初始化项目，由ModelHub XC社区提供模型
Model: KKHYA/qwen3-1.7b-fft-math 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
|
||||
61
README.md
Normal file
61
README.md
Normal file
@@ -0,0 +1,61 @@
|
||||
---
|
||||
library_name: transformers
|
||||
license: apache-2.0
|
||||
base_model: Qwen/Qwen3-1.7B
|
||||
tags:
|
||||
- llama-factory
|
||||
- full
|
||||
- generated_from_trainer
|
||||
model-index:
|
||||
- name: qwen3-1.7b-fft-math
|
||||
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. -->
|
||||
|
||||
# qwen3-1.7b-fft-math
|
||||
|
||||
This model is a fine-tuned version of [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) on the mft_metamath, the mft_numinamath_tir, the mft_numinamath_cot, the mft_tulu3_personas_math, the mft_tulu3_personas_math_grade and the mft_tulu3_personas_algebra datasets.
|
||||
|
||||
## 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: 1e-05
|
||||
- train_batch_size: 2
|
||||
- eval_batch_size: 8
|
||||
- seed: 42
|
||||
- distributed_type: multi-GPU
|
||||
- num_devices: 8
|
||||
- gradient_accumulation_steps: 8
|
||||
- total_train_batch_size: 128
|
||||
- total_eval_batch_size: 64
|
||||
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
||||
- lr_scheduler_type: cosine
|
||||
- lr_scheduler_warmup_steps: 0.1
|
||||
- num_epochs: 2.0
|
||||
|
||||
### Training results
|
||||
|
||||
|
||||
|
||||
### Framework versions
|
||||
|
||||
- Transformers 5.2.0
|
||||
- Pytorch 2.10.0+cu130
|
||||
- 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": 2.0,
|
||||
"total_flos": 1.2200945171646382e+18,
|
||||
"train_loss": 0.2963919332032519,
|
||||
"train_runtime": 3915.3376,
|
||||
"train_samples_per_second": 30.649,
|
||||
"train_steps_per_second": 0.24
|
||||
}
|
||||
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 %}
|
||||
70
config.json
Normal file
70
config.json
Normal file
@@ -0,0 +1,70 @@
|
||||
{
|
||||
"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": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 6144,
|
||||
"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"
|
||||
],
|
||||
"masked_layers": null,
|
||||
"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,
|
||||
"sparsity_attn": null,
|
||||
"sparsity_mlp": null,
|
||||
"subnet_mode": null,
|
||||
"subnet_type": null,
|
||||
"threshold_attn": null,
|
||||
"threshold_mlp": null,
|
||||
"tie_word_embeddings": true,
|
||||
"transformers_version": "5.2.0",
|
||||
"use_cache": false,
|
||||
"use_sliding_window": false,
|
||||
"vocab_size": 151936
|
||||
}
|
||||
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": "5.2.0"
|
||||
}
|
||||
3
model.safetensors
Normal file
3
model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:969ad94985c2c70a61c87a23ebf9c7028163014421cf750139e9a5197f25c82f
|
||||
size 4063515640
|
||||
2843
qwen3_1.7b_math_fft_20260429_154203.log
Normal file
2843
qwen3_1.7b_math_fft_20260429_154203.log
Normal file
File diff suppressed because one or more lines are too long
3
tokenizer.json
Normal file
3
tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||
size 11422650
|
||||
30
tokenizer_config.json
Normal file
30
tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
||||
{
|
||||
"add_prefix_space": false,
|
||||
"backend": "tokenizers",
|
||||
"bos_token": null,
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "<|im_end|>",
|
||||
"errors": "replace",
|
||||
"extra_special_tokens": [
|
||||
"<|im_start|>",
|
||||
"<|im_end|>",
|
||||
"<|object_ref_start|>",
|
||||
"<|object_ref_end|>",
|
||||
"<|box_start|>",
|
||||
"<|box_end|>",
|
||||
"<|quad_start|>",
|
||||
"<|quad_end|>",
|
||||
"<|vision_start|>",
|
||||
"<|vision_end|>",
|
||||
"<|vision_pad|>",
|
||||
"<|image_pad|>",
|
||||
"<|video_pad|>"
|
||||
],
|
||||
"is_local": 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": 2.0,
|
||||
"total_flos": 1.2200945171646382e+18,
|
||||
"train_loss": 0.2963919332032519,
|
||||
"train_runtime": 3915.3376,
|
||||
"train_samples_per_second": 30.649,
|
||||
"train_steps_per_second": 0.24
|
||||
}
|
||||
94
trainer_log.jsonl
Normal file
94
trainer_log.jsonl
Normal file
@@ -0,0 +1,94 @@
|
||||
{"current_steps": 10, "total_steps": 938, "loss": 0.6841894149780273, "lr": 9.574468085106384e-07, "epoch": 0.021333333333333333, "percentage": 1.07, "elapsed_time": "0:00:43", "remaining_time": "1:07:39"}
|
||||
{"current_steps": 20, "total_steps": 938, "loss": 0.6096980094909668, "lr": 2.021276595744681e-06, "epoch": 0.042666666666666665, "percentage": 2.13, "elapsed_time": "0:01:23", "remaining_time": "1:04:01"}
|
||||
{"current_steps": 30, "total_steps": 938, "loss": 0.46523351669311525, "lr": 3.0851063829787237e-06, "epoch": 0.064, "percentage": 3.2, "elapsed_time": "0:02:03", "remaining_time": "1:02:05"}
|
||||
{"current_steps": 40, "total_steps": 938, "loss": 0.38690838813781736, "lr": 4.148936170212766e-06, "epoch": 0.08533333333333333, "percentage": 4.26, "elapsed_time": "0:02:43", "remaining_time": "1:01:20"}
|
||||
{"current_steps": 50, "total_steps": 938, "loss": 0.3463115692138672, "lr": 5.212765957446809e-06, "epoch": 0.10666666666666667, "percentage": 5.33, "elapsed_time": "0:03:26", "remaining_time": "1:01:14"}
|
||||
{"current_steps": 60, "total_steps": 938, "loss": 0.3305189609527588, "lr": 6.276595744680851e-06, "epoch": 0.128, "percentage": 6.4, "elapsed_time": "0:04:07", "remaining_time": "1:00:18"}
|
||||
