38950/43939 [62:49:41<7:54:10, 5.70s/it]
{'loss': 0.6573, 'grad_norm': 1.826428771018982, 'learning_rate': 1.9387019886818515e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.56, 'epoch': 3.55}
89%|████████▊ | 38950/43939 [62:49:41<7:54:10, 5.70s/it]
89%|████████▊ | 38951/43939 [62:49:47<7:53:55, 5.70s/it]
89%|████████▊ | 38952/43939 [62:49:53<7:55:10, 5.72s/it]
89%|████████▊ | 38953/43939 [62:49:59<7:56:19, 5.73s/it]
89%|████████▊ | 38954/43939 [62:50:04<7:54:34, 5.71s/it]
89%|████████▊ | 38955/43939 [62:50:10<7:54:14, 5.71s/it]
89%|████████▊ | 38956/43939 [62:50:16<7:54:35, 5.71s/it]
89%|████████▊ | 38957/43939 [62:50:21<7:54:51, 5.72s/it]
89%|████████▊ | 38958/43939 [62:50:27<7:59:22, 5.77s/it]
89%|████████▊ | 38959/43939 [62:50:33<7:57:45, 5.76s/it]
89%|████████▊ | 38960/43939 [62:50:39<7:56:45, 5.75s/it]
89%|████████▊ | 38961/43939 [62:50:44<7:55:24, 5.73s/it]
89%|████████▊ | 38962/43939 [62:50:50<7:53:51, 5.71s/it]
89%|████████▊ | 38963/43939 [62:50:56<7:52:49, 5.70s/it]
89%|████████▊ | 38964/43939 [62:51:01<7:52:02, 5.69s/it]
89%|████████▊ | 38965/43939 [62:51:07<7:52:28, 5.70s/it]
89%|████████▊ | 38966/43939 [62:51:13<7:53:47, 5.72s/it]
89%|████████▊ | 38967/43939 [62:51:19<7:54:12, 5.72s/it]
89%|████████▊ | 38968/43939 [62:51:24<7:55:57, 5.74s/it]
89%|████████▊ | 38969/43939 [62:51:30<7:54:01, 5.72s/it]
89%|████████▊ | 38970/43939 [62:51:36<7:54:24, 5.73s/it]
89%|████████▊ | 38971/43939 [62:51:42<7:54:03, 5.73s/it]
89%|████████▊ | 38972/43939 [62:51:47<7:53:43, 5.72s/it]
89%|████████▊ | 38973/43939 [62:51:53<7:53:44, 5.72s/it]
89%|████████▊ | 38974/43939 [62:51:59<7:52:47, 5.71s/it]
89%|████████▊ | 38975/43939 [62:52:04<7:52:35, 5.71s/it]
89%|████████▊ | 38976/43939 [62:52:10<7:52:02, 5.71s/it]
89%|████████▊ | 38977/43939 [62:52:16<7:51:22, 5.70s/it]
89%|████████▊ | 38978/43939 [62:52:22<7:51:24, 5.70s/it]
89%|████████▊ | 38979/43939 [62:52:27<7:51:16, 5.70s/it]
89%|████████▊ | 38980/43939 [62:52:33<7:56:45, 5.77s/it]
89%|████████▊ | 38981/43939 [62:52:39<7:56:43, 5.77s/it]
89%|████████▊ | 38982/43939 [62:52:45<7:55:15, 5.75s/it]
89%|████████▊ | 38983/43939 [62:52:50<7:53:05, 5.73s/it]
89%|████████▊ | 38984/43939 [62:52:56<7:53:00, 5.73s/it]
89%|████████▊ | 38985/43939 [62:53:02<7:52:11, 5.72s/it]
89%|████████▊ | 38986/43939 [62:53:07<7:51:36, 5.71s/it]
89%|████████▊ | 38987/43939 [62:53:13<7:51:05, 5.71s/it]
89%|████████▊ | 38988/43939 [62:53:19<7:50:51, 5.71s/it]
89%|████████▊ | 38989/43939 [62:53:25<7:51:55, 5.72s/it]
89%|████████▊ | 38990/43939 [62:53:30<7:51:56, 5.72s/it]
89%|████████▊ | 38991/43939 [62:53:36<7:50:52, 5.71s/it]
89%|████████▊ | 38992/43939 [62:53:42<7:51:21, 5.72s/it]
89%|████████▊ | 38993/43939 [62:53:47<7:50:19, 5.71s/it]
89%|████████▊ | 38994/43939 [62:53:53<7:50:11, 5.70s/it]
89%|████████▊ | 38995/43939 [62:53:59<7:51:32, 5.72s/it]
89%|████████▉ | 38996/43939 [62:54:05<7:50:43, 5.71s/it]
89%|████████▉ | 38997/43939 [62:54:10<7:50:02, 5.71s/it]
89%|████████▉ | 38998/43939 [62:54:16<7:50:51, 5.72s/it]
89%|████████▉ | 38999/43939 [62:54:22<7:51:16, 5.72s/it]
89%|████████▉ | 39000/43939 [62:54:28<7:56:26, 5.79s/it]
{'loss': 0.6643, 'grad_norm': 1.4561930894851685, 'learning_rate': 1.9005423809189959e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.81, 'epoch': 3.55}
89%|████████▉ | 39000/43939 [62:54:28<7:56:26, 5.79s/it]
89%|████████▉ | 39001/43939 [62:54:33<7:55:51, 5.78s/it]
89%|████████▉ | 39002/43939 [62:54:39<7:54:13, 5.76s/it]
89%|████████▉ | 39003/43939 [62:54:45<7:52:41, 5.75s/it]
89%|████████▉ | 39004/43939 [62:54:51<7:51:07, 5.73s/it]
89%|████████▉ | 39005/43939 [62:54:56<7:50:53, 5.73s/it]
89%|████████▉ | 39006/43939 [62:55:02<7:50:51, 5.73s/it]
89%|████████▉ | 39007/43939 [62:55:08<7:49:50, 5.72s/it]
89%|████████▉ | 39008/43939 [62:55:13<7:50:23, 5.72s/it]
89%|████████▉ | 39009/43939 [62:55:19<7:49:29, 5.71s/it]
89%|████████▉ | 39010/43939 [62:55:25<7:48:32, 5.70s/it]
89%|████████▉ | 39011/43939 [62:55:31<7:48:11, 5.70s/it]
89%|████████▉ | 39012/43939 [62:55:36<7:49:08, 5.71s/it]
89%|████████▉ | 39013/43939 [62:55:42<7:48:42, 5.71s/it]
89%|████████▉ | 39014/43939 [62:55:48<7:49:57, 5.73s/it]
89%|████████▉ | 39015/43939 [62:55:53<7:48:29, 5.71s/it]
89%|████████▉ | 39016/43939 [62:55:59<7:48:25, 5.71s/it]
89%|████████▉ | 39017/43939 [62:56:05<7:48:36, 5.71s/it]
89%|████████▉ | 39018/43939 [62:56:11<7:48:48, 5.72s/it]
89%|████████▉ | 39019/43939 [62:56:16<7:48:43, 5.72s/it]
89%|████████▉ | 39020/43939 [62:56:22<7:47:57, 5.71s/it]
89%|████████▉ | 39021/43939 [62:56:28<7:47:09, 5.70s/it]
89%|████████▉ | 39022/43939 [62:56:34<7:53:16, 5.78s/it]
89%|████████▉ | 39023/43939 [62:56:39<7:51:49, 5.76s/it]
89%|████████▉ | 39024/43939 [62:56:45<7:52:07, 5.76s/it]
89%|████████▉ | 39025/43939 [62:56:51<7:50:18, 5.74s/it]
89%|████████▉ | 39026/43939 [62:56:56<7:49:29, 5.73s/it]
89%|████████▉ | 39027/43939 [62:57:02<7:48:21, 5.72s/it]
89%|████████▉ | 39028/43939 [62:57:08<7:47:47, 5.72s/it]
89%|████████▉ | 39029/43939 [62:57:14<7:47:01, 5.71s/it]
89%|████████▉ | 39030/43939 [62:57:19<7:47:55, 5.72s/it]
89%|████████▉ | 39031/43939 [62:57:25<7:47:05, 5.71s/it]
89%|████████▉ | 39032/43939 [62:57:31<7:46:33, 5.70s/it]
89%|████████▉ | 39033/43939 [62:57:36<7:46:52, 5.71s/it]
89%|████████▉ | 39034/43939 [62:57:42<7:46:45, 5.71s/it]
89%|████████▉ | 39035/43939 [62:57:48<7:46:44, 5.71s/it]
89%|████████▉ | 39036/43939 [62:57:54<7:47:33, 5.72s/it]
89%|████████▉ | 39037/43939 [62:57:59<7:47:05, 5.72s/it]
89%|████████▉ | 39038/43939 [62:58:05<7:46:23, 5.71s/it]
89%|████████▉ | 39039/43939 [62:58:11<7:46:44, 5.72s/it]
89%|████████▉ | 39040/43939 [62:58:16<7:46:35, 5.71s/it]
89%|████████▉ | 39041/43939 [62:58:22<7:45:47, 5.71s/it]
89%|████████▉ | 39042/43939 [62:58:28<7:47:15, 5.73s/it]
89%|████████▉ | 39043/43939 [62:58:34<7:47:08, 5.72s/it]
89%|████████▉ | 39044/43939 [62:58:40<7:52:21, 5.79s/it]
89%|████████▉ | 39045/43939 [62:58:45<7:49:37, 5.76s/it]
89%|████████▉ | 39046/43939 [62:58:51<7:48:33, 5.75s/it]
89%|████████▉ | 39047/43939 [62:58:57<7:46:37, 5.72s/it]
89%|████████▉ | 39048/43939 [62:59:02<7:45:28, 5.71s/it]
89%|████████▉ | 39049/43939 [62:59:08<7:45:07, 5.71s/it]
89%|████████▉ | 39050/43939 [62:59:14<7:45:13, 5.71s/it]
{'loss': 0.6523, 'grad_norm': 1.8014180660247803, 'learning_rate': 1.8627472213830572e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.78, 'epoch': 3.55}
89%|████████▉ | 39050/43939 [62:59:14<7:45:13, 5.71s/it]
89%|████████▉ | 39051/43939 [62:59:19<7:46:09, 5.72s/it]
89%|████████▉ | 39052/43939 [62:59:25<7:46:57, 5.73s/it]
89%|████████▉ | 39053/43939 [62:59:31<7:45:36, 5.72s/it]
89%|████████▉ | 39054/43939 [62:59:37<7:44:15, 5.70s/it]
89%|████████▉ | 39055/43939 [62:59:42<7:45:07, 5.71s/it]
89%|████████▉ | 39056/43939 [62:59:48<7:43:53, 5.70s/it]
89%|████████▉ | 39057/43939 [62:59:54<7:43:48, 5.70s/it]
89%|████████▉ | 39058/43939 [62:59:59<7:42:51, 5.69s/it]
89%|████████▉ | 39059/43939 [63:00:05<7:44:00, 5.70s/it]
89%|████████▉ | 39060/43939 [63:00:11<7:43:19, 5.70s/it]
89%|████████▉ | 39061/43939 [63:00:16<7:42:53, 5.69s/it]
89%|████████▉ | 39062/43939 [63:00:22<7:42:30, 5.69s/it]
89%|████████▉ | 39063/43939 [63:00:28<7:42:49, 5.70s/it]
89%|████████▉ | 39064/43939 [63:00:33<7:42:05, 5.69s/it]
89%|████████▉ | 39065/43939 [63:00:39<7:47:16, 5.75s/it]
89%|████████▉ | 39066/43939 [63:00:45<7:46:21, 5.74s/it]
89%|████████▉ | 39067/43939 [63:00:51<7:44:48, 5.72s/it]
89%|████████▉ | 39068/43939 [63:00:56<7:43:56, 5.71s/it]
89%|████████▉ | 39069/43939 [63:01:02<7:43:08, 5.71s/it]
89%|████████▉ | 39070/43939 [63:01:08<7:43:09, 5.71s/it]
89%|████████▉ | 39071/43939 [63:01:14<7:44:01, 5.72s/it]
89%|████████▉ | 39072/43939 [63:01:19<7:43:10, 5.71s/it]
89%|████████▉ | 39073/43939 [63:01:25<7:43:06, 5.71s/it]
89%|████████▉ | 39074/43939 [63:01:31<7:43:11, 5.71s/it]
89%|████████▉ | 39075/43939 [63:01:36<7:42:05, 5.70s/it]
89%|████████▉ | 39076/43939 [63:01:42<7:42:34, 5.71s/it]
89%|████████▉ | 39077/43939 [63:01:48<7:44:18, 5.73s/it]
89%|████████▉ | 39078/43939 [63:01:54<7:43:28, 5.72s/it]
89%|████████▉ | 39079/43939 [63:01:59<7:42:27, 5.71s/it]
89%|████████▉ | 39080/43939 [63:02:05<7:43:18, 5.72s/it]
89%|████████▉ | 39081/43939 [63:02:11<7:44:42, 5.74s/it]
89%|████████▉ | 39082/43939 [63:02:17<7:43:10, 5.72s/it]
89%|████████▉ | 39083/43939 [63:02:22<7:42:03, 5.71s/it]
89%|████████▉ | 39084/43939 [63:02:28<7:43:53, 5.73s/it]
89%|████████▉ | 39085/43939 [63:02:34<7:42:59, 5.72s/it]
89%|████████▉ | 39086/43939 [63:02:39<7:42:33, 5.72s/it]
89%|████████▉ | 39087/43939 [63:02:45<7:48:30, 5.79s/it]
89%|████████▉ | 39088/43939 [63:02:51<7:46:04, 5.76s/it]
89%|████████▉ | 39089/43939 [63:02:57<7:45:29, 5.76s/it]
89%|████████▉ | 39090/43939 [63:03:02<7:43:39, 5.74s/it]
89%|████████▉ | 39091/43939 [63:03:08<7:42:41, 5.73s/it]
89%|████████▉ | 39092/43939 [63:03:14<7:43:09, 5.73s/it]
89%|████████▉ | 39093/43939 [63:03:20<7:42:17, 5.72s/it]
89%|████████▉ | 39094/43939 [63:03:25<7:41:52, 5.72s/it]
89%|████████▉ | 39095/43939 [63:03:31<7:41:49, 5.72s/it]
89%|████████▉ | 39096/43939 [63:03:37<7:41:52, 5.72s/it]
89%|████████▉ | 39097/43939 [63:03:43<7:41:14, 5.72s/it]
89%|████████▉ | 39098/43939 [63:03:48<7:41:02, 5.71s/it]
89%|████████▉ | 39099/43939 [63:03:54<7:40:02, 5.70s/it]
89%|████████▉ | 39100/43939 [63:04:00<7:39:18, 5.70s/it]
{'loss': 0.6537, 'grad_norm': 1.6044520139694214, 'learning_rate': 1.8253171063815023e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.64, 'epoch': 3.56}
89%|████████▉ | 39100/43939 [63:04:00<7:39:18, 5.70s/it]
89%|████████▉ | 39101/43939 [63:04:05<7:39:05, 5.69s/it]
89%|████████▉ | 39102/43939 [63:04:11<7:38:47, 5.69s/it]
89%|████████▉ | 39103/43939 [63:04:17<7:38:50, 5.69s/it]
89%|████████▉ | 39104/43939 [63:04:22<7:39:19, 5.70s/it]
89%|████████▉ | 39105/43939 [63:04:28<7:39:11, 5.70s/it]
89%|████████▉ | 39106/43939 [63:04:34<7:38:39, 5.69s/it]
89%|████████▉ | 39107/43939 [63:04:39<7:39:22, 5.70s/it]
89%|████████▉ | 39108/43939 [63:04:45<7:44:57, 5.77s/it]
89%|████████▉ | 39109/43939 [63:04:51<7:43:47, 5.76s/it]
89%|████████▉ | 39110/43939 [63:04:57<7:43:59, 5.77s/it]
89%|████████▉ | 39111/43939 [63:05:03<7:42:10, 5.74s/it]
89%|████████▉ | 39112/43939 [63:05:08<7:41:30, 5.74s/it]
89%|████████▉ | 39113/43939 [63:05:14<7:41:02, 5.73s/it]
89%|████████▉ | 39114/43939 [63:05:20<7:39:58, 5.72s/it]
89%|████████▉ | 39115/43939 [63:05:25<7:39:40, 5.72s/it]
89%|████████▉ | 39116/43939 [63:05:31<7:38:52, 5.71s/it]
89%|████████▉ | 39117/43939 [63:05:37<7:37:47, 5.70s/it]
89%|████████▉ | 39118/43939 [63:05:43<7:38:37, 5.71s/it]
89%|████████▉ | 39119/43939 [63:05:48<7:38:58, 5.71s/it]
89%|████████▉ | 39120/43939 [63:05:54<7:38:15, 5.71s/it]
89%|████████▉ | 39121/43939 [63:06:00<7:38:22, 5.71s/it]
89%|████████▉ | 39122/43939 [63:06:05<7:37:40, 5.70s/it]
89%|████████▉ | 39123/43939 [63:06:11<7:36:53, 5.69s/it]
89%|████████▉ | 39124/43939 [63:06:17<7:36:55, 5.69s/it]
89%|████████▉ | 39125/43939 [63:06:22<7:37:06, 5.70s/it]
89%|████████▉ | 39126/43939 [63:06:28<7:36:30, 5.69s/it]
89%|████████▉ | 39127/43939 [63:06:34<7:35:49, 5.68s/it]
89%|████████▉ | 39128/43939 [63:06:39<7:36:21, 5.69s/it]
89%|████████▉ | 39129/43939 [63:06:45<7:35:59, 5.69s/it]
89%|████████▉ | 39130/43939 [63:06:51<7:40:59, 5.75s/it]
89%|████████▉ | 39131/43939 [63:06:57<7:39:02, 5.73s/it]
89%|████████▉ | 39132/43939 [63:07:02<7:38:15, 5.72s/it]
89%|████████▉ | 39133/43939 [63:07:08<7:37:52, 5.72s/it]
89%|████████▉ | 39134/43939 [63:07:14<7:37:07, 5.71s/it]
89%|████████▉ | 39135/43939 [63:07:20<7:37:36, 5.72s/it]
89%|████████▉ | 39136/43939 [63:07:25<7:37:52, 5.72s/it]
89%|████████▉ | 39137/43939 [63:07:31<7:38:18, 5.73s/it]
89%|████████▉ | 39138/43939 [63:07:37<7:37:22, 5.72s/it]
89%|████████▉ | 39139/43939 [63:07:42<7:37:02, 5.71s/it]
89%|████████▉ | 39140/43939 [63:07:48<7:36:11, 5.70s/it]
89%|████████▉ | 39141/43939 [63:07:54<7:35:13, 5.69s/it]
89%|████████▉ | 39142/43939 [63:07:59<7:34:50, 5.69s/it]
89%|████████▉ | 39143/43939 [63:08:05<7:34:31, 5.69s/it]
89%|████████▉ | 39144/43939 [63:08:11<7:33:55, 5.68s/it]
89%|████████▉ | 39145/43939 [63:08:17<7:34:38, 5.69s/it]
89%|████████▉ | 39146/43939 [63:08:22<7:34:29, 5.69s/it]
89%|████████▉ | 39147/43939 [63:08:28<7:33:58, 5.68s/it]
89%|████████▉ | 39148/43939 [63:08:34<7:33:14, 5.68s/it]
89%|████████▉ | 39149/43939 [63:08:39<7:36:42, 5.72s/it]
89%|████████▉ | 39150/43939 [63:08:45<7:36:32, 5.72s/it]
{'loss': 0.6527, 'grad_norm': 1.4636297225952148, 'learning_rate': 1.78825262646235e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.22, 'epoch': 3.56}
89%|████████▉ | 39150/43939 [63:08:45<7:36:32, 5.72s/it]
89%|████████▉ | 39151/43939 [63:08:51<7:42:26, 5.79s/it]
89%|████████▉ | 39152/43939 [63:08:57<7:40:00, 5.77s/it]
89%|████████▉ | 39153/43939 [63:09:02<7:38:48, 5.75s/it]
89%|████████▉ | 39154/43939 [63:09:08<7:37:56, 5.74s/it]
89%|████████▉ | 39155/43939 [63:09:14<7:36:24, 5.72s/it]
89%|████████▉ | 39156/43939 [63:09:20<7:35:14, 5.71s/it]
89%|████████▉ | 39157/43939 [63:09:25<7:34:09, 5.70s/it]
89%|████████▉ | 39158/43939 [63:09:31<7:33:13, 5.69s/it]
89%|████████▉ | 39159/43939 [63:09:37<7:34:39, 5.71s/it]
89%|████████▉ | 39160/43939 [63:09:42<7:34:27, 5.71s/it]
89%|████████▉ | 39161/43939 [63:09:48<7:35:26, 5.72s/it]
89%|████████▉ | 39162/43939 [63:09:54<7:35:33, 5.72s/it]
89%|████████▉ | 39163/43939 [63:10:00<7:36:13, 5.73s/it]
89%|████████▉ | 39164/43939 [63:10:05<7:35:10, 5.72s/it]
89%|████████▉ | 39165/43939 [63:10:11<7:35:26, 5.72s/it]
89%|████████▉ | 39166/43939 [63:10:17<7:35:54, 5.73s/it]
89%|████████▉ | 39167/43939 [63:10:22<7:34:49, 5.72s/it]
89%|████████▉ | 39168/43939 [63:10:28<7:34:21, 5.71s/it]
89%|████████▉ | 39169/43939 [63:10:34<7:34:15, 5.71s/it]
89%|████████▉ | 39170/43939 [63:10:40<7:34:35, 5.72s/it]
89%|████████▉ | 39171/43939 [63:10:45<7:34:55, 5.72s/it]
89%|████████▉ | 39172/43939 [63:10:51<7:34:25, 5.72s/it]
89%|████████▉ | 39173/43939 [63:10:57<7:39:18, 5.78s/it]
89%|████████▉ | 39174/43939 [63:11:03<7:36:42, 5.75s/it]
89%|████████▉ | 39175/43939 [63:11:08<7:35:20, 5.73s/it]
89%|████████▉ | 39176/43939 [63:11:14<7:34:00, 5.72s/it]
89%|████████▉ | 39177/43939 [63:11:20<7:33:32, 5.71s/it]
89%|████████▉ | 39178/43939 [63:11:25<7:33:37, 5.72s/it]
89%|████████▉ | 39179/43939 [63:11:31<7:33:06, 5.71s/it]
89%|████████▉ | 39180/43939 [63:11:37<7:32:42, 5.71s/it]
89%|████████▉ | 39181/43939 [63:11:43<7:33:32, 5.72s/it]
89%|████████▉ | 39182/43939 [63:11:48<7:32:31, 5.71s/it]
89%|████████▉ | 39183/43939 [63:11:54<7:32:23, 5.71s/it]
89%|████████▉ | 39184/43939 [63:12:00<7:32:26, 5.71s/it]
89%|████████▉ | 39185/43939 [63:12:05<7:32:07, 5.71s/it]
89%|████████▉ | 39186/43939 [63:12:11<7:31:51, 5.70s/it]
89%|████████▉ | 39187/43939 [63:12:17<7:32:02, 5.71s/it]
89%|████████▉ | 39188/43939 [63:12:23<7:33:14, 5.72s/it]
89%|████████▉ | 39189/43939 [63:12:28<7:32:21, 5.71s/it]
89%|████████▉ | 39190/43939 [63:12:34<7:31:20, 5.70s/it]
89%|████████▉ | 39191/43939 [63:12:40<7:31:52, 5.71s/it]
89%|████████▉ | 39192/43939 [63:12:45<7:31:49, 5.71s/it]
89%|████████▉ | 39193/43939 [63:12:51<7:32:42, 5.72s/it]
89%|████████▉ | 39194/43939 [63:12:57<7:31:50, 5.71s/it]
89%|████████▉ | 39195/43939 [63:13:03<7:36:18, 5.77s/it]
89%|████████▉ | 39196/43939 [63:13:08<7:34:42, 5.75s/it]
89%|████████▉ | 39197/43939 [63:13:14<7:32:50, 5.73s/it]
89%|████████▉ | 39198/43939 [63:13:20<7:32:04, 5.72s/it]
89%|████████▉ | 39199/43939 [63:13:26<7:31:51, 5.72s/it]
89%|████████▉ | 39200/43939 [63:13:31<7:30:59, 5.71s/it]
{'loss': 0.6779, 'grad_norm': 1.907880187034607, 'learning_rate': 1.7515543664048728e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.82, 'epoch': 3.57}
89%|████████▉ | 39200/43939 [63:13:31<7:30:59, 5.71s/it]
89%|████████▉ | 39201/43939 [63:13:37<7:31:44, 5.72s/it]
89%|████████▉ | 39202/43939 [63:13:43<7:31:20, 5.72s/it]
89%|████████▉ | 39203/43939 [63:13:48<7:31:40, 5.72s/it]
89%|████████▉ | 39204/43939 [63:13:54<7:31:14, 5.72s/it]
89%|████████▉ | 39205/43939 [63:14:00<7:30:30, 5.71s/it]
89%|████████▉ | 39206/43939 [63:14:05<7:30:04, 5.71s/it]
89%|████████▉ | 39207/43939 [63:14:11<7:29:51, 5.70s/it]
89%|████████▉ | 39208/43939 [63:14:17<7:29:49, 5.70s/it]
89%|████████▉ | 39209/43939 [63:14:23<7:29:48, 5.71s/it]
89%|████████▉ | 39210/43939 [63:14:28<7:29:34, 5.70s/it]
89%|████████▉ | 39211/43939 [63:14:34<7:29:20, 5.70s/it]
89%|████████▉ | 39212/43939 [63:14:40<7:29:56, 5.71s/it]
89%|████████▉ | 39213/43939 [63:14:45<7:30:19, 5.72s/it]
89%|████████▉ | 39214/43939 [63:14:51<7:29:56, 5.71s/it]
89%|████████▉ | 39215/43939 [63:14:57<7:28:35, 5.70s/it]
89%|████████▉ | 39216/43939 [63:15:03<7:33:55, 5.77s/it]
89%|████████▉ | 39217/43939 [63:15:08<7:32:43, 5.75s/it]
89%|████████▉ | 39218/43939 [63:15:14<7:31:45, 5.74s/it]
89%|████████▉ | 39219/43939 [63:15:20<7:29:40, 5.72s/it]
89%|████████▉ | 39220/43939 [63:15:26<7:29:25, 5.71s/it]
89%|████████▉ | 39221/43939 [63:15:31<7:28:42, 5.71s/it]
89%|████████▉ | 39222/43939 [63:15:37<7:28:10, 5.70s/it]
89%|████████▉ | 39223/43939 [63:15:43<7:28:15, 5.70s/it]
89%|████████▉ | 39224/43939 [63:15:48<7:27:40, 5.70s/it]
89%|████████▉ | 39225/43939 [63:15:54<7:27:26, 5.70s/it]
89%|████████▉ | 39226/43939 [63:16:00<7:27:28, 5.70s/it]
89%|████████▉ | 39227/43939 [63:16:05<7:28:44, 5.71s/it]
89%|████████▉ | 39228/43939 [63:16:11<7:27:29, 5.70s/it]
89%|████████▉ | 39229/43939 [63:16:17<7:27:30, 5.70s/it]
89%|████████▉ | 39230/43939 [63:16:23<7:27:35, 5.70s/it]
89%|████████▉ | 39231/43939 [63:16:28<7:26:31, 5.69s/it]
89%|████████▉ | 39232/43939 [63:16:34<7:26:36, 5.69s/it]
89%|████████▉ | 39233/43939 [63:16:40<7:25:56, 5.69s/it]
89%|████████▉ | 39234/43939 [63:16:45<7:25:08, 5.68s/it]
89%|████████▉ | 39235/43939 [63:16:51<7:25:22, 5.68s/it]
89%|████████▉ | 39236/43939 [63:16:57<7:25:04, 5.68s/it]
89%|████████▉ | 39237/43939 [63:17:02<7:30:09, 5.74s/it]
89%|████████▉ | 39238/43939 [63:17:08<7:28:56, 5.73s/it]
89%|████████▉ | 39239/43939 [63:17:14<7:28:07, 5.72s/it]
89%|████████▉ | 39240/43939 [63:17:20<7:28:12, 5.72s/it]
89%|████████▉ | 39241/43939 [63:17:25<7:26:36, 5.70s/it]
89%|████████▉ | 39242/43939 [63:17:31<7:26:14, 5.70s/it]
89%|████████▉ | 39243/43939 [63:17:37<7:25:55, 5.70s/it]
89%|████████▉ | 39244/43939 [63:17:42<7:26:00, 5.70s/it]
89%|████████▉ | 39245/43939 [63:17:48<7:25:56, 5.70s/it]
89%|████████▉ | 39246/43939 [63:17:54<7:26:31, 5.71s/it]
89%|████████▉ | 39247/43939 [63:17:59<7:25:27, 5.70s/it]
89%|████████▉ | 39248/43939 [63:18:05<7:25:15, 5.69s/it]
89%|████████▉ | 39249/43939 [63:18:11<7:25:08, 5.69s/it]
89%|████████▉ | 39250/43939 [63:18:17<7:26:26, 5.71s/it]
{'loss': 0.641, 'grad_norm': 1.7787017822265625, 'learning_rate': 1.715222905210362e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 457.56, 'epoch': 3.57}
89%|████████▉ | 39250/43939 [63:18:17<7:26:26, 5.71s/it]
89%|████████▉ | 39251/43939 [63:18:22<7:27:01, 5.72s/it]
89%|████████▉ | 39252/43939 [63:18:28<7:26:28, 5.72s/it]
89%|████████▉ | 39253/43939 [63:18:34<7:26:08, 5.71s/it]
89%|████████▉ | 39254/43939 [63:18:39<7:26:40, 5.72s/it]
89%|████████▉ | 39255/43939 [63:18:45<7:26:42, 5.72s/it]
89%|████████▉ | 39256/43939 [63:18:51<7:25:42, 5.71s/it]
89%|████████▉ | 39257/43939 [63:18:57<7:26:21, 5.72s/it]
89%|████████▉ | 39258/43939 [63:19:02<7:25:33, 5.71s/it]
89%|████████▉ | 39259/43939 [63:19:08<7:30:47, 5.78s/it]
89%|████████▉ | 39260/43939 [63:19:14<7:29:37, 5.77s/it]
89%|████████▉ | 39261/43939 [63:19:20<7:28:38, 5.75s/it]
89%|████████▉ | 39262/43939 [63:19:25<7:27:32, 5.74s/it]
89%|████████▉ | 39263/43939 [63:19:31<7:26:00, 5.72s/it]
89%|████████▉ | 39264/43939 [63:19:37<7:24:41, 5.71s/it]
89%|████████▉ | 39265/43939 [63:19:42<7:24:25, 5.71s/it]
89%|████████▉ | 39266/43939 [63:19:48<7:23:45, 5.70s/it]
89%|████████▉ | 39267/43939 [63:19:54<7:23:38, 5.70s/it]
89%|████████▉ | 39268/43939 [63:20:00<7:24:03, 5.70s/it]
89%|████████▉ | 39269/43939 [63:20:05<7:24:01, 5.70s/it]
89%|████████▉ | 39270/43939 [63:20:11<7:25:01, 5.72s/it]
89%|████████▉ | 39271/43939 [63:20:17<7:25:42, 5.73s/it]
89%|████████▉ | 39272/43939 [63:20:22<7:24:48, 5.72s/it]
89%|████████▉ | 39273/43939 [63:20:28<7:24:06, 5.71s/it]
89%|████████▉ | 39274/43939 [63:20:34<7:22:50, 5.70s/it]
89%|████████▉ | 39275/43939 [63:20:40<7:22:44, 5.70s/it]
89%|████████▉ | 39276/43939 [63:20:45<7:22:22, 5.69s/it]
89%|████████▉ | 39277/43939 [63:20:51<7:21:49, 5.69s/it]
89%|████████▉ | 39278/43939 [63:20:57<7:23:22, 5.71s/it]
89%|████████▉ | 39279/43939 [63:21:02<7:23:05, 5.71s/it]
89%|████████▉ | 39280/43939 [63:21:08<7:22:58, 5.70s/it]
89%|████████▉ | 39281/43939 [63:21:14<7:29:21, 5.79s/it]
89%|████████▉ | 39282/43939 [63:21:20<7:26:17, 5.75s/it]
89%|████████▉ | 39283/43939 [63:21:25<7:25:12, 5.74s/it]
89%|████████▉ | 39284/43939 [63:21:31<7:24:03, 5.72s/it]
89%|████████▉ | 39285/43939 [63:21:37<7:23:58, 5.72s/it]
89%|████████▉ | 39286/43939 [63:21:43<7:24:22, 5.73s/it]
89%|████████▉ | 39287/43939 [63:21:48<7:23:25, 5.72s/it]
89%|████████▉ | 39288/43939 [63:21:54<7:24:08, 5.73s/it]
89%|████████▉ | 39289/43939 [63:22:00<7:23:46, 5.73s/it]
89%|████████▉ | 39290/43939 [63:22:05<7:22:51, 5.72s/it]
89%|████████▉ | 39291/43939 [63:22:11<7:21:52, 5.70s/it]
89%|████████▉ | 39292/43939 [63:22:17<7:22:58, 5.72s/it]
89%|████████▉ | 39293/43939 [63:22:23<7:22:35, 5.72s/it]
89%|████████▉ | 39294/43939 [63:22:28<7:23:20, 5.73s/it]
89%|████████▉ | 39295/43939 [63:22:34<7:21:58, 5.71s/it]
89%|████████▉ | 39296/43939 [63:22:40<7:21:31, 5.71s/it]
89%|████████▉ | 39297/43939 [63:22:45<7:21:22, 5.70s/it]
89%|████████▉ | 39298/43939 [63:22:51<7:21:00, 5.70s/it]
89%|████████▉ | 39299/43939 [63:22:57<7:21:07, 5.70s/it]
89%|████████▉ | 39300/43939 [63:23:03<7:21:15, 5.71s/it]
{'loss': 0.6339, 'grad_norm': 1.5393340587615967, 'learning_rate': 1.679258816092999e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.29, 'epoch': 3.58}
89%|████████▉ | 39300/43939 [63:23:03<7:21:15, 5.71s/it]
89%|████████▉ | 39301/43939 [63:23:08<7:20:39, 5.70s/it]
89%|████████▉ | 39302/43939 [63:23:14<7:25:39, 5.77s/it]
89%|████████▉ | 39303/43939 [63:23:20<7:25:35, 5.77s/it]
89%|████████▉ | 39304/43939 [63:23:26<7:24:30, 5.75s/it]
89%|████████▉ | 39305/43939 [63:23:31<7:22:59, 5.74s/it]
89%|████████▉ | 39306/43939 [63:23:37<7:22:18, 5.73s/it]
89%|████████▉ | 39307/43939 [63:23:43<7:22:14, 5.73s/it]
89%|████████▉ | 39308/43939 [63:23:48<7:21:13, 5.72s/it]
89%|████████▉ | 39309/43939 [63:23:54<7:21:26, 5.72s/it]
89%|████████▉ | 39310/43939 [63:24:00<7:19:57, 5.70s/it]
89%|████████▉ | 39311/43939 [63:24:06<7:20:33, 5.71s/it]
89%|████████▉ | 39312/43939 [63:24:11<7:21:17, 5.72s/it]
89%|████████▉ | 39313/43939 [63:24:17<7:22:47, 5.74s/it]
89%|████████▉ | 39314/43939 [63:24:23<7:21:51, 5.73s/it]
89%|████████▉ | 39315/43939 [63:24:29<7:21:18, 5.73s/it]
89%|████████▉ | 39316/43939 [63:24:34<7:20:36, 5.72s/it]
89%|████████▉ | 39317/43939 [63:24:40<7:20:16, 5.72s/it]
89%|████████▉ | 39318/43939 [63:24:46<7:19:17, 5.70s/it]
89%|████████▉ | 39319/43939 [63:24:51<7:18:20, 5.69s/it]
89%|████████▉ | 39320/43939 [63:24:57<7:18:37, 5.70s/it]
89%|████████▉ | 39321/43939 [63:25:03<7:19:19, 5.71s/it]
89%|████████▉ | 39322/43939 [63:25:08<7:19:03, 5.71s/it]
89%|████████▉ | 39323/43939 [63:25:14<7:24:25, 5.78s/it]
89%|████████▉ | 39324/43939 [63:25:20<7:22:28, 5.75s/it]
89%|████████▉ | 39325/43939 [63:25:26<7:21:26, 5.74s/it]
90%|████████▉ | 39326/43939 [63:25:31<7:19:46, 5.72s/it]
90%|████████▉ | 39327/43939 [63:25:37<7:18:57, 5.71s/it]
90%|████████▉ | 39328/43939 [63:25:43<7:18:09, 5.70s/it]
90%|████████▉ | 39329/43939 [63:25:48<7:17:43, 5.70s/it]
90%|████████▉ | 39330/43939 [63:25:54<7:17:46, 5.70s/it]
90%|████████▉ | 39331/43939 [63:26:00<7:17:09, 5.69s/it]
90%|████████▉ | 39332/43939 [63:26:06<7:16:49, 5.69s/it]
90%|████████▉ | 39333/43939 [63:26:11<7:17:50, 5.70s/it]
90%|████████▉ | 39334/43939 [63:26:17<7:18:05, 5.71s/it]
90%|████████▉ | 39335/43939 [63:26:23<7:18:18, 5.71s/it]
90%|████████▉ | 39336/43939 [63:26:28<7:18:53, 5.72s/it]
90%|████████▉ | 39337/43939 [63:26:34<7:18:51, 5.72s/it]
90%|████████▉ | 39338/43939 [63:26:40<7:18:32, 5.72s/it]
90%|████████▉ | 39339/43939 [63:26:46<7:18:15, 5.72s/it]
90%|████████▉ | 39340/43939 [63:26:51<7:18:15, 5.72s/it]
90%|████████▉ | 39341/43939 [63:26:57<7:16:39, 5.70s/it]
90%|████████▉ | 39342/43939 [63:27:03<7:15:38, 5.69s/it]
90%|████████▉ | 39343/43939 [63:27:08<7:14:50, 5.68s/it]
90%|████████▉ | 39344/43939 [63:27:14<7:15:44, 5.69s/it]
90%|████████▉ | 39345/43939 [63:27:20<7:22:06, 5.77s/it]
90%|████████▉ | 39346/43939 [63:27:26<7:19:00, 5.73s/it]
90%|████████▉ | 39347/43939 [63:27:31<7:18:00, 5.72s/it]
90%|████████▉ | 39348/43939 [63:27:37<7:16:50, 5.71s/it]
90%|████████▉ | 39349/43939 [63:27:43<7:16:22, 5.70s/it]
90%|████████▉ | 39350/43939 [63:27:48<7:16:36, 5.71s/it]
{'loss': 0.6608, 'grad_norm': 1.6022719144821167, 'learning_rate': 1.6436626664707995e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.19, 'epoch': 3.58}
90%|████████▉ | 39350/43939 [63:27:48<7:16:36, 5.71s/it]
90%|████████▉ | 39351/43939 [63:27:54<7:16:56, 5.71s/it]
90%|████████▉ | 39352/43939 [63:28:00<7:16:32, 5.71s/it]
90%|████████▉ | 39353/43939 [63:28:06<7:16:36, 5.71s/it]
90%|████████▉ | 39354/43939 [63:28:11<7:17:16, 5.72s/it]
90%|████████▉ | 39355/43939 [63:28:17<7:17:00, 5.72s/it]
90%|████████▉ | 39356/43939 [63:28:23<7:16:16, 5.71s/it]
90%|████████▉ | 39357/43939 [63:28:28<7:14:43, 5.69s/it]
90%|████████▉ | 39358/43939 [63:28:34<7:15:04, 5.70s/it]
90%|████████▉ | 39359/43939 [63:28:40<7:14:55, 5.70s/it]
90%|████████▉ | 39360/43939 [63:28:45<7:15:28, 5.71s/it]
90%|████████▉ | 39361/43939 [63:28:51<7:15:41, 5.71s/it]
90%|████████▉ | 39362/43939 [63:28:57<7:15:01, 5.70s/it]
90%|████████▉ | 39363/43939 [63:29:03<7:16:26, 5.72s/it]
90%|████████▉ | 39364/43939 [63:29:08<7:15:56, 5.72s/it]
90%|████████▉ | 39365/43939 [63:29:14<7:14:55, 5.71s/it]
90%|████████▉ | 39366/43939 [63:29:20<7:21:06, 5.79s/it]
90%|████████▉ | 39367/43939 [63:29:26<7:19:05, 5.76s/it]
90%|████████▉ | 39368/43939 [63:29:31<7:18:39, 5.76s/it]
90%|████████▉ | 39369/43939 [63:29:37<7:17:10, 5.74s/it]
90%|████████▉ | 39370/43939 [63:29:43<7:15:29, 5.72s/it]
90%|████████▉ | 39371/43939 [63:29:49<7:15:42, 5.72s/it]
90%|████████▉ | 39372/43939 [63:29:54<7:16:49, 5.74s/it]
90%|████████▉ | 39373/43939 [63:30:00<7:16:05, 5.73s/it]
90%|████████▉ | 39374/43939 [63:30:06<7:14:56, 5.72s/it]
90%|████████▉ | 39375/43939 [63:30:11<7:14:35, 5.71s/it]
90%|████████▉ | 39376/43939 [63:30:17<7:14:22, 5.71s/it]
90%|████████▉ | 39377/43939 [63:30:23<7:14:22, 5.71s/it]
90%|████████▉ | 39378/43939 [63:30:29<7:13:41, 5.71s/it]
90%|████████▉ | 39379/43939 [63:30:34<7:13:30, 5.70s/it]
90%|████████▉ | 39380/43939 [63:30:40<7:12:31, 5.69s/it]
90%|████████▉ | 39381/43939 [63:30:46<7:12:49, 5.70s/it]
90%|████████▉ | 39382/43939 [63:30:51<7:12:52, 5.70s/it]
90%|████████▉ | 39383/43939 [63:30:57<7:13:15, 5.71s/it]
90%|████████▉ | 39384/43939 [63:31:03<7:12:59, 5.70s/it]
90%|████████▉ | 39385/43939 [63:31:08<7:13:18, 5.71s/it]
90%|████████▉ | 39386/43939 [63:31:14<7:13:28, 5.71s/it]
90%|████████▉ | 39387/43939 [63:31:20<7:18:40, 5.78s/it]
90%|████████▉ | 39388/43939 [63:31:26<7:17:47, 5.77s/it]
90%|████████▉ | 39389/43939 [63:31:32<7:16:29, 5.76s/it]
90%|████████▉ | 39390/43939 [63:31:37<7:14:47, 5.73s/it]
90%|████████▉ | 39391/43939 [63:31:43<7:13:17, 5.72s/it]
90%|████████▉ | 39392/43939 [63:31:49<7:13:50, 5.72s/it]
90%|████████▉ | 39393/43939 [63:31:54<7:13:21, 5.72s/it]
90%|████████▉ | 39394/43939 [63:32:00<7:12:51, 5.71s/it]
90%|████████▉ | 39395/43939 [63:32:06<7:12:08, 5.71s/it]
90%|████████▉ | 39396/43939 [63:32:12<7:11:21, 5.70s/it]
90%|████████▉ | 39397/43939 [63:32:17<7:13:12, 5.72s/it]
90%|████████▉ | 39398/43939 [63:32:23<7:11:53, 5.71s/it]
90%|████████▉ | 39399/43939 [63:32:29<7:11:41, 5.71s/it]
90%|████████▉ | 39400/43939 [63:32:34<7:11:46, 5.71s/it]
{'loss': 0.6446, 'grad_norm': 1.6085164546966553, 'learning_rate': 1.608435017956672e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.76, 'epoch': 3.59}
90%|████████▉ | 39400/43939 [63:32:34<7:11:46, 5.71s/it]
90%|████████▉ | 39401/43939 [63:32:40<7:10:59, 5.70s/it]
90%|████████▉ | 39402/43939 [63:32:46<7:10:37, 5.69s/it]
90%|████████▉ | 39403/43939 [63:32:51<7:10:25, 5.69s/it]
90%|████████▉ | 39404/43939 [63:32:57<7:09:59, 5.69s/it]
90%|████████▉ | 39405/43939 [63:33:03<7:10:30, 5.70s/it]
90%|████████▉ | 39406/43939 [63:33:09<7:11:34, 5.71s/it]
90%|████████▉ | 39407/43939 [63:33:14<7:11:02, 5.71s/it]
90%|████████▉ | 39408/43939 [63:33:20<7:16:55, 5.79s/it]
90%|████████▉ | 39409/43939 [63:33:26<7:15:28, 5.77s/it]
90%|████████▉ | 39410/43939 [63:33:32<7:13:18, 5.74s/it]
90%|████████▉ | 39411/43939 [63:33:37<7:12:45, 5.73s/it]
90%|████████▉ | 39412/43939 [63:33:43<7:12:50, 5.74s/it]
90%|████████▉ | 39413/43939 [63:33:49<7:11:40, 5.72s/it]
90%|████████▉ | 39414/43939 [63:33:55<7:11:46, 5.73s/it]
90%|████████▉ | 39415/43939 [63:34:00<7:10:46, 5.71s/it]
90%|████████▉ | 39416/43939 [63:34:06<7:10:31, 5.71s/it]
90%|████████▉ | 39417/43939 [63:34:12<7:10:33, 5.71s/it]
90%|████████▉ | 39418/43939 [63:34:17<7:09:11, 5.70s/it]
90%|████████▉ | 39419/43939 [63:34:23<7:09:50, 5.71s/it]
90%|████████▉ | 39420/43939 [63:34:29<7:09:52, 5.71s/it]
90%|████████▉ | 39421/43939 [63:34:34<7:09:36, 5.71s/it]
90%|████████▉ | 39422/43939 [63:34:40<7:09:02, 5.70s/it]
90%|████████▉ | 39423/43939 [63:34:46<7:08:45, 5.70s/it]
90%|████████▉ | 39424/43939 [63:34:52<7:09:42, 5.71s/it]
90%|████████▉ | 39425/43939 [63:34:57<7:09:47, 5.71s/it]
90%|████████▉ | 39426/43939 [63:35:03<7:08:41, 5.70s/it]
90%|████████▉ | 39427/43939 [63:35:09<7:09:09, 5.71s/it]
90%|████████▉ | 39428/43939 [63:35:14<7:09:19, 5.71s/it]
90%|████████▉ | 39429/43939 [63:35:20<7:09:30, 5.71s/it]
90%|████████▉ | 39430/43939 [63:35:26<7:14:29, 5.78s/it]
90%|████████▉ | 39431/43939 [63:35:32<7:11:46, 5.75s/it]
90%|████████▉ | 39432/43939 [63:35:37<7:11:01, 5.74s/it]
90%|████████▉ | 39433/43939 [63:35:43<7:11:23, 5.74s/it]
90%|████████▉ | 39434/43939 [63:35:49<7:10:48, 5.74s/it]
90%|████████▉ | 39435/43939 [63:35:55<7:09:02, 5.72s/it]
90%|████████▉ | 39436/43939 [63:36:00<7:08:13, 5.71s/it]
90%|████████▉ | 39437/43939 [63:36:06<7:08:26, 5.71s/it]
90%|████████▉ | 39438/43939 [63:36:12<7:08:09, 5.71s/it]
90%|████████▉ | 39439/43939 [63:36:17<7:08:41, 5.72s/it]
90%|████████▉ | 39440/43939 [63:36:23<7:09:51, 5.73s/it]
90%|████████▉ | 39441/43939 [63:36:29<7:08:15, 5.71s/it]
90%|████████▉ | 39442/43939 [63:36:35<7:07:32, 5.70s/it]
90%|████████▉ | 39443/43939 [63:36:40<7:07:56, 5.71s/it]
90%|████████▉ | 39444/43939 [63:36:46<7:06:35, 5.69s/it]
90%|████████▉ | 39445/43939 [63:36:52<7:07:14, 5.70s/it]
90%|████████▉ | 39446/43939 [63:36:57<7:07:28, 5.71s/it]
90%|████████▉ | 39447/43939 [63:37:03<7:06:36, 5.70s/it]
90%|████████▉ | 39448/43939 [63:37:09<7:06:51, 5.70s/it]
90%|████████▉ | 39449/43939 [63:37:14<7:06:54, 5.70s/it]
90%|████████▉ | 39450/43939 [63:37:20<7:06:27, 5.70s/it]
{'loss': 0.6571, 'grad_norm': 1.5960315465927124, 'learning_rate': 1.5735764263495655e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.71, 'epoch': 3.59}
90%|████████▉ | 39450/43939 [63:37:20<7:06:27, 5.70s/it]
90%|████████▉ | 39451/43939 [63:37:26<7:06:24, 5.70s/it]
90%|████████▉ | 39452/43939 [63:37:32<7:10:53, 5.76s/it]
90%|████████▉ | 39453/43939 [63:37:37<7:10:05, 5.75s/it]
90%|████████▉ | 39454/43939 [63:37:43<7:08:40, 5.73s/it]
90%|████████▉ | 39455/43939 [63:37:49<7:07:26, 5.72s/it]
90%|████████▉ | 39456/43939 [63:37:55<7:06:35, 5.71s/it]
90%|████████▉ | 39457/43939 [63:38:00<7:06:25, 5.71s/it]
90%|████████▉ | 39458/43939 [63:38:06<7:06:30, 5.71s/it]
90%|████████▉ | 39459/43939 [63:38:12<7:06:46, 5.72s/it]
90%|████████▉ | 39460/43939 [63:38:17<7:06:17, 5.71s/it]
90%|████████▉ | 39461/43939 [63:38:23<7:06:02, 5.71s/it]
90%|████████▉ | 39462/43939 [63:38:29<7:06:08, 5.71s/it]
90%|████████▉ | 39463/43939 [63:38:35<7:06:30, 5.72s/it]
90%|████████▉ | 39464/43939 [63:38:40<7:05:47, 5.71s/it]
90%|████████▉ | 39465/43939 [63:38:46<7:05:50, 5.71s/it]
90%|████████▉ | 39466/43939 [63:38:52<7:05:20, 5.71s/it]
90%|████████▉ | 39467/43939 [63:38:57<7:05:39, 5.71s/it]
90%|████████▉ | 39468/43939 [63:39:03<7:04:41, 5.70s/it]
90%|████████▉ | 39469/43939 [63:39:09<7:05:28, 5.71s/it]
90%|████████▉ | 39470/43939 [63:39:14<7:04:32, 5.70s/it]
90%|████████▉ | 39471/43939 [63:39:20<7:05:22, 5.71s/it]
90%|████████▉ | 39472/43939 [63:39:26<7:04:58, 5.71s/it]
90%|████████▉ | 39473/43939 [63:39:32<7:09:10, 5.77s/it]
90%|████████▉ | 39474/43939 [63:39:37<7:07:10, 5.74s/it]
90%|████████▉ | 39475/43939 [63:39:43<7:06:38, 5.73s/it]
90%|████████▉ | 39476/43939 [63:39:49<7:06:36, 5.74s/it]
90%|████████▉ | 39477/43939 [63:39:55<7:07:04, 5.74s/it]
90%|████████▉ | 39478/43939 [63:40:00<7:07:34, 5.75s/it]
90%|████████▉ | 39479/43939 [63:40:06<7:06:06, 5.73s/it]
90%|████████▉ | 39480/43939 [63:40:12<7:05:31, 5.73s/it]
90%|████████▉ | 39481/43939 [63:40:18<7:05:20, 5.72s/it]
90%|████████▉ | 39482/43939 [63:40:23<7:05:23, 5.73s/it]
90%|████████▉ | 39483/43939 [63:40:29<7:04:49, 5.72s/it]
90%|████████▉ | 39484/43939 [63:40:35<7:03:48, 5.71s/it]
90%|████████▉ | 39485/43939 [63:40:40<7:03:41, 5.71s/it]
90%|████████▉ | 39486/43939 [63:40:46<7:03:13, 5.70s/it]
90%|████████▉ | 39487/43939 [63:40:52<7:04:33, 5.72s/it]
90%|████████▉ | 39488/43939 [63:40:58<7:03:54, 5.71s/it]
90%|████████▉ | 39489/43939 [63:41:03<7:04:34, 5.72s/it]
90%|████████▉ | 39490/43939 [63:41:09<7:04:00, 5.72s/it]
90%|████████▉ | 39491/43939 [63:41:15<7:03:51, 5.72s/it]
90%|████████▉ | 39492/43939 [63:41:20<7:03:02, 5.71s/it]
90%|████████▉ | 39493/43939 [63:41:26<7:02:25, 5.70s/it]
90%|████████▉ | 39494/43939 [63:41:32<7:01:51, 5.69s/it]
90%|████████▉ | 39495/43939 [63:41:38<7:07:15, 5.77s/it]
90%|████████▉ | 39496/43939 [63:41:43<7:06:01, 5.75s/it]
90%|████████▉ | 39497/43939 [63:41:49<7:05:17, 5.74s/it]
90%|████████▉ | 39498/43939 [63:41:55<7:03:54, 5.73s/it]
90%|████████▉ | 39499/43939 [63:42:01<7:02:59, 5.72s/it]
90%|████████▉ | 39500/43939 [63:42:06<7:02:39, 5.71s/it]
{'loss': 0.6744, 'grad_norm': 1.590531826019287, 'learning_rate': 1.5390874416256694e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 515.42, 'epoch': 3.6}
90%|████████▉ | 39500/43939 [63:42:06<7:02:39, 5.71s/it]
90%|████████▉ | 39501/43939 [63:42:12<7:02:21, 5.71s/it]
90%|████████▉ | 39502/43939 [63:42:18<7:01:55, 5.71s/it]
90%|████████▉ | 39503/43939 [63:42:23<7:01:21, 5.70s/it]
90%|████████▉ | 39504/43939 [63:42:29<7:00:53, 5.69s/it]
90%|████████▉ | 39505/43939 [63:42:35<7:01:04, 5.70s/it]
90%|████████▉ | 39506/43939 [63:42:40<7:01:00, 5.70s/it]
90%|████████▉ | 39507/43939 [63:42:46<7:01:09, 5.70s/it]
90%|████████▉ | 39508/43939 [63:42:52<7:01:13, 5.70s/it]
90%|████████▉ | 39509/43939 [63:42:58<7:01:47, 5.71s/it]
90%|████████▉ | 39510/43939 [63:43:03<7:00:37, 5.70s/it]
90%|████████▉ | 39511/43939 [63:43:09<7:00:40, 5.70s/it]
90%|████████▉ | 39512/43939 [63:43:15<7:00:33, 5.70s/it]
90%|████████▉ | 39513/43939 [63:43:20<7:00:52, 5.71s/it]
90%|████████▉ | 39514/43939 [63:43:26<7:02:03, 5.72s/it]
90%|████████▉ | 39515/43939 [63:43:32<7:01:00, 5.71s/it]
90%|████████▉ | 39516/43939 [63:43:38<7:05:37, 5.77s/it]
90%|████████▉ | 39517/43939 [63:43:43<7:03:47, 5.75s/it]
90%|████████▉ | 39518/43939 [63:43:49<7:03:01, 5.74s/it]
90%|████████▉ | 39519/43939 [63:43:55<7:01:06, 5.72s/it]
90%|████████▉ | 39520/43939 [63:44:00<7:00:10, 5.70s/it]
90%|████████▉ | 39521/43939 [63:44:06<7:00:00, 5.70s/it]
90%|████████▉ | 39522/43939 [63:44:12<6:59:49, 5.70s/it]
90%|████████▉ | 39523/43939 [63:44:18<6:59:58, 5.71s/it]
90%|████████▉ | 39524/43939 [63:44:23<7:00:00, 5.71s/it]
90%|████████▉ | 39525/43939 [63:44:29<6:59:55, 5.71s/it]
90%|████████▉ | 39526/43939 [63:44:35<6:59:28, 5.70s/it]
90%|████████▉ | 39527/43939 [63:44:40<6:59:08, 5.70s/it]
90%|████████▉ | 39528/43939 [63:44:46<6:59:28, 5.71s/it]
90%|████████▉ | 39529/43939 [63:44:52<7:00:31, 5.72s/it]
90%|████████▉ | 39530/43939 [63:44:58<7:00:24, 5.72s/it]
90%|████████▉ | 39531/43939 [63:45:03<6:59:56, 5.72s/it]
90%|████████▉ | 39532/43939 [63:45:09<7:01:18, 5.74s/it]
90%|████████▉ | 39533/43939 [63:45:15<7:01:33, 5.74s/it]
90%|████████▉ | 39534/43939 [63:45:21<7:00:57, 5.73s/it]
90%|████████▉ | 39535/43939 [63:45:26<6:59:28, 5.71s/it]
90%|████████▉ | 39536/43939 [63:45:32<6:59:21, 5.71s/it]
90%|████████▉ | 39537/43939 [63:45:38<6:59:30, 5.72s/it]
90%|████████▉ | 39538/43939 [63:45:44<7:05:41, 5.80s/it]
90%|████████▉ | 39539/43939 [63:45:49<7:03:22, 5.77s/it]
90%|████████▉ | 39540/43939 [63:45:55<7:01:45, 5.75s/it]
90%|████████▉ | 39541/43939 [63:46:01<7:00:41, 5.74s/it]
90%|████████▉ | 39542/43939 [63:46:06<7:00:16, 5.74s/it]
90%|████████▉ | 39543/43939 [63:46:12<6:59:15, 5.72s/it]
90%|████████▉ | 39544/43939 [63:46:18<6:58:31, 5.71s/it]
90%|████████▉ | 39545/43939 [63:46:24<6:57:50, 5.71s/it]
90%|█████████ | 39546/43939 [63:46:29<6:57:37, 5.70s/it][2025-12-01 09:50:43,079] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:35, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.93s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.22s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:45, 2.67s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.69s/it][A
17%|█▋ | 12/72 [00:29<02:42, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.70s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.73s/it][A
29%|██▉ | 21/72 [00:54<02:19, 2.73s/it][A
31%|███ | 22/72 [00:57<02:16, 2.73s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.73s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.73s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.73s/it][A
39%|███▉ | 28/72 [01:13<01:59, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.73s/it][A
42%|████▏ | 30/72 [01:19<01:54, 2.73s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.73s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.73s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.73s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:30, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.73s/it][A
57%|█████▋ | 41/72 [01:49<01:24, 2.73s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.73s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.73s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.72s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.73s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.73s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.73s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.72s/it][A
72%|███████▏ | 52/72 [02:19<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.73s/it][A
75%|███████▌ | 54/72 [02:24<00:49, 2.73s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.73s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.73s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.73s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.73s/it][A
82%|████████▏ | 59/72 [02:38<00:35, 2.73s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.73s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:49<00:24, 2.73s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.73s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:08<00:05, 2.72s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A{'eval_loss': 1.015249490737915, 'eval_runtime': 196.7067, 'eval_samples_per_second': 5.079, 'eval_steps_per_second': 0.366, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 3.6}
90%|█████████ | 39546/43939 [63:49:46<6:57:37, 5.70s/it]
100%|██████████| 72/72 [03:12<00:00, 2.25s/it][A
[A
90%|█████████ | 39547/43939 [63:49:52<78:59:41, 64.75s/it]
