初始化项目，由ModelHub XC社区提供模型
Model: shubhamrgandhi/qwen3-4b-instruct-2507-full-sft-prm-r2egym-swebench-instructions-k5-qwen-only 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
|
||||||
60
README.md
Normal file
60
README.md
Normal file
@@ -0,0 +1,60 @@
|
|||||||
|
---
|
||||||
|
library_name: transformers
|
||||||
|
license: other
|
||||||
|
base_model: Qwen/Qwen3-4B-Instruct-2507
|
||||||
|
tags:
|
||||||
|
- llama-factory
|
||||||
|
- full
|
||||||
|
- generated_from_trainer
|
||||||
|
model-index:
|
||||||
|
- name: qwen3-4b-instruct-2507-full-sft-prm-r2egym-swebench-instructions-k5-qwen-only-opus-distill-32k-multiturn
|
||||||
|
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. -->
|
||||||
|
|
||||||
|
# qwen3-4b-instruct-2507-full-sft-prm-r2egym-swebench-instructions-k5-qwen-only-opus-distill-32k-multiturn
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-4B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-4B-Instruct-2507) on the prm_sft_train 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: 5e-06
|
||||||
|
- train_batch_size: 1
|
||||||
|
- eval_batch_size: 8
|
||||||
|
- seed: 42
|
||||||
|
- distributed_type: multi-GPU
|
||||||
|
- num_devices: 8
|
||||||
|
- total_train_batch_size: 8
|
||||||
|
- total_eval_batch_size: 64
|
||||||
|
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.1
|
||||||
|
- num_epochs: 3.0
|
||||||
|
|
||||||
|
### Training results
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- Transformers 5.2.0
|
||||||
|
- Pytorch 2.9.1+cu128
|
||||||
|
- 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": 3.0,
|
||||||
|
"total_flos": 3.143410755085271e+17,
|
||||||
|
"train_loss": 0.2539728307061725,
|
||||||
|
"train_runtime": 24192.1432,
|
||||||
|
"train_samples_per_second": 0.237,
|
||||||
|
"train_steps_per_second": 0.03
|
||||||
|
}
|
||||||
61
chat_template.jinja
Normal file
61
chat_template.jinja
Normal file
@@ -0,0 +1,61 @@
|
|||||||
|
{%- 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 %}
|
||||||
|
{%- 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" %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- 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' }}
|
||||||
|
{%- endif %}
|
||||||
71
config.json
Normal file
71
config.json
Normal file
@@ -0,0 +1,71 @@
|
|||||||
|
{
|
||||||
|
"architectures": [
|
||||||
|
"Qwen3ForCausalLM"
|
||||||
|
],
|
||||||
|
"attention_bias": false,
|
||||||
|
"attention_dropout": 0.0,
|
||||||
|
"bos_token_id": null,
|
||||||
|
"dtype": "float32",
|
||||||
|
"eos_token_id": 151645,
|
||||||
|
"head_dim": 128,
|
||||||
|
"hidden_act": "silu",
|
||||||
|
"hidden_size": 2560,
|
||||||
|
"initializer_range": 0.02,
|
||||||
|
"intermediate_size": 9728,
|
||||||
|
"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",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention",
|
||||||
|
"full_attention"
|
||||||
|
],
|
||||||
|
"max_position_embeddings": 262144,
|
||||||
|
"max_window_layers": 36,
|
||||||
|
"model_type": "qwen3",
|
||||||
|
"num_attention_heads": 32,
|
||||||
|
"num_hidden_layers": 36,
|
||||||
|
"num_key_value_heads": 8,
|
||||||
|
"pad_token_id": 151643,
|
||||||
|
"rms_norm_eps": 1e-06,
|
||||||
|
"rope_parameters": {
|
||||||
|
"rope_theta": 5000000,
|
||||||
|
"rope_type": "default"
|
||||||
|
},
|
||||||
|
"sliding_window": 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": 0.7,
|
||||||
|
"top_k": 20,
|
||||||
|
"top_p": 0.8,
|
||||||
|
"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 sha256:0cd52a1835f8562bbbeeab1ef4cb7e511239a7e33369fe4bad811f73f8a4053a
|
||||||
|
size 17645743048
|
||||||
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": 1010000,
|
||||||
|
"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": 3.0,
|
||||||
|
"total_flos": 3.143410755085271e+17,
|
||||||
|
"train_loss": 0.2539728307061725,
|
||||||
|
"train_runtime": 24192.1432,
|
||||||
|
"train_samples_per_second": 0.237,
|
||||||
|
"train_steps_per_second": 0.03
|
||||||
|
}
|
||||||
77
trainer_log.jsonl
Normal file
77
trainer_log.jsonl
Normal file
@@ -0,0 +1,77 @@
|
|||||||
|
{"current_steps": 10, "total_steps": 720, "loss": 1.3478851318359375, "lr": 6.25e-07, "epoch": 0.041666666666666664, "percentage": 1.39, "elapsed_time": "0:06:44", "remaining_time": "7:58:59"}
|
||||||
|
{"current_steps": 20, "total_steps": 720, "loss": 1.1414000511169433, "lr": 1.3194444444444446e-06, "epoch": 0.08333333333333333, "percentage": 2.78, "elapsed_time": "0:14:36", "remaining_time": "8:31:10"}