{"current_steps": 70, "total_steps": 938, "loss": 0.33223772048950195, "lr": 7.340425531914894e-06, "epoch": 0.14933333333333335, "percentage": 7.46, "elapsed_time": "0:04:49", "remaining_time": "0:59:44"}
|
||||
{"current_steps": 80, "total_steps": 938, "loss": 0.31659588813781736, "lr": 8.404255319148937e-06, "epoch": 0.17066666666666666, "percentage": 8.53, "elapsed_time": "0:05:31", "remaining_time": "0:59:11"}
|
||||
{"current_steps": 90, "total_steps": 938, "loss": 0.31022181510925295, "lr": 9.46808510638298e-06, "epoch": 0.192, "percentage": 9.59, "elapsed_time": "0:06:11", "remaining_time": "0:58:21"}
|
||||
{"current_steps": 100, "total_steps": 938, "loss": 0.3155974388122559, "lr": 9.999134070902206e-06, "epoch": 0.21333333333333335, "percentage": 10.66, "elapsed_time": "0:06:52", "remaining_time": "0:57:34"}
|
||||
{"current_steps": 110, "total_steps": 938, "loss": 0.31209163665771483, "lr": 9.992208437615651e-06, "epoch": 0.23466666666666666, "percentage": 11.73, "elapsed_time": "0:07:33", "remaining_time": "0:56:57"}
|
||||
{"current_steps": 120, "total_steps": 938, "loss": 0.30946059226989747, "lr": 9.97836676558346e-06, "epoch": 0.256, "percentage": 12.79, "elapsed_time": "0:08:14", "remaining_time": "0:56:12"}
|
||||
{"current_steps": 130, "total_steps": 938, "loss": 0.3128779888153076, "lr": 9.957628230595527e-06, "epoch": 0.2773333333333333, "percentage": 13.86, "elapsed_time": "0:08:54", "remaining_time": "0:55:20"}
|
||||
{"current_steps": 140, "total_steps": 938, "loss": 0.3062174320220947, "lr": 9.930021563125204e-06, "epoch": 0.2986666666666667, "percentage": 14.93, "elapsed_time": "0:09:35", "remaining_time": "0:54:39"}
|
||||
{"current_steps": 150, "total_steps": 938, "loss": 0.30686254501342775, "lr": 9.895585008527075e-06, "epoch": 0.32, "percentage": 15.99, "elapsed_time": "0:10:15", "remaining_time": "0:53:53"}
|
||||
{"current_steps": 160, "total_steps": 938, "loss": 0.3146007299423218, "lr": 9.854366274053125e-06, "epoch": 0.3413333333333333, "percentage": 17.06, "elapsed_time": "0:10:56", "remaining_time": "0:53:11"}
|
||||
{"current_steps": 170, "total_steps": 938, "loss": 0.30956292152404785, "lr": 9.806422462760687e-06, "epoch": 0.3626666666666667, "percentage": 18.12, "elapsed_time": "0:11:36", "remaining_time": "0:52:27"}
|
||||
{"current_steps": 180, "total_steps": 938, "loss": 0.30700514316558836, "lr": 9.751819994403802e-06, "epoch": 0.384, "percentage": 19.19, "elapsed_time": "0:12:19", "remaining_time": "0:51:54"}
|
||||
{"current_steps": 190, "total_steps": 938, "loss": 0.31289896965026853, "lr": 9.690634513417487e-06, "epoch": 0.4053333333333333, "percentage": 20.26, "elapsed_time": "0:12:59", "remaining_time": "0:51:09"}
|
||||
{"current_steps": 200, "total_steps": 938, "loss": 0.3092350959777832, "lr": 9.622950784122473e-06, "epoch": 0.4266666666666667, "percentage": 21.32, "elapsed_time": "0:13:40", "remaining_time": "0:50:26"}
|
||||
{"current_steps": 210, "total_steps": 938, "loss": 0.3032077312469482, "lr": 9.548862573295552e-06, "epoch": 0.448, "percentage": 22.39, "elapsed_time": "0:14:22", "remaining_time": "0:49:50"}
|
||||
{"current_steps": 220, "total_steps": 938, "loss": 0.30856895446777344, "lr": 9.468472520268207e-06, "epoch": 0.4693333333333333, "percentage": 23.45, "elapsed_time": "0:15:04", "remaining_time": "0:49:13"}
|
||||
{"current_steps": 230, "total_steps": 938, "loss": 0.3001561641693115, "lr": 9.38189199473352e-06, "epoch": 0.49066666666666664, "percentage": 24.52, "elapsed_time": "0:15:45", "remaining_time": "0:48:31"}
|
||||
{"current_steps": 240, "total_steps": 938, "loss": 0.2990098476409912, "lr": 9.289240942458322e-06, "epoch": 0.512, "percentage": 25.59, "elapsed_time": "0:16:25", "remaining_time": "0:47:45"}
|
||||
{"current_steps": 250, "total_steps": 938, "loss": 0.2924530029296875, "lr": 9.190647719114328e-06, "epoch": 0.5333333333333333, "percentage": 26.65, "elapsed_time": "0:17:08", "remaining_time": "0:47:09"}
|
||||
{"current_steps": 260, "total_steps": 938, "loss": 0.3162691116333008, "lr": 9.086248912458484e-06, "epoch": 0.5546666666666666, "percentage": 27.72, "elapsed_time": "0:17:48", "remaining_time": "0:46:25"}
|
||||
{"current_steps": 270, "total_steps": 938, "loss": 0.30709133148193357, "lr": 8.976189153108853e-06, "epoch": 0.576, "percentage": 28.78, "elapsed_time": "0:18:28", "remaining_time": "0:45:42"}
|
||||
{"current_steps": 280, "total_steps": 938, "loss": 0.3032949924468994, "lr": 8.860620914178188e-06, "epoch": 0.5973333333333334, "percentage": 29.85, "elapsed_time": "0:19:08", "remaining_time": "0:44:59"}
|
||||
{"current_steps": 290, "total_steps": 938, "loss": 0.29636149406433104, "lr": 8.73970430004278e-06, "epoch": 0.6186666666666667, "percentage": 30.92, "elapsed_time": "0:19:49", "remaining_time": "0:44:18"}
|
||||
{"current_steps": 300, "total_steps": 938, "loss": 0.3070181369781494, "lr": 8.613606824539198e-06, "epoch": 0.64, "percentage": 31.98, "elapsed_time": "0:20:32", "remaining_time": "0:43:41"}
|
||||
{"current_steps": 310, "total_steps": 938, "loss": 0.3074164867401123, "lr": 8.482503178896227e-06, "epoch": 0.6613333333333333, "percentage": 33.05, "elapsed_time": "0:21:12", "remaining_time": "0:42:57"}
|
||||
{"current_steps": 320, "total_steps": 938, "loss": 0.2806987762451172, "lr": 8.34657498972347e-06, "epoch": 0.6826666666666666, "percentage": 34.12, "elapsed_time": "0:21:55", "remaining_time": "0:42:20"}
|
||||
