初始化项目，由ModelHub XC社区提供模型
Model: lldois/v05_user_item_up_lr2e5 Source: Original Platform
This commit is contained in:
ModelHub XC
36
.gitattributes
vendored
Normal file
36
.gitattributes
vendored
Normal file
@@ -0,0 +1,36 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
58
README.md
Normal file
58
README.md
Normal file
@@ -0,0 +1,58 @@
|
||||
---
|
||||
library_name: transformers
|
||||
license: other
|
||||
base_model: OpenOneRec/OneReason-0.8B-pretrain-competition
|
||||
tags:
|
||||
- llama-factory
|
||||
- full
|
||||
- generated_from_trainer
|
||||
model-index:
|
||||
- name: v05_user_item_up_lr2e5
|
||||
results: []
|
||||
---
|
||||
|
||||
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
||||
should probably proofread and complete it, then remove this comment. -->
|
||||
|
||||
# v05_user_item_up_lr2e5
|
||||
|
||||
This model is a fine-tuned version of [OpenOneRec/OneReason-0.8B-pretrain-competition](https://huggingface.co/OpenOneRec/OneReason-0.8B-pretrain-competition) on the overnight_user_item_up dataset.
|
||||
|
||||
## Model description
|
||||
|
||||
More information needed
|
||||
|
||||
## Intended uses & limitations
|
||||
|
||||
More information needed
|
||||
|
||||
## Training and evaluation data
|
||||
|
||||
More information needed
|
||||
|
||||
## Training procedure
|
||||
|
||||
### Training hyperparameters
|
||||
|
||||
The following hyperparameters were used during training:
|
||||
- learning_rate: 2e-05
|
||||
- train_batch_size: 1
|
||||
- eval_batch_size: 8
|
||||
- seed: 202607065
|
||||
- gradient_accumulation_steps: 4
|
||||
- total_train_batch_size: 4
|
||||
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
||||
- lr_scheduler_type: cosine
|
||||
- lr_scheduler_warmup_steps: 0.03
|
||||
- num_epochs: 1
|
||||
|
||||
### Training results
|
||||
|
||||
|
||||
|
||||
### Framework versions
|
||||
|
||||
- Transformers 5.6.0
|
||||
- Pytorch 2.7.1+cu126
|
||||
- Datasets 4.0.0
|
||||
- Tokenizers 0.22.2
|
||||
8
all_results.json
Normal file
8
all_results.json
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"epoch": 1.0,
|
||||
"total_flos": 3.5281465280347546e+17,
|
||||
"train_loss": 1.4733744319910553,
|
||||
"train_runtime": 6290.4242,
|
||||
"train_samples_per_second": 0.465,
|
||||
"train_steps_per_second": 0.116
|
||||
}
|
||||
89
chat_template.jinja
Normal file
89
chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
||||
{%- if tools %}
|
||||
{{- '<|im_start|>system\n' }}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- messages[0].content + '\n\n' }}
|
||||
{%- endif %}
|
||||
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||
{%- for tool in tools %}
|
||||
{{- "\n" }}
|
||||
{{- tool | tojson }}
|
||||
{%- endfor %}
|
||||
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||
{%- else %}
|
||||
{%- if messages[0].role == 'system' %}
|
||||
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||
{%- for message in messages[::-1] %}
|
||||
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||
{%- set ns.multi_step_tool = false %}
|
||||
{%- set ns.last_query_index = index %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- for message in messages %}
|
||||
{%- if message.content is string %}
|
||||
{%- set content = message.content %}
|
||||
{%- else %}
|
||||
{%- set content = '' %}
|
||||
{%- endif %}
|
||||
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||
{%- elif message.role == "assistant" %}
|
||||
{%- set reasoning_content = '' %}
|
||||
{%- if message.reasoning_content is string %}
|
||||
{%- set reasoning_content = message.reasoning_content %}
|
||||
{%- else %}
|
||||
{%- if '</think>' in content %}
|
||||
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- if loop.index0 > ns.last_query_index %}
|
||||
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- else %}
|
||||
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||
{%- endif %}
|
||||
{%- if message.tool_calls %}
|
||||
{%- for tool_call in message.tool_calls %}
|
||||
{%- if (loop.first and content) or (not loop.first) %}
|
||||
{{- '\n' }}
|
||||
{%- endif %}
|
||||
{%- if tool_call.function %}
|
||||
{%- set tool_call = tool_call.function %}
|
||||
{%- endif %}
|
||||
{{- '<tool_call>\n{"name": "' }}
|
||||
{{- tool_call.name }}
|
||||
{{- '", "arguments": ' }}
|
||||
{%- if tool_call.arguments is string %}
|
||||
{{- tool_call.arguments }}
|
||||
{%- else %}
|
||||
{{- tool_call.arguments | tojson }}
|
||||
{%- endif %}
|
||||
{{- '}\n</tool_call>' }}
|
||||
{%- endfor %}
|
||||
{%- endif %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- elif message.role == "tool" %}
|
||||
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||
{{- '<|im_start|>user' }}
|
||||
{%- endif %}
|
||||
{{- '\n<tool_response>\n' }}
|
||||
{{- content }}
|
||||
{{- '\n</tool_response>' }}
|
||||
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||
{{- '<|im_end|>\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