|
||||||
|
{"current_steps": 30, "total_steps": 720, "loss": 0.8360883712768554, "lr": 2.0138888888888893e-06, "epoch": 0.125, "percentage": 4.17, "elapsed_time": "0:22:48", "remaining_time": "8:44:34"}
|
||||||
|
{"current_steps": 40, "total_steps": 720, "loss": 0.7460173606872559, "lr": 2.7083333333333334e-06, "epoch": 0.16666666666666666, "percentage": 5.56, "elapsed_time": "0:29:47", "remaining_time": "8:26:29"}
|
||||||
|
{"current_steps": 50, "total_steps": 720, "loss": 0.7003425598144531, "lr": 3.4027777777777783e-06, "epoch": 0.20833333333333334, "percentage": 6.94, "elapsed_time": "0:37:14", "remaining_time": "8:19:02"}
|
||||||
|
{"current_steps": 60, "total_steps": 720, "loss": 0.6781439304351806, "lr": 4.097222222222222e-06, "epoch": 0.25, "percentage": 8.33, "elapsed_time": "0:44:22", "remaining_time": "8:08:07"}
|
||||||
|
{"current_steps": 70, "total_steps": 720, "loss": 0.6273435592651367, "lr": 4.791666666666668e-06, "epoch": 0.2916666666666667, "percentage": 9.72, "elapsed_time": "0:51:44", "remaining_time": "8:00:29"}
|
||||||
|
{"current_steps": 80, "total_steps": 720, "loss": 0.6232743263244629, "lr": 4.998560491796288e-06, "epoch": 0.3333333333333333, "percentage": 11.11, "elapsed_time": "0:59:17", "remaining_time": "7:54:21"}
|
||||||
|
{"current_steps": 90, "total_steps": 720, "loss": 0.5979775428771973, "lr": 4.991513829823945e-06, "epoch": 0.375, "percentage": 12.5, "elapsed_time": "1:07:01", "remaining_time": "7:49:07"}
|
||||||
|
{"current_steps": 100, "total_steps": 720, "loss": 0.5922712326049805, "lr": 4.978612153434527e-06, "epoch": 0.4166666666666667, "percentage": 13.89, "elapsed_time": "1:13:20", "remaining_time": "7:34:40"}
|
||||||
|
{"current_steps": 110, "total_steps": 720, "loss": 0.602515983581543, "lr": 4.9598857813444524e-06, "epoch": 0.4583333333333333, "percentage": 15.28, "elapsed_time": "1:20:50", "remaining_time": "7:28:17"}
|
||||||
|
{"current_steps": 120, "total_steps": 720, "loss": 0.5727059364318847, "lr": 4.935378720204319e-06, "epoch": 0.5, "percentage": 16.67, "elapsed_time": "1:28:22", "remaining_time": "7:21:52"}
|
||||||
|
{"current_steps": 130, "total_steps": 720, "loss": 0.5602549552917481, "lr": 4.905148561184033e-06, "epoch": 0.5416666666666666, "percentage": 18.06, "elapsed_time": "1:36:28", "remaining_time": "7:17:48"}
|
||||||
|
{"current_steps": 140, "total_steps": 720, "loss": 0.5857479095458984, "lr": 4.869266344634556e-06, "epoch": 0.5833333333333334, "percentage": 19.44, "elapsed_time": "1:43:54", "remaining_time": "7:10:29"}
|
||||||
|
{"current_steps": 150, "total_steps": 720, "loss": 0.5755341529846192, "lr": 4.827816393144305e-06, "epoch": 0.625, "percentage": 20.83, "elapsed_time": "1:51:14", "remaining_time": "7:02:44"}
|
||||||
|
{"current_steps": 160, "total_steps": 720, "loss": 0.5534672737121582, "lr": 4.780896113382536e-06, "epoch": 0.6666666666666666, "percentage": 22.22, "elapsed_time": "1:58:59", "remaining_time": "6:56:29"}
|
||||||
|
{"current_steps": 170, "total_steps": 720, "loss": 0.553608751296997, "lr": 4.728615767195347e-06, "epoch": 0.7083333333333334, "percentage": 23.61, "elapsed_time": "2:06:32", "remaining_time": "6:49:25"}
|
||||||
|
{"current_steps": 180, "total_steps": 720, "loss": 0.5626870155334472, "lr": 4.671098212492251e-06, "epoch": 0.75, "percentage": 25.0, "elapsed_time": "2:14:28", "remaining_time": "6:43:24"}
|
||||||
|
{"current_steps": 190, "total_steps": 720, "loss": 0.5459296226501464, "lr": 4.608478614532215e-06, "epoch": 0.7916666666666666, "percentage": 26.39, "elapsed_time": "2:22:07", "remaining_time": "6:36:25"}
|
||||||
|
{"current_steps": 200, "total_steps": 720, "loss": 0.5788500785827637, "lr": 4.54090412828763e-06, "epoch": 0.8333333333333334, "percentage": 27.78, "elapsed_time": "2:29:27", "remaining_time": "6:28:35"}
|
||||||
|
{"current_steps": 210, "total_steps": 720, "loss": 0.5325177192687989, "lr": 4.468533552632681e-06, "epoch": 0.875, "percentage": 29.17, "elapsed_time": "2:39:25", "remaining_time": "6:27:09"}
|
||||||
|
{"current_steps": 220, "total_steps": 720, "loss": 0.5544806003570557, "lr": 4.391536957168733e-06, "epoch": 0.9166666666666666, "percentage": 30.56, "elapsed_time": "2:47:28", "remaining_time": "6:20:36"}
|
||||||
|