90%|█████████ | 39548/43939 [63:49:57<57:21:27, 47.03s/it]
90%|█████████ | 39549/43939 [63:50:03<42:13:14, 34.62s/it]
90%|█████████ | 39550/43939 [63:50:09<31:38:15, 25.95s/it]
{'loss': 0.6493, 'grad_norm': 1.5634604692459106, 'learning_rate': 1.5049686079297715e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 32.91, 'epoch': 3.6}
90%|█████████ | 39550/43939 [63:50:09<31:38:15, 25.95s/it]
90%|█████████ | 39551/43939 [63:50:15<24:14:29, 19.89s/it]
90%|█████████ | 39552/43939 [63:50:20<19:02:21, 15.62s/it]
90%|█████████ | 39553/43939 [63:50:26<15:25:48, 12.66s/it]
90%|█████████ | 39554/43939 [63:50:32<12:52:24, 10.57s/it]
90%|█████████ | 39555/43939 [63:50:37<11:05:59, 9.11s/it]
90%|█████████ | 39556/43939 [63:50:43<9:51:29, 8.10s/it]
90%|█████████ | 39557/43939 [63:50:49<8:59:22, 7.39s/it]
90%|█████████ | 39558/43939 [63:50:55<8:29:08, 6.97s/it]
90%|█████████ | 39559/43939 [63:51:01<8:01:38, 6.60s/it]
90%|█████████ | 39560/43939 [63:51:06<7:41:54, 6.33s/it]
90%|█████████ | 39561/43939 [63:51:12<7:28:11, 6.14s/it]
90%|█████████ | 39562/43939 [63:51:18<7:18:24, 6.01s/it]
90%|█████████ | 39563/43939 [63:51:23<7:11:30, 5.92s/it]
90%|█████████ | 39564/43939 [63:51:29<7:06:43, 5.85s/it]
90%|█████████ | 39565/43939 [63:51:35<7:04:02, 5.82s/it]
90%|█████████ | 39566/43939 [63:51:41<7:01:03, 5.78s/it]
90%|█████████ | 39567/43939 [63:51:46<6:59:35, 5.76s/it]
90%|█████████ | 39568/43939 [63:51:52<6:59:34, 5.76s/it]
90%|█████████ | 39569/43939 [63:51:58<6:58:04, 5.74s/it]
90%|█████████ | 39570/43939 [63:52:03<6:57:27, 5.73s/it]
90%|█████████ | 39571/43939 [63:52:09<6:57:26, 5.73s/it]
90%|█████████ | 39572/43939 [63:52:15<6:56:37, 5.72s/it]
90%|█████████ | 39573/43939 [63:52:21<6:56:54, 5.73s/it]
90%|█████████ | 39574/43939 [63:52:26<6:56:15, 5.72s/it]
90%|█████████ | 39575/43939 [63:52:32<6:55:36, 5.71s/it]
90%|█████████ | 39576/43939 [63:52:38<6:55:42, 5.72s/it]
90%|█████████ | 39577/43939 [63:52:43<6:54:46, 5.71s/it]
90%|█████████ | 39578/43939 [63:52:49<6:54:02, 5.70s/it]
90%|█████████ | 39579/43939 [63:52:55<6:55:14, 5.71s/it]
90%|█████████ | 39580/43939 [63:53:01<7:00:34, 5.79s/it]
90%|█████████ | 39581/43939 [63:53:07<6:59:02, 5.77s/it]
90%|█████████ | 39582/43939 [63:53:12<6:58:05, 5.76s/it]
90%|█████████ | 39583/43939 [63:53:18<6:55:48, 5.73s/it]
90%|█████████ | 39584/43939 [63:53:24<6:55:36, 5.73s/it]
90%|█████████ | 39585/43939 [63:53:29<6:54:32, 5.71s/it]
90%|█████████ | 39586/43939 [63:53:35<6:53:46, 5.70s/it]
90%|█████████ | 39587/43939 [63:53:41<6:53:58, 5.71s/it]
90%|█████████ | 39588/43939 [63:53:46<6:53:28, 5.70s/it]
90%|█████████ | 39589/43939 [63:53:52<6:53:33, 5.70s/it]
90%|█████████ | 39590/43939 [63:53:58<6:53:03, 5.70s/it]
90%|█████████ | 39591/43939 [63:54:04<6:53:16, 5.70s/it]
90%|█████████ | 39592/43939 [63:54:09<6:53:33, 5.71s/it]
90%|█████████ | 39593/43939 [63:54:15<6:53:56, 5.71s/it]
90%|█████████ | 39594/43939 [63:54:21<6:53:34, 5.71s/it]
90%|█████████ | 39595/43939 [63:54:26<6:53:05, 5.71s/it]
90%|█████████ | 39596/43939 [63:54:32<6:52:22, 5.70s/it]
90%|█████████ | 39597/43939 [63:54:38<6:53:18, 5.71s/it]
90%|█████████ | 39598/43939 [63:54:44<6:53:27, 5.71s/it]
90%|█████████ | 39599/43939 [63:54:49<6:53:15, 5.71s/it]
90%|█████████ | 39600/43939 [63:54:55<6:53:47, 5.72s/it]
{'loss': 0.6562, 'grad_norm': 1.8618580102920532, 'learning_rate': 1.4712204635666498e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 488.16, 'epoch': 3.6}
90%|█████████ | 39600/43939 [63:54:55<6:53:47, 5.72s/it]
90%|█████████ | 39601/43939 [63:55:01<6:58:32, 5.79s/it]
90%|█████████ | 39602/43939 [63:55:07<6:56:42, 5.76s/it]
90%|█████████ | 39603/43939 [63:55:12<6:55:05, 5.74s/it]
90%|█████████ | 39604/43939 [63:55:18<6:54:01, 5.73s/it]
90%|█████████ | 39605/43939 [63:55:24<6:54:12, 5.73s/it]
90%|█████████ | 39606/43939 [63:55:29<6:52:57, 5.72s/it]
90%|█████████ | 39607/43939 [63:55:35<6:52:15, 5.71s/it]
90%|█████████ | 39608/43939 [63:55:41<6:51:56, 5.71s/it]
90%|█████████ | 39609/43939 [63:55:47<6:51:35, 5.70s/it]
90%|█████████ | 39610/43939 [63:55:52<6:50:56, 5.70s/it]
90%|█████████ | 39611/43939 [63:55:58<6:51:09, 5.70s/it]
90%|█████████ | 39612/43939 [63:56:04<6:51:19, 5.70s/it]
90%|█████████ | 39613/43939 [63:56:09<6:50:48, 5.70s/it]
90%|█████████ | 39614/43939 [63:56:15<6:50:40, 5.70s/it]
90%|█████████ | 39615/43939 [63:56:21<6:51:22, 5.71s/it]
90%|█████████ | 39616/43939 [63:56:26<6:52:00, 5.72s/it]
90%|█████████ | 39617/43939 [63:56:32<6:51:09, 5.71s/it]
90%|█████████ | 39618/43939 [63:56:38<6:51:31, 5.71s/it]
90%|█████████ | 39619/43939 [63:56:44<6:51:39, 5.72s/it]
90%|█████████ | 39620/43939 [63:56:49<6:53:12, 5.74s/it]
90%|█████████ | 39621/43939 [63:56:55<6:52:15, 5.73s/it]
90%|█████████ | 39622/43939 [63:57:01<6:56:21, 5.79s/it]
90%|█████████ | 39623/43939 [63:57:07<6:54:02, 5.76s/it]
90%|█████████ | 39624/43939 [63:57:12<6:52:08, 5.73s/it]
90%|█████████ | 39625/43939 [63:57:18<6:51:14, 5.72s/it]
90%|█████████ | 39626/43939 [63:57:24<6:51:30, 5.72s/it]
90%|█████████ | 39627/43939 [63:57:30<6:51:06, 5.72s/it]
90%|█████████ | 39628/43939 [63:57:35<6:50:49, 5.72s/it]
90%|█████████ | 39629/43939 [63:57:41<6:50:15, 5.71s/it]
90%|█████████ | 39630/43939 [63:57:47<6:50:47, 5.72s/it]
90%|█████████ | 39631/43939 [63:57:52<6:50:27, 5.72s/it]
90%|█████████ | 39632/43939 [63:57:58<6:50:41, 5.72s/it]
90%|█████████ | 39633/43939 [63:58:04<6:50:43, 5.72s/it]
90%|█████████ | 39634/43939 [63:58:10<6:49:35, 5.71s/it]
90%|█████████ | 39635/43939 [63:58:15<6:50:13, 5.72s/it]
90%|█████████ | 39636/43939 [63:58:21<6:50:34, 5.72s/it]
90%|█████████ | 39637/43939 [63:58:27<6:50:11, 5.72s/it]
90%|█████████ | 39638/43939 [63:58:32<6:50:21, 5.72s/it]
90%|█████████ | 39639/43939 [63:58:38<6:49:39, 5.72s/it]
90%|█████████ | 39640/43939 [63:58:44<6:49:11, 5.71s/it]
90%|█████████ | 39641/43939 [63:58:50<6:48:44, 5.71s/it]
90%|█████████ | 39642/43939 [63:58:55<6:48:48, 5.71s/it]
90%|█████████ | 39643/43939 [63:59:01<6:48:11, 5.70s/it]
90%|█████████ | 39644/43939 [63:59:07<6:54:15, 5.79s/it]
90%|█████████ | 39645/43939 [63:59:13<6:52:27, 5.76s/it]
90%|█████████ | 39646/43939 [63:59:18<6:51:07, 5.75s/it]
90%|█████████ | 39647/43939 [63:59:24<6:49:52, 5.73s/it]
90%|█████████ | 39648/43939 [63:59:30<6:49:19, 5.72s/it]
90%|█████████ | 39649/43939 [63:59:35<6:48:24, 5.71s/it]
90%|█████████ | 39650/43939 [63:59:41<6:49:17, 5.73s/it]
{'loss': 0.6469, 'grad_norm': 1.4708662033081055, 'learning_rate': 1.4378435409925867e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.84, 'epoch': 3.61}
90%|█████████ | 39650/43939 [63:59:41<6:49:17, 5.73s/it]
90%|█████████ | 39651/43939 [63:59:47<6:48:54, 5.72s/it]
90%|█████████ | 39652/43939 [63:59:53<6:48:27, 5.72s/it]
90%|█████████ | 39653/43939 [63:59:58<6:49:38, 5.73s/it]
90%|█████████ | 39654/43939 [64:00:04<6:48:55, 5.73s/it]
90%|█████████ | 39655/43939 [64:00:10<6:50:02, 5.74s/it]
90%|█████████ | 39656/43939 [64:00:16<6:49:54, 5.74s/it]
90%|█████████ | 39657/43939 [64:00:21<6:49:48, 5.74s/it]
90%|█████████ | 39658/43939 [64:00:27<6:48:47, 5.73s/it]
90%|█████████ | 39659/43939 [64:00:33<6:48:13, 5.72s/it]
90%|█████████ | 39660/43939 [64:00:38<6:47:56, 5.72s/it]
90%|█████████ | 39661/43939 [64:00:44<6:47:59, 5.72s/it]
90%|█████████ | 39662/43939 [64:00:50<6:46:31, 5.70s/it]
90%|█████████ | 39663/43939 [64:00:56<6:47:12, 5.71s/it]
90%|█████████ | 39664/43939 [64:01:01<6:48:01, 5.73s/it]
90%|█████████ | 39665/43939 [64:01:07<6:53:34, 5.81s/it]
90%|█████████ | 39666/43939 [64:01:13<6:51:35, 5.78s/it]
90%|█████████ | 39667/43939 [64:01:19<6:50:44, 5.77s/it]
90%|█████████ | 39668/43939 [64:01:25<6:49:47, 5.76s/it]
90%|█████████ | 39669/43939 [64:01:30<6:48:53, 5.75s/it]
90%|█████████ | 39670/43939 [64:01:36<6:47:35, 5.73s/it]
90%|█████████ | 39671/43939 [64:01:42<6:47:20, 5.73s/it]
90%|█████████ | 39672/43939 [64:01:47<6:46:31, 5.72s/it]
90%|█████████ | 39673/43939 [64:01:53<6:47:25, 5.73s/it]
90%|█████████ | 39674/43939 [64:01:59<6:46:49, 5.72s/it]
90%|█████████ | 39675/43939 [64:02:05<6:47:12, 5.73s/it]
90%|█████████ | 39676/43939 [64:02:10<6:46:50, 5.73s/it]
90%|█████████ | 39677/43939 [64:02:16<6:46:15, 5.72s/it]
90%|█████████ | 39678/43939 [64:02:22<6:45:43, 5.71s/it]
90%|█████████ | 39679/43939 [64:02:27<6:45:10, 5.71s/it]
90%|█████████ | 39680/43939 [64:02:33<6:44:20, 5.70s/it]
90%|█████████ | 39681/43939 [64:02:39<6:44:31, 5.70s/it]
90%|█████████ | 39682/43939 [64:02:44<6:44:43, 5.70s/it]
90%|█████████ | 39683/43939 [64:02:50<6:44:07, 5.70s/it]
90%|█████████ | 39684/43939 [64:02:56<6:45:33, 5.72s/it]
90%|█████████ | 39685/43939 [64:03:02<6:45:00, 5.71s/it]
90%|█████████ | 39686/43939 [64:03:08<6:50:18, 5.79s/it]
90%|█████████ | 39687/43939 [64:03:13<6:48:23, 5.76s/it]
90%|█████████ | 39688/43939 [64:03:19<6:47:59, 5.76s/it]
90%|█████████ | 39689/43939 [64:03:25<6:47:13, 5.75s/it]
90%|█████████ | 39690/43939 [64:03:30<6:45:51, 5.73s/it]
90%|█████████ | 39691/43939 [64:03:36<6:44:23, 5.71s/it]
90%|█████████ | 39692/43939 [64:03:42<6:44:13, 5.71s/it]
90%|█████████ | 39693/43939 [64:03:48<6:44:06, 5.71s/it]
90%|█████████ | 39694/43939 [64:03:53<6:43:58, 5.71s/it]
90%|█████████ | 39695/43939 [64:03:59<6:43:26, 5.70s/it]
90%|█████████ | 39696/43939 [64:04:05<6:43:41, 5.71s/it]
90%|█████████ | 39697/43939 [64:04:10<6:43:22, 5.71s/it]
90%|█████████ | 39698/43939 [64:04:16<6:43:33, 5.71s/it]
90%|█████████ | 39699/43939 [64:04:22<6:43:05, 5.70s/it]
90%|█████████ | 39700/43939 [64:04:27<6:42:14, 5.69s/it]
{'loss': 0.6733, 'grad_norm': 1.8748801946640015, 'learning_rate': 1.404838366806971e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 519.82, 'epoch': 3.61}
90%|█████████ | 39700/43939 [64:04:27<6:42:14, 5.69s/it]
90%|█████████ | 39701/43939 [64:04:33<6:42:34, 5.70s/it]
90%|█████████ | 39702/43939 [64:04:39<6:42:39, 5.70s/it]
90%|█████████ | 39703/43939 [64:04:45<6:42:40, 5.70s/it]
90%|█████████ | 39704/43939 [64:04:50<6:42:25, 5.70s/it]
90%|█████████ | 39705/43939 [64:04:56<6:43:32, 5.72s/it]
90%|█████████ | 39706/43939 [64:05:02<6:43:01, 5.71s/it]
90%|█████████ | 39707/43939 [64:05:08<6:48:11, 5.79s/it]
90%|█████████ | 39708/43939 [64:05:13<6:47:12, 5.77s/it]
90%|█████████ | 39709/43939 [64:05:19<6:45:38, 5.75s/it]
90%|█████████ | 39710/43939 [64:05:25<6:44:34, 5.74s/it]
90%|█████████ | 39711/43939 [64:05:31<6:45:01, 5.75s/it]
90%|█████████ | 39712/43939 [64:05:36<6:44:02, 5.74s/it]
90%|█████████ | 39713/43939 [64:05:42<6:43:53, 5.73s/it]
90%|█████████ | 39714/43939 [64:05:48<6:42:04, 5.71s/it]
90%|█████████ | 39715/43939 [64:05:53<6:41:37, 5.70s/it]
90%|█████████ | 39716/43939 [64:05:59<6:43:56, 5.74s/it]
90%|█████████ | 39717/43939 [64:06:05<6:44:35, 5.75s/it]
90%|█████████ | 39718/43939 [64:06:11<6:43:36, 5.74s/it]
90%|█████████ | 39719/43939 [64:06:16<6:42:45, 5.73s/it]
90%|█████████ | 39720/43939 [64:06:22<6:42:24, 5.72s/it]
90%|█████████ | 39721/43939 [64:06:28<6:41:58, 5.72s/it]
90%|█████████ | 39722/43939 [64:06:34<6:42:26, 5.73s/it]
90%|█████████ | 39723/43939 [64:06:39<6:43:49, 5.75s/it]
90%|█████████ | 39724/43939 [64:06:45<6:41:57, 5.72s/it]
90%|█████████ | 39725/43939 [64:06:51<6:40:59, 5.71s/it]
90%|█████████ | 39726/43939 [64:06:56<6:40:54, 5.71s/it]
90%|█████████ | 39727/43939 [64:07:02<6:40:25, 5.70s/it]
90%|█████████ | 39728/43939 [64:07:08<6:40:21, 5.70s/it]
90%|█████████ | 39729/43939 [64:07:14<6:45:14, 5.78s/it]
90%|█████████ | 39730/43939 [64:07:19<6:43:34, 5.75s/it]
90%|█████████ | 39731/43939 [64:07:25<6:41:39, 5.73s/it]
90%|█████████ | 39732/43939 [64:07:31<6:41:00, 5.72s/it]
90%|█████████ | 39733/43939 [64:07:37<6:40:36, 5.71s/it]
90%|█████████ | 39734/43939 [64:07:42<6:40:55, 5.72s/it]
90%|█████████ | 39735/43939 [64:07:48<6:40:37, 5.72s/it]
90%|█████████ | 39736/43939 [64:07:54<6:40:04, 5.71s/it]
90%|█████████ | 39737/43939 [64:07:59<6:41:18, 5.73s/it]
90%|█████████ | 39738/43939 [64:08:05<6:40:29, 5.72s/it]
90%|█████████ | 39739/43939 [64:08:11<6:39:20, 5.70s/it]
90%|█████████ | 39740/43939 [64:08:16<6:38:47, 5.70s/it]
90%|█████████ | 39741/43939 [64:08:22<6:38:14, 5.69s/it]
90%|█████████ | 39742/43939 [64:08:28<6:39:28, 5.71s/it]
90%|█████████ | 39743/43939 [64:08:34<6:39:40, 5.72s/it]
90%|█████████ | 39744/43939 [64:08:39<6:39:29, 5.71s/it]
90%|█████████ | 39745/43939 [64:08:45<6:39:50, 5.72s/it]
90%|█████████ | 39746/43939 [64:08:51<6:39:20, 5.71s/it]
90%|█████████ | 39747/43939 [64:08:56<6:39:13, 5.71s/it]
90%|█████████ | 39748/43939 [64:09:02<6:39:13, 5.72s/it]
90%|█████████ | 39749/43939 [64:09:08<6:38:20, 5.70s/it]
90%|█████████ | 39750/43939 [64:09:14<6:38:15, 5.70s/it]
{'loss': 0.6537, 'grad_norm': 2.1891469955444336, 'learning_rate': 1.3722054617439916e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.04, 'epoch': 3.62}
90%|█████████ | 39750/43939 [64:09:14<6:38:15, 5.70s/it]
90%|█████████ | 39751/43939 [64:09:20<6:42:54, 5.77s/it]
90%|█████████ | 39752/43939 [64:09:25<6:41:01, 5.75s/it]
90%|█████████ | 39753/43939 [64:09:31<6:40:22, 5.74s/it]
90%|█████████ | 39754/43939 [64:09:37<6:40:07, 5.74s/it]
90%|█████████ | 39755/43939 [64:09:42<6:39:35, 5.73s/it]
90%|█████████ | 39756/43939 [64:09:48<6:38:29, 5.72s/it]
90%|█████████ | 39757/43939 [64:09:54<6:38:24, 5.72s/it]
90%|█████████ | 39758/43939 [64:10:00<6:38:59, 5.73s/it]
90%|█████████ | 39759/43939 [64:10:05<6:38:12, 5.72s/it]
90%|█████████ | 39760/43939 [64:10:11<6:37:58, 5.71s/it]
90%|█████████ | 39761/43939 [64:10:17<6:36:59, 5.70s/it]
90%|█████████ | 39762/43939 [64:10:22<6:37:50, 5.71s/it]
90%|█████████ | 39763/43939 [64:10:28<6:38:17, 5.72s/it]
90%|█████████ | 39764/43939 [64:10:34<6:37:23, 5.71s/it]
91%|█████████ | 39765/43939 [64:10:39<6:36:56, 5.71s/it]
91%|█████████ | 39766/43939 [64:10:45<6:37:07, 5.71s/it]
91%|█████████ | 39767/43939 [64:10:51<6:37:02, 5.71s/it]
91%|█████████ | 39768/43939 [64:10:57<6:37:53, 5.72s/it]
91%|█████████ | 39769/43939 [64:11:02<6:37:49, 5.72s/it]
91%|█████████ | 39770/43939 [64:11:08<6:36:19, 5.70s/it]
91%|█████████ | 39771/43939 [64:11:14<6:36:16, 5.70s/it]
91%|█████████ | 39772/43939 [64:11:20<6:40:20, 5.76s/it]
91%|█████████ | 39773/43939 [64:11:25<6:37:38, 5.73s/it]
91%|█████████ | 39774/43939 [64:11:31<6:37:47, 5.73s/it]
91%|█████████ | 39775/43939 [64:11:37<6:37:07, 5.72s/it]
91%|█████████ | 39776/43939 [64:11:42<6:37:22, 5.73s/it]
91%|█████████ | 39777/43939 [64:11:48<6:36:21, 5.71s/it]
91%|█████████ | 39778/43939 [64:11:54<6:36:09, 5.71s/it]
91%|█████████ | 39779/43939 [64:12:00<6:36:18, 5.72s/it]
91%|█████████ | 39780/43939 [64:12:05<6:35:46, 5.71s/it]
91%|█████████ | 39781/43939 [64:12:11<6:35:19, 5.70s/it]
91%|█████████ | 39782/43939 [64:12:17<6:34:23, 5.69s/it]
91%|█████████ | 39783/43939 [64:12:22<6:34:50, 5.70s/it]
91%|█████████ | 39784/43939 [64:12:28<6:34:42, 5.70s/it]
91%|█████████ | 39785/43939 [64:12:34<6:35:45, 5.72s/it]
91%|█████████ | 39786/43939 [64:12:40<6:36:00, 5.72s/it]
91%|█████████ | 39787/43939 [64:12:45<6:35:03, 5.71s/it]
91%|█████████ | 39788/43939 [64:12:51<6:34:18, 5.70s/it]
91%|█████████ | 39789/43939 [64:12:57<6:34:15, 5.70s/it]
91%|█████████ | 39790/43939 [64:13:02<6:35:27, 5.72s/it]
91%|█████████ | 39791/43939 [64:13:08<6:35:19, 5.72s/it]
91%|█████████ | 39792/43939 [64:13:14<6:34:28, 5.71s/it]
91%|█████████ | 39793/43939 [64:13:19<6:34:53, 5.71s/it]
91%|█████████ | 39794/43939 [64:13:25<6:38:45, 5.77s/it]
91%|█████████ | 39795/43939 [64:13:31<6:37:20, 5.75s/it]
91%|█████████ | 39796/43939 [64:13:37<6:36:15, 5.74s/it]
91%|█████████ | 39797/43939 [64:13:43<6:35:39, 5.73s/it]
91%|█████████ | 39798/43939 [64:13:48<6:34:10, 5.71s/it]
91%|█████████ | 39799/43939 [64:13:54<6:34:22, 5.72s/it]
91%|█████████ | 39800/43939 [64:14:00<6:33:18, 5.70s/it]
{'loss': 0.6531, 'grad_norm': 1.809812307357788, 'learning_rate': 1.3399453406644059e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 506.87, 'epoch': 3.62}
91%|█████████ | 39800/43939 [64:14:00<6:33:18, 5.70s/it]
91%|█████████ | 39801/43939 [64:14:05<6:33:07, 5.70s/it]
91%|█████████ | 39802/43939 [64:14:11<6:33:22, 5.71s/it]
91%|█████████ | 39803/43939 [64:14:17<6:32:22, 5.69s/it]
91%|█████████ | 39804/43939 [64:14:22<6:33:14, 5.71s/it]
91%|█████████ | 39805/43939 [64:14:28<6:32:59, 5.70s/it]
91%|█████████ | 39806/43939 [64:14:34<6:33:54, 5.72s/it]
91%|█████████ | 39807/43939 [64:14:40<6:34:02, 5.72s/it]
91%|█████████ | 39808/43939 [64:14:45<6:34:21, 5.73s/it]
91%|█████████ | 39809/43939 [64:14:51<6:33:38, 5.72s/it]
91%|█████████ | 39810/43939 [64:14:57<6:34:03, 5.73s/it]
91%|█████████ | 39811/43939 [64:15:02<6:33:01, 5.71s/it]
91%|█████████ | 39812/43939 [64:15:08<6:32:15, 5.70s/it]
91%|█████████ | 39813/43939 [64:15:14<6:32:31, 5.71s/it]
91%|█████████ | 39814/43939 [64:15:20<6:33:06, 5.72s/it]
91%|█████████ | 39815/43939 [64:15:26<6:37:31, 5.78s/it]
91%|█████████ | 39816/43939 [64:15:31<6:36:04, 5.76s/it]
91%|█████████ | 39817/43939 [64:15:37<6:34:26, 5.74s/it]
91%|█████████ | 39818/43939 [64:15:43<6:34:02, 5.74s/it]
91%|█████████ | 39819/43939 [64:15:48<6:32:43, 5.72s/it]
91%|█████████ | 39820/43939 [64:15:54<6:31:52, 5.71s/it]
91%|█████████ | 39821/43939 [64:16:00<6:31:31, 5.70s/it]
91%|█████████ | 39822/43939 [64:16:05<6:31:07, 5.70s/it]
91%|█████████ | 39823/43939 [64:16:11<6:30:33, 5.69s/it]
91%|█████████ | 39824/43939 [64:16:17<6:30:18, 5.69s/it]
91%|█████████ | 39825/43939 [64:16:22<6:29:29, 5.68s/it]
91%|█████████ | 39826/43939 [64:16:28<6:29:49, 5.69s/it]
91%|█████████ | 39827/43939 [64:16:34<6:30:32, 5.70s/it]
91%|█████████ | 39828/43939 [64:16:40<6:31:00, 5.71s/it]
91%|█████████ | 39829/43939 [64:16:45<6:31:29, 5.72s/it]
91%|█████████ | 39830/43939 [64:16:51<6:31:37, 5.72s/it]
91%|█████████ | 39831/43939 [64:16:57<6:30:53, 5.71s/it]
91%|█████████ | 39832/43939 [64:17:03<6:33:12, 5.74s/it]
91%|█████████ | 39833/43939 [64:17:08<6:32:36, 5.74s/it]
91%|█████████ | 39834/43939 [64:17:14<6:32:22, 5.74s/it]
91%|█████████ | 39835/43939 [64:17:20<6:31:50, 5.73s/it]
91%|█████████ | 39836/43939 [64:17:25<6:30:32, 5.71s/it]
91%|█████████ | 39837/43939 [64:17:31<6:34:36, 5.77s/it]
91%|█████████ | 39838/43939 [64:17:37<6:32:20, 5.74s/it]
91%|█████████ | 39839/43939 [64:17:43<6:31:30, 5.73s/it]
91%|█████████ | 39840/43939 [64:17:48<6:31:20, 5.73s/it]
91%|█████████ | 39841/43939 [64:17:54<6:30:56, 5.72s/it]
91%|█████████ | 39842/43939 [64:18:00<6:30:01, 5.71s/it]
91%|█████████ | 39843/43939 [64:18:05<6:29:49, 5.71s/it]
91%|█████████ | 39844/43939 [64:18:11<6:29:31, 5.71s/it]
91%|█████████ | 39845/43939 [64:18:17<6:28:59, 5.70s/it]
91%|█████████ | 39846/43939 [64:18:23<6:28:46, 5.70s/it]
91%|█████████ | 39847/43939 [64:18:28<6:28:32, 5.70s/it]
91%|█████████ | 39848/43939 [64:18:34<6:28:25, 5.70s/it]
91%|█████████ | 39849/43939 [64:18:40<6:28:19, 5.70s/it]
91%|█████████ | 39850/43939 [64:18:45<6:28:33, 5.70s/it]
{'loss': 0.6608, 'grad_norm': 1.682203769683838, 'learning_rate': 1.308058512547436e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 502.02, 'epoch': 3.63}
91%|█████████ | 39850/43939 [64:18:45<6:28:33, 5.70s/it]
91%|█████████ | 39851/43939 [64:18:51<6:27:41, 5.69s/it]
91%|█████████ | 39852/43939 [64:18:57<6:27:22, 5.69s/it]
91%|█████████ | 39853/43939 [64:19:02<6:27:23, 5.69s/it]
91%|█████████ | 39854/43939 [64:19:08<6:26:47, 5.68s/it]
91%|█████████ | 39855/43939 [64:19:14<6:26:23, 5.68s/it]
91%|█████████ | 39856/43939 [64:19:19<6:26:46, 5.68s/it]
91%|█████████ | 39857/43939 [64:19:25<6:26:43, 5.68s/it]
91%|█████████ | 39858/43939 [64:19:31<6:26:28, 5.68s/it]
91%|█████████ | 39859/43939 [64:19:37<6:32:19, 5.77s/it]
91%|█████████ | 39860/43939 [64:19:42<6:30:52, 5.75s/it]
91%|█████████ | 39861/43939 [64:19:48<6:29:25, 5.73s/it]
91%|█████████ | 39862/43939 [64:19:54<6:28:59, 5.72s/it]
91%|█████████ | 39863/43939 [64:20:00<6:28:17, 5.72s/it]
91%|█████████ | 39864/43939 [64:20:05<6:27:01, 5.70s/it]
91%|█████████ | 39865/43939 [64:20:11<6:26:47, 5.70s/it]
91%|█████████ | 39866/43939 [64:20:17<6:26:53, 5.70s/it]
91%|█████████ | 39867/43939 [64:20:22<6:27:14, 5.71s/it]
91%|█████████ | 39868/43939 [64:20:28<6:26:22, 5.69s/it]
91%|█████████ | 39869/43939 [64:20:34<6:26:18, 5.70s/it]
91%|█████████ | 39870/43939 [64:20:39<6:25:27, 5.68s/it]
91%|█████████ | 39871/43939 [64:20:45<6:25:32, 5.69s/it]
91%|█████████ | 39872/43939 [64:20:51<6:24:56, 5.68s/it]
91%|█████████ | 39873/43939 [64:20:56<6:25:07, 5.68s/it]
91%|█████████ | 39874/43939 [64:21:02<6:25:53, 5.70s/it]
91%|█████████ | 39875/43939 [64:21:08<6:26:14, 5.70s/it]
91%|█████████ | 39876/43939 [64:21:14<6:26:10, 5.70s/it]
91%|█████████ | 39877/43939 [64:21:19<6:25:29, 5.69s/it]
91%|█████████ | 39878/43939 [64:21:25<6:24:42, 5.68s/it]
91%|█████████ | 39879/43939 [64:21:31<6:25:21, 5.69s/it]
91%|█████████ | 39880/43939 [64:21:37<6:29:27, 5.76s/it]
91%|█████████ | 39881/43939 [64:21:42<6:27:50, 5.73s/it]
91%|█████████ | 39882/43939 [64:21:48<6:26:09, 5.71s/it]
91%|█████████ | 39883/43939 [64:21:54<6:26:28, 5.72s/it]
91%|█████████ | 39884/43939 [64:21:59<6:26:01, 5.71s/it]
91%|█████████ | 39885/43939 [64:22:05<6:25:03, 5.70s/it]
91%|█████████ | 39886/43939 [64:22:11<6:25:24, 5.71s/it]
91%|█████████ | 39887/43939 [64:22:16<6:25:41, 5.71s/it]
91%|█████████ | 39888/43939 [64:22:22<6:26:12, 5.72s/it]
91%|█████████ | 39889/43939 [64:22:28<6:28:10, 5.75s/it]
91%|█████████ | 39890/43939 [64:22:34<6:27:38, 5.74s/it]
91%|█████████ | 39891/43939 [64:22:39<6:25:51, 5.72s/it]
91%|█████████ | 39892/43939 [64:22:45<6:24:59, 5.71s/it]
91%|█████████ | 39893/43939 [64:22:51<6:24:20, 5.70s/it]
91%|█████████ | 39894/43939 [64:22:56<6:24:12, 5.70s/it]
91%|█████████ | 39895/43939 [64:23:02<6:23:57, 5.70s/it]
91%|█████████ | 39896/43939 [64:23:08<6:23:19, 5.69s/it]
91%|█████████ | 39897/43939 [64:23:13<6:22:35, 5.68s/it]
91%|█████████ | 39898/43939 [64:23:19<6:23:06, 5.69s/it]
91%|█████████ | 39899/43939 [64:23:25<6:23:07, 5.69s/it]
91%|█████████ | 39900/43939 [64:23:31<6:24:05, 5.71s/it]
{'loss': 0.6698, 'grad_norm': 1.541329026222229, 'learning_rate': 1.276545480482727e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.38, 'epoch': 3.63}
91%|█████████ | 39900/43939 [64:23:31<6:24:05, 5.71s/it]
91%|█████████ | 39901/43939 [64:23:36<6:28:31, 5.77s/it]
91%|█████████ | 39902/43939 [64:23:42<6:27:08, 5.75s/it]
91%|█████████ | 39903/43939 [64:23:48<6:26:02, 5.74s/it]
91%|█████████ | 39904/43939 [64:23:54<6:24:45, 5.72s/it]
91%|█████████ | 39905/43939 [64:23:59<6:23:38, 5.71s/it]
91%|█████████ | 39906/43939 [64:24:05<6:23:21, 5.70s/it]
91%|█████████ | 39907/43939 [64:24:11<6:23:14, 5.70s/it]
91%|█████████ | 39908/43939 [64:24:16<6:22:34, 5.69s/it]
91%|█████████ | 39909/43939 [64:24:22<6:22:06, 5.69s/it]
91%|█████████ | 39910/43939 [64:24:28<6:22:51, 5.70s/it]
91%|█████████ | 39911/43939 [64:24:33<6:23:06, 5.71s/it]
91%|█████████ | 39912/43939 [64:24:39<6:23:00, 5.71s/it]
91%|█████████ | 39913/43939 [64:24:45<6:22:35, 5.70s/it]
91%|█████████ | 39914/43939 [64:24:51<6:22:12, 5.70s/it]
91%|█████████ | 39915/43939 [64:24:56<6:22:51, 5.71s/it]
91%|█████████ | 39916/43939 [64:25:02<6:22:51, 5.71s/it]
91%|█████████ | 39917/43939 [64:25:08<6:23:05, 5.71s/it]
91%|█████████ | 39918/43939 [64:25:13<6:22:09, 5.70s/it]
91%|█████████ | 39919/43939 [64:25:19<6:22:03, 5.70s/it]
91%|█████████ | 39920/43939 [64:25:25<6:22:08, 5.71s/it]
91%|█████████ | 39921/43939 [64:25:31<6:22:04, 5.71s/it]
91%|█████████ | 39922/43939 [64:25:36<6:21:39, 5.70s/it]
91%|█████████ | 39923/43939 [64:25:42<6:25:06, 5.75s/it]
91%|█████████ | 39924/43939 [64:25:48<6:23:16, 5.73s/it]
91%|█████████ | 39925/43939 [64:25:53<6:22:46, 5.72s/it]
91%|█████████ | 39926/43939 [64:25:59<6:24:04, 5.74s/it]
91%|█████████ | 39927/43939 [64:26:05<6:22:49, 5.73s/it]
91%|█████████ | 39928/43939 [64:26:11<6:22:14, 5.72s/it]
91%|█████████ | 39929/43939 [64:26:16<6:21:29, 5.71s/it]
91%|█████████ | 39930/43939 [64:26:22<6:20:57, 5.70s/it]
91%|█████████ | 39931/43939 [64:26:28<6:21:31, 5.71s/it]
91%|█████████ | 39932/43939 [64:26:33<6:21:42, 5.72s/it]
91%|█████████ | 39933/43939 [64:26:39<6:21:15, 5.71s/it]
91%|█████████ | 39934/43939 [64:26:45<6:20:49, 5.71s/it]
91%|█████████ | 39935/43939 [64:26:51<6:20:56, 5.71s/it]
91%|█████████ | 39936/43939 [64:26:56<6:20:42, 5.71s/it]
91%|█████████ | 39937/43939 [64:27:02<6:20:10, 5.70s/it]
91%|█████████ | 39938/43939 [64:27:08<6:20:41, 5.71s/it]
91%|█████████ | 39939/43939 [64:27:13<6:20:20, 5.71s/it]
91%|█████████ | 39940/43939 [64:27:19<6:20:00, 5.70s/it]
91%|█████████ | 39941/43939 [64:27:25<6:20:36, 5.71s/it]
91%|█████████ | 39942/43939 [64:27:31<6:20:12, 5.71s/it]
91%|█████████ | 39943/43939 [64:27:36<6:20:11, 5.71s/it]
91%|█████████ | 39944/43939 [64:27:42<6:24:20, 5.77s/it]
91%|█████████ | 39945/43939 [64:27:48<6:22:33, 5.75s/it]
91%|█████████ | 39946/43939 [64:27:54<6:21:27, 5.73s/it]
91%|█████████ | 39947/43939 [64:27:59<6:21:42, 5.74s/it]
91%|█████████ | 39948/43939 [64:28:05<6:22:10, 5.75s/it]
91%|█████████ | 39949/43939 [64:28:11<6:21:33, 5.74s/it]
91%|█████████ | 39950/43939 [64:28:16<6:19:55, 5.71s/it]
{'loss': 0.6437, 'grad_norm': 1.6913994550704956, 'learning_rate': 1.2454067416624193e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.39, 'epoch': 3.64}
91%|█████████ | 39950/43939 [64:28:16<6:19:55, 5.71s/it]
91%|█████████ | 39951/43939 [64:28:22<6:19:39, 5.71s/it]
91%|█████████ | 39952/43939 [64:28:28<6:19:37, 5.71s/it]
91%|█████████ | 39953/43939 [64:28:34<6:19:46, 5.72s/it]
91%|█████████ | 39954/43939 [64:28:39<6:19:22, 5.71s/it]
91%|█████████ | 39955/43939 [64:28:45<6:19:00, 5.71s/it]
91%|█████████ | 39956/43939 [64:28:51<6:20:01, 5.72s/it]
91%|█████████ | 39957/43939 [64:28:56<6:19:59, 5.73s/it]
91%|█████████ | 39958/43939 [64:29:02<6:19:14, 5.72s/it]
91%|█████████ | 39959/43939 [64:29:08<6:18:33, 5.71s/it]
91%|█████████ | 39960/43939 [64:29:14<6:17:46, 5.70s/it]
91%|█████████ | 39961/43939 [64:29:19<6:17:09, 5.69s/it]
91%|█████████ | 39962/43939 [64:29:25<6:17:17, 5.69s/it]
91%|█████████ | 39963/43939 [64:29:31<6:17:04, 5.69s/it]
91%|█████████ | 39964/43939 [64:29:36<6:17:28, 5.70s/it]
91%|█████████ | 39965/43939 [64:29:42<6:16:59, 5.69s/it]
91%|█████████ | 39966/43939 [64:29:48<6:21:13, 5.76s/it]
91%|█████████ | 39967/43939 [64:29:54<6:19:19, 5.73s/it]
91%|█████████ | 39968/43939 [64:29:59<6:18:29, 5.72s/it]
91%|█████████ | 39969/43939 [64:30:05<6:19:02, 5.73s/it]
91%|█████████ | 39970/43939 [64:30:11<6:18:08, 5.72s/it]
91%|█████████ | 39971/43939 [64:30:16<6:18:01, 5.72s/it]
91%|█████████ | 39972/43939 [64:30:22<6:18:27, 5.72s/it]
91%|█████████ | 39973/43939 [64:30:28<6:18:29, 5.73s/it]
91%|█████████ | 39974/43939 [64:30:34<6:17:30, 5.71s/it]
91%|█████████ | 39975/43939 [64:30:39<6:17:21, 5.71s/it]
91%|█████████ | 39976/43939 [64:30:45<6:17:22, 5.71s/it]
91%|█████████ | 39977/43939 [64:30:51<6:17:05, 5.71s/it]
91%|█████████ | 39978/43939 [64:30:56<6:16:35, 5.70s/it]
91%|█████████ | 39979/43939 [64:31:02<6:17:14, 5.72s/it]
91%|█████████ | 39980/43939 [64:31:08<6:17:07, 5.72s/it]
91%|█████████ | 39981/43939 [64:31:14<6:16:48, 5.71s/it]
91%|█████████ | 39982/43939 [64:31:19<6:16:29, 5.71s/it]
91%|█████████ | 39983/43939 [64:31:25<6:16:57, 5.72s/it]
91%|█████████ | 39984/43939 [64:31:31<6:16:56, 5.72s/it]
91%|█████████ | 39985/43939 [64:31:36<6:17:07, 5.72s/it]
91%|█████████ | 39986/43939 [64:31:42<6:17:17, 5.73s/it]
91%|█████████ | 39987/43939 [64:31:48<6:16:36, 5.72s/it]
91%|█████████ | 39988/43939 [64:31:54<6:20:44, 5.78s/it]
91%|█████████ | 39989/43939 [64:31:59<6:18:15, 5.75s/it]
91%|█████████ | 39990/43939 [64:32:05<6:17:39, 5.74s/it]
91%|█████████ | 39991/43939 [64:32:11<6:16:45, 5.73s/it]
91%|█████████ | 39992/43939 [64:32:17<6:16:04, 5.72s/it]
91%|█████████ | 39993/43939 [64:32:22<6:15:17, 5.71s/it]
91%|█████████ | 39994/43939 [64:32:28<6:14:13, 5.69s/it]
91%|█████████ | 39995/43939 [64:32:34<6:14:54, 5.70s/it]
91%|█████████ | 39996/43939 [64:32:39<6:15:40, 5.72s/it]
91%|█████████ | 39997/43939 [64:32:45<6:14:43, 5.70s/it]
91%|█████████ | 39998/43939 [64:32:51<6:14:37, 5.70s/it]
91%|█████████ | 39999/43939 [64:32:56<6:14:35, 5.70s/it]
91%|█████████ | 40000/43939 [64:33:02<6:13:47, 5.69s/it]
{'loss': 0.6496, 'grad_norm': 1.6575472354888916, 'learning_rate': 1.2146427873732852e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.07, 'epoch': 3.64}
91%|█████████ | 40000/43939 [64:33:02<6:13:47, 5.69s/it]
91%|█████████ | 40001/43939 [64:33:08<6:13:38, 5.69s/it]
91%|█████████ | 40002/43939 [64:33:14<6:13:38, 5.69s/it]
91%|█████████ | 40003/43939 [64:33:19<6:13:18, 5.69s/it]
91%|█████████ | 40004/43939 [64:33:25<6:13:22, 5.69s/it]
91%|█████████ | 40005/43939 [64:33:31<6:13:12, 5.69s/it]
91%|█████████ | 40006/43939 [64:33:36<6:12:05, 5.68s/it]
91%|█████████ | 40007/43939 [64:33:42<6:12:12, 5.68s/it]
91%|█████████ | 40008/43939 [64:33:48<6:12:15, 5.68s/it]
91%|█████████ | 40009/43939 [64:33:53<6:12:58, 5.69s/it]
91%|█████████ | 40010/43939 [64:33:59<6:16:20, 5.75s/it]
91%|█████████ | 40011/43939 [64:34:05<6:15:32, 5.74s/it]
91%|█████████ | 40012/43939 [64:34:11<6:14:50, 5.73s/it]
91%|█████████ | 40013/43939 [64:34:16<6:14:19, 5.72s/it]
91%|█████████ | 40014/43939 [64:34:22<6:14:01, 5.72s/it]
91%|█████████ | 40015/43939 [64:34:28<6:13:46, 5.72s/it]
91%|█████████ | 40016/43939 [64:34:33<6:13:48, 5.72s/it]
91%|█████████ | 40017/43939 [64:34:39<6:13:21, 5.71s/it]
91%|█████████ | 40018/43939 [64:34:45<6:12:58, 5.71s/it]
91%|█████████ | 40019/43939 [64:34:51<6:12:49, 5.71s/it]
91%|█████████ | 40020/43939 [64:34:56<6:11:52, 5.69s/it]
91%|█████████ | 40021/43939 [64:35:02<6:12:10, 5.70s/it]
91%|█████████ | 40022/43939 [64:35:08<6:13:46, 5.73s/it]
91%|█████████ | 40023/43939 [64:35:13<6:13:03, 5.72s/it]
91%|█████████ | 40024/43939 [64:35:19<6:12:48, 5.71s/it]
91%|█████████ | 40025/43939 [64:35:25<6:12:15, 5.71s/it]
91%|█████████ | 40026/43939 [64:35:31<6:12:46, 5.72s/it]
91%|█████████ | 40027/43939 [64:35:36<6:12:38, 5.72s/it]
91%|█████████ | 40028/43939 [64:35:42<6:12:12, 5.71s/it]
91%|█████████ | 40029/43939 [64:35:48<6:12:40, 5.72s/it]
91%|█████████ | 40030/43939 [64:35:53<6:12:08, 5.71s/it]
91%|█████████ | 40031/43939 [64:35:59<6:16:47, 5.78s/it]
91%|█████████ | 40032/43939 [64:36:05<6:15:06, 5.76s/it]
91%|█████████ | 40033/43939 [64:36:11<6:13:49, 5.74s/it]
91%|█████████ | 40034/43939 [64:36:16<6:13:11, 5.73s/it]
91%|█████████ | 40035/43939 [64:36:22<6:12:33, 5.73s/it]
91%|█████████ | 40036/43939 [64:36:28<6:12:20, 5.72s/it]
91%|█████████ | 40037/43939 [64:36:34<6:12:17, 5.72s/it]
91%|█████████ | 40038/43939 [64:36:39<6:11:24, 5.71s/it]
91%|█████████ | 40039/43939 [64:36:45<6:10:11, 5.70s/it]
91%|█████████ | 40040/43939 [64:36:51<6:10:11, 5.70s/it]
91%|█████████ | 40041/43939 [64:36:56<6:10:07, 5.70s/it]
91%|█████████ | 40042/43939 [64:37:02<6:09:43, 5.69s/it]
91%|█████████ | 40043/43939 [64:37:08<6:09:11, 5.69s/it]
91%|█████████ | 40044/43939 [64:37:13<6:09:03, 5.69s/it]
91%|█████████ | 40045/43939 [64:37:19<6:09:22, 5.69s/it]
91%|█████████ | 40046/43939 [64:37:25<6:09:36, 5.70s/it]
91%|█████████ | 40047/43939 [64:37:31<6:10:44, 5.72s/it]
91%|█████████ | 40048/43939 [64:37:36<6:10:44, 5.72s/it]
91%|█████████ | 40049/43939 [64:37:42<6:09:14, 5.70s/it]
91%|█████████ | 40050/43939 [64:37:48<6:09:04, 5.69s/it]
{'loss': 0.6698, 'grad_norm': 1.321257472038269, 'learning_rate': 1.1842541029890048e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 491.1, 'epoch': 3.65}
91%|█████████ | 40050/43939 [64:37:48<6:09:04, 5.69s/it]
91%|█████████ | 40051/43939 [64:37:53<6:09:19, 5.70s/it]
91%|█████████ | 40052/43939 [64:37:59<6:09:29, 5.70s/it]
91%|█████████ | 40053/43939 [64:38:05<6:14:03, 5.78s/it]
91%|█████████ | 40054/43939 [64:38:11<6:12:21, 5.75s/it]
91%|█████████ | 40055/43939 [64:38:16<6:10:39, 5.73s/it]
91%|█████████ | 40056/43939 [64:38:22<6:10:16, 5.72s/it]
91%|█████████ | 40057/43939 [64:38:28<6:10:04, 5.72s/it]
91%|█████████ | 40058/43939 [64:38:33<6:09:42, 5.72s/it]
91%|█████████ | 40059/43939 [64:38:39<6:09:34, 5.71s/it]
91%|█████████ | 40060/43939 [64:38:45<6:09:55, 5.72s/it]
91%|█████████ | 40061/43939 [64:38:51<6:09:48, 5.72s/it]
91%|█████████ | 40062/43939 [64:38:56<6:09:20, 5.72s/it]
91%|█████████ | 40063/43939 [64:39:02<6:07:51, 5.69s/it]
91%|█████████ | 40064/43939 [64:39:08<6:08:09, 5.70s/it]
91%|█████████ | 40065/43939 [64:39:13<6:08:00, 5.70s/it]
91%|█████████ | 40066/43939 [64:39:19<6:07:26, 5.69s/it]
91%|█████████ | 40067/43939 [64:39:25<6:07:26, 5.69s/it]
91%|█████████ | 40068/43939 [64:39:30<6:07:15, 5.69s/it]
91%|█████████ | 40069/43939 [64:39:36<6:07:25, 5.70s/it]
91%|█████████ | 40070/43939 [64:39:42<6:07:48, 5.70s/it]
91%|█████████ | 40071/43939 [64:39:48<6:07:54, 5.71s/it]
91%|█████████ | 40072/43939 [64:39:53<6:07:34, 5.70s/it]
91%|█████████ | 40073/43939 [64:39:59<6:07:22, 5.70s/it]
91%|█████████ | 40074/43939 [64:40:05<6:07:37, 5.71s/it]
91%|█████████ | 40075/43939 [64:40:11<6:12:11, 5.78s/it]
91%|█████████ | 40076/43939 [64:40:16<6:10:17, 5.75s/it]
91%|█████████ | 40077/43939 [64:40:22<6:09:20, 5.74s/it]
91%|█████████ | 40078/43939 [64:40:28<6:08:13, 5.72s/it]
91%|█████████ | 40079/43939 [64:40:34<6:08:55, 5.73s/it]
91%|█████████ | 40080/43939 [64:40:39<6:08:21, 5.73s/it]
91%|█████████ | 40081/43939 [64:40:45<6:07:52, 5.72s/it]
91%|█████████ | 40082/43939 [64:40:51<6:06:51, 5.71s/it]
91%|█████████ | 40083/43939 [64:40:56<6:07:03, 5.71s/it]
91%|█████████ | 40084/43939 [64:41:02<6:07:28, 5.72s/it]
91%|█████████ | 40085/43939 [64:41:08<6:07:15, 5.72s/it]
91%|█████████ | 40086/43939 [64:41:13<6:06:59, 5.71s/it]
91%|█████████ | 40087/43939 [64:41:19<6:07:08, 5.72s/it]
91%|█████████ | 40088/43939 [64:41:25<6:07:59, 5.73s/it]
91%|█████████ | 40089/43939 [64:41:31<6:07:42, 5.73s/it]
91%|█████████ | 40090/43939 [64:41:36<6:07:34, 5.73s/it]
91%|█████████ | 40091/43939 [64:41:42<6:06:47, 5.72s/it]
91%|█████████ | 40092/43939 [64:41:48<6:05:51, 5.71s/it]
91%|█████████ | 40093/43939 [64:41:54<6:05:55, 5.71s/it]
91%|█████████ | 40094/43939 [64:41:59<6:05:11, 5.70s/it]
91%|█████████▏| 40095/43939 [64:42:05<6:04:37, 5.69s/it]
91%|█████████▏| 40096/43939 [64:42:11<6:09:46, 5.77s/it]
91%|█████████▏| 40097/43939 [64:42:17<6:08:20, 5.75s/it]
91%|█████████▏| 40098/43939 [64:42:22<6:07:56, 5.75s/it]
91%|█████████▏| 40099/43939 [64:42:28<6:06:53, 5.73s/it]
91%|█████████▏| 40100/43939 [64:42:34<6:06:26, 5.73s/it]
{'loss': 0.6591, 'grad_norm': 1.3233212232589722, 'learning_rate': 1.154241167962486e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.64, 'epoch': 3.65}
91%|█████████▏| 40100/43939 [64:42:34<6:06:26, 5.73s/it]
91%|█████████▏| 40101/43939 [64:42:39<6:07:23, 5.74s/it]
91%|█████████▏| 40102/43939 [64:42:45<6:06:28, 5.73s/it]
91%|█████████▏| 40103/43939 [64:42:51<6:06:32, 5.73s/it]
91%|█████████▏| 40104/43939 [64:42:57<6:06:18, 5.73s/it]
91%|█████████▏| 40105/43939 [64:43:02<6:06:00, 5.73s/it]
91%|█████████▏| 40106/43939 [64:43:08<6:05:20, 5.72s/it]
91%|█████████▏| 40107/43939 [64:43:14<6:04:57, 5.71s/it]
91%|█████████▏| 40108/43939 [64:43:19<6:04:09, 5.70s/it]
91%|█████████▏| 40109/43939 [64:43:25<6:04:11, 5.71s/it]
91%|█████████▏| 40110/43939 [64:43:31<6:04:22, 5.71s/it]
91%|█████████▏| 40111/43939 [64:43:37<6:03:48, 5.70s/it]
91%|█████████▏| 40112/43939 [64:43:42<6:03:12, 5.69s/it]
91%|█████████▏| 40113/43939 [64:43:48<6:04:03, 5.71s/it]
91%|█████████▏| 40114/43939 [64:43:54<6:04:23, 5.72s/it]
91%|█████████▏| 40115/43939 [64:43:59<6:04:14, 5.71s/it]
91%|█████████▏| 40116/43939 [64:44:05<6:03:13, 5.70s/it]
91%|█████████▏| 40117/43939 [64:44:11<6:02:58, 5.70s/it]
91%|█████████▏| 40118/43939 [64:44:17<6:07:13, 5.77s/it]
91%|█████████▏| 40119/43939 [64:44:22<6:06:15, 5.75s/it]
91%|█████████▏| 40120/43939 [64:44:28<6:05:01, 5.73s/it]
91%|█████████▏| 40121/43939 [64:44:34<6:04:16, 5.72s/it]
91%|█████████▏| 40122/43939 [64:44:40<6:04:18, 5.73s/it]
91%|█████████▏| 40123/43939 [64:44:45<6:03:37, 5.72s/it]
91%|█████████▏| 40124/43939 [64:44:51<6:03:46, 5.72s/it]
91%|█████████▏| 40125/43939 [64:44:57<6:03:23, 5.72s/it]
91%|█████████▏| 40126/43939 [64:45:02<6:03:02, 5.71s/it]
91%|█████████▏| 40127/43939 [64:45:08<6:02:44, 5.71s/it]
91%|█████████▏| 40128/43939 [64:45:14<6:01:59, 5.70s/it]
91%|█████████▏| 40129/43939 [64:45:20<6:02:38, 5.71s/it]
91%|█████████▏| 40130/43939 [64:45:25<6:01:32, 5.70s/it]
91%|█████████▏| 40131/43939 [64:45:31<6:01:52, 5.70s/it]
91%|█████████▏| 40132/43939 [64:45:37<6:00:43, 5.69s/it]
91%|█████████▏| 40133/43939 [64:45:42<6:00:49, 5.69s/it]
91%|█████████▏| 40134/43939 [64:45:48<6:01:21, 5.70s/it]
91%|█████████▏| 40135/43939 [64:45:54<6:01:06, 5.70s/it]
91%|█████████▏| 40136/43939 [64:45:59<6:00:36, 5.69s/it]
91%|█████████▏| 40137/43939 [64:46:05<6:01:59, 5.71s/it]
91%|█████████▏| 40138/43939 [64:46:11<6:02:03, 5.72s/it]
91%|█████████▏| 40139/43939 [64:46:17<6:06:21, 5.78s/it]
91%|█████████▏| 40140/43939 [64:46:22<6:04:58, 5.76s/it]
91%|█████████▏| 40141/43939 [64:46:28<6:03:46, 5.75s/it]
91%|█████████▏| 40142/43939 [64:46:34<6:02:37, 5.73s/it]
91%|█████████▏| 40143/43939 [64:46:40<6:02:28, 5.73s/it]
91%|█████████▏| 40144/43939 [64:46:45<6:01:55, 5.72s/it]
91%|█████████▏| 40145/43939 [64:46:51<6:00:50, 5.71s/it]
91%|█████████▏| 40146/43939 [64:46:57<6:00:49, 5.71s/it]
91%|█████████▏| 40147/43939 [64:47:02<6:00:37, 5.71s/it]
91%|█████████▏| 40148/43939 [64:47:08<6:00:38, 5.71s/it]
91%|█████████▏| 40149/43939 [64:47:14<6:00:49, 5.71s/it]
91%|█████████▏| 40150/43939 [64:47:19<6:00:01, 5.70s/it]
{'loss': 0.6715, 'grad_norm': 1.799234390258789, 'learning_rate': 1.1246044558183066e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.27, 'epoch': 3.65}
91%|█████████▏| 40150/43939 [64:47:19<6:00:01, 5.70s/it]
91%|█████████▏| 40151/43939 [64:47:25<5:59:50, 5.70s/it]
91%|█████████▏| 40152/43939 [64:47:31<6:00:54, 5.72s/it]
91%|█████████▏| 40153/43939 [64:47:37<6:01:19, 5.73s/it]
91%|█████████▏| 40154/43939 [64:47:42<6:01:18, 5.73s/it]
91%|█████████▏| 40155/43939 [64:47:48<6:00:14, 5.71s/it]
91%|█████████▏| 40156/43939 [64:47:54<5:59:17, 5.70s/it]
91%|█████████▏| 40157/43939 [64:47:59<5:58:33, 5.69s/it]
91%|█████████▏| 40158/43939 [64:48:05<5:58:52, 5.69s/it]
91%|█████████▏| 40159/43939 [64:48:11<5:58:39, 5.69s/it]
91%|█████████▏| 40160/43939 [64:48:17<5:58:35, 5.69s/it]
91%|█████████▏| 40161/43939 [64:48:22<6:03:30, 5.77s/it]
91%|█████████▏| 40162/43939 [64:48:28<6:02:06, 5.75s/it]
91%|█████████▏| 40163/43939 [64:48:34<6:01:01, 5.74s/it]
91%|█████████▏| 40164/43939 [64:48:40<5:59:29, 5.71s/it]
91%|█████████▏| 40165/43939 [64:48:45<5:58:27, 5.70s/it]
91%|█████████▏| 40166/43939 [64:48:51<5:58:02, 5.69s/it]
91%|█████████▏| 40167/43939 [64:48:57<5:58:40, 5.71s/it]
91%|█████████▏| 40168/43939 [64:49:02<5:59:05, 5.71s/it]
91%|█████████▏| 40169/43939 [64:49:08<5:59:14, 5.72s/it]
91%|█████████▏| 40170/43939 [64:49:14<5:58:59, 5.71s/it]
91%|█████████▏| 40171/43939 [64:49:20<5:59:16, 5.72s/it]
91%|█████████▏| 40172/43939 [64:49:25<5:58:57, 5.72s/it]
91%|█████████▏| 40173/43939 [64:49:31<5:58:14, 5.71s/it]
91%|█████████▏| 40174/43939 [64:49:37<5:57:44, 5.70s/it]
91%|█████████▏| 40175/43939 [64:49:42<5:57:35, 5.70s/it]
91%|█████████▏| 40176/43939 [64:49:48<5:58:08, 5.71s/it]
91%|█████████▏| 40177/43939 [64:49:54<5:58:43, 5.72s/it]
91%|█████████▏| 40178/43939 [64:49:59<5:57:08, 5.70s/it]
91%|█████████▏| 40179/43939 [64:50:05<5:57:01, 5.70s/it]
91%|█████████▏| 40180/43939 [64:50:11<5:56:50, 5.70s/it]
91%|█████████▏| 40181/43939 [64:50:17<5:57:38, 5.71s/it]
91%|█████████▏| 40182/43939 [64:50:22<5:57:34, 5.71s/it]
91%|█████████▏| 40183/43939 [64:50:28<6:02:20, 5.79s/it]
91%|█████████▏| 40184/43939 [64:50:34<6:01:06, 5.77s/it]
91%|█████████▏| 40185/43939 [64:50:40<5:59:40, 5.75s/it]
91%|█████████▏| 40186/43939 [64:50:45<5:59:14, 5.74s/it]