{%- endfor %}
|
||||
{%- if add_generation_prompt %}
|
||||
{{- '<|im_start|>assistant\n' }}
|
||||
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||
{{- '<think>\n\n</think>\n\n' }}
|
||||
{%- endif %}
|
||||
{%- endif %}
|
||||
63
config.json
Normal file
63
config.json
Normal file
@@ -0,0 +1,63 @@
|
||||
{
|
||||
"architectures": [
|
||||
"Qwen3ForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": null,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 151645,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 1024,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 3072,
|
||||
"layer_types": [
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention",
|
||||
"full_attention"
|
||||
],
|
||||
"max_position_embeddings": 40960,
|
||||
"max_window_layers": 28,
|
||||
"model_type": "qwen3",
|
||||
"num_attention_heads": 16,
|
||||
"num_hidden_layers": 28,
|
||||
"num_key_value_heads": 8,
|
||||
"pad_token_id": 151643,
|
||||
"rms_norm_eps": 1e-06,
|
||||
"rope_parameters": {
|
||||
"rope_theta": 1000000,
|
||||
"rope_type": "default"
|
||||
},
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": false,
|
||||
"transformers_version": "5.6.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 176253
|
||||
}
|
||||
12
generation_config.json
Normal file
12
generation_config.json
Normal file
@@ -0,0 +1,12 @@
|
||||
{
|
||||
"do_sample": true,
|
||||
"eos_token_id": [
|
||||
151645,
|
||||
151643
|
||||
],
|
||||
"pad_token_id": 151643,
|
||||
"temperature": 0.6,
|
||||
"top_k": 20,
|
||||
"top_p": 0.95,
|
||||
"transformers_version": "5.6.0"
|
||||
}
|
||||
3
model.safetensors
Normal file
3
model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:912c559b24c10118f8bc0a637821c72c4a0d2edef99bc8d05d3bb1013d6bda23
|
||||
size 1602902832
|
||||
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f6d4269b975b40bccbbbe27389b955ee55265969ea8741b7da4c69644ee356ff
|
||||
size 37026
|
||||
3
tokenizer.json
Normal file
3
tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:cd4d15f596979aecbc11ba12668cb21c9fb8452ce1235cd9d7878cc950170421
|
||||
size 16016554
|
||||
16
tokenizer_config.json
Normal file
16
tokenizer_config.json
Normal file
@@ -0,0 +1,16 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"is_local": false,
|
||||
"local_files_only": false,
|
||||
"model_max_length": 131072,
|
||||
"pad_token": "<|endoftext|>",
|
||||
"padding_side": "right",
|
||||
"split_special_tokens": false,
|
||||
"tokenizer_class": "Qwen2Tokenizer",
|
||||
"unk_token": null
|
||||
}
|
||||
8
train_results.json
Normal file
8
train_results.json
Normal file
@@ -0,0 +1,8 @@
|
||||
{
|
||||
"epoch": 1.0,
|
||||
"total_flos": 3.5281465280347546e+17,
|
||||
"train_loss": 1.4733744319910553,
|
||||
"train_runtime": 6290.4242,
|
||||
"train_samples_per_second": 0.465,
|
||||
"train_steps_per_second": 0.116
|
||||
}
|
||||
147
trainer_log.jsonl
Normal file
147
trainer_log.jsonl
Normal file
@@ -0,0 +1,147 @@
|
||||
{"current_steps": 5, "total_steps": 731, "loss": 2.809466743469238, "lr": 3.6363636363636366e-06, "epoch": 0.006842285323297981, "percentage": 0.68, "elapsed_time": "0:00:42", "remaining_time": "1:42:44"}
|
||||
{"current_steps": 10, "total_steps": 731, "loss": 2.548305892944336, "lr": 8.181818181818183e-06, "epoch": 0.013684570646595963, "percentage": 1.37, "elapsed_time": "0:01:26", "remaining_time": "1:44:20"}
|
||||
{"current_steps": 15, "total_steps": 731, "loss": 2.402619743347168, "lr": 1.2727272727272728e-05, "epoch": 0.020526855969893943, "percentage": 2.05, "elapsed_time": "0:02:08", "remaining_time": "1:42:27"}
|
||||
{"current_steps": 20, "total_steps": 731, "loss": 1.9256172180175781, "lr": 1.7272727272727274e-05, "epoch": 0.027369141293191925, "percentage": 2.74, "elapsed_time": "0:02:52", "remaining_time": "1:42:15"}
|
||||
{"current_steps": 25, "total_steps": 731, "loss": 2.0962652206420898, "lr": 1.9999607323936722e-05, "epoch": 0.03421142661648991, "percentage": 3.42, "elapsed_time": "0:03:33", "remaining_time": "1:40:22"}
|
||||
{"current_steps": 30, "total_steps": 731, "loss": 1.8469963073730469, "lr": 1.9995190072382676e-05, "epoch": 0.041053711939787886, "percentage": 4.1, "elapsed_time": "0:04:17", "remaining_time": "1:40:05"}
|
||||
{"current_steps": 35, "total_steps": 731, "loss": 1.8669818878173827, "lr": 1.9985866899535076e-05, "epoch": 0.04789599726308587, "percentage": 4.79, "elapsed_time": "0:04:59", "remaining_time": "1:39:07"}
|
||||
{"current_steps": 40, "total_steps": 731, "loss": 1.7637271881103516, "lr": 1.9971642381470207e-05, "epoch": 0.05473828258638385, "percentage": 5.47, "elapsed_time": "0:05:40", "remaining_time": "1:38:08"}
|
||||
{"current_steps": 45, "total_steps": 731, "loss": 1.705878448486328, "lr": 1.9952523499982866e-05, "epoch": 0.06158056790968183, "percentage": 6.16, "elapsed_time": "0:06:24", "remaining_time": "1:37:41"}
|
||||