{"current_steps": 330, "total_steps": 938, "loss": 0.28669133186340334, "lr": 8.206010567391918e-06, "epoch": 0.704, "percentage": 35.18, "elapsed_time": "0:22:37", "remaining_time": "0:41:41"}
|
||||
{"current_steps": 340, "total_steps": 938, "loss": 0.30783905982971194, "lr": 8.061004645155049e-06, "epoch": 0.7253333333333334, "percentage": 36.25, "elapsed_time": "0:23:23", "remaining_time": "0:41:08"}
|
||||
{"current_steps": 350, "total_steps": 938, "loss": 0.29372529983520507, "lr": 7.91175810937189e-06, "epoch": 0.7466666666666667, "percentage": 37.31, "elapsed_time": "0:24:06", "remaining_time": "0:40:30"}
|
||||
{"current_steps": 360, "total_steps": 938, "loss": 0.2857055187225342, "lr": 7.758477721205767e-06, "epoch": 0.768, "percentage": 38.38, "elapsed_time": "0:24:50", "remaining_time": "0:39:52"}
|
||||
{"current_steps": 370, "total_steps": 938, "loss": 0.2911552429199219, "lr": 7.601375830184297e-06, "epoch": 0.7893333333333333, "percentage": 39.45, "elapsed_time": "0:25:32", "remaining_time": "0:39:12"}
|
||||
{"current_steps": 380, "total_steps": 938, "loss": 0.28858585357666017, "lr": 7.4406700800174545e-06, "epoch": 0.8106666666666666, "percentage": 40.51, "elapsed_time": "0:26:18", "remaining_time": "0:38:37"}
|
||||
{"current_steps": 390, "total_steps": 938, "loss": 0.2948747158050537, "lr": 7.276583107081243e-06, "epoch": 0.832, "percentage": 41.58, "elapsed_time": "0:26:59", "remaining_time": "0:37:56"}
|
||||
{"current_steps": 400, "total_steps": 938, "loss": 0.28877460956573486, "lr": 7.109342231984698e-06, "epoch": 0.8533333333333334, "percentage": 42.64, "elapsed_time": "0:27:43", "remaining_time": "0:37:16"}
|
||||
{"current_steps": 410, "total_steps": 938, "loss": 0.3021462202072144, "lr": 6.939179144647516e-06, "epoch": 0.8746666666666667, "percentage": 43.71, "elapsed_time": "0:28:23", "remaining_time": "0:36:34"}
|
||||
{"current_steps": 420, "total_steps": 938, "loss": 0.2928939342498779, "lr": 6.766329583324582e-06, "epoch": 0.896, "percentage": 44.78, "elapsed_time": "0:29:06", "remaining_time": "0:35:53"}
|
||||
{"current_steps": 430, "total_steps": 938, "loss": 0.28879172801971437, "lr": 6.591033008022067e-06, "epoch": 0.9173333333333333, "percentage": 45.84, "elapsed_time": "0:29:46", "remaining_time": "0:35:10"}
|
||||
{"current_steps": 440, "total_steps": 938, "loss": 0.2961143493652344, "lr": 6.413532268757537e-06, "epoch": 0.9386666666666666, "percentage": 46.91, "elapsed_time": "0:30:28", "remaining_time": "0:34:29"}
|
||||
{"current_steps": 450, "total_steps": 938, "loss": 0.2954572200775146, "lr": 6.234073269123654e-06, "epoch": 0.96, "percentage": 47.97, "elapsed_time": "0:31:09", "remaining_time": "0:33:47"}
|
||||
{"current_steps": 460, "total_steps": 938, "loss": 0.2948294162750244, "lr": 6.052904625621556e-06, "epoch": 0.9813333333333333, "percentage": 49.04, "elapsed_time": "0:31:54", "remaining_time": "0:33:09"}
|
||||
{"current_steps": 470, "total_steps": 938, "loss": 0.2971747875213623, "lr": 5.870277323235871e-06, "epoch": 1.0021333333333333, "percentage": 50.11, "elapsed_time": "0:32:34", "remaining_time": "0:32:26"}
|
||||
{"current_steps": 480, "total_steps": 938, "loss": 0.26477513313293455, "lr": 5.686444367728494e-06, "epoch": 1.0234666666666667, "percentage": 51.17, "elapsed_time": "0:33:15", "remaining_time": "0:31:44"}
|
||||
{"current_steps": 490, "total_steps": 938, "loss": 0.2709561824798584, "lr": 5.501660435132871e-06, "epoch": 1.0448, "percentage": 52.24, "elapsed_time": "0:33:57", "remaining_time": "0:31:02"}
|
||||
{"current_steps": 500, "total_steps": 938, "loss": 0.27056255340576174, "lr": 5.316181518934319e-06, "epoch": 1.0661333333333334, "percentage": 53.3, "elapsed_time": "0:34:38", "remaining_time": "0:30:20"}
|
||||
{"current_steps": 510, "total_steps": 938, "loss": 0.27835679054260254, "lr": 5.130264575425225e-06, "epoch": 1.0874666666666666, "percentage": 54.37, "elapsed_time": "0:35:30", "remaining_time": "0:29:48"}
|
||||
{"current_steps": 520, "total_steps": 938, "loss": 0.26178646087646484, "lr": 4.944167167726367e-06, "epoch": 1.1088, "percentage": 55.44, "elapsed_time": "0:36:13", "remaining_time": "0:29:07"}
|
||||
{"current_steps": 530, "total_steps": 938, "loss": 0.2711505889892578, "lr": 4.758147108967585e-06, "epoch": 1.1301333333333332, "percentage": 56.5, "elapsed_time": "0:36:54", "remaining_time": "0:28:24"}
|
||||
{"current_steps": 540, "total_steps": 938, "loss": 0.262709903717041, "lr": 4.572462105122078e-06, "epoch": 1.1514666666666666, "percentage": 57.57, "elapsed_time": "0:37:37", "remaining_time": "0:27:43"}
|
||||
{"current_steps": 550, "total_steps": 938, "loss": 0.2738328218460083, "lr": 4.3873693979891705e-06, "epoch": 1.1728, "percentage": 58.64, "elapsed_time": "0:38:18", "remaining_time": "0:27:01"}
|
||||
{"current_steps": 560, "total_steps": 938, "loss": 0.26540517807006836, "lr": 4.203125408820106e-06, "epoch": 1.1941333333333333, "percentage": 59.7, "elapsed_time": "0:38:58", "remaining_time": "0:26:18"}
|
||||
{"current_steps": 570, "total_steps": 938, "loss": 0.26726350784301756, "lr": 4.019985383080632e-06, "epoch": 1.2154666666666667, "percentage": 60.77, "elapsed_time": "0:39:40", "remaining_time": "0:25:37"}
|
||||
{"current_steps": 580, "total_steps": 938, "loss": 0.2629551887512207, "lr": 3.838203036842446e-06, "epoch": 1.2368000000000001, "percentage": 61.83, "elapsed_time": "0:40:20", "remaining_time": "0:24:53"}