91%|█████████▏| 40187/43939 [64:50:51<5:59:33, 5.75s/it]
91%|█████████▏| 40188/43939 [64:50:57<5:59:17, 5.75s/it]
91%|█████████▏| 40189/43939 [64:51:03<5:58:26, 5.74s/it]
91%|█████████▏| 40190/43939 [64:51:08<5:57:31, 5.72s/it]
91%|█████████▏| 40191/43939 [64:51:14<5:56:56, 5.71s/it]
91%|█████████▏| 40192/43939 [64:51:20<5:56:25, 5.71s/it]
91%|█████████▏| 40193/43939 [64:51:25<5:56:16, 5.71s/it]
91%|█████████▏| 40194/43939 [64:51:31<5:56:01, 5.70s/it]
91%|█████████▏| 40195/43939 [64:51:37<5:56:04, 5.71s/it]
91%|█████████▏| 40196/43939 [64:51:43<5:56:49, 5.72s/it]
91%|█████████▏| 40197/43939 [64:51:48<5:55:44, 5.70s/it]
91%|█████████▏| 40198/43939 [64:51:54<5:55:14, 5.70s/it]
91%|█████████▏| 40199/43939 [64:52:00<5:55:44, 5.71s/it]
91%|█████████▏| 40200/43939 [64:52:05<5:54:58, 5.70s/it]
{'loss': 0.6682, 'grad_norm': 2.0476598739624023, 'learning_rate': 1.0953444341452458e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 505.54, 'epoch': 3.66}
91%|█████████▏| 40200/43939 [64:52:05<5:54:58, 5.70s/it]
91%|█████████▏| 40201/43939 [64:52:11<5:54:43, 5.69s/it]
91%|█████████▏| 40202/43939 [64:52:17<5:54:02, 5.68s/it]
91%|█████████▏| 40203/43939 [64:52:22<5:54:08, 5.69s/it]
91%|█████████▏| 40204/43939 [64:52:28<5:59:50, 5.78s/it]
92%|█████████▏| 40205/43939 [64:52:34<5:58:02, 5.75s/it]
92%|█████████▏| 40206/43939 [64:52:40<5:56:26, 5.73s/it]
92%|█████████▏| 40207/43939 [64:52:45<5:56:17, 5.73s/it]
92%|█████████▏| 40208/43939 [64:52:51<5:55:24, 5.72s/it]
92%|█████████▏| 40209/43939 [64:52:57<5:54:20, 5.70s/it]
92%|█████████▏| 40210/43939 [64:53:02<5:54:15, 5.70s/it]
92%|█████████▏| 40211/43939 [64:53:08<5:53:44, 5.69s/it]
92%|█████████▏| 40212/43939 [64:53:14<5:53:53, 5.70s/it]
92%|█████████▏| 40213/43939 [64:53:20<5:54:34, 5.71s/it]
92%|█████████▏| 40214/43939 [64:53:25<5:54:14, 5.71s/it]
92%|█████████▏| 40215/43939 [64:53:31<5:54:00, 5.70s/it]
92%|█████████▏| 40216/43939 [64:53:37<5:54:18, 5.71s/it]
92%|█████████▏| 40217/43939 [64:53:42<5:53:49, 5.70s/it]
92%|█████████▏| 40218/43939 [64:53:48<5:53:48, 5.70s/it]
92%|█████████▏| 40219/43939 [64:53:54<5:53:14, 5.70s/it]
92%|█████████▏| 40220/43939 [64:54:00<5:53:24, 5.70s/it]
92%|█████████▏| 40221/43939 [64:54:05<5:53:41, 5.71s/it]
92%|█████████▏| 40222/43939 [64:54:11<5:53:01, 5.70s/it]
92%|█████████▏| 40223/43939 [64:54:17<5:52:16, 5.69s/it]
92%|█████████▏| 40224/43939 [64:54:22<5:53:51, 5.72s/it]
92%|█████████▏| 40225/43939 [64:54:28<5:53:20, 5.71s/it]
92%|█████████▏| 40226/43939 [64:54:34<5:57:21, 5.77s/it]
92%|█████████▏| 40227/43939 [64:54:40<5:56:13, 5.76s/it]
92%|█████████▏| 40228/43939 [64:54:45<5:55:58, 5.76s/it]
92%|█████████▏| 40229/43939 [64:54:51<5:54:39, 5.74s/it]
92%|█████████▏| 40230/43939 [64:54:57<5:53:24, 5.72s/it]
92%|█████████▏| 40231/43939 [64:55:03<5:53:14, 5.72s/it]
92%|█████████▏| 40232/43939 [64:55:08<5:52:32, 5.71s/it]
92%|█████████▏| 40233/43939 [64:55:14<5:51:55, 5.70s/it]
92%|█████████▏| 40234/43939 [64:55:20<5:51:33, 5.69s/it]
92%|█████████▏| 40235/43939 [64:55:25<5:53:04, 5.72s/it]
92%|█████████▏| 40236/43939 [64:55:31<5:52:40, 5.71s/it]
92%|█████████▏| 40237/43939 [64:55:37<5:52:43, 5.72s/it]
92%|█████████▏| 40238/43939 [64:55:42<5:52:17, 5.71s/it]
92%|█████████▏| 40239/43939 [64:55:48<5:52:45, 5.72s/it]
92%|█████████▏| 40240/43939 [64:55:54<5:53:08, 5.73s/it]
92%|█████████▏| 40241/43939 [64:56:00<5:52:24, 5.72s/it]
92%|█████████▏| 40242/43939 [64:56:05<5:52:03, 5.71s/it]
92%|█████████▏| 40243/43939 [64:56:11<5:51:24, 5.70s/it]
92%|█████████▏| 40244/43939 [64:56:17<5:51:24, 5.71s/it]
92%|█████████▏| 40245/43939 [64:56:22<5:50:50, 5.70s/it]
92%|█████████▏| 40246/43939 [64:56:28<5:51:38, 5.71s/it]
92%|█████████▏| 40247/43939 [64:56:34<5:50:44, 5.70s/it]
92%|█████████▏| 40248/43939 [64:56:40<5:54:59, 5.77s/it]
92%|█████████▏| 40249/43939 [64:56:45<5:53:30, 5.75s/it]
92%|█████████▏| 40250/43939 [64:56:51<5:53:15, 5.75s/it]
{'loss': 0.6373, 'grad_norm': 1.5130788087844849, 'learning_rate': 1.0664615645889054e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 465.58, 'epoch': 3.66}
92%|█████████▏| 40250/43939 [64:56:51<5:53:15, 5.75s/it]
92%|█████████▏| 40251/43939 [64:56:57<5:52:52, 5.74s/it]
92%|█████████▏| 40252/43939 [64:57:03<5:52:10, 5.73s/it]
92%|█████████▏| 40253/43939 [64:57:08<5:51:24, 5.72s/it]
92%|█████████▏| 40254/43939 [64:57:14<5:51:22, 5.72s/it]
92%|█████████▏| 40255/43939 [64:57:20<5:51:08, 5.72s/it]
92%|█████████▏| 40256/43939 [64:57:25<5:50:30, 5.71s/it]
92%|█████████▏| 40257/43939 [64:57:31<5:50:24, 5.71s/it]
92%|█████████▏| 40258/43939 [64:57:37<5:50:42, 5.72s/it]
92%|█████████▏| 40259/43939 [64:57:43<5:50:42, 5.72s/it]
92%|█████████▏| 40260/43939 [64:57:48<5:51:01, 5.72s/it]
92%|█████████▏| 40261/43939 [64:57:54<5:51:17, 5.73s/it]
92%|█████████▏| 40262/43939 [64:58:00<5:49:59, 5.71s/it]
92%|█████████▏| 40263/43939 [64:58:06<5:50:14, 5.72s/it]
92%|█████████▏| 40264/43939 [64:58:11<5:50:56, 5.73s/it]
92%|█████████▏| 40265/43939 [64:58:17<5:50:05, 5.72s/it]
92%|█████████▏| 40266/43939 [64:58:23<5:49:51, 5.72s/it]
92%|█████████▏| 40267/43939 [64:58:28<5:48:51, 5.70s/it]
92%|█████████▏| 40268/43939 [64:58:34<5:48:49, 5.70s/it]
92%|█████████▏| 40269/43939 [64:58:40<5:53:35, 5.78s/it]
92%|█████████▏| 40270/43939 [64:58:46<5:52:59, 5.77s/it]
92%|█████████▏| 40271/43939 [64:58:52<5:52:32, 5.77s/it]
92%|█████████▏| 40272/43939 [64:58:57<5:51:12, 5.75s/it]
92%|█████████▏| 40273/43939 [64:59:03<5:50:20, 5.73s/it]
92%|█████████▏| 40274/43939 [64:59:09<5:50:20, 5.74s/it]
92%|█████████▏| 40275/43939 [64:59:14<5:49:47, 5.73s/it]
92%|█████████▏| 40276/43939 [64:59:20<5:50:14, 5.74s/it]
92%|█████████▏| 40277/43939 [64:59:26<5:48:59, 5.72s/it]
92%|█████████▏| 40278/43939 [64:59:31<5:48:33, 5.71s/it]
92%|█████████▏| 40279/43939 [64:59:37<5:48:26, 5.71s/it]
92%|█████████▏| 40280/43939 [64:59:43<5:47:55, 5.71s/it]
92%|█████████▏| 40281/43939 [64:59:49<5:48:27, 5.72s/it]
92%|█████████▏| 40282/43939 [64:59:54<5:48:04, 5.71s/it]
92%|█████████▏| 40283/43939 [65:00:00<5:47:46, 5.71s/it]
92%|█████████▏| 40284/43939 [65:00:06<5:48:00, 5.71s/it]
92%|█████████▏| 40285/43939 [65:00:11<5:47:35, 5.71s/it]
92%|█████████▏| 40286/43939 [65:00:17<5:47:47, 5.71s/it]
92%|█████████▏| 40287/43939 [65:00:23<5:48:00, 5.72s/it]
92%|█████████▏| 40288/43939 [65:00:29<5:48:21, 5.72s/it]
92%|█████████▏| 40289/43939 [65:00:34<5:47:48, 5.72s/it]
92%|█████████▏| 40290/43939 [65:00:40<5:51:26, 5.78s/it]
92%|█████████▏| 40291/43939 [65:00:46<5:49:59, 5.76s/it]
92%|█████████▏| 40292/43939 [65:00:52<5:48:21, 5.73s/it]
92%|█████████▏| 40293/43939 [65:00:57<5:48:03, 5.73s/it]
92%|█████████▏| 40294/43939 [65:01:03<5:47:00, 5.71s/it]
92%|█████████▏| 40295/43939 [65:01:09<5:46:08, 5.70s/it]
92%|█████████▏| 40296/43939 [65:01:14<5:47:11, 5.72s/it]
92%|█████████▏| 40297/43939 [65:01:20<5:46:44, 5.71s/it]
92%|█████████▏| 40298/43939 [65:01:26<5:46:08, 5.70s/it]
92%|█████████▏| 40299/43939 [65:01:32<5:46:28, 5.71s/it]
92%|█████████▏| 40300/43939 [65:01:37<5:46:19, 5.71s/it]
{'loss': 0.6325, 'grad_norm': 1.7768909931182861, 'learning_rate': 1.0379563028444255e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 481.66, 'epoch': 3.67}
92%|█████████▏| 40300/43939 [65:01:37<5:46:19, 5.71s/it]
92%|█████████▏| 40301/43939 [65:01:43<5:46:04, 5.71s/it]
92%|█████████▏| 40302/43939 [65:01:49<5:46:49, 5.72s/it]
92%|█████████▏| 40303/43939 [65:01:54<5:46:25, 5.72s/it]
92%|█████████▏| 40304/43939 [65:02:00<5:46:02, 5.71s/it]
92%|█████████▏| 40305/43939 [65:02:06<5:45:35, 5.71s/it]
92%|█████████▏| 40306/43939 [65:02:12<5:46:08, 5.72s/it]
92%|█████████▏| 40307/43939 [65:02:17<5:46:45, 5.73s/it]
92%|█████████▏| 40308/43939 [65:02:23<5:46:20, 5.72s/it]
92%|█████████▏| 40309/43939 [65:02:29<5:45:30, 5.71s/it]
92%|█████████▏| 40310/43939 [65:02:34<5:45:14, 5.71s/it]
92%|█████████▏| 40311/43939 [65:02:40<5:45:28, 5.71s/it]
92%|█████████▏| 40312/43939 [65:02:46<5:49:12, 5.78s/it]
92%|█████████▏| 40313/43939 [65:02:52<5:47:41, 5.75s/it]
92%|█████████▏| 40314/43939 [65:02:58<5:46:48, 5.74s/it]
92%|█████████▏| 40315/43939 [65:03:03<5:46:30, 5.74s/it]
92%|█████████▏| 40316/43939 [65:03:09<5:45:43, 5.73s/it]
92%|█████████▏| 40317/43939 [65:03:15<5:45:13, 5.72s/it]
92%|█████████▏| 40318/43939 [65:03:20<5:45:00, 5.72s/it]
92%|█████████▏| 40319/43939 [65:03:26<5:44:52, 5.72s/it]
92%|█████████▏| 40320/43939 [65:03:32<5:44:42, 5.72s/it]
92%|█████████▏| 40321/43939 [65:03:38<5:44:50, 5.72s/it]
92%|█████████▏| 40322/43939 [65:03:43<5:44:21, 5.71s/it]
92%|█████████▏| 40323/43939 [65:03:49<5:43:43, 5.70s/it]
92%|█████████▏| 40324/43939 [65:03:55<5:44:10, 5.71s/it]
92%|█████████▏| 40325/43939 [65:04:00<5:44:00, 5.71s/it]
92%|█████████▏| 40326/43939 [65:04:06<5:43:09, 5.70s/it]
92%|█████████▏| 40327/43939 [65:04:12<5:43:34, 5.71s/it]
92%|█████████▏| 40328/43939 [65:04:17<5:42:47, 5.70s/it]
92%|█████████▏| 40329/43939 [65:04:23<5:42:28, 5.69s/it]
92%|█████████▏| 40330/43939 [65:04:29<5:42:25, 5.69s/it]
92%|█████████▏| 40331/43939 [65:04:34<5:42:27, 5.69s/it]
92%|█████████▏| 40332/43939 [65:04:40<5:43:31, 5.71s/it]
92%|█████████▏| 40333/43939 [65:04:46<5:43:11, 5.71s/it]
92%|█████████▏| 40334/43939 [65:04:52<5:46:43, 5.77s/it]
92%|█████████▏| 40335/43939 [65:04:58<5:45:36, 5.75s/it]
92%|█████████▏| 40336/43939 [65:05:03<5:44:39, 5.74s/it]
92%|█████████▏| 40337/43939 [65:05:09<5:43:59, 5.73s/it]
92%|█████████▏| 40338/43939 [65:05:15<5:43:12, 5.72s/it]
92%|█████████▏| 40339/43939 [65:05:20<5:42:53, 5.71s/it]
92%|█████████▏| 40340/43939 [65:05:26<5:42:17, 5.71s/it]
92%|█████████▏| 40341/43939 [65:05:32<5:41:44, 5.70s/it]
92%|█████████▏| 40342/43939 [65:05:37<5:41:31, 5.70s/it]
92%|█████████▏| 40343/43939 [65:05:43<5:41:18, 5.69s/it]
92%|█████████▏| 40344/43939 [65:05:49<5:42:02, 5.71s/it]
92%|█████████▏| 40345/43939 [65:05:55<5:41:40, 5.70s/it]
92%|█████████▏| 40346/43939 [65:06:00<5:41:49, 5.71s/it]
92%|█████████▏| 40347/43939 [65:06:06<5:41:45, 5.71s/it]
92%|█████████▏| 40348/43939 [65:06:12<5:41:21, 5.70s/it]
92%|█████████▏| 40349/43939 [65:06:17<5:41:46, 5.71s/it]
92%|█████████▏| 40350/43939 [65:06:23<5:41:09, 5.70s/it]
{'loss': 0.64, 'grad_norm': 1.909771203994751, 'learning_rate': 1.0098290986493009e-06, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 474.29, 'epoch': 3.67}
92%|█████████▏| 40350/43939 [65:06:23<5:41:09, 5.70s/it]
92%|█████████▏| 40351/43939 [65:06:29<5:40:55, 5.70s/it]
92%|█████████▏| 40352/43939 [65:06:35<5:41:17, 5.71s/it]
92%|█████████▏| 40353/43939 [65:06:40<5:41:02, 5.71s/it]
92%|█████████▏| 40354/43939 [65:06:46<5:41:41, 5.72s/it]
92%|█████████▏| 40355/43939 [65:06:52<5:45:52, 5.79s/it]
92%|█████████▏| 40356/43939 [65:06:58<5:44:43, 5.77s/it]
92%|█████████▏| 40357/43939 [65:07:03<5:43:43, 5.76s/it]
92%|█████████▏| 40358/43939 [65:07:09<5:42:08, 5.73s/it]
92%|█████████▏| 40359/43939 [65:07:15<5:42:13, 5.74s/it]
92%|█████████▏| 40360/43939 [65:07:20<5:41:24, 5.72s/it]
92%|█████████▏| 40361/43939 [65:07:26<5:40:33, 5.71s/it]
92%|█████████▏| 40362/43939 [65:07:32<5:40:09, 5.71s/it]
92%|█████████▏| 40363/43939 [65:07:38<5:40:27, 5.71s/it]
92%|█████████▏| 40364/43939 [65:07:43<5:40:23, 5.71s/it]
92%|█████████▏| 40365/43939 [65:07:49<5:40:17, 5.71s/it]
92%|█████████▏| 40366/43939 [65:07:55<5:40:01, 5.71s/it]
92%|█████████▏| 40367/43939 [65:08:00<5:40:48, 5.72s/it]
92%|█████████▏| 40368/43939 [65:08:06<5:40:31, 5.72s/it]
92%|█████████▏| 40369/43939 [65:08:12<5:40:06, 5.72s/it]
92%|█████████▏| 40370/43939 [65:08:18<5:39:31, 5.71s/it]
92%|█████████▏| 40371/43939 [65:08:23<5:39:00, 5.70s/it]
92%|█████████▏| 40372/43939 [65:08:29<5:38:53, 5.70s/it]
92%|█████████▏| 40373/43939 [65:08:35<5:38:26, 5.69s/it]
92%|█████████▏| 40374/43939 [65:08:40<5:38:30, 5.70s/it]
92%|█████████▏| 40375/43939 [65:08:46<5:38:37, 5.70s/it]
92%|█████████▏| 40376/43939 [65:08:52<5:38:14, 5.70s/it]
92%|█████████▏| 40377/43939 [65:08:58<5:42:36, 5.77s/it]
92%|█████████▏| 40378/43939 [65:09:03<5:41:11, 5.75s/it]
92%|█████████▏| 40379/43939 [65:09:09<5:39:59, 5.73s/it]
92%|█████████▏| 40380/43939 [65:09:15<5:38:37, 5.71s/it]
92%|█████████▏| 40381/43939 [65:09:21<5:39:50, 5.73s/it]
92%|█████████▏| 40382/43939 [65:09:26<5:39:02, 5.72s/it]
92%|█████████▏| 40383/43939 [65:09:32<5:39:03, 5.72s/it]
92%|█████████▏| 40384/43939 [65:09:38<5:38:24, 5.71s/it]
92%|█████████▏| 40385/43939 [65:09:43<5:37:51, 5.70s/it]
92%|█████████▏| 40386/43939 [65:09:49<5:37:57, 5.71s/it]
92%|█████████▏| 40387/43939 [65:09:55<5:38:19, 5.71s/it]
92%|█████████▏| 40388/43939 [65:10:00<5:38:13, 5.71s/it]
92%|█████████▏| 40389/43939 [65:10:06<5:37:26, 5.70s/it]
92%|█████████▏| 40390/43939 [65:10:12<5:37:16, 5.70s/it]
92%|█████████▏| 40391/43939 [65:10:18<5:37:10, 5.70s/it]
92%|█████████▏| 40392/43939 [65:10:23<5:36:41, 5.70s/it]
92%|█████████▏| 40393/43939 [65:10:29<5:37:26, 5.71s/it]
92%|█████████▏| 40394/43939 [65:10:35<5:37:00, 5.70s/it]
92%|█████████▏| 40395/43939 [65:10:40<5:36:35, 5.70s/it]
92%|█████████▏| 40396/43939 [65:10:46<5:36:20, 5.70s/it]
92%|█████████▏| 40397/43939 [65:10:52<5:36:22, 5.70s/it]
92%|█████████▏| 40398/43939 [65:10:58<5:40:26, 5.77s/it]
92%|█████████▏| 40399/43939 [65:11:03<5:39:18, 5.75s/it]
92%|█████████▏| 40400/43939 [65:11:09<5:38:27, 5.74s/it]
{'loss': 0.6608, 'grad_norm': 1.581221580505371, 'learning_rate': 9.820803957762693e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.98, 'epoch': 3.68}
92%|█████████▏| 40400/43939 [65:11:09<5:38:27, 5.74s/it]
92%|█████████▏| 40401/43939 [65:11:15<5:37:52, 5.73s/it]
92%|█████████▏| 40402/43939 [65:11:20<5:36:39, 5.71s/it]
92%|█████████▏| 40403/43939 [65:11:26<5:35:27, 5.69s/it]
92%|█████████▏| 40404/43939 [65:11:32<5:34:51, 5.68s/it]
92%|█████████▏| 40405/43939 [65:11:37<5:34:29, 5.68s/it]
92%|█████████▏| 40406/43939 [65:11:43<5:35:07, 5.69s/it]
92%|█████████▏| 40407/43939 [65:11:49<5:35:27, 5.70s/it]
92%|█████████▏| 40408/43939 [65:11:55<5:35:34, 5.70s/it]
92%|█████████▏| 40409/43939 [65:12:00<5:35:06, 5.70s/it]
92%|█████████▏| 40410/43939 [65:12:06<5:35:41, 5.71s/it]
92%|█████████▏| 40411/43939 [65:12:12<5:35:25, 5.70s/it]
92%|█████████▏| 40412/43939 [65:12:17<5:35:41, 5.71s/it]
92%|█████████▏| 40413/43939 [65:12:23<5:34:49, 5.70s/it]
92%|█████████▏| 40414/43939 [65:12:29<5:35:03, 5.70s/it]
92%|█████████▏| 40415/43939 [65:12:35<5:35:03, 5.70s/it]
92%|█████████▏| 40416/43939 [65:12:40<5:35:16, 5.71s/it]
92%|█████████▏| 40417/43939 [65:12:46<5:34:33, 5.70s/it]
92%|█████████▏| 40418/43939 [65:12:52<5:34:20, 5.70s/it]
92%|█████████▏| 40419/43939 [65:12:57<5:34:30, 5.70s/it]
92%|█████████▏| 40420/43939 [65:13:03<5:37:33, 5.76s/it]
92%|█████████▏| 40421/43939 [65:13:09<5:37:02, 5.75s/it]
92%|█████████▏| 40422/43939 [65:13:15<5:36:23, 5.74s/it]
92%|█████████▏| 40423/43939 [65:13:20<5:35:40, 5.73s/it]
92%|█████████▏| 40424/43939 [65:13:26<5:34:42, 5.71s/it]
92%|█████████▏| 40425/43939 [65:13:32<5:34:19, 5.71s/it]
92%|█████████▏| 40426/43939 [65:13:37<5:34:11, 5.71s/it]
92%|█████████▏| 40427/43939 [65:13:43<5:34:25, 5.71s/it]
92%|█████████▏| 40428/43939 [65:13:49<5:34:14, 5.71s/it]
92%|█████████▏| 40429/43939 [65:13:55<5:34:57, 5.73s/it]
92%|█████████▏| 40430/43939 [65:14:00<5:34:24, 5.72s/it]
92%|█████████▏| 40431/43939 [65:14:06<5:33:34, 5.71s/it]
92%|█████████▏| 40432/43939 [65:14:12<5:32:43, 5.69s/it]
92%|█████████▏| 40433/43939 [65:14:17<5:31:59, 5.68s/it]
92%|█████████▏| 40434/43939 [65:14:23<5:32:36, 5.69s/it]
92%|█████████▏| 40435/43939 [65:14:29<5:33:34, 5.71s/it]
92%|█████████▏| 40436/43939 [65:14:35<5:33:21, 5.71s/it]
92%|█████████▏| 40437/43939 [65:14:40<5:33:33, 5.71s/it]
92%|█████████▏| 40438/43939 [65:14:46<5:32:48, 5.70s/it]
92%|█████████▏| 40439/43939 [65:14:52<5:32:34, 5.70s/it]
92%|█████████▏| 40440/43939 [65:14:57<5:32:17, 5.70s/it]
92%|█████████▏| 40441/43939 [65:15:03<5:35:55, 5.76s/it]
92%|█████████▏| 40442/43939 [65:15:09<5:34:49, 5.74s/it]
92%|█████████▏| 40443/43939 [65:15:15<5:33:55, 5.73s/it]
92%|█████████▏| 40444/43939 [65:15:20<5:33:42, 5.73s/it]
92%|█████████▏| 40445/43939 [65:15:26<5:33:15, 5.72s/it]
92%|█████████▏| 40446/43939 [65:15:32<5:33:38, 5.73s/it]
92%|█████████▏| 40447/43939 [65:15:37<5:32:29, 5.71s/it]
92%|█████████▏| 40448/43939 [65:15:43<5:32:29, 5.71s/it]
92%|█████████▏| 40449/43939 [65:15:49<5:32:32, 5.72s/it]
92%|█████████▏| 40450/43939 [65:15:55<5:32:11, 5.71s/it]
{'loss': 0.6397, 'grad_norm': 1.7531181573867798, 'learning_rate': 9.547106320263344e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 447.41, 'epoch': 3.68}
92%|█████████▏| 40450/43939 [65:15:55<5:32:11, 5.71s/it]
92%|█████████▏| 40451/43939 [65:16:00<5:32:48, 5.72s/it]
92%|█████████▏| 40452/43939 [65:16:06<5:32:22, 5.72s/it]
92%|█████████▏| 40453/43939 [65:16:12<5:31:55, 5.71s/it]
92%|█████████▏| 40454/43939 [65:16:17<5:31:41, 5.71s/it]
92%|█████████▏| 40455/43939 [65:16:23<5:30:55, 5.70s/it]
92%|█████████▏| 40456/43939 [65:16:29<5:31:15, 5.71s/it]
92%|█████████▏| 40457/43939 [65:16:35<5:31:34, 5.71s/it]
92%|█████████▏| 40458/43939 [65:16:40<5:32:10, 5.73s/it]
92%|█████████▏| 40459/43939 [65:16:46<5:31:13, 5.71s/it]
92%|█████████▏| 40460/43939 [65:16:52<5:31:53, 5.72s/it]
92%|█████████▏| 40461/43939 [65:16:57<5:30:58, 5.71s/it]
92%|█████████▏| 40462/43939 [65:17:03<5:30:43, 5.71s/it]
92%|█████████▏| 40463/43939 [65:17:09<5:34:21, 5.77s/it]
92%|█████████▏| 40464/43939 [65:17:15<5:33:07, 5.75s/it]
92%|█████████▏| 40465/43939 [65:17:20<5:31:50, 5.73s/it]
92%|█████████▏| 40466/43939 [65:17:26<5:30:43, 5.71s/it]
92%|█████████▏| 40467/43939 [65:17:32<5:31:15, 5.72s/it]
92%|█████████▏| 40468/43939 [65:17:38<5:30:45, 5.72s/it]
92%|█████████▏| 40469/43939 [65:17:43<5:30:21, 5.71s/it]
92%|█████████▏| 40470/43939 [65:17:49<5:29:53, 5.71s/it]
92%|█████████▏| 40471/43939 [65:17:55<5:30:29, 5.72s/it]
92%|█████████▏| 40472/43939 [65:18:00<5:30:05, 5.71s/it]
92%|█████████▏| 40473/43939 [65:18:06<5:29:15, 5.70s/it]
92%|█████████▏| 40474/43939 [65:18:12<5:29:40, 5.71s/it]
92%|█████████▏| 40475/43939 [65:18:18<5:29:41, 5.71s/it]
92%|█████████▏| 40476/43939 [65:18:23<5:29:17, 5.71s/it]
92%|█████████▏| 40477/43939 [65:18:29<5:29:30, 5.71s/it]
92%|█████████▏| 40478/43939 [65:18:35<5:28:44, 5.70s/it]
92%|█████████▏| 40479/43939 [65:18:40<5:28:47, 5.70s/it]
92%|█████████▏| 40480/43939 [65:18:46<5:28:55, 5.71s/it]
92%|█████████▏| 40481/43939 [65:18:52<5:28:43, 5.70s/it]
92%|█████████▏| 40482/43939 [65:18:57<5:28:02, 5.69s/it]
92%|█████████▏| 40483/43939 [65:19:03<5:28:25, 5.70s/it]
92%|█████████▏| 40484/43939 [65:19:09<5:32:42, 5.78s/it]
92%|█████████▏| 40485/43939 [65:19:15<5:31:00, 5.75s/it]
92%|█████████▏| 40486/43939 [65:19:20<5:30:08, 5.74s/it]
92%|█████████▏| 40487/43939 [65:19:26<5:29:58, 5.74s/it]
92%|█████████▏| 40488/43939 [65:19:32<5:29:18, 5.73s/it]
92%|█████████▏| 40489/43939 [65:19:38<5:28:21, 5.71s/it]
92%|█████████▏| 40490/43939 [65:19:43<5:27:37, 5.70s/it]
92%|█████████▏| 40491/43939 [65:19:49<5:27:30, 5.70s/it]
92%|█████████▏| 40492/43939 [65:19:55<5:27:24, 5.70s/it]
92%|█████████▏| 40493/43939 [65:20:00<5:26:55, 5.69s/it]
92%|█████████▏| 40494/43939 [65:20:06<5:27:21, 5.70s/it]
92%|█████████▏| 40495/43939 [65:20:12<5:27:15, 5.70s/it]
92%|█████████▏| 40496/43939 [65:20:18<5:28:03, 5.72s/it]
92%|█████████▏| 40497/43939 [65:20:23<5:28:00, 5.72s/it]
92%|█████████▏| 40498/43939 [65:20:29<5:27:41, 5.71s/it]
92%|█████████▏| 40499/43939 [65:20:35<5:27:12, 5.71s/it]
92%|█████████▏| 40500/43939 [65:20:40<5:26:55, 5.70s/it]
{'loss': 0.6586, 'grad_norm': 1.3387778997421265, 'learning_rate': 9.277202392218298e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 493.88, 'epoch': 3.69}
92%|█████████▏| 40500/43939 [65:20:40<5:26:55, 5.70s/it]
92%|█████████▏| 40501/43939 [65:20:46<5:27:21, 5.71s/it]
92%|█████████▏| 40502/43939 [65:20:52<5:26:32, 5.70s/it]
92%|█████████▏| 40503/43939 [65:20:57<5:25:46, 5.69s/it]
92%|█████████▏| 40504/43939 [65:21:03<5:25:22, 5.68s/it]
92%|█████████▏| 40505/43939 [65:21:09<5:25:22, 5.69s/it]
92%|█████████▏| 40506/43939 [65:21:15<5:29:45, 5.76s/it]
92%|█████████▏| 40507/43939 [65:21:20<5:27:59, 5.73s/it]
92%|█████████▏| 40508/43939 [65:21:26<5:26:26, 5.71s/it]
92%|█████████▏| 40509/43939 [65:21:32<5:25:48, 5.70s/it]
92%|█████████▏| 40510/43939 [65:21:37<5:25:46, 5.70s/it]
92%|█████████▏| 40511/43939 [65:21:43<5:25:10, 5.69s/it]
92%|█████████▏| 40512/43939 [65:21:49<5:25:01, 5.69s/it]
92%|█████████▏| 40513/43939 [65:21:54<5:24:57, 5.69s/it]
92%|█████████▏| 40514/43939 [65:22:00<5:24:51, 5.69s/it]
92%|█████████▏| 40515/43939 [65:22:06<5:25:44, 5.71s/it]
92%|█████████▏| 40516/43939 [65:22:12<5:26:12, 5.72s/it]
92%|█████████▏| 40517/43939 [65:22:17<5:26:10, 5.72s/it]
92%|█████████▏| 40518/43939 [65:22:23<5:26:26, 5.73s/it]
92%|█████████▏| 40519/43939 [65:22:29<5:26:00, 5.72s/it]
92%|█████████▏| 40520/43939 [65:22:35<5:26:12, 5.72s/it]
92%|█████████▏| 40521/43939 [65:22:40<5:26:25, 5.73s/it]
92%|█████████▏| 40522/43939 [65:22:46<5:26:29, 5.73s/it]
92%|█████████▏| 40523/43939 [65:22:52<5:25:23, 5.72s/it]
92%|█████████▏| 40524/43939 [65:22:57<5:25:26, 5.72s/it]
92%|█████████▏| 40525/43939 [65:23:03<5:24:49, 5.71s/it]
92%|█████████▏| 40526/43939 [65:23:09<5:24:13, 5.70s/it]
92%|█████████▏| 40527/43939 [65:23:15<5:29:09, 5.79s/it]
92%|█████████▏| 40528/43939 [65:23:20<5:27:16, 5.76s/it]
92%|█████████▏| 40529/43939 [65:23:26<5:26:06, 5.74s/it]
92%|█████████▏| 40530/43939 [65:23:32<5:25:36, 5.73s/it]
92%|█████████▏| 40531/43939 [65:23:38<5:25:05, 5.72s/it]
92%|█████████▏| 40532/43939 [65:23:43<5:25:15, 5.73s/it]
92%|█████████▏| 40533/43939 [65:23:49<5:25:00, 5.73s/it]
92%|█████████▏| 40534/43939 [65:23:55<5:24:59, 5.73s/it]
92%|█████████▏| 40535/43939 [65:24:00<5:24:48, 5.73s/it]
92%|█████████▏| 40536/43939 [65:24:06<5:23:52, 5.71s/it]
92%|█████████▏| 40537/43939 [65:24:12<5:23:21, 5.70s/it]
92%|█████████▏| 40538/43939 [65:24:18<5:23:14, 5.70s/it]
92%|█████████▏| 40539/43939 [65:24:23<5:23:44, 5.71s/it]
92%|█████████▏| 40540/43939 [65:24:29<5:23:41, 5.71s/it]
92%|█████████▏| 40541/43939 [65:24:35<5:24:19, 5.73s/it]
92%|█████████▏| 40542/43939 [65:24:40<5:23:31, 5.71s/it]
92%|█████████▏| 40543/43939 [65:24:46<5:23:18, 5.71s/it]
92%|█████████▏| 40544/43939 [65:24:52<5:23:12, 5.71s/it]
92%|█████████▏| 40545/43939 [65:24:58<5:22:56, 5.71s/it]
92%|█████████▏| 40546/43939 [65:25:03<5:22:33, 5.70s/it]
92%|█████████▏| 40547/43939 [65:25:09<5:21:57, 5.69s/it]
92%|█████████▏| 40548/43939 [65:25:15<5:22:41, 5.71s/it]
92%|█████████▏| 40549/43939 [65:25:21<5:26:48, 5.78s/it]
92%|█████████▏| 40550/43939 [65:25:26<5:24:52, 5.75s/it]
{'loss': 0.6548, 'grad_norm': 1.9285058975219727, 'learning_rate': 9.011096431996352e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 504.07, 'epoch': 3.69}
92%|█████████▏| 40550/43939 [65:25:26<5:24:52, 5.75s/it]
92%|█████████▏| 40551/43939 [65:25:32<5:23:27, 5.73s/it]
92%|█████████▏| 40552/43939 [65:25:38<5:23:05, 5.72s/it]
92%|█████████▏| 40553/43939 [65:25:43<5:21:59, 5.71s/it]
92%|█████████▏| 40554/43939 [65:25:49<5:22:01, 5.71s/it]
92%|█████████▏| 40555/43939 [65:25:55<5:21:03, 5.69s/it]
92%|█████████▏| 40556/43939 [65:26:00<5:21:35, 5.70s/it]
92%|█████████▏| 40557/43939 [65:26:06<5:21:16, 5.70s/it]
92%|█████████▏| 40558/43939 [65:26:12<5:21:13, 5.70s/it]
92%|█████████▏| 40559/43939 [65:26:18<5:21:13, 5.70s/it]
92%|█████████▏| 40560/43939 [65:26:23<5:20:37, 5.69s/it]
92%|█████████▏| 40561/43939 [65:26:29<5:21:51, 5.72s/it]
92%|█████████▏| 40562/43939 [65:26:35<5:22:15, 5.73s/it]
92%|█████████▏| 40563/43939 [65:26:40<5:21:19, 5.71s/it]
92%|█████████▏| 40564/43939 [65:26:46<5:20:53, 5.70s/it]
92%|█████████▏| 40565/43939 [65:26:52<5:20:16, 5.70s/it]
92%|█████████▏| 40566/43939 [65:26:58<5:21:35, 5.72s/it]
92%|█████████▏| 40567/43939 [65:27:03<5:21:27, 5.72s/it]
92%|█████████▏| 40568/43939 [65:27:09<5:21:14, 5.72s/it]
92%|█████████▏| 40569/43939 [65:27:15<5:22:13, 5.74s/it]
92%|█████████▏| 40570/43939 [65:27:21<5:25:04, 5.79s/it]
92%|█████████▏| 40571/43939 [65:27:26<5:23:00, 5.75s/it]
92%|█████████▏| 40572/43939 [65:27:32<5:22:03, 5.74s/it]
92%|█████████▏| 40573/43939 [65:27:38<5:21:28, 5.73s/it]
92%|█████████▏| 40574/43939 [65:27:44<5:21:40, 5.74s/it]
92%|█████████▏| 40575/43939 [65:27:49<5:20:29, 5.72s/it]
92%|█████████▏| 40576/43939 [65:27:55<5:19:23, 5.70s/it]
92%|█████████▏| 40577/43939 [65:28:01<5:19:29, 5.70s/it]
92%|█████████▏| 40578/43939 [65:28:06<5:18:45, 5.69s/it]
92%|█████████▏| 40579/43939 [65:28:12<5:20:07, 5.72s/it]
92%|█████████▏| 40580/43939 [65:28:18<5:19:21, 5.70s/it]
92%|█████████▏| 40581/43939 [65:28:23<5:19:22, 5.71s/it]
92%|█████████▏| 40582/43939 [65:28:29<5:19:52, 5.72s/it]
92%|█████████▏| 40583/43939 [65:28:35<5:19:31, 5.71s/it]
92%|█████████▏| 40584/43939 [65:28:41<5:19:23, 5.71s/it]
92%|█████████▏| 40585/43939 [65:28:46<5:19:07, 5.71s/it]
92%|█████████▏| 40586/43939 [65:28:52<5:18:24, 5.70s/it]
92%|█████████▏| 40587/43939 [65:28:58<5:19:07, 5.71s/it]
92%|█████████▏| 40588/43939 [65:29:03<5:18:30, 5.70s/it]
92%|█████████▏| 40589/43939 [65:29:09<5:18:05, 5.70s/it]
92%|█████████▏| 40590/43939 [65:29:15<5:17:46, 5.69s/it]
92%|█████████▏| 40591/43939 [65:29:20<5:18:48, 5.71s/it]
92%|█████████▏| 40592/43939 [65:29:26<5:22:36, 5.78s/it]
92%|█████████▏| 40593/43939 [65:29:32<5:20:41, 5.75s/it]
92%|█████████▏| 40594/43939 [65:29:38<5:20:04, 5.74s/it]
92%|█████████▏| 40595/43939 [65:29:44<5:19:14, 5.73s/it]
92%|█████████▏| 40596/43939 [65:29:49<5:18:45, 5.72s/it]
92%|█████████▏| 40597/43939 [65:29:55<5:18:58, 5.73s/it]
92%|█████████▏| 40598/43939 [65:30:01<5:18:34, 5.72s/it]
92%|█████████▏| 40599/43939 [65:30:06<5:17:37, 5.71s/it]
92%|█████████▏| 40600/43939 [65:30:12<5:17:07, 5.70s/it]
{'loss': 0.6782, 'grad_norm': 1.5096101760864258, 'learning_rate': 8.748792638044289e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 523.39, 'epoch': 3.7}
92%|█████████▏| 40600/43939 [65:30:12<5:17:07, 5.70s/it]
92%|█████████▏| 40601/43939 [65:30:18<5:17:23, 5.71s/it]
92%|█████████▏| 40602/43939 [65:30:23<5:17:20, 5.71s/it]
92%|█████████▏| 40603/43939 [65:30:29<5:17:29, 5.71s/it]
92%|█████████▏| 40604/43939 [65:30:35<5:16:31, 5.69s/it]
92%|█████████▏| 40605/43939 [65:30:41<5:16:41, 5.70s/it]
92%|█████████▏| 40606/43939 [65:30:46<5:17:44, 5.72s/it]
92%|█████████▏| 40607/43939 [65:30:52<5:17:00, 5.71s/it]
92%|█████████▏| 40608/43939 [65:30:58<5:17:07, 5.71s/it]
92%|█████████▏| 40609/43939 [65:31:03<5:17:28, 5.72s/it]
92%|█████████▏| 40610/43939 [65:31:09<5:16:56, 5.71s/it]
92%|█████████▏| 40611/43939 [65:31:15<5:16:03, 5.70s/it]
92%|█████████▏| 40612/43939 [65:31:21<5:16:30, 5.71s/it]
92%|█████████▏| 40613/43939 [65:31:26<5:20:18, 5.78s/it]
92%|█████████▏| 40614/43939 [65:31:32<5:19:13, 5.76s/it]
92%|█████████▏| 40615/43939 [65:31:38<5:18:00, 5.74s/it]
92%|█████████▏| 40616/43939 [65:31:44<5:17:40, 5.74s/it]
92%|█████████▏| 40617/43939 [65:31:49<5:18:08, 5.75s/it]
92%|█████████▏| 40618/43939 [65:31:55<5:17:08, 5.73s/it]
92%|█████████▏| 40619/43939 [65:32:01<5:16:13, 5.71s/it]
92%|█████████▏| 40620/43939 [65:32:06<5:16:10, 5.72s/it]
92%|█████████▏| 40621/43939 [65:32:12<5:15:46, 5.71s/it]
92%|█████████▏| 40622/43939 [65:32:18<5:15:29, 5.71s/it]
92%|█████████▏| 40623/43939 [65:32:24<5:15:26, 5.71s/it]
92%|█████████▏| 40624/43939 [65:32:29<5:15:11, 5.70s/it]
92%|█████████▏| 40625/43939 [65:32:35<5:14:55, 5.70s/it]
92%|█████████▏| 40626/43939 [65:32:41<5:14:06, 5.69s/it]
92%|█████████▏| 40627/43939 [65:32:46<5:13:57, 5.69s/it]
92%|█████████▏| 40628/43939 [65:32:52<5:14:25, 5.70s/it]
92%|█████████▏| 40629/43939 [65:32:58<5:13:45, 5.69s/it]
92%|█████████▏| 40630/43939 [65:33:03<5:13:29, 5.68s/it]
92%|█████████▏| 40631/43939 [65:33:09<5:13:12, 5.68s/it]
92%|█████████▏| 40632/43939 [65:33:15<5:13:57, 5.70s/it]
92%|█████████▏| 40633/43939 [65:33:21<5:14:28, 5.71s/it]
92%|█████████▏| 40634/43939 [65:33:26<5:14:54, 5.72s/it]
92%|█████████▏| 40635/43939 [65:33:32<5:18:27, 5.78s/it]
92%|█████████▏| 40636/43939 [65:33:38<5:17:18, 5.76s/it]
92%|█████████▏| 40637/43939 [65:33:44<5:15:42, 5.74s/it]
92%|█████████▏| 40638/43939 [65:33:49<5:14:28, 5.72s/it]
92%|█████████▏| 40639/43939 [65:33:55<5:14:19, 5.72s/it]
92%|█████████▏| 40640/43939 [65:34:01<5:13:59, 5.71s/it]
92%|█████████▏| 40641/43939 [65:34:06<5:14:25, 5.72s/it]
92%|█████████▏| 40642/43939 [65:34:12<5:14:48, 5.73s/it]
92%|█████████▏| 40643/43939 [65:34:18<5:13:56, 5.72s/it]
93%|█████████▎| 40644/43939 [65:34:24<5:13:30, 5.71s/it]
93%|█████████▎| 40645/43939 [65:34:29<5:13:15, 5.71s/it]
93%|█████████▎| 40646/43939 [65:34:35<5:12:58, 5.70s/it]
93%|█████████▎| 40647/43939 [65:34:41<5:12:57, 5.70s/it]
93%|█████████▎| 40648/43939 [65:34:46<5:13:15, 5.71s/it]
93%|█████████▎| 40649/43939 [65:34:52<5:13:37, 5.72s/it]
93%|█████████▎| 40650/43939 [65:34:58<5:13:21, 5.72s/it]
{'loss': 0.6532, 'grad_norm': 1.5004130601882935, 'learning_rate': 8.490295148820965e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.85, 'epoch': 3.7}
93%|█████████▎| 40650/43939 [65:34:58<5:13:21, 5.72s/it]
93%|█████████▎| 40651/43939 [65:35:04<5:12:53, 5.71s/it]
93%|█████████▎| 40652/43939 [65:35:09<5:13:22, 5.72s/it]
93%|█████████▎| 40653/43939 [65:35:15<5:12:39, 5.71s/it]
93%|█████████▎| 40654/43939 [65:35:21<5:12:29, 5.71s/it]
93%|█████████▎| 40655/43939 [65:35:26<5:12:09, 5.70s/it]
93%|█████████▎| 40656/43939 [65:35:32<5:15:46, 5.77s/it]
93%|█████████▎| 40657/43939 [65:35:38<5:14:58, 5.76s/it]
93%|█████████▎| 40658/43939 [65:35:44<5:13:52, 5.74s/it]
93%|█████████▎| 40659/43939 [65:35:49<5:12:17, 5.71s/it]
93%|█████████▎| 40660/43939 [65:35:55<5:11:35, 5.70s/it]
93%|█████████▎| 40661/43939 [65:36:01<5:11:18, 5.70s/it]
93%|█████████▎| 40662/43939 [65:36:06<5:11:53, 5.71s/it]
93%|█████████▎| 40663/43939 [65:36:12<5:11:57, 5.71s/it]
93%|█████████▎| 40664/43939 [65:36:18<5:11:51, 5.71s/it]
93%|█████████▎| 40665/43939 [65:36:24<5:11:06, 5.70s/it]
93%|█████████▎| 40666/43939 [65:36:29<5:11:46, 5.72s/it]
93%|█████████▎| 40667/43939 [65:36:35<5:10:59, 5.70s/it]
93%|█████████▎| 40668/43939 [65:36:41<5:10:19, 5.69s/it]
93%|█████████▎| 40669/43939 [65:36:46<5:10:10, 5.69s/it]
93%|█████████▎| 40670/43939 [65:36:52<5:10:02, 5.69s/it]
93%|█████████▎| 40671/43939 [65:36:58<5:10:17, 5.70s/it]
93%|█████████▎| 40672/43939 [65:37:03<5:10:21, 5.70s/it]
93%|█████████▎| 40673/43939 [65:37:09<5:09:39, 5.69s/it]
93%|█████████▎| 40674/43939 [65:37:15<5:09:20, 5.68s/it]
93%|█████████▎| 40675/43939 [65:37:20<5:09:25, 5.69s/it]
93%|█████████▎| 40676/43939 [65:37:26<5:09:10, 5.69s/it]
93%|█████████▎| 40677/43939 [65:37:32<5:09:09, 5.69s/it]
93%|█████████▎| 40678/43939 [65:37:38<5:12:58, 5.76s/it]
93%|█████████▎| 40679/43939 [65:37:43<5:12:05, 5.74s/it]
93%|█████████▎| 40680/43939 [65:37:49<5:11:12, 5.73s/it]
93%|█████████▎| 40681/43939 [65:37:55<5:10:38, 5.72s/it]
93%|█████████▎| 40682/43939 [65:38:01<5:10:10, 5.71s/it]
93%|█████████▎| 40683/43939 [65:38:06<5:10:28, 5.72s/it]
93%|█████████▎| 40684/43939 [65:38:12<5:09:51, 5.71s/it]
93%|█████████▎| 40685/43939 [65:38:18<5:10:51, 5.73s/it]
93%|█████████▎| 40686/43939 [65:38:24<5:10:51, 5.73s/it]
93%|█████████▎| 40687/43939 [65:38:29<5:10:29, 5.73s/it]
93%|█████████▎| 40688/43939 [65:38:35<5:09:56, 5.72s/it]
93%|█████████▎| 40689/43939 [65:38:41<5:09:34, 5.72s/it]
93%|█████████▎| 40690/43939 [65:38:46<5:08:53, 5.70s/it]
93%|█████████▎| 40691/43939 [65:38:52<5:08:38, 5.70s/it]
93%|█████████▎| 40692/43939 [65:38:58<5:08:53, 5.71s/it]
93%|█████████▎| 40693/43939 [65:39:03<5:08:38, 5.70s/it]
93%|█████████▎| 40694/43939 [65:39:09<5:08:28, 5.70s/it]
93%|█████████▎| 40695/43939 [65:39:15<5:09:40, 5.73s/it]
93%|█████████▎| 40696/43939 [65:39:21<5:09:09, 5.72s/it]
93%|█████████▎| 40697/43939 [65:39:26<5:09:00, 5.72s/it]
93%|█████████▎| 40698/43939 [65:39:32<5:08:35, 5.71s/it]
93%|█████████▎| 40699/43939 [65:39:38<5:12:30, 5.79s/it]
93%|█████████▎| 40700/43939 [65:39:44<5:10:51, 5.76s/it]
{'loss': 0.6537, 'grad_norm': 1.642987608909607, 'learning_rate': 8.235608042731685e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.68, 'epoch': 3.71}
93%|█████████▎| 40700/43939 [65:39:44<5:10:51, 5.76s/it]
93%|█████████▎| 40701/43939 [65:39:49<5:09:58, 5.74s/it]
93%|█████████▎| 40702/43939 [65:39:55<5:09:36, 5.74s/it]
93%|█████████▎| 40703/43939 [65:40:01<5:08:31, 5.72s/it]
93%|█████████▎| 40704/43939 [65:40:06<5:08:00, 5.71s/it]
93%|█████████▎| 40705/43939 [65:40:12<5:07:58, 5.71s/it]
93%|█████████▎| 40706/43939 [65:40:18<5:07:48, 5.71s/it]
93%|█████████▎| 40707/43939 [65:40:24<5:07:49, 5.71s/it]
93%|█████████▎| 40708/43939 [65:40:29<5:08:05, 5.72s/it]
93%|█████████▎| 40709/43939 [65:40:35<5:07:01, 5.70s/it]
93%|█████████▎| 40710/43939 [65:40:41<5:06:57, 5.70s/it]
93%|█████████▎| 40711/43939 [65:40:46<5:06:52, 5.70s/it]
93%|█████████▎| 40712/43939 [65:40:52<5:06:22, 5.70s/it]
93%|█████████▎| 40713/43939 [65:40:58<5:06:33, 5.70s/it]
93%|█████████▎| 40714/43939 [65:41:04<5:07:11, 5.72s/it]
93%|█████████▎| 40715/43939 [65:41:09<5:06:50, 5.71s/it]
93%|█████████▎| 40716/43939 [65:41:15<5:06:20, 5.70s/it]
93%|█████████▎| 40717/43939 [65:41:21<5:07:04, 5.72s/it]
93%|█████████▎| 40718/43939 [65:41:26<5:06:46, 5.71s/it]
93%|█████████▎| 40719/43939 [65:41:32<5:06:30, 5.71s/it]
93%|█████████▎| 40720/43939 [65:41:38<5:05:49, 5.70s/it]
93%|█████████▎| 40721/43939 [65:41:44<5:09:31, 5.77s/it]
93%|█████████▎| 40722/43939 [65:41:49<5:08:13, 5.75s/it]
93%|█████████▎| 40723/43939 [65:41:55<5:07:31, 5.74s/it]
93%|█████████▎| 40724/43939 [65:42:01<5:06:50, 5.73s/it]
93%|█████████▎| 40725/43939 [65:42:07<5:06:32, 5.72s/it]
93%|█████████▎| 40726/43939 [65:42:12<5:05:33, 5.71s/it]
93%|█████████▎| 40727/43939 [65:42:18<5:05:26, 5.71s/it]
93%|█████████▎| 40728/43939 [65:42:24<5:06:13, 5.72s/it]
93%|█████████▎| 40729/43939 [65:42:29<5:05:47, 5.72s/it]
93%|█████████▎| 40730/43939 [65:42:35<5:06:05, 5.72s/it]
93%|█████████▎| 40731/43939 [65:42:41<5:06:24, 5.73s/it]
93%|█████████▎| 40732/43939 [65:42:47<5:05:44, 5.72s/it]
93%|█████████▎| 40733/43939 [65:42:52<5:05:15, 5.71s/it]
93%|█████████▎| 40734/43939 [65:42:58<5:05:08, 5.71s/it]
93%|█████████▎| 40735/43939 [65:43:04<5:05:42, 5.72s/it]
93%|█████████▎| 40736/43939 [65:43:09<5:05:36, 5.72s/it]
93%|█████████▎| 40737/43939 [65:43:15<5:06:18, 5.74s/it]
93%|█████████▎| 40738/43939 [65:43:21<5:05:51, 5.73s/it]
93%|█████████▎| 40739/43939 [65:43:27<5:06:00, 5.74s/it]
93%|█████████▎| 40740/43939 [65:43:32<5:05:27, 5.73s/it]
93%|█████████▎| 40741/43939 [65:43:38<5:05:36, 5.73s/it]
93%|█████████▎| 40742/43939 [65:43:44<5:05:17, 5.73s/it]
93%|█████████▎| 40743/43939 [65:43:50<5:08:21, 5.79s/it]
93%|█████████▎| 40744/43939 [65:43:55<5:06:11, 5.75s/it]
93%|█████████▎| 40745/43939 [65:44:01<5:04:30, 5.72s/it]
93%|█████████▎| 40746/43939 [65:44:07<5:03:49, 5.71s/it]
93%|█████████▎| 40747/43939 [65:44:13<5:04:14, 5.72s/it]
93%|█████████▎| 40748/43939 [65:44:18<5:03:46, 5.71s/it]
93%|█████████▎| 40749/43939 [65:44:24<5:04:11, 5.72s/it]
93%|█████████▎| 40750/43939 [65:44:30<5:04:21, 5.73s/it]
{'loss': 0.6464, 'grad_norm': 1.7471836805343628, 'learning_rate': 7.984735338064043e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.71, 'epoch': 3.71}
93%|█████████▎| 40750/43939 [65:44:30<5:04:21, 5.73s/it]
93%|█████████▎| 40751/43939 [65:44:35<5:04:06, 5.72s/it]
93%|█████████▎| 40752/43939 [65:44:41<5:04:21, 5.73s/it]
93%|█████████▎| 40753/43939 [65:44:47<5:04:01, 5.73s/it]
93%|█████████▎| 40754/43939 [65:44:53<5:03:22, 5.71s/it]
93%|█████████▎| 40755/43939 [65:44:58<5:02:28, 5.70s/it]
93%|█████████▎| 40756/43939 [65:45:04<5:02:37, 5.70s/it]
93%|█████████▎| 40757/43939 [65:45:10<5:02:29, 5.70s/it]
93%|█████████▎| 40758/43939 [65:45:15<5:02:32, 5.71s/it]
93%|█████████▎| 40759/43939 [65:45:21<5:02:08, 5.70s/it]
93%|█████████▎| 40760/43939 [65:45:27<5:03:27, 5.73s/it]
93%|█████████▎| 40761/43939 [65:45:33<5:03:24, 5.73s/it]
93%|█████████▎| 40762/43939 [65:45:38<5:02:44, 5.72s/it]
93%|█████████▎| 40763/43939 [65:45:44<5:02:11, 5.71s/it]
93%|█████████▎| 40764/43939 [65:45:50<5:05:44, 5.78s/it]
93%|█████████▎| 40765/43939 [65:45:56<5:04:24, 5.75s/it]
93%|█████████▎| 40766/43939 [65:46:01<5:03:28, 5.74s/it]
93%|█████████▎| 40767/43939 [65:46:07<5:03:03, 5.73s/it]
93%|█████████▎| 40768/43939 [65:46:13<5:02:58, 5.73s/it]
93%|█████████▎| 40769/43939 [65:46:19<5:03:11, 5.74s/it]
93%|█████████▎| 40770/43939 [65:46:24<5:02:23, 5.73s/it]
93%|█████████▎| 40771/43939 [65:46:30<5:01:39, 5.71s/it]
93%|█████████▎| 40772/43939 [65:46:36<5:01:31, 5.71s/it]
93%|█████████▎| 40773/43939 [65:46:41<5:01:51, 5.72s/it]
93%|█████████▎| 40774/43939 [65:46:47<5:01:03, 5.71s/it]
93%|█████████▎| 40775/43939 [65:46:53<5:01:13, 5.71s/it]
93%|█████████▎| 40776/43939 [65:46:59<5:01:57, 5.73s/it]
93%|█████████▎| 40777/43939 [65:47:04<5:01:11, 5.72s/it]
93%|█████████▎| 40778/43939 [65:47:10<5:00:52, 5.71s/it]
93%|█████████▎| 40779/43939 [65:47:16<5:00:49, 5.71s/it]
93%|█████████▎| 40780/43939 [65:47:21<5:00:48, 5.71s/it]
93%|█████████▎| 40781/43939 [65:47:27<5:00:37, 5.71s/it]
93%|█████████▎| 40782/43939 [65:47:33<5:00:23, 5.71s/it]
93%|█████████▎| 40783/43939 [65:47:38<5:00:28, 5.71s/it]
93%|█████████▎| 40784/43939 [65:47:44<5:00:28, 5.71s/it]
93%|█████████▎| 40785/43939 [65:47:50<5:03:59, 5.78s/it]
93%|█████████▎| 40786/43939 [65:47:56<5:03:11, 5.77s/it]
93%|█████████▎| 40787/43939 [65:48:02<5:02:36, 5.76s/it]
93%|█████████▎| 40788/43939 [65:48:07<5:01:41, 5.74s/it]
93%|█████████▎| 40789/43939 [65:48:13<5:00:54, 5.73s/it]
93%|█████████▎| 40790/43939 [65:48:19<5:00:23, 5.72s/it]
93%|█████████▎| 40791/43939 [65:48:24<4:59:55, 5.72s/it]
93%|█████████▎| 40792/43939 [65:48:30<4:59:37, 5.71s/it]
93%|█████████▎| 40793/43939 [65:48:36<4:59:47, 5.72s/it]
93%|█████████▎| 40794/43939 [65:48:42<4:59:39, 5.72s/it]
93%|█████████▎| 40795/43939 [65:48:47<4:58:59, 5.71s/it]
93%|█████████▎| 40796/43939 [65:48:53<4:59:31, 5.72s/it]
93%|█████████▎| 40797/43939 [65:48:59<4:59:09, 5.71s/it]
93%|█████████▎| 40798/43939 [65:49:04<4:58:18, 5.70s/it]
93%|█████████▎| 40799/43939 [65:49:10<4:58:47, 5.71s/it]
93%|█████████▎| 40800/43939 [65:49:16<4:59:40, 5.73s/it]
{'loss': 0.6789, 'grad_norm': 1.4065643548965454, 'learning_rate': 7.737680992924489e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 517.25, 'epoch': 3.71}
93%|█████████▎| 40800/43939 [65:49:16<4:59:40, 5.73s/it]
93%|█████████▎| 40801/43939 [65:49:22<4:59:28, 5.73s/it]
93%|█████████▎| 40802/43939 [65:49:27<4:59:15, 5.72s/it]
93%|█████████▎| 40803/43939 [65:49:33<4:58:33, 5.71s/it]
93%|█████████▎| 40804/43939 [65:49:39<4:57:51, 5.70s/it]
93%|█████████▎| 40805/43939 [65:49:44<4:57:53, 5.70s/it]
93%|█████████▎| 40806/43939 [65:49:50<4:57:48, 5.70s/it]
93%|█████████▎| 40807/43939 [65:49:56<5:01:50, 5.78s/it]
93%|█████████▎| 40808/43939 [65:50:02<5:00:59, 5.77s/it]
93%|█████████▎| 40809/43939 [65:50:07<5:00:02, 5.75s/it]
93%|█████████▎| 40810/43939 [65:50:13<4:59:19, 5.74s/it]
93%|█████████▎| 40811/43939 [65:50:19<4:58:33, 5.73s/it]
93%|█████████▎| 40812/43939 [65:50:25<4:57:14, 5.70s/it]
93%|█████████▎| 40813/43939 [65:50:30<4:56:38, 5.69s/it]
93%|█████████▎| 40814/43939 [65:50:36<4:57:13, 5.71s/it]
93%|█████████▎| 40815/43939 [65:50:42<4:57:10, 5.71s/it]
93%|█████████▎| 40816/43939 [65:50:47<4:57:11, 5.71s/it]
93%|█████████▎| 40817/43939 [65:50:53<4:57:15, 5.71s/it]
93%|█████████▎| 40818/43939 [65:50:59<4:56:58, 5.71s/it]
93%|█████████▎| 40819/43939 [65:51:04<4:56:38, 5.70s/it]
93%|█████████▎| 40820/43939 [65:51:10<4:57:00, 5.71s/it]
93%|█████████▎| 40821/43939 [65:51:16<4:57:19, 5.72s/it]
93%|█████████▎| 40822/43939 [65:51:22<4:56:30, 5.71s/it]
93%|█████████▎| 40823/43939 [65:51:27<4:56:20, 5.71s/it]
93%|█████████▎| 40824/43939 [65:51:33<4:56:59, 5.72s/it]
93%|█████████▎| 40825/43939 [65:51:39<4:56:48, 5.72s/it]