{"current_steps": 50, "total_steps": 731, "loss": 1.6680316925048828, "lr": 1.9928519639159508e-05, "epoch": 0.06842285323297982, "percentage": 6.84, "elapsed_time": "0:07:07", "remaining_time": "1:37:06"}
|
||||
{"current_steps": 55, "total_steps": 731, "loss": 1.7056447982788085, "lr": 1.9899642580772274e-05, "epoch": 0.0752651385562778, "percentage": 7.52, "elapsed_time": "0:07:50", "remaining_time": "1:36:21"}
|
||||
{"current_steps": 60, "total_steps": 731, "loss": 1.636207389831543, "lr": 1.9865906498496164e-05, "epoch": 0.08210742387957577, "percentage": 8.21, "elapsed_time": "0:08:31", "remaining_time": "1:35:17"}
|
||||
{"current_steps": 65, "total_steps": 731, "loss": 1.623453712463379, "lr": 1.9827327950952197e-05, "epoch": 0.08894970920287376, "percentage": 8.89, "elapsed_time": "0:09:14", "remaining_time": "1:34:37"}
|
||||
{"current_steps": 70, "total_steps": 731, "loss": 1.6949512481689453, "lr": 1.9783925873579967e-05, "epoch": 0.09579199452617174, "percentage": 9.58, "elapsed_time": "0:09:55", "remaining_time": "1:33:39"}
|
||||
{"current_steps": 75, "total_steps": 731, "loss": 1.5341100692749023, "lr": 1.97357215693436e-05, "epoch": 0.10263427984946973, "percentage": 10.26, "elapsed_time": "0:10:39", "remaining_time": "1:33:09"}
|
||||
{"current_steps": 80, "total_steps": 731, "loss": 1.606492805480957, "lr": 1.9682738698275663e-05, "epoch": 0.1094765651727677, "percentage": 10.94, "elapsed_time": "0:11:21", "remaining_time": "1:32:24"}
|
||||
{"current_steps": 85, "total_steps": 731, "loss": 1.4498618125915528, "lr": 1.9625003265864148e-05, "epoch": 0.11631885049606569, "percentage": 11.63, "elapsed_time": "0:12:05", "remaining_time": "1:31:55"}
|
||||
{"current_steps": 90, "total_steps": 731, "loss": 1.5422866821289063, "lr": 1.956254361028823e-05, "epoch": 0.12316113581936367, "percentage": 12.31, "elapsed_time": "0:12:48", "remaining_time": "1:31:14"}
|
||||
{"current_steps": 95, "total_steps": 731, "loss": 1.5427611351013184, "lr": 1.9495390388509123e-05, "epoch": 0.13000342114266164, "percentage": 13.0, "elapsed_time": "0:13:31", "remaining_time": "1:30:32"}
|
||||
{"current_steps": 100, "total_steps": 731, "loss": 1.564542007446289, "lr": 1.9423576561222742e-05, "epoch": 0.13684570646595964, "percentage": 13.68, "elapsed_time": "0:14:16", "remaining_time": "1:30:02"}
|
||||
{"current_steps": 105, "total_steps": 731, "loss": 1.4948160171508789, "lr": 1.9347137376681692e-05, "epoch": 0.14368799178925762, "percentage": 14.36, "elapsed_time": "0:14:58", "remaining_time": "1:29:19"}
|
||||
{"current_steps": 110, "total_steps": 731, "loss": 1.539006805419922, "lr": 1.926611035339441e-05, "epoch": 0.1505302771125556, "percentage": 15.05, "elapsed_time": "0:15:41", "remaining_time": "1:28:36"}
|
||||
{"current_steps": 115, "total_steps": 731, "loss": 1.542184543609619, "lr": 1.918053526171004e-05, "epoch": 0.15737256243585357, "percentage": 15.73, "elapsed_time": "0:16:23", "remaining_time": "1:27:47"}
|
||||
{"current_steps": 120, "total_steps": 731, "loss": 1.6334993362426757, "lr": 1.9090454104298003e-05, "epoch": 0.16421484775915154, "percentage": 16.42, "elapsed_time": "0:17:06", "remaining_time": "1:27:04"}
|
||||
{"current_steps": 125, "total_steps": 731, "loss": 1.4427576065063477, "lr": 1.8995911095531932e-05, "epoch": 0.17105713308244955, "percentage": 17.1, "elapsed_time": "0:17:49", "remaining_time": "1:26:24"}
|
||||
{"current_steps": 130, "total_steps": 731, "loss": 1.5040811538696288, "lr": 1.8896952639787977e-05, "epoch": 0.17789941840574752, "percentage": 17.78, "elapsed_time": "0:18:32", "remaining_time": "1:25:44"}
|
||||
{"current_steps": 135, "total_steps": 731, "loss": 1.5777132034301757, "lr": 1.879362730866825e-05, "epoch": 0.1847417037290455, "percentage": 18.47, "elapsed_time": "0:19:14", "remaining_time": "1:24:56"}
|
||||
{"current_steps": 140, "total_steps": 731, "loss": 1.5796485900878907, "lr": 1.8685985817160503e-05, "epoch": 0.19158398905234347, "percentage": 19.15, "elapsed_time": "0:19:56", "remaining_time": "1:24:09"}
|
||||
{"current_steps": 145, "total_steps": 731, "loss": 1.5201278686523438, "lr": 1.8574080998745785e-05, "epoch": 0.19842627437564148, "percentage": 19.84, "elapsed_time": "0:20:38", "remaining_time": "1:23:26"}
|
||||
{"current_steps": 150, "total_steps": 731, "loss": 1.5156174659729005, "lr": 1.845796777946627e-05, "epoch": 0.20526855969893945, "percentage": 20.52, "elapsed_time": "0:21:22", "remaining_time": "1:22:48"}
|
||||
{"current_steps": 155, "total_steps": 731, "loss": 1.494316291809082, "lr": 1.8337703150966025e-05, "epoch": 0.21211084502223743, "percentage": 21.2, "elapsed_time": "0:22:05", "remaining_time": "1:22:04"}
|
||||
{"current_steps": 160, "total_steps": 731, "loss": 1.4805891036987304, "lr": 1.8213346142517884e-05, "epoch": 0.2189531303455354, "percentage": 21.89, "elapsed_time": "0:22:48", "remaining_time": "1:21:25"}
|
||||
{"current_steps": 165, "total_steps": 731, "loss": 1.5021228790283203, "lr": 1.8084957792050226e-05, "epoch": 0.22579541566883338, "percentage": 22.57, "elapsed_time": "0:23:27", "remaining_time": "1:20:27"}