{"current_steps": 230, "total_steps": 720, "loss": 0.5456794261932373, "lr": 4.3100952825637064e-06, "epoch": 0.9583333333333334, "percentage": 31.94, "elapsed_time": "2:55:36", "remaining_time": "6:14:06"}
|
||||||
|
{"current_steps": 240, "total_steps": 720, "loss": 0.5358950138092041, "lr": 4.2243999153446445e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "3:03:01", "remaining_time": "6:06:02"}
|
||||||
|
{"current_steps": 210, "total_steps": 720, "loss": 0.5325778007507325, "lr": 4.468533552632681e-06, "epoch": 0.875, "percentage": 29.17, "elapsed_time": "0:08:16", "remaining_time": "0:20:06"}
|
||||||
|
{"current_steps": 220, "total_steps": 720, "loss": 0.5544434547424316, "lr": 4.391536957168733e-06, "epoch": 0.9166666666666666, "percentage": 30.56, "elapsed_time": "0:16:19", "remaining_time": "0:37:07"}
|
||||||
|
{"current_steps": 230, "total_steps": 720, "loss": 0.5456916332244873, "lr": 4.3100952825637064e-06, "epoch": 0.9583333333333334, "percentage": 31.94, "elapsed_time": "0:24:28", "remaining_time": "0:52:08"}
|
||||||
|
{"current_steps": 240, "total_steps": 720, "loss": 0.535283899307251, "lr": 4.2243999153446445e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "0:31:53", "remaining_time": "1:03:46"}
|
||||||
|
{"current_steps": 250, "total_steps": 720, "loss": 0.4393170356750488, "lr": 4.134652238142674e-06, "epoch": 1.0416666666666667, "percentage": 34.72, "elapsed_time": "0:39:59", "remaining_time": "1:15:11"}
|
||||||
|
{"current_steps": 260, "total_steps": 720, "loss": 0.4061892032623291, "lr": 4.041063156447314e-06, "epoch": 1.0833333333333333, "percentage": 36.11, "elapsed_time": "0:47:07", "remaining_time": "1:23:22"}
|
||||||
|
{"current_steps": 270, "total_steps": 720, "loss": 0.413454532623291, "lr": 3.943852602982213e-06, "epoch": 1.125, "percentage": 37.5, "elapsed_time": "0:55:06", "remaining_time": "1:31:50"}
|
||||||
|
{"current_steps": 280, "total_steps": 720, "loss": 0.41506023406982423, "lr": 3.8432490208670605e-06, "epoch": 1.1666666666666667, "percentage": 38.89, "elapsed_time": "1:02:31", "remaining_time": "1:38:15"}
|
||||||
|
{"current_steps": 290, "total_steps": 720, "loss": 0.3757039785385132, "lr": 3.739488826780199e-06, "epoch": 1.2083333333333333, "percentage": 40.28, "elapsed_time": "1:10:27", "remaining_time": "1:44:29"}
|
||||||
|
{"current_steps": 300, "total_steps": 720, "loss": 0.38776512145996095, "lr": 3.6328158553835206e-06, "epoch": 1.25, "percentage": 41.67, "elapsed_time": "1:17:30", "remaining_time": "1:48:31"}
|
||||||
|
{"current_steps": 310, "total_steps": 720, "loss": 0.3929452896118164, "lr": 3.5234807863152316e-06, "epoch": 1.2916666666666667, "percentage": 43.06, "elapsed_time": "1:25:43", "remaining_time": "1:53:23"}
|
||||||
|
{"current_steps": 320, "total_steps": 720, "loss": 0.41544904708862307, "lr": 3.41174055509703e-06, "epoch": 1.3333333333333333, "percentage": 44.44, "elapsed_time": "1:33:38", "remaining_time": "1:57:03"}
|
||||||
|
{"current_steps": 330, "total_steps": 720, "loss": 0.4201189517974854, "lr": 3.2978577493400775e-06, "epoch": 1.375, "percentage": 45.83, "elapsed_time": "1:41:33", "remaining_time": "2:00:01"}
|
||||||
|
{"current_steps": 340, "total_steps": 720, "loss": 0.3872462272644043, "lr": 3.182099991668653e-06, "epoch": 1.4166666666666667, "percentage": 47.22, "elapsed_time": "1:48:51", "remaining_time": "2:01:40"}
|
||||||
|
{"current_steps": 350, "total_steps": 720, "loss": 0.3856550931930542, "lr": 3.064739310811612e-06, "epoch": 1.4583333333333333, "percentage": 48.61, "elapsed_time": "1:56:03", "remaining_time": "2:02:41"}
|
||||||
|
{"current_steps": 360, "total_steps": 720, "loss": 0.4167680263519287, "lr": 2.9460515023395807e-06, "epoch": 1.5, "percentage": 50.0, "elapsed_time": "2:04:12", "remaining_time": "2:04:12"}
|
||||||
|
{"current_steps": 370, "total_steps": 720, "loss": 0.40921745300292967, "lr": 2.82631548055013e-06, "epoch": 1.5416666666666665, "percentage": 51.39, "elapsed_time": "2:10:18", "remaining_time": "2:03:15"}
|
||||||
|
{"current_steps": 380, "total_steps": 720, "loss": 0.39605083465576174, "lr": 2.705812623024e-06, "epoch": 1.5833333333333335, "percentage": 52.78, "elapsed_time": "2:17:49", "remaining_time": "2:03:18"}
|
||||||
|
{"current_steps": 390, "total_steps": 720, "loss": 0.4075204372406006, "lr": 2.5848261093926565e-06, "epoch": 1.625, "percentage": 54.17, "elapsed_time": "2:25:45", "remaining_time": "2:03:19"}
|
||||||
|