93%|█████████▎| 40826/43939 [65:51:44<4:56:07, 5.71s/it]
93%|█████████▎| 40827/43939 [65:51:50<4:55:26, 5.70s/it]
93%|█████████▎| 40828/43939 [65:51:56<4:55:23, 5.70s/it]
93%|█████████▎| 40829/43939 [65:52:02<4:59:32, 5.78s/it]
93%|█████████▎| 40830/43939 [65:52:08<4:58:42, 5.76s/it]
93%|█████████▎| 40831/43939 [65:52:13<4:57:19, 5.74s/it]
93%|█████████▎| 40832/43939 [65:52:19<4:56:23, 5.72s/it]
93%|█████████▎| 40833/43939 [65:52:25<4:56:31, 5.73s/it]
93%|█████████▎| 40834/43939 [65:52:30<4:56:04, 5.72s/it]
93%|█████████▎| 40835/43939 [65:52:36<4:56:17, 5.73s/it]
93%|█████████▎| 40836/43939 [65:52:42<4:56:00, 5.72s/it]
93%|█████████▎| 40837/43939 [65:52:48<4:55:38, 5.72s/it]
93%|█████████▎| 40838/43939 [65:52:53<4:55:26, 5.72s/it]
93%|█████████▎| 40839/43939 [65:52:59<4:55:36, 5.72s/it]
93%|█████████▎| 40840/43939 [65:53:05<4:55:03, 5.71s/it]
93%|█████████▎| 40841/43939 [65:53:10<4:55:10, 5.72s/it]
93%|█████████▎| 40842/43939 [65:53:16<4:55:38, 5.73s/it]
93%|█████████▎| 40843/43939 [65:53:22<4:55:22, 5.72s/it]
93%|█████████▎| 40844/43939 [65:53:28<4:54:44, 5.71s/it]
93%|█████████▎| 40845/43939 [65:53:33<4:54:33, 5.71s/it]
93%|█████████▎| 40846/43939 [65:53:39<4:54:50, 5.72s/it]
93%|█████████▎| 40847/43939 [65:53:45<4:55:33, 5.74s/it]
93%|█████████▎| 40848/43939 [65:53:50<4:54:44, 5.72s/it]
93%|█████████▎| 40849/43939 [65:53:56<4:54:42, 5.72s/it]
93%|█████████▎| 40850/43939 [65:54:02<4:57:47, 5.78s/it]
{'loss': 0.6507, 'grad_norm': 1.5657329559326172, 'learning_rate': 7.494448905175922e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.29, 'epoch': 3.72}
93%|█████████▎| 40850/43939 [65:54:02<4:57:47, 5.78s/it]
93%|█████████▎| 40851/43939 [65:54:08<4:56:34, 5.76s/it]
93%|█████████▎| 40852/43939 [65:54:14<4:55:36, 5.75s/it]
93%|█████████▎| 40853/43939 [65:54:19<4:54:12, 5.72s/it]
93%|█████████▎| 40854/43939 [65:54:25<4:53:39, 5.71s/it]
93%|█████████▎| 40855/43939 [65:54:31<4:54:19, 5.73s/it]
93%|█████████▎| 40856/43939 [65:54:36<4:54:19, 5.73s/it]
93%|█████████▎| 40857/43939 [65:54:42<4:54:07, 5.73s/it]
93%|█████████▎| 40858/43939 [65:54:48<4:53:42, 5.72s/it]
93%|█████████▎| 40859/43939 [65:54:54<4:53:34, 5.72s/it]
93%|█████████▎| 40860/43939 [65:54:59<4:53:24, 5.72s/it]
93%|█████████▎| 40861/43939 [65:55:05<4:52:39, 5.70s/it]
93%|█████████▎| 40862/43939 [65:55:11<4:52:51, 5.71s/it]
93%|█████████▎| 40863/43939 [65:55:16<4:52:52, 5.71s/it]
93%|█████████▎| 40864/43939 [65:55:22<4:52:52, 5.71s/it]
93%|█████████▎| 40865/43939 [65:55:28<4:53:23, 5.73s/it]
93%|█████████▎| 40866/43939 [65:55:34<4:52:59, 5.72s/it]
93%|█████████▎| 40867/43939 [65:55:39<4:53:08, 5.73s/it]
93%|█████████▎| 40868/43939 [65:55:45<4:52:36, 5.72s/it]
93%|█████████▎| 40869/43939 [65:55:51<4:53:19, 5.73s/it]
93%|█████████▎| 40870/43939 [65:55:56<4:52:55, 5.73s/it]
93%|█████████▎| 40871/43939 [65:56:02<4:56:20, 5.80s/it]
93%|█████████▎| 40872/43939 [65:56:08<4:54:51, 5.77s/it]
93%|█████████▎| 40873/43939 [65:56:14<4:54:03, 5.75s/it]
93%|█████████▎| 40874/43939 [65:56:20<4:53:36, 5.75s/it]
93%|█████████▎| 40875/43939 [65:56:25<4:53:01, 5.74s/it]
93%|█████████▎| 40876/43939 [65:56:31<4:52:11, 5.72s/it]
93%|█████████▎| 40877/43939 [65:56:37<4:52:16, 5.73s/it]
93%|█████████▎| 40878/43939 [65:56:42<4:51:25, 5.71s/it]
93%|█████████▎| 40879/43939 [65:56:48<4:51:05, 5.71s/it]
93%|█████████▎| 40880/43939 [65:56:54<4:51:54, 5.73s/it]
93%|█████████▎| 40881/43939 [65:57:00<4:51:49, 5.73s/it]
93%|█████████▎| 40882/43939 [65:57:05<4:51:43, 5.73s/it]
93%|█████████▎| 40883/43939 [65:57:11<4:51:39, 5.73s/it]
93%|█████████▎| 40884/43939 [65:57:17<4:52:15, 5.74s/it]
93%|█████████▎| 40885/43939 [65:57:23<4:51:47, 5.73s/it]
93%|█████████▎| 40886/43939 [65:57:28<4:51:38, 5.73s/it]
93%|█████████▎| 40887/43939 [65:57:34<4:51:31, 5.73s/it]
93%|█████████▎| 40888/43939 [65:57:40<4:51:03, 5.72s/it]
93%|█████████▎| 40889/43939 [65:57:45<4:50:26, 5.71s/it]
93%|█████████▎| 40890/43939 [65:57:51<4:50:32, 5.72s/it]
93%|█████████▎| 40891/43939 [65:57:57<4:50:01, 5.71s/it]
93%|█████████▎| 40892/43939 [65:58:02<4:49:52, 5.71s/it]
93%|█████████▎| 40893/43939 [65:58:08<4:53:09, 5.77s/it]
93%|█████████▎| 40894/43939 [65:58:14<4:51:59, 5.75s/it]
93%|█████████▎| 40895/43939 [65:58:20<4:52:04, 5.76s/it]
93%|█████████▎| 40896/43939 [65:58:26<4:51:22, 5.75s/it]
93%|█████████▎| 40897/43939 [65:58:31<4:51:35, 5.75s/it]
93%|█████████▎| 40898/43939 [65:58:37<4:50:36, 5.73s/it]
93%|█████████▎| 40899/43939 [65:58:43<4:49:26, 5.71s/it]
93%|█████████▎| 40900/43939 [65:58:48<4:48:28, 5.70s/it]
{'loss': 0.6466, 'grad_norm': 1.9447294473648071, 'learning_rate': 7.255042912376108e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.68, 'epoch': 3.72}
93%|█████████▎| 40900/43939 [65:58:48<4:48:28, 5.70s/it]
93%|█████████▎| 40901/43939 [65:58:54<4:49:04, 5.71s/it]
93%|█████████▎| 40902/43939 [65:59:00<4:48:44, 5.70s/it]
93%|█████████▎| 40903/43939 [65:59:06<4:48:23, 5.70s/it]
93%|█████████▎| 40904/43939 [65:59:11<4:48:00, 5.69s/it]
93%|█████████▎| 40905/43939 [65:59:17<4:49:15, 5.72s/it]
93%|█████████▎| 40906/43939 [65:59:23<4:49:00, 5.72s/it]
93%|█████████▎| 40907/43939 [65:59:28<4:49:26, 5.73s/it]
93%|█████████▎| 40908/43939 [65:59:34<4:48:33, 5.71s/it]
93%|█████████▎| 40909/43939 [65:59:40<4:48:40, 5.72s/it]
93%|█████████▎| 40910/43939 [65:59:46<4:48:11, 5.71s/it]
93%|█████████▎| 40911/43939 [65:59:51<4:48:35, 5.72s/it]
93%|█████████▎| 40912/43939 [65:59:57<4:48:08, 5.71s/it]
93%|█████████▎| 40913/43939 [66:00:03<4:47:31, 5.70s/it]
93%|█████████▎| 40914/43939 [66:00:08<4:47:22, 5.70s/it]
93%|█████████▎| 40915/43939 [66:00:14<4:51:05, 5.78s/it]
93%|█████████▎| 40916/43939 [66:00:20<4:50:26, 5.76s/it]
93%|█████████▎| 40917/43939 [66:00:26<4:49:09, 5.74s/it]
93%|█████████▎| 40918/43939 [66:00:31<4:48:33, 5.73s/it]
93%|█████████▎| 40919/43939 [66:00:37<4:48:32, 5.73s/it]
93%|█████████▎| 40920/43939 [66:00:43<4:47:33, 5.72s/it]
93%|█████████▎| 40921/43939 [66:00:49<4:48:12, 5.73s/it]
93%|█████████▎| 40922/43939 [66:00:54<4:47:36, 5.72s/it]
93%|█████████▎| 40923/43939 [66:01:00<4:46:42, 5.70s/it]
93%|█████████▎| 40924/43939 [66:01:06<4:46:42, 5.71s/it]
93%|█████████▎| 40925/43939 [66:01:11<4:46:23, 5.70s/it]
93%|█████████▎| 40926/43939 [66:01:17<4:46:17, 5.70s/it]
93%|█████████▎| 40927/43939 [66:01:23<4:46:16, 5.70s/it]
93%|█████████▎| 40928/43939 [66:01:28<4:46:17, 5.70s/it]
93%|█████████▎| 40929/43939 [66:01:34<4:45:58, 5.70s/it]
93%|█████████▎| 40930/43939 [66:01:40<4:45:55, 5.70s/it]
93%|█████████▎| 40931/43939 [66:01:46<4:45:42, 5.70s/it]
93%|█████████▎| 40932/43939 [66:01:51<4:45:24, 5.69s/it]
93%|█████████▎| 40933/43939 [66:01:57<4:45:28, 5.70s/it]
93%|█████████▎| 40934/43939 [66:02:03<4:45:10, 5.69s/it]
93%|█████████▎| 40935/43939 [66:02:08<4:45:24, 5.70s/it]
93%|█████████▎| 40936/43939 [66:02:14<4:48:37, 5.77s/it]
93%|█████████▎| 40937/43939 [66:02:20<4:47:11, 5.74s/it]
93%|█████████▎| 40938/43939 [66:02:26<4:46:56, 5.74s/it]
93%|█████████▎| 40939/43939 [66:02:31<4:46:31, 5.73s/it]
93%|█████████▎| 40940/43939 [66:02:37<4:46:14, 5.73s/it]
93%|█████████▎| 40941/43939 [66:02:43<4:45:41, 5.72s/it]
93%|█████████▎| 40942/43939 [66:02:49<4:45:26, 5.71s/it]
93%|█████████▎| 40943/43939 [66:02:54<4:44:33, 5.70s/it]
93%|█████████▎| 40944/43939 [66:03:00<4:44:20, 5.70s/it]
93%|█████████▎| 40945/43939 [66:03:06<4:44:04, 5.69s/it]
93%|█████████▎| 40946/43939 [66:03:11<4:44:23, 5.70s/it]
93%|█████████▎| 40947/43939 [66:03:17<4:44:13, 5.70s/it]
93%|█████████▎| 40948/43939 [66:03:23<4:43:49, 5.69s/it]
93%|█████████▎| 40949/43939 [66:03:28<4:43:55, 5.70s/it]
93%|█████████▎| 40950/43939 [66:03:34<4:44:40, 5.71s/it]
{'loss': 0.6551, 'grad_norm': 1.7239102125167847, 'learning_rate': 7.019466791717166e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 472.22, 'epoch': 3.73}
93%|█████████▎| 40950/43939 [66:03:34<4:44:40, 5.71s/it]
93%|█████████▎| 40951/43939 [66:03:40<4:44:23, 5.71s/it]
93%|█████████▎| 40952/43939 [66:03:46<4:44:34, 5.72s/it]
93%|█████████▎| 40953/43939 [66:03:51<4:44:25, 5.72s/it]
93%|█████████▎| 40954/43939 [66:03:57<4:45:01, 5.73s/it]
93%|█████████▎| 40955/43939 [66:04:03<4:44:56, 5.73s/it]
93%|█████████▎| 40956/43939 [66:04:08<4:45:01, 5.73s/it]
93%|█████████▎| 40957/43939 [66:04:14<4:48:14, 5.80s/it]
93%|█████████▎| 40958/43939 [66:04:20<4:45:59, 5.76s/it]
93%|█████████▎| 40959/43939 [66:04:26<4:44:21, 5.73s/it]
93%|█████████▎| 40960/43939 [66:04:31<4:44:02, 5.72s/it]
93%|█████████▎| 40961/43939 [66:04:37<4:43:33, 5.71s/it]
93%|█████████▎| 40962/43939 [66:04:43<4:43:01, 5.70s/it]
93%|█████████▎| 40963/43939 [66:04:49<4:43:18, 5.71s/it]
93%|█████████▎| 40964/43939 [66:04:54<4:42:43, 5.70s/it]
93%|█████████▎| 40965/43939 [66:05:00<4:42:45, 5.70s/it]
93%|█████████▎| 40966/43939 [66:05:06<4:42:33, 5.70s/it]
93%|█████████▎| 40967/43939 [66:05:11<4:42:29, 5.70s/it]
93%|█████████▎| 40968/43939 [66:05:17<4:42:50, 5.71s/it]
93%|█████████▎| 40969/43939 [66:05:23<4:43:01, 5.72s/it]
93%|█████████▎| 40970/43939 [66:05:29<4:43:09, 5.72s/it]
93%|█████████▎| 40971/43939 [66:05:34<4:44:07, 5.74s/it]
93%|█████████▎| 40972/43939 [66:05:40<4:43:24, 5.73s/it]
93%|█████████▎| 40973/43939 [66:05:46<4:42:14, 5.71s/it]
93%|█████████▎| 40974/43939 [66:05:51<4:41:59, 5.71s/it]
93%|█████████▎| 40975/43939 [66:05:57<4:41:40, 5.70s/it]
93%|█████████▎| 40976/43939 [66:06:03<4:42:35, 5.72s/it]
93%|█████████▎| 40977/43939 [66:06:09<4:42:46, 5.73s/it]
93%|█████████▎| 40978/43939 [66:06:14<4:42:44, 5.73s/it]
93%|█████████▎| 40979/43939 [66:06:20<4:45:59, 5.80s/it]
93%|█████████▎| 40980/43939 [66:06:26<4:44:42, 5.77s/it]
93%|█████████▎| 40981/43939 [66:06:32<4:43:17, 5.75s/it]
93%|█████████▎| 40982/43939 [66:06:37<4:43:25, 5.75s/it]
93%|█████████▎| 40983/43939 [66:06:43<4:42:47, 5.74s/it]
93%|█████████▎| 40984/43939 [66:06:49<4:42:08, 5.73s/it]
93%|█████████▎| 40985/43939 [66:06:55<4:41:15, 5.71s/it]
93%|█████████▎| 40986/43939 [66:07:00<4:41:41, 5.72s/it]
93%|█████████▎| 40987/43939 [66:07:06<4:41:15, 5.72s/it]
93%|█████████▎| 40988/43939 [66:07:12<4:40:42, 5.71s/it]
93%|█████████▎| 40989/43939 [66:07:17<4:40:05, 5.70s/it]
93%|█████████▎| 40990/43939 [66:07:23<4:39:58, 5.70s/it]
93%|█████████▎| 40991/43939 [66:07:29<4:39:58, 5.70s/it]
93%|█████████▎| 40992/43939 [66:07:35<4:40:35, 5.71s/it]
93%|█████████▎| 40993/43939 [66:07:40<4:40:51, 5.72s/it]
93%|█████████▎| 40994/43939 [66:07:46<4:41:17, 5.73s/it]
93%|█████████▎| 40995/43939 [66:07:52<4:40:42, 5.72s/it]
93%|█████████▎| 40996/43939 [66:07:57<4:39:32, 5.70s/it]
93%|█████████▎| 40997/43939 [66:08:03<4:39:12, 5.69s/it]
93%|█████████▎| 40998/43939 [66:08:09<4:39:18, 5.70s/it]
93%|█████████▎| 40999/43939 [66:08:14<4:39:08, 5.70s/it]
93%|█████████▎| 41000/43939 [66:08:20<4:39:21, 5.70s/it]
{'loss': 0.6464, 'grad_norm': 1.4075065851211548, 'learning_rate': 6.787724259965988e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.84, 'epoch': 3.73}
93%|█████████▎| 41000/43939 [66:08:20<4:39:21, 5.70s/it]
93%|█████████▎| 41001/43939 [66:08:26<4:42:18, 5.77s/it]
93%|█████████▎| 41002/43939 [66:08:32<4:40:57, 5.74s/it]
93%|█████████▎| 41003/43939 [66:08:37<4:40:19, 5.73s/it]
93%|█████████▎| 41004/43939 [66:08:43<4:39:38, 5.72s/it]
93%|█████████▎| 41005/43939 [66:08:49<4:39:07, 5.71s/it]
93%|█████████▎| 41006/43939 [66:08:55<4:39:39, 5.72s/it]
93%|█████████▎| 41007/43939 [66:09:00<4:39:29, 5.72s/it]
93%|█████████▎| 41008/43939 [66:09:06<4:39:25, 5.72s/it]
93%|█████████▎| 41009/43939 [66:09:12<4:39:33, 5.72s/it]
93%|█████████▎| 41010/43939 [66:09:17<4:39:14, 5.72s/it]
93%|█████████▎| 41011/43939 [66:09:23<4:39:25, 5.73s/it]
93%|█████████▎| 41012/43939 [66:09:29<4:38:27, 5.71s/it]
93%|█████████▎| 41013/43939 [66:09:35<4:38:40, 5.71s/it]
93%|█████████▎| 41014/43939 [66:09:40<4:38:00, 5.70s/it]
93%|█████████▎| 41015/43939 [66:09:46<4:37:59, 5.70s/it]
93%|█████████▎| 41016/43939 [66:09:52<4:38:19, 5.71s/it]
93%|█████████▎| 41017/43939 [66:09:57<4:38:58, 5.73s/it]
93%|█████████▎| 41018/43939 [66:10:03<4:38:32, 5.72s/it]
93%|█████████▎| 41019/43939 [66:10:09<4:38:31, 5.72s/it]
93%|█████████▎| 41020/43939 [66:10:15<4:37:25, 5.70s/it]
93%|█████████▎| 41021/43939 [66:10:20<4:37:28, 5.71s/it]
93%|█████████▎| 41022/43939 [66:10:26<4:39:49, 5.76s/it]
93%|█████████▎| 41023/43939 [66:10:32<4:38:44, 5.74s/it]
93%|█████████▎| 41024/43939 [66:10:38<4:37:48, 5.72s/it]
93%|█████████▎| 41025/43939 [66:10:43<4:37:32, 5.71s/it]
93%|█████████▎| 41026/43939 [66:10:49<4:37:37, 5.72s/it]
93%|█████████▎| 41027/43939 [66:10:55<4:37:22, 5.72s/it]
93%|█████████▎| 41028/43939 [66:11:00<4:37:22, 5.72s/it]
93%|█████████▎| 41029/43939 [66:11:06<4:38:10, 5.74s/it]
93%|█████████▎| 41030/43939 [66:11:12<4:37:38, 5.73s/it]
93%|█████████▎| 41031/43939 [66:11:18<4:37:12, 5.72s/it]
93%|█████████▎| 41032/43939 [66:11:23<4:37:07, 5.72s/it]
93%|█████████▎| 41033/43939 [66:11:29<4:36:46, 5.71s/it]
93%|█████████▎| 41034/43939 [66:11:35<4:36:28, 5.71s/it]
93%|█████████▎| 41035/43939 [66:11:40<4:36:25, 5.71s/it]
93%|█████████▎| 41036/43939 [66:11:46<4:36:54, 5.72s/it]
93%|█████████▎| 41037/43939 [66:11:52<4:37:04, 5.73s/it]
93%|█████████▎| 41038/43939 [66:11:58<4:36:28, 5.72s/it]
93%|█████████▎| 41039/43939 [66:12:03<4:35:56, 5.71s/it]
93%|█████████▎| 41040/43939 [66:12:09<4:35:27, 5.70s/it]
93%|█████████▎| 41041/43939 [66:12:15<4:34:59, 5.69s/it]
93%|█████████▎| 41042/43939 [66:12:20<4:35:10, 5.70s/it]
93%|█████████▎| 41043/43939 [66:12:26<4:34:57, 5.70s/it]
93%|█████████▎| 41044/43939 [66:12:32<4:37:52, 5.76s/it]
93%|█████████▎| 41045/43939 [66:12:38<4:37:26, 5.75s/it]
93%|█████████▎| 41046/43939 [66:12:43<4:36:50, 5.74s/it]
93%|█████████▎| 41047/43939 [66:12:49<4:36:02, 5.73s/it]
93%|█████████▎| 41048/43939 [66:12:55<4:35:34, 5.72s/it]
93%|█████████▎| 41049/43939 [66:13:00<4:35:06, 5.71s/it]
93%|█████████▎| 41050/43939 [66:13:06<4:35:14, 5.72s/it]
{'loss': 0.6604, 'grad_norm': 1.4332797527313232, 'learning_rate': 6.55981897340563e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.3, 'epoch': 3.74}
93%|█████████▎| 41050/43939 [66:13:06<4:35:14, 5.72s/it]
93%|█████████▎| 41051/43939 [66:13:12<4:35:16, 5.72s/it]
93%|█████████▎| 41052/43939 [66:13:18<4:34:54, 5.71s/it]
93%|█████████▎| 41053/43939 [66:13:23<4:35:15, 5.72s/it]
93%|█████████▎| 41054/43939 [66:13:29<4:34:27, 5.71s/it]
93%|█████████▎| 41055/43939 [66:13:35<4:35:22, 5.73s/it]
93%|█████████▎| 41056/43939 [66:13:41<4:34:35, 5.71s/it]
93%|█████████▎| 41057/43939 [66:13:46<4:34:18, 5.71s/it]
93%|█████████▎| 41058/43939 [66:13:52<4:34:07, 5.71s/it]
93%|█████████▎| 41059/43939 [66:13:58<4:34:20, 5.72s/it]
93%|█████████▎| 41060/43939 [66:14:03<4:33:36, 5.70s/it]
93%|█████████▎| 41061/43939 [66:14:09<4:33:23, 5.70s/it]
93%|█████████▎| 41062/43939 [66:14:15<4:33:22, 5.70s/it]
93%|█████████▎| 41063/43939 [66:14:20<4:33:05, 5.70s/it]
93%|█████████▎| 41064/43939 [66:14:26<4:32:30, 5.69s/it]
93%|█████████▎| 41065/43939 [66:14:32<4:36:21, 5.77s/it]
93%|█████████▎| 41066/43939 [66:14:38<4:35:08, 5.75s/it]
93%|█████████▎| 41067/43939 [66:14:43<4:34:05, 5.73s/it]
93%|█████████▎| 41068/43939 [66:14:49<4:33:57, 5.73s/it]
93%|█████████▎| 41069/43939 [66:14:55<4:33:41, 5.72s/it]
93%|█████████▎| 41070/43939 [66:15:01<4:33:43, 5.72s/it]
93%|█████████▎| 41071/43939 [66:15:06<4:34:01, 5.73s/it]
93%|█████████▎| 41072/43939 [66:15:12<4:33:28, 5.72s/it]
93%|█████████▎| 41073/43939 [66:15:18<4:33:04, 5.72s/it]
93%|█████████▎| 41074/43939 [66:15:23<4:33:26, 5.73s/it]
93%|█████████▎| 41075/43939 [66:15:29<4:32:56, 5.72s/it]
93%|█████████▎| 41076/43939 [66:15:35<4:32:29, 5.71s/it]
93%|█████████▎| 41077/43939 [66:15:41<4:31:59, 5.70s/it]
93%|█████████▎| 41078/43939 [66:15:46<4:31:42, 5.70s/it]
93%|█████████▎| 41079/43939 [66:15:52<4:31:35, 5.70s/it]
93%|█████████▎| 41080/43939 [66:15:58<4:31:48, 5.70s/it]
93%|█████████▎| 41081/43939 [66:16:03<4:31:40, 5.70s/it]
93%|█████████▎| 41082/43939 [66:16:09<4:31:27, 5.70s/it]
94%|█████████▎| 41083/43939 [66:16:15<4:30:47, 5.69s/it]
94%|█████████▎| 41084/43939 [66:16:20<4:31:15, 5.70s/it]
94%|█████████▎| 41085/43939 [66:16:26<4:31:11, 5.70s/it]
94%|█████████▎| 41086/43939 [66:16:32<4:31:10, 5.70s/it]
94%|█████████▎| 41087/43939 [66:16:38<4:33:54, 5.76s/it]
94%|█████████▎| 41088/43939 [66:16:43<4:32:40, 5.74s/it]
94%|█████████▎| 41089/43939 [66:16:49<4:32:34, 5.74s/it]
94%|█████████▎| 41090/43939 [66:16:55<4:32:10, 5.73s/it]
94%|█████████▎| 41091/43939 [66:17:01<4:31:35, 5.72s/it]
94%|█████████▎| 41092/43939 [66:17:06<4:31:21, 5.72s/it]
94%|█████████▎| 41093/43939 [66:17:12<4:30:43, 5.71s/it]
94%|█████████▎| 41094/43939 [66:17:18<4:30:10, 5.70s/it]
94%|█████████▎| 41095/43939 [66:17:23<4:29:27, 5.68s/it]
94%|█████████▎| 41096/43939 [66:17:29<4:29:24, 5.69s/it]
94%|█████████▎| 41097/43939 [66:17:35<4:29:33, 5.69s/it]
94%|█████████▎| 41098/43939 [66:17:40<4:29:17, 5.69s/it]
94%|█████████▎| 41099/43939 [66:17:46<4:29:25, 5.69s/it]
94%|█████████▎| 41100/43939 [66:17:52<4:28:50, 5.68s/it]
{'loss': 0.655, 'grad_norm': 1.563614845275879, 'learning_rate': 6.335754527777515e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.4, 'epoch': 3.74}
94%|█████████▎| 41100/43939 [66:17:52<4:28:50, 5.68s/it]
94%|█████████▎| 41101/43939 [66:17:57<4:29:11, 5.69s/it]
94%|█████████▎| 41102/43939 [66:18:03<4:29:32, 5.70s/it]
94%|█████████▎| 41103/43939 [66:18:09<4:29:42, 5.71s/it]
94%|█████████▎| 41104/43939 [66:18:15<4:29:15, 5.70s/it]
94%|█████████▎| 41105/43939 [66:18:20<4:28:54, 5.69s/it]
94%|█████████▎| 41106/43939 [66:18:26<4:29:05, 5.70s/it]
94%|█████████▎| 41107/43939 [66:18:32<4:29:14, 5.70s/it]
94%|█████████▎| 41108/43939 [66:18:38<4:32:38, 5.78s/it]
94%|█████████▎| 41109/43939 [66:18:43<4:31:56, 5.77s/it]
94%|█████████▎| 41110/43939 [66:18:49<4:31:35, 5.76s/it]
94%|█████████▎| 41111/43939 [66:18:55<4:31:27, 5.76s/it]
94%|█████████▎| 41112/43939 [66:19:01<4:30:42, 5.75s/it]
94%|█████████▎| 41113/43939 [66:19:06<4:30:08, 5.74s/it]
94%|█████████▎| 41114/43939 [66:19:12<4:30:14, 5.74s/it]
94%|█████████▎| 41115/43939 [66:19:18<4:30:07, 5.74s/it]
94%|█████████▎| 41116/43939 [66:19:24<4:30:35, 5.75s/it]
94%|█████████▎| 41117/43939 [66:19:29<4:30:07, 5.74s/it]
94%|█████████▎| 41118/43939 [66:19:35<4:29:33, 5.73s/it]
94%|█████████▎| 41119/43939 [66:19:41<4:29:39, 5.74s/it]
94%|█████████▎| 41120/43939 [66:19:46<4:28:54, 5.72s/it]
94%|█████████▎| 41121/43939 [66:19:52<4:28:31, 5.72s/it]
94%|█████████▎| 41122/43939 [66:19:58<4:29:03, 5.73s/it]
94%|█████████▎| 41123/43939 [66:20:04<4:28:58, 5.73s/it]
94%|█████████▎| 41124/43939 [66:20:09<4:29:06, 5.74s/it]
94%|█████████▎| 41125/43939 [66:20:15<4:29:26, 5.75s/it]
94%|█████████▎| 41126/43939 [66:20:21<4:29:09, 5.74s/it]
94%|█████████▎| 41127/43939 [66:20:27<4:28:48, 5.74s/it]
94%|█████████▎| 41128/43939 [66:20:32<4:28:14, 5.73s/it]
94%|█████████▎| 41129/43939 [66:20:38<4:27:52, 5.72s/it]
94%|█████████▎| 41130/43939 [66:20:44<4:30:51, 5.79s/it]
94%|█████████▎| 41131/43939 [66:20:50<4:30:01, 5.77s/it]
94%|█████████▎| 41132/43939 [66:20:55<4:29:28, 5.76s/it]
94%|█████████▎| 41133/43939 [66:21:01<4:28:39, 5.74s/it]
94%|█████████▎| 41134/43939 [66:21:07<4:28:04, 5.73s/it]
94%|█████████▎| 41135/43939 [66:21:13<4:27:45, 5.73s/it]
94%|█████████▎| 41136/43939 [66:21:18<4:26:54, 5.71s/it]
94%|█████████▎| 41137/43939 [66:21:24<4:26:06, 5.70s/it]
94%|█████████▎| 41138/43939 [66:21:30<4:26:06, 5.70s/it]
94%|█████████▎| 41139/43939 [66:21:35<4:25:57, 5.70s/it]
94%|█████████▎| 41140/43939 [66:21:41<4:25:43, 5.70s/it]
94%|█████████▎| 41141/43939 [66:21:47<4:25:38, 5.70s/it]
94%|█████████▎| 41142/43939 [66:21:52<4:25:34, 5.70s/it]
94%|█████████▎| 41143/43939 [66:21:58<4:25:08, 5.69s/it]
94%|█████████▎| 41144/43939 [66:22:04<4:25:06, 5.69s/it]
94%|█████████▎| 41145/43939 [66:22:10<4:26:22, 5.72s/it]
94%|█████████▎| 41146/43939 [66:22:15<4:26:12, 5.72s/it]
94%|█████████▎| 41147/43939 [66:22:21<4:25:37, 5.71s/it]
94%|█████████▎| 41148/43939 [66:22:27<4:25:17, 5.70s/it]
94%|█████████▎| 41149/43939 [66:22:32<4:24:39, 5.69s/it]
94%|█████████▎| 41150/43939 [66:22:38<4:24:25, 5.69s/it]
{'loss': 0.6556, 'grad_norm': 1.7577075958251953, 'learning_rate': 6.115534458224853e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 530.38, 'epoch': 3.75}
94%|█████████▎| 41150/43939 [66:22:38<4:24:25, 5.69s/it]
94%|█████████▎| 41151/43939 [66:22:44<4:24:23, 5.69s/it]
94%|█████████▎| 41152/43939 [66:22:50<4:27:31, 5.76s/it]
94%|█████████▎| 41153/43939 [66:22:55<4:26:31, 5.74s/it]
94%|█████████▎| 41154/43939 [66:23:01<4:26:39, 5.74s/it]
94%|█████████▎| 41155/43939 [66:23:07<4:25:45, 5.73s/it]
94%|█████████▎| 41156/43939 [66:23:12<4:25:15, 5.72s/it]
94%|█████████▎| 41157/43939 [66:23:18<4:25:10, 5.72s/it]
94%|█████████▎| 41158/43939 [66:23:24<4:24:31, 5.71s/it]
94%|█████████▎| 41159/43939 [66:23:30<4:24:49, 5.72s/it]
94%|█████████▎| 41160/43939 [66:23:35<4:24:33, 5.71s/it]
94%|█████████▎| 41161/43939 [66:23:41<4:24:29, 5.71s/it]
94%|█████████▎| 41162/43939 [66:23:47<4:24:47, 5.72s/it]
94%|█████████▎| 41163/43939 [66:23:52<4:24:11, 5.71s/it]
94%|█████████▎| 41164/43939 [66:23:58<4:24:06, 5.71s/it]
94%|█████████▎| 41165/43939 [66:24:04<4:23:41, 5.70s/it]
94%|█████████▎| 41166/43939 [66:24:10<4:23:49, 5.71s/it]
94%|█████████▎| 41167/43939 [66:24:15<4:23:40, 5.71s/it]
94%|█████████▎| 41168/43939 [66:24:21<4:24:04, 5.72s/it]
94%|█████████▎| 41169/43939 [66:24:27<4:23:57, 5.72s/it]
94%|█████████▎| 41170/43939 [66:24:32<4:23:24, 5.71s/it]
94%|█████████▎| 41171/43939 [66:24:38<4:23:43, 5.72s/it]
94%|█████████▎| 41172/43939 [66:24:44<4:23:14, 5.71s/it]
94%|█████████▎| 41173/43939 [66:24:50<4:26:24, 5.78s/it]
94%|█████████▎| 41174/43939 [66:24:55<4:25:28, 5.76s/it]
94%|█████████▎| 41175/43939 [66:25:01<4:24:11, 5.73s/it]
94%|█████████▎| 41176/43939 [66:25:07<4:24:00, 5.73s/it]
94%|█████████▎| 41177/43939 [66:25:13<4:23:23, 5.72s/it]
94%|█████████▎| 41178/43939 [66:25:18<4:22:51, 5.71s/it]
94%|█████████▎| 41179/43939 [66:25:24<4:24:10, 5.74s/it]
94%|█████████▎| 41180/43939 [66:25:30<4:23:37, 5.73s/it]
94%|█████████▎| 41181/43939 [66:25:36<4:23:29, 5.73s/it]
94%|█████████▎| 41182/43939 [66:25:41<4:22:36, 5.72s/it]
94%|█████████▎| 41183/43939 [66:25:47<4:22:03, 5.71s/it]
94%|█████████▎| 41184/43939 [66:25:53<4:21:28, 5.69s/it]
94%|█████████▎| 41185/43939 [66:25:58<4:21:34, 5.70s/it]
94%|█████████▎| 41186/43939 [66:26:04<4:21:30, 5.70s/it]
94%|█████████▎| 41187/43939 [66:26:10<4:21:03, 5.69s/it]
94%|█████████▎| 41188/43939 [66:26:15<4:21:04, 5.69s/it]
94%|█████████▎| 41189/43939 [66:26:21<4:20:36, 5.69s/it]
94%|█████████▎| 41190/43939 [66:26:27<4:20:14, 5.68s/it]
94%|█████████▎| 41191/43939 [66:26:32<4:20:27, 5.69s/it]
94%|█████████▎| 41192/43939 [66:26:38<4:20:46, 5.70s/it]
94%|█████████▍| 41193/43939 [66:26:44<4:20:57, 5.70s/it]
94%|█████████▍| 41194/43939 [66:26:49<4:20:45, 5.70s/it]
94%|█████████▍| 41195/43939 [66:26:55<4:23:35, 5.76s/it]
94%|█████████▍| 41196/43939 [66:27:01<4:22:41, 5.75s/it]
94%|█████████▍| 41197/43939 [66:27:07<4:21:58, 5.73s/it]
94%|█████████▍| 41198/43939 [66:27:13<4:22:14, 5.74s/it]
94%|█████████▍| 41199/43939 [66:27:18<4:21:41, 5.73s/it]
94%|█████████▍| 41200/43939 [66:27:24<4:21:15, 5.72s/it]
{'loss': 0.6657, 'grad_norm': 1.8939036130905151, 'learning_rate': 5.899162239236727e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 514.09, 'epoch': 3.75}
94%|█████████▍| 41200/43939 [66:27:24<4:21:15, 5.72s/it]
94%|█████████▍| 41201/43939 [66:27:30<4:20:45, 5.71s/it]
94%|█████████▍| 41202/43939 [66:27:35<4:20:25, 5.71s/it]
94%|█████████▍| 41203/43939 [66:27:41<4:20:12, 5.71s/it]
94%|█████████▍| 41204/43939 [66:27:47<4:20:15, 5.71s/it]
94%|█████████▍| 41205/43939 [66:27:52<4:20:19, 5.71s/it]
94%|█████████▍| 41206/43939 [66:27:58<4:19:26, 5.70s/it]
94%|█████████▍| 41207/43939 [66:28:04<4:19:30, 5.70s/it]
94%|█████████▍| 41208/43939 [66:28:10<4:19:38, 5.70s/it]
94%|█████████▍| 41209/43939 [66:28:15<4:19:40, 5.71s/it]
94%|█████████▍| 41210/43939 [66:28:21<4:19:27, 5.70s/it]
94%|█████████▍| 41211/43939 [66:28:27<4:18:48, 5.69s/it]
94%|█████████▍| 41212/43939 [66:28:32<4:19:09, 5.70s/it]
94%|█████████▍| 41213/43939 [66:28:38<4:18:47, 5.70s/it]
94%|█████████▍| 41214/43939 [66:28:44<4:18:45, 5.70s/it]
94%|█████████▍| 41215/43939 [66:28:50<4:19:19, 5.71s/it]
94%|█████████▍| 41216/43939 [66:28:55<4:22:18, 5.78s/it]
94%|█████████▍| 41217/43939 [66:29:01<4:20:59, 5.75s/it]
94%|█████████▍| 41218/43939 [66:29:07<4:20:09, 5.74s/it]
94%|█████████▍| 41219/43939 [66:29:13<4:20:09, 5.74s/it]
94%|█████████▍| 41220/43939 [66:29:18<4:19:20, 5.72s/it]
94%|█████████▍| 41221/43939 [66:29:24<4:19:07, 5.72s/it]
94%|█████████▍| 41222/43939 [66:29:30<4:19:31, 5.73s/it]
94%|█████████▍| 41223/43939 [66:29:35<4:19:00, 5.72s/it]
94%|█████████▍| 41224/43939 [66:29:41<4:18:35, 5.71s/it]
94%|█████████▍| 41225/43939 [66:29:47<4:18:20, 5.71s/it]
94%|█████████▍| 41226/43939 [66:29:53<4:18:27, 5.72s/it]
94%|█████████▍| 41227/43939 [66:29:58<4:17:54, 5.71s/it]
94%|█████████▍| 41228/43939 [66:30:04<4:17:26, 5.70s/it]
94%|█████████▍| 41229/43939 [66:30:10<4:17:30, 5.70s/it]
94%|█████████▍| 41230/43939 [66:30:15<4:17:15, 5.70s/it]
94%|█████████▍| 41231/43939 [66:30:21<4:17:06, 5.70s/it]
94%|█████████▍| 41232/43939 [66:30:27<4:17:03, 5.70s/it]
94%|█████████▍| 41233/43939 [66:30:32<4:17:22, 5.71s/it]
94%|█████████▍| 41234/43939 [66:30:38<4:17:15, 5.71s/it]
94%|█████████▍| 41235/43939 [66:30:44<4:16:35, 5.69s/it]
94%|█████████▍| 41236/43939 [66:30:49<4:16:20, 5.69s/it]
94%|█████████▍| 41237/43939 [66:30:55<4:16:42, 5.70s/it]
94%|█████████▍| 41238/43939 [66:31:01<4:19:27, 5.76s/it]
94%|█████████▍| 41239/43939 [66:31:07<4:18:18, 5.74s/it]
94%|█████████▍| 41240/43939 [66:31:13<4:18:18, 5.74s/it]
94%|█████████▍| 41241/43939 [66:31:18<4:18:38, 5.75s/it]
94%|█████████▍| 41242/43939 [66:31:24<4:18:12, 5.74s/it]
94%|█████████▍| 41243/43939 [66:31:30<4:17:52, 5.74s/it]
94%|█████████▍| 41244/43939 [66:31:35<4:16:59, 5.72s/it]
94%|█████████▍| 41245/43939 [66:31:41<4:16:29, 5.71s/it]
94%|█████████▍| 41246/43939 [66:31:47<4:16:09, 5.71s/it]
94%|█████████▍| 41247/43939 [66:31:53<4:16:25, 5.72s/it]
94%|█████████▍| 41248/43939 [66:31:58<4:15:52, 5.71s/it]
94%|█████████▍| 41249/43939 [66:32:04<4:16:36, 5.72s/it]
94%|█████████▍| 41250/43939 [66:32:10<4:16:31, 5.72s/it]
{'loss': 0.6634, 'grad_norm': 1.8021842241287231, 'learning_rate': 5.686641284593369e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 462.24, 'epoch': 3.76}
94%|█████████▍| 41250/43939 [66:32:10<4:16:31, 5.72s/it]
94%|█████████▍| 41251/43939 [66:32:15<4:16:04, 5.72s/it]
94%|█████████▍| 41252/43939 [66:32:21<4:16:36, 5.73s/it]
94%|█████████▍| 41253/43939 [66:32:27<4:16:34, 5.73s/it]
94%|█████████▍| 41254/43939 [66:32:33<4:16:42, 5.74s/it]
94%|█████████▍| 41255/43939 [66:32:38<4:16:09, 5.73s/it]
94%|█████████▍| 41256/43939 [66:32:44<4:15:54, 5.72s/it]
94%|█████████▍| 41257/43939 [66:32:50<4:15:34, 5.72s/it]
94%|█████████▍| 41258/43939 [66:32:56<4:15:27, 5.72s/it]
94%|█████████▍| 41259/43939 [66:33:01<4:17:51, 5.77s/it]
94%|█████████▍| 41260/43939 [66:33:07<4:16:32, 5.75s/it]
94%|█████████▍| 41261/43939 [66:33:13<4:15:27, 5.72s/it]
94%|█████████▍| 41262/43939 [66:33:19<4:16:13, 5.74s/it]
94%|█████████▍| 41263/43939 [66:33:24<4:15:19, 5.72s/it]
94%|█████████▍| 41264/43939 [66:33:30<4:15:08, 5.72s/it]
94%|█████████▍| 41265/43939 [66:33:36<4:14:55, 5.72s/it]
94%|█████████▍| 41266/43939 [66:33:41<4:14:18, 5.71s/it]
94%|█████████▍| 41267/43939 [66:33:47<4:14:18, 5.71s/it]
94%|█████████▍| 41268/43939 [66:33:53<4:14:06, 5.71s/it]
94%|█████████▍| 41269/43939 [66:33:59<4:14:23, 5.72s/it]
94%|█████████▍| 41270/43939 [66:34:04<4:14:38, 5.72s/it]
94%|█████████▍| 41271/43939 [66:34:10<4:14:03, 5.71s/it]
94%|█████████▍| 41272/43939 [66:34:16<4:14:18, 5.72s/it]
94%|█████████▍| 41273/43939 [66:34:21<4:13:51, 5.71s/it]
94%|█████████▍| 41274/43939 [66:34:27<4:13:42, 5.71s/it]
94%|█████████▍| 41275/43939 [66:34:33<4:13:40, 5.71s/it]
94%|█████████▍| 41276/43939 [66:34:39<4:14:26, 5.73s/it]
94%|█████████▍| 41277/43939 [66:34:44<4:13:49, 5.72s/it]
94%|█████████▍| 41278/43939 [66:34:50<4:13:09, 5.71s/it]
94%|█████████▍| 41279/43939 [66:34:56<4:13:16, 5.71s/it]
94%|█████████▍| 41280/43939 [66:35:01<4:13:04, 5.71s/it]
94%|█████████▍| 41281/43939 [66:35:07<4:16:03, 5.78s/it]
94%|█████████▍| 41282/43939 [66:35:13<4:15:04, 5.76s/it]
94%|█████████▍| 41283/43939 [66:35:19<4:14:11, 5.74s/it]
94%|█████████▍| 41284/43939 [66:35:24<4:13:29, 5.73s/it]
94%|█████████▍| 41285/43939 [66:35:30<4:13:20, 5.73s/it]
94%|█████████▍| 41286/43939 [66:35:36<4:13:00, 5.72s/it]
94%|█████████▍| 41287/43939 [66:35:42<4:12:43, 5.72s/it]
94%|█████████▍| 41288/43939 [66:35:47<4:13:13, 5.73s/it]
94%|█████████▍| 41289/43939 [66:35:53<4:12:27, 5.72s/it]
94%|█████████▍| 41290/43939 [66:35:59<4:12:23, 5.72s/it]
94%|█████████▍| 41291/43939 [66:36:04<4:11:46, 5.71s/it]
94%|█████████▍| 41292/43939 [66:36:10<4:12:16, 5.72s/it]
94%|█████████▍| 41293/43939 [66:36:16<4:12:08, 5.72s/it]
94%|█████████▍| 41294/43939 [66:36:22<4:11:24, 5.70s/it]
94%|█████████▍| 41295/43939 [66:36:27<4:11:16, 5.70s/it]
94%|█████████▍| 41296/43939 [66:36:33<4:10:42, 5.69s/it]
94%|█████████▍| 41297/43939 [66:36:39<4:11:10, 5.70s/it]
94%|█████████▍| 41298/43939 [66:36:44<4:11:23, 5.71s/it]
94%|█████████▍| 41299/43939 [66:36:50<4:11:17, 5.71s/it]
94%|█████████▍| 41300/43939 [66:36:56<4:11:24, 5.72s/it]
{'loss': 0.6425, 'grad_norm': 1.4221422672271729, 'learning_rate': 5.477974947312281e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 471.62, 'epoch': 3.76}
94%|█████████▍| 41300/43939 [66:36:56<4:11:24, 5.72s/it]
94%|█████████▍| 41301/43939 [66:37:02<4:11:33, 5.72s/it]
94%|█████████▍| 41302/43939 [66:37:07<4:13:56, 5.78s/it]
94%|█████████▍| 41303/43939 [66:37:13<4:13:01, 5.76s/it]
94%|█████████▍| 41304/43939 [66:37:19<4:11:46, 5.73s/it]
94%|█████████▍| 41305/43939 [66:37:25<4:11:03, 5.72s/it]
94%|█████████▍| 41306/43939 [66:37:30<4:10:48, 5.72s/it]
94%|█████████▍| 41307/43939 [66:37:36<4:11:09, 5.73s/it]
94%|█████████▍| 41308/43939 [66:37:42<4:10:49, 5.72s/it]
94%|█████████▍| 41309/43939 [66:37:47<4:11:03, 5.73s/it]
94%|█████████▍| 41310/43939 [66:37:53<4:09:58, 5.71s/it]
94%|█████████▍| 41311/43939 [66:37:59<4:09:34, 5.70s/it]
94%|█████████▍| 41312/43939 [66:38:04<4:09:14, 5.69s/it]
94%|█████████▍| 41313/43939 [66:38:10<4:09:05, 5.69s/it]
94%|█████████▍| 41314/43939 [66:38:16<4:09:09, 5.70s/it]
94%|█████████▍| 41315/43939 [66:38:22<4:09:13, 5.70s/it]
94%|█████████▍| 41316/43939 [66:38:27<4:09:01, 5.70s/it]
94%|█████████▍| 41317/43939 [66:38:33<4:09:06, 5.70s/it]
94%|█████████▍| 41318/43939 [66:38:39<4:08:59, 5.70s/it]
94%|█████████▍| 41319/43939 [66:38:44<4:08:56, 5.70s/it]
94%|█████████▍| 41320/43939 [66:38:50<4:08:53, 5.70s/it]
94%|█████████▍| 41321/43939 [66:38:56<4:08:45, 5.70s/it]
94%|█████████▍| 41322/43939 [66:39:01<4:08:35, 5.70s/it]
94%|█████████▍| 41323/43939 [66:39:07<4:08:39, 5.70s/it]
94%|█████████▍| 41324/43939 [66:39:13<4:11:23, 5.77s/it]
94%|█████████▍| 41325/43939 [66:39:19<4:10:30, 5.75s/it]
94%|█████████▍| 41326/43939 [66:39:25<4:10:09, 5.74s/it]
94%|█████████▍| 41327/43939 [66:39:30<4:09:41, 5.74s/it]
94%|█████████▍| 41328/43939 [66:39:36<4:09:25, 5.73s/it]
94%|█████████▍| 41329/43939 [66:39:42<4:09:03, 5.73s/it]
94%|█████████▍| 41330/43939 [66:39:47<4:08:53, 5.72s/it]
94%|█████████▍| 41331/43939 [66:39:53<4:08:19, 5.71s/it]
94%|█████████▍| 41332/43939 [66:39:59<4:09:10, 5.73s/it]
94%|█████████▍| 41333/43939 [66:40:05<4:08:48, 5.73s/it]
94%|█████████▍| 41334/43939 [66:40:10<4:07:58, 5.71s/it]
94%|█████████▍| 41335/43939 [66:40:16<4:08:09, 5.72s/it]
94%|█████████▍| 41336/43939 [66:40:22<4:07:34, 5.71s/it]
94%|█████████▍| 41337/43939 [66:40:27<4:06:57, 5.69s/it]
94%|█████████▍| 41338/43939 [66:40:33<4:07:48, 5.72s/it]
94%|█████████▍| 41339/43939 [66:40:39<4:07:46, 5.72s/it]
94%|█████████▍| 41340/43939 [66:40:45<4:07:33, 5.72s/it]
94%|█████████▍| 41341/43939 [66:40:50<4:07:18, 5.71s/it]
94%|█████████▍| 41342/43939 [66:40:56<4:07:58, 5.73s/it]
94%|█████████▍| 41343/43939 [66:41:02<4:07:10, 5.71s/it]
94%|█████████▍| 41344/43939 [66:41:07<4:06:58, 5.71s/it]
94%|█████████▍| 41345/43939 [66:41:13<4:10:16, 5.79s/it]
94%|█████████▍| 41346/43939 [66:41:19<4:08:41, 5.75s/it]
94%|█████████▍| 41347/43939 [66:41:25<4:07:58, 5.74s/it]
94%|█████████▍| 41348/43939 [66:41:30<4:07:29, 5.73s/it]
94%|█████████▍| 41349/43939 [66:41:36<4:06:32, 5.71s/it]
94%|█████████▍| 41350/43939 [66:41:42<4:06:19, 5.71s/it]
{'loss': 0.6421, 'grad_norm': 1.4905974864959717, 'learning_rate': 5.273166519595251e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.04, 'epoch': 3.76}
94%|█████████▍| 41350/43939 [66:41:42<4:06:19, 5.71s/it]
94%|█████████▍| 41351/43939 [66:41:48<4:06:10, 5.71s/it]
94%|█████████▍| 41352/43939 [66:41:53<4:05:56, 5.70s/it]
94%|█████████▍| 41353/43939 [66:41:59<4:06:27, 5.72s/it]
94%|█████████▍| 41354/43939 [66:42:05<4:05:53, 5.71s/it]
94%|█████████▍| 41355/43939 [66:42:10<4:06:09, 5.72s/it]
94%|█████████▍| 41356/43939 [66:42:16<4:06:02, 5.72s/it]
94%|█████████▍| 41357/43939 [66:42:22<4:05:46, 5.71s/it]
94%|█████████▍| 41358/43939 [66:42:28<4:05:49, 5.71s/it]
94%|█████████▍| 41359/43939 [66:42:33<4:05:42, 5.71s/it]
94%|█████████▍| 41360/43939 [66:42:39<4:05:46, 5.72s/it]
94%|█████████▍| 41361/43939 [66:42:45<4:05:18, 5.71s/it]
94%|█████████▍| 41362/43939 [66:42:51<4:06:35, 5.74s/it]
94%|█████████▍| 41363/43939 [66:42:56<4:05:53, 5.73s/it]
94%|█████████▍| 41364/43939 [66:43:02<4:05:15, 5.71s/it]
94%|█████████▍| 41365/43939 [66:43:08<4:04:52, 5.71s/it]
94%|█████████▍| 41366/43939 [66:43:13<4:04:32, 5.70s/it]
94%|█████████▍| 41367/43939 [66:43:19<4:07:35, 5.78s/it]
94%|█████████▍| 41368/43939 [66:43:25<4:07:21, 5.77s/it]
94%|█████████▍| 41369/43939 [66:43:31<4:06:15, 5.75s/it]
94%|█████████▍| 41370/43939 [66:43:36<4:05:09, 5.73s/it]
94%|█████████▍| 41371/43939 [66:43:42<4:04:44, 5.72s/it]
94%|█████████▍| 41372/43939 [66:43:48<4:04:13, 5.71s/it]
94%|█████████▍| 41373/43939 [66:43:53<4:04:08, 5.71s/it]
94%|█████████▍| 41374/43939 [66:43:59<4:05:09, 5.73s/it]
94%|█████████▍| 41375/43939 [66:44:05<4:04:34, 5.72s/it]
94%|█████████▍| 41376/43939 [66:44:11<4:04:20, 5.72s/it]
94%|█████████▍| 41377/43939 [66:44:16<4:04:03, 5.72s/it]
94%|█████████▍| 41378/43939 [66:44:22<4:03:25, 5.70s/it]
94%|█████████▍| 41379/43939 [66:44:28<4:03:49, 5.71s/it]
94%|█████████▍| 41380/43939 [66:44:33<4:03:56, 5.72s/it]
94%|█████████▍| 41381/43939 [66:44:39<4:03:37, 5.71s/it]
94%|█████████▍| 41382/43939 [66:44:45<4:03:22, 5.71s/it]
94%|█████████▍| 41383/43939 [66:44:51<4:02:53, 5.70s/it]
94%|█████████▍| 41384/43939 [66:44:56<4:02:44, 5.70s/it]
94%|█████████▍| 41385/43939 [66:45:02<4:02:38, 5.70s/it]
94%|█████████▍| 41386/43939 [66:45:08<4:02:46, 5.71s/it]
94%|█████████▍| 41387/43939 [66:45:13<4:02:47, 5.71s/it]
94%|█████████▍| 41388/43939 [66:45:19<4:02:35, 5.71s/it]
94%|█████████▍| 41389/43939 [66:45:25<4:05:05, 5.77s/it]
94%|█████████▍| 41390/43939 [66:45:31<4:04:18, 5.75s/it]
94%|█████████▍| 41391/43939 [66:45:36<4:03:29, 5.73s/it]
94%|█████████▍| 41392/43939 [66:45:42<4:03:35, 5.74s/it]
94%|█████████▍| 41393/43939 [66:45:48<4:03:23, 5.74s/it]
94%|█████████▍| 41394/43939 [66:45:54<4:03:12, 5.73s/it]
94%|█████████▍| 41395/43939 [66:45:59<4:02:41, 5.72s/it]
94%|█████████▍| 41396/43939 [66:46:05<4:02:05, 5.71s/it]
94%|█████████▍| 41397/43939 [66:46:11<4:02:25, 5.72s/it]
94%|█████████▍| 41398/43939 [66:46:16<4:01:48, 5.71s/it]
94%|█████████▍| 41399/43939 [66:46:22<4:01:48, 5.71s/it]
94%|█████████▍| 41400/43939 [66:46:28<4:01:11, 5.70s/it]
{'loss': 0.6639, 'grad_norm': 1.8478672504425049, 'learning_rate': 5.072219232776564e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.03, 'epoch': 3.77}
94%|█████████▍| 41400/43939 [66:46:28<4:01:11, 5.70s/it]
94%|█████████▍| 41401/43939 [66:46:34<4:01:24, 5.71s/it]
94%|█████████▍| 41402/43939 [66:46:39<4:01:16, 5.71s/it]
94%|█████████▍| 41403/43939 [66:46:45<4:01:24, 5.71s/it]
94%|█████████▍| 41404/43939 [66:46:51<4:00:59, 5.70s/it]
94%|█████████▍| 41405/43939 [66:46:56<4:00:58, 5.71s/it]
94%|█████████▍| 41406/43939 [66:47:02<4:00:46, 5.70s/it]
94%|█████████▍| 41407/43939 [66:47:08<4:00:10, 5.69s/it]
94%|█████████▍| 41408/43939 [66:47:13<4:00:05, 5.69s/it]
94%|█████████▍| 41409/43939 [66:47:19<3:59:57, 5.69s/it]
94%|█████████▍| 41410/43939 [66:47:25<4:03:11, 5.77s/it]
94%|█████████▍| 41411/43939 [66:47:31<4:02:12, 5.75s/it]
94%|█████████▍| 41412/43939 [66:47:37<4:01:58, 5.75s/it]
94%|█████████▍| 41413/43939 [66:47:42<4:01:23, 5.73s/it]
94%|█████████▍| 41414/43939 [66:47:48<4:01:01, 5.73s/it]
94%|█████████▍| 41415/43939 [66:47:54<4:00:18, 5.71s/it]
94%|█████████▍| 41416/43939 [66:47:59<3:59:59, 5.71s/it]
94%|█████████▍| 41417/43939 [66:48:05<3:59:54, 5.71s/it]
94%|█████████▍| 41418/43939 [66:48:11<3:59:37, 5.70s/it]
94%|█████████▍| 41419/43939 [66:48:16<3:59:43, 5.71s/it]
94%|█████████▍| 41420/43939 [66:48:22<3:59:29, 5.70s/it]
94%|█████████▍| 41421/43939 [66:48:28<3:59:14, 5.70s/it]
94%|█████████▍| 41422/43939 [66:48:34<3:59:18, 5.70s/it]
94%|█████████▍| 41423/43939 [66:48:39<3:59:21, 5.71s/it]
94%|█████████▍| 41424/43939 [66:48:45<3:59:35, 5.72s/it]
94%|█████████▍| 41425/43939 [66:48:51<3:59:24, 5.71s/it]
94%|█████████▍| 41426/43939 [66:48:56<3:58:55, 5.70s/it]
94%|█████████▍| 41427/43939 [66:49:02<3:59:19, 5.72s/it]
94%|█████████▍| 41428/43939 [66:49:08<3:58:49, 5.71s/it]
94%|█████████▍| 41429/43939 [66:49:13<3:58:18, 5.70s/it]
94%|█████████▍| 41430/43939 [66:49:19<3:59:03, 5.72s/it]
94%|█████████▍| 41431/43939 [66:49:25<4:01:54, 5.79s/it]
94%|█████████▍| 41432/43939 [66:49:31<4:00:51, 5.76s/it]
94%|█████████▍| 41433/43939 [66:49:37<3:59:55, 5.74s/it]
94%|█████████▍| 41434/43939 [66:49:42<3:59:50, 5.74s/it]
94%|█████████▍| 41435/43939 [66:49:48<3:59:49, 5.75s/it]
94%|█████████▍| 41436/43939 [66:49:54<3:59:06, 5.73s/it]
94%|█████████▍| 41437/43939 [66:50:00<3:58:59, 5.73s/it]
94%|█████████▍| 41438/43939 [66:50:05<3:58:33, 5.72s/it]
94%|█████████▍| 41439/43939 [66:50:11<3:58:42, 5.73s/it]
94%|█████████▍| 41440/43939 [66:50:17<3:59:04, 5.74s/it]
94%|█████████▍| 41441/43939 [66:50:22<3:58:17, 5.72s/it]
94%|█████████▍| 41442/43939 [66:50:28<3:58:19, 5.73s/it]
94%|█████████▍| 41443/43939 [66:50:34<3:57:33, 5.71s/it]
94%|█████████▍| 41444/43939 [66:50:40<3:57:22, 5.71s/it]
94%|█████████▍| 41445/43939 [66:50:45<3:57:29, 5.71s/it]
94%|█████████▍| 41446/43939 [66:50:51<3:57:24, 5.71s/it]
94%|█████████▍| 41447/43939 [66:50:57<3:57:39, 5.72s/it]
94%|█████████▍| 41448/43939 [66:51:02<3:57:42, 5.73s/it]
94%|█████████▍| 41449/43939 [66:51:08<3:57:55, 5.73s/it]
94%|█████████▍| 41450/43939 [66:51:14<3:57:25, 5.72s/it]
{'loss': 0.6629, 'grad_norm': 1.469612717628479, 'learning_rate': 4.875136257271845e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.73, 'epoch': 3.77}
94%|█████████▍| 41450/43939 [66:51:14<3:57:25, 5.72s/it]
94%|█████████▍| 41451/43939 [66:51:20<3:57:35, 5.73s/it]
94%|█████████▍| 41452/43939 [66:51:25<3:57:44, 5.74s/it]
94%|█████████▍| 41453/43939 [66:51:31<4:00:13, 5.80s/it]
94%|█████████▍| 41454/43939 [66:51:37<3:58:26, 5.76s/it]
94%|█████████▍| 41455/43939 [66:51:43<3:58:27, 5.76s/it]
94%|█████████▍| 41456/43939 [66:51:49<3:58:39, 5.77s/it]
94%|█████████▍| 41457/43939 [66:51:54<3:58:05, 5.76s/it]
94%|█████████▍| 41458/43939 [66:52:00<3:57:48, 5.75s/it]
94%|█████████▍| 41459/43939 [66:52:06<3:57:10, 5.74s/it]
94%|█████████▍| 41460/43939 [66:52:11<3:56:50, 5.73s/it]
94%|█████████▍| 41461/43939 [66:52:17<3:56:21, 5.72s/it]
94%|█████████▍| 41462/43939 [66:52:23<3:56:10, 5.72s/it]
94%|█████████▍| 41463/43939 [66:52:29<3:55:44, 5.71s/it]
94%|█████████▍| 41464/43939 [66:52:34<3:55:53, 5.72s/it]
94%|█████████▍| 41465/43939 [66:52:40<3:55:19, 5.71s/it]