|
||||
{"current_steps": 590, "total_steps": 938, "loss": 0.2717682600021362, "lr": 3.65803020529343e-06, "epoch": 1.2581333333333333, "percentage": 62.9, "elapsed_time": "0:41:04", "remaining_time": "0:24:13"}
|
||||
{"current_steps": 600, "total_steps": 938, "loss": 0.2709221601486206, "lr": 3.4797164938536113e-06, "epoch": 1.2794666666666665, "percentage": 63.97, "elapsed_time": "0:41:46", "remaining_time": "0:23:32"}
|
||||
{"current_steps": 610, "total_steps": 938, "loss": 0.26828670501708984, "lr": 3.3035089323801323e-06, "epoch": 1.3008, "percentage": 65.03, "elapsed_time": "0:42:29", "remaining_time": "0:22:50"}
|
||||
{"current_steps": 620, "total_steps": 938, "loss": 0.27147223949432375, "lr": 3.1296516329403626e-06, "epoch": 1.3221333333333334, "percentage": 66.1, "elapsed_time": "0:43:10", "remaining_time": "0:22:08"}
|
||||
{"current_steps": 630, "total_steps": 938, "loss": 0.272589111328125, "lr": 2.9583854516271815e-06, "epoch": 1.3434666666666666, "percentage": 67.16, "elapsed_time": "0:43:51", "remaining_time": "0:21:26"}
|
||||
{"current_steps": 640, "total_steps": 938, "loss": 0.2664657115936279, "lr": 2.7899476548850046e-06, "epoch": 1.3648, "percentage": 68.23, "elapsed_time": "0:44:32", "remaining_time": "0:20:44"}
|
||||
{"current_steps": 650, "total_steps": 938, "loss": 0.2647489786148071, "lr": 2.624571590808781e-06, "epoch": 1.3861333333333334, "percentage": 69.3, "elapsed_time": "0:45:12", "remaining_time": "0:20:01"}
|
||||
{"current_steps": 660, "total_steps": 938, "loss": 0.2691283941268921, "lr": 2.4624863658713383e-06, "epoch": 1.4074666666666666, "percentage": 70.36, "elapsed_time": "0:45:53", "remaining_time": "0:19:19"}
|
||||
{"current_steps": 670, "total_steps": 938, "loss": 0.2635840892791748, "lr": 2.3039165275269214e-06, "epoch": 1.4288, "percentage": 71.43, "elapsed_time": "0:46:34", "remaining_time": "0:18:37"}
|
||||
{"current_steps": 680, "total_steps": 938, "loss": 0.2607363224029541, "lr": 2.1490817531306778e-06, "epoch": 1.4501333333333333, "percentage": 72.49, "elapsed_time": "0:47:15", "remaining_time": "0:17:55"}
|
||||
{"current_steps": 690, "total_steps": 938, "loss": 0.27231812477111816, "lr": 1.99819654560496e-06, "epoch": 1.4714666666666667, "percentage": 73.56, "elapsed_time": "0:47:54", "remaining_time": "0:17:13"}
|
||||
{"current_steps": 700, "total_steps": 938, "loss": 0.2777836561203003, "lr": 1.851469936274174e-06, "epoch": 1.4928, "percentage": 74.63, "elapsed_time": "0:48:35", "remaining_time": "0:16:31"}
|
||||
{"current_steps": 710, "total_steps": 938, "loss": 0.2632488965988159, "lr": 1.7091051952797404e-06, "epoch": 1.5141333333333333, "percentage": 75.69, "elapsed_time": "0:49:18", "remaining_time": "0:15:50"}
|
||||
{"current_steps": 720, "total_steps": 938, "loss": 0.2699136257171631, "lr": 1.571299549976456e-06, "epoch": 1.5354666666666668, "percentage": 76.76, "elapsed_time": "0:49:58", "remaining_time": "0:15:07"}
|
||||
{"current_steps": 730, "total_steps": 938, "loss": 0.25556948184967043, "lr": 1.4382439117002938e-06, "epoch": 1.5568, "percentage": 77.83, "elapsed_time": "0:50:41", "remaining_time": "0:14:26"}
|
||||
{"current_steps": 740, "total_steps": 938, "loss": 0.26279048919677733, "lr": 1.3101226112862232e-06, "epoch": 1.5781333333333334, "percentage": 78.89, "elapsed_time": "0:51:22", "remaining_time": "0:13:44"}
|
||||
{"current_steps": 750, "total_steps": 938, "loss": 0.2726590633392334, "lr": 1.1871131437024291e-06, "epoch": 1.5994666666666668, "percentage": 79.96, "elapsed_time": "0:52:02", "remaining_time": "0:13:02"}
|
||||
{"current_steps": 760, "total_steps": 938, "loss": 0.2752861499786377, "lr": 1.0693859221547115e-06, "epoch": 1.6208, "percentage": 81.02, "elapsed_time": "0:52:45", "remaining_time": "0:12:21"}
|
||||
{"current_steps": 770, "total_steps": 938, "loss": 0.2579826354980469, "lr": 9.571040420017324e-07, "epoch": 1.6421333333333332, "percentage": 82.09, "elapsed_time": "0:53:28", "remaining_time": "0:11:39"}
|
||||
{"current_steps": 780, "total_steps": 938, "loss": 0.26852080821990965, "lr": 8.504230548081499e-07, "epoch": 1.6634666666666666, "percentage": 83.16, "elapsed_time": "0:54:10", "remaining_time": "0:10:58"}
|
||||
{"current_steps": 790, "total_steps": 938, "loss": 0.26603677272796633, "lr": 7.4949075284868e-07, "epoch": 1.6848, "percentage": 84.22, "elapsed_time": "0:54:52", "remaining_time": "0:10:16"}
|
||||
{"current_steps": 800, "total_steps": 938, "loss": 0.25719456672668456, "lr": 6.544469643616192e-07, "epoch": 1.7061333333333333, "percentage": 85.29, "elapsed_time": "0:55:34", "remaining_time": "0:09:35"}
|
||||
{"current_steps": 810, "total_steps": 938, "loss": 0.2663470506668091, "lr": 5.654233598354752e-07, "epoch": 1.7274666666666667, "percentage": 86.35, "elapsed_time": "0:56:15", "remaining_time": "0:08:53"}
|
||||
{"current_steps": 820, "total_steps": 938, "loss": 0.26163082122802733, "lr": 4.825432695970673e-07, "epoch": 1.7488000000000001, "percentage": 87.42, "elapsed_time": "0:57:00", "remaining_time": "0:08:12"}
|
||||
{"current_steps": 830, "total_steps": 938, "loss": 0.26932682991027834, "lr": 4.0592151295382464e-07, "epoch": 1.7701333333333333, "percentage": 88.49, "elapsed_time": "0:57:40", "remaining_time": "0:07:30"}
|
||||
{"current_steps": 840, "total_steps": 938, "loss": 0.27272663116455076, "lr": 3.356642391269405e-07, "epoch": 1.7914666666666665, "percentage": 89.55, "elapsed_time": "0:58:21", "remaining_time": "0:06:48"}