{"current_steps": 400, "total_steps": 720, "loss": 0.40534563064575196, "lr": 2.4636402558710434e-06, "epoch": 1.6666666666666665, "percentage": 55.56, "elapsed_time": "2:33:40", "remaining_time": "2:02:56"}
|
||||||
|
{"current_steps": 410, "total_steps": 720, "loss": 0.39900033473968505, "lr": 2.3425398471194056e-06, "epoch": 1.7083333333333335, "percentage": 56.94, "elapsed_time": "2:44:10", "remaining_time": "2:04:08"}
|
||||||
|
{"current_steps": 420, "total_steps": 720, "loss": 0.40346851348876955, "lr": 2.221809467004251e-06, "epoch": 1.75, "percentage": 58.33, "elapsed_time": "2:51:16", "remaining_time": "2:02:20"}
|
||||||
|
{"current_steps": 430, "total_steps": 720, "loss": 0.4041299819946289, "lr": 2.101732829831194e-06, "epoch": 1.7916666666666665, "percentage": 59.72, "elapsed_time": "2:58:58", "remaining_time": "2:00:41"}
|
||||||
|
{"current_steps": 440, "total_steps": 720, "loss": 0.3829362392425537, "lr": 1.9825921136212372e-06, "epoch": 1.8333333333333335, "percentage": 61.11, "elapsed_time": "3:07:14", "remaining_time": "1:59:09"}
|
||||||
|
{"current_steps": 450, "total_steps": 720, "loss": 0.41110787391662595, "lr": 1.8646672969973034e-06, "epoch": 1.875, "percentage": 62.5, "elapsed_time": "3:15:22", "remaining_time": "1:57:13"}
|
||||||
|
{"current_steps": 460, "total_steps": 720, "loss": 0.376175594329834, "lr": 1.7482355012393177e-06, "epoch": 1.9166666666666665, "percentage": 63.89, "elapsed_time": "3:22:44", "remaining_time": "1:54:35"}
|
||||||
|
{"current_steps": 470, "total_steps": 720, "loss": 0.3938215017318726, "lr": 1.633570339053982e-06, "epoch": 1.9583333333333335, "percentage": 65.28, "elapsed_time": "3:30:17", "remaining_time": "1:51:51"}
|
||||||
|
{"current_steps": 480, "total_steps": 720, "loss": 0.41222400665283204, "lr": 1.5209412715896527e-06, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "3:38:21", "remaining_time": "1:49:10"}
|
||||||
|
{"current_steps": 490, "total_steps": 720, "loss": 0.2820391893386841, "lr": 1.4106129752073023e-06, "epoch": 2.0416666666666665, "percentage": 68.06, "elapsed_time": "3:44:41", "remaining_time": "1:45:27"}
|
||||||
|
{"current_steps": 500, "total_steps": 720, "loss": 0.2755776882171631, "lr": 1.3028447194956413e-06, "epoch": 2.0833333333333335, "percentage": 69.44, "elapsed_time": "3:51:56", "remaining_time": "1:42:03"}
|
||||||
|
{"current_steps": 510, "total_steps": 720, "loss": 0.26791958808898925, "lr": 1.197889757992076e-06, "epoch": 2.125, "percentage": 70.83, "elapsed_time": "3:59:26", "remaining_time": "1:38:35"}
|
||||||
|
{"current_steps": 520, "total_steps": 720, "loss": 0.27331366539001467, "lr": 1.0959947330412681e-06, "epoch": 2.1666666666666665, "percentage": 72.22, "elapsed_time": "4:07:09", "remaining_time": "1:35:03"}
|
||||||
|
{"current_steps": 530, "total_steps": 720, "loss": 0.27553110122680663, "lr": 9.973990961898928e-07, "epoch": 2.2083333333333335, "percentage": 73.61, "elapsed_time": "4:15:22", "remaining_time": "1:31:32"}
|
||||||
|
{"current_steps": 540, "total_steps": 720, "loss": 0.2681769132614136, "lr": 9.023345454796459e-07, "epoch": 2.25, "percentage": 75.0, "elapsed_time": "4:22:50", "remaining_time": "1:27:36"}
|
||||||
|
{"current_steps": 550, "total_steps": 720, "loss": 0.25668916702270506, "lr": 8.110244809608494e-07, "epoch": 2.2916666666666665, "percentage": 76.39, "elapsed_time": "4:30:01", "remaining_time": "1:23:27"}
|
||||||
|
{"current_steps": 560, "total_steps": 720, "loss": 0.27220232486724855, "lr": 7.236834797061979e-07, "epoch": 2.3333333333333335, "percentage": 77.78, "elapsed_time": "4:37:28", "remaining_time": "1:19:16"}
|
||||||
|
{"current_steps": 570, "total_steps": 720, "loss": 0.28708949089050295, "lr": 6.405167915583477e-07, "epoch": 2.375, "percentage": 79.17, "elapsed_time": "4:45:13", "remaining_time": "1:15:03"}
|
||||||
|
{"current_steps": 580, "total_steps": 720, "loss": 0.2636965274810791, "lr": 5.617198567963353e-07, "epoch": 2.4166666666666665, "percentage": 80.56, "elapsed_time": "4:52:33", "remaining_time": "1:10:37"}
|
||||||
|
{"current_steps": 590, "total_steps": 720, "loss": 0.2697884559631348, "lr": 4.874778468543043e-07, "epoch": 2.4583333333333335, "percentage": 81.94, "elapsed_time": "5:00:38", "remaining_time": "1:06:14"}
|
||||||
|
{"current_steps": 600, "total_steps": 720, "loss": 0.2729990005493164, "lr": 4.1796522917183226e-07, "epoch": 2.5, "percentage": 83.33, "elapsed_time": "5:08:34", "remaining_time": "1:01:42"}