94%|█████████▍| 41466/43939 [66:52:46<3:55:46, 5.72s/it]
94%|█████████▍| 41467/43939 [66:52:51<3:55:25, 5.71s/it]
94%|█████████▍| 41468/43939 [66:52:57<3:55:08, 5.71s/it]
94%|█████████▍| 41469/43939 [66:53:03<3:54:29, 5.70s/it]
94%|█████████▍| 41470/43939 [66:53:08<3:54:14, 5.69s/it]
94%|█████████▍| 41471/43939 [66:53:14<3:55:12, 5.72s/it]
94%|█████████▍| 41472/43939 [66:53:20<3:55:03, 5.72s/it]
94%|█████████▍| 41473/43939 [66:53:26<3:55:03, 5.72s/it]
94%|█████████▍| 41474/43939 [66:53:31<3:54:17, 5.70s/it]
94%|█████████▍| 41475/43939 [66:53:37<3:57:33, 5.78s/it]
94%|█████████▍| 41476/43939 [66:53:43<3:56:20, 5.76s/it]
94%|█████████▍| 41477/43939 [66:53:49<3:55:39, 5.74s/it]
94%|█████████▍| 41478/43939 [66:53:54<3:54:34, 5.72s/it]
94%|█████████▍| 41479/43939 [66:54:00<3:54:56, 5.73s/it]
94%|█████████▍| 41480/43939 [66:54:06<3:54:26, 5.72s/it]
94%|█████████▍| 41481/43939 [66:54:12<3:54:25, 5.72s/it]
94%|█████████▍| 41482/43939 [66:54:17<3:54:14, 5.72s/it]
94%|█████████▍| 41483/43939 [66:54:23<3:53:57, 5.72s/it]
94%|█████████▍| 41484/43939 [66:54:29<3:54:02, 5.72s/it]
94%|█████████▍| 41485/43939 [66:54:34<3:53:55, 5.72s/it]
94%|█████████▍| 41486/43939 [66:54:40<3:53:10, 5.70s/it]
94%|█████████▍| 41487/43939 [66:54:46<3:53:07, 5.70s/it]
94%|█████████▍| 41488/43939 [66:54:52<3:53:18, 5.71s/it]
94%|█████████▍| 41489/43939 [66:54:57<3:52:55, 5.70s/it]
94%|█████████▍| 41490/43939 [66:55:03<3:52:55, 5.71s/it]
94%|█████████▍| 41491/43939 [66:55:09<3:52:43, 5.70s/it]
94%|█████████▍| 41492/43939 [66:55:14<3:52:38, 5.70s/it]
94%|█████████▍| 41493/43939 [66:55:20<3:52:27, 5.70s/it]
94%|█████████▍| 41494/43939 [66:55:26<3:52:42, 5.71s/it]
94%|█████████▍| 41495/43939 [66:55:31<3:52:51, 5.72s/it]
94%|█████████▍| 41496/43939 [66:55:37<3:53:06, 5.72s/it]
94%|█████████▍| 41497/43939 [66:55:43<3:55:45, 5.79s/it]
94%|█████████▍| 41498/43939 [66:55:49<3:54:01, 5.75s/it]
94%|█████████▍| 41499/43939 [66:55:55<3:53:21, 5.74s/it]
94%|█████████▍| 41500/43939 [66:56:00<3:52:39, 5.72s/it]
{'loss': 0.6474, 'grad_norm': 1.693228840827942, 'learning_rate': 4.6819207025281806e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.12, 'epoch': 3.78}
94%|█████████▍| 41500/43939 [66:56:00<3:52:39, 5.72s/it]
94%|█████████▍| 41501/43939 [66:56:06<3:52:13, 5.72s/it]
94%|█████████▍| 41502/43939 [66:56:12<3:53:15, 5.74s/it]
94%|█████████▍| 41503/43939 [66:56:17<3:52:21, 5.72s/it]
94%|█████████▍| 41504/43939 [66:56:23<3:51:50, 5.71s/it]
94%|█████████▍| 41505/43939 [66:56:29<3:51:32, 5.71s/it]
94%|█████████▍| 41506/43939 [66:56:35<3:51:19, 5.70s/it]
94%|█████████▍| 41507/43939 [66:56:40<3:51:20, 5.71s/it]
94%|█████████▍| 41508/43939 [66:56:46<3:51:53, 5.72s/it]
94%|█████████▍| 41509/43939 [66:56:52<3:51:55, 5.73s/it]
94%|█████████▍| 41510/43939 [66:56:57<3:51:51, 5.73s/it]
94%|█████████▍| 41511/43939 [66:57:03<3:50:59, 5.71s/it]
94%|█████████▍| 41512/43939 [66:57:09<3:50:43, 5.70s/it]
94%|█████████▍| 41513/43939 [66:57:14<3:50:24, 5.70s/it]
94%|█████████▍| 41514/43939 [66:57:20<3:50:28, 5.70s/it]
94%|█████████▍| 41515/43939 [66:57:26<3:50:19, 5.70s/it]
94%|█████████▍| 41516/43939 [66:57:32<3:50:21, 5.70s/it]
94%|█████████▍| 41517/43939 [66:57:37<3:50:02, 5.70s/it]
94%|█████████▍| 41518/43939 [66:57:43<3:50:12, 5.71s/it]
94%|█████████▍| 41519/43939 [66:57:49<3:52:21, 5.76s/it]
94%|█████████▍| 41520/43939 [66:57:55<3:51:26, 5.74s/it]
94%|█████████▍| 41521/43939 [66:58:00<3:50:52, 5.73s/it]
94%|█████████▍| 41522/43939 [66:58:06<3:50:10, 5.71s/it]
95%|█████████▍| 41523/43939 [66:58:12<3:49:55, 5.71s/it]
95%|█████████▍| 41524/43939 [66:58:17<3:49:48, 5.71s/it]
95%|█████████▍| 41525/43939 [66:58:23<3:49:44, 5.71s/it]
95%|█████████▍| 41526/43939 [66:58:29<3:49:23, 5.70s/it]
95%|█████████▍| 41527/43939 [66:58:34<3:49:22, 5.71s/it]
95%|█████████▍| 41528/43939 [66:58:40<3:49:19, 5.71s/it]
95%|█████████▍| 41529/43939 [66:58:46<3:48:59, 5.70s/it]
95%|█████████▍| 41530/43939 [66:58:52<3:48:50, 5.70s/it]
95%|█████████▍| 41531/43939 [66:58:57<3:48:57, 5.71s/it]
95%|█████████▍| 41532/43939 [66:59:03<3:48:54, 5.71s/it]
95%|█████████▍| 41533/43939 [66:59:09<3:48:36, 5.70s/it]
95%|█████████▍| 41534/43939 [66:59:14<3:48:53, 5.71s/it]
95%|█████████▍| 41535/43939 [66:59:20<3:48:33, 5.70s/it]
95%|█████████▍| 41536/43939 [66:59:26<3:48:36, 5.71s/it]
95%|█████████▍| 41537/43939 [66:59:32<3:48:47, 5.71s/it]
95%|█████████▍| 41538/43939 [66:59:37<3:48:26, 5.71s/it]
95%|█████████▍| 41539/43939 [66:59:43<3:48:25, 5.71s/it]
95%|█████████▍| 41540/43939 [66:59:49<3:51:12, 5.78s/it]
95%|█████████▍| 41541/43939 [66:59:55<3:50:35, 5.77s/it]
95%|█████████▍| 41542/43939 [67:00:00<3:49:54, 5.75s/it]
95%|█████████▍| 41543/43939 [67:00:06<3:49:32, 5.75s/it]
95%|█████████▍| 41544/43939 [67:00:12<3:49:10, 5.74s/it]
95%|█████████▍| 41545/43939 [67:00:18<3:48:16, 5.72s/it]
95%|█████████▍| 41546/43939 [67:00:23<3:47:44, 5.71s/it]
95%|█████████▍| 41547/43939 [67:00:29<3:47:21, 5.70s/it]
95%|█████████▍| 41548/43939 [67:00:35<3:46:55, 5.69s/it]
95%|█████████▍| 41549/43939 [67:00:40<3:47:03, 5.70s/it]
95%|█████████▍| 41550/43939 [67:00:46<3:47:10, 5.71s/it]
{'loss': 0.658, 'grad_norm': 1.7729908227920532, 'learning_rate': 4.4925756169750255e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.74, 'epoch': 3.78}
95%|█████████▍| 41550/43939 [67:00:46<3:47:10, 5.71s/it]
95%|█████████▍| 41551/43939 [67:00:52<3:47:11, 5.71s/it]
95%|█████████▍| 41552/43939 [67:00:57<3:46:54, 5.70s/it]
95%|█████████▍| 41553/43939 [67:01:03<3:46:39, 5.70s/it]
95%|█████████▍| 41554/43939 [67:01:09<3:46:32, 5.70s/it]
95%|█████████▍| 41555/43939 [67:01:14<3:46:26, 5.70s/it]
95%|█████████▍| 41556/43939 [67:01:20<3:46:13, 5.70s/it]
95%|█████████▍| 41557/43939 [67:01:26<3:46:31, 5.71s/it]
95%|█████████▍| 41558/43939 [67:01:32<3:46:04, 5.70s/it]
95%|█████████▍| 41559/43939 [67:01:37<3:46:03, 5.70s/it]
95%|█████████▍| 41560/43939 [67:01:43<3:46:08, 5.70s/it]
95%|█████████▍| 41561/43939 [67:01:49<3:49:04, 5.78s/it]
95%|█████████▍| 41562/43939 [67:01:55<3:48:06, 5.76s/it]
95%|█████████▍| 41563/43939 [67:02:00<3:47:37, 5.75s/it]
95%|█████████▍| 41564/43939 [67:02:06<3:47:31, 5.75s/it]
95%|█████████▍| 41565/43939 [67:02:12<3:46:17, 5.72s/it]
95%|█████████▍| 41566/43939 [67:02:18<3:45:58, 5.71s/it]
95%|█████████▍| 41567/43939 [67:02:23<3:45:37, 5.71s/it]
95%|█████████▍| 41568/43939 [67:02:29<3:45:27, 5.71s/it]
95%|█████████▍| 41569/43939 [67:02:35<3:45:22, 5.71s/it]
95%|█████████▍| 41570/43939 [67:02:40<3:44:51, 5.70s/it]
95%|█████████▍| 41571/43939 [67:02:46<3:44:47, 5.70s/it]
95%|█████████▍| 41572/43939 [67:02:52<3:44:35, 5.69s/it]
95%|█████████▍| 41573/43939 [67:02:57<3:44:38, 5.70s/it]
95%|█████████▍| 41574/43939 [67:03:03<3:44:24, 5.69s/it]
95%|█████████▍| 41575/43939 [67:03:09<3:44:13, 5.69s/it]
95%|█████████▍| 41576/43939 [67:03:14<3:44:17, 5.69s/it]
95%|█████████▍| 41577/43939 [67:03:20<3:44:22, 5.70s/it]
95%|█████████▍| 41578/43939 [67:03:26<3:44:20, 5.70s/it]
95%|█████████▍| 41579/43939 [67:03:32<3:44:29, 5.71s/it]
95%|█████████▍| 41580/43939 [67:03:37<3:44:34, 5.71s/it]
95%|█████████▍| 41581/43939 [67:03:43<3:44:24, 5.71s/it]
95%|█████████▍| 41582/43939 [67:03:49<3:44:17, 5.71s/it]
95%|█████████▍| 41583/43939 [67:03:55<3:47:08, 5.78s/it]
95%|█████████▍| 41584/43939 [67:04:00<3:46:37, 5.77s/it]
95%|█████████▍| 41585/43939 [67:04:06<3:45:52, 5.76s/it]
95%|█████████▍| 41586/43939 [67:04:12<3:45:20, 5.75s/it]
95%|█████████▍| 41587/43939 [67:04:18<3:44:27, 5.73s/it]
95%|█████████▍| 41588/43939 [67:04:23<3:44:06, 5.72s/it]
95%|█████████▍| 41589/43939 [67:04:29<3:43:59, 5.72s/it]
95%|█████████▍| 41590/43939 [67:04:35<3:44:01, 5.72s/it]
95%|█████████▍| 41591/43939 [67:04:40<3:43:28, 5.71s/it]
95%|█████████▍| 41592/43939 [67:04:46<3:42:53, 5.70s/it]
95%|█████████▍| 41593/43939 [67:04:52<3:43:27, 5.71s/it]
95%|█████████▍| 41594/43939 [67:04:58<3:43:22, 5.72s/it]
95%|█████████▍| 41595/43939 [67:05:03<3:42:44, 5.70s/it]
95%|█████████▍| 41596/43939 [67:05:09<3:42:34, 5.70s/it]
95%|█████████▍| 41597/43939 [67:05:15<3:43:01, 5.71s/it]
95%|█████████▍| 41598/43939 [67:05:20<3:43:48, 5.74s/it]
95%|█████████▍| 41599/43939 [67:05:26<3:42:59, 5.72s/it]
95%|█████████▍| 41600/43939 [67:05:32<3:42:29, 5.71s/it]
{'loss': 0.6405, 'grad_norm': 1.680176019668579, 'learning_rate': 4.307103987976041e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 469.94, 'epoch': 3.79}
95%|█████████▍| 41600/43939 [67:05:32<3:42:29, 5.71s/it]
95%|█████████▍| 41601/43939 [67:05:37<3:42:12, 5.70s/it]
95%|█████████▍| 41602/43939 [67:05:43<3:42:20, 5.71s/it]
95%|█████████▍| 41603/43939 [67:05:49<3:42:24, 5.71s/it]
95%|█████████▍| 41604/43939 [67:05:55<3:45:13, 5.79s/it]
95%|█████████▍| 41605/43939 [67:06:01<3:44:29, 5.77s/it]
95%|█████████▍| 41606/43939 [67:06:06<3:43:40, 5.75s/it]
95%|█████████▍| 41607/43939 [67:06:12<3:43:10, 5.74s/it]
95%|█████████▍| 41608/43939 [67:06:18<3:42:56, 5.74s/it]
95%|█████████▍| 41609/43939 [67:06:23<3:42:31, 5.73s/it]
95%|█████████▍| 41610/43939 [67:06:29<3:42:18, 5.73s/it]
95%|█████████▍| 41611/43939 [67:06:35<3:41:43, 5.71s/it]
95%|█████████▍| 41612/43939 [67:06:41<3:41:16, 5.71s/it]
95%|█████████▍| 41613/43939 [67:06:46<3:41:21, 5.71s/it]
95%|█████████▍| 41614/43939 [67:06:52<3:41:21, 5.71s/it]
95%|█████████▍| 41615/43939 [67:06:58<3:41:26, 5.72s/it]
95%|█████████▍| 41616/43939 [67:07:03<3:41:04, 5.71s/it]
95%|█████████▍| 41617/43939 [67:07:09<3:41:05, 5.71s/it]
95%|█████████▍| 41618/43939 [67:07:15<3:40:55, 5.71s/it]
95%|█████████▍| 41619/43939 [67:07:21<3:40:51, 5.71s/it]
95%|█████████▍| 41620/43939 [67:07:26<3:40:27, 5.70s/it]
95%|█████████▍| 41621/43939 [67:07:32<3:40:05, 5.70s/it]
95%|█████████▍| 41622/43939 [67:07:38<3:39:57, 5.70s/it]
95%|█████████▍| 41623/43939 [67:07:43<3:39:24, 5.68s/it]
95%|█████████▍| 41624/43939 [67:07:49<3:39:19, 5.68s/it]
95%|█████████▍| 41625/43939 [67:07:55<3:40:00, 5.70s/it]
95%|█████████▍| 41626/43939 [67:08:01<3:42:48, 5.78s/it]
95%|█████████▍| 41627/43939 [67:08:06<3:41:46, 5.76s/it]
95%|█████████▍| 41628/43939 [67:08:12<3:40:56, 5.74s/it]
95%|█████████▍| 41629/43939 [67:08:18<3:40:30, 5.73s/it]
95%|█████████▍| 41630/43939 [67:08:24<3:40:38, 5.73s/it]
95%|█████████▍| 41631/43939 [67:08:29<3:40:27, 5.73s/it]
95%|█████████▍| 41632/43939 [67:08:35<3:40:11, 5.73s/it]
95%|█████████▍| 41633/43939 [67:08:41<3:39:54, 5.72s/it]
95%|█████████▍| 41634/43939 [67:08:46<3:39:37, 5.72s/it]
95%|█████████▍| 41635/43939 [67:08:52<3:39:25, 5.71s/it]
95%|█████████▍| 41636/43939 [67:08:58<3:38:42, 5.70s/it]
95%|█████████▍| 41637/43939 [67:09:03<3:38:43, 5.70s/it]
95%|█████████▍| 41638/43939 [67:09:09<3:39:09, 5.71s/it]
95%|█████████▍| 41639/43939 [67:09:15<3:39:03, 5.71s/it]
95%|█████████▍| 41640/43939 [67:09:21<3:38:46, 5.71s/it]
95%|█████████▍| 41641/43939 [67:09:26<3:38:44, 5.71s/it]
95%|█████████▍| 41642/43939 [67:09:32<3:38:29, 5.71s/it]
95%|█████████▍| 41643/43939 [67:09:38<3:38:23, 5.71s/it]
95%|█████████▍| 41644/43939 [67:09:43<3:38:08, 5.70s/it]
95%|█████████▍| 41645/43939 [67:09:49<3:37:53, 5.70s/it]
95%|█████████▍| 41646/43939 [67:09:55<3:37:51, 5.70s/it]
95%|█████████▍| 41647/43939 [67:10:00<3:37:37, 5.70s/it]
95%|█████████▍| 41648/43939 [67:10:06<3:40:17, 5.77s/it]
95%|█████████▍| 41649/43939 [67:10:12<3:39:54, 5.76s/it]
95%|█████████▍| 41650/43939 [67:10:18<3:39:15, 5.75s/it]
{'loss': 0.6665, 'grad_norm': 1.6599911451339722, 'learning_rate': 4.125508741781997e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 501.12, 'epoch': 3.79}
95%|█████████▍| 41650/43939 [67:10:18<3:39:15, 5.75s/it]
95%|█████████▍| 41651/43939 [67:10:24<3:38:39, 5.73s/it]
95%|█████████▍| 41652/43939 [67:10:29<3:38:11, 5.72s/it]
95%|█████████▍| 41653/43939 [67:10:35<3:37:47, 5.72s/it]
95%|█████████▍| 41654/43939 [67:10:41<3:37:34, 5.71s/it]
95%|█████████▍| 41655/43939 [67:10:46<3:37:21, 5.71s/it]
95%|█████████▍| 41656/43939 [67:10:52<3:37:30, 5.72s/it]
95%|█████████▍| 41657/43939 [67:10:58<3:37:34, 5.72s/it]
95%|█████████▍| 41658/43939 [67:11:04<3:37:56, 5.73s/it]
95%|█████████▍| 41659/43939 [67:11:09<3:37:23, 5.72s/it]
95%|█████████▍| 41660/43939 [67:11:15<3:36:40, 5.70s/it]
95%|█████████▍| 41661/43939 [67:11:21<3:36:41, 5.71s/it]
95%|█████████▍| 41662/43939 [67:11:26<3:36:46, 5.71s/it]
95%|█████████▍| 41663/43939 [67:11:32<3:36:24, 5.71s/it]
95%|█████████▍| 41664/43939 [67:11:38<3:36:41, 5.71s/it]
95%|█████████▍| 41665/43939 [67:11:44<3:36:50, 5.72s/it]
95%|█████████▍| 41666/43939 [67:11:49<3:36:37, 5.72s/it]
95%|█████████▍| 41667/43939 [67:11:55<3:36:35, 5.72s/it]
95%|█████████▍| 41668/43939 [67:12:01<3:36:14, 5.71s/it]
95%|█████████▍| 41669/43939 [67:12:07<3:39:06, 5.79s/it]
95%|█████████▍| 41670/43939 [67:12:12<3:38:17, 5.77s/it]
95%|█████████▍| 41671/43939 [67:12:18<3:37:20, 5.75s/it]
95%|█████████▍| 41672/43939 [67:12:24<3:36:49, 5.74s/it]
95%|█████████▍| 41673/43939 [67:12:30<3:36:17, 5.73s/it]
95%|█████████▍| 41674/43939 [67:12:35<3:35:43, 5.71s/it]
95%|█████████▍| 41675/43939 [67:12:41<3:35:30, 5.71s/it]
95%|█████████▍| 41676/43939 [67:12:47<3:35:18, 5.71s/it]
95%|█████████▍| 41677/43939 [67:12:52<3:35:26, 5.71s/it]
95%|█████████▍| 41678/43939 [67:12:58<3:35:22, 5.72s/it]
95%|█████████▍| 41679/43939 [67:13:04<3:35:04, 5.71s/it]
95%|█████████▍| 41680/43939 [67:13:09<3:35:04, 5.71s/it]
95%|█████████▍| 41681/43939 [67:13:15<3:34:54, 5.71s/it]
95%|█████████▍| 41682/43939 [67:13:21<3:34:20, 5.70s/it]
95%|█████████▍| 41683/43939 [67:13:27<3:34:25, 5.70s/it]
95%|█████████▍| 41684/43939 [67:13:32<3:34:37, 5.71s/it]
95%|█████████▍| 41685/43939 [67:13:38<3:34:18, 5.70s/it]
95%|█████████▍| 41686/43939 [67:13:44<3:34:52, 5.72s/it]
95%|█████████▍| 41687/43939 [67:13:49<3:34:38, 5.72s/it]
95%|█████████▍| 41688/43939 [67:13:55<3:34:42, 5.72s/it]
95%|█████████▍| 41689/43939 [67:14:01<3:33:51, 5.70s/it]
95%|█████████▍| 41690/43939 [67:14:07<3:36:30, 5.78s/it]
95%|█████████▍| 41691/43939 [67:14:12<3:35:31, 5.75s/it]
95%|█████████▍| 41692/43939 [67:14:18<3:34:33, 5.73s/it]
95%|█████████▍| 41693/43939 [67:14:24<3:34:06, 5.72s/it]
95%|█████████▍| 41694/43939 [67:14:30<3:33:44, 5.71s/it]
95%|█████████▍| 41695/43939 [67:14:35<3:34:07, 5.73s/it]
95%|█████████▍| 41696/43939 [67:14:41<3:33:50, 5.72s/it]
95%|█████████▍| 41697/43939 [67:14:47<3:33:11, 5.71s/it]
95%|█████████▍| 41698/43939 [67:14:52<3:33:14, 5.71s/it]
95%|█████████▍| 41699/43939 [67:14:58<3:33:00, 5.71s/it]
95%|█████████▍| 41700/43939 [67:15:04<3:32:19, 5.69s/it]
{'loss': 0.6398, 'grad_norm': 1.6246556043624878, 'learning_rate': 3.947792743484696e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 487.38, 'epoch': 3.8}
95%|█████████▍| 41700/43939 [67:15:04<3:32:19, 5.69s/it]
95%|█████████▍| 41701/43939 [67:15:09<3:32:17, 5.69s/it]
95%|█████████▍| 41702/43939 [67:15:15<3:32:49, 5.71s/it]
95%|█████████▍| 41703/43939 [67:15:21<3:32:29, 5.70s/it]
95%|█████████▍| 41704/43939 [67:15:27<3:32:42, 5.71s/it]
95%|█████████▍| 41705/43939 [67:15:32<3:33:05, 5.72s/it]
95%|█████████▍| 41706/43939 [67:15:38<3:33:18, 5.73s/it]
95%|█████████▍| 41707/43939 [67:15:44<3:32:54, 5.72s/it]
95%|█████████▍| 41708/43939 [67:15:50<3:32:40, 5.72s/it]
95%|█████████▍| 41709/43939 [67:15:55<3:32:17, 5.71s/it]
95%|█████████▍| 41710/43939 [67:16:01<3:32:10, 5.71s/it]
95%|█████████▍| 41711/43939 [67:16:07<3:31:37, 5.70s/it]
95%|█████████▍| 41712/43939 [67:16:13<3:34:28, 5.78s/it]
95%|█████████▍| 41713/43939 [67:16:18<3:33:35, 5.76s/it]
95%|█████████▍| 41714/43939 [67:16:24<3:33:04, 5.75s/it]
95%|█████████▍| 41715/43939 [67:16:30<3:32:27, 5.73s/it]
95%|█████████▍| 41716/43939 [67:16:35<3:31:51, 5.72s/it]
95%|█████████▍| 41717/43939 [67:16:41<3:32:21, 5.73s/it]
95%|█████████▍| 41718/43939 [67:16:47<3:32:28, 5.74s/it]
95%|█████████▍| 41719/43939 [67:16:53<3:32:13, 5.74s/it]
95%|█████████▍| 41720/43939 [67:16:58<3:31:36, 5.72s/it]
95%|█████████▍| 41721/43939 [67:17:04<3:31:07, 5.71s/it]
95%|█████████▍| 41722/43939 [67:17:10<3:30:32, 5.70s/it]
95%|█████████▍| 41723/43939 [67:17:15<3:30:27, 5.70s/it]
95%|█████████▍| 41724/43939 [67:17:21<3:29:51, 5.68s/it]
95%|█████████▍| 41725/43939 [67:17:27<3:30:13, 5.70s/it]
95%|█████████▍| 41726/43939 [67:17:32<3:30:02, 5.69s/it]
95%|█████████▍| 41727/43939 [67:17:38<3:29:32, 5.68s/it]
95%|█████████▍| 41728/43939 [67:17:44<3:29:48, 5.69s/it]
95%|█████████▍| 41729/43939 [67:17:50<3:30:05, 5.70s/it]
95%|█████████▍| 41730/43939 [67:17:55<3:30:27, 5.72s/it]
95%|█████████▍| 41731/43939 [67:18:01<3:30:51, 5.73s/it]
95%|█████████▍| 41732/43939 [67:18:07<3:30:16, 5.72s/it]
95%|█████████▍| 41733/43939 [67:18:12<3:30:23, 5.72s/it]
95%|█████████▍| 41734/43939 [67:18:18<3:33:01, 5.80s/it]
95%|█████████▍| 41735/43939 [67:18:24<3:32:00, 5.77s/it]
95%|█████████▍| 41736/43939 [67:18:30<3:31:14, 5.75s/it]
95%|█████████▍| 41737/43939 [67:18:36<3:30:33, 5.74s/it]
95%|█████████▍| 41738/43939 [67:18:41<3:30:16, 5.73s/it]
95%|█████████▍| 41739/43939 [67:18:47<3:29:58, 5.73s/it]
95%|█████████▍| 41740/43939 [67:18:53<3:30:07, 5.73s/it]
95%|█████████▍| 41741/43939 [67:18:59<3:30:08, 5.74s/it]
95%|█████████▍| 41742/43939 [67:19:04<3:29:16, 5.72s/it]
95%|█████████▌| 41743/43939 [67:19:10<3:28:52, 5.71s/it][2025-12-01 13:23:23,682] [INFO] [axolotl.core.trainers.base.evaluate:377] [PID:3633661] Running evaluation step...
0%| | 0/72 [00:00, ?it/s][A
3%|▎ | 2/72 [00:02<01:36, 1.37s/it][A
4%|▍ | 3/72 [00:05<02:13, 1.94s/it][A
6%|▌ | 4/72 [00:08<02:31, 2.23s/it][A
7%|▋ | 5/72 [00:10<02:40, 2.40s/it][A
8%|▊ | 6/72 [00:13<02:45, 2.51s/it][A
10%|▉ | 7/72 [00:16<02:47, 2.58s/it][A
11%|█ | 8/72 [00:19<02:48, 2.63s/it][A
12%|█▎ | 9/72 [00:21<02:47, 2.66s/it][A
14%|█▍ | 10/72 [00:24<02:45, 2.67s/it][A
15%|█▌ | 11/72 [00:27<02:43, 2.68s/it][A
17%|█▋ | 12/72 [00:29<02:41, 2.70s/it][A
18%|█▊ | 13/72 [00:32<02:39, 2.71s/it][A
19%|█▉ | 14/72 [00:35<02:37, 2.71s/it][A
21%|██ | 15/72 [00:38<02:34, 2.71s/it][A
22%|██▏ | 16/72 [00:40<02:31, 2.71s/it][A
24%|██▎ | 17/72 [00:43<02:29, 2.72s/it][A
25%|██▌ | 18/72 [00:46<02:26, 2.72s/it][A
26%|██▋ | 19/72 [00:49<02:24, 2.72s/it][A
28%|██▊ | 20/72 [00:51<02:21, 2.72s/it][A
29%|██▉ | 21/72 [00:54<02:18, 2.72s/it][A
31%|███ | 22/72 [00:57<02:16, 2.72s/it][A
32%|███▏ | 23/72 [00:59<02:13, 2.72s/it][A
33%|███▎ | 24/72 [01:02<02:10, 2.72s/it][A
35%|███▍ | 25/72 [01:05<02:08, 2.72s/it][A
36%|███▌ | 26/72 [01:08<02:05, 2.72s/it][A
38%|███▊ | 27/72 [01:10<02:02, 2.72s/it][A
39%|███▉ | 28/72 [01:13<02:00, 2.73s/it][A
40%|████ | 29/72 [01:16<01:57, 2.72s/it][A
42%|████▏ | 30/72 [01:18<01:54, 2.72s/it][A
43%|████▎ | 31/72 [01:21<01:51, 2.72s/it][A
44%|████▍ | 32/72 [01:24<01:49, 2.73s/it][A
46%|████▌ | 33/72 [01:27<01:46, 2.73s/it][A
47%|████▋ | 34/72 [01:29<01:43, 2.72s/it][A
49%|████▊ | 35/72 [01:32<01:40, 2.72s/it][A
50%|█████ | 36/72 [01:35<01:38, 2.73s/it][A
51%|█████▏ | 37/72 [01:38<01:35, 2.73s/it][A
53%|█████▎ | 38/72 [01:40<01:32, 2.73s/it][A
54%|█████▍ | 39/72 [01:43<01:29, 2.73s/it][A
56%|█████▌ | 40/72 [01:46<01:27, 2.72s/it][A
57%|█████▋ | 41/72 [01:48<01:24, 2.72s/it][A
58%|█████▊ | 42/72 [01:51<01:21, 2.72s/it][A
60%|█████▉ | 43/72 [01:54<01:19, 2.72s/it][A
61%|██████ | 44/72 [01:57<01:16, 2.73s/it][A
62%|██████▎ | 45/72 [01:59<01:13, 2.73s/it][A
64%|██████▍ | 46/72 [02:02<01:10, 2.73s/it][A
65%|██████▌ | 47/72 [02:05<01:08, 2.72s/it][A
67%|██████▋ | 48/72 [02:08<01:05, 2.72s/it][A
68%|██████▊ | 49/72 [02:10<01:02, 2.72s/it][A
69%|██████▉ | 50/72 [02:13<00:59, 2.72s/it][A
71%|███████ | 51/72 [02:16<00:57, 2.73s/it][A
72%|███████▏ | 52/72 [02:18<00:54, 2.72s/it][A
74%|███████▎ | 53/72 [02:21<00:51, 2.72s/it][A
75%|███████▌ | 54/72 [02:24<00:48, 2.72s/it][A
76%|███████▋ | 55/72 [02:27<00:46, 2.72s/it][A
78%|███████▊ | 56/72 [02:29<00:43, 2.72s/it][A
79%|███████▉ | 57/72 [02:32<00:40, 2.72s/it][A
81%|████████ | 58/72 [02:35<00:38, 2.72s/it][A
82%|████████▏ | 59/72 [02:37<00:35, 2.72s/it][A
83%|████████▎ | 60/72 [02:40<00:32, 2.72s/it][A
85%|████████▍ | 61/72 [02:43<00:30, 2.73s/it][A
86%|████████▌ | 62/72 [02:46<00:27, 2.73s/it][A
88%|████████▊ | 63/72 [02:48<00:24, 2.72s/it][A
89%|████████▉ | 64/72 [02:51<00:21, 2.72s/it][A
90%|█████████ | 65/72 [02:54<00:19, 2.72s/it][A
92%|█████████▏| 66/72 [02:57<00:16, 2.72s/it][A
93%|█████████▎| 67/72 [02:59<00:13, 2.72s/it][A
94%|█████████▍| 68/72 [03:02<00:10, 2.72s/it][A
96%|█████████▌| 69/72 [03:05<00:08, 2.72s/it][A
97%|█████████▋| 70/72 [03:07<00:05, 2.72s/it][A
99%|█████████▊| 71/72 [03:10<00:02, 2.72s/it][A
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A{'eval_loss': 1.0149483680725098, 'eval_runtime': 195.9333, 'eval_samples_per_second': 5.099, 'eval_steps_per_second': 0.367, 'memory/max_active (GiB)': 77.11, 'memory/max_allocated (GiB)': 77.11, 'memory/device_reserved (GiB)': 77.96, 'epoch': 3.8}
95%|█████████▌| 41743/43939 [67:22:26<3:28:52, 5.71s/it]
100%|██████████| 72/72 [03:11<00:00, 2.25s/it][A
[A
95%|█████████▌| 41744/43939 [67:22:32<39:20:35, 64.53s/it]
95%|█████████▌| 41745/43939 [67:22:37<28:33:40, 46.86s/it]
95%|█████████▌| 41746/43939 [67:22:43<21:01:27, 34.51s/it]
95%|█████████▌| 41747/43939 [67:22:49<15:44:53, 25.86s/it]
95%|█████████▌| 41748/43939 [67:22:54<12:03:39, 19.82s/it]
95%|█████████▌| 41749/43939 [67:23:00<9:29:10, 15.59s/it]
95%|█████████▌| 41750/43939 [67:23:06<7:40:15, 12.62s/it]
{'loss': 0.6515, 'grad_norm': 1.7122375965118408, 'learning_rate': 3.773958796971622e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 56.73, 'epoch': 3.8}
95%|█████████▌| 41750/43939 [67:23:06<7:40:15, 12.62s/it]
95%|█████████▌| 41751/43939 [67:23:11<6:24:28, 10.54s/it]
95%|█████████▌| 41752/43939 [67:23:17<5:31:23, 9.09s/it]
95%|█████████▌| 41753/43939 [67:23:23<4:54:06, 8.07s/it]
95%|█████████▌| 41754/43939 [67:23:29<4:28:39, 7.38s/it]
95%|█████████▌| 41755/43939 [67:23:35<4:13:35, 6.97s/it]
95%|█████████▌| 41756/43939 [67:23:40<3:59:50, 6.59s/it]
95%|█████████▌| 41757/43939 [67:23:46<3:49:59, 6.32s/it]
95%|█████████▌| 41758/43939 [67:23:52<3:43:27, 6.15s/it]
95%|█████████▌| 41759/43939 [67:23:57<3:38:17, 6.01s/it]
95%|█████████▌| 41760/43939 [67:24:03<3:35:10, 5.93s/it]
95%|█████████▌| 41761/43939 [67:24:09<3:32:37, 5.86s/it]
95%|█████████▌| 41762/43939 [67:24:15<3:31:06, 5.82s/it]
95%|█████████▌| 41763/43939 [67:24:20<3:29:18, 5.77s/it]
95%|█████████▌| 41764/43939 [67:24:26<3:28:09, 5.74s/it]
95%|█████████▌| 41765/43939 [67:24:32<3:27:10, 5.72s/it]
95%|█████████▌| 41766/43939 [67:24:37<3:26:40, 5.71s/it]
95%|█████████▌| 41767/43939 [67:24:43<3:26:43, 5.71s/it]
95%|█████████▌| 41768/43939 [67:24:49<3:26:31, 5.71s/it]
95%|█████████▌| 41769/43939 [67:24:54<3:26:18, 5.70s/it]
95%|█████████▌| 41770/43939 [67:25:00<3:26:02, 5.70s/it]
95%|█████████▌| 41771/43939 [67:25:06<3:25:56, 5.70s/it]
95%|█████████▌| 41772/43939 [67:25:12<3:25:53, 5.70s/it]
95%|█████████▌| 41773/43939 [67:25:17<3:26:07, 5.71s/it]
95%|█████████▌| 41774/43939 [67:25:23<3:25:55, 5.71s/it]
95%|█████████▌| 41775/43939 [67:25:29<3:25:58, 5.71s/it]
95%|█████████▌| 41776/43939 [67:25:34<3:26:25, 5.73s/it]
95%|█████████▌| 41777/43939 [67:25:40<3:28:52, 5.80s/it]
95%|█████████▌| 41778/43939 [67:25:46<3:28:14, 5.78s/it]
95%|█████████▌| 41779/43939 [67:25:52<3:27:24, 5.76s/it]
95%|█████████▌| 41780/43939 [67:25:58<3:26:29, 5.74s/it]
95%|█████████▌| 41781/43939 [67:26:03<3:26:10, 5.73s/it]
95%|█████████▌| 41782/43939 [67:26:09<3:26:03, 5.73s/it]
95%|█████████▌| 41783/43939 [67:26:15<3:25:27, 5.72s/it]
95%|█████████▌| 41784/43939 [67:26:20<3:25:13, 5.71s/it]
95%|█████████▌| 41785/43939 [67:26:26<3:25:28, 5.72s/it]
95%|█████████▌| 41786/43939 [67:26:32<3:25:03, 5.71s/it]
95%|█████████▌| 41787/43939 [67:26:38<3:24:55, 5.71s/it]
95%|█████████▌| 41788/43939 [67:26:43<3:24:39, 5.71s/it]
95%|█████████▌| 41789/43939 [67:26:49<3:25:02, 5.72s/it]
95%|█████████▌| 41790/43939 [67:26:55<3:24:40, 5.71s/it]
95%|█████████▌| 41791/43939 [67:27:00<3:24:31, 5.71s/it]
95%|█████████▌| 41792/43939 [67:27:06<3:24:26, 5.71s/it]
95%|█████████▌| 41793/43939 [67:27:12<3:24:05, 5.71s/it]
95%|█████████▌| 41794/43939 [67:27:18<3:23:59, 5.71s/it]
95%|█████████▌| 41795/43939 [67:27:23<3:24:09, 5.71s/it]
95%|█████████▌| 41796/43939 [67:27:29<3:24:26, 5.72s/it]
95%|█████████▌| 41797/43939 [67:27:35<3:24:11, 5.72s/it]
95%|█████████▌| 41798/43939 [67:27:40<3:23:48, 5.71s/it]
95%|█████████▌| 41799/43939 [67:27:46<3:25:41, 5.77s/it]
95%|█████████▌| 41800/43939 [67:27:52<3:24:50, 5.75s/it]
{'loss': 0.6878, 'grad_norm': 1.7058682441711426, 'learning_rate': 3.604009644881751e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.56, 'epoch': 3.81}
95%|█████████▌| 41800/43939 [67:27:52<3:24:50, 5.75s/it]
95%|█████████▌| 41801/43939 [67:27:58<3:24:20, 5.73s/it]
95%|█████████▌| 41802/43939 [67:28:03<3:23:52, 5.72s/it]
95%|█████████▌| 41803/43939 [67:28:09<3:23:19, 5.71s/it]
95%|█████████▌| 41804/43939 [67:28:15<3:23:27, 5.72s/it]
95%|█████████▌| 41805/43939 [67:28:20<3:22:58, 5.71s/it]
95%|█████████▌| 41806/43939 [67:28:26<3:22:46, 5.70s/it]
95%|█████████▌| 41807/43939 [67:28:32<3:22:44, 5.71s/it]
95%|█████████▌| 41808/43939 [67:28:38<3:22:49, 5.71s/it]
95%|█████████▌| 41809/43939 [67:28:43<3:22:30, 5.70s/it]
95%|█████████▌| 41810/43939 [67:28:49<3:22:34, 5.71s/it]
95%|█████████▌| 41811/43939 [67:28:55<3:22:18, 5.70s/it]
95%|█████████▌| 41812/43939 [67:29:00<3:22:07, 5.70s/it]
95%|█████████▌| 41813/43939 [67:29:06<3:21:59, 5.70s/it]
95%|█████████▌| 41814/43939 [67:29:12<3:21:49, 5.70s/it]
95%|█████████▌| 41815/43939 [67:29:17<3:21:40, 5.70s/it]
95%|█████████▌| 41816/43939 [67:29:23<3:22:16, 5.72s/it]
95%|█████████▌| 41817/43939 [67:29:29<3:22:20, 5.72s/it]
95%|█████████▌| 41818/43939 [67:29:35<3:22:09, 5.72s/it]
95%|█████████▌| 41819/43939 [67:29:40<3:21:29, 5.70s/it]
95%|█████████▌| 41820/43939 [67:29:46<3:24:08, 5.78s/it]
95%|█████████▌| 41821/43939 [67:29:52<3:23:11, 5.76s/it]
95%|█████████▌| 41822/43939 [67:29:58<3:22:12, 5.73s/it]
95%|█████████▌| 41823/43939 [67:30:03<3:21:36, 5.72s/it]
95%|█████████▌| 41824/43939 [67:30:09<3:21:30, 5.72s/it]
95%|█████████▌| 41825/43939 [67:30:15<3:21:14, 5.71s/it]
95%|█████████▌| 41826/43939 [67:30:21<3:21:01, 5.71s/it]
95%|█████████▌| 41827/43939 [67:30:26<3:20:50, 5.71s/it]
95%|█████████▌| 41828/43939 [67:30:32<3:20:34, 5.70s/it]
95%|█████████▌| 41829/43939 [67:30:38<3:20:23, 5.70s/it]
95%|█████████▌| 41830/43939 [67:30:43<3:20:08, 5.69s/it]
95%|█████████▌| 41831/43939 [67:30:49<3:20:03, 5.69s/it]
95%|█████████▌| 41832/43939 [67:30:55<3:20:15, 5.70s/it]
95%|█████████▌| 41833/43939 [67:31:00<3:20:10, 5.70s/it]
95%|█████████▌| 41834/43939 [67:31:06<3:19:47, 5.69s/it]
95%|█████████▌| 41835/43939 [67:31:12<3:19:41, 5.69s/it]
95%|█████████▌| 41836/43939 [67:31:17<3:19:39, 5.70s/it]
95%|█████████▌| 41837/43939 [67:31:23<3:20:03, 5.71s/it]
95%|█████████▌| 41838/43939 [67:31:29<3:20:26, 5.72s/it]
95%|█████████▌| 41839/43939 [67:31:35<3:20:03, 5.72s/it]
95%|█████████▌| 41840/43939 [67:31:40<3:19:21, 5.70s/it]
95%|█████████▌| 41841/43939 [67:31:46<3:21:08, 5.75s/it]
95%|█████████▌| 41842/43939 [67:31:52<3:20:19, 5.73s/it]
95%|█████████▌| 41843/43939 [67:31:58<3:19:34, 5.71s/it]
95%|█████████▌| 41844/43939 [67:32:03<3:19:41, 5.72s/it]
95%|█████████▌| 41845/43939 [67:32:09<3:19:27, 5.72s/it]
95%|█████████▌| 41846/43939 [67:32:15<3:19:09, 5.71s/it]
95%|█████████▌| 41847/43939 [67:32:20<3:18:34, 5.70s/it]
95%|█████████▌| 41848/43939 [67:32:26<3:18:50, 5.71s/it]
95%|█████████▌| 41849/43939 [67:32:32<3:18:47, 5.71s/it]
95%|█████████▌| 41850/43939 [67:32:37<3:18:37, 5.70s/it]
{'loss': 0.6574, 'grad_norm': 1.3675822019577026, 'learning_rate': 3.437947968562366e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 503.0, 'epoch': 3.81}
95%|█████████▌| 41850/43939 [67:32:37<3:18:37, 5.70s/it]
95%|█████████▌| 41851/43939 [67:32:43<3:19:04, 5.72s/it]
95%|█████████▌| 41852/43939 [67:32:49<3:18:48, 5.72s/it]
95%|█████████▌| 41853/43939 [67:32:55<3:18:19, 5.70s/it]
95%|█████████▌| 41854/43939 [67:33:00<3:18:47, 5.72s/it]
95%|█████████▌| 41855/43939 [67:33:06<3:18:24, 5.71s/it]
95%|█████████▌| 41856/43939 [67:33:12<3:18:22, 5.71s/it]
95%|█████████▌| 41857/43939 [67:33:17<3:17:47, 5.70s/it]
95%|█████████▌| 41858/43939 [67:33:23<3:18:07, 5.71s/it]
95%|█████████▌| 41859/43939 [67:33:29<3:17:52, 5.71s/it]
95%|█████████▌| 41860/43939 [67:33:35<3:17:14, 5.69s/it]
95%|█████████▌| 41861/43939 [67:33:40<3:16:57, 5.69s/it]
95%|█████████▌| 41862/43939 [67:33:46<3:16:44, 5.68s/it]
95%|█████████▌| 41863/43939 [67:33:52<3:19:21, 5.76s/it]
95%|█████████▌| 41864/43939 [67:33:58<3:18:10, 5.73s/it]
95%|█████████▌| 41865/43939 [67:34:03<3:18:01, 5.73s/it]
95%|█████████▌| 41866/43939 [67:34:09<3:17:48, 5.73s/it]
95%|█████████▌| 41867/43939 [67:34:15<3:17:21, 5.72s/it]
95%|█████████▌| 41868/43939 [67:34:20<3:17:07, 5.71s/it]
95%|█████████▌| 41869/43939 [67:34:26<3:16:34, 5.70s/it]
95%|█████████▌| 41870/43939 [67:34:32<3:16:07, 5.69s/it]
95%|█████████▌| 41871/43939 [67:34:37<3:16:06, 5.69s/it]
95%|█████████▌| 41872/43939 [67:34:43<3:15:55, 5.69s/it]
95%|█████████▌| 41873/43939 [67:34:49<3:16:19, 5.70s/it]
95%|█████████▌| 41874/43939 [67:34:54<3:16:18, 5.70s/it]
95%|█████████▌| 41875/43939 [67:35:00<3:16:06, 5.70s/it]
95%|█████████▌| 41876/43939 [67:35:06<3:15:59, 5.70s/it]
95%|█████████▌| 41877/43939 [67:35:12<3:16:05, 5.71s/it]
95%|█████████▌| 41878/43939 [67:35:17<3:15:41, 5.70s/it]
95%|█████████▌| 41879/43939 [67:35:23<3:15:19, 5.69s/it]
95%|█████████▌| 41880/43939 [67:35:29<3:15:36, 5.70s/it]
95%|█████████▌| 41881/43939 [67:35:34<3:15:41, 5.71s/it]
95%|█████████▌| 41882/43939 [67:35:40<3:15:03, 5.69s/it]
95%|█████████▌| 41883/43939 [67:35:46<3:15:16, 5.70s/it]
95%|█████████▌| 41884/43939 [67:35:51<3:15:29, 5.71s/it]
95%|█████████▌| 41885/43939 [67:35:57<3:18:25, 5.80s/it]
95%|█████████▌| 41886/43939 [67:36:03<3:17:13, 5.76s/it]
95%|█████████▌| 41887/43939 [67:36:09<3:16:21, 5.74s/it]
95%|█████████▌| 41888/43939 [67:36:15<3:16:04, 5.74s/it]
95%|█████████▌| 41889/43939 [67:36:20<3:15:44, 5.73s/it]
95%|█████████▌| 41890/43939 [67:36:26<3:15:17, 5.72s/it]
95%|█████████▌| 41891/43939 [67:36:32<3:15:47, 5.74s/it]
95%|█████████▌| 41892/43939 [67:36:37<3:15:26, 5.73s/it]
95%|█████████▌| 41893/43939 [67:36:43<3:15:15, 5.73s/it]
95%|█████████▌| 41894/43939 [67:36:49<3:15:14, 5.73s/it]
95%|█████████▌| 41895/43939 [67:36:55<3:15:16, 5.73s/it]
95%|█████████▌| 41896/43939 [67:37:00<3:14:35, 5.71s/it]
95%|█████████▌| 41897/43939 [67:37:06<3:14:47, 5.72s/it]
95%|█████████▌| 41898/43939 [67:37:12<3:14:32, 5.72s/it]
95%|█████████▌| 41899/43939 [67:37:17<3:14:01, 5.71s/it]
95%|█████████▌| 41900/43939 [67:37:23<3:14:08, 5.71s/it]
{'loss': 0.657, 'grad_norm': 1.6166784763336182, 'learning_rate': 3.2757763880266746e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.53, 'epoch': 3.81}
95%|█████████▌| 41900/43939 [67:37:23<3:14:08, 5.71s/it]
95%|█████████▌| 41901/43939 [67:37:29<3:14:08, 5.72s/it]
95%|█████████▌| 41902/43939 [67:37:35<3:14:00, 5.71s/it]
95%|█████████▌| 41903/43939 [67:37:40<3:14:18, 5.73s/it]
95%|█████████▌| 41904/43939 [67:37:46<3:13:56, 5.72s/it]
95%|█████████▌| 41905/43939 [67:37:52<3:13:17, 5.70s/it]
95%|█████████▌| 41906/43939 [67:37:57<3:13:29, 5.71s/it]
95%|█████████▌| 41907/43939 [67:38:03<3:15:53, 5.78s/it]
95%|█████████▌| 41908/43939 [67:38:09<3:15:14, 5.77s/it]
95%|█████████▌| 41909/43939 [67:38:15<3:14:21, 5.74s/it]
95%|█████████▌| 41910/43939 [67:38:21<3:13:45, 5.73s/it]
95%|█████████▌| 41911/43939 [67:38:26<3:13:17, 5.72s/it]
95%|█████████▌| 41912/43939 [67:38:32<3:13:33, 5.73s/it]
95%|█████████▌| 41913/43939 [67:38:38<3:12:53, 5.71s/it]
95%|█████████▌| 41914/43939 [67:38:43<3:12:35, 5.71s/it]
95%|█████████▌| 41915/43939 [67:38:49<3:12:57, 5.72s/it]
95%|█████████▌| 41916/43939 [67:38:55<3:12:12, 5.70s/it]
95%|█████████▌| 41917/43939 [67:39:00<3:11:48, 5.69s/it]
95%|█████████▌| 41918/43939 [67:39:06<3:11:50, 5.70s/it]
95%|█████████▌| 41919/43939 [67:39:12<3:11:43, 5.69s/it]
95%|█████████▌| 41920/43939 [67:39:18<3:11:40, 5.70s/it]
95%|█████████▌| 41921/43939 [67:39:23<3:11:31, 5.69s/it]
95%|█████████▌| 41922/43939 [67:39:29<3:11:22, 5.69s/it]
95%|█████████▌| 41923/43939 [67:39:35<3:11:02, 5.69s/it]
95%|█████████▌| 41924/43939 [67:39:40<3:10:54, 5.68s/it]
95%|█████████▌| 41925/43939 [67:39:46<3:11:05, 5.69s/it]
95%|█████████▌| 41926/43939 [67:39:52<3:11:34, 5.71s/it]
95%|█████████▌| 41927/43939 [67:39:57<3:11:07, 5.70s/it]
95%|█████████▌| 41928/43939 [67:40:03<3:13:45, 5.78s/it]
95%|█████████▌| 41929/43939 [67:40:09<3:12:52, 5.76s/it]
95%|█████████▌| 41930/43939 [67:40:15<3:11:59, 5.73s/it]
95%|█████████▌| 41931/43939 [67:40:21<3:12:23, 5.75s/it]
95%|█████████▌| 41932/43939 [67:40:26<3:11:37, 5.73s/it]
95%|█████████▌| 41933/43939 [67:40:32<3:11:40, 5.73s/it]
95%|█████████▌| 41934/43939 [67:40:38<3:11:18, 5.72s/it]
95%|█████████▌| 41935/43939 [67:40:43<3:10:55, 5.72s/it]
95%|█████████▌| 41936/43939 [67:40:49<3:10:31, 5.71s/it]
95%|█████████▌| 41937/43939 [67:40:55<3:10:18, 5.70s/it]
95%|█████████▌| 41938/43939 [67:41:00<3:10:11, 5.70s/it]
95%|█████████▌| 41939/43939 [67:41:06<3:10:37, 5.72s/it]
95%|█████████▌| 41940/43939 [67:41:12<3:10:43, 5.72s/it]
95%|█████████▌| 41941/43939 [67:41:18<3:10:17, 5.71s/it]
95%|█████████▌| 41942/43939 [67:41:23<3:10:08, 5.71s/it]
95%|█████████▌| 41943/43939 [67:41:29<3:10:02, 5.71s/it]
95%|█████████▌| 41944/43939 [67:41:35<3:09:45, 5.71s/it]
95%|█████████▌| 41945/43939 [67:41:40<3:09:24, 5.70s/it]
95%|█████████▌| 41946/43939 [67:41:46<3:09:26, 5.70s/it]
95%|█████████▌| 41947/43939 [67:41:52<3:09:13, 5.70s/it]
95%|█████████▌| 41948/43939 [67:41:58<3:09:04, 5.70s/it]
95%|█████████▌| 41949/43939 [67:42:03<3:09:05, 5.70s/it]
95%|█████████▌| 41950/43939 [67:42:09<3:11:04, 5.76s/it]
{'loss': 0.6499, 'grad_norm': 1.4668976068496704, 'learning_rate': 3.117497461912422e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.17, 'epoch': 3.82}
95%|█████████▌| 41950/43939 [67:42:09<3:11:04, 5.76s/it]
95%|█████████▌| 41951/43939 [67:42:15<3:10:36, 5.75s/it]
95%|█████████▌| 41952/43939 [67:42:21<3:10:21, 5.75s/it]
95%|█████████▌| 41953/43939 [67:42:26<3:10:37, 5.76s/it]
95%|█████████▌| 41954/43939 [67:42:32<3:10:06, 5.75s/it]
95%|█████████▌| 41955/43939 [67:42:38<3:09:47, 5.74s/it]
95%|█████████▌| 41956/43939 [67:42:44<3:09:04, 5.72s/it]
95%|█████████▌| 41957/43939 [67:42:49<3:08:47, 5.72s/it]
95%|█████████▌| 41958/43939 [67:42:55<3:08:33, 5.71s/it]
95%|█████████▌| 41959/43939 [67:43:01<3:08:25, 5.71s/it]
95%|█████████▌| 41960/43939 [67:43:06<3:07:49, 5.69s/it]
95%|█████████▌| 41961/43939 [67:43:12<3:07:42, 5.69s/it]
96%|█████████▌| 41962/43939 [67:43:18<3:07:20, 5.69s/it]
96%|█████████▌| 41963/43939 [67:43:23<3:07:42, 5.70s/it]
96%|█████████▌| 41964/43939 [67:43:29<3:07:56, 5.71s/it]
96%|█████████▌| 41965/43939 [67:43:35<3:07:28, 5.70s/it]
96%|█████████▌| 41966/43939 [67:43:40<3:07:06, 5.69s/it]
96%|█████████▌| 41967/43939 [67:43:46<3:06:58, 5.69s/it]
96%|█████████▌| 41968/43939 [67:43:52<3:07:05, 5.70s/it]
96%|█████████▌| 41969/43939 [67:43:58<3:07:22, 5.71s/it]
96%|█████████▌| 41970/43939 [67:44:03<3:07:17, 5.71s/it]
96%|█████████▌| 41971/43939 [67:44:09<3:06:54, 5.70s/it]
96%|█████████▌| 41972/43939 [67:44:15<3:09:35, 5.78s/it]
96%|█████████▌| 41973/43939 [67:44:21<3:09:28, 5.78s/it]
96%|█████████▌| 41974/43939 [67:44:26<3:08:32, 5.76s/it]
96%|█████████▌| 41975/43939 [67:44:32<3:07:59, 5.74s/it]
96%|█████████▌| 41976/43939 [67:44:38<3:07:30, 5.73s/it]
96%|█████████▌| 41977/43939 [67:44:44<3:07:11, 5.72s/it]
96%|█████████▌| 41978/43939 [67:44:49<3:07:00, 5.72s/it]
96%|█████████▌| 41979/43939 [67:44:55<3:06:34, 5.71s/it]
96%|█████████▌| 41980/43939 [67:45:01<3:06:06, 5.70s/it]
96%|█████████▌| 41981/43939 [67:45:06<3:06:17, 5.71s/it]
96%|█████████▌| 41982/43939 [67:45:12<3:07:12, 5.74s/it]
96%|█████████▌| 41983/43939 [67:45:18<3:07:24, 5.75s/it]
96%|█████████▌| 41984/43939 [67:45:24<3:06:45, 5.73s/it]
96%|█████████▌| 41985/43939 [67:45:29<3:06:18, 5.72s/it]
96%|█████████▌| 41986/43939 [67:45:35<3:05:40, 5.70s/it]
96%|█████████▌| 41987/43939 [67:45:41<3:06:08, 5.72s/it]
96%|█████████▌| 41988/43939 [67:45:46<3:05:49, 5.71s/it]
96%|█████████▌| 41989/43939 [67:45:52<3:05:49, 5.72s/it]
96%|█████████▌| 41990/43939 [67:45:58<3:05:40, 5.72s/it]
96%|█████████▌| 41991/43939 [67:46:04<3:05:58, 5.73s/it]
96%|█████████▌| 41992/43939 [67:46:09<3:05:42, 5.72s/it]
96%|█████████▌| 41993/43939 [67:46:15<3:07:46, 5.79s/it]
96%|█████████▌| 41994/43939 [67:46:21<3:07:07, 5.77s/it]
96%|█████████▌| 41995/43939 [67:46:27<3:06:23, 5.75s/it]
96%|█████████▌| 41996/43939 [67:46:32<3:05:52, 5.74s/it]
96%|█████████▌| 41997/43939 [67:46:38<3:05:32, 5.73s/it]
96%|█████████▌| 41998/43939 [67:46:44<3:04:57, 5.72s/it]
96%|█████████▌| 41999/43939 [67:46:50<3:04:44, 5.71s/it]
96%|█████████▌| 42000/43939 [67:46:55<3:04:49, 5.72s/it]
{'loss': 0.6454, 'grad_norm': 1.5816718339920044, 'learning_rate': 2.9631136874415934e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 478.53, 'epoch': 3.82}
96%|█████████▌| 42000/43939 [67:46:55<3:04:49, 5.72s/it]
96%|█████████▌| 42001/43939 [67:47:01<3:04:51, 5.72s/it]
96%|█████████▌| 42002/43939 [67:47:07<3:05:09, 5.74s/it]
96%|█████████▌| 42003/43939 [67:47:13<3:05:14, 5.74s/it]
96%|█████████▌| 42004/43939 [67:47:18<3:04:43, 5.73s/it]
96%|█████████▌| 42005/43939 [67:47:24<3:04:17, 5.72s/it]
96%|█████████▌| 42006/43939 [67:47:30<3:04:41, 5.73s/it]
96%|█████████▌| 42007/43939 [67:47:35<3:04:07, 5.72s/it]
96%|█████████▌| 42008/43939 [67:47:41<3:03:34, 5.70s/it]
96%|█████████▌| 42009/43939 [67:47:47<3:03:52, 5.72s/it]
96%|█████████▌| 42010/43939 [67:47:53<3:04:11, 5.73s/it]
96%|█████████▌| 42011/43939 [67:47:58<3:03:56, 5.72s/it]
96%|█████████▌| 42012/43939 [67:48:04<3:04:07, 5.73s/it]
96%|█████████▌| 42013/43939 [67:48:10<3:04:22, 5.74s/it]
96%|█████████▌| 42014/43939 [67:48:15<3:03:39, 5.72s/it]
96%|█████████▌| 42015/43939 [67:48:21<3:05:33, 5.79s/it]
96%|█████████▌| 42016/43939 [67:48:27<3:04:46, 5.76s/it]
96%|█████████▌| 42017/43939 [67:48:33<3:04:21, 5.76s/it]
96%|█████████▌| 42018/43939 [67:48:39<3:03:36, 5.73s/it]
96%|█████████▌| 42019/43939 [67:48:44<3:03:16, 5.73s/it]
96%|█████████▌| 42020/43939 [67:48:50<3:02:50, 5.72s/it]
96%|█████████▌| 42021/43939 [67:48:56<3:02:20, 5.70s/it]
96%|█████████▌| 42022/43939 [67:49:01<3:01:53, 5.69s/it]
96%|█████████▌| 42023/43939 [67:49:07<3:01:48, 5.69s/it]
96%|█████████▌| 42024/43939 [67:49:13<3:01:32, 5.69s/it]
96%|█████████▌| 42025/43939 [67:49:18<3:01:49, 5.70s/it]
96%|█████████▌| 42026/43939 [67:49:24<3:02:05, 5.71s/it]
96%|█████████▌| 42027/43939 [67:49:30<3:02:18, 5.72s/it]
96%|█████████▌| 42028/43939 [67:49:36<3:02:26, 5.73s/it]
96%|█████████▌| 42029/43939 [67:49:41<3:02:09, 5.72s/it]
96%|█████████▌| 42030/43939 [67:49:47<3:01:41, 5.71s/it]
96%|█████████▌| 42031/43939 [67:49:53<3:01:01, 5.69s/it]
96%|█████████▌| 42032/43939 [67:49:58<3:01:01, 5.70s/it]
96%|█████████▌| 42033/43939 [67:50:04<3:01:27, 5.71s/it]
96%|█████████▌| 42034/43939 [67:50:10<3:01:07, 5.70s/it]
96%|█████████▌| 42035/43939 [67:50:16<3:01:13, 5.71s/it]
96%|█████████▌| 42036/43939 [67:50:21<3:03:29, 5.79s/it]
96%|█████████▌| 42037/43939 [67:50:27<3:02:33, 5.76s/it]
96%|█████████▌| 42038/43939 [67:50:33<3:01:53, 5.74s/it]
96%|█████████▌| 42039/43939 [67:50:39<3:02:02, 5.75s/it]
96%|█████████▌| 42040/43939 [67:50:44<3:01:26, 5.73s/it]
96%|█████████▌| 42041/43939 [67:50:50<3:01:12, 5.73s/it]
96%|█████████▌| 42042/43939 [67:50:56<3:01:04, 5.73s/it]
96%|█████████▌| 42043/43939 [67:51:02<3:00:57, 5.73s/it]
96%|█████████▌| 42044/43939 [67:51:07<3:00:48, 5.72s/it]
96%|█████████▌| 42045/43939 [67:51:13<3:00:56, 5.73s/it]
96%|█████████▌| 42046/43939 [67:51:19<3:00:48, 5.73s/it]
96%|█████████▌| 42047/43939 [67:51:24<3:00:17, 5.72s/it]
96%|█████████▌| 42048/43939 [67:51:30<3:00:07, 5.72s/it]