|
||||
{"current_steps": 850, "total_steps": 938, "loss": 0.2759810447692871, "lr": 2.7186878019580197e-07, "epoch": 1.8128, "percentage": 90.62, "elapsed_time": "0:59:02", "remaining_time": "0:06:06"}
|
||||
{"current_steps": 860, "total_steps": 938, "loss": 0.2736822128295898, "lr": 2.1462351625736677e-07, "epoch": 1.8341333333333334, "percentage": 91.68, "elapsed_time": "0:59:43", "remaining_time": "0:05:24"}
|
||||
{"current_steps": 870, "total_steps": 938, "loss": 0.2506378173828125, "lr": 1.6400775298734017e-07, "epoch": 1.8554666666666666, "percentage": 92.75, "elapsed_time": "1:00:28", "remaining_time": "0:04:43"}
|
||||
{"current_steps": 880, "total_steps": 938, "loss": 0.2735074520111084, "lr": 1.2009161177273742e-07, "epoch": 1.8768, "percentage": 93.82, "elapsed_time": "1:01:08", "remaining_time": "0:04:01"}
|
||||
{"current_steps": 890, "total_steps": 938, "loss": 0.26666769981384275, "lr": 8.293593256805842e-08, "epoch": 1.8981333333333335, "percentage": 94.88, "elapsed_time": "1:01:49", "remaining_time": "0:03:20"}
|
||||
{"current_steps": 900, "total_steps": 938, "loss": 0.2759205102920532, "lr": 5.259218960964874e-08, "epoch": 1.9194666666666667, "percentage": 95.95, "elapsed_time": "1:02:31", "remaining_time": "0:02:38"}
|
||||
{"current_steps": 910, "total_steps": 938, "loss": 0.27174739837646483, "lr": 2.9102420105009966e-08, "epoch": 1.9407999999999999, "percentage": 97.01, "elapsed_time": "1:03:12", "remaining_time": "0:01:56"}
|
||||
{"current_steps": 920, "total_steps": 938, "loss": 0.26730976104736326, "lr": 1.249916599585954e-08, "epoch": 1.9621333333333333, "percentage": 98.08, "elapsed_time": "1:03:50", "remaining_time": "0:01:14"}
|
||||
{"current_steps": 930, "total_steps": 938, "loss": 0.2772205352783203, "lr": 2.80542887561297e-09, "epoch": 1.9834666666666667, "percentage": 99.15, "elapsed_time": "1:04:31", "remaining_time": "0:00:33"}
|
||||
{"current_steps": 938, "total_steps": 938, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "1:05:13", "remaining_time": "0:00:00"}
|
||||
694
trainer_state.json
Normal file
694
trainer_state.json
Normal file
@@ -0,0 +1,694 @@
|
||||
{
|
||||
"best_global_step": null,
|
||||
"best_metric": null,
|
||||
"best_model_checkpoint": null,
|
||||
"epoch": 2.0,
|
||||
"eval_steps": 500,
|
||||
"global_step": 938,
|
||||
"is_hyper_param_search": false,
|
||||
"is_local_process_zero": true,
|
||||
"is_world_process_zero": true,
|
||||
"log_history": [
|
||||
{
|
||||
"epoch": 0.021333333333333333,
|
||||
"grad_norm": 11.143438339233398,
|
||||
"learning_rate": 9.574468085106384e-07,
|
||||
"loss": 0.6841894149780273,
|
||||
"step": 10
|
||||
},
|
||||
{
|
||||
"epoch": 0.042666666666666665,
|
||||
"grad_norm": 4.5582380294799805,
|
||||
"learning_rate": 2.021276595744681e-06,
|
||||
"loss": 0.6096980094909668,
|
||||
"step": 20
|
||||
},
|
||||
{
|
||||
"epoch": 0.064,
|
||||
"grad_norm": 1.4977697134017944,
|
||||
"learning_rate": 3.0851063829787237e-06,
|
||||
"loss": 0.46523351669311525,
|
||||
"step": 30
|
||||
},
|
||||
{
|
||||
"epoch": 0.08533333333333333,
|
||||
"grad_norm": 0.7574602961540222,
|
||||
"learning_rate": 4.148936170212766e-06,
|
||||
"loss": 0.38690838813781736,
|
||||
"step": 40
|
||||
},
|
||||
{
|
||||
"epoch": 0.10666666666666667,
|
||||
"grad_norm": 0.6511842012405396,
|
||||
"learning_rate": 5.212765957446809e-06,
|
||||
"loss": 0.3463115692138672,
|
||||
"step": 50
|
||||
},
|
||||
{
|
||||
"epoch": 0.128,
|
||||
"grad_norm": 0.5265598297119141,
|
||||
"learning_rate": 6.276595744680851e-06,
|
||||
"loss": 0.3305189609527588,
|
||||
"step": 60
|
||||
},
|
||||
{
|
||||
"epoch": 0.14933333333333335,
|
||||
"grad_norm": 0.5384871363639832,
|
||||
"learning_rate": 7.340425531914894e-06,
|
||||
"loss": 0.33223772048950195,
|
||||
"step": 70
|
||||
},
|
||||
{
|
||||
"epoch": 0.17066666666666666,
|
||||
"grad_norm": 0.5109423995018005,
|
||||
"learning_rate": 8.404255319148937e-06,
|
||||
"loss": 0.31659588813781736,
|
||||
"step": 80
|
||||
},
|
||||
{
|
||||
"epoch": 0.192,
|
||||
"grad_norm": 0.47720399498939514,
|
||||
"learning_rate": 9.46808510638298e-06,
|
||||
"loss": 0.31022181510925295,
|
||||
"step": 90
|
||||
},
|
||||
{
|
||||
"epoch": 0.21333333333333335,
|
||||
"grad_norm": 0.49756094813346863,
|
||||
"learning_rate": 9.999134070902206e-06,
|
||||
"loss": 0.3155974388122559,
|
||||
"step": 100
|
||||
},
|
||||
{
|
||||
"epoch": 0.23466666666666666,
|
||||
"grad_norm": 0.4745752215385437,
|
||||
"learning_rate": 9.992208437615651e-06,
|
||||
"loss": 0.31209163665771483,
|
||||
"step": 110
|
||||
},
|
||||
{
|
||||
"epoch": 0.256,
|
||||
"grad_norm": 0.4676297903060913,
|
||||
"learning_rate": 9.97836676558346e-06,
|
||||
"loss": 0.30946059226989747,
|
||||
"step": 120
|
||||
},
|
||||
{
|
||||
"epoch": 0.2773333333333333,
|
||||
"grad_norm": 0.4762783944606781,
|
||||
"learning_rate": 9.957628230595527e-06,
|
||||
"loss": 0.3128779888153076,
|
||||
"step": 130
|
||||
},
|
||||
{
|
||||
"epoch": 0.2986666666666667,
|
||||
"grad_norm": 0.4929763078689575,
|
||||
"learning_rate": 9.930021563125204e-06,
|
||||
"loss": 0.3062174320220947,
|
||||
"step": 140
|
||||
},
|
||||
{
|
||||
"epoch": 0.32,
|
||||
"grad_norm": 0.4954136312007904,
|
||||
"learning_rate": 9.895585008527075e-06,
|
||||
"loss": 0.30686254501342775,
|
||||
"step": 150