|
||||
{"current_steps": 170, "total_steps": 731, "loss": 1.4761942863464355, "lr": 1.7952601116187822e-05, "epoch": 0.23263770099213138, "percentage": 23.26, "elapsed_time": "0:24:11", "remaining_time": "1:19:50"}
|
||||
{"current_steps": 175, "total_steps": 731, "loss": 1.5559394836425782, "lr": 1.781634107932153e-05, "epoch": 0.23947998631542936, "percentage": 23.94, "elapsed_time": "0:24:52", "remaining_time": "1:19:01"}
|
||||
{"current_steps": 180, "total_steps": 731, "loss": 1.4734925270080566, "lr": 1.7676244561721905e-05, "epoch": 0.24632227163872733, "percentage": 24.62, "elapsed_time": "0:25:35", "remaining_time": "1:18:21"}
|
||||
{"current_steps": 185, "total_steps": 731, "loss": 1.4294141769409179, "lr": 1.7532380326712487e-05, "epoch": 0.25316455696202533, "percentage": 25.31, "elapsed_time": "0:26:19", "remaining_time": "1:17:43"}
|
||||
{"current_steps": 190, "total_steps": 731, "loss": 1.4497101783752442, "lr": 1.738481898691884e-05, "epoch": 0.2600068422853233, "percentage": 25.99, "elapsed_time": "0:27:01", "remaining_time": "1:16:56"}
|
||||
{"current_steps": 195, "total_steps": 731, "loss": 1.4012206077575684, "lr": 1.7233632969609844e-05, "epoch": 0.2668491276086213, "percentage": 26.68, "elapsed_time": "0:27:46", "remaining_time": "1:16:20"}
|
||||
{"current_steps": 200, "total_steps": 731, "loss": 1.4317466735839843, "lr": 1.707889648114839e-05, "epoch": 0.2736914129319193, "percentage": 27.36, "elapsed_time": "0:28:29", "remaining_time": "1:15:38"}
|
||||
{"current_steps": 205, "total_steps": 731, "loss": 1.4276908874511718, "lr": 1.6920685470568775e-05, "epoch": 0.28053369825521723, "percentage": 28.04, "elapsed_time": "0:29:12", "remaining_time": "1:14:55"}
|
||||
{"current_steps": 210, "total_steps": 731, "loss": 1.4698185920715332, "lr": 1.675907759229879e-05, "epoch": 0.28737598357851524, "percentage": 28.73, "elapsed_time": "0:29:56", "remaining_time": "1:14:16"}
|
||||
{"current_steps": 215, "total_steps": 731, "loss": 1.389587688446045, "lr": 1.659415216804471e-05, "epoch": 0.2942182689018132, "percentage": 29.41, "elapsed_time": "0:30:40", "remaining_time": "1:13:37"}
|
||||
{"current_steps": 220, "total_steps": 731, "loss": 1.4244729042053224, "lr": 1.6425990147858004e-05, "epoch": 0.3010605542251112, "percentage": 30.1, "elapsed_time": "0:31:24", "remaining_time": "1:12:56"}
|
||||
{"current_steps": 225, "total_steps": 731, "loss": 1.423227310180664, "lr": 1.625467407040273e-05, "epoch": 0.3079028395484092, "percentage": 30.78, "elapsed_time": "0:32:06", "remaining_time": "1:12:13"}
|
||||
{"current_steps": 230, "total_steps": 731, "loss": 1.4495688438415528, "lr": 1.6080288022443242e-05, "epoch": 0.31474512487170714, "percentage": 31.46, "elapsed_time": "0:32:51", "remaining_time": "1:11:34"}
|
||||
{"current_steps": 235, "total_steps": 731, "loss": 1.4790194511413575, "lr": 1.590291759757204e-05, "epoch": 0.32158741019500514, "percentage": 32.15, "elapsed_time": "0:33:32", "remaining_time": "1:10:47"}
|
||||
{"current_steps": 240, "total_steps": 731, "loss": 1.3981167793273925, "lr": 1.5722649854198007e-05, "epoch": 0.3284296955183031, "percentage": 32.83, "elapsed_time": "0:34:14", "remaining_time": "1:10:02"}
|
||||
{"current_steps": 245, "total_steps": 731, "loss": 1.3656198501586914, "lr": 1.5539573272815698e-05, "epoch": 0.3352719808416011, "percentage": 33.52, "elapsed_time": "0:34:58", "remaining_time": "1:09:23"}
|
||||
{"current_steps": 250, "total_steps": 731, "loss": 1.4632407188415528, "lr": 1.5353777712576613e-05, "epoch": 0.3421142661648991, "percentage": 34.2, "elapsed_time": "0:35:39", "remaining_time": "1:08:36"}
|
||||
{"current_steps": 255, "total_steps": 731, "loss": 1.5673239707946778, "lr": 1.5165354367183778e-05, "epoch": 0.34895655148819704, "percentage": 34.88, "elapsed_time": "0:36:21", "remaining_time": "1:07:52"}
|
||||
{"current_steps": 260, "total_steps": 731, "loss": 1.4822378158569336, "lr": 1.4974395720131328e-05, "epoch": 0.35579883681149505, "percentage": 35.57, "elapsed_time": "0:37:04", "remaining_time": "1:07:09"}
|
||||
{"current_steps": 265, "total_steps": 731, "loss": 1.4012256622314454, "lr": 1.4780995499310973e-05, "epoch": 0.36264112213479305, "percentage": 36.25, "elapsed_time": "0:37:45", "remaining_time": "1:06:23"}
|
||||
{"current_steps": 270, "total_steps": 731, "loss": 1.3955447196960449, "lr": 1.458524863100772e-05, "epoch": 0.369483407458091, "percentage": 36.94, "elapsed_time": "0:38:29", "remaining_time": "1:05:43"}
|
||||
{"current_steps": 275, "total_steps": 731, "loss": 1.4113948822021485, "lr": 1.4387251193307369e-05, "epoch": 0.376325692781389, "percentage": 37.62, "elapsed_time": "0:39:13", "remaining_time": "1:05:02"}
|
||||
{"current_steps": 280, "total_steps": 731, "loss": 1.4017261505126952, "lr": 1.4187100368938678e-05, "epoch": 0.38316797810468695, "percentage": 38.3, "elapsed_time": "0:39:56", "remaining_time": "1:04:20"}