96%|█████████▌| 42049/43939 [67:51:36<3:00:26, 5.73s/it]
96%|█████████▌| 42050/43939 [67:51:42<3:00:13, 5.72s/it]
{'loss': 0.6687, 'grad_norm': 1.4421151876449585, 'learning_rate': 2.812627500381082e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.74, 'epoch': 3.83}
96%|█████████▌| 42050/43939 [67:51:42<3:00:13, 5.72s/it]
96%|█████████▌| 42051/43939 [67:51:47<3:00:21, 5.73s/it]
96%|█████████▌| 42052/43939 [67:51:53<2:59:58, 5.72s/it]
96%|█████████▌| 42053/43939 [67:51:59<2:59:41, 5.72s/it]
96%|█████████▌| 42054/43939 [67:52:04<2:59:18, 5.71s/it]
96%|█████████▌| 42055/43939 [67:52:10<2:59:28, 5.72s/it]
96%|█████████▌| 42056/43939 [67:52:16<2:59:27, 5.72s/it]
96%|█████████▌| 42057/43939 [67:52:22<2:59:15, 5.72s/it]
96%|█████████▌| 42058/43939 [67:52:28<3:01:12, 5.78s/it]
96%|█████████▌| 42059/43939 [67:52:33<3:00:11, 5.75s/it]
96%|█████████▌| 42060/43939 [67:52:39<2:59:36, 5.74s/it]
96%|█████████▌| 42061/43939 [67:52:45<2:59:00, 5.72s/it]
96%|█████████▌| 42062/43939 [67:52:50<2:59:11, 5.73s/it]
96%|█████████▌| 42063/43939 [67:52:56<2:58:49, 5.72s/it]
96%|█████████▌| 42064/43939 [67:53:02<2:59:40, 5.75s/it]
96%|█████████▌| 42065/43939 [67:53:08<2:59:26, 5.74s/it]
96%|█████████▌| 42066/43939 [67:53:13<2:58:50, 5.73s/it]
96%|█████████▌| 42067/43939 [67:53:19<2:58:47, 5.73s/it]
96%|█████████▌| 42068/43939 [67:53:25<2:58:27, 5.72s/it]
96%|█████████▌| 42069/43939 [67:53:30<2:58:08, 5.72s/it]
96%|█████████▌| 42070/43939 [67:53:36<2:58:09, 5.72s/it]
96%|█████████▌| 42071/43939 [67:53:42<2:57:53, 5.71s/it]
96%|█████████▌| 42072/43939 [67:53:48<2:57:41, 5.71s/it]
96%|█████████▌| 42073/43939 [67:53:53<2:57:38, 5.71s/it]
96%|█████████▌| 42074/43939 [67:53:59<2:57:40, 5.72s/it]
96%|█████████▌| 42075/43939 [67:54:05<2:57:30, 5.71s/it]
96%|█████████▌| 42076/43939 [67:54:10<2:57:35, 5.72s/it]
96%|█████████▌| 42077/43939 [67:54:16<2:57:39, 5.72s/it]
96%|█████████▌| 42078/43939 [67:54:22<2:57:15, 5.72s/it]
96%|█████████▌| 42079/43939 [67:54:28<2:57:13, 5.72s/it]
96%|█████████▌| 42080/43939 [67:54:33<2:58:57, 5.78s/it]
96%|█████████▌| 42081/43939 [67:54:39<2:58:43, 5.77s/it]
96%|█████████▌| 42082/43939 [67:54:45<2:58:28, 5.77s/it]
96%|█████████▌| 42083/43939 [67:54:51<2:57:39, 5.74s/it]
96%|█████████▌| 42084/43939 [67:54:56<2:57:20, 5.74s/it]
96%|█████████▌| 42085/43939 [67:55:02<2:57:22, 5.74s/it]
96%|█████████▌| 42086/43939 [67:55:08<2:57:01, 5.73s/it]
96%|█████████▌| 42087/43939 [67:55:14<2:57:08, 5.74s/it]
96%|█████████▌| 42088/43939 [67:55:19<2:56:52, 5.73s/it]
96%|█████████▌| 42089/43939 [67:55:25<2:56:25, 5.72s/it]
96%|█████████▌| 42090/43939 [67:55:31<2:56:15, 5.72s/it]
96%|█████████▌| 42091/43939 [67:55:36<2:56:04, 5.72s/it]
96%|█████████▌| 42092/43939 [67:55:42<2:55:47, 5.71s/it]
96%|█████████▌| 42093/43939 [67:55:48<2:55:22, 5.70s/it]
96%|█████████▌| 42094/43939 [67:55:54<2:55:12, 5.70s/it]
96%|█████████▌| 42095/43939 [67:55:59<2:55:16, 5.70s/it]
96%|█████████▌| 42096/43939 [67:56:05<2:55:08, 5.70s/it]
96%|█████████▌| 42097/43939 [67:56:11<2:54:41, 5.69s/it]
96%|█████████▌| 42098/43939 [67:56:16<2:54:27, 5.69s/it]
96%|█████████▌| 42099/43939 [67:56:22<2:54:34, 5.69s/it]
96%|█████████▌| 42100/43939 [67:56:28<2:54:57, 5.71s/it]
{'loss': 0.6691, 'grad_norm': 1.2845165729522705, 'learning_rate': 2.666041275004111e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 524.17, 'epoch': 3.83}
96%|█████████▌| 42100/43939 [67:56:28<2:54:57, 5.71s/it]
96%|█████████▌| 42101/43939 [67:56:34<2:57:03, 5.78s/it]
96%|█████████▌| 42102/43939 [67:56:39<2:56:42, 5.77s/it]
96%|█████████▌| 42103/43939 [67:56:45<2:56:08, 5.76s/it]
96%|█████████▌| 42104/43939 [67:56:51<2:55:33, 5.74s/it]
96%|█████████▌| 42105/43939 [67:56:57<2:55:12, 5.73s/it]
96%|█████████▌| 42106/43939 [67:57:02<2:55:04, 5.73s/it]
96%|█████████▌| 42107/43939 [67:57:08<2:54:46, 5.72s/it]
96%|█████████▌| 42108/43939 [67:57:14<2:54:16, 5.71s/it]
96%|█████████▌| 42109/43939 [67:57:19<2:53:50, 5.70s/it]
96%|█████████▌| 42110/43939 [67:57:25<2:53:33, 5.69s/it]
96%|█████████▌| 42111/43939 [67:57:31<2:53:29, 5.69s/it]
96%|█████████▌| 42112/43939 [67:57:36<2:53:26, 5.70s/it]
96%|█████████▌| 42113/43939 [67:57:42<2:53:13, 5.69s/it]
96%|█████████▌| 42114/43939 [67:57:48<2:53:10, 5.69s/it]
96%|█████████▌| 42115/43939 [67:57:54<2:53:09, 5.70s/it]
96%|█████████▌| 42116/43939 [67:57:59<2:52:51, 5.69s/it]
96%|█████████▌| 42117/43939 [67:58:05<2:52:58, 5.70s/it]
96%|█████████▌| 42118/43939 [67:58:11<2:52:48, 5.69s/it]
96%|█████████▌| 42119/43939 [67:58:16<2:53:01, 5.70s/it]
96%|█████████▌| 42120/43939 [67:58:22<2:52:53, 5.70s/it]
96%|█████████▌| 42121/43939 [67:58:28<2:53:43, 5.73s/it]
96%|█████████▌| 42122/43939 [67:58:34<2:53:07, 5.72s/it]
96%|█████████▌| 42123/43939 [67:58:39<2:55:07, 5.79s/it]
96%|█████████▌| 42124/43939 [67:58:45<2:55:06, 5.79s/it]
96%|█████████▌| 42125/43939 [67:58:51<2:54:21, 5.77s/it]
96%|█████████▌| 42126/43939 [67:58:57<2:53:44, 5.75s/it]
96%|█████████▌| 42127/43939 [67:59:02<2:53:07, 5.73s/it]
96%|█████████▌| 42128/43939 [67:59:08<2:52:43, 5.72s/it]
96%|█████████▌| 42129/43939 [67:59:14<2:52:21, 5.71s/it]
96%|█████████▌| 42130/43939 [67:59:20<2:52:38, 5.73s/it]
96%|█████████▌| 42131/43939 [67:59:25<2:52:20, 5.72s/it]
96%|█████████▌| 42132/43939 [67:59:31<2:52:18, 5.72s/it]
96%|█████████▌| 42133/43939 [67:59:37<2:51:50, 5.71s/it]
96%|█████████▌| 42134/43939 [67:59:42<2:51:32, 5.70s/it]
96%|█████████▌| 42135/43939 [67:59:48<2:51:53, 5.72s/it]
96%|█████████▌| 42136/43939 [67:59:54<2:52:04, 5.73s/it]
96%|█████████▌| 42137/43939 [68:00:00<2:52:01, 5.73s/it]
96%|█████████▌| 42138/43939 [68:00:05<2:51:47, 5.72s/it]
96%|█████████▌| 42139/43939 [68:00:11<2:51:33, 5.72s/it]
96%|█████████▌| 42140/43939 [68:00:17<2:51:21, 5.72s/it]
96%|█████████▌| 42141/43939 [68:00:22<2:51:14, 5.71s/it]
96%|█████████▌| 42142/43939 [68:00:28<2:51:20, 5.72s/it]
96%|█████████▌| 42143/43939 [68:00:34<2:51:33, 5.73s/it]
96%|█████████▌| 42144/43939 [68:00:40<2:50:54, 5.71s/it]
96%|█████████▌| 42145/43939 [68:00:45<2:52:43, 5.78s/it]
96%|█████████▌| 42146/43939 [68:00:51<2:51:55, 5.75s/it]
96%|█████████▌| 42147/43939 [68:00:57<2:51:30, 5.74s/it]
96%|█████████▌| 42148/43939 [68:01:03<2:50:59, 5.73s/it]
96%|█████████▌| 42149/43939 [68:01:08<2:50:54, 5.73s/it]
96%|█████████▌| 42150/43939 [68:01:14<2:50:28, 5.72s/it]
{'loss': 0.6557, 'grad_norm': 1.8156490325927734, 'learning_rate': 2.5233573240528443e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.51, 'epoch': 3.84}
96%|█████████▌| 42150/43939 [68:01:14<2:50:28, 5.72s/it]
96%|█████████▌| 42151/43939 [68:01:20<2:50:13, 5.71s/it]
96%|█████████▌| 42152/43939 [68:01:25<2:49:55, 5.71s/it]
96%|█████████▌| 42153/43939 [68:01:31<2:50:29, 5.73s/it]
96%|█████████▌| 42154/43939 [68:01:37<2:49:55, 5.71s/it]
96%|█████████▌| 42155/43939 [68:01:43<2:49:45, 5.71s/it]
96%|█████████▌| 42156/43939 [68:01:48<2:49:58, 5.72s/it]
96%|█████████▌| 42157/43939 [68:01:54<2:49:43, 5.71s/it]
96%|█████████▌| 42158/43939 [68:02:00<2:49:36, 5.71s/it]
96%|█████████▌| 42159/43939 [68:02:05<2:49:05, 5.70s/it]
96%|█████████▌| 42160/43939 [68:02:11<2:49:05, 5.70s/it]
96%|█████████▌| 42161/43939 [68:02:17<2:49:16, 5.71s/it]
96%|█████████▌| 42162/43939 [68:02:22<2:48:46, 5.70s/it]
96%|█████████▌| 42163/43939 [68:02:28<2:48:48, 5.70s/it]
96%|█████████▌| 42164/43939 [68:02:34<2:48:39, 5.70s/it]
96%|█████████▌| 42165/43939 [68:02:40<2:48:31, 5.70s/it]
96%|█████████▌| 42166/43939 [68:02:46<2:50:29, 5.77s/it]
96%|█████████▌| 42167/43939 [68:02:51<2:49:34, 5.74s/it]
96%|█████████▌| 42168/43939 [68:02:57<2:49:07, 5.73s/it]
96%|█████████▌| 42169/43939 [68:03:03<2:49:00, 5.73s/it]
96%|█████████▌| 42170/43939 [68:03:08<2:48:25, 5.71s/it]
96%|█████████▌| 42171/43939 [68:03:14<2:48:06, 5.70s/it]
96%|█████████▌| 42172/43939 [68:03:20<2:47:47, 5.70s/it]
96%|█████████▌| 42173/43939 [68:03:25<2:47:44, 5.70s/it]
96%|█████████▌| 42174/43939 [68:03:31<2:47:36, 5.70s/it]
96%|█████████▌| 42175/43939 [68:03:37<2:47:22, 5.69s/it]
96%|█████████▌| 42176/43939 [68:03:42<2:47:20, 5.70s/it]
96%|█████████▌| 42177/43939 [68:03:48<2:47:27, 5.70s/it]
96%|█████████▌| 42178/43939 [68:03:54<2:47:20, 5.70s/it]
96%|█████████▌| 42179/43939 [68:04:00<2:47:36, 5.71s/it]
96%|█████████▌| 42180/43939 [68:04:05<2:47:09, 5.70s/it]
96%|█████████▌| 42181/43939 [68:04:11<2:47:22, 5.71s/it]
96%|█████████▌| 42182/43939 [68:04:17<2:47:36, 5.72s/it]
96%|█████████▌| 42183/43939 [68:04:22<2:47:18, 5.72s/it]
96%|█████████▌| 42184/43939 [68:04:28<2:47:15, 5.72s/it]
96%|█████████▌| 42185/43939 [68:04:34<2:46:54, 5.71s/it]
96%|█████████▌| 42186/43939 [68:04:40<2:46:42, 5.71s/it]
96%|█████████▌| 42187/43939 [68:04:46<2:48:36, 5.77s/it]
96%|█████████▌| 42188/43939 [68:04:51<2:47:51, 5.75s/it]
96%|█████████▌| 42189/43939 [68:04:57<2:47:07, 5.73s/it]
96%|█████████▌| 42190/43939 [68:05:03<2:46:46, 5.72s/it]
96%|█████████▌| 42191/43939 [68:05:08<2:46:48, 5.73s/it]
96%|█████████▌| 42192/43939 [68:05:14<2:46:09, 5.71s/it]
96%|█████████▌| 42193/43939 [68:05:20<2:46:01, 5.71s/it]
96%|█████████▌| 42194/43939 [68:05:25<2:46:06, 5.71s/it]
96%|█████████▌| 42195/43939 [68:05:31<2:45:52, 5.71s/it]
96%|█████████▌| 42196/43939 [68:05:37<2:45:34, 5.70s/it]
96%|█████████▌| 42197/43939 [68:05:43<2:45:33, 5.70s/it]
96%|█████████▌| 42198/43939 [68:05:48<2:45:45, 5.71s/it]
96%|█████████▌| 42199/43939 [68:05:54<2:45:36, 5.71s/it]
96%|█████████▌| 42200/43939 [68:06:00<2:45:56, 5.73s/it]
{'loss': 0.6619, 'grad_norm': 1.5084383487701416, 'learning_rate': 2.3845778987019173e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 470.58, 'epoch': 3.84}
96%|█████████▌| 42200/43939 [68:06:00<2:45:56, 5.73s/it]
96%|█████████▌| 42201/43939 [68:06:05<2:45:30, 5.71s/it]
96%|█████████▌| 42202/43939 [68:06:11<2:45:54, 5.73s/it]
96%|█████████▌| 42203/43939 [68:06:17<2:45:29, 5.72s/it]
96%|█████████▌| 42204/43939 [68:06:23<2:45:35, 5.73s/it]
96%|█████████▌| 42205/43939 [68:06:28<2:45:00, 5.71s/it]
96%|█████████▌| 42206/43939 [68:06:34<2:44:31, 5.70s/it]
96%|█████████▌| 42207/43939 [68:06:40<2:44:15, 5.69s/it]
96%|█████████▌| 42208/43939 [68:06:45<2:44:11, 5.69s/it]
96%|█████████▌| 42209/43939 [68:06:51<2:46:18, 5.77s/it]
96%|█████████▌| 42210/43939 [68:06:57<2:45:55, 5.76s/it]
96%|█████████▌| 42211/43939 [68:07:03<2:45:27, 5.74s/it]
96%|█████████▌| 42212/43939 [68:07:08<2:45:01, 5.73s/it]
96%|█████████▌| 42213/43939 [68:07:14<2:44:48, 5.73s/it]
96%|█████████▌| 42214/43939 [68:07:20<2:44:27, 5.72s/it]
96%|█████████▌| 42215/43939 [68:07:26<2:44:31, 5.73s/it]
96%|█████████▌| 42216/43939 [68:07:31<2:44:03, 5.71s/it]
96%|█████████▌| 42217/43939 [68:07:37<2:43:48, 5.71s/it]
96%|█████████▌| 42218/43939 [68:07:43<2:43:38, 5.71s/it]
96%|█████████▌| 42219/43939 [68:07:48<2:43:36, 5.71s/it]
96%|█████████▌| 42220/43939 [68:07:54<2:43:14, 5.70s/it]
96%|█████████▌| 42221/43939 [68:08:00<2:43:34, 5.71s/it]
96%|█████████▌| 42222/43939 [68:08:06<2:43:33, 5.72s/it]
96%|█████████▌| 42223/43939 [68:08:11<2:43:14, 5.71s/it]
96%|█████████▌| 42224/43939 [68:08:17<2:43:26, 5.72s/it]
96%|█████████▌| 42225/43939 [68:08:23<2:42:52, 5.70s/it]
96%|█████████▌| 42226/43939 [68:08:28<2:43:00, 5.71s/it]
96%|█████████▌| 42227/43939 [68:08:34<2:42:58, 5.71s/it]
96%|█████████▌| 42228/43939 [68:08:40<2:42:48, 5.71s/it]
96%|█████████▌| 42229/43939 [68:08:45<2:42:37, 5.71s/it]
96%|█████████▌| 42230/43939 [68:08:51<2:42:58, 5.72s/it]
96%|█████████▌| 42231/43939 [68:08:57<2:44:43, 5.79s/it]
96%|█████████▌| 42232/43939 [68:09:03<2:43:43, 5.75s/it]
96%|█████████▌| 42233/43939 [68:09:09<2:43:04, 5.74s/it]
96%|█████████▌| 42234/43939 [68:09:14<2:42:44, 5.73s/it]
96%|█████████▌| 42235/43939 [68:09:20<2:42:28, 5.72s/it]
96%|█████████▌| 42236/43939 [68:09:26<2:42:53, 5.74s/it]
96%|█████████▌| 42237/43939 [68:09:31<2:42:46, 5.74s/it]
96%|█████████▌| 42238/43939 [68:09:37<2:42:26, 5.73s/it]
96%|█████████▌| 42239/43939 [68:09:43<2:42:18, 5.73s/it]
96%|█████████▌| 42240/43939 [68:09:49<2:42:07, 5.73s/it]
96%|█████████▌| 42241/43939 [68:09:54<2:41:52, 5.72s/it]
96%|█████████▌| 42242/43939 [68:10:00<2:41:31, 5.71s/it]
96%|█████████▌| 42243/43939 [68:10:06<2:40:55, 5.69s/it]
96%|█████████▌| 42244/43939 [68:10:11<2:41:01, 5.70s/it]
96%|█████████▌| 42245/43939 [68:10:17<2:40:56, 5.70s/it]
96%|█████████▌| 42246/43939 [68:10:23<2:41:00, 5.71s/it]
96%|█████████▌| 42247/43939 [68:10:28<2:40:49, 5.70s/it]
96%|█████████▌| 42248/43939 [68:10:34<2:40:52, 5.71s/it]
96%|█████████▌| 42249/43939 [68:10:40<2:40:47, 5.71s/it]
96%|█████████▌| 42250/43939 [68:10:46<2:40:30, 5.70s/it]
{'loss': 0.6493, 'grad_norm': 1.76215398311615, 'learning_rate': 2.2497051885228827e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 494.36, 'epoch': 3.85}
96%|█████████▌| 42250/43939 [68:10:46<2:40:30, 5.70s/it]
96%|█████████▌| 42251/43939 [68:10:51<2:40:39, 5.71s/it]
96%|█████████▌| 42252/43939 [68:10:57<2:42:41, 5.79s/it]
96%|█████████▌| 42253/43939 [68:11:03<2:41:38, 5.75s/it]
96%|█████████▌| 42254/43939 [68:11:09<2:41:04, 5.74s/it]
96%|█████████▌| 42255/43939 [68:11:14<2:40:35, 5.72s/it]
96%|█████████▌| 42256/43939 [68:11:20<2:39:58, 5.70s/it]
96%|█████████▌| 42257/43939 [68:11:26<2:39:54, 5.70s/it]
96%|█████████▌| 42258/43939 [68:11:31<2:39:35, 5.70s/it]
96%|█████████▌| 42259/43939 [68:11:37<2:39:35, 5.70s/it]
96%|█████████▌| 42260/43939 [68:11:43<2:39:53, 5.71s/it]
96%|█████████▌| 42261/43939 [68:11:49<2:39:37, 5.71s/it]
96%|█████████▌| 42262/43939 [68:11:54<2:39:49, 5.72s/it]
96%|█████████▌| 42263/43939 [68:12:00<2:39:23, 5.71s/it]
96%|█████████▌| 42264/43939 [68:12:06<2:39:10, 5.70s/it]
96%|█████████▌| 42265/43939 [68:12:11<2:39:39, 5.72s/it]
96%|█████████▌| 42266/43939 [68:12:17<2:39:11, 5.71s/it]
96%|█████████▌| 42267/43939 [68:12:23<2:39:08, 5.71s/it]
96%|█████████▌| 42268/43939 [68:12:29<2:38:45, 5.70s/it]
96%|█████████▌| 42269/43939 [68:12:34<2:38:45, 5.70s/it]
96%|█████████▌| 42270/43939 [68:12:40<2:38:47, 5.71s/it]
96%|█████████▌| 42271/43939 [68:12:46<2:38:36, 5.71s/it]
96%|█████████▌| 42272/43939 [68:12:51<2:38:27, 5.70s/it]
96%|█████████▌| 42273/43939 [68:12:57<2:38:06, 5.69s/it]
96%|█████████▌| 42274/43939 [68:13:03<2:39:54, 5.76s/it]
96%|█████████▌| 42275/43939 [68:13:09<2:39:21, 5.75s/it]
96%|█████████▌| 42276/43939 [68:13:14<2:38:50, 5.73s/it]
96%|█████████▌| 42277/43939 [68:13:20<2:38:31, 5.72s/it]
96%|█████████▌| 42278/43939 [68:13:26<2:38:20, 5.72s/it]
96%|█████████▌| 42279/43939 [68:13:31<2:38:05, 5.71s/it]
96%|█████████▌| 42280/43939 [68:13:37<2:37:59, 5.71s/it]
96%|█████████▌| 42281/43939 [68:13:43<2:37:38, 5.70s/it]
96%|█████████▌| 42282/43939 [68:13:49<2:37:14, 5.69s/it]
96%|█████████▌| 42283/43939 [68:13:54<2:37:14, 5.70s/it]
96%|█████████▌| 42284/43939 [68:14:00<2:37:26, 5.71s/it]
96%|█████████▌| 42285/43939 [68:14:06<2:37:19, 5.71s/it]
96%|█████████▌| 42286/43939 [68:14:11<2:36:44, 5.69s/it]
96%|█████████▌| 42287/43939 [68:14:17<2:36:57, 5.70s/it]
96%|█████████▌| 42288/43939 [68:14:23<2:36:42, 5.70s/it]
96%|█████████▌| 42289/43939 [68:14:28<2:36:56, 5.71s/it]
96%|█████████▌| 42290/43939 [68:14:34<2:36:47, 5.70s/it]
96%|█████████▌| 42291/43939 [68:14:40<2:37:09, 5.72s/it]
96%|█████████▋| 42292/43939 [68:14:46<2:37:20, 5.73s/it]
96%|█████████▋| 42293/43939 [68:14:51<2:36:45, 5.71s/it]
96%|█████████▋| 42294/43939 [68:14:57<2:36:25, 5.71s/it]
96%|█████████▋| 42295/43939 [68:15:03<2:38:11, 5.77s/it]
96%|█████████▋| 42296/43939 [68:15:09<2:37:40, 5.76s/it]
96%|█████████▋| 42297/43939 [68:15:14<2:37:15, 5.75s/it]
96%|█████████▋| 42298/43939 [68:15:20<2:36:39, 5.73s/it]
96%|█████████▋| 42299/43939 [68:15:26<2:36:38, 5.73s/it]
96%|█████████▋| 42300/43939 [68:15:32<2:36:16, 5.72s/it]
{'loss': 0.6345, 'grad_norm': 1.689795970916748, 'learning_rate': 2.1187413214497076e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 446.55, 'epoch': 3.85}
96%|█████████▋| 42300/43939 [68:15:32<2:36:16, 5.72s/it]
96%|█████████▋| 42301/43939 [68:15:37<2:36:41, 5.74s/it]
96%|█████████▋| 42302/43939 [68:15:43<2:35:57, 5.72s/it]
96%|█████████▋| 42303/43939 [68:15:49<2:35:54, 5.72s/it]
96%|█████████▋| 42304/43939 [68:15:54<2:35:55, 5.72s/it]
96%|█████████▋| 42305/43939 [68:16:00<2:35:43, 5.72s/it]
96%|█████████▋| 42306/43939 [68:16:06<2:35:25, 5.71s/it]
96%|█████████▋| 42307/43939 [68:16:12<2:35:09, 5.70s/it]
96%|█████████▋| 42308/43939 [68:16:17<2:34:59, 5.70s/it]
96%|█████████▋| 42309/43939 [68:16:23<2:34:38, 5.69s/it]
96%|█████████▋| 42310/43939 [68:16:29<2:35:00, 5.71s/it]
96%|█████████▋| 42311/43939 [68:16:34<2:34:38, 5.70s/it]
96%|█████████▋| 42312/43939 [68:16:40<2:35:01, 5.72s/it]
96%|█████████▋| 42313/43939 [68:16:46<2:34:50, 5.71s/it]
96%|█████████▋| 42314/43939 [68:16:52<2:35:03, 5.72s/it]
96%|█████████▋| 42315/43939 [68:16:57<2:34:57, 5.72s/it]
96%|█████████▋| 42316/43939 [68:17:03<2:34:37, 5.72s/it]
96%|█████████▋| 42317/43939 [68:17:09<2:36:37, 5.79s/it]
96%|█████████▋| 42318/43939 [68:17:15<2:35:37, 5.76s/it]
96%|█████████▋| 42319/43939 [68:17:20<2:34:56, 5.74s/it]
96%|█████████▋| 42320/43939 [68:17:26<2:34:39, 5.73s/it]
96%|█████████▋| 42321/43939 [68:17:32<2:34:37, 5.73s/it]
96%|█████████▋| 42322/43939 [68:17:37<2:34:07, 5.72s/it]
96%|█████████▋| 42323/43939 [68:17:43<2:34:39, 5.74s/it]
96%|█████████▋| 42324/43939 [68:17:49<2:34:08, 5.73s/it]
96%|█████████▋| 42325/43939 [68:17:55<2:33:35, 5.71s/it]
96%|█████████▋| 42326/43939 [68:18:00<2:33:36, 5.71s/it]
96%|█████████▋| 42327/43939 [68:18:06<2:33:14, 5.70s/it]
96%|█████████▋| 42328/43939 [68:18:12<2:33:24, 5.71s/it]
96%|█████████▋| 42329/43939 [68:18:17<2:33:05, 5.71s/it]
96%|█████████▋| 42330/43939 [68:18:23<2:32:44, 5.70s/it]
96%|█████████▋| 42331/43939 [68:18:29<2:32:52, 5.70s/it]
96%|█████████▋| 42332/43939 [68:18:35<2:32:51, 5.71s/it]
96%|█████████▋| 42333/43939 [68:18:40<2:32:42, 5.71s/it]
96%|█████████▋| 42334/43939 [68:18:46<2:32:37, 5.71s/it]
96%|█████████▋| 42335/43939 [68:18:52<2:32:27, 5.70s/it]
96%|█████████▋| 42336/43939 [68:18:57<2:32:11, 5.70s/it]
96%|█████████▋| 42337/43939 [68:19:03<2:32:07, 5.70s/it]
96%|█████████▋| 42338/43939 [68:19:09<2:33:43, 5.76s/it]
96%|█████████▋| 42339/43939 [68:19:15<2:33:25, 5.75s/it]
96%|█████████▋| 42340/43939 [68:19:20<2:33:05, 5.74s/it]
96%|█████████▋| 42341/43939 [68:19:26<2:32:33, 5.73s/it]
96%|█████████▋| 42342/43939 [68:19:32<2:32:09, 5.72s/it]
96%|█████████▋| 42343/43939 [68:19:38<2:32:26, 5.73s/it]
96%|█████████▋| 42344/43939 [68:19:43<2:32:01, 5.72s/it]
96%|█████████▋| 42345/43939 [68:19:49<2:31:24, 5.70s/it]
96%|█████████▋| 42346/43939 [68:19:55<2:31:00, 5.69s/it]
96%|█████████▋| 42347/43939 [68:20:00<2:31:15, 5.70s/it]
96%|█████████▋| 42348/43939 [68:20:06<2:31:14, 5.70s/it]
96%|█████████▋| 42349/43939 [68:20:12<2:30:48, 5.69s/it]
96%|█████████▋| 42350/43939 [68:20:17<2:30:38, 5.69s/it]
{'loss': 0.638, 'grad_norm': 1.707517385482788, 'learning_rate': 1.9916883637451645e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 448.16, 'epoch': 3.86}
96%|█████████▋| 42350/43939 [68:20:17<2:30:38, 5.69s/it]
96%|█████████▋| 42351/43939 [68:20:23<2:30:56, 5.70s/it]
96%|█████████▋| 42352/43939 [68:20:29<2:30:58, 5.71s/it]
96%|█████████▋| 42353/43939 [68:20:34<2:30:42, 5.70s/it]
96%|█████████▋| 42354/43939 [68:20:40<2:30:35, 5.70s/it]
96%|█████████▋| 42355/43939 [68:20:46<2:30:53, 5.72s/it]
96%|█████████▋| 42356/43939 [68:20:52<2:30:44, 5.71s/it]
96%|█████████▋| 42357/43939 [68:20:57<2:30:21, 5.70s/it]
96%|█████████▋| 42358/43939 [68:21:03<2:30:39, 5.72s/it]
96%|█████████▋| 42359/43939 [68:21:09<2:32:24, 5.79s/it]
96%|█████████▋| 42360/43939 [68:21:15<2:31:30, 5.76s/it]
96%|█████████▋| 42361/43939 [68:21:20<2:31:06, 5.75s/it]
96%|█████████▋| 42362/43939 [68:21:26<2:30:35, 5.73s/it]
96%|█████████▋| 42363/43939 [68:21:32<2:30:20, 5.72s/it]
96%|█████████▋| 42364/43939 [68:21:38<2:30:26, 5.73s/it]
96%|█████████▋| 42365/43939 [68:21:43<2:30:08, 5.72s/it]
96%|█████████▋| 42366/43939 [68:21:49<2:29:59, 5.72s/it]
96%|█████████▋| 42367/43939 [68:21:55<2:29:26, 5.70s/it]
96%|█████████▋| 42368/43939 [68:22:00<2:29:20, 5.70s/it]
96%|█████████▋| 42369/43939 [68:22:06<2:29:15, 5.70s/it]
96%|█████████▋| 42370/43939 [68:22:12<2:29:00, 5.70s/it]
96%|█████████▋| 42371/43939 [68:22:17<2:28:41, 5.69s/it]
96%|█████████▋| 42372/43939 [68:22:23<2:28:44, 5.70s/it]
96%|█████████▋| 42373/43939 [68:22:29<2:28:43, 5.70s/it]
96%|█████████▋| 42374/43939 [68:22:34<2:28:28, 5.69s/it]
96%|█████████▋| 42375/43939 [68:22:40<2:28:27, 5.70s/it]
96%|█████████▋| 42376/43939 [68:22:46<2:28:31, 5.70s/it]
96%|█████████▋| 42377/43939 [68:22:52<2:28:27, 5.70s/it]
96%|█████████▋| 42378/43939 [68:22:57<2:28:13, 5.70s/it]
96%|█████████▋| 42379/43939 [68:23:03<2:28:02, 5.69s/it]
96%|█████████▋| 42380/43939 [68:23:09<2:27:53, 5.69s/it]
96%|█████████▋| 42381/43939 [68:23:15<2:29:38, 5.76s/it]
96%|█████████▋| 42382/43939 [68:23:20<2:29:06, 5.75s/it]
96%|█████████▋| 42383/43939 [68:23:26<2:28:38, 5.73s/it]
96%|█████████▋| 42384/43939 [68:23:32<2:28:39, 5.74s/it]
96%|█████████▋| 42385/43939 [68:23:37<2:28:06, 5.72s/it]
96%|█████████▋| 42386/43939 [68:23:43<2:27:56, 5.72s/it]
96%|█████████▋| 42387/43939 [68:23:49<2:27:42, 5.71s/it]
96%|█████████▋| 42388/43939 [68:23:55<2:27:20, 5.70s/it]
96%|█████████▋| 42389/43939 [68:24:00<2:27:17, 5.70s/it]
96%|█████████▋| 42390/43939 [68:24:06<2:27:12, 5.70s/it]
96%|█████████▋| 42391/43939 [68:24:12<2:27:03, 5.70s/it]
96%|█████████▋| 42392/43939 [68:24:17<2:27:19, 5.71s/it]
96%|█████████▋| 42393/43939 [68:24:23<2:26:52, 5.70s/it]
96%|█████████▋| 42394/43939 [68:24:29<2:26:49, 5.70s/it]
96%|█████████▋| 42395/43939 [68:24:34<2:26:39, 5.70s/it]
96%|█████████▋| 42396/43939 [68:24:40<2:26:21, 5.69s/it]
96%|█████████▋| 42397/43939 [68:24:46<2:26:23, 5.70s/it]
96%|█████████▋| 42398/43939 [68:24:51<2:26:13, 5.69s/it]
96%|█████████▋| 42399/43939 [68:24:57<2:25:51, 5.68s/it]
96%|█████████▋| 42400/43939 [68:25:03<2:25:52, 5.69s/it]
{'loss': 0.6511, 'grad_norm': 1.4168503284454346, 'learning_rate': 1.8685483199681607e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.8, 'epoch': 3.86}
96%|█████████▋| 42400/43939 [68:25:03<2:25:52, 5.69s/it]
96%|█████████▋| 42401/43939 [68:25:09<2:26:03, 5.70s/it]
97%|█████████▋| 42402/43939 [68:25:14<2:26:16, 5.71s/it]
97%|█████████▋| 42403/43939 [68:25:20<2:27:54, 5.78s/it]
97%|█████████▋| 42404/43939 [68:25:26<2:27:31, 5.77s/it]
97%|█████████▋| 42405/43939 [68:25:32<2:26:59, 5.75s/it]
97%|█████████▋| 42406/43939 [68:25:37<2:26:26, 5.73s/it]
97%|█████████▋| 42407/43939 [68:25:43<2:26:16, 5.73s/it]
97%|█████████▋| 42408/43939 [68:25:49<2:26:17, 5.73s/it]
97%|█████████▋| 42409/43939 [68:25:55<2:25:49, 5.72s/it]
97%|█████████▋| 42410/43939 [68:26:00<2:25:22, 5.70s/it]
97%|█████████▋| 42411/43939 [68:26:06<2:25:35, 5.72s/it]
97%|█████████▋| 42412/43939 [68:26:12<2:25:28, 5.72s/it]
97%|█████████▋| 42413/43939 [68:26:17<2:25:23, 5.72s/it]
97%|█████████▋| 42414/43939 [68:26:23<2:25:14, 5.71s/it]
97%|█████████▋| 42415/43939 [68:26:29<2:25:06, 5.71s/it]
97%|█████████▋| 42416/43939 [68:26:34<2:24:54, 5.71s/it]
97%|█████████▋| 42417/43939 [68:26:40<2:24:34, 5.70s/it]
97%|█████████▋| 42418/43939 [68:26:46<2:24:26, 5.70s/it]
97%|█████████▋| 42419/43939 [68:26:52<2:24:25, 5.70s/it]
97%|█████████▋| 42420/43939 [68:26:57<2:24:19, 5.70s/it]
97%|█████████▋| 42421/43939 [68:27:03<2:24:45, 5.72s/it]
97%|█████████▋| 42422/43939 [68:27:09<2:24:50, 5.73s/it]
97%|█████████▋| 42423/43939 [68:27:14<2:24:34, 5.72s/it]
97%|█████████▋| 42424/43939 [68:27:20<2:26:10, 5.79s/it]
97%|█████████▋| 42425/43939 [68:27:26<2:25:46, 5.78s/it]
97%|█████████▋| 42426/43939 [68:27:32<2:25:33, 5.77s/it]
97%|█████████▋| 42427/43939 [68:27:38<2:24:53, 5.75s/it]
97%|█████████▋| 42428/43939 [68:27:43<2:24:18, 5.73s/it]
97%|█████████▋| 42429/43939 [68:27:49<2:23:46, 5.71s/it]
97%|█████████▋| 42430/43939 [68:27:55<2:23:29, 5.71s/it]
97%|█████████▋| 42431/43939 [68:28:00<2:23:24, 5.71s/it]
97%|█████████▋| 42432/43939 [68:28:06<2:23:13, 5.70s/it]
97%|█████████▋| 42433/43939 [68:28:12<2:23:19, 5.71s/it]
97%|█████████▋| 42434/43939 [68:28:18<2:23:15, 5.71s/it]
97%|█████████▋| 42435/43939 [68:28:23<2:23:07, 5.71s/it]
97%|█████████▋| 42436/43939 [68:28:29<2:23:00, 5.71s/it]
97%|█████████▋| 42437/43939 [68:28:35<2:23:04, 5.72s/it]
97%|█████████▋| 42438/43939 [68:28:40<2:22:37, 5.70s/it]
97%|█████████▋| 42439/43939 [68:28:46<2:22:49, 5.71s/it]
97%|█████████▋| 42440/43939 [68:28:52<2:22:40, 5.71s/it]
97%|█████████▋| 42441/43939 [68:28:57<2:22:30, 5.71s/it]
97%|█████████▋| 42442/43939 [68:29:03<2:22:42, 5.72s/it]
97%|█████████▋| 42443/43939 [68:29:09<2:22:37, 5.72s/it]
97%|█████████▋| 42444/43939 [68:29:15<2:22:22, 5.71s/it]
97%|█████████▋| 42445/43939 [68:29:20<2:21:47, 5.69s/it]
97%|█████████▋| 42446/43939 [68:29:26<2:23:24, 5.76s/it]
97%|█████████▋| 42447/43939 [68:29:32<2:22:40, 5.74s/it]
97%|█████████▋| 42448/43939 [68:29:38<2:22:19, 5.73s/it]
97%|█████████▋| 42449/43939 [68:29:43<2:22:03, 5.72s/it]
97%|█████████▋| 42450/43939 [68:29:49<2:21:45, 5.71s/it]
{'loss': 0.6677, 'grad_norm': 1.635582447052002, 'learning_rate': 1.7493231329423764e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 496.41, 'epoch': 3.86}
97%|█████████▋| 42450/43939 [68:29:49<2:21:45, 5.71s/it]
97%|█████████▋| 42451/43939 [68:29:55<2:21:39, 5.71s/it]
97%|█████████▋| 42452/43939 [68:30:00<2:21:26, 5.71s/it]
97%|█████████▋| 42453/43939 [68:30:06<2:21:08, 5.70s/it]
97%|█████████▋| 42454/43939 [68:30:12<2:21:07, 5.70s/it]
97%|█████████▋| 42455/43939 [68:30:17<2:20:52, 5.70s/it]
97%|█████████▋| 42456/43939 [68:30:23<2:20:50, 5.70s/it]
97%|█████████▋| 42457/43939 [68:30:29<2:20:41, 5.70s/it]
97%|█████████▋| 42458/43939 [68:30:35<2:20:47, 5.70s/it]
97%|█████████▋| 42459/43939 [68:30:40<2:20:53, 5.71s/it]
97%|█████████▋| 42460/43939 [68:30:46<2:20:34, 5.70s/it]
97%|█████████▋| 42461/43939 [68:30:52<2:20:15, 5.69s/it]
97%|█████████▋| 42462/43939 [68:30:57<2:20:12, 5.70s/it]
97%|█████████▋| 42463/43939 [68:31:03<2:20:15, 5.70s/it]
97%|█████████▋| 42464/43939 [68:31:09<2:19:50, 5.69s/it]
97%|█████████▋| 42465/43939 [68:31:14<2:19:40, 5.69s/it]
97%|█████████▋| 42466/43939 [68:31:20<2:19:24, 5.68s/it]
97%|█████████▋| 42467/43939 [68:31:26<2:21:13, 5.76s/it]
97%|█████████▋| 42468/43939 [68:31:32<2:20:38, 5.74s/it]
97%|█████████▋| 42469/43939 [68:31:37<2:19:51, 5.71s/it]
97%|█████████▋| 42470/43939 [68:31:43<2:19:51, 5.71s/it]
97%|█████████▋| 42471/43939 [68:31:49<2:19:52, 5.72s/it]
97%|█████████▋| 42472/43939 [68:31:54<2:19:22, 5.70s/it]
97%|█████████▋| 42473/43939 [68:32:00<2:19:18, 5.70s/it]
97%|█████████▋| 42474/43939 [68:32:06<2:19:09, 5.70s/it]
97%|█████████▋| 42475/43939 [68:32:12<2:19:00, 5.70s/it]
97%|█████████▋| 42476/43939 [68:32:17<2:19:00, 5.70s/it]
97%|█████████▋| 42477/43939 [68:32:23<2:18:38, 5.69s/it]
97%|█████████▋| 42478/43939 [68:32:29<2:18:49, 5.70s/it]
97%|█████████▋| 42479/43939 [68:32:34<2:18:40, 5.70s/it]
97%|█████████▋| 42480/43939 [68:32:40<2:18:16, 5.69s/it]
97%|█████████▋| 42481/43939 [68:32:46<2:18:22, 5.69s/it]
97%|█████████▋| 42482/43939 [68:32:51<2:18:20, 5.70s/it]
97%|█████████▋| 42483/43939 [68:32:57<2:18:12, 5.70s/it]
97%|█████████▋| 42484/43939 [68:33:03<2:18:06, 5.70s/it]
97%|█████████▋| 42485/43939 [68:33:09<2:17:58, 5.69s/it]
97%|█████████▋| 42486/43939 [68:33:14<2:17:43, 5.69s/it]
97%|█████████▋| 42487/43939 [68:33:20<2:17:52, 5.70s/it]
97%|█████████▋| 42488/43939 [68:33:26<2:17:55, 5.70s/it]
97%|█████████▋| 42489/43939 [68:33:32<2:19:32, 5.77s/it]
97%|█████████▋| 42490/43939 [68:33:37<2:18:38, 5.74s/it]
97%|█████████▋| 42491/43939 [68:33:43<2:18:22, 5.73s/it]
97%|█████████▋| 42492/43939 [68:33:49<2:18:14, 5.73s/it]
97%|█████████▋| 42493/43939 [68:33:54<2:17:51, 5.72s/it]
97%|█████████▋| 42494/43939 [68:34:00<2:17:40, 5.72s/it]
97%|█████████▋| 42495/43939 [68:34:06<2:17:22, 5.71s/it]
97%|█████████▋| 42496/43939 [68:34:11<2:17:15, 5.71s/it]
97%|█████████▋| 42497/43939 [68:34:17<2:17:14, 5.71s/it]
97%|█████████▋| 42498/43939 [68:34:23<2:17:29, 5.72s/it]
97%|█████████▋| 42499/43939 [68:34:29<2:16:49, 5.70s/it]
97%|█████████▋| 42500/43939 [68:34:34<2:16:39, 5.70s/it]
{'loss': 0.6595, 'grad_norm': 1.5602213144302368, 'learning_rate': 1.6340146837252324e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 482.26, 'epoch': 3.87}
97%|█████████▋| 42500/43939 [68:34:34<2:16:39, 5.70s/it]
97%|█████████▋| 42501/43939 [68:34:40<2:16:23, 5.69s/it]
97%|█████████▋| 42502/43939 [68:34:46<2:16:14, 5.69s/it]
97%|█████████▋| 42503/43939 [68:34:51<2:16:06, 5.69s/it]
97%|█████████▋| 42504/43939 [68:34:57<2:16:06, 5.69s/it]
97%|█████████▋| 42505/43939 [68:35:03<2:16:01, 5.69s/it]
97%|█████████▋| 42506/43939 [68:35:08<2:16:07, 5.70s/it]
97%|█████████▋| 42507/43939 [68:35:14<2:16:07, 5.70s/it]
97%|█████████▋| 42508/43939 [68:35:20<2:15:50, 5.70s/it]
97%|█████████▋| 42509/43939 [68:35:26<2:15:47, 5.70s/it]
97%|█████████▋| 42510/43939 [68:35:31<2:17:32, 5.78s/it]
97%|█████████▋| 42511/43939 [68:35:37<2:16:57, 5.75s/it]
97%|█████████▋| 42512/43939 [68:35:43<2:16:39, 5.75s/it]
97%|█████████▋| 42513/43939 [68:35:49<2:16:02, 5.72s/it]
97%|█████████▋| 42514/43939 [68:35:54<2:15:45, 5.72s/it]
97%|█████████▋| 42515/43939 [68:36:00<2:15:32, 5.71s/it]
97%|█████████▋| 42516/43939 [68:36:06<2:15:32, 5.71s/it]
97%|█████████▋| 42517/43939 [68:36:11<2:15:19, 5.71s/it]
97%|█████████▋| 42518/43939 [68:36:17<2:15:12, 5.71s/it]
97%|█████████▋| 42519/43939 [68:36:23<2:14:53, 5.70s/it]
97%|█████████▋| 42520/43939 [68:36:29<2:14:51, 5.70s/it]
97%|█████████▋| 42521/43939 [68:36:34<2:14:46, 5.70s/it]
97%|█████████▋| 42522/43939 [68:36:40<2:14:34, 5.70s/it]
97%|█████████▋| 42523/43939 [68:36:46<2:14:19, 5.69s/it]
97%|█████████▋| 42524/43939 [68:36:51<2:14:04, 5.69s/it]
97%|█████████▋| 42525/43939 [68:36:57<2:13:53, 5.68s/it]
97%|█████████▋| 42526/43939 [68:37:03<2:13:47, 5.68s/it]
97%|█████████▋| 42527/43939 [68:37:08<2:14:06, 5.70s/it]
97%|█████████▋| 42528/43939 [68:37:14<2:14:28, 5.72s/it]
97%|█████████▋| 42529/43939 [68:37:20<2:14:41, 5.73s/it]
97%|█████████▋| 42530/43939 [68:37:26<2:14:29, 5.73s/it]
97%|█████████▋| 42531/43939 [68:37:32<2:15:55, 5.79s/it]
97%|█████████▋| 42532/43939 [68:37:37<2:15:15, 5.77s/it]
97%|█████████▋| 42533/43939 [68:37:43<2:14:40, 5.75s/it]
97%|█████████▋| 42534/43939 [68:37:49<2:14:14, 5.73s/it]
97%|█████████▋| 42535/43939 [68:37:54<2:13:54, 5.72s/it]
97%|█████████▋| 42536/43939 [68:38:00<2:13:32, 5.71s/it]
97%|█████████▋| 42537/43939 [68:38:06<2:13:30, 5.71s/it]
97%|█████████▋| 42538/43939 [68:38:11<2:13:31, 5.72s/it]
97%|█████████▋| 42539/43939 [68:38:17<2:13:26, 5.72s/it]
97%|█████████▋| 42540/43939 [68:38:23<2:13:16, 5.72s/it]
97%|█████████▋| 42541/43939 [68:38:29<2:13:01, 5.71s/it]
97%|█████████▋| 42542/43939 [68:38:34<2:12:45, 5.70s/it]
97%|█████████▋| 42543/43939 [68:38:40<2:12:24, 5.69s/it]
97%|█████████▋| 42544/43939 [68:38:46<2:12:49, 5.71s/it]
97%|█████████▋| 42545/43939 [68:38:51<2:12:40, 5.71s/it]
97%|█████████▋| 42546/43939 [68:38:57<2:12:30, 5.71s/it]
97%|█████████▋| 42547/43939 [68:39:03<2:12:37, 5.72s/it]
97%|█████████▋| 42548/43939 [68:39:09<2:12:30, 5.72s/it]
97%|█████████▋| 42549/43939 [68:39:14<2:12:19, 5.71s/it]
97%|█████████▋| 42550/43939 [68:39:20<2:12:10, 5.71s/it]
{'loss': 0.6504, 'grad_norm': 1.3467036485671997, 'learning_rate': 1.5226247915784976e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.94, 'epoch': 3.87}
97%|█████████▋| 42550/43939 [68:39:20<2:12:10, 5.71s/it]
97%|█████████▋| 42551/43939 [68:39:26<2:12:27, 5.73s/it]
97%|█████████▋| 42552/43939 [68:39:31<2:12:17, 5.72s/it]
97%|█████████▋| 42553/43939 [68:39:37<2:13:41, 5.79s/it]
97%|█████████▋| 42554/43939 [68:39:43<2:13:15, 5.77s/it]
97%|█████████▋| 42555/43939 [68:39:49<2:12:51, 5.76s/it]
97%|█████████▋| 42556/43939 [68:39:55<2:12:16, 5.74s/it]
97%|█████████▋| 42557/43939 [68:40:00<2:11:54, 5.73s/it]
97%|█████████▋| 42558/43939 [68:40:06<2:11:26, 5.71s/it]
97%|█████████▋| 42559/43939 [68:40:12<2:11:26, 5.71s/it]
97%|█████████▋| 42560/43939 [68:40:17<2:11:09, 5.71s/it]
97%|█████████▋| 42561/43939 [68:40:23<2:11:05, 5.71s/it]
97%|█████████▋| 42562/43939 [68:40:29<2:10:48, 5.70s/it]
97%|█████████▋| 42563/43939 [68:40:34<2:10:39, 5.70s/it]
97%|█████████▋| 42564/43939 [68:40:40<2:10:51, 5.71s/it]
97%|█████████▋| 42565/43939 [68:40:46<2:11:04, 5.72s/it]
97%|█████████▋| 42566/43939 [68:40:52<2:10:39, 5.71s/it]
97%|█████████▋| 42567/43939 [68:40:57<2:10:52, 5.72s/it]
97%|█████████▋| 42568/43939 [68:41:03<2:10:36, 5.72s/it]
97%|█████████▋| 42569/43939 [68:41:09<2:10:34, 5.72s/it]
97%|█████████▋| 42570/43939 [68:41:14<2:10:33, 5.72s/it]
97%|█████████▋| 42571/43939 [68:41:20<2:10:28, 5.72s/it]
97%|█████████▋| 42572/43939 [68:41:26<2:10:16, 5.72s/it]
97%|█████████▋| 42573/43939 [68:41:32<2:10:12, 5.72s/it]
97%|█████████▋| 42574/43939 [68:41:37<2:09:54, 5.71s/it]
97%|█████████▋| 42575/43939 [68:41:43<2:11:28, 5.78s/it]
97%|█████████▋| 42576/43939 [68:41:49<2:11:00, 5.77s/it]
97%|█████████▋| 42577/43939 [68:41:55<2:10:40, 5.76s/it]
97%|█████████▋| 42578/43939 [68:42:00<2:10:12, 5.74s/it]
97%|█████████▋| 42579/43939 [68:42:06<2:09:55, 5.73s/it]
97%|█████████▋| 42580/43939 [68:42:12<2:09:48, 5.73s/it]
97%|█████████▋| 42581/43939 [68:42:18<2:09:26, 5.72s/it]
97%|█████████▋| 42582/43939 [68:42:23<2:09:10, 5.71s/it]
97%|█████████▋| 42583/43939 [68:42:29<2:09:12, 5.72s/it]
97%|█████████▋| 42584/43939 [68:42:35<2:09:38, 5.74s/it]
97%|█████████▋| 42585/43939 [68:42:40<2:09:00, 5.72s/it]
97%|█████████▋| 42586/43939 [68:42:46<2:08:43, 5.71s/it]
97%|█████████▋| 42587/43939 [68:42:52<2:08:59, 5.72s/it]
97%|█████████▋| 42588/43939 [68:42:58<2:08:44, 5.72s/it]
97%|█████████▋| 42589/43939 [68:43:03<2:08:25, 5.71s/it]
97%|█████████▋| 42590/43939 [68:43:09<2:08:06, 5.70s/it]
97%|█████████▋| 42591/43939 [68:43:15<2:07:54, 5.69s/it]
97%|█████████▋| 42592/43939 [68:43:20<2:08:22, 5.72s/it]
97%|█████████▋| 42593/43939 [68:43:26<2:07:52, 5.70s/it]
97%|█████████▋| 42594/43939 [68:43:32<2:07:44, 5.70s/it]
97%|█████████▋| 42595/43939 [68:43:38<2:07:59, 5.71s/it]
97%|█████████▋| 42596/43939 [68:43:44<2:09:32, 5.79s/it]
97%|█████████▋| 42597/43939 [68:43:49<2:09:20, 5.78s/it]
97%|█████████▋| 42598/43939 [68:43:55<2:08:48, 5.76s/it]
97%|█████████▋| 42599/43939 [68:44:01<2:08:25, 5.75s/it]
97%|█████████▋| 42600/43939 [68:44:06<2:08:05, 5.74s/it]
{'loss': 0.6667, 'grad_norm': 1.7891490459442139, 'learning_rate': 1.4151552139394787e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.72, 'epoch': 3.88}
97%|█████████▋| 42600/43939 [68:44:06<2:08:05, 5.74s/it]
97%|█████████▋| 42601/43939 [68:44:12<2:07:39, 5.72s/it]
97%|█████████▋| 42602/43939 [68:44:18<2:07:23, 5.72s/it]
97%|█████████▋| 42603/43939 [68:44:24<2:07:14, 5.71s/it]
97%|█████████▋| 42604/43939 [68:44:29<2:07:05, 5.71s/it]
97%|█████████▋| 42605/43939 [68:44:35<2:06:58, 5.71s/it]
97%|█████████▋| 42606/43939 [68:44:41<2:06:49, 5.71s/it]
97%|█████████▋| 42607/43939 [68:44:46<2:06:56, 5.72s/it]
97%|█████████▋| 42608/43939 [68:44:52<2:07:02, 5.73s/it]
97%|█████████▋| 42609/43939 [68:44:58<2:06:43, 5.72s/it]
97%|█████████▋| 42610/43939 [68:45:04<2:06:39, 5.72s/it]
97%|█████████▋| 42611/43939 [68:45:09<2:06:19, 5.71s/it]
97%|█████████▋| 42612/43939 [68:45:15<2:06:27, 5.72s/it]
97%|█████████▋| 42613/43939 [68:45:21<2:06:11, 5.71s/it]
97%|█████████▋| 42614/43939 [68:45:26<2:06:11, 5.71s/it]
97%|█████████▋| 42615/43939 [68:45:32<2:06:07, 5.72s/it]
97%|█████████▋| 42616/43939 [68:45:38<2:05:58, 5.71s/it]
97%|█████████▋| 42617/43939 [68:45:44<2:05:55, 5.71s/it]
97%|█████████▋| 42618/43939 [68:45:49<2:07:13, 5.78s/it]
97%|█████████▋| 42619/43939 [68:45:55<2:06:46, 5.76s/it]
97%|█████████▋| 42620/43939 [68:46:01<2:06:13, 5.74s/it]
97%|█████████▋| 42621/43939 [68:46:07<2:05:52, 5.73s/it]
97%|█████████▋| 42622/43939 [68:46:12<2:05:41, 5.73s/it]
97%|█████████▋| 42623/43939 [68:46:18<2:05:52, 5.74s/it]
97%|█████████▋| 42624/43939 [68:46:24<2:05:18, 5.72s/it]
97%|█████████▋| 42625/43939 [68:46:29<2:04:59, 5.71s/it]
97%|█████████▋| 42626/43939 [68:46:35<2:05:03, 5.72s/it]
97%|█████████▋| 42627/43939 [68:46:41<2:05:00, 5.72s/it]
97%|█████████▋| 42628/43939 [68:46:47<2:04:40, 5.71s/it]
97%|█████████▋| 42629/43939 [68:46:52<2:04:51, 5.72s/it]
97%|█████████▋| 42630/43939 [68:46:58<2:04:40, 5.71s/it]
97%|█████████▋| 42631/43939 [68:47:04<2:04:33, 5.71s/it]
97%|█████████▋| 42632/43939 [68:47:09<2:04:33, 5.72s/it]
97%|█████████▋| 42633/43939 [68:47:15<2:04:21, 5.71s/it]
97%|█████████▋| 42634/43939 [68:47:21<2:03:58, 5.70s/it]
97%|█████████▋| 42635/43939 [68:47:27<2:04:06, 5.71s/it]
97%|█████████▋| 42636/43939 [68:47:32<2:03:59, 5.71s/it]
97%|█████████▋| 42637/43939 [68:47:38<2:04:05, 5.72s/it]
97%|█████████▋| 42638/43939 [68:47:44<2:04:09, 5.73s/it]
97%|█████████▋| 42639/43939 [68:47:50<2:05:42, 5.80s/it]
97%|█████████▋| 42640/43939 [68:47:55<2:04:48, 5.76s/it]
97%|█████████▋| 42641/43939 [68:48:01<2:04:12, 5.74s/it]
97%|█████████▋| 42642/43939 [68:48:07<2:04:00, 5.74s/it]
97%|█████████▋| 42643/43939 [68:48:13<2:03:42, 5.73s/it]
97%|█████████▋| 42644/43939 [68:48:18<2:03:39, 5.73s/it]
97%|█████████▋| 42645/43939 [68:48:24<2:03:24, 5.72s/it]
97%|█████████▋| 42646/43939 [68:48:30<2:02:49, 5.70s/it]
97%|█████████▋| 42647/43939 [68:48:35<2:02:37, 5.69s/it]
97%|█████████▋| 42648/43939 [68:48:41<2:02:37, 5.70s/it]
97%|█████████▋| 42649/43939 [68:48:47<2:02:41, 5.71s/it]
97%|█████████▋| 42650/43939 [68:48:52<2:02:44, 5.71s/it]
{'loss': 0.6394, 'grad_norm': 1.572636365890503, 'learning_rate': 1.311607646393348e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 473.64, 'epoch': 3.88}
97%|█████████▋| 42650/43939 [68:48:52<2:02:44, 5.71s/it]
97%|█████████▋| 42651/43939 [68:48:58<2:02:43, 5.72s/it]
97%|█████████▋| 42652/43939 [68:49:04<2:02:43, 5.72s/it]
97%|█████████▋| 42653/43939 [68:49:10<2:02:32, 5.72s/it]
97%|█████████▋| 42654/43939 [68:49:15<2:02:20, 5.71s/it]
97%|█████████▋| 42655/43939 [68:49:21<2:02:22, 5.72s/it]
97%|█████████▋| 42656/43939 [68:49:27<2:02:05, 5.71s/it]
97%|█████████▋| 42657/43939 [68:49:32<2:01:54, 5.71s/it]
97%|█████████▋| 42658/43939 [68:49:38<2:01:37, 5.70s/it]
97%|█████████▋| 42659/43939 [68:49:44<2:01:32, 5.70s/it]
97%|█████████▋| 42660/43939 [68:49:49<2:01:19, 5.69s/it]
97%|█████████▋| 42661/43939 [68:49:55<2:02:52, 5.77s/it]
97%|█████████▋| 42662/43939 [68:50:01<2:02:17, 5.75s/it]
97%|█████████▋| 42663/43939 [68:50:07<2:01:54, 5.73s/it]
97%|█████████▋| 42664/43939 [68:50:13<2:01:42, 5.73s/it]
97%|█████████▋| 42665/43939 [68:50:18<2:01:21, 5.72s/it]
97%|█████████▋| 42666/43939 [68:50:24<2:01:11, 5.71s/it]
97%|█████████▋| 42667/43939 [68:50:30<2:00:46, 5.70s/it]
97%|█████████▋| 42668/43939 [68:50:35<2:00:29, 5.69s/it]
97%|█████████▋| 42669/43939 [68:50:41<2:00:23, 5.69s/it]
97%|█████████▋| 42670/43939 [68:50:47<2:00:20, 5.69s/it]
97%|█████████▋| 42671/43939 [68:50:52<2:00:17, 5.69s/it]
97%|█████████▋| 42672/43939 [68:50:58<1:59:55, 5.68s/it]
97%|█████████▋| 42673/43939 [68:51:04<1:59:49, 5.68s/it]
97%|█████████▋| 42674/43939 [68:51:09<1:59:42, 5.68s/it]
97%|█████████▋| 42675/43939 [68:51:15<1:59:37, 5.68s/it]
97%|█████████▋| 42676/43939 [68:51:21<1:59:29, 5.68s/it]
97%|█████████▋| 42677/43939 [68:51:26<1:59:23, 5.68s/it]
97%|█████████▋| 42678/43939 [68:51:32<1:59:28, 5.69s/it]
97%|█████████▋| 42679/43939 [68:51:38<1:59:35, 5.70s/it]
97%|█████████▋| 42680/43939 [68:51:43<1:59:23, 5.69s/it]
97%|█████████▋| 42681/43939 [68:51:49<1:59:25, 5.70s/it]
97%|█████████▋| 42682/43939 [68:51:55<1:59:06, 5.69s/it]
97%|█████████▋| 42683/43939 [68:52:01<2:00:37, 5.76s/it]
97%|█████████▋| 42684/43939 [68:52:06<1:59:57, 5.73s/it]
97%|█████████▋| 42685/43939 [68:52:12<1:59:33, 5.72s/it]
97%|█████████▋| 42686/43939 [68:52:18<1:59:24, 5.72s/it]
97%|█████████▋| 42687/43939 [68:52:24<1:59:09, 5.71s/it]