|
||||
},
|
||||
{
|
||||
"epoch": 0.3413333333333333,
|
||||
"grad_norm": 0.48502442240715027,
|
||||
"learning_rate": 9.854366274053125e-06,
|
||||
"loss": 0.3146007299423218,
|
||||
"step": 160
|
||||
},
|
||||
{
|
||||
"epoch": 0.3626666666666667,
|
||||
"grad_norm": 0.4601064622402191,
|
||||
"learning_rate": 9.806422462760687e-06,
|
||||
"loss": 0.30956292152404785,
|
||||
"step": 170
|
||||
},
|
||||
{
|
||||
"epoch": 0.384,
|
||||
"grad_norm": 0.4399697780609131,
|
||||
"learning_rate": 9.751819994403802e-06,
|
||||
"loss": 0.30700514316558836,
|
||||
"step": 180
|
||||
},
|
||||
{
|
||||
"epoch": 0.4053333333333333,
|
||||
"grad_norm": 0.5086579322814941,
|
||||
"learning_rate": 9.690634513417487e-06,
|
||||
"loss": 0.31289896965026853,
|
||||
"step": 190
|
||||
},
|
||||
{
|
||||
"epoch": 0.4266666666666667,
|
||||
"grad_norm": 0.4878501892089844,
|
||||
"learning_rate": 9.622950784122473e-06,
|
||||
"loss": 0.3092350959777832,
|
||||
"step": 200
|
||||
},
|
||||
{
|
||||
"epoch": 0.448,
|
||||
"grad_norm": 0.47980615496635437,
|
||||
"learning_rate": 9.548862573295552e-06,
|
||||
"loss": 0.3032077312469482,
|
||||
"step": 210
|
||||
},
|
||||
{
|
||||
"epoch": 0.4693333333333333,
|
||||
"grad_norm": 0.4887213110923767,
|
||||
"learning_rate": 9.468472520268207e-06,
|
||||
"loss": 0.30856895446777344,
|
||||
"step": 220
|
||||
},
|
||||
{
|
||||
"epoch": 0.49066666666666664,
|
||||
"grad_norm": 0.48731157183647156,
|
||||
"learning_rate": 9.38189199473352e-06,
|
||||
"loss": 0.3001561641693115,
|
||||
"step": 230
|
||||
},
|
||||
{
|
||||
"epoch": 0.512,
|
||||
"grad_norm": 0.5213120579719543,
|
||||
"learning_rate": 9.289240942458322e-06,
|
||||
"loss": 0.2990098476409912,
|
||||
"step": 240
|
||||
},
|
||||
{
|
||||
"epoch": 0.5333333333333333,
|
||||
"grad_norm": 0.46841681003570557,
|
||||
"learning_rate": 9.190647719114328e-06,
|
||||
"loss": 0.2924530029296875,
|
||||
"step": 250
|
||||
},
|
||||
{
|
||||
"epoch": 0.5546666666666666,
|
||||
"grad_norm": 0.45176810026168823,
|
||||
"learning_rate": 9.086248912458484e-06,
|
||||
"loss": 0.3162691116333008,
|
||||
"step": 260
|
||||
},
|
||||
{
|
||||
"epoch": 0.576,
|
||||
"grad_norm": 0.4480721056461334,
|
||||
"learning_rate": 8.976189153108853e-06,
|
||||
"loss": 0.30709133148193357,
|
||||
"step": 270
|
||||
},
|
||||
{
|
||||
"epoch": 0.5973333333333334,
|
||||
"grad_norm": 0.4673004150390625,
|
||||
"learning_rate": 8.860620914178188e-06,
|
||||
"loss": 0.3032949924468994,
|
||||
"step": 280
|
||||
},
|
||||
{
|
||||
"epoch": 0.6186666666666667,
|
||||
"grad_norm": 0.4601098299026489,
|
||||
"learning_rate": 8.73970430004278e-06,
|
||||
"loss": 0.29636149406433104,
|
||||
"step": 290
|
||||
},
|
||||
{
|
||||
"epoch": 0.64,
|
||||
"grad_norm": 0.4462672173976898,
|
||||
"learning_rate": 8.613606824539198e-06,
|
||||
"loss": 0.3070181369781494,
|
||||
"step": 300
|
||||
},
|
||||
{
|
||||
"epoch": 0.6613333333333333,
|
||||
"grad_norm": 0.48093080520629883,
|
||||
"learning_rate": 8.482503178896227e-06,
|
||||
"loss": 0.3074164867401123,
|
||||
"step": 310
|
||||
},
|
||||
{
|
||||
"epoch": 0.6826666666666666,
|
||||
"grad_norm": 0.5080389380455017,
|
||||
"learning_rate": 8.34657498972347e-06,
|
||||
"loss": 0.2806987762451172,
|
||||
"step": 320
|
||||
},
|
||||
{
|
||||
"epoch": 0.704,
|
||||
"grad_norm": 0.46668171882629395,
|
||||
"learning_rate": 8.206010567391918e-06,
|
||||
"loss": 0.28669133186340334,
|
||||
"step": 330
|
||||
},
|
||||
{
|
||||
"epoch": 0.7253333333333334,
|
||||
"grad_norm": 0.4524853527545929,
|
||||
"learning_rate": 8.061004645155049e-06,
|
||||
"loss": 0.30783905982971194,
|
||||
"step": 340
|
||||
},
|
||||
{
|
||||
"epoch": 0.7466666666666667,
|
||||
"grad_norm": 0.5166858434677124,
|
||||
"learning_rate": 7.91175810937189e-06,
|
||||
"loss": 0.29372529983520507,
|
||||
"step": 350
|
||||
},
|
||||
{
|
||||
"epoch": 0.768,
|
||||
"grad_norm": 0.46780723333358765,
|
||||
"learning_rate": 7.758477721205767e-06,
|
||||
"loss": 0.2857055187225342,
|
||||
"step": 360
|
||||
},
|
||||
{
|
||||
"epoch": 0.7893333333333333,
|
||||
"grad_norm": 0.47541725635528564,
|
||||
"learning_rate": 7.601375830184297e-06,
|
||||
"loss": 0.2911552429199219,
|
||||
"step": 370
|
||||
},
|
||||
{
|
||||
"epoch": 0.8106666666666666,
|
||||
"grad_norm": 0.44671547412872314,
|
||||
"learning_rate": 7.4406700800174545e-06,
|
||||
"loss": 0.28858585357666017,
|
||||
"step": 380
|
||||
},
|
||||
{
|
||||
"epoch": 0.832,
|
||||
"grad_norm": 0.44009825587272644,
|
||||
"learning_rate": 7.276583107081243e-06,
|
||||
"loss": 0.2948747158050537,
|
||||
"step": 390
|
||||
},
|
||||
{
|
||||
"epoch": 0.8533333333333334,
|
||||
"grad_norm": 0.4630362093448639,
|
||||
"learning_rate": 7.109342231984698e-06,
|
||||
"loss": 0.28877460956573486,
|
||||
"step": 400
|
||||
},
|
||||
{
|
||||
"epoch": 0.8746666666666667,
|
||||
"grad_norm": 0.506441593170166,
|
||||
"learning_rate": 6.939179144647516e-06,
|
||||
"loss": 0.3021462202072144,
|
||||
"step": 410
|
||||
},
|
||||
{
|
||||
"epoch": 0.896,
|
||||
"grad_norm": 0.4811643660068512,
|
||||
"learning_rate": 6.766329583324582e-06,
|
||||
"loss": 0.2928939342498779,
|
||||
"step": 420
|
||||
},
|
||||
{
|
||||
"epoch": 0.9173333333333333,
|
||||
"grad_norm": 0.4531882107257843,
|
||||
"learning_rate": 6.591033008022067e-06,
|
||||
"loss": 0.28879172801971437,
|
||||
"step": 430
|
||||
},
|
||||
{
|