|
||||||
|
{"current_steps": 610, "total_steps": 720, "loss": 0.2704613208770752, "lr": 3.5334535719846767e-07, "epoch": 2.5416666666666665, "percentage": 84.72, "elapsed_time": "5:19:12", "remaining_time": "0:57:33"}
|
||||||
|
{"current_steps": 620, "total_steps": 720, "loss": 0.26971731185913084, "lr": 2.9377008651594737e-07, "epoch": 2.5833333333333335, "percentage": 86.11, "elapsed_time": "5:26:26", "remaining_time": "0:52:39"}
|
||||||
|
{"current_steps": 630, "total_steps": 720, "loss": 0.266050910949707, "lr": 2.393794179802053e-07, "epoch": 2.625, "percentage": 87.5, "elapsed_time": "5:33:27", "remaining_time": "0:47:38"}
|
||||||
|
{"current_steps": 640, "total_steps": 720, "loss": 0.27685208320617677, "lr": 1.9030116872178317e-07, "epoch": 2.6666666666666665, "percentage": 88.89, "elapsed_time": "5:40:03", "remaining_time": "0:42:30"}
|
||||||
|
{"current_steps": 650, "total_steps": 720, "loss": 0.2522747993469238, "lr": 1.4665067177778187e-07, "epoch": 2.7083333333333335, "percentage": 90.28, "elapsed_time": "5:47:50", "remaining_time": "0:37:27"}
|
||||||
|
{"current_steps": 660, "total_steps": 720, "loss": 0.25966610908508303, "lr": 1.0853050506121927e-07, "epoch": 2.75, "percentage": 91.67, "elapsed_time": "5:55:01", "remaining_time": "0:32:16"}
|
||||||
|
{"current_steps": 670, "total_steps": 720, "loss": 0.27193603515625, "lr": 7.603025030471001e-08, "epoch": 2.7916666666666665, "percentage": 93.06, "elapsed_time": "6:02:51", "remaining_time": "0:27:04"}
|
||||||
|
{"current_steps": 680, "total_steps": 720, "loss": 0.27923831939697263, "lr": 4.922628254493994e-08, "epoch": 2.8333333333333335, "percentage": 94.44, "elapsed_time": "6:10:16", "remaining_time": "0:21:46"}
|
||||||
|
{"current_steps": 690, "total_steps": 720, "loss": 0.26351704597473147, "lr": 2.81815906426533e-08, "epoch": 2.875, "percentage": 95.83, "elapsed_time": "6:18:03", "remaining_time": "0:16:26"}
|
||||||
|
{"current_steps": 700, "total_steps": 720, "loss": 0.25306267738342286, "lr": 1.2945629259917547e-08, "epoch": 2.9166666666666665, "percentage": 97.22, "elapsed_time": "6:25:46", "remaining_time": "0:11:01"}
|
||||||
|
{"current_steps": 710, "total_steps": 720, "loss": 0.26765923500061034, "lr": 3.5542026425233746e-09, "epoch": 2.9583333333333335, "percentage": 98.61, "elapsed_time": "6:33:28", "remaining_time": "0:05:32"}
|
||||||
|
{"current_steps": 720, "total_steps": 720, "loss": 0.26591691970825193, "lr": 2.938048062617327e-11, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "6:40:39", "remaining_time": "0:00:00"}
|
||||||
|
{"current_steps": 720, "total_steps": 720, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "6:43:12", "remaining_time": "0:00:00"}
|
||||||
547
trainer_state.json
Normal file
547
trainer_state.json
Normal file
@@ -0,0 +1,547 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 3.0,
|
||||||
|
"eval_steps": 500,
|
||||||
|
"global_step": 720,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"epoch": 0.041666666666666664,
|
||||||
|
"grad_norm": 29.499774932861328,
|
||||||
|
"learning_rate": 6.25e-07,
|
||||||
|
"loss": 1.3478851318359375,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.08333333333333333,
|
||||||
|
"grad_norm": 10.380602836608887,
|
||||||
|
"learning_rate": 1.3194444444444446e-06,
|
||||||
|
"loss": 1.1414000511169433,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.125,
|
||||||
|
"grad_norm": 4.048536777496338,
|
||||||
|
"learning_rate": 2.0138888888888893e-06,
|
||||||
|
"loss": 0.8360883712768554,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.16666666666666666,
|
||||||
|
"grad_norm": 3.7858238220214844,
|
||||||
|
"learning_rate": 2.7083333333333334e-06,
|
||||||
|
"loss": 0.7460173606872559,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.20833333333333334,
|
||||||
|
"grad_norm": 3.232109546661377,
|
||||||
|
"learning_rate": 3.4027777777777783e-06,
|
||||||
|
"loss": 0.7003425598144531,
|
||||||
|
"step": 50
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.25,
|
||||||
|
"grad_norm": 3.959477424621582,
|
||||||
|
"learning_rate": 4.097222222222222e-06,
|
||||||
|
"loss": 0.6781439304351806,
|
||||||
|
"step": 60
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.2916666666666667,
|
||||||
|
"grad_norm": 3.6397626399993896,
|
||||||
|
"learning_rate": 4.791666666666668e-06,