97%|█████████▋| 42688/43939 [68:52:29<1:58:55, 5.70s/it]
97%|█████████▋| 42689/43939 [68:52:35<1:58:42, 5.70s/it]
97%|█████████▋| 42690/43939 [68:52:41<1:58:33, 5.70s/it]
97%|█████████▋| 42691/43939 [68:52:46<1:58:37, 5.70s/it]
97%|█████████▋| 42692/43939 [68:52:52<1:58:24, 5.70s/it]
97%|█████████▋| 42693/43939 [68:52:58<1:58:36, 5.71s/it]
97%|█████████▋| 42694/43939 [68:53:03<1:58:21, 5.70s/it]
97%|█████████▋| 42695/43939 [68:53:09<1:58:02, 5.69s/it]
97%|█████████▋| 42696/43939 [68:53:15<1:58:03, 5.70s/it]
97%|█████████▋| 42697/43939 [68:53:21<1:58:03, 5.70s/it]
97%|█████████▋| 42698/43939 [68:53:26<1:57:53, 5.70s/it]
97%|█████████▋| 42699/43939 [68:53:32<1:57:47, 5.70s/it]
97%|█████████▋| 42700/43939 [68:53:38<1:57:39, 5.70s/it]
{'loss': 0.6621, 'grad_norm': 1.6758222579956055, 'learning_rate': 1.2119837226463037e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 461.1, 'epoch': 3.89}
97%|█████████▋| 42700/43939 [68:53:38<1:57:39, 5.70s/it]
97%|█████████▋| 42701/43939 [68:53:43<1:57:27, 5.69s/it]
97%|█████████▋| 42702/43939 [68:53:49<1:57:33, 5.70s/it]
97%|█████████▋| 42703/43939 [68:53:55<1:57:28, 5.70s/it]
97%|█████████▋| 42704/43939 [68:54:01<1:58:44, 5.77s/it]
97%|█████████▋| 42705/43939 [68:54:06<1:57:52, 5.73s/it]
97%|█████████▋| 42706/43939 [68:54:12<1:57:29, 5.72s/it]
97%|█████████▋| 42707/43939 [68:54:18<1:57:14, 5.71s/it]
97%|█████████▋| 42708/43939 [68:54:23<1:56:59, 5.70s/it]
97%|█████████▋| 42709/43939 [68:54:29<1:56:45, 5.70s/it]
97%|█████████▋| 42710/43939 [68:54:35<1:56:24, 5.68s/it]
97%|█████████▋| 42711/43939 [68:54:40<1:56:28, 5.69s/it]
97%|█████████▋| 42712/43939 [68:54:46<1:56:25, 5.69s/it]
97%|█████████▋| 42713/43939 [68:54:52<1:56:23, 5.70s/it]
97%|█████████▋| 42714/43939 [68:54:58<1:56:23, 5.70s/it]
97%|█████████▋| 42715/43939 [68:55:03<1:56:33, 5.71s/it]
97%|█████████▋| 42716/43939 [68:55:09<1:56:21, 5.71s/it]
97%|█████████▋| 42717/43939 [68:55:15<1:56:21, 5.71s/it]
97%|█████████▋| 42718/43939 [68:55:20<1:56:20, 5.72s/it]
97%|█████████▋| 42719/43939 [68:55:26<1:56:21, 5.72s/it]
97%|█████████▋| 42720/43939 [68:55:32<1:56:26, 5.73s/it]
97%|█████████▋| 42721/43939 [68:55:38<1:56:17, 5.73s/it]
97%|█████████▋| 42722/43939 [68:55:43<1:56:03, 5.72s/it]
97%|█████████▋| 42723/43939 [68:55:49<1:55:48, 5.71s/it]
97%|█████████▋| 42724/43939 [68:55:55<1:55:38, 5.71s/it]
97%|█████████▋| 42725/43939 [68:56:00<1:55:37, 5.71s/it]
97%|█████████▋| 42726/43939 [68:56:06<1:57:10, 5.80s/it]
97%|█████████▋| 42727/43939 [68:56:12<1:56:32, 5.77s/it]
97%|█████████▋| 42728/43939 [68:56:18<1:55:58, 5.75s/it]
97%|█████████▋| 42729/43939 [68:56:24<1:55:33, 5.73s/it]
97%|█████████▋| 42730/43939 [68:56:29<1:55:11, 5.72s/it]
97%|█████████▋| 42731/43939 [68:56:35<1:54:54, 5.71s/it]
97%|█████████▋| 42732/43939 [68:56:41<1:54:45, 5.71s/it]
97%|█████████▋| 42733/43939 [68:56:46<1:54:34, 5.70s/it]
97%|█████████▋| 42734/43939 [68:56:52<1:54:35, 5.71s/it]
97%|█████████▋| 42735/43939 [68:56:58<1:54:23, 5.70s/it]
97%|█████████▋| 42736/43939 [68:57:03<1:54:27, 5.71s/it]
97%|█████████▋| 42737/43939 [68:57:09<1:54:27, 5.71s/it]
97%|█████████▋| 42738/43939 [68:57:15<1:54:22, 5.71s/it]
97%|█████████▋| 42739/43939 [68:57:21<1:54:13, 5.71s/it]
97%|█████████▋| 42740/43939 [68:57:26<1:54:26, 5.73s/it]
97%|█████████▋| 42741/43939 [68:57:32<1:54:08, 5.72s/it]
97%|█████████▋| 42742/43939 [68:57:38<1:54:08, 5.72s/it]
97%|█████████▋| 42743/43939 [68:57:43<1:54:03, 5.72s/it]
97%|█████████▋| 42744/43939 [68:57:49<1:53:46, 5.71s/it]
97%|█████████▋| 42745/43939 [68:57:55<1:53:33, 5.71s/it]
97%|█████████▋| 42746/43939 [68:58:01<1:53:25, 5.70s/it]
97%|█████████▋| 42747/43939 [68:58:06<1:53:27, 5.71s/it]
97%|█████████▋| 42748/43939 [68:58:12<1:55:03, 5.80s/it]
97%|█████████▋| 42749/43939 [68:58:18<1:54:38, 5.78s/it]
97%|█████████▋| 42750/43939 [68:58:24<1:54:20, 5.77s/it]
{'loss': 0.6546, 'grad_norm': 1.3241678476333618, 'learning_rate': 1.1162850144998405e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.78, 'epoch': 3.89}
97%|█████████▋| 42750/43939 [68:58:24<1:54:20, 5.77s/it]
97%|█████████▋| 42751/43939 [68:58:29<1:53:53, 5.75s/it]
97%|█████████▋| 42752/43939 [68:58:35<1:53:49, 5.75s/it]
97%|█████████▋| 42753/43939 [68:58:41<1:53:18, 5.73s/it]
97%|█████████▋| 42754/43939 [68:58:47<1:52:52, 5.72s/it]
97%|█████████▋| 42755/43939 [68:58:52<1:52:39, 5.71s/it]
97%|█████████▋| 42756/43939 [68:58:58<1:52:22, 5.70s/it]
97%|█████████▋| 42757/43939 [68:59:04<1:52:16, 5.70s/it]
97%|█████████▋| 42758/43939 [68:59:09<1:52:13, 5.70s/it]
97%|█████████▋| 42759/43939 [68:59:15<1:52:18, 5.71s/it]
97%|█████████▋| 42760/43939 [68:59:21<1:52:29, 5.72s/it]
97%|█████████▋| 42761/43939 [68:59:27<1:52:17, 5.72s/it]
97%|█████████▋| 42762/43939 [68:59:32<1:52:17, 5.72s/it]
97%|█████████▋| 42763/43939 [68:59:38<1:52:12, 5.72s/it]
97%|█████████▋| 42764/43939 [68:59:44<1:52:00, 5.72s/it]
97%|█████████▋| 42765/43939 [68:59:49<1:51:57, 5.72s/it]
97%|█████████▋| 42766/43939 [68:59:55<1:51:32, 5.71s/it]
97%|█████████▋| 42767/43939 [69:00:01<1:51:28, 5.71s/it]
97%|█████████▋| 42768/43939 [69:00:07<1:51:12, 5.70s/it]
97%|█████████▋| 42769/43939 [69:00:12<1:50:57, 5.69s/it]
97%|█████████▋| 42770/43939 [69:00:18<1:52:22, 5.77s/it]
97%|█████████▋| 42771/43939 [69:00:24<1:52:06, 5.76s/it]
97%|█████████▋| 42772/43939 [69:00:30<1:51:40, 5.74s/it]
97%|█████████▋| 42773/43939 [69:00:35<1:51:26, 5.73s/it]
97%|█████████▋| 42774/43939 [69:00:41<1:50:56, 5.71s/it]
97%|█████████▋| 42775/43939 [69:00:47<1:50:52, 5.72s/it]
97%|█████████▋| 42776/43939 [69:00:52<1:51:05, 5.73s/it]
97%|█████████▋| 42777/43939 [69:00:58<1:50:49, 5.72s/it]
97%|█████████▋| 42778/43939 [69:01:04<1:50:28, 5.71s/it]
97%|█████████▋| 42779/43939 [69:01:10<1:50:28, 5.71s/it]
97%|█████████▋| 42780/43939 [69:01:15<1:50:21, 5.71s/it]
97%|█████████▋| 42781/43939 [69:01:21<1:50:24, 5.72s/it]
97%|█████████▋| 42782/43939 [69:01:27<1:50:22, 5.72s/it]
97%|█████████▋| 42783/43939 [69:01:32<1:50:06, 5.71s/it]
97%|█████████▋| 42784/43939 [69:01:38<1:49:55, 5.71s/it]
97%|█████████▋| 42785/43939 [69:01:44<1:49:50, 5.71s/it]
97%|█████████▋| 42786/43939 [69:01:50<1:49:39, 5.71s/it]
97%|█████████▋| 42787/43939 [69:01:55<1:49:35, 5.71s/it]
97%|█████████▋| 42788/43939 [69:02:01<1:49:37, 5.71s/it]
97%|█████████▋| 42789/43939 [69:02:07<1:49:27, 5.71s/it]
97%|█████████▋| 42790/43939 [69:02:12<1:49:18, 5.71s/it]
97%|█████████▋| 42791/43939 [69:02:18<1:50:33, 5.78s/it]
97%|█████████▋| 42792/43939 [69:02:24<1:49:52, 5.75s/it]
97%|█████████▋| 42793/43939 [69:02:30<1:49:38, 5.74s/it]
97%|█████████▋| 42794/43939 [69:02:35<1:49:21, 5.73s/it]
97%|█████████▋| 42795/43939 [69:02:41<1:49:15, 5.73s/it]
97%|█████████▋| 42796/43939 [69:02:47<1:49:00, 5.72s/it]
97%|█████████▋| 42797/43939 [69:02:53<1:49:15, 5.74s/it]
97%|█████████▋| 42798/43939 [69:02:58<1:49:00, 5.73s/it]
97%|█████████▋| 42799/43939 [69:03:04<1:48:42, 5.72s/it]
97%|█████████▋| 42800/43939 [69:03:10<1:48:58, 5.74s/it]
{'loss': 0.6472, 'grad_norm': 1.2928262948989868, 'learning_rate': 1.0245130318259633e-07, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 490.58, 'epoch': 3.9}
97%|█████████▋| 42800/43939 [69:03:10<1:48:58, 5.74s/it]
97%|█████████▋| 42801/43939 [69:03:16<1:48:32, 5.72s/it]
97%|█████████▋| 42802/43939 [69:03:21<1:48:20, 5.72s/it]
97%|█████████▋| 42803/43939 [69:03:27<1:48:01, 5.71s/it]
97%|█████████▋| 42804/43939 [69:03:33<1:47:58, 5.71s/it]
97%|█████████▋| 42805/43939 [69:03:38<1:48:07, 5.72s/it]
97%|█████████▋| 42806/43939 [69:03:44<1:48:16, 5.73s/it]
97%|█████████▋| 42807/43939 [69:03:50<1:48:03, 5.73s/it]
97%|█████████▋| 42808/43939 [69:03:56<1:47:35, 5.71s/it]
97%|█████████▋| 42809/43939 [69:04:01<1:47:59, 5.73s/it]
97%|█████████▋| 42810/43939 [69:04:07<1:47:41, 5.72s/it]
97%|█████████▋| 42811/43939 [69:04:13<1:47:24, 5.71s/it]
97%|█████████▋| 42812/43939 [69:04:18<1:47:14, 5.71s/it]
97%|█████████▋| 42813/43939 [69:04:24<1:48:21, 5.77s/it]
97%|█████████▋| 42814/43939 [69:04:30<1:47:55, 5.76s/it]
97%|█████████▋| 42815/43939 [69:04:36<1:47:31, 5.74s/it]
97%|█████████▋| 42816/43939 [69:04:41<1:47:12, 5.73s/it]
97%|█████████▋| 42817/43939 [69:04:47<1:46:49, 5.71s/it]
97%|█████████▋| 42818/43939 [69:04:53<1:46:39, 5.71s/it]
97%|█████████▋| 42819/43939 [69:04:59<1:46:33, 5.71s/it]
97%|█████████▋| 42820/43939 [69:05:04<1:46:32, 5.71s/it]
97%|█████████▋| 42821/43939 [69:05:10<1:46:21, 5.71s/it]
97%|█████████▋| 42822/43939 [69:05:16<1:45:59, 5.69s/it]
97%|█████████▋| 42823/43939 [69:05:21<1:46:08, 5.71s/it]
97%|█████████▋| 42824/43939 [69:05:27<1:45:59, 5.70s/it]
97%|█████████▋| 42825/43939 [69:05:33<1:46:04, 5.71s/it]
97%|█████████▋| 42826/43939 [69:05:38<1:45:46, 5.70s/it]
97%|█████████▋| 42827/43939 [69:05:44<1:45:38, 5.70s/it]
97%|█████████▋| 42828/43939 [69:05:50<1:45:16, 5.69s/it]
97%|█████████▋| 42829/43939 [69:05:55<1:45:12, 5.69s/it]
97%|█████████▋| 42830/43939 [69:06:01<1:45:15, 5.69s/it]
97%|█████████▋| 42831/43939 [69:06:07<1:45:22, 5.71s/it]
97%|█████████▋| 42832/43939 [69:06:13<1:45:12, 5.70s/it]
97%|█████████▋| 42833/43939 [69:06:18<1:45:05, 5.70s/it]
97%|█████████▋| 42834/43939 [69:06:24<1:46:15, 5.77s/it]
97%|█████████▋| 42835/43939 [69:06:30<1:45:34, 5.74s/it]
97%|█████████▋| 42836/43939 [69:06:36<1:45:12, 5.72s/it]
97%|█████████▋| 42837/43939 [69:06:41<1:45:00, 5.72s/it]
97%|█████████▋| 42838/43939 [69:06:47<1:45:09, 5.73s/it]
97%|█████████▋| 42839/43939 [69:06:53<1:45:02, 5.73s/it]
97%|█████████▋| 42840/43939 [69:06:59<1:44:50, 5.72s/it]
98%|█████████▊| 42841/43939 [69:07:04<1:44:33, 5.71s/it]
98%|█████████▊| 42842/43939 [69:07:10<1:44:16, 5.70s/it]
98%|█████████▊| 42843/43939 [69:07:16<1:44:14, 5.71s/it]
98%|█████████▊| 42844/43939 [69:07:21<1:44:15, 5.71s/it]
98%|█████████▊| 42845/43939 [69:07:27<1:44:15, 5.72s/it]
98%|█████████▊| 42846/43939 [69:07:33<1:43:54, 5.70s/it]
98%|█████████▊| 42847/43939 [69:07:38<1:43:47, 5.70s/it]
98%|█████████▊| 42848/43939 [69:07:44<1:43:47, 5.71s/it]
98%|█████████▊| 42849/43939 [69:07:50<1:43:56, 5.72s/it]
98%|█████████▊| 42850/43939 [69:07:56<1:43:45, 5.72s/it]
{'loss': 0.6634, 'grad_norm': 1.594543695449829, 'learning_rate': 9.366692225433737e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 508.98, 'epoch': 3.9}
98%|█████████▊| 42850/43939 [69:07:56<1:43:45, 5.72s/it]
98%|█████████▊| 42851/43939 [69:08:01<1:43:39, 5.72s/it]
98%|█████████▊| 42852/43939 [69:08:07<1:43:41, 5.72s/it]
98%|█████████▊| 42853/43939 [69:08:13<1:43:31, 5.72s/it]
98%|█████████▊| 42854/43939 [69:08:18<1:43:22, 5.72s/it]
98%|█████████▊| 42855/43939 [69:08:24<1:43:12, 5.71s/it]
98%|█████████▊| 42856/43939 [69:08:30<1:44:26, 5.79s/it]
98%|█████████▊| 42857/43939 [69:08:36<1:43:51, 5.76s/it]
98%|█████████▊| 42858/43939 [69:08:42<1:43:26, 5.74s/it]
98%|█████████▊| 42859/43939 [69:08:47<1:42:54, 5.72s/it]
98%|█████████▊| 42860/43939 [69:08:53<1:42:43, 5.71s/it]
98%|█████████▊| 42861/43939 [69:08:59<1:42:44, 5.72s/it]
98%|█████████▊| 42862/43939 [69:09:04<1:42:36, 5.72s/it]
98%|█████████▊| 42863/43939 [69:09:10<1:42:23, 5.71s/it]
98%|█████████▊| 42864/43939 [69:09:16<1:42:16, 5.71s/it]
98%|█████████▊| 42865/43939 [69:09:21<1:42:08, 5.71s/it]
98%|█████████▊| 42866/43939 [69:09:27<1:41:58, 5.70s/it]
98%|█████████▊| 42867/43939 [69:09:33<1:41:54, 5.70s/it]
98%|█████████▊| 42868/43939 [69:09:39<1:41:35, 5.69s/it]
98%|█████████▊| 42869/43939 [69:09:44<1:41:31, 5.69s/it]
98%|█████████▊| 42870/43939 [69:09:50<1:41:20, 5.69s/it]
98%|█████████▊| 42871/43939 [69:09:56<1:41:32, 5.70s/it]
98%|█████████▊| 42872/43939 [69:10:01<1:42:02, 5.74s/it]
98%|█████████▊| 42873/43939 [69:10:07<1:41:45, 5.73s/it]
98%|█████████▊| 42874/43939 [69:10:13<1:41:34, 5.72s/it]
98%|█████████▊| 42875/43939 [69:10:19<1:41:27, 5.72s/it]
98%|█████████▊| 42876/43939 [69:10:24<1:41:26, 5.73s/it]
98%|█████████▊| 42877/43939 [69:10:30<1:42:36, 5.80s/it]
98%|█████████▊| 42878/43939 [69:10:36<1:41:58, 5.77s/it]
98%|█████████▊| 42879/43939 [69:10:42<1:41:36, 5.75s/it]
98%|█████████▊| 42880/43939 [69:10:47<1:41:16, 5.74s/it]
98%|█████████▊| 42881/43939 [69:10:53<1:40:56, 5.72s/it]
98%|█████████▊| 42882/43939 [69:10:59<1:40:41, 5.72s/it]
98%|█████████▊| 42883/43939 [69:11:04<1:40:27, 5.71s/it]
98%|█████████▊| 42884/43939 [69:11:10<1:40:20, 5.71s/it]
98%|█████████▊| 42885/43939 [69:11:16<1:40:22, 5.71s/it]
98%|█████████▊| 42886/43939 [69:11:22<1:40:17, 5.71s/it]
98%|█████████▊| 42887/43939 [69:11:27<1:40:09, 5.71s/it]
98%|█████████▊| 42888/43939 [69:11:33<1:39:59, 5.71s/it]
98%|█████████▊| 42889/43939 [69:11:39<1:39:44, 5.70s/it]
98%|█████████▊| 42890/43939 [69:11:44<1:39:47, 5.71s/it]
98%|█████████▊| 42891/43939 [69:11:50<1:39:52, 5.72s/it]
98%|█████████▊| 42892/43939 [69:11:56<1:39:55, 5.73s/it]
98%|█████████▊| 42893/43939 [69:12:02<1:39:43, 5.72s/it]
98%|█████████▊| 42894/43939 [69:12:07<1:39:38, 5.72s/it]
98%|█████████▊| 42895/43939 [69:12:13<1:39:23, 5.71s/it]
98%|█████████▊| 42896/43939 [69:12:19<1:39:25, 5.72s/it]
98%|█████████▊| 42897/43939 [69:12:24<1:39:17, 5.72s/it]
98%|█████████▊| 42898/43939 [69:12:30<1:39:02, 5.71s/it]
98%|█████████▊| 42899/43939 [69:12:36<1:40:06, 5.78s/it]
98%|█████████▊| 42900/43939 [69:12:42<1:39:36, 5.75s/it]
{'loss': 0.6558, 'grad_norm': 1.7149603366851807, 'learning_rate': 8.527549725945993e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 489.84, 'epoch': 3.91}
98%|█████████▊| 42900/43939 [69:12:42<1:39:36, 5.75s/it]
98%|█████████▊| 42901/43939 [69:12:48<1:39:32, 5.75s/it]
98%|█████████▊| 42902/43939 [69:12:53<1:39:16, 5.74s/it]
98%|█████████▊| 42903/43939 [69:12:59<1:38:44, 5.72s/it]
98%|█████████▊| 42904/43939 [69:13:05<1:38:34, 5.71s/it]
98%|█████████▊| 42905/43939 [69:13:10<1:38:37, 5.72s/it]
98%|█████████▊| 42906/43939 [69:13:16<1:38:26, 5.72s/it]
98%|█████████▊| 42907/43939 [69:13:22<1:38:10, 5.71s/it]
98%|█████████▊| 42908/43939 [69:13:28<1:38:11, 5.71s/it]
98%|█████████▊| 42909/43939 [69:13:33<1:38:06, 5.71s/it]
98%|█████████▊| 42910/43939 [69:13:39<1:37:56, 5.71s/it]
98%|█████████▊| 42911/43939 [69:13:45<1:37:53, 5.71s/it]
98%|█████████▊| 42912/43939 [69:13:50<1:37:40, 5.71s/it]
98%|█████████▊| 42913/43939 [69:13:56<1:37:42, 5.71s/it]
98%|█████████▊| 42914/43939 [69:14:02<1:37:49, 5.73s/it]
98%|█████████▊| 42915/43939 [69:14:08<1:37:55, 5.74s/it]
98%|█████████▊| 42916/43939 [69:14:13<1:37:47, 5.74s/it]
98%|█████████▊| 42917/43939 [69:14:19<1:37:36, 5.73s/it]
98%|█████████▊| 42918/43939 [69:14:25<1:37:22, 5.72s/it]
98%|█████████▊| 42919/43939 [69:14:30<1:37:01, 5.71s/it]
98%|█████████▊| 42920/43939 [69:14:36<1:38:16, 5.79s/it]
98%|█████████▊| 42921/43939 [69:14:42<1:37:50, 5.77s/it]
98%|█████████▊| 42922/43939 [69:14:48<1:37:21, 5.74s/it]
98%|█████████▊| 42923/43939 [69:14:54<1:37:19, 5.75s/it]
98%|█████████▊| 42924/43939 [69:14:59<1:36:58, 5.73s/it]
98%|█████████▊| 42925/43939 [69:15:05<1:36:43, 5.72s/it]
98%|█████████▊| 42926/43939 [69:15:11<1:36:31, 5.72s/it]
98%|█████████▊| 42927/43939 [69:15:16<1:36:18, 5.71s/it]
98%|█████████▊| 42928/43939 [69:15:22<1:36:08, 5.71s/it]
98%|█████████▊| 42929/43939 [69:15:28<1:36:09, 5.71s/it]
98%|█████████▊| 42930/43939 [69:15:33<1:36:03, 5.71s/it]
98%|█████████▊| 42931/43939 [69:15:39<1:35:57, 5.71s/it]
98%|█████████▊| 42932/43939 [69:15:45<1:35:47, 5.71s/it]
98%|█████████▊| 42933/43939 [69:15:51<1:35:40, 5.71s/it]
98%|█████████▊| 42934/43939 [69:15:56<1:35:47, 5.72s/it]
98%|█████████▊| 42935/43939 [69:16:02<1:35:37, 5.71s/it]
98%|█████████▊| 42936/43939 [69:16:08<1:35:31, 5.71s/it]
98%|█████████▊| 42937/43939 [69:16:13<1:35:26, 5.71s/it]
98%|█████████▊| 42938/43939 [69:16:19<1:35:14, 5.71s/it]
98%|█████████▊| 42939/43939 [69:16:25<1:34:58, 5.70s/it]
98%|█████████▊| 42940/43939 [69:16:31<1:34:56, 5.70s/it]
98%|█████████▊| 42941/43939 [69:16:36<1:34:49, 5.70s/it]
98%|█████████▊| 42942/43939 [69:16:42<1:36:05, 5.78s/it]
98%|█████████▊| 42943/43939 [69:16:48<1:35:29, 5.75s/it]
98%|█████████▊| 42944/43939 [69:16:54<1:35:27, 5.76s/it]
98%|█████████▊| 42945/43939 [69:16:59<1:35:07, 5.74s/it]
98%|█████████▊| 42946/43939 [69:17:05<1:34:40, 5.72s/it]
98%|█████████▊| 42947/43939 [69:17:11<1:34:31, 5.72s/it]
98%|█████████▊| 42948/43939 [69:17:16<1:34:24, 5.72s/it]
98%|█████████▊| 42949/43939 [69:17:22<1:34:15, 5.71s/it]
98%|█████████▊| 42950/43939 [69:17:28<1:34:09, 5.71s/it]
{'loss': 0.6671, 'grad_norm': 1.5452629327774048, 'learning_rate': 7.727716059240941e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.62, 'epoch': 3.91}
98%|█████████▊| 42950/43939 [69:17:28<1:34:09, 5.71s/it]
98%|█████████▊| 42951/43939 [69:17:34<1:34:04, 5.71s/it]
98%|█████████▊| 42952/43939 [69:17:39<1:33:50, 5.70s/it]
98%|█████████▊| 42953/43939 [69:17:45<1:33:34, 5.69s/it]
98%|█████████▊| 42954/43939 [69:17:51<1:33:30, 5.70s/it]
98%|█████████▊| 42955/43939 [69:17:56<1:33:20, 5.69s/it]
98%|█████████▊| 42956/43939 [69:18:02<1:33:20, 5.70s/it]
98%|█████████▊| 42957/43939 [69:18:08<1:33:17, 5.70s/it]
98%|█████████▊| 42958/43939 [69:18:13<1:33:20, 5.71s/it]
98%|█████████▊| 42959/43939 [69:18:19<1:33:13, 5.71s/it]
98%|█████████▊| 42960/43939 [69:18:25<1:33:14, 5.71s/it]
98%|█████████▊| 42961/43939 [69:18:31<1:32:55, 5.70s/it]
98%|█████████▊| 42962/43939 [69:18:36<1:32:45, 5.70s/it]
98%|█████████▊| 42963/43939 [69:18:42<1:33:54, 5.77s/it]
98%|█████████▊| 42964/43939 [69:18:48<1:33:30, 5.75s/it]
98%|█████████▊| 42965/43939 [69:18:54<1:33:03, 5.73s/it]
98%|█████████▊| 42966/43939 [69:18:59<1:32:43, 5.72s/it]
98%|█████████▊| 42967/43939 [69:19:05<1:32:30, 5.71s/it]
98%|█████████▊| 42968/43939 [69:19:11<1:32:19, 5.70s/it]
98%|█████████▊| 42969/43939 [69:19:16<1:32:08, 5.70s/it]
98%|█████████▊| 42970/43939 [69:19:22<1:31:58, 5.69s/it]
98%|█████████▊| 42971/43939 [69:19:28<1:31:59, 5.70s/it]
98%|█████████▊| 42972/43939 [69:19:33<1:31:52, 5.70s/it]
98%|█████████▊| 42973/43939 [69:19:39<1:31:47, 5.70s/it]
98%|█████████▊| 42974/43939 [69:19:45<1:31:42, 5.70s/it]
98%|█████████▊| 42975/43939 [69:19:51<1:31:34, 5.70s/it]
98%|█████████▊| 42976/43939 [69:19:56<1:31:34, 5.71s/it]
98%|█████████▊| 42977/43939 [69:20:02<1:31:29, 5.71s/it]
98%|█████████▊| 42978/43939 [69:20:08<1:31:18, 5.70s/it]
98%|█████████▊| 42979/43939 [69:20:13<1:31:08, 5.70s/it]
98%|█████████▊| 42980/43939 [69:20:19<1:31:06, 5.70s/it]
98%|█████████▊| 42981/43939 [69:20:25<1:30:59, 5.70s/it]
98%|█████████▊| 42982/43939 [69:20:30<1:30:58, 5.70s/it]
98%|█████████▊| 42983/43939 [69:20:36<1:30:54, 5.71s/it]
98%|█████████▊| 42984/43939 [69:20:42<1:30:59, 5.72s/it]
98%|█████████▊| 42985/43939 [69:20:48<1:31:56, 5.78s/it]
98%|█████████▊| 42986/43939 [69:20:54<1:31:22, 5.75s/it]
98%|█████████▊| 42987/43939 [69:20:59<1:31:06, 5.74s/it]
98%|█████████▊| 42988/43939 [69:21:05<1:30:42, 5.72s/it]
98%|█████████▊| 42989/43939 [69:21:11<1:30:32, 5.72s/it]
98%|█████████▊| 42990/43939 [69:21:16<1:30:28, 5.72s/it]
98%|█████████▊| 42991/43939 [69:21:22<1:30:16, 5.71s/it]
98%|█████████▊| 42992/43939 [69:21:28<1:29:59, 5.70s/it]
98%|█████████▊| 42993/43939 [69:21:34<1:30:04, 5.71s/it]
98%|█████████▊| 42994/43939 [69:21:39<1:30:00, 5.71s/it]
98%|█████████▊| 42995/43939 [69:21:45<1:29:59, 5.72s/it]
98%|█████████▊| 42996/43939 [69:21:51<1:29:49, 5.72s/it]
98%|█████████▊| 42997/43939 [69:21:56<1:29:52, 5.72s/it]
98%|█████████▊| 42998/43939 [69:22:02<1:29:43, 5.72s/it]
98%|█████████▊| 42999/43939 [69:22:08<1:29:26, 5.71s/it]
98%|█████████▊| 43000/43939 [69:22:13<1:29:08, 5.70s/it]
{'loss': 0.635, 'grad_norm': 1.6672520637512207, 'learning_rate': 6.967203844574776e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 449.67, 'epoch': 3.91}
98%|█████████▊| 43000/43939 [69:22:13<1:29:08, 5.70s/it]
98%|█████████▊| 43001/43939 [69:22:19<1:29:09, 5.70s/it]
98%|█████████▊| 43002/43939 [69:22:25<1:29:05, 5.70s/it]
98%|█████████▊| 43003/43939 [69:22:31<1:29:14, 5.72s/it]
98%|█████████▊| 43004/43939 [69:22:36<1:29:13, 5.73s/it]
98%|█████████▊| 43005/43939 [69:22:42<1:28:57, 5.71s/it]
98%|█████████▊| 43006/43939 [69:22:48<1:29:18, 5.74s/it]
98%|█████████▊| 43007/43939 [69:22:54<1:30:08, 5.80s/it]
98%|█████████▊| 43008/43939 [69:23:00<1:29:38, 5.78s/it]
98%|█████████▊| 43009/43939 [69:23:05<1:29:08, 5.75s/it]
98%|█████████▊| 43010/43939 [69:23:11<1:29:12, 5.76s/it]
98%|█████████▊| 43011/43939 [69:23:17<1:28:59, 5.75s/it]
98%|█████████▊| 43012/43939 [69:23:23<1:28:54, 5.75s/it]
98%|█████████▊| 43013/43939 [69:23:28<1:28:37, 5.74s/it]
98%|█████████▊| 43014/43939 [69:23:34<1:28:36, 5.75s/it]
98%|█████████▊| 43015/43939 [69:23:40<1:28:23, 5.74s/it]
98%|█████████▊| 43016/43939 [69:23:45<1:28:01, 5.72s/it]
98%|█████████▊| 43017/43939 [69:23:51<1:28:03, 5.73s/it]
98%|█████████▊| 43018/43939 [69:23:57<1:27:41, 5.71s/it]
98%|█████████▊| 43019/43939 [69:24:03<1:27:43, 5.72s/it]
98%|█████████▊| 43020/43939 [69:24:08<1:27:39, 5.72s/it]
98%|█████████▊| 43021/43939 [69:24:14<1:27:30, 5.72s/it]
98%|█████████▊| 43022/43939 [69:24:20<1:27:34, 5.73s/it]
98%|█████████▊| 43023/43939 [69:24:25<1:27:18, 5.72s/it]
98%|█████████▊| 43024/43939 [69:24:31<1:26:59, 5.70s/it]
98%|█████████▊| 43025/43939 [69:24:37<1:27:05, 5.72s/it]
98%|█████████▊| 43026/43939 [69:24:43<1:27:19, 5.74s/it]
98%|█████████▊| 43027/43939 [69:24:48<1:26:53, 5.72s/it]
98%|█████████▊| 43028/43939 [69:24:54<1:27:51, 5.79s/it]
98%|█████████▊| 43029/43939 [69:25:00<1:27:21, 5.76s/it]
98%|█████████▊| 43030/43939 [69:25:06<1:26:59, 5.74s/it]
98%|█████████▊| 43031/43939 [69:25:11<1:26:44, 5.73s/it]
98%|█████████▊| 43032/43939 [69:25:17<1:26:35, 5.73s/it]
98%|█████████▊| 43033/43939 [69:25:23<1:26:13, 5.71s/it]
98%|█████████▊| 43034/43939 [69:25:28<1:25:58, 5.70s/it]
98%|█████████▊| 43035/43939 [69:25:34<1:25:54, 5.70s/it]
98%|█████████▊| 43036/43939 [69:25:40<1:25:58, 5.71s/it]
98%|█████████▊| 43037/43939 [69:25:46<1:25:40, 5.70s/it]
98%|█████████▊| 43038/43939 [69:25:51<1:25:35, 5.70s/it]
98%|█████████▊| 43039/43939 [69:25:57<1:25:32, 5.70s/it]
98%|█████████▊| 43040/43939 [69:26:03<1:25:30, 5.71s/it]
98%|█████████▊| 43041/43939 [69:26:08<1:25:28, 5.71s/it]
98%|█████████▊| 43042/43939 [69:26:14<1:25:13, 5.70s/it]
98%|█████████▊| 43043/43939 [69:26:20<1:25:00, 5.69s/it]
98%|█████████▊| 43044/43939 [69:26:25<1:24:49, 5.69s/it]
98%|█████████▊| 43045/43939 [69:26:31<1:24:43, 5.69s/it]
98%|█████████▊| 43046/43939 [69:26:37<1:24:50, 5.70s/it]
98%|█████████▊| 43047/43939 [69:26:43<1:24:38, 5.69s/it]
98%|█████████▊| 43048/43939 [69:26:48<1:24:50, 5.71s/it]
98%|█████████▊| 43049/43939 [69:26:54<1:24:41, 5.71s/it]
98%|█████████▊| 43050/43939 [69:27:00<1:25:50, 5.79s/it]
{'loss': 0.6692, 'grad_norm': 1.5347639322280884, 'learning_rate': 6.246025080814677e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 507.3, 'epoch': 3.92}
98%|█████████▊| 43050/43939 [69:27:00<1:25:50, 5.79s/it]
98%|█████████▊| 43051/43939 [69:27:06<1:25:19, 5.77s/it]
98%|█████████▊| 43052/43939 [69:27:11<1:24:51, 5.74s/it]
98%|█████████▊| 43053/43939 [69:27:17<1:24:39, 5.73s/it]
98%|█████████▊| 43054/43939 [69:27:23<1:24:23, 5.72s/it]
98%|█████████▊| 43055/43939 [69:27:28<1:24:07, 5.71s/it]
98%|█████████▊| 43056/43939 [69:27:34<1:23:55, 5.70s/it]
98%|█████████▊| 43057/43939 [69:27:40<1:23:48, 5.70s/it]
98%|█████████▊| 43058/43939 [69:27:46<1:23:47, 5.71s/it]
98%|█████████▊| 43059/43939 [69:27:51<1:23:50, 5.72s/it]
98%|█████████▊| 43060/43939 [69:27:57<1:23:32, 5.70s/it]
98%|█████████▊| 43061/43939 [69:28:03<1:23:25, 5.70s/it]
98%|█████████▊| 43062/43939 [69:28:08<1:23:19, 5.70s/it]
98%|█████████▊| 43063/43939 [69:28:14<1:23:23, 5.71s/it]
98%|█████████▊| 43064/43939 [69:28:20<1:23:15, 5.71s/it]
98%|█████████▊| 43065/43939 [69:28:26<1:23:15, 5.72s/it]
98%|█████████▊| 43066/43939 [69:28:31<1:23:04, 5.71s/it]
98%|█████████▊| 43067/43939 [69:28:37<1:22:53, 5.70s/it]
98%|█████████▊| 43068/43939 [69:28:43<1:22:56, 5.71s/it]
98%|█████████▊| 43069/43939 [69:28:48<1:22:47, 5.71s/it]
98%|█████████▊| 43070/43939 [69:28:54<1:22:39, 5.71s/it]
98%|█████████▊| 43071/43939 [69:29:00<1:23:45, 5.79s/it]
98%|█████████▊| 43072/43939 [69:29:06<1:23:18, 5.77s/it]
98%|█████████▊| 43073/43939 [69:29:11<1:22:53, 5.74s/it]
98%|█████████▊| 43074/43939 [69:29:17<1:22:31, 5.72s/it]
98%|█████████▊| 43075/43939 [69:29:23<1:22:07, 5.70s/it]
98%|█████████▊| 43076/43939 [69:29:28<1:22:03, 5.71s/it]
98%|█████████▊| 43077/43939 [69:29:34<1:21:51, 5.70s/it]
98%|█████████▊| 43078/43939 [69:29:40<1:21:43, 5.70s/it]
98%|█████████▊| 43079/43939 [69:29:46<1:21:30, 5.69s/it]
98%|█████████▊| 43080/43939 [69:29:51<1:21:26, 5.69s/it]
98%|█████████▊| 43081/43939 [69:29:57<1:21:13, 5.68s/it]
98%|█████████▊| 43082/43939 [69:30:03<1:21:11, 5.68s/it]
98%|█████████▊| 43083/43939 [69:30:08<1:21:10, 5.69s/it]
98%|█████████▊| 43084/43939 [69:30:14<1:21:06, 5.69s/it]
98%|█████████▊| 43085/43939 [69:30:20<1:20:56, 5.69s/it]
98%|█████████▊| 43086/43939 [69:30:25<1:20:55, 5.69s/it]
98%|█████████▊| 43087/43939 [69:30:31<1:20:55, 5.70s/it]
98%|█████████▊| 43088/43939 [69:30:37<1:20:51, 5.70s/it]
98%|█████████▊| 43089/43939 [69:30:42<1:20:41, 5.70s/it]
98%|█████████▊| 43090/43939 [69:30:48<1:20:28, 5.69s/it]
98%|█████████▊| 43091/43939 [69:30:54<1:20:24, 5.69s/it]
98%|█████████▊| 43092/43939 [69:31:00<1:20:38, 5.71s/it]
98%|█████████▊| 43093/43939 [69:31:06<1:21:34, 5.79s/it]
98%|█████████▊| 43094/43939 [69:31:11<1:21:00, 5.75s/it]
98%|█████████▊| 43095/43939 [69:31:17<1:20:38, 5.73s/it]
98%|█████████▊| 43096/43939 [69:31:23<1:20:30, 5.73s/it]
98%|█████████▊| 43097/43939 [69:31:28<1:20:17, 5.72s/it]
98%|█████████▊| 43098/43939 [69:31:34<1:20:09, 5.72s/it]
98%|█████████▊| 43099/43939 [69:31:40<1:20:00, 5.72s/it]
98%|█████████▊| 43100/43939 [69:31:45<1:19:54, 5.71s/it]
{'loss': 0.6508, 'grad_norm': 1.4704046249389648, 'learning_rate': 5.564191146250064e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 458.86, 'epoch': 3.92}
98%|█████████▊| 43100/43939 [69:31:45<1:19:54, 5.71s/it]
98%|█████████▊| 43101/43939 [69:31:51<1:19:43, 5.71s/it]
98%|█████████▊| 43102/43939 [69:31:57<1:19:43, 5.72s/it]
98%|█████████▊| 43103/43939 [69:32:03<1:19:29, 5.70s/it]
98%|█████████▊| 43104/43939 [69:32:08<1:19:16, 5.70s/it]
98%|█████████▊| 43105/43939 [69:32:14<1:19:07, 5.69s/it]
98%|█████████▊| 43106/43939 [69:32:20<1:18:55, 5.68s/it]
98%|█████████▊| 43107/43939 [69:32:25<1:18:54, 5.69s/it]
98%|█████████▊| 43108/43939 [69:32:31<1:19:03, 5.71s/it]
98%|█████████▊| 43109/43939 [69:32:37<1:18:52, 5.70s/it]
98%|█████████▊| 43110/43939 [69:32:42<1:19:00, 5.72s/it]
98%|█████████▊| 43111/43939 [69:32:48<1:18:52, 5.72s/it]
98%|█████████▊| 43112/43939 [69:32:54<1:18:46, 5.72s/it]
98%|█████████▊| 43113/43939 [69:33:00<1:18:32, 5.70s/it]
98%|█████████▊| 43114/43939 [69:33:05<1:19:17, 5.77s/it]
98%|█████████▊| 43115/43939 [69:33:11<1:18:54, 5.75s/it]
98%|█████████▊| 43116/43939 [69:33:17<1:18:39, 5.73s/it]
98%|█████████▊| 43117/43939 [69:33:23<1:18:27, 5.73s/it]
98%|█████████▊| 43118/43939 [69:33:28<1:18:30, 5.74s/it]
98%|█████████▊| 43119/43939 [69:33:34<1:18:23, 5.74s/it]
98%|█████████▊| 43120/43939 [69:33:40<1:18:01, 5.72s/it]
98%|█████████▊| 43121/43939 [69:33:45<1:17:56, 5.72s/it]
98%|█████████▊| 43122/43939 [69:33:51<1:17:49, 5.72s/it]
98%|█████████▊| 43123/43939 [69:33:57<1:17:38, 5.71s/it]
98%|█████████▊| 43124/43939 [69:34:03<1:17:18, 5.69s/it]
98%|█████████▊| 43125/43939 [69:34:08<1:17:17, 5.70s/it]
98%|█████████▊| 43126/43939 [69:34:14<1:17:36, 5.73s/it]
98%|█████████▊| 43127/43939 [69:34:20<1:17:24, 5.72s/it]
98%|█████████▊| 43128/43939 [69:34:26<1:17:34, 5.74s/it]
98%|█████████▊| 43129/43939 [69:34:31<1:17:20, 5.73s/it]
98%|█████████▊| 43130/43939 [69:34:37<1:17:05, 5.72s/it]
98%|█████████▊| 43131/43939 [69:34:43<1:16:53, 5.71s/it]
98%|█████████▊| 43132/43939 [69:34:48<1:16:59, 5.72s/it]
98%|█████████▊| 43133/43939 [69:34:54<1:16:41, 5.71s/it]
98%|█████████▊| 43134/43939 [69:35:00<1:16:34, 5.71s/it]
98%|█████████▊| 43135/43939 [69:35:06<1:17:35, 5.79s/it]
98%|█████████▊| 43136/43939 [69:35:11<1:17:10, 5.77s/it]
98%|█████████▊| 43137/43939 [69:35:17<1:16:52, 5.75s/it]
98%|█████████▊| 43138/43939 [69:35:23<1:16:46, 5.75s/it]
98%|█████████▊| 43139/43939 [69:35:29<1:16:22, 5.73s/it]
98%|█████████▊| 43140/43939 [69:35:34<1:16:09, 5.72s/it]
98%|█████████▊| 43141/43939 [69:35:40<1:15:48, 5.70s/it]
98%|█████████▊| 43142/43939 [69:35:46<1:15:45, 5.70s/it]
98%|█████████▊| 43143/43939 [69:35:51<1:15:29, 5.69s/it]
98%|█████████▊| 43144/43939 [69:35:57<1:15:30, 5.70s/it]
98%|█████████▊| 43145/43939 [69:36:03<1:15:27, 5.70s/it]
98%|█████████▊| 43146/43939 [69:36:08<1:15:26, 5.71s/it]
98%|█████████▊| 43147/43939 [69:36:14<1:15:11, 5.70s/it]
98%|█████████▊| 43148/43939 [69:36:20<1:15:10, 5.70s/it]
98%|█████████▊| 43149/43939 [69:36:26<1:14:57, 5.69s/it]
98%|█████████▊| 43150/43939 [69:36:31<1:14:53, 5.69s/it]
{'loss': 0.6524, 'grad_norm': 1.8729603290557861, 'learning_rate': 4.921712798414135e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.25, 'epoch': 3.93}
98%|█████████▊| 43150/43939 [69:36:31<1:14:53, 5.69s/it]
98%|█████████▊| 43151/43939 [69:36:37<1:15:00, 5.71s/it]
98%|█████████▊| 43152/43939 [69:36:43<1:14:49, 5.70s/it]
98%|█████████▊| 43153/43939 [69:36:48<1:14:39, 5.70s/it]
98%|█████████▊| 43154/43939 [69:36:54<1:14:36, 5.70s/it]
98%|█████████▊| 43155/43939 [69:37:00<1:14:33, 5.71s/it]
98%|█████████▊| 43156/43939 [69:37:06<1:14:34, 5.71s/it]
98%|█████████▊| 43157/43939 [69:37:11<1:15:31, 5.79s/it]
98%|█████████▊| 43158/43939 [69:37:17<1:14:58, 5.76s/it]
98%|█████████▊| 43159/43939 [69:37:23<1:14:46, 5.75s/it]
98%|█████████▊| 43160/43939 [69:37:29<1:14:32, 5.74s/it]
98%|█████████▊| 43161/43939 [69:37:34<1:14:14, 5.73s/it]
98%|█████████▊| 43162/43939 [69:37:40<1:14:20, 5.74s/it]
98%|█████████▊| 43163/43939 [69:37:46<1:14:04, 5.73s/it]
98%|█████████▊| 43164/43939 [69:37:51<1:13:55, 5.72s/it]
98%|█████████▊| 43165/43939 [69:37:57<1:13:49, 5.72s/it]
98%|█████████▊| 43166/43939 [69:38:03<1:13:41, 5.72s/it]
98%|█████████▊| 43167/43939 [69:38:09<1:13:29, 5.71s/it]
98%|█████████▊| 43168/43939 [69:38:14<1:13:20, 5.71s/it]
98%|█████████▊| 43169/43939 [69:38:20<1:13:11, 5.70s/it]
98%|█████████▊| 43170/43939 [69:38:26<1:12:58, 5.69s/it]
98%|█████████▊| 43171/43939 [69:38:31<1:12:56, 5.70s/it]
98%|█████████▊| 43172/43939 [69:38:37<1:12:55, 5.71s/it]
98%|█████████▊| 43173/43939 [69:38:43<1:12:59, 5.72s/it]
98%|█████████▊| 43174/43939 [69:38:49<1:12:58, 5.72s/it]
98%|█████████▊| 43175/43939 [69:38:54<1:12:49, 5.72s/it]
98%|█████████▊| 43176/43939 [69:39:00<1:12:39, 5.71s/it]
98%|█████████▊| 43177/43939 [69:39:06<1:12:19, 5.69s/it]
98%|█████████▊| 43178/43939 [69:39:11<1:12:17, 5.70s/it]
98%|█████████▊| 43179/43939 [69:39:17<1:13:12, 5.78s/it]
98%|█████████▊| 43180/43939 [69:39:23<1:12:50, 5.76s/it]
98%|█████████▊| 43181/43939 [69:39:29<1:12:26, 5.73s/it]
98%|█████████▊| 43182/43939 [69:39:34<1:12:14, 5.73s/it]
98%|█████████▊| 43183/43939 [69:39:40<1:12:06, 5.72s/it]
98%|█████████▊| 43184/43939 [69:39:46<1:11:54, 5.71s/it]
98%|█████████▊| 43185/43939 [69:39:52<1:11:44, 5.71s/it]
98%|█████████▊| 43186/43939 [69:39:57<1:11:42, 5.71s/it]
98%|█████████▊| 43187/43939 [69:40:03<1:11:35, 5.71s/it]
98%|█████████▊| 43188/43939 [69:40:09<1:11:27, 5.71s/it]
98%|█████████▊| 43189/43939 [69:40:14<1:11:16, 5.70s/it]
98%|█████████▊| 43190/43939 [69:40:20<1:11:13, 5.71s/it]
98%|█████████▊| 43191/43939 [69:40:26<1:11:13, 5.71s/it]
98%|█████████▊| 43192/43939 [69:40:32<1:11:11, 5.72s/it]
98%|█████████▊| 43193/43939 [69:40:37<1:11:14, 5.73s/it]
98%|█████████▊| 43194/43939 [69:40:43<1:11:03, 5.72s/it]
98%|█████████▊| 43195/43939 [69:40:49<1:10:58, 5.72s/it]
98%|█████████▊| 43196/43939 [69:40:54<1:10:53, 5.72s/it]
98%|█████████▊| 43197/43939 [69:41:00<1:10:34, 5.71s/it]
98%|█████████▊| 43198/43939 [69:41:06<1:10:22, 5.70s/it]
98%|█████████▊| 43199/43939 [69:41:12<1:10:20, 5.70s/it]
98%|█████████▊| 43200/43939 [69:41:17<1:11:09, 5.78s/it]
{'loss': 0.662, 'grad_norm': 1.5762940645217896, 'learning_rate': 4.3186001739120577e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 483.57, 'epoch': 3.93}
98%|█████████▊| 43200/43939 [69:41:17<1:11:09, 5.78s/it]
98%|█████████▊| 43201/43939 [69:41:23<1:10:49, 5.76s/it]
98%|█████████▊| 43202/43939 [69:41:29<1:10:32, 5.74s/it]
98%|█████████▊| 43203/43939 [69:41:35<1:10:19, 5.73s/it]
98%|█████████▊| 43204/43939 [69:41:40<1:10:15, 5.73s/it]
98%|█████████▊| 43205/43939 [69:41:46<1:10:03, 5.73s/it]
98%|█████████▊| 43206/43939 [69:41:52<1:09:56, 5.72s/it]
98%|█████████▊| 43207/43939 [69:41:57<1:09:46, 5.72s/it]
98%|█████████▊| 43208/43939 [69:42:03<1:09:40, 5.72s/it]
98%|█████████▊| 43209/43939 [69:42:09<1:09:25, 5.71s/it]
98%|█████████▊| 43210/43939 [69:42:15<1:09:22, 5.71s/it]
98%|█████████▊| 43211/43939 [69:42:20<1:09:06, 5.70s/it]
98%|█████████▊| 43212/43939 [69:42:26<1:08:57, 5.69s/it]
98%|█████████▊| 43213/43939 [69:42:32<1:08:52, 5.69s/it]
98%|█████████▊| 43214/43939 [69:42:37<1:08:37, 5.68s/it]
98%|█████████▊| 43215/43939 [69:42:43<1:08:43, 5.70s/it]
98%|█████████▊| 43216/43939 [69:42:49<1:08:36, 5.69s/it]
98%|█████████▊| 43217/43939 [69:42:54<1:08:28, 5.69s/it]
98%|█████████▊| 43218/43939 [69:43:00<1:08:20, 5.69s/it]
98%|█████████▊| 43219/43939 [69:43:06<1:08:18, 5.69s/it]
98%|█████████▊| 43220/43939 [69:43:11<1:08:13, 5.69s/it]
98%|█████████▊| 43221/43939 [69:43:17<1:08:13, 5.70s/it]
98%|█████████▊| 43222/43939 [69:43:23<1:09:02, 5.78s/it]
98%|█████████▊| 43223/43939 [69:43:29<1:08:48, 5.77s/it]
98%|█████████▊| 43224/43939 [69:43:35<1:08:28, 5.75s/it]
98%|█████████▊| 43225/43939 [69:43:40<1:08:09, 5.73s/it]
98%|█████████▊| 43226/43939 [69:43:46<1:07:58, 5.72s/it]
98%|█████████▊| 43227/43939 [69:43:52<1:07:54, 5.72s/it]
98%|█████████▊| 43228/43939 [69:43:57<1:07:47, 5.72s/it]
98%|█████████▊| 43229/43939 [69:44:03<1:07:43, 5.72s/it]
98%|█████████▊| 43230/43939 [69:44:09<1:07:41, 5.73s/it]
98%|█████████▊| 43231/43939 [69:44:15<1:07:24, 5.71s/it]
98%|█████████▊| 43232/43939 [69:44:20<1:07:15, 5.71s/it]
98%|█████████▊| 43233/43939 [69:44:26<1:07:06, 5.70s/it]
98%|█████████▊| 43234/43939 [69:44:32<1:06:53, 5.69s/it]
98%|█████████▊| 43235/43939 [69:44:37<1:06:47, 5.69s/it]
98%|█████████▊| 43236/43939 [69:44:43<1:06:43, 5.69s/it]
98%|█████████▊| 43237/43939 [69:44:49<1:06:45, 5.71s/it]
98%|█████████▊| 43238/43939 [69:44:54<1:06:37, 5.70s/it]
98%|█████████▊| 43239/43939 [69:45:00<1:06:33, 5.70s/it]
98%|█████████▊| 43240/43939 [69:45:06<1:06:39, 5.72s/it]
98%|█████████▊| 43241/43939 [69:45:12<1:06:29, 5.72s/it]
98%|█████████▊| 43242/43939 [69:45:17<1:06:20, 5.71s/it]
98%|█████████▊| 43243/43939 [69:45:23<1:07:17, 5.80s/it]
98%|█████████▊| 43244/43939 [69:45:29<1:06:44, 5.76s/it]
98%|█████████▊| 43245/43939 [69:45:35<1:06:20, 5.74s/it]
98%|█████████▊| 43246/43939 [69:45:40<1:06:06, 5.72s/it]
98%|█████████▊| 43247/43939 [69:45:46<1:05:49, 5.71s/it]
98%|█████████▊| 43248/43939 [69:45:52<1:05:53, 5.72s/it]
98%|█████████▊| 43249/43939 [69:45:57<1:05:42, 5.71s/it]
98%|█████████▊| 43250/43939 [69:46:03<1:05:35, 5.71s/it]
{'loss': 0.6711, 'grad_norm': 1.548627495765686, 'learning_rate': 3.754862788263036e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 485.09, 'epoch': 3.94}
98%|█████████▊| 43250/43939 [69:46:03<1:05:35, 5.71s/it]
98%|█████████▊| 43251/43939 [69:46:09<1:05:20, 5.70s/it]
98%|█████████▊| 43252/43939 [69:46:15<1:05:14, 5.70s/it]
98%|█████████▊| 43253/43939 [69:46:20<1:05:12, 5.70s/it]
98%|█████████▊| 43254/43939 [69:46:26<1:05:18, 5.72s/it]
98%|█████████▊| 43255/43939 [69:46:32<1:05:20, 5.73s/it]
98%|█████████▊| 43256/43939 [69:46:38<1:05:16, 5.73s/it]
98%|█████████▊| 43257/43939 [69:46:43<1:05:06, 5.73s/it]
98%|█████████▊| 43258/43939 [69:46:49<1:05:00, 5.73s/it]
98%|█████████▊| 43259/43939 [69:46:55<1:04:54, 5.73s/it]
98%|█████████▊| 43260/43939 [69:47:00<1:04:46, 5.72s/it]
98%|█████████▊| 43261/43939 [69:47:06<1:04:39, 5.72s/it]
98%|█████████▊| 43262/43939 [69:47:12<1:04:31, 5.72s/it]
98%|█████████▊| 43263/43939 [69:47:18<1:04:32, 5.73s/it]
98%|█████████▊| 43264/43939 [69:47:23<1:04:21, 5.72s/it]
98%|█████████▊| 43265/43939 [69:47:29<1:04:47, 5.77s/it]
98%|█████████▊| 43266/43939 [69:47:35<1:04:29, 5.75s/it]
98%|█████████▊| 43267/43939 [69:47:41<1:04:13, 5.74s/it]
98%|█████████▊| 43268/43939 [69:47:46<1:04:02, 5.73s/it]
98%|█████████▊| 43269/43939 [69:47:52<1:03:52, 5.72s/it]
98%|█████████▊| 43270/43939 [69:47:58<1:03:39, 5.71s/it]
98%|█████████▊| 43271/43939 [69:48:03<1:03:32, 5.71s/it]
98%|█████████▊| 43272/43939 [69:48:09<1:03:23, 5.70s/it]
98%|█████████▊| 43273/43939 [69:48:15<1:03:15, 5.70s/it]
98%|█████████▊| 43274/43939 [69:48:20<1:03:12, 5.70s/it]
98%|█████████▊| 43275/43939 [69:48:26<1:03:07, 5.70s/it]
98%|█████████▊| 43276/43939 [69:48:32<1:03:07, 5.71s/it]
98%|█████████▊| 43277/43939 [69:48:38<1:03:04, 5.72s/it]
98%|█████████▊| 43278/43939 [69:48:43<1:02:51, 5.71s/it]
98%|█████████▊| 43279/43939 [69:48:49<1:02:38, 5.69s/it]
99%|█████████▊| 43280/43939 [69:48:55<1:02:38, 5.70s/it]
99%|█████████▊| 43281/43939 [69:49:00<1:02:34, 5.71s/it]
99%|█████████▊| 43282/43939 [69:49:06<1:02:37, 5.72s/it]
99%|█████████▊| 43283/43939 [69:49:12<1:02:39, 5.73s/it]
99%|█████████▊| 43284/43939 [69:49:18<1:02:23, 5.72s/it]
99%|█████████▊| 43285/43939 [69:49:23<1:02:08, 5.70s/it]
99%|█████████▊| 43286/43939 [69:49:29<1:02:49, 5.77s/it]
99%|█████████▊| 43287/43939 [69:49:35<1:02:41, 5.77s/it]
99%|█████████▊| 43288/43939 [69:49:41<1:02:14, 5.74s/it]
99%|█████████▊| 43289/43939 [69:49:46<1:02:04, 5.73s/it]
99%|█████████▊| 43290/43939 [69:49:52<1:02:02, 5.74s/it]
99%|█████████▊| 43291/43939 [69:49:58<1:01:50, 5.73s/it]
99%|█████████▊| 43292/43939 [69:50:03<1:01:40, 5.72s/it]
99%|█████████▊| 43293/43939 [69:50:09<1:01:29, 5.71s/it]
99%|█████████▊| 43294/43939 [69:50:15<1:01:21, 5.71s/it]
99%|█████████▊| 43295/43939 [69:50:21<1:01:08, 5.70s/it]
99%|█████████▊| 43296/43939 [69:50:26<1:00:57, 5.69s/it]
99%|█████████▊| 43297/43939 [69:50:32<1:00:59, 5.70s/it]
99%|█████████▊| 43298/43939 [69:50:38<1:00:55, 5.70s/it]
99%|█████████▊| 43299/43939 [69:50:43<1:00:55, 5.71s/it]
99%|█████████▊| 43300/43939 [69:50:49<1:00:49, 5.71s/it]
{'loss': 0.6532, 'grad_norm': 1.9532479047775269, 'learning_rate': 3.230509535749049e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 486.72, 'epoch': 3.94}
99%|█████████▊| 43300/43939 [69:50:49<1:00:49, 5.71s/it]
99%|█████████▊| 43301/43939 [69:50:55<1:00:51, 5.72s/it]
99%|█████████▊| 43302/43939 [69:51:01<1:00:36, 5.71s/it]
99%|█████████▊| 43303/43939 [69:51:06<1:00:32, 5.71s/it]
99%|█████████▊| 43304/43939 [69:51:12<1:00:18, 5.70s/it]
99%|█████████▊| 43305/43939 [69:51:18<1:00:15, 5.70s/it]
99%|█████████▊| 43306/43939 [69:51:23<1:00:06, 5.70s/it]
99%|█████████▊| 43307/43939 [69:51:29<1:00:06, 5.71s/it]
99%|█████████▊| 43308/43939 [69:51:35<1:00:42, 5.77s/it]
99%|█████████▊| 43309/43939 [69:51:41<1:00:25, 5.75s/it]
99%|█████████▊| 43310/43939 [69:51:46<1:00:15, 5.75s/it]
99%|█████████▊| 43311/43939 [69:51:52<1:00:00, 5.73s/it]
99%|█████████▊| 43312/43939 [69:51:58<1:00:09, 5.76s/it]
99%|█████████▊| 43313/43939 [69:52:04<59:52, 5.74s/it]
99%|█████████▊| 43314/43939 [69:52:09<59:37, 5.72s/it]
99%|█████████▊| 43315/43939 [69:52:15<59:22, 5.71s/it]
99%|█████████▊| 43316/43939 [69:52:21<59:13, 5.70s/it]
99%|█████████▊| 43317/43939 [69:52:26<59:10, 5.71s/it]
99%|█████████▊| 43318/43939 [69:52:32<59:04, 5.71s/it]
99%|█████████▊| 43319/43939 [69:52:38<59:00, 5.71s/it]
99%|█████████▊| 43320/43939 [69:52:44<58:59, 5.72s/it]
99%|█████████▊| 43321/43939 [69:52:49<58:46, 5.71s/it]
99%|█████████▊| 43322/43939 [69:52:55<58:30, 5.69s/it]
99%|█████████▊| 43323/43939 [69:53:01<58:30, 5.70s/it]
99%|█████████▊| 43324/43939 [69:53:06<58:26, 5.70s/it]
99%|█████████▊| 43325/43939 [69:53:12<58:24, 5.71s/it]
99%|█████████▊| 43326/43939 [69:53:18<58:12, 5.70s/it]
99%|█████████▊| 43327/43939 [69:53:23<58:06, 5.70s/it]