|
||||
{"current_steps": 285, "total_steps": 731, "loss": 1.5206121444702148, "lr": 1.3984894397573332e-05, "epoch": 0.39001026342798495, "percentage": 38.99, "elapsed_time": "0:40:38", "remaining_time": "1:03:35"}
|
||||
{"current_steps": 290, "total_steps": 731, "loss": 1.322197723388672, "lr": 1.3780732527607157e-05, "epoch": 0.39685254875128295, "percentage": 39.67, "elapsed_time": "0:41:22", "remaining_time": "1:02:55"}
|
||||
{"current_steps": 295, "total_steps": 731, "loss": 1.3588292121887207, "lr": 1.3574714967446193e-05, "epoch": 0.4036948340745809, "percentage": 40.36, "elapsed_time": "0:42:07", "remaining_time": "1:02:15"}
|
||||
{"current_steps": 300, "total_steps": 731, "loss": 1.3987003326416017, "lr": 1.3366942836321574e-05, "epoch": 0.4105371193978789, "percentage": 41.04, "elapsed_time": "0:42:50", "remaining_time": "1:01:32"}
|
||||
{"current_steps": 305, "total_steps": 731, "loss": 1.3717717170715331, "lr": 1.3157518114657382e-05, "epoch": 0.41737940472117685, "percentage": 41.72, "elapsed_time": "0:43:34", "remaining_time": "1:00:51"}
|
||||
{"current_steps": 310, "total_steps": 731, "loss": 1.4352380752563476, "lr": 1.2946543594015752e-05, "epoch": 0.42422169004447485, "percentage": 42.41, "elapsed_time": "0:44:19", "remaining_time": "1:00:11"}
|
||||
{"current_steps": 315, "total_steps": 731, "loss": 1.46458158493042, "lr": 1.2734122826643884e-05, "epoch": 0.43106397536777286, "percentage": 43.09, "elapsed_time": "0:45:03", "remaining_time": "0:59:30"}
|
||||
{"current_steps": 320, "total_steps": 731, "loss": 1.4723733901977538, "lr": 1.2520360074647686e-05, "epoch": 0.4379062606910708, "percentage": 43.78, "elapsed_time": "0:45:44", "remaining_time": "0:58:45"}
|
||||
{"current_steps": 325, "total_steps": 731, "loss": 1.448530387878418, "lr": 1.2305360258816976e-05, "epoch": 0.4447485460143688, "percentage": 44.46, "elapsed_time": "0:46:27", "remaining_time": "0:58:02"}
|
||||
{"current_steps": 330, "total_steps": 731, "loss": 1.4141918182373048, "lr": 1.2089228907127403e-05, "epoch": 0.45159083133766675, "percentage": 45.14, "elapsed_time": "0:47:11", "remaining_time": "0:57:20"}
|
||||
{"current_steps": 335, "total_steps": 731, "loss": 1.4687261581420898, "lr": 1.1872072102944345e-05, "epoch": 0.45843311666096476, "percentage": 45.83, "elapsed_time": "0:47:53", "remaining_time": "0:56:36"}
|
||||
{"current_steps": 340, "total_steps": 731, "loss": 1.381917667388916, "lr": 1.1653996432954194e-05, "epoch": 0.46527540198426276, "percentage": 46.51, "elapsed_time": "0:48:37", "remaining_time": "0:55:55"}
|
||||
{"current_steps": 345, "total_steps": 731, "loss": 1.398417091369629, "lr": 1.1435108934848609e-05, "epoch": 0.4721176873075607, "percentage": 47.2, "elapsed_time": "0:49:20", "remaining_time": "0:55:12"}
|
||||
{"current_steps": 350, "total_steps": 731, "loss": 1.3287519454956054, "lr": 1.1215517044787387e-05, "epoch": 0.4789599726308587, "percentage": 47.88, "elapsed_time": "0:50:03", "remaining_time": "0:54:29"}
|
||||
{"current_steps": 355, "total_steps": 731, "loss": 1.4714726448059081, "lr": 1.0995328544665776e-05, "epoch": 0.4858022579541567, "percentage": 48.56, "elapsed_time": "0:50:45", "remaining_time": "0:53:46"}
|
||||
{"current_steps": 360, "total_steps": 731, "loss": 1.431112003326416, "lr": 1.077465150921207e-05, "epoch": 0.49264454327745466, "percentage": 49.25, "elapsed_time": "0:51:28", "remaining_time": "0:53:03"}
|
||||
{"current_steps": 365, "total_steps": 731, "loss": 1.4446621894836427, "lr": 1.0553594252941489e-05, "epoch": 0.49948682860075266, "percentage": 49.93, "elapsed_time": "0:52:09", "remaining_time": "0:52:18"}
|
||||
{"current_steps": 370, "total_steps": 731, "loss": 1.3496983528137207, "lr": 1.0332265276992361e-05, "epoch": 0.5063291139240507, "percentage": 50.62, "elapsed_time": "0:52:53", "remaining_time": "0:51:36"}
|
||||
{"current_steps": 375, "total_steps": 731, "loss": 1.324970531463623, "lr": 1.0110773215870695e-05, "epoch": 0.5131713992473487, "percentage": 51.3, "elapsed_time": "0:53:37", "remaining_time": "0:50:54"}
|
||||
{"current_steps": 380, "total_steps": 731, "loss": 1.3443440437316894, "lr": 9.889226784129307e-06, "epoch": 0.5200136845706466, "percentage": 51.98, "elapsed_time": "0:54:20", "remaining_time": "0:50:12"}
|
||||
{"current_steps": 385, "total_steps": 731, "loss": 1.4248334884643554, "lr": 9.66773472300764e-06, "epoch": 0.5268559698939446, "percentage": 52.67, "elapsed_time": "0:55:03", "remaining_time": "0:49:29"}
|
||||
{"current_steps": 390, "total_steps": 731, "loss": 1.416274070739746, "lr": 9.446405747058513e-06, "epoch": 0.5336982552172426, "percentage": 53.35, "elapsed_time": "0:55:46", "remaining_time": "0:48:45"}
|
||||
{"current_steps": 395, "total_steps": 731, "loss": 1.4464293479919434, "lr": 9.225348490787935e-06, "epoch": 0.5405405405405406, "percentage": 54.04, "elapsed_time": "0:56:29", "remaining_time": "0:48:03"}