||||
"epoch": 0.9386666666666666,
|
||||
"grad_norm": 0.4446716904640198,
|
||||
"learning_rate": 6.413532268757537e-06,
|
||||
"loss": 0.2961143493652344,
|
||||
"step": 440
|
||||
},
|
||||
{
|
||||
"epoch": 0.96,
|
||||
"grad_norm": 0.41976165771484375,
|
||||
"learning_rate": 6.234073269123654e-06,
|
||||
"loss": 0.2954572200775146,
|
||||
"step": 450
|
||||
},
|
||||
{
|
||||
"epoch": 0.9813333333333333,
|
||||
"grad_norm": 0.4504052698612213,
|
||||
"learning_rate": 6.052904625621556e-06,
|
||||
"loss": 0.2948294162750244,
|
||||
"step": 460
|
||||
},
|
||||
{
|
||||
"epoch": 1.0021333333333333,
|
||||
"grad_norm": 0.5018347501754761,
|
||||
"learning_rate": 5.870277323235871e-06,
|
||||
"loss": 0.2971747875213623,
|
||||
"step": 470
|
||||
},
|
||||
{
|
||||
"epoch": 1.0234666666666667,
|
||||
"grad_norm": 0.48416292667388916,
|
||||
"learning_rate": 5.686444367728494e-06,
|
||||
"loss": 0.26477513313293455,
|
||||
"step": 480
|
||||
},
|
||||
{
|
||||
"epoch": 1.0448,
|
||||
"grad_norm": 0.4401777386665344,
|
||||
"learning_rate": 5.501660435132871e-06,
|
||||
"loss": 0.2709561824798584,
|
||||
"step": 490
|
||||
},
|
||||
{
|
||||
"epoch": 1.0661333333333334,
|
||||
"grad_norm": 0.4749163091182709,
|
||||
"learning_rate": 5.316181518934319e-06,
|
||||
"loss": 0.27056255340576174,
|
||||
"step": 500
|
||||
},
|
||||
{
|
||||
"epoch": 1.0874666666666666,
|
||||
"grad_norm": 0.4591522216796875,
|
||||
"learning_rate": 5.130264575425225e-06,
|
||||
"loss": 0.27835679054260254,
|
||||
"step": 510
|
||||
},
|
||||
{
|
||||
"epoch": 1.1088,
|
||||
"grad_norm": 0.4608944356441498,
|
||||
"learning_rate": 4.944167167726367e-06,
|
||||
"loss": 0.26178646087646484,
|
||||
"step": 520
|
||||
},
|
||||
{
|
||||
"epoch": 1.1301333333333332,
|
||||
"grad_norm": 0.48433825373649597,
|
||||
"learning_rate": 4.758147108967585e-06,
|
||||
"loss": 0.2711505889892578,
|
||||
"step": 530
|
||||
},
|
||||
{
|
||||
"epoch": 1.1514666666666666,
|
||||
"grad_norm": 0.5324186682701111,
|
||||
"learning_rate": 4.572462105122078e-06,
|
||||
"loss": 0.262709903717041,
|
||||
"step": 540
|
||||
},
|
||||
{
|
||||
"epoch": 1.1728,
|
||||
"grad_norm": 0.4596474766731262,
|
||||
"learning_rate": 4.3873693979891705e-06,
|
||||
"loss": 0.2738328218460083,
|
||||
"step": 550
|
||||
},
|
||||
{
|
||||
"epoch": 1.1941333333333333,
|
||||
"grad_norm": 0.48839670419692993,
|
||||
"learning_rate": 4.203125408820106e-06,
|
||||
"loss": 0.26540517807006836,
|
||||
"step": 560
|
||||
},
|
||||
{
|
||||
"epoch": 1.2154666666666667,
|
||||
"grad_norm": 0.4470955729484558,
|
||||
"learning_rate": 4.019985383080632e-06,
|
||||
"loss": 0.26726350784301756,
|
||||
"step": 570
|
||||
},
|
||||
{
|
||||
"epoch": 1.2368000000000001,
|
||||
"grad_norm": 0.44690802693367004,
|
||||
"learning_rate": 3.838203036842446e-06,
|
||||
"loss": 0.2629551887512207,
|
||||
"step": 580
|
||||
},
|
||||
{
|
||||
"epoch": 1.2581333333333333,
|
||||
"grad_norm": 0.4668348431587219,
|
||||
"learning_rate": 3.65803020529343e-06,
|
||||
"loss": 0.2717682600021362,
|
||||
"step": 590
|
||||
},
|
||||
{
|
||||
"epoch": 1.2794666666666665,
|
||||
"grad_norm": 0.440378338098526,
|
||||
"learning_rate": 3.4797164938536113e-06,
|
||||
"loss": 0.2709221601486206,
|
||||
"step": 600
|
||||
},
|
||||
{
|
||||
"epoch": 1.3008,
|
||||
"grad_norm": 0.508844792842865,
|
||||
"learning_rate": 3.3035089323801323e-06,
|
||||
"loss": 0.26828670501708984,
|
||||
"step": 610
|
||||
},
|
||||
{
|
||||
"epoch": 1.3221333333333334,
|
||||
"grad_norm": 0.49634313583374023,
|
||||
"learning_rate": 3.1296516329403626e-06,
|
||||
"loss": 0.27147223949432375,
|
||||
"step": 620
|
||||
},
|
||||
{
|
||||
"epoch": 1.3434666666666666,
|
||||
"grad_norm": 0.4920685887336731,
|
||||
"learning_rate": 2.9583854516271815e-06,
|
||||
"loss": 0.272589111328125,
|
||||
"step": 630
|
||||
},
|
||||
{
|
||||
"epoch": 1.3648,
|
||||
"grad_norm": 0.4857788681983948,
|
||||
"learning_rate": 2.7899476548850046e-06,
|
||||
"loss": 0.2664657115936279,
|
||||
"step": 640
|
||||
},
|
||||
{
|
||||
"epoch": 1.3861333333333334,
|
||||
"grad_norm": 0.45807984471321106,
|
||||
"learning_rate": 2.624571590808781e-06,
|
||||
"loss": 0.2647489786148071,
|
||||
"step": 650
|
||||
},
|
||||
{
|
||||
"epoch": 1.4074666666666666,
|
||||
"grad_norm": 0.4540899097919464,
|
||||
"learning_rate": 2.4624863658713383e-06,
|
||||
"loss": 0.2691283941268921,
|
||||
"step": 660
|
||||
},
|
||||
{
|
||||
"epoch": 1.4288,
|
||||
"grad_norm": 0.5011265873908997,
|
||||
"learning_rate": 2.3039165275269214e-06,
|
||||
"loss": 0.2635840892791748,
|
||||
"step": 670
|
||||
},
|
||||
{
|
||||
"epoch": 1.4501333333333333,
|
||||
"grad_norm": 0.441933810710907,
|
||||
"learning_rate": 2.1490817531306778e-06,
|
||||
"loss": 0.2607363224029541,
|
||||
"step": 680
|
||||
},
|
||||
{
|
||||
"epoch": 1.4714666666666667,
|
||||
"grad_norm": 0.44827017188072205,
|
||||
"learning_rate": 1.99819654560496e-06,
|
||||
"loss": 0.27231812477111816,
|
||||
"step": 690
|
||||
},
|
||||
{
|
||||
"epoch": 1.4928,
|
||||
"grad_norm": 0.4514150619506836,
|
||||
"learning_rate": 1.851469936274174e-06,
|
||||
"loss": 0.2777836561203003,
|
||||
"step": 700
|
||||
},
|
||||
{
|
||||
"epoch": 1.5141333333333333,
|
||||
"grad_norm": 0.47360625863075256,
|
||||
"learning_rate": 1.7091051952797404e-06,
|
||||
"loss": 0.2632488965988159,
|
||||
"step": 710
|
||||
},
|
||||
{
|