99%|█████████▊| 43328/43939 [69:53:29<57:58, 5.69s/it]
99%|█████████▊| 43329/43939 [69:53:35<58:43, 5.78s/it]
99%|█████████▊| 43330/43939 [69:53:41<58:23, 5.75s/it]
99%|█████████▊| 43331/43939 [69:53:46<57:59, 5.72s/it]
99%|█████████▊| 43332/43939 [69:53:52<57:54, 5.72s/it]
99%|█████████▊| 43333/43939 [69:53:58<57:43, 5.71s/it]
99%|█████████▊| 43334/43939 [69:54:04<57:39, 5.72s/it]
99%|█████████▊| 43335/43939 [69:54:09<57:38, 5.73s/it]
99%|█████████▊| 43336/43939 [69:54:15<57:29, 5.72s/it]
99%|█████████▊| 43337/43939 [69:54:21<57:25, 5.72s/it]
99%|█████████▊| 43338/43939 [69:54:26<57:13, 5.71s/it]
99%|█████████▊| 43339/43939 [69:54:32<57:04, 5.71s/it]
99%|█████████▊| 43340/43939 [69:54:38<56:56, 5.70s/it]
99%|█████████▊| 43341/43939 [69:54:44<56:50, 5.70s/it]
99%|█████████▊| 43342/43939 [69:54:49<56:45, 5.71s/it]
99%|█████████▊| 43343/43939 [69:54:55<56:35, 5.70s/it]
99%|█████████▊| 43344/43939 [69:55:01<56:37, 5.71s/it]
99%|█████████▊| 43345/43939 [69:55:06<56:31, 5.71s/it]
99%|█████████▊| 43346/43939 [69:55:12<56:29, 5.72s/it]
99%|█████████▊| 43347/43939 [69:55:18<56:14, 5.70s/it]
99%|█████████▊| 43348/43939 [69:55:23<56:11, 5.70s/it]
99%|█████████▊| 43349/43939 [69:55:29<56:12, 5.72s/it]
99%|█████████▊| 43350/43939 [69:55:35<56:13, 5.73s/it]
{'loss': 0.6476, 'grad_norm': 1.5838146209716797, 'learning_rate': 2.745548689274957e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 468.18, 'epoch': 3.95}
99%|█████████▊| 43350/43939 [69:55:35<56:13, 5.73s/it]
99%|█████████▊| 43351/43939 [69:55:41<56:38, 5.78s/it]
99%|█████████▊| 43352/43939 [69:55:47<56:17, 5.75s/it]
99%|█████████▊| 43353/43939 [69:55:52<56:05, 5.74s/it]
99%|█████████▊| 43354/43939 [69:55:58<55:55, 5.74s/it]
99%|█████████▊| 43355/43939 [69:56:04<55:40, 5.72s/it]
99%|█████████▊| 43356/43939 [69:56:09<55:29, 5.71s/it]
99%|█████████▊| 43357/43939 [69:56:15<55:25, 5.71s/it]
99%|█████████▊| 43358/43939 [69:56:21<55:28, 5.73s/it]
99%|█████████▊| 43359/43939 [69:56:27<55:19, 5.72s/it]
99%|█████████▊| 43360/43939 [69:56:32<55:10, 5.72s/it]
99%|█████████▊| 43361/43939 [69:56:38<55:01, 5.71s/it]
99%|█████████▊| 43362/43939 [69:56:44<54:55, 5.71s/it]
99%|█████████▊| 43363/43939 [69:56:49<54:44, 5.70s/it]
99%|█████████▊| 43364/43939 [69:56:55<54:40, 5.71s/it]
99%|█████████▊| 43365/43939 [69:57:01<54:36, 5.71s/it]
99%|█████████▊| 43366/43939 [69:57:06<54:27, 5.70s/it]
99%|█████████▊| 43367/43939 [69:57:12<54:23, 5.71s/it]
99%|█████████▊| 43368/43939 [69:57:18<54:11, 5.69s/it]
99%|█████████▊| 43369/43939 [69:57:24<54:05, 5.69s/it]
99%|█████████▊| 43370/43939 [69:57:29<53:58, 5.69s/it]
99%|█████████▊| 43371/43939 [69:57:35<53:59, 5.70s/it]
99%|█████████▊| 43372/43939 [69:57:41<54:37, 5.78s/it]
99%|█████████▊| 43373/43939 [69:57:47<54:15, 5.75s/it]
99%|█████████▊| 43374/43939 [69:57:52<53:56, 5.73s/it]
99%|█████████▊| 43375/43939 [69:57:58<53:44, 5.72s/it]
99%|█████████▊| 43376/43939 [69:58:04<53:32, 5.71s/it]
99%|█████████▊| 43377/43939 [69:58:09<53:26, 5.71s/it]
99%|█████████▊| 43378/43939 [69:58:15<53:19, 5.70s/it]
99%|█████████▊| 43379/43939 [69:58:21<53:15, 5.71s/it]
99%|█████████▊| 43380/43939 [69:58:26<53:04, 5.70s/it]
99%|█████████▊| 43381/43939 [69:58:32<53:01, 5.70s/it]
99%|█████████▊| 43382/43939 [69:58:38<52:53, 5.70s/it]
99%|█████████▊| 43383/43939 [69:58:44<52:44, 5.69s/it]
99%|█████████▊| 43384/43939 [69:58:49<52:57, 5.73s/it]
99%|█████████▊| 43385/43939 [69:58:55<52:48, 5.72s/it]
99%|█████████▊| 43386/43939 [69:59:01<52:34, 5.70s/it]
99%|█████████▊| 43387/43939 [69:59:06<52:27, 5.70s/it]
99%|█████████▊| 43388/43939 [69:59:12<52:22, 5.70s/it]
99%|█████████▊| 43389/43939 [69:59:18<52:25, 5.72s/it]
99%|█████████▉| 43390/43939 [69:59:24<52:23, 5.72s/it]
99%|█████████▉| 43391/43939 [69:59:29<52:19, 5.73s/it]
99%|█████████▉| 43392/43939 [69:59:35<52:19, 5.74s/it]
99%|█████████▉| 43393/43939 [69:59:41<52:17, 5.75s/it]
99%|█████████▉| 43394/43939 [69:59:47<52:46, 5.81s/it]
99%|█████████▉| 43395/43939 [69:59:52<52:17, 5.77s/it]
99%|█████████▉| 43396/43939 [69:59:58<52:01, 5.75s/it]
99%|█████████▉| 43397/43939 [70:00:04<51:48, 5.74s/it]
99%|█████████▉| 43398/43939 [70:00:10<51:36, 5.72s/it]
99%|█████████▉| 43399/43939 [70:00:15<51:37, 5.74s/it]
99%|█████████▉| 43400/43939 [70:00:21<51:23, 5.72s/it]
{'loss': 0.6457, 'grad_norm': 1.6701834201812744, 'learning_rate': 2.2999879002377766e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.61, 'epoch': 3.95}
99%|█████████▉| 43400/43939 [70:00:21<51:23, 5.72s/it]
99%|█████████▉| 43401/43939 [70:00:27<51:21, 5.73s/it]
99%|█████████▉| 43402/43939 [70:00:32<51:08, 5.71s/it]
99%|█████████▉| 43403/43939 [70:00:38<51:12, 5.73s/it]
99%|█████████▉| 43404/43939 [70:00:44<51:11, 5.74s/it]
99%|█████████▉| 43405/43939 [70:00:50<51:03, 5.74s/it]
99%|█████████▉| 43406/43939 [70:00:55<50:51, 5.73s/it]
99%|█████████▉| 43407/43939 [70:01:01<50:41, 5.72s/it]
99%|█████████▉| 43408/43939 [70:01:07<50:34, 5.72s/it]
99%|█████████▉| 43409/43939 [70:01:13<50:32, 5.72s/it]
99%|█████████▉| 43410/43939 [70:01:18<50:25, 5.72s/it]
99%|█████████▉| 43411/43939 [70:01:24<50:13, 5.71s/it]
99%|█████████▉| 43412/43939 [70:01:30<50:02, 5.70s/it]
99%|█████████▉| 43413/43939 [70:01:35<50:04, 5.71s/it]
99%|█████████▉| 43414/43939 [70:01:41<49:59, 5.71s/it]
99%|█████████▉| 43415/43939 [70:01:47<50:29, 5.78s/it]
99%|█████████▉| 43416/43939 [70:01:53<50:11, 5.76s/it]
99%|█████████▉| 43417/43939 [70:01:58<49:57, 5.74s/it]
99%|█████████▉| 43418/43939 [70:02:04<49:39, 5.72s/it]
99%|█████████▉| 43419/43939 [70:02:10<49:27, 5.71s/it]
99%|█████████▉| 43420/43939 [70:02:16<49:22, 5.71s/it]
99%|█████████▉| 43421/43939 [70:02:21<49:15, 5.71s/it]
99%|█████████▉| 43422/43939 [70:02:27<49:11, 5.71s/it]
99%|█████████▉| 43423/43939 [70:02:33<49:03, 5.70s/it]
99%|█████████▉| 43424/43939 [70:02:38<48:54, 5.70s/it]
99%|█████████▉| 43425/43939 [70:02:44<48:46, 5.69s/it]
99%|█████████▉| 43426/43939 [70:02:50<48:40, 5.69s/it]
99%|█████████▉| 43427/43939 [70:02:55<48:30, 5.68s/it]
99%|█████████▉| 43428/43939 [70:03:01<48:25, 5.69s/it]
99%|█████████▉| 43429/43939 [70:03:07<48:17, 5.68s/it]
99%|█████████▉| 43430/43939 [70:03:12<48:15, 5.69s/it]
99%|█████████▉| 43431/43939 [70:03:18<48:16, 5.70s/it]
99%|█████████▉| 43432/43939 [70:03:24<48:10, 5.70s/it]
99%|█████████▉| 43433/43939 [70:03:30<48:08, 5.71s/it]
99%|█████████▉| 43434/43939 [70:03:35<47:57, 5.70s/it]
99%|█████████▉| 43435/43939 [70:03:41<47:49, 5.69s/it]
99%|█████████▉| 43436/43939 [70:03:47<47:49, 5.71s/it]
99%|█████████▉| 43437/43939 [70:03:53<48:19, 5.78s/it]
99%|█████████▉| 43438/43939 [70:03:58<48:04, 5.76s/it]
99%|█████████▉| 43439/43939 [70:04:04<47:48, 5.74s/it]
99%|█████████▉| 43440/43939 [70:04:10<47:47, 5.75s/it]
99%|█████████▉| 43441/43939 [70:04:15<47:31, 5.73s/it]
99%|█████████▉| 43442/43939 [70:04:21<47:25, 5.73s/it]
99%|█████████▉| 43443/43939 [70:04:27<47:20, 5.73s/it]
99%|█████████▉| 43444/43939 [70:04:33<47:10, 5.72s/it]
99%|█████████▉| 43445/43939 [70:04:38<47:01, 5.71s/it]
99%|█████████▉| 43446/43939 [70:04:44<46:49, 5.70s/it]
99%|█████████▉| 43447/43939 [70:04:50<46:52, 5.72s/it]
99%|█████████▉| 43448/43939 [70:04:55<46:41, 5.71s/it]
99%|█████████▉| 43449/43939 [70:05:01<46:41, 5.72s/it]
99%|█████████▉| 43450/43939 [70:05:07<46:30, 5.71s/it]
{'loss': 0.6521, 'grad_norm': 1.8635591268539429, 'learning_rate': 1.893834198405664e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.68, 'epoch': 3.96}
99%|█████████▉| 43450/43939 [70:05:07<46:30, 5.71s/it]
99%|█████████▉| 43451/43939 [70:05:13<46:32, 5.72s/it]
99%|█████████▉| 43452/43939 [70:05:18<46:33, 5.74s/it]
99%|█████████▉| 43453/43939 [70:05:24<46:27, 5.73s/it]
99%|█████████▉| 43454/43939 [70:05:30<46:13, 5.72s/it]
99%|█████████▉| 43455/43939 [70:05:35<46:05, 5.71s/it]
99%|█████████▉| 43456/43939 [70:05:41<45:58, 5.71s/it]
99%|█████████▉| 43457/43939 [70:05:47<45:52, 5.71s/it]
99%|█████████▉| 43458/43939 [70:05:53<46:19, 5.78s/it]
99%|█████████▉| 43459/43939 [70:05:59<46:05, 5.76s/it]
99%|█████████▉| 43460/43939 [70:06:04<45:53, 5.75s/it]
99%|█████████▉| 43461/43939 [70:06:10<45:36, 5.73s/it]
99%|█████████▉| 43462/43939 [70:06:16<45:31, 5.73s/it]
99%|█████████▉| 43463/43939 [70:06:21<45:17, 5.71s/it]
99%|█████████▉| 43464/43939 [70:06:27<45:19, 5.73s/it]
99%|█████████▉| 43465/43939 [70:06:33<45:05, 5.71s/it]
99%|█████████▉| 43466/43939 [70:06:38<45:00, 5.71s/it]
99%|█████████▉| 43467/43939 [70:06:44<44:57, 5.72s/it]
99%|█████████▉| 43468/43939 [70:06:50<44:54, 5.72s/it]
99%|█████████▉| 43469/43939 [70:06:56<44:43, 5.71s/it]
99%|█████████▉| 43470/43939 [70:07:01<44:32, 5.70s/it]
99%|█████████▉| 43471/43939 [70:07:07<44:29, 5.70s/it]
99%|█████████▉| 43472/43939 [70:07:13<44:22, 5.70s/it]
99%|█████████▉| 43473/43939 [70:07:18<44:13, 5.69s/it]
99%|█████████▉| 43474/43939 [70:07:24<44:07, 5.69s/it]
99%|█████████▉| 43475/43939 [70:07:30<43:57, 5.68s/it]
99%|█████████▉| 43476/43939 [70:07:35<43:57, 5.70s/it]
99%|█████████▉| 43477/43939 [70:07:41<44:01, 5.72s/it]
99%|█████████▉| 43478/43939 [70:07:47<43:54, 5.71s/it]
99%|█████████▉| 43479/43939 [70:07:53<43:48, 5.71s/it]
99%|█████████▉| 43480/43939 [70:07:59<44:07, 5.77s/it]
99%|█████████▉| 43481/43939 [70:08:04<43:57, 5.76s/it]
99%|█████████▉| 43482/43939 [70:08:10<43:43, 5.74s/it]
99%|█████████▉| 43483/43939 [70:08:16<43:27, 5.72s/it]
99%|█████████▉| 43484/43939 [70:08:21<43:23, 5.72s/it]
99%|█████████▉| 43485/43939 [70:08:27<43:09, 5.70s/it]
99%|█████████▉| 43486/43939 [70:08:33<43:00, 5.70s/it]
99%|█████████▉| 43487/43939 [70:08:38<42:59, 5.71s/it]
99%|█████████▉| 43488/43939 [70:08:44<42:55, 5.71s/it]
99%|█████████▉| 43489/43939 [70:08:50<42:49, 5.71s/it]
99%|█████████▉| 43490/43939 [70:08:56<42:38, 5.70s/it]
99%|█████████▉| 43491/43939 [70:09:01<42:37, 5.71s/it]
99%|█████████▉| 43492/43939 [70:09:07<42:30, 5.71s/it]
99%|█████████▉| 43493/43939 [70:09:13<42:30, 5.72s/it]
99%|█████████▉| 43494/43939 [70:09:18<42:29, 5.73s/it]
99%|█████████▉| 43495/43939 [70:09:24<42:16, 5.71s/it]
99%|█████████▉| 43496/43939 [70:09:30<42:10, 5.71s/it]
99%|█████████▉| 43497/43939 [70:09:36<42:04, 5.71s/it]
99%|█████████▉| 43498/43939 [70:09:41<42:01, 5.72s/it]
99%|█████████▉| 43499/43939 [70:09:47<41:56, 5.72s/it]
99%|█████████▉| 43500/43939 [70:09:53<41:51, 5.72s/it]
{'loss': 0.6624, 'grad_norm': 1.4792386293411255, 'learning_rate': 1.5270939918082817e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 499.59, 'epoch': 3.96}
99%|█████████▉| 43500/43939 [70:09:53<41:51, 5.72s/it]
99%|█████████▉| 43501/43939 [70:09:58<41:45, 5.72s/it]
99%|█████████▉| 43502/43939 [70:10:04<42:10, 5.79s/it]
99%|█████████▉| 43503/43939 [70:10:10<41:54, 5.77s/it]
99%|█████████▉| 43504/43939 [70:10:16<41:49, 5.77s/it]
99%|█████████▉| 43505/43939 [70:10:22<41:34, 5.75s/it]
99%|█████████▉| 43506/43939 [70:10:27<41:24, 5.74s/it]
99%|█████████▉| 43507/43939 [70:10:33<41:15, 5.73s/it]
99%|█████████▉| 43508/43939 [70:10:39<41:08, 5.73s/it]
99%|█████████▉| 43509/43939 [70:10:44<41:00, 5.72s/it]
99%|█████████▉| 43510/43939 [70:10:50<40:55, 5.72s/it]
99%|█████████▉| 43511/43939 [70:10:56<40:47, 5.72s/it]
99%|█████████▉| 43512/43939 [70:11:02<40:38, 5.71s/it]
99%|█████████▉| 43513/43939 [70:11:07<40:31, 5.71s/it]
99%|█████████▉| 43514/43939 [70:11:13<40:25, 5.71s/it]
99%|█████████▉| 43515/43939 [70:11:19<40:16, 5.70s/it]
99%|█████████▉| 43516/43939 [70:11:24<40:16, 5.71s/it]
99%|█████████▉| 43517/43939 [70:11:30<40:13, 5.72s/it]
99%|█████████▉| 43518/43939 [70:11:36<40:09, 5.72s/it]
99%|█████████▉| 43519/43939 [70:11:42<40:00, 5.72s/it]
99%|█████████▉| 43520/43939 [70:11:47<39:50, 5.70s/it]
99%|█████████▉| 43521/43939 [70:11:53<39:46, 5.71s/it]
99%|█████████▉| 43522/43939 [70:11:59<39:46, 5.72s/it]
99%|█████████▉| 43523/43939 [70:12:05<40:02, 5.77s/it]
99%|█████████▉| 43524/43939 [70:12:10<39:49, 5.76s/it]
99%|█████████▉| 43525/43939 [70:12:16<39:36, 5.74s/it]
99%|█████████▉| 43526/43939 [70:12:22<39:25, 5.73s/it]
99%|█████████▉| 43527/43939 [70:12:27<39:18, 5.72s/it]
99%|█████████▉| 43528/43939 [70:12:33<39:07, 5.71s/it]
99%|█████████▉| 43529/43939 [70:12:39<38:55, 5.70s/it]
99%|█████████▉| 43530/43939 [70:12:45<38:50, 5.70s/it]
99%|█████████▉| 43531/43939 [70:12:50<38:47, 5.71s/it]
99%|█████████▉| 43532/43939 [70:12:56<38:37, 5.69s/it]
99%|█████████▉| 43533/43939 [70:13:02<38:32, 5.70s/it]
99%|█████████▉| 43534/43939 [70:13:07<38:26, 5.70s/it]
99%|█████████▉| 43535/43939 [70:13:13<38:17, 5.69s/it]
99%|█████████▉| 43536/43939 [70:13:19<38:13, 5.69s/it]
99%|█████████▉| 43537/43939 [70:13:24<38:01, 5.68s/it]
99%|█████████▉| 43538/43939 [70:13:30<37:57, 5.68s/it]
99%|█████████▉| 43539/43939 [70:13:36<37:56, 5.69s/it]
99%|█████████▉| 43540/43939 [70:13:41<37:55, 5.70s/it]
99%|█████████▉| 43541/43939 [70:13:47<37:48, 5.70s/it]
99%|█████████▉| 43542/43939 [70:13:53<37:43, 5.70s/it]
99%|█████████▉| 43543/43939 [70:13:59<37:35, 5.70s/it]
99%|█████████▉| 43544/43939 [70:14:04<37:30, 5.70s/it]
99%|█████████▉| 43545/43939 [70:14:10<37:49, 5.76s/it]
99%|█████████▉| 43546/43939 [70:14:16<37:40, 5.75s/it]
99%|█████████▉| 43547/43939 [70:14:22<37:31, 5.74s/it]
99%|█████████▉| 43548/43939 [70:14:27<37:18, 5.72s/it]
99%|█████████▉| 43549/43939 [70:14:33<37:09, 5.72s/it]
99%|█████████▉| 43550/43939 [70:14:39<37:00, 5.71s/it]
{'loss': 0.6517, 'grad_norm': 1.7554465532302856, 'learning_rate': 1.1997730666338247e-08, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 500.97, 'epoch': 3.96}
99%|█████████▉| 43550/43939 [70:14:39<37:00, 5.71s/it]
99%|█████████▉| 43551/43939 [70:14:44<36:51, 5.70s/it]
99%|█████████▉| 43552/43939 [70:14:50<36:47, 5.70s/it]
99%|█████████▉| 43553/43939 [70:14:56<36:42, 5.71s/it]
99%|█████████▉| 43554/43939 [70:15:01<36:35, 5.70s/it]
99%|█████████▉| 43555/43939 [70:15:07<36:30, 5.70s/it]
99%|█████████▉| 43556/43939 [70:15:13<36:21, 5.70s/it]
99%|█████████▉| 43557/43939 [70:15:19<36:20, 5.71s/it]
99%|█████████▉| 43558/43939 [70:15:24<36:15, 5.71s/it]
99%|█████████▉| 43559/43939 [70:15:30<36:05, 5.70s/it]
99%|█████████▉| 43560/43939 [70:15:36<35:57, 5.69s/it]
99%|█████████▉| 43561/43939 [70:15:41<35:49, 5.69s/it]
99%|█████████▉| 43562/43939 [70:15:47<35:44, 5.69s/it]
99%|█████████▉| 43563/43939 [70:15:53<35:39, 5.69s/it]
99%|█████████▉| 43564/43939 [70:15:58<35:31, 5.68s/it]
99%|█████████▉| 43565/43939 [70:16:04<35:26, 5.69s/it]
99%|█████████▉| 43566/43939 [70:16:10<35:52, 5.77s/it]
99%|█████████▉| 43567/43939 [70:16:16<35:41, 5.76s/it]
99%|█████████▉| 43568/43939 [70:16:21<35:32, 5.75s/it]
99%|█████████▉| 43569/43939 [70:16:27<35:17, 5.72s/it]
99%|█████████▉| 43570/43939 [70:16:33<35:11, 5.72s/it]
99%|█████████▉| 43571/43939 [70:16:39<35:02, 5.71s/it]
99%|█████████▉| 43572/43939 [70:16:44<34:54, 5.71s/it]
99%|█████████▉| 43573/43939 [70:16:50<34:50, 5.71s/it]
99%|█████████▉| 43574/43939 [70:16:56<34:40, 5.70s/it]
99%|█████████▉| 43575/43939 [70:17:01<34:34, 5.70s/it]
99%|█████████▉| 43576/43939 [70:17:07<34:29, 5.70s/it]
99%|█████████▉| 43577/43939 [70:17:13<34:24, 5.70s/it]
99%|█████████▉| 43578/43939 [70:17:18<34:14, 5.69s/it]
99%|█████████▉| 43579/43939 [70:17:24<34:13, 5.70s/it]
99%|█████████▉| 43580/43939 [70:17:30<34:06, 5.70s/it]
99%|█████████▉| 43581/43939 [70:17:36<34:00, 5.70s/it]
99%|█████████▉| 43582/43939 [70:17:41<33:52, 5.69s/it]
99%|█████████▉| 43583/43939 [70:17:47<33:48, 5.70s/it]
99%|█████████▉| 43584/43939 [70:17:53<33:44, 5.70s/it]
99%|█████████▉| 43585/43939 [70:17:58<33:35, 5.69s/it]
99%|█████████▉| 43586/43939 [70:18:04<33:30, 5.70s/it]
99%|█████████▉| 43587/43939 [70:18:10<33:25, 5.70s/it]
99%|█████████▉| 43588/43939 [70:18:16<33:40, 5.76s/it]
99%|█████████▉| 43589/43939 [70:18:21<33:29, 5.74s/it]
99%|█████████▉| 43590/43939 [70:18:27<33:20, 5.73s/it]
99%|█████████▉| 43591/43939 [70:18:33<33:12, 5.72s/it]
99%|█████████▉| 43592/43939 [70:18:38<33:00, 5.71s/it]
99%|█████████▉| 43593/43939 [70:18:44<32:52, 5.70s/it]
99%|█████████▉| 43594/43939 [70:18:50<32:51, 5.72s/it]
99%|█████████▉| 43595/43939 [70:18:56<32:45, 5.71s/it]
99%|█████████▉| 43596/43939 [70:19:01<32:38, 5.71s/it]
99%|█████████▉| 43597/43939 [70:19:07<32:34, 5.72s/it]
99%|█████████▉| 43598/43939 [70:19:13<32:31, 5.72s/it]
99%|█████████▉| 43599/43939 [70:19:18<32:28, 5.73s/it]
99%|█████████▉| 43600/43939 [70:19:24<32:25, 5.74s/it]
{'loss': 0.6545, 'grad_norm': 1.477016568183899, 'learning_rate': 9.118765871390933e-09, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 475.26, 'epoch': 3.97}
99%|█████████▉| 43600/43939 [70:19:24<32:25, 5.74s/it]
99%|█████████▉| 43601/43939 [70:19:30<32:21, 5.74s/it]
99%|█████████▉| 43602/43939 [70:19:36<32:12, 5.73s/it]
99%|█████████▉| 43603/43939 [70:19:41<32:00, 5.72s/it]
99%|█████████▉| 43604/43939 [70:19:47<31:53, 5.71s/it]
99%|█████████▉| 43605/43939 [70:19:53<31:47, 5.71s/it]
99%|█████████▉| 43606/43939 [70:19:59<31:48, 5.73s/it]
99%|█████████▉| 43607/43939 [70:20:04<31:39, 5.72s/it]
99%|█████████▉| 43608/43939 [70:20:10<31:31, 5.71s/it]
99%|█████████▉| 43609/43939 [70:20:16<31:48, 5.78s/it]
99%|█████████▉| 43610/43939 [70:20:22<31:36, 5.76s/it]
99%|█████████▉| 43611/43939 [70:20:27<31:25, 5.75s/it]
99%|█████████▉| 43612/43939 [70:20:33<31:14, 5.73s/it]
99%|█████████▉| 43613/43939 [70:20:39<31:06, 5.73s/it]
99%|█████████▉| 43614/43939 [70:20:44<31:00, 5.72s/it]
99%|█████████▉| 43615/43939 [70:20:50<30:54, 5.72s/it]
99%|█████████▉| 43616/43939 [70:20:56<30:46, 5.72s/it]
99%|█████████▉| 43617/43939 [70:21:02<30:42, 5.72s/it]
99%|█████████▉| 43618/43939 [70:21:07<30:34, 5.72s/it]
99%|█████████▉| 43619/43939 [70:21:13<30:30, 5.72s/it]
99%|█████████▉| 43620/43939 [70:21:19<30:24, 5.72s/it]
99%|█████████▉| 43621/43939 [70:21:24<30:18, 5.72s/it]
99%|█████████▉| 43622/43939 [70:21:30<30:11, 5.72s/it]
99%|█████████▉| 43623/43939 [70:21:36<30:04, 5.71s/it]
99%|█████████▉| 43624/43939 [70:21:42<29:58, 5.71s/it]
99%|█████████▉| 43625/43939 [70:21:47<29:52, 5.71s/it]
99%|█████████▉| 43626/43939 [70:21:53<29:48, 5.71s/it]
99%|█████████▉| 43627/43939 [70:21:59<29:40, 5.71s/it]
99%|█████████▉| 43628/43939 [70:22:04<29:34, 5.71s/it]
99%|█████████▉| 43629/43939 [70:22:10<29:28, 5.70s/it]
99%|█████████▉| 43630/43939 [70:22:16<29:23, 5.71s/it]
99%|█████████▉| 43631/43939 [70:22:22<29:38, 5.77s/it]
99%|█████████▉| 43632/43939 [70:22:27<29:27, 5.76s/it]
99%|█████████▉| 43633/43939 [70:22:33<29:18, 5.75s/it]
99%|█████████▉| 43634/43939 [70:22:39<29:11, 5.74s/it]
99%|█████████▉| 43635/43939 [70:22:45<29:01, 5.73s/it]
99%|█████████▉| 43636/43939 [70:22:50<28:55, 5.73s/it]
99%|█████████▉| 43637/43939 [70:22:56<28:47, 5.72s/it]
99%|█████████▉| 43638/43939 [70:23:02<28:38, 5.71s/it]
99%|█████████▉| 43639/43939 [70:23:07<28:35, 5.72s/it]
99%|█████████▉| 43640/43939 [70:23:13<28:28, 5.71s/it]
99%|█████████▉| 43641/43939 [70:23:19<28:21, 5.71s/it]
99%|█████████▉| 43642/43939 [70:23:25<28:13, 5.70s/it]
99%|█████████▉| 43643/43939 [70:23:30<28:20, 5.74s/it]
99%|█████████▉| 43644/43939 [70:23:36<28:12, 5.74s/it]
99%|█████████▉| 43645/43939 [70:23:42<28:08, 5.74s/it]
99%|█████████▉| 43646/43939 [70:23:48<27:59, 5.73s/it]
99%|█████████▉| 43647/43939 [70:23:53<27:54, 5.73s/it]
99%|█████████▉| 43648/43939 [70:23:59<27:47, 5.73s/it]
99%|█████████▉| 43649/43939 [70:24:05<27:41, 5.73s/it]
99%|█████████▉| 43650/43939 [70:24:11<27:37, 5.73s/it]
{'loss': 0.6709, 'grad_norm': 1.7487385272979736, 'learning_rate': 6.634090955676131e-09, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 497.31, 'epoch': 3.97}
99%|█████████▉| 43650/43939 [70:24:11<27:37, 5.73s/it]
99%|█████████▉| 43651/43939 [70:24:16<27:29, 5.73s/it]
99%|█████████▉| 43652/43939 [70:24:22<27:42, 5.79s/it]
99%|█████████▉| 43653/43939 [70:24:28<27:30, 5.77s/it]
99%|█████████▉| 43654/43939 [70:24:34<27:21, 5.76s/it]
99%|█████████▉| 43655/43939 [70:24:39<27:10, 5.74s/it]
99%|█████████▉| 43656/43939 [70:24:45<26:58, 5.72s/it]
99%|█████████▉| 43657/43939 [70:24:51<26:52, 5.72s/it]
99%|█████████▉| 43658/43939 [70:24:56<26:46, 5.72s/it]
99%|█████████▉| 43659/43939 [70:25:02<26:37, 5.70s/it]
99%|█████████▉| 43660/43939 [70:25:08<26:31, 5.70s/it]
99%|█████████▉| 43661/43939 [70:25:14<26:26, 5.71s/it]
99%|█████████▉| 43662/43939 [70:25:19<26:20, 5.71s/it]
99%|█████████▉| 43663/43939 [70:25:25<26:12, 5.70s/it]
99%|█████████▉| 43664/43939 [70:25:31<26:07, 5.70s/it]
99%|█████████▉| 43665/43939 [70:25:36<26:02, 5.70s/it]
99%|█████████▉| 43666/43939 [70:25:42<25:56, 5.70s/it]
99%|█████████▉| 43667/43939 [70:25:48<25:51, 5.70s/it]
99%|█████████▉| 43668/43939 [70:25:53<25:46, 5.71s/it]
99%|█████████▉| 43669/43939 [70:25:59<25:40, 5.70s/it]
99%|█████████▉| 43670/43939 [70:26:05<25:38, 5.72s/it]
99%|█████████▉| 43671/43939 [70:26:11<25:30, 5.71s/it]
99%|█████████▉| 43672/43939 [70:26:16<25:23, 5.70s/it]
99%|█████████▉| 43673/43939 [70:26:22<25:13, 5.69s/it]
99%|█████████▉| 43674/43939 [70:26:28<25:29, 5.77s/it]
99%|█████████▉| 43675/43939 [70:26:34<25:20, 5.76s/it]
99%|█████████▉| 43676/43939 [70:26:39<25:12, 5.75s/it]
99%|█████████▉| 43677/43939 [70:26:45<25:01, 5.73s/it]
99%|█████████▉| 43678/43939 [70:26:51<24:54, 5.73s/it]
99%|█████████▉| 43679/43939 [70:26:56<24:46, 5.72s/it]
99%|█████████▉| 43680/43939 [70:27:02<24:42, 5.72s/it]
99%|█████████▉| 43681/43939 [70:27:08<24:32, 5.71s/it]
99%|█████████▉| 43682/43939 [70:27:14<24:24, 5.70s/it]
99%|█████████▉| 43683/43939 [70:27:19<24:18, 5.70s/it]
99%|█████████▉| 43684/43939 [70:27:25<24:10, 5.69s/it]
99%|█████████▉| 43685/43939 [70:27:31<24:04, 5.69s/it]
99%|█████████▉| 43686/43939 [70:27:36<24:00, 5.69s/it]
99%|█████████▉| 43687/43939 [70:27:42<23:55, 5.70s/it]
99%|█████████▉| 43688/43939 [70:27:48<23:48, 5.69s/it]
99%|█████████▉| 43689/43939 [70:27:53<23:44, 5.70s/it]
99%|█████████▉| 43690/43939 [70:27:59<23:40, 5.70s/it]
99%|█████████▉| 43691/43939 [70:28:05<23:33, 5.70s/it]
99%|█████████▉| 43692/43939 [70:28:10<23:26, 5.69s/it]
99%|█████████▉| 43693/43939 [70:28:16<23:22, 5.70s/it]
99%|█████████▉| 43694/43939 [70:28:22<23:17, 5.70s/it]
99%|█████████▉| 43695/43939 [70:28:28<23:28, 5.77s/it]
99%|█████████▉| 43696/43939 [70:28:34<23:15, 5.74s/it]
99%|█████████▉| 43697/43939 [70:28:39<23:06, 5.73s/it]
99%|█████████▉| 43698/43939 [70:28:45<23:00, 5.73s/it]
99%|█████████▉| 43699/43939 [70:28:51<22:55, 5.73s/it]
99%|█████████▉| 43700/43939 [70:28:56<22:46, 5.72s/it]
{'loss': 0.661, 'grad_norm': 1.6341969966888428, 'learning_rate': 4.5437451207802624e-09, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 480.91, 'epoch': 3.98}
99%|█████████▉| 43700/43939 [70:28:56<22:46, 5.72s/it]
99%|█████████▉| 43701/43939 [70:29:02<22:41, 5.72s/it]
99%|█████████▉| 43702/43939 [70:29:08<22:32, 5.71s/it]
99%|█████████▉| 43703/43939 [70:29:13<22:26, 5.70s/it]
99%|█████████▉| 43704/43939 [70:29:19<22:17, 5.69s/it]
99%|█████████▉| 43705/43939 [70:29:25<22:10, 5.69s/it]
99%|█████████▉| 43706/43939 [70:29:31<22:07, 5.70s/it]
99%|█████████▉| 43707/43939 [70:29:36<21:58, 5.68s/it]
99%|█████████▉| 43708/43939 [70:29:42<21:53, 5.69s/it]
99%|█████████▉| 43709/43939 [70:29:48<21:47, 5.69s/it]
99%|█████████▉| 43710/43939 [70:29:53<21:42, 5.69s/it]
99%|█████████▉| 43711/43939 [70:29:59<21:38, 5.69s/it]
99%|█████████▉| 43712/43939 [70:30:05<21:33, 5.70s/it]
99%|█████████▉| 43713/43939 [70:30:10<21:28, 5.70s/it]
99%|█████████▉| 43714/43939 [70:30:16<21:21, 5.70s/it]
99%|█████████▉| 43715/43939 [70:30:22<21:16, 5.70s/it]
99%|█████████▉| 43716/43939 [70:30:27<21:07, 5.68s/it]
99%|█████████▉| 43717/43939 [70:30:33<21:19, 5.76s/it]
99%|█████████▉| 43718/43939 [70:30:39<21:11, 5.75s/it]
99%|█████████▉| 43719/43939 [70:30:45<21:00, 5.73s/it]
100%|█████████▉| 43720/43939 [70:30:51<20:57, 5.74s/it]
100%|█████████▉| 43721/43939 [70:30:56<20:49, 5.73s/it]
100%|█████████▉| 43722/43939 [70:31:02<20:42, 5.73s/it]
100%|█████████▉| 43723/43939 [70:31:08<20:35, 5.72s/it]
100%|█████████▉| 43724/43939 [70:31:13<20:31, 5.73s/it]
100%|█████████▉| 43725/43939 [70:31:19<20:23, 5.72s/it]
100%|█████████▉| 43726/43939 [70:31:25<20:18, 5.72s/it]
100%|█████████▉| 43727/43939 [70:31:31<20:11, 5.71s/it]
100%|█████████▉| 43728/43939 [70:31:36<20:03, 5.71s/it]
100%|█████████▉| 43729/43939 [70:31:42<19:58, 5.71s/it]
100%|█████████▉| 43730/43939 [70:31:48<19:52, 5.71s/it]
100%|█████████▉| 43731/43939 [70:31:53<19:48, 5.71s/it]
100%|█████████▉| 43732/43939 [70:31:59<19:44, 5.72s/it]
100%|█████████▉| 43733/43939 [70:32:05<19:39, 5.72s/it]
100%|█████████▉| 43734/43939 [70:32:11<19:32, 5.72s/it]
100%|█████████▉| 43735/43939 [70:32:16<19:32, 5.75s/it]
100%|█████████▉| 43736/43939 [70:32:22<19:26, 5.75s/it]
100%|█████████▉| 43737/43939 [70:32:28<19:19, 5.74s/it]
100%|█████████▉| 43738/43939 [70:32:34<19:27, 5.81s/it]
100%|█████████▉| 43739/43939 [70:32:40<19:17, 5.79s/it]
100%|█████████▉| 43740/43939 [70:32:45<19:06, 5.76s/it]
100%|█████████▉| 43741/43939 [70:32:51<18:56, 5.74s/it]
100%|█████████▉| 43742/43939 [70:32:57<18:45, 5.71s/it]
100%|█████████▉| 43743/43939 [70:33:02<18:37, 5.70s/it]
100%|█████████▉| 43744/43939 [70:33:08<18:33, 5.71s/it]
100%|█████████▉| 43745/43939 [70:33:14<18:32, 5.73s/it]
100%|█████████▉| 43746/43939 [70:33:19<18:25, 5.73s/it]
100%|█████████▉| 43747/43939 [70:33:25<18:18, 5.72s/it]
100%|█████████▉| 43748/43939 [70:33:31<18:12, 5.72s/it]
100%|█████████▉| 43749/43939 [70:33:37<18:07, 5.72s/it]
100%|█████████▉| 43750/43939 [70:33:42<18:02, 5.73s/it]
{'loss': 0.6514, 'grad_norm': 1.680708885192871, 'learning_rate': 2.8477613468219643e-09, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 509.85, 'epoch': 3.98}
100%|█████████▉| 43750/43939 [70:33:42<18:02, 5.73s/it]
100%|█████████▉| 43751/43939 [70:33:48<17:56, 5.73s/it]
100%|█████████▉| 43752/43939 [70:33:54<17:49, 5.72s/it]
100%|█████████▉| 43753/43939 [70:33:59<17:42, 5.71s/it]
100%|█████████▉| 43754/43939 [70:34:05<17:33, 5.70s/it]
100%|█████████▉| 43755/43939 [70:34:11<17:28, 5.70s/it]
100%|█████████▉| 43756/43939 [70:34:17<17:23, 5.70s/it]
100%|█████████▉| 43757/43939 [70:34:22<17:14, 5.69s/it]
100%|█████████▉| 43758/43939 [70:34:28<17:14, 5.71s/it]
100%|█████████▉| 43759/43939 [70:34:34<17:08, 5.71s/it]
100%|█████████▉| 43760/43939 [70:34:40<17:13, 5.78s/it]
100%|█████████▉| 43761/43939 [70:34:45<17:03, 5.75s/it]
100%|█████████▉| 43762/43939 [70:34:51<16:56, 5.75s/it]
100%|█████████▉| 43763/43939 [70:34:57<16:47, 5.72s/it]
100%|█████████▉| 43764/43939 [70:35:02<16:38, 5.71s/it]
100%|█████████▉| 43765/43939 [70:35:08<16:32, 5.70s/it]
100%|█████████▉| 43766/43939 [70:35:14<16:24, 5.69s/it]
100%|█████████▉| 43767/43939 [70:35:19<16:19, 5.69s/it]
100%|█████████▉| 43768/43939 [70:35:25<16:14, 5.70s/it]
100%|█████████▉| 43769/43939 [70:35:31<16:11, 5.71s/it]
100%|█████████▉| 43770/43939 [70:35:37<16:04, 5.70s/it]
100%|█████████▉| 43771/43939 [70:35:42<15:58, 5.71s/it]
100%|█████████▉| 43772/43939 [70:35:48<15:52, 5.70s/it]
100%|█████████▉| 43773/43939 [70:35:54<15:45, 5.70s/it]
100%|█████████▉| 43774/43939 [70:35:59<15:40, 5.70s/it]
100%|█████████▉| 43775/43939 [70:36:05<15:35, 5.70s/it]
100%|█████████▉| 43776/43939 [70:36:11<15:28, 5.70s/it]
100%|█████████▉| 43777/43939 [70:36:17<15:23, 5.70s/it]
100%|█████████▉| 43778/43939 [70:36:22<15:18, 5.70s/it]
100%|█████████▉| 43779/43939 [70:36:28<15:10, 5.69s/it]
100%|█████████▉| 43780/43939 [70:36:34<15:03, 5.68s/it]
100%|█████████▉| 43781/43939 [70:36:39<15:09, 5.76s/it]
100%|█████████▉| 43782/43939 [70:36:45<15:00, 5.74s/it]
100%|█████████▉| 43783/43939 [70:36:51<14:55, 5.74s/it]
100%|█████████▉| 43784/43939 [70:36:57<14:48, 5.73s/it]
100%|█████████▉| 43785/43939 [70:37:02<14:41, 5.72s/it]
100%|█████████▉| 43786/43939 [70:37:08<14:34, 5.71s/it]
100%|█████████▉| 43787/43939 [70:37:14<14:27, 5.71s/it]
100%|█████████▉| 43788/43939 [70:37:19<14:23, 5.72s/it]
100%|█████████▉| 43789/43939 [70:37:25<14:17, 5.72s/it]
100%|█████████▉| 43790/43939 [70:37:31<14:11, 5.71s/it]
100%|█████████▉| 43791/43939 [70:37:37<14:06, 5.72s/it]
100%|█████████▉| 43792/43939 [70:37:42<13:59, 5.71s/it]
100%|█████████▉| 43793/43939 [70:37:48<13:53, 5.71s/it]
100%|█████████▉| 43794/43939 [70:37:54<13:48, 5.72s/it]
100%|█████████▉| 43795/43939 [70:37:59<13:41, 5.70s/it]
100%|█████████▉| 43796/43939 [70:38:05<13:35, 5.71s/it]
100%|█████████▉| 43797/43939 [70:38:11<13:30, 5.71s/it]
100%|█████████▉| 43798/43939 [70:38:17<13:24, 5.71s/it]
100%|█████████▉| 43799/43939 [70:38:22<13:19, 5.71s/it]
100%|█████████▉| 43800/43939 [70:38:28<13:13, 5.71s/it]
{'loss': 0.6397, 'grad_norm': 1.7303528785705566, 'learning_rate': 1.5461663919302815e-09, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 484.5, 'epoch': 3.99}
100%|█████████▉| 43800/43939 [70:38:28<13:13, 5.71s/it]
100%|█████████▉| 43801/43939 [70:38:34<13:08, 5.71s/it]
100%|█████████▉| 43802/43939 [70:38:39<13:01, 5.70s/it]
100%|█████████▉| 43803/43939 [70:38:45<13:04, 5.77s/it]
100%|█████████▉| 43804/43939 [70:38:51<12:56, 5.75s/it]
100%|█████████▉| 43805/43939 [70:38:57<12:51, 5.76s/it]
100%|█████████▉| 43806/43939 [70:39:02<12:42, 5.73s/it]
100%|█████████▉| 43807/43939 [70:39:08<12:36, 5.73s/it]
100%|█████████▉| 43808/43939 [70:39:14<12:31, 5.74s/it]
100%|█████████▉| 43809/43939 [70:39:20<12:26, 5.74s/it]
100%|█████████▉| 43810/43939 [70:39:25<12:17, 5.72s/it]
100%|█████████▉| 43811/43939 [70:39:31<12:13, 5.73s/it]
100%|█████████▉| 43812/43939 [70:39:37<12:07, 5.73s/it]
100%|█████████▉| 43813/43939 [70:39:42<12:00, 5.72s/it]
100%|█████████▉| 43814/43939 [70:39:48<11:54, 5.72s/it]
100%|█████████▉| 43815/43939 [70:39:54<11:47, 5.71s/it]
100%|█████████▉| 43816/43939 [70:40:00<11:41, 5.70s/it]
100%|█████████▉| 43817/43939 [70:40:05<11:36, 5.71s/it]
100%|█████████▉| 43818/43939 [70:40:11<11:29, 5.70s/it]
100%|█████████▉| 43819/43939 [70:40:17<11:23, 5.70s/it]
100%|█████████▉| 43820/43939 [70:40:22<11:20, 5.72s/it]
100%|█████████▉| 43821/43939 [70:40:28<11:13, 5.71s/it]
100%|█████████▉| 43822/43939 [70:40:34<11:08, 5.71s/it]
100%|█████████▉| 43823/43939 [70:40:40<11:01, 5.71s/it]
100%|█████████▉| 43824/43939 [70:40:45<10:55, 5.70s/it]
100%|█████████▉| 43825/43939 [70:40:51<10:59, 5.79s/it]
100%|█████████▉| 43826/43939 [70:40:57<10:52, 5.78s/it]
100%|█████████▉| 43827/43939 [70:41:03<10:43, 5.74s/it]
100%|█████████▉| 43828/43939 [70:41:08<10:35, 5.72s/it]
100%|█████████▉| 43829/43939 [70:41:14<10:28, 5.72s/it]
100%|█████████▉| 43830/43939 [70:41:20<10:22, 5.71s/it]
100%|█████████▉| 43831/43939 [70:41:25<10:17, 5.71s/it]
100%|█████████▉| 43832/43939 [70:41:31<10:11, 5.71s/it]
100%|█████████▉| 43833/43939 [70:41:37<10:06, 5.72s/it]
100%|█████████▉| 43834/43939 [70:41:43<10:00, 5.72s/it]
100%|█████████▉| 43835/43939 [70:41:48<09:53, 5.71s/it]
100%|█████████▉| 43836/43939 [70:41:54<09:45, 5.69s/it]
100%|█████████▉| 43837/43939 [70:42:00<09:42, 5.71s/it]
100%|█████████▉| 43838/43939 [70:42:05<09:36, 5.71s/it]
100%|█████████▉| 43839/43939 [70:42:11<09:29, 5.70s/it]
100%|█████████▉| 43840/43939 [70:42:17<09:23, 5.70s/it]
100%|█████████▉| 43841/43939 [70:42:22<09:18, 5.70s/it]
100%|█████████▉| 43842/43939 [70:42:28<09:12, 5.70s/it]
100%|█████████▉| 43843/43939 [70:42:34<09:07, 5.70s/it]
100%|█████████▉| 43844/43939 [70:42:40<09:01, 5.70s/it]
100%|█████████▉| 43845/43939 [70:42:45<08:55, 5.69s/it]
100%|█████████▉| 43846/43939 [70:42:51<08:54, 5.75s/it]
100%|█████████▉| 43847/43939 [70:42:57<08:47, 5.73s/it]
100%|█████████▉| 43848/43939 [70:43:03<08:41, 5.73s/it]
100%|█████████▉| 43849/43939 [70:43:08<08:36, 5.74s/it]
100%|█████████▉| 43850/43939 [70:43:14<08:30, 5.73s/it]
{'loss': 0.671, 'grad_norm': 1.6371973752975464, 'learning_rate': 6.389807918227852e-10, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 510.08, 'epoch': 3.99}
100%|█████████▉| 43850/43939 [70:43:14<08:30, 5.73s/it]
100%|█████████▉| 43851/43939 [70:43:20<08:24, 5.73s/it]
100%|█████████▉| 43852/43939 [70:43:25<08:17, 5.72s/it]
100%|█████████▉| 43853/43939 [70:43:31<08:11, 5.71s/it]
100%|█████████▉| 43854/43939 [70:43:37<08:05, 5.72s/it]
100%|█████████▉| 43855/43939 [70:43:43<07:59, 5.71s/it]
100%|█████████▉| 43856/43939 [70:43:48<07:53, 5.70s/it]
100%|█████████▉| 43857/43939 [70:43:54<07:47, 5.70s/it]
100%|█████████▉| 43858/43939 [70:44:00<07:42, 5.71s/it]
100%|█████████▉| 43859/43939 [70:44:05<07:38, 5.74s/it]
100%|█████████▉| 43860/43939 [70:44:11<07:31, 5.72s/it]
100%|█████████▉| 43861/43939 [70:44:17<07:25, 5.71s/it]
100%|█████████▉| 43862/43939 [70:44:23<07:19, 5.71s/it]
100%|█████████▉| 43863/43939 [70:44:28<07:13, 5.70s/it]
100%|█████████▉| 43864/43939 [70:44:34<07:07, 5.71s/it]
100%|█████████▉| 43865/43939 [70:44:40<07:01, 5.70s/it]
100%|█████████▉| 43866/43939 [70:44:45<06:55, 5.69s/it]
100%|█████████▉| 43867/43939 [70:44:51<06:49, 5.69s/it]
100%|█████████▉| 43868/43939 [70:44:57<06:49, 5.77s/it]
100%|█████████▉| 43869/43939 [70:45:03<06:42, 5.75s/it]
100%|█████████▉| 43870/43939 [70:45:08<06:35, 5.74s/it]
100%|█████████▉| 43871/43939 [70:45:14<06:29, 5.72s/it]
100%|█████████▉| 43872/43939 [70:45:20<06:23, 5.72s/it]
100%|█████████▉| 43873/43939 [70:45:25<06:16, 5.71s/it]
100%|█████████▉| 43874/43939 [70:45:31<06:11, 5.72s/it]
100%|█████████▉| 43875/43939 [70:45:37<06:06, 5.72s/it]
100%|█████████▉| 43876/43939 [70:45:43<05:59, 5.71s/it]
100%|█████████▉| 43877/43939 [70:45:48<05:54, 5.71s/it]
100%|█████████▉| 43878/43939 [70:45:54<05:47, 5.70s/it]
100%|█████████▉| 43879/43939 [70:46:00<05:41, 5.70s/it]
100%|█████████▉| 43880/43939 [70:46:05<05:36, 5.70s/it]
100%|█████████▉| 43881/43939 [70:46:11<05:29, 5.69s/it]
100%|█████████▉| 43882/43939 [70:46:17<05:24, 5.69s/it]
100%|█████████▉| 43883/43939 [70:46:22<05:18, 5.69s/it]
100%|█████████▉| 43884/43939 [70:46:28<05:12, 5.69s/it]
100%|█████████▉| 43885/43939 [70:46:34<05:07, 5.69s/it]
100%|█████████▉| 43886/43939 [70:46:39<05:01, 5.69s/it]
100%|█████████▉| 43887/43939 [70:46:45<04:55, 5.69s/it]
100%|█████████▉| 43888/43939 [70:46:51<04:49, 5.68s/it]
100%|█████████▉| 43889/43939 [70:46:57<04:47, 5.75s/it]
100%|█████████▉| 43890/43939 [70:47:02<04:40, 5.73s/it]
100%|█████████▉| 43891/43939 [70:47:08<04:34, 5.71s/it]
100%|█████████▉| 43892/43939 [70:47:14<04:29, 5.74s/it]
100%|█████████▉| 43893/43939 [70:47:20<04:23, 5.72s/it]
100%|█████████▉| 43894/43939 [70:47:25<04:17, 5.72s/it]
100%|█████████▉| 43895/43939 [70:47:31<04:11, 5.72s/it]
100%|█████████▉| 43896/43939 [70:47:37<04:05, 5.71s/it]
100%|█████████▉| 43897/43939 [70:47:42<04:00, 5.73s/it]
100%|█████████▉| 43898/43939 [70:47:48<03:54, 5.73s/it]
100%|█████████▉| 43899/43939 [70:47:54<03:48, 5.72s/it]
100%|█████████▉| 43900/43939 [70:48:00<03:43, 5.72s/it]
{'loss': 0.635, 'grad_norm': 1.7206001281738281, 'learning_rate': 1.2621885948360667e-10, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'tokens_per_second_per_gpu': 460.19, 'epoch': 4.0}
100%|█████████▉| 43900/43939 [70:48:00<03:43, 5.72s/it]
100%|█████████▉| 43901/43939 [70:48:05<03:36, 5.71s/it]
100%|█████████▉| 43902/43939 [70:48:11<03:31, 5.72s/it]
100%|█████████▉| 43903/43939 [70:48:17<03:25, 5.71s/it]
100%|█████████▉| 43904/43939 [70:48:22<03:19, 5.71s/it]
100%|█████████▉| 43905/43939 [70:48:28<03:14, 5.72s/it]
100%|█████████▉| 43906/43939 [70:48:34<03:08, 5.72s/it]
100%|█████████▉| 43907/43939 [70:48:40<03:02, 5.70s/it]
100%|█████████▉| 43908/43939 [70:48:45<02:56, 5.70s/it]
100%|█████████▉| 43909/43939 [70:48:51<02:50, 5.70s/it]
100%|█████████▉| 43910/43939 [70:48:57<02:45, 5.69s/it]
100%|█████████▉| 43911/43939 [70:49:03<02:41, 5.76s/it]
100%|█████████▉| 43912/43939 [70:49:08<02:35, 5.75s/it]
100%|█████████▉| 43913/43939 [70:49:14<02:28, 5.73s/it]
100%|█████████▉| 43914/43939 [70:49:20<02:22, 5.71s/it]
100%|█████████▉| 43915/43939 [70:49:25<02:17, 5.71s/it]
100%|█████████▉| 43916/43939 [70:49:31<02:11, 5.70s/it]
100%|█████████▉| 43917/43939 [70:49:37<02:05, 5.70s/it]
100%|█████████▉| 43918/43939 [70:49:42<01:59, 5.71s/it]
100%|█████████▉| 43919/43939 [70:49:48<01:53, 5.69s/it]
100%|█████████▉| 43920/43939 [70:49:54<01:48, 5.70s/it]
100%|█████████▉| 43921/43939 [70:50:00<01:42, 5.71s/it]
100%|█████████▉| 43922/43939 [70:50:05<01:37, 5.71s/it]
100%|█████████▉| 43923/43939 [70:50:11<01:31, 5.70s/it]
100%|█████████▉| 43924/43939 [70:50:17<01:25, 5.71s/it]
100%|█████████▉| 43925/43939 [70:50:22<01:20, 5.72s/it]
100%|█████████▉| 43926/43939 [70:50:28<01:14, 5.73s/it]
100%|█████████▉| 43927/43939 [70:50:34<01:08, 5.72s/it]
100%|█████████▉| 43928/43939 [70:50:40<01:02, 5.71s/it]
100%|█████████▉| 43929/43939 [70:50:45<00:57, 5.73s/it]
100%|█████████▉| 43930/43939 [70:50:51<00:51, 5.72s/it]
100%|█████████▉| 43931/43939 [70:50:57<00:45, 5.72s/it]
100%|█████████▉| 43932/43939 [70:51:03<00:40, 5.77s/it]
100%|█████████▉| 43933/43939 [70:51:08<00:34, 5.76s/it]
100%|█████████▉| 43934/43939 [70:51:14<00:28, 5.75s/it]
100%|█████████▉| 43935/43939 [70:51:20<00:22, 5.74s/it]
100%|█████████▉| 43936/43939 [70:51:26<00:17, 5.74s/it]
100%|█████████▉| 43937/43939 [70:51:31<00:11, 5.72s/it]
100%|█████████▉| 43938/43939 [70:51:37<00:05, 5.72s/it]
100%|██████████| 43939/43939 [70:51:43<00:00, 5.71s/it][2025-12-01 16:55:56,475] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft/checkpoint-43939
{'train_runtime': 255105.2639, 'train_samples_per_second': 1.55, 'train_steps_per_second': 0.172, 'train_loss': 0.864465983982538, 'memory/max_active (GiB)': 75.64, 'memory/max_allocated (GiB)': 75.64, 'memory/device_reserved (GiB)': 77.07, 'epoch': 4.0}
100%|██████████| 43939/43939 [70:51:45<00:00, 5.71s/it]
100%|██████████| 43939/43939 [70:51:45<00:00, 5.81s/it]
[2025-12-01 16:56:13,832] [INFO] [axolotl.train.save_trained_model:218] [PID:3633661] Training completed! Saving trained model to /model_out/mistral-nemo-12b_sft.
[2025-12-01 16:56:14,638] [INFO] [axolotl.train.save_trained_model:336] [PID:3633661] Model successfully saved to /model_out/mistral-nemo-12b_sft
[2025-12-01 16:56:14,762] [INFO] [axolotl.core.trainers.base._save:665] [PID:3633661] Saving model checkpoint to /model_out/mistral-nemo-12b_sft
Processing Files (0 / 0) : | | 0.00B / 0.00B
New Data Upload : | | 0.00B / 0.00B [A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 26%|██▌ | 117MB / 456MB [A[A[A[A[A[A[A[A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 26%|██▌ | 117MB / 456MB [A[A[A[A[A[A[A[A
Processing Files (6 / 7) : 28%|██▊ | 135MB / 474MB, ???B/s
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 51%|█████▏ | 235MB / 456MB [A[A[A[A[A[A[A[A
Processing Files (6 / 7) : 53%|█████▎ | 252MB / 474MB, 586MB/s
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 75%|███████▌ | 344MB / 456MB [A[A[A[A[A[A[A[A
Processing Files (6 / 7) : 76%|███████▋ | 361MB / 474MB, 565MB/s
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
Processing Files (7 / 7) : 100%|██████████| 474MB / 474MB, 564MB/s
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB [A[A
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB [A[A[A
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB [A[A[A[A
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB [A[A[A[A[A
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB [A[A[A[A[A[A
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB [A[A[A[A[A[A[A
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB [A[A[A[A[A[A[A[A
Processing Files (7 / 7) : 100%|██████████| 474MB / 474MB, 242MB/s
New Data Upload : | | 0.00B / 0.00B, 0.00B/s
...12b_sft/training_args.bin: 100%|██████████| 11.4kB / 11.4kB
...v.alces.network.1563957.0: 100%|██████████| 13.2kB / 13.2kB
...v.alces.network.3509172.0: 100%|██████████| 11.7kB / 11.7kB
...v.alces.network.3588845.0: 100%|██████████| 15.7kB / 15.7kB
...v.alces.network.3633661.0: 100%|██████████| 458kB / 458kB
...mo-12b_sft/tokenizer.json: 100%|██████████| 17.1MB / 17.1MB
...adapter_model.safetensors: 100%|██████████| 456MB / 456MB