|
||||
{"current_steps": 400, "total_steps": 731, "loss": 1.3776466369628906, "lr": 9.004671455334227e-06, "epoch": 0.5473828258638386, "percentage": 54.72, "elapsed_time": "0:57:13", "remaining_time": "0:47:21"}
|
||||
{"current_steps": 405, "total_steps": 731, "loss": 1.3454496383666992, "lr": 8.784482955212615e-06, "epoch": 0.5542251111871365, "percentage": 55.4, "elapsed_time": "0:57:57", "remaining_time": "0:46:38"}
|
||||
{"current_steps": 410, "total_steps": 731, "loss": 1.3769558906555175, "lr": 8.564891065151391e-06, "epoch": 0.5610673965104345, "percentage": 56.09, "elapsed_time": "0:58:39", "remaining_time": "0:45:55"}
|
||||
{"current_steps": 415, "total_steps": 731, "loss": 1.4825267791748047, "lr": 8.346003567045806e-06, "epoch": 0.5679096818337325, "percentage": 56.77, "elapsed_time": "0:59:22", "remaining_time": "0:45:12"}
|
||||
{"current_steps": 420, "total_steps": 731, "loss": 1.3498231887817382, "lr": 8.127927897055658e-06, "epoch": 0.5747519671570305, "percentage": 57.46, "elapsed_time": "1:00:06", "remaining_time": "0:44:30"}
|
||||
{"current_steps": 425, "total_steps": 731, "loss": 1.4994674682617188, "lr": 7.910771092872598e-06, "epoch": 0.5815942524803285, "percentage": 58.14, "elapsed_time": "1:00:50", "remaining_time": "0:43:48"}
|
||||
{"current_steps": 430, "total_steps": 731, "loss": 1.406801700592041, "lr": 7.694639741183027e-06, "epoch": 0.5884365378036264, "percentage": 58.82, "elapsed_time": "1:01:33", "remaining_time": "0:43:05"}
|
||||
{"current_steps": 435, "total_steps": 731, "loss": 1.57269344329834, "lr": 7.479639925352318e-06, "epoch": 0.5952788231269244, "percentage": 59.51, "elapsed_time": "1:02:14", "remaining_time": "0:42:21"}
|
||||
{"current_steps": 440, "total_steps": 731, "loss": 1.3171756744384766, "lr": 7.2658771733561175e-06, "epoch": 0.6021211084502224, "percentage": 60.19, "elapsed_time": "1:02:58", "remaining_time": "0:41:38"}
|
||||
{"current_steps": 445, "total_steps": 731, "loss": 1.4658644676208497, "lr": 7.053456405984254e-06, "epoch": 0.6089633937735204, "percentage": 60.88, "elapsed_time": "1:03:40", "remaining_time": "0:40:55"}
|
||||
{"current_steps": 450, "total_steps": 731, "loss": 1.4110689163208008, "lr": 6.842481885342624e-06, "epoch": 0.6158056790968184, "percentage": 61.56, "elapsed_time": "1:04:23", "remaining_time": "0:40:12"}
|
||||
{"current_steps": 455, "total_steps": 731, "loss": 1.3724404335021974, "lr": 6.633057163678427e-06, "epoch": 0.6226479644201163, "percentage": 62.24, "elapsed_time": "1:05:07", "remaining_time": "0:39:30"}
|
||||
{"current_steps": 460, "total_steps": 731, "loss": 1.3259340286254884, "lr": 6.42528503255381e-06, "epoch": 0.6294902497434143, "percentage": 62.93, "elapsed_time": "1:05:51", "remaining_time": "0:38:48"}
|
||||
{"current_steps": 465, "total_steps": 731, "loss": 1.4074788093566895, "lr": 6.2192674723928425e-06, "epoch": 0.6363325350667123, "percentage": 63.61, "elapsed_time": "1:06:34", "remaining_time": "0:38:05"}
|
||||
{"current_steps": 470, "total_steps": 731, "loss": 1.4261545181274413, "lr": 6.01510560242667e-06, "epoch": 0.6431748203900103, "percentage": 64.3, "elapsed_time": "1:07:17", "remaining_time": "0:37:22"}
|
||||
{"current_steps": 475, "total_steps": 731, "loss": 1.411759376525879, "lr": 5.812899631061328e-06, "epoch": 0.6500171057133083, "percentage": 64.98, "elapsed_time": "1:08:00", "remaining_time": "0:36:38"}
|
||||
{"current_steps": 480, "total_steps": 731, "loss": 1.4686535835266112, "lr": 5.612748806692631e-06, "epoch": 0.6568593910366062, "percentage": 65.66, "elapsed_time": "1:08:41", "remaining_time": "0:35:55"}
|
||||
{"current_steps": 485, "total_steps": 731, "loss": 1.3639310836791991, "lr": 5.414751368992281e-06, "epoch": 0.6637016763599042, "percentage": 66.35, "elapsed_time": "1:09:25", "remaining_time": "0:35:12"}
|
||||
{"current_steps": 490, "total_steps": 731, "loss": 1.370129108428955, "lr": 5.219004500689031e-06, "epoch": 0.6705439616832022, "percentage": 67.03, "elapsed_time": "1:10:09", "remaining_time": "0:34:30"}
|
||||
{"current_steps": 495, "total_steps": 731, "loss": 1.3223713874816894, "lr": 5.025604279868677e-06, "epoch": 0.6773862470065002, "percentage": 67.72, "elapsed_time": "1:10:52", "remaining_time": "0:33:47"}
|
||||
{"current_steps": 500, "total_steps": 731, "loss": 1.4085097312927246, "lr": 4.834645632816227e-06, "epoch": 0.6842285323297982, "percentage": 68.4, "elapsed_time": "1:11:34", "remaining_time": "0:33:04"}
|
||||
{"current_steps": 505, "total_steps": 731, "loss": 1.3346193313598633, "lr": 4.646222287423391e-06, "epoch": 0.6910708176530961, "percentage": 69.08, "elapsed_time": "1:12:20", "remaining_time": "0:32:22"}
|
||||
{"current_steps": 510, "total_steps": 731, "loss": 1.3574554443359375, "lr": 4.460426727184305e-06, "epoch": 0.6979131029763941, "percentage": 69.77, "elapsed_time": "1:13:04", "remaining_time": "0:31:40"}
|
||||