|
||||||
|
"loss": 0.6273435592651367,
|
||||||
|
"step": 70
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.3333333333333333,
|
||||||
|
"grad_norm": 3.215730905532837,
|
||||||
|
"learning_rate": 4.998560491796288e-06,
|
||||||
|
"loss": 0.6232743263244629,
|
||||||
|
"step": 80
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.375,
|
||||||
|
"grad_norm": 3.3890368938446045,
|
||||||
|
"learning_rate": 4.991513829823945e-06,
|
||||||
|
"loss": 0.5979775428771973,
|
||||||
|
"step": 90
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4166666666666667,
|
||||||
|
"grad_norm": 3.129528522491455,
|
||||||
|
"learning_rate": 4.978612153434527e-06,
|
||||||
|
"loss": 0.5922712326049805,
|
||||||
|
"step": 100
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4583333333333333,
|
||||||
|
"grad_norm": 3.7336618900299072,
|
||||||
|
"learning_rate": 4.9598857813444524e-06,
|
||||||
|
"loss": 0.602515983581543,
|
||||||
|
"step": 110
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.5,
|
||||||
|
"grad_norm": 3.025545120239258,
|
||||||
|
"learning_rate": 4.935378720204319e-06,
|
||||||
|
"loss": 0.5727059364318847,
|
||||||
|
"step": 120
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.5416666666666666,
|
||||||
|
"grad_norm": 3.025608539581299,
|
||||||
|
"learning_rate": 4.905148561184033e-06,
|
||||||
|
"loss": 0.5602549552917481,
|
||||||
|
"step": 130
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.5833333333333334,
|
||||||
|
"grad_norm": 3.1320998668670654,
|
||||||
|
"learning_rate": 4.869266344634556e-06,
|
||||||
|
"loss": 0.5857479095458984,
|
||||||
|
"step": 140
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.625,
|
||||||
|
"grad_norm": 2.8681347370147705,
|
||||||
|
"learning_rate": 4.827816393144305e-06,
|
||||||
|
"loss": 0.5755341529846192,
|
||||||
|
"step": 150
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"grad_norm": 3.168527841567993,
|
||||||
|
"learning_rate": 4.780896113382536e-06,
|
||||||
|
"loss": 0.5534672737121582,
|
||||||
|
"step": 160
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.7083333333333334,
|
||||||
|
"grad_norm": 2.962265729904175,
|
||||||
|
"learning_rate": 4.728615767195347e-06,
|
||||||
|
"loss": 0.553608751296997,
|
||||||
|
"step": 170
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.75,
|
||||||
|
"grad_norm": 3.3806610107421875,
|
||||||
|
"learning_rate": 4.671098212492251e-06,
|
||||||
|
"loss": 0.5626870155334472,
|
||||||
|
"step": 180
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.7916666666666666,
|
||||||
|
"grad_norm": 3.1959733963012695,
|
||||||
|
"learning_rate": 4.608478614532215e-06,
|
||||||
|
"loss": 0.5459296226501464,
|
||||||
|
"step": 190
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8333333333333334,
|
||||||
|
"grad_norm": 2.7560014724731445,
|
||||||
|
"learning_rate": 4.54090412828763e-06,
|
||||||
|
"loss": 0.5788500785827637,
|
||||||
|
"step": 200
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.875,
|
||||||
|
"grad_norm": 2.802913188934326,
|
||||||
|
"learning_rate": 4.468533552632681e-06,
|
||||||
|
"loss": 0.5325778007507325,
|
||||||
|
"step": 210
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.9166666666666666,
|
||||||
|
"grad_norm": 3.1139848232269287,
|
||||||
|
"learning_rate": 4.391536957168733e-06,
|
||||||
|
"loss": 0.5544434547424316,
|
||||||
|
"step": 220
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.9583333333333334,
|
||||||
|
"grad_norm": 2.9709322452545166,
|
||||||
|
"learning_rate": 4.3100952825637064e-06,
|
||||||
|
"loss": 0.5456916332244873,
|
||||||
|
"step": 230
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 2.8750693798065186,
|
||||||
|
"learning_rate": 4.2243999153446445e-06,
|
||||||
|
"loss": 0.535283899307251,
|
||||||
|
"step": 240
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0416666666666667,
|
||||||
|
"grad_norm": 2.816533327102661,
|
||||||
|
"learning_rate": 4.134652238142674e-06,
|
||||||
|
"loss": 0.4393170356750488,
|
||||||
|
"step": 250
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0833333333333333,
|
||||||
|
"grad_norm": 3.158930540084839,
|
||||||
|
"learning_rate": 4.041063156447314e-06,
|
||||||
|
"loss": 0.4061892032623291,
|
||||||
|
"step": 260
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.125,
|
||||||
|
"grad_norm": 3.3394505977630615,
|
||||||
|