||||
"epoch": 1.5354666666666668,
|
||||
"grad_norm": 0.4452371895313263,
|
||||
"learning_rate": 1.571299549976456e-06,
|
||||
"loss": 0.2699136257171631,
|
||||
"step": 720
|
||||
},
|
||||
{
|
||||
"epoch": 1.5568,
|
||||
"grad_norm": 0.45347362756729126,
|
||||
"learning_rate": 1.4382439117002938e-06,
|
||||
"loss": 0.25556948184967043,
|
||||
"step": 730
|
||||
},
|
||||
{
|
||||
"epoch": 1.5781333333333334,
|
||||
"grad_norm": 0.45227858424186707,
|
||||
"learning_rate": 1.3101226112862232e-06,
|
||||
"loss": 0.26279048919677733,
|
||||
"step": 740
|
||||
},
|
||||
{
|
||||
"epoch": 1.5994666666666668,
|
||||
"grad_norm": 0.4334966838359833,
|
||||
"learning_rate": 1.1871131437024291e-06,
|
||||
"loss": 0.2726590633392334,
|
||||
"step": 750
|
||||
},
|
||||
{
|
||||
"epoch": 1.6208,
|
||||
"grad_norm": 0.4579485058784485,
|
||||
"learning_rate": 1.0693859221547115e-06,
|
||||
"loss": 0.2752861499786377,
|
||||
"step": 760
|
||||
},
|
||||
{
|
||||
"epoch": 1.6421333333333332,
|
||||
"grad_norm": 0.48792654275894165,
|
||||
"learning_rate": 9.571040420017324e-07,
|
||||
"loss": 0.2579826354980469,
|
||||
"step": 770
|
||||
},
|
||||
{
|
||||
"epoch": 1.6634666666666666,
|
||||
"grad_norm": 0.4276820421218872,
|
||||
"learning_rate": 8.504230548081499e-07,
|
||||
"loss": 0.26852080821990965,
|
||||
"step": 780
|
||||
},
|
||||
{
|
||||
"epoch": 1.6848,
|
||||
"grad_norm": 0.4804341793060303,
|
||||
"learning_rate": 7.4949075284868e-07,
|
||||
"loss": 0.26603677272796633,
|
||||
"step": 790
|
||||
},
|
||||
{
|
||||
"epoch": 1.7061333333333333,
|
||||
"grad_norm": 0.4831051230430603,
|
||||
"learning_rate": 6.544469643616192e-07,
|
||||
"loss": 0.25719456672668456,
|
||||
"step": 800
|
||||
},
|
||||
{
|
||||
"epoch": 1.7274666666666667,
|
||||
"grad_norm": 0.508553683757782,
|
||||
"learning_rate": 5.654233598354752e-07,
|
||||
"loss": 0.2663470506668091,
|
||||
"step": 810
|
||||
},
|
||||
{
|
||||
"epoch": 1.7488000000000001,
|
||||
"grad_norm": 0.4779980480670929,
|
||||
"learning_rate": 4.825432695970673e-07,
|
||||
"loss": 0.26163082122802733,
|
||||
"step": 820
|
||||
},
|
||||
{
|
||||
"epoch": 1.7701333333333333,
|
||||
"grad_norm": 0.41447553038597107,
|
||||
"learning_rate": 4.0592151295382464e-07,
|
||||
"loss": 0.26932682991027834,
|
||||
"step": 830
|
||||
},
|
||||
{
|
||||
"epoch": 1.7914666666666665,
|
||||
"grad_norm": 0.43118005990982056,
|
||||
"learning_rate": 3.356642391269405e-07,
|
||||
"loss": 0.27272663116455076,
|
||||
"step": 840
|
||||
},
|
||||
{
|
||||
"epoch": 1.8128,
|
||||
"grad_norm": 0.4650283753871918,
|
||||
"learning_rate": 2.7186878019580197e-07,
|
||||
"loss": 0.2759810447692871,
|
||||
"step": 850
|
||||
},
|
||||
{
|
||||
"epoch": 1.8341333333333334,
|
||||
"grad_norm": 0.45311886072158813,
|
||||
"learning_rate": 2.1462351625736677e-07,
|
||||
"loss": 0.2736822128295898,
|
||||
"step": 860
|
||||
},
|
||||
{
|
||||
"epoch": 1.8554666666666666,
|
||||
"grad_norm": 0.43716010451316833,
|
||||
"learning_rate": 1.6400775298734017e-07,
|
||||
"loss": 0.2506378173828125,
|
||||
"step": 870
|
||||
},
|
||||
{
|
||||
"epoch": 1.8768,
|
||||
"grad_norm": 0.46398675441741943,
|
||||
"learning_rate": 1.2009161177273742e-07,
|
||||
"loss": 0.2735074520111084,
|
||||
"step": 880
|
||||
},
|
||||
{
|
||||
"epoch": 1.8981333333333335,
|
||||
"grad_norm": 0.4555061161518097,
|
||||
"learning_rate": 8.293593256805842e-08,
|
||||
"loss": 0.26666769981384275,
|
||||
"step": 890
|
||||
},
|
||||
{
|
||||
"epoch": 1.9194666666666667,
|
||||
"grad_norm": 0.4673771560192108,
|
||||
"learning_rate": 5.259218960964874e-08,
|
||||
"loss": 0.2759205102920532,
|
||||
"step": 900
|
||||
},
|
||||
{
|
||||
"epoch": 1.9407999999999999,
|
||||
"grad_norm": 0.46401599049568176,
|
||||
"learning_rate": 2.9102420105009966e-08,
|
||||
"loss": 0.27174739837646483,
|
||||
"step": 910
|
||||
},
|
||||
{
|
||||
"epoch": 1.9621333333333333,
|
||||
"grad_norm": 0.45422956347465515,
|
||||
"learning_rate": 1.249916599585954e-08,
|
||||
"loss": 0.26730976104736326,
|
||||
"step": 920
|
||||
},
|
||||
{
|
||||
"epoch": 1.9834666666666667,
|
||||
"grad_norm": 0.4789884686470032,
|
||||
"learning_rate": 2.80542887561297e-09,
|
||||
"loss": 0.2772205352783203,
|
||||
"step": 930
|
||||
},
|
||||
{
|
||||
"epoch": 2.0,
|
||||
"step": 938,
|
||||
"total_flos": 1.2200945171646382e+18,
|
||||
"train_loss": 0.2963919332032519,
|
||||
"train_runtime": 3915.3376,
|
||||
"train_samples_per_second": 30.649,
|
||||
"train_steps_per_second": 0.24
|
||||
}
|
||||
],
|
||||
"logging_steps": 10,
|
||||
"max_steps": 938,
|
||||
"num_input_tokens_seen": 0,
|
||||
"num_train_epochs": 2,
|
||||
"save_steps": 500,
|
||||
"stateful_callbacks": {
|
||||
"TrainerControl": {
|
||||
"args": {
|
||||
"should_epoch_stop": false,
|
||||
"should_evaluate": false,
|
||||
"should_log": false,
|
||||
"should_save": true,
|
||||
"should_training_stop": true
|
||||
},
|
||||
"attributes": {}
|
||||
}
|
||||
},
|
||||
"total_flos": 1.2200945171646382e+18,
|
||||
"train_batch_size": 2,
|
||||
"trial_name": null,
|
||||
"trial_params": null
|
||||
}
|
||||
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:cceab43f5c9cf083825a08d73354a3abf6cb2e33b74d60669f851f401d270ebd
|
||||
size 7185
|
||||
BIN
training_loss.png
Normal file
BIN
training_loss.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 34 KiB |
Reference in New Issue
Block a user