{"current_steps": 515, "total_steps": 731, "loss": 1.3625916481018066, "lr": 4.277350145801994e-06, "epoch": 0.7047553882996921, "percentage": 70.45, "elapsed_time": "1:13:47", "remaining_time": "0:30:56"}
|
||||
{"current_steps": 520, "total_steps": 731, "loss": 1.4155165672302246, "lr": 4.097082402427962e-06, "epoch": 0.7115976736229901, "percentage": 71.14, "elapsed_time": "1:14:30", "remaining_time": "0:30:14"}
|
||||
{"current_steps": 525, "total_steps": 731, "loss": 1.4300453186035156, "lr": 3.9197119775567595e-06, "epoch": 0.7184399589462881, "percentage": 71.82, "elapsed_time": "1:15:12", "remaining_time": "0:29:30"}
|
||||
{"current_steps": 530, "total_steps": 731, "loss": 1.369569969177246, "lr": 3.745325929597272e-06, "epoch": 0.7252822442695861, "percentage": 72.5, "elapsed_time": "1:15:57", "remaining_time": "0:28:48"}
|
||||
{"current_steps": 535, "total_steps": 731, "loss": 1.4022644996643066, "lr": 3.5740098521419985e-06, "epoch": 0.732124529592884, "percentage": 73.19, "elapsed_time": "1:16:40", "remaining_time": "0:28:05"}
|
||||
{"current_steps": 540, "total_steps": 731, "loss": 1.4104683876037598, "lr": 3.4058478319552935e-06, "epoch": 0.738966814916182, "percentage": 73.87, "elapsed_time": "1:17:23", "remaining_time": "0:27:22"}
|
||||
{"current_steps": 545, "total_steps": 731, "loss": 1.4028725624084473, "lr": 3.2409224077012135e-06, "epoch": 0.74580910023948, "percentage": 74.56, "elapsed_time": "1:18:07", "remaining_time": "0:26:39"}
|
||||
{"current_steps": 550, "total_steps": 731, "loss": 1.4290534019470216, "lr": 3.0793145294312253e-06, "epoch": 0.752651385562778, "percentage": 75.24, "elapsed_time": "1:18:49", "remaining_time": "0:25:56"}
|
||||
{"current_steps": 555, "total_steps": 731, "loss": 1.4063751220703125, "lr": 2.921103518851609e-06, "epoch": 0.759493670886076, "percentage": 75.92, "elapsed_time": "1:19:32", "remaining_time": "0:25:13"}
|
||||
{"current_steps": 560, "total_steps": 731, "loss": 1.3523286819458007, "lr": 2.766367030390157e-06, "epoch": 0.7663359562093739, "percentage": 76.61, "elapsed_time": "1:20:16", "remaining_time": "0:24:30"}
|
||||
{"current_steps": 565, "total_steps": 731, "loss": 1.397650909423828, "lr": 2.615181013081164e-06, "epoch": 0.7731782415326719, "percentage": 77.29, "elapsed_time": "1:20:58", "remaining_time": "0:23:47"}
|
||||
{"current_steps": 570, "total_steps": 731, "loss": 1.4020781517028809, "lr": 2.4676196732875147e-06, "epoch": 0.7800205268559699, "percentage": 77.98, "elapsed_time": "1:21:40", "remaining_time": "0:23:04"}
|
||||
{"current_steps": 575, "total_steps": 731, "loss": 1.4961904525756835, "lr": 2.3237554382781004e-06, "epoch": 0.7868628121792679, "percentage": 78.66, "elapsed_time": "1:22:23", "remaining_time": "0:22:21"}
|
||||
{"current_steps": 580, "total_steps": 731, "loss": 1.451511764526367, "lr": 2.1836589206784742e-06, "epoch": 0.7937050975025659, "percentage": 79.34, "elapsed_time": "1:23:05", "remaining_time": "0:21:38"}
|
||||
{"current_steps": 585, "total_steps": 731, "loss": 1.3788551330566405, "lr": 2.0473988838121783e-06, "epoch": 0.8005473828258638, "percentage": 80.03, "elapsed_time": "1:23:49", "remaining_time": "0:20:55"}
|
||||
{"current_steps": 590, "total_steps": 731, "loss": 1.410994529724121, "lr": 1.9150422079497787e-06, "epoch": 0.8073896681491618, "percentage": 80.71, "elapsed_time": "1:24:32", "remaining_time": "0:20:12"}
|
||||
{"current_steps": 595, "total_steps": 731, "loss": 1.3450016975402832, "lr": 1.786653857482118e-06, "epoch": 0.8142319534724598, "percentage": 81.4, "elapsed_time": "1:25:16", "remaining_time": "0:19:29"}
|
||||
{"current_steps": 600, "total_steps": 731, "loss": 1.4508660316467286, "lr": 1.6622968490339775e-06, "epoch": 0.8210742387957578, "percentage": 82.08, "elapsed_time": "1:25:57", "remaining_time": "0:18:46"}
|
||||
{"current_steps": 605, "total_steps": 731, "loss": 1.3347270965576172, "lr": 1.5420322205337335e-06, "epoch": 0.8279165241190558, "percentage": 82.76, "elapsed_time": "1:26:42", "remaining_time": "0:18:03"}
|
||||
{"current_steps": 610, "total_steps": 731, "loss": 1.2560535430908204, "lr": 1.425919001254219e-06, "epoch": 0.8347588094423537, "percentage": 83.45, "elapsed_time": "1:27:26", "remaining_time": "0:17:20"}
|
||||
{"current_steps": 615, "total_steps": 731, "loss": 1.355188751220703, "lr": 1.3140141828394993e-06, "epoch": 0.8416010947656517, "percentage": 84.13, "elapsed_time": "1:28:09", "remaining_time": "0:16:37"}
|
||||
{"current_steps": 620, "total_steps": 731, "loss": 1.397140884399414, "lr": 1.2063726913317508e-06, "epoch": 0.8484433800889497, "percentage": 84.82, "elapsed_time": "1:28:51", "remaining_time": "0:15:54"}
|
||||
{"current_steps": 625, "total_steps": 731, "loss": 1.3479557037353516, "lr": 1.103047360212024e-06, "epoch": 0.8552856654122477, "percentage": 85.5, "elapsed_time": "1:29:36", "remaining_time": "0:15:11"}
|
||||