"learning_rate": 3.943852602982213e-06,
|
||||||
|
"loss": 0.413454532623291,
|
||||||
|
"step": 270
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.1666666666666667,
|
||||||
|
"grad_norm": 3.0901780128479004,
|
||||||
|
"learning_rate": 3.8432490208670605e-06,
|
||||||
|
"loss": 0.41506023406982423,
|
||||||
|
"step": 280
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.2083333333333333,
|
||||||
|
"grad_norm": 3.2386937141418457,
|
||||||
|
"learning_rate": 3.739488826780199e-06,
|
||||||
|
"loss": 0.3757039785385132,
|
||||||
|
"step": 290
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.25,
|
||||||
|
"grad_norm": 3.1895928382873535,
|
||||||
|
"learning_rate": 3.6328158553835206e-06,
|
||||||
|
"loss": 0.38776512145996095,
|
||||||
|
"step": 300
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.2916666666666667,
|
||||||
|
"grad_norm": 3.395063638687134,
|
||||||
|
"learning_rate": 3.5234807863152316e-06,
|
||||||
|
"loss": 0.3929452896118164,
|
||||||
|
"step": 310
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.3333333333333333,
|
||||||
|
"grad_norm": 2.6497585773468018,
|
||||||
|
"learning_rate": 3.41174055509703e-06,
|
||||||
|
"loss": 0.41544904708862307,
|
||||||
|
"step": 320
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.375,
|
||||||
|
"grad_norm": 3.4338808059692383,
|
||||||
|
"learning_rate": 3.2978577493400775e-06,
|
||||||
|
"loss": 0.4201189517974854,
|
||||||
|
"step": 330
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.4166666666666667,
|
||||||
|
"grad_norm": 5.390817642211914,
|
||||||
|
"learning_rate": 3.182099991668653e-06,
|
||||||
|
"loss": 0.3872462272644043,
|
||||||
|
"step": 340
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.4583333333333333,
|
||||||
|
"grad_norm": 3.0678157806396484,
|
||||||
|
"learning_rate": 3.064739310811612e-06,
|
||||||
|
"loss": 0.3856550931930542,
|
||||||
|
"step": 350
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.5,
|
||||||
|
"grad_norm": 2.928752899169922,
|
||||||
|
"learning_rate": 2.9460515023395807e-06,
|
||||||
|
"loss": 0.4167680263519287,
|
||||||
|
"step": 360
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.5416666666666665,
|
||||||
|
"grad_norm": 3.0667483806610107,
|
||||||
|
"learning_rate": 2.82631548055013e-06,
|
||||||
|
"loss": 0.40921745300292967,
|
||||||
|
"step": 370
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.5833333333333335,
|
||||||
|
"grad_norm": 3.30924391746521,
|
||||||
|
"learning_rate": 2.705812623024e-06,
|
||||||
|
"loss": 0.39605083465576174,
|
||||||
|
"step": 380
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.625,
|
||||||
|
"grad_norm": 3.206585168838501,
|
||||||
|
"learning_rate": 2.5848261093926565e-06,
|
||||||
|
"loss": 0.4075204372406006,
|
||||||
|
"step": 390
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.6666666666666665,
|
||||||
|
"grad_norm": 3.0673396587371826,
|
||||||
|
"learning_rate": 2.4636402558710434e-06,
|
||||||
|
"loss": 0.40534563064575196,
|
||||||
|
"step": 400
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.7083333333333335,
|
||||||
|
"grad_norm": 3.4459261894226074,
|
||||||
|
"learning_rate": 2.3425398471194056e-06,
|
||||||
|
"loss": 0.39900033473968505,
|
||||||
|
"step": 410
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.75,
|
||||||
|
"grad_norm": 3.114468574523926,
|
||||||
|
"learning_rate": 2.221809467004251e-06,
|
||||||
|
"loss": 0.40346851348876955,
|
||||||
|
"step": 420
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.7916666666666665,
|
||||||
|
"grad_norm": 3.425462007522583,
|
||||||
|
"learning_rate": 2.101732829831194e-06,
|
||||||
|
"loss": 0.4041299819946289,
|
||||||
|
"step": 430
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.8333333333333335,
|
||||||
|
"grad_norm": 3.256767511367798,
|
||||||
|
"learning_rate": 1.9825921136212372e-06,
|
||||||
|
"loss": 0.3829362392425537,
|
||||||
|
"step": 440
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.875,
|
||||||
|
"grad_norm": 3.315276861190796,
|
||||||
|
"learning_rate": 1.8646672969973034e-06,
|
||||||
|
"loss": 0.41110787391662595,
|
||||||
|
"step": 450
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.9166666666666665,
|
||||||
|
"grad_norm": 3.0989630222320557,
|
||||||
|
"learning_rate": 1.7482355012393177e-06,
|
||||||
|
"loss": 0.376175594329834,
|
||||||
|
"step": 460