{"current_steps": 630, "total_steps": 731, "loss": 1.3707120895385743, "lr": 1.0040889044680702e-06, "epoch": 0.8621279507355457, "percentage": 86.18, "elapsed_time": "1:30:20", "remaining_time": "0:14:29"}
|
||||
{"current_steps": 635, "total_steps": 731, "loss": 1.3732416152954101, "lr": 9.095458957019986e-07, "epoch": 0.8689702360588436, "percentage": 86.87, "elapsed_time": "1:31:02", "remaining_time": "0:13:45"}
|
||||
{"current_steps": 640, "total_steps": 731, "loss": 1.3615639686584473, "lr": 8.194647382899657e-07, "epoch": 0.8758125213821416, "percentage": 87.55, "elapsed_time": "1:31:47", "remaining_time": "0:13:03"}
|
||||
{"current_steps": 645, "total_steps": 731, "loss": 1.4400001525878907, "lr": 7.338896466055934e-07, "epoch": 0.8826548067054396, "percentage": 88.24, "elapsed_time": "1:32:28", "remaining_time": "0:12:19"}
|
||||
{"current_steps": 650, "total_steps": 731, "loss": 1.3915739059448242, "lr": 6.52862623318311e-07, "epoch": 0.8894970920287376, "percentage": 88.92, "elapsed_time": "1:33:11", "remaining_time": "0:11:36"}
|
||||
{"current_steps": 655, "total_steps": 731, "loss": 1.3268877029418946, "lr": 5.764234387772593e-07, "epoch": 0.8963393773520356, "percentage": 89.6, "elapsed_time": "1:33:56", "remaining_time": "0:10:54"}
|
||||
{"current_steps": 660, "total_steps": 731, "loss": 1.3429359436035155, "lr": 5.046096114908783e-07, "epoch": 0.9031816626753335, "percentage": 90.29, "elapsed_time": "1:34:40", "remaining_time": "0:10:11"}
|
||||
{"current_steps": 665, "total_steps": 731, "loss": 1.4380492210388183, "lr": 4.374563897117701e-07, "epoch": 0.9100239479986315, "percentage": 90.97, "elapsed_time": "1:35:22", "remaining_time": "0:09:27"}
|
||||
{"current_steps": 670, "total_steps": 731, "loss": 1.3397421836853027, "lr": 3.7499673413585516e-07, "epoch": 0.9168662333219295, "percentage": 91.66, "elapsed_time": "1:36:07", "remaining_time": "0:08:45"}
|
||||
{"current_steps": 675, "total_steps": 731, "loss": 1.4501882553100587, "lr": 3.1726130172433756e-07, "epoch": 0.9237085186452275, "percentage": 92.34, "elapsed_time": "1:36:49", "remaining_time": "0:08:01"}
|
||||
{"current_steps": 680, "total_steps": 731, "loss": 1.3810162544250488, "lr": 2.642784306564017e-07, "epoch": 0.9305508039685255, "percentage": 93.02, "elapsed_time": "1:37:31", "remaining_time": "0:07:18"}
|
||||
{"current_steps": 685, "total_steps": 731, "loss": 1.3060113906860351, "lr": 2.160741264200361e-07, "epoch": 0.9373930892918235, "percentage": 93.71, "elapsed_time": "1:38:15", "remaining_time": "0:06:35"}
|
||||
{"current_steps": 690, "total_steps": 731, "loss": 1.4677888870239257, "lr": 1.7267204904780621e-07, "epoch": 0.9442353746151214, "percentage": 94.39, "elapsed_time": "1:38:57", "remaining_time": "0:05:52"}
|
||||
{"current_steps": 695, "total_steps": 731, "loss": 1.3917272567749024, "lr": 1.3409350150383804e-07, "epoch": 0.9510776599384194, "percentage": 95.08, "elapsed_time": "1:39:41", "remaining_time": "0:05:09"}
|
||||
{"current_steps": 700, "total_steps": 731, "loss": 1.4361152648925781, "lr": 1.0035741922772902e-07, "epoch": 0.9579199452617174, "percentage": 95.76, "elapsed_time": "1:40:22", "remaining_time": "0:04:26"}
|
||||
{"current_steps": 705, "total_steps": 731, "loss": 1.452861213684082, "lr": 7.14803608404957e-08, "epoch": 0.9647622305850154, "percentage": 96.44, "elapsed_time": "1:41:03", "remaining_time": "0:03:43"}
|
||||
{"current_steps": 710, "total_steps": 731, "loss": 1.3411725044250489, "lr": 4.747650001713622e-08, "epoch": 0.9716045159083134, "percentage": 97.13, "elapsed_time": "1:41:46", "remaining_time": "0:03:00"}
|
||||
{"current_steps": 715, "total_steps": 731, "loss": 1.3774123191833496, "lr": 2.835761852979291e-08, "epoch": 0.9784468012316113, "percentage": 97.81, "elapsed_time": "1:42:28", "remaining_time": "0:02:17"}
|
||||
{"current_steps": 720, "total_steps": 731, "loss": 1.3300366401672363, "lr": 1.4133100464922556e-08, "epoch": 0.9852890865549093, "percentage": 98.5, "elapsed_time": "1:43:12", "remaining_time": "0:01:34"}
|
||||
{"current_steps": 725, "total_steps": 731, "loss": 1.295484733581543, "lr": 4.809927617324128e-09, "epoch": 0.9921313718782073, "percentage": 99.18, "elapsed_time": "1:43:57", "remaining_time": "0:00:51"}
|
||||
{"current_steps": 730, "total_steps": 731, "loss": 1.3570124626159668, "lr": 3.9267606327775266e-10, "epoch": 0.9989736572015053, "percentage": 99.86, "elapsed_time": "1:44:41", "remaining_time": "0:00:08"}
|
||||
{"current_steps": 731, "total_steps": 731, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "1:44:50", "remaining_time": "0:00:00"}
|
||||
1065
trainer_state.json
Normal file
1065
trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
training_args.bin
Normal file
3
training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5ab633c6efd46966ccbc35b6bb7056f8e5aac8c0cae73a7c8386ea42e9dadfa6
|
||||
size 6033
|
||||
BIN
training_loss.png
Normal file
BIN
training_loss.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 47 KiB |
Reference in New Issue
Block a user