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.9583333333333335,
|
||||||
|
"grad_norm": 3.8257453441619873,
|
||||||
|
"learning_rate": 1.633570339053982e-06,
|
||||||
|
"loss": 0.3938215017318726,
|
||||||
|
"step": 470
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.0,
|
||||||
|
"grad_norm": 2.900068998336792,
|
||||||
|
"learning_rate": 1.5209412715896527e-06,
|
||||||
|
"loss": 0.41222400665283204,
|
||||||
|
"step": 480
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.0416666666666665,
|
||||||
|
"grad_norm": 2.951523780822754,
|
||||||
|
"learning_rate": 1.4106129752073023e-06,
|
||||||
|
"loss": 0.2820391893386841,
|
||||||
|
"step": 490
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.0833333333333335,
|
||||||
|
"grad_norm": 3.1139001846313477,
|
||||||
|
"learning_rate": 1.3028447194956413e-06,
|
||||||
|
"loss": 0.2755776882171631,
|
||||||
|
"step": 500
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.125,
|
||||||
|
"grad_norm": 3.254406452178955,
|
||||||
|
"learning_rate": 1.197889757992076e-06,
|
||||||
|
"loss": 0.26791958808898925,
|
||||||
|
"step": 510
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.1666666666666665,
|
||||||
|
"grad_norm": 3.350184917449951,
|
||||||
|
"learning_rate": 1.0959947330412681e-06,
|
||||||
|
"loss": 0.27331366539001467,
|
||||||
|
"step": 520
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.2083333333333335,
|
||||||
|
"grad_norm": 3.6832144260406494,
|
||||||
|
"learning_rate": 9.973990961898928e-07,
|
||||||
|
"loss": 0.27553110122680663,
|
||||||
|
"step": 530
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.25,
|
||||||
|
"grad_norm": 3.2515552043914795,
|
||||||
|
"learning_rate": 9.023345454796459e-07,
|
||||||
|
"loss": 0.2681769132614136,
|
||||||
|
"step": 540
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.2916666666666665,
|
||||||
|
"grad_norm": 3.410749673843384,
|
||||||
|
"learning_rate": 8.110244809608494e-07,
|
||||||
|
"loss": 0.25668916702270506,
|
||||||
|
"step": 550
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.3333333333333335,
|
||||||
|
"grad_norm": 3.052957773208618,
|
||||||
|
"learning_rate": 7.236834797061979e-07,
|
||||||
|
"loss": 0.27220232486724855,
|
||||||
|
"step": 560
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.375,
|
||||||
|
"grad_norm": 4.0483808517456055,
|
||||||
|
"learning_rate": 6.405167915583477e-07,
|
||||||
|
"loss": 0.28708949089050295,
|
||||||
|
"step": 570
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.4166666666666665,
|
||||||
|
"grad_norm": 3.3622629642486572,
|
||||||
|
"learning_rate": 5.617198567963353e-07,
|
||||||
|
"loss": 0.2636965274810791,
|
||||||
|
"step": 580
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.4583333333333335,
|
||||||
|
"grad_norm": 3.524901866912842,
|
||||||
|
"learning_rate": 4.874778468543043e-07,
|
||||||
|
"loss": 0.2697884559631348,
|
||||||
|
"step": 590
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.5,
|
||||||
|
"grad_norm": 3.72314453125,
|
||||||
|
"learning_rate": 4.1796522917183226e-07,
|
||||||
|
"loss": 0.2729990005493164,
|
||||||
|
"step": 600
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.5416666666666665,
|
||||||
|
"grad_norm": 3.488933563232422,
|
||||||
|
"learning_rate": 3.5334535719846767e-07,
|
||||||
|
"loss": 0.2704613208770752,
|
||||||
|
"step": 610
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.5833333333333335,
|
||||||
|
"grad_norm": 3.4488272666931152,
|
||||||
|
"learning_rate": 2.9377008651594737e-07,
|
||||||
|
"loss": 0.26971731185913084,
|
||||||
|
"step": 620
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.625,
|
||||||
|
"grad_norm": 3.6721582412719727,
|
||||||
|
"learning_rate": 2.393794179802053e-07,
|
||||||
|
"loss": 0.266050910949707,
|
||||||
|
"step": 630
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.6666666666666665,
|
||||||
|
"grad_norm": 3.4804670810699463,
|
||||||
|
"learning_rate": 1.9030116872178317e-07,
|
||||||
|
"loss": 0.27685208320617677,
|
||||||
|
"step": 640
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.7083333333333335,
|
||||||
|
"grad_norm": 3.309819221496582,
|
||||||
|
"learning_rate": 1.4665067177778187e-07,
|
||||||
|
"loss": 0.2522747993469238,
|
||||||
|
"step": 650
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.75,
|
||||||
|
"grad_norm": 3.369265079498291,
|
||||||
|
"learning_rate": 1.0853050506121927e-07,
|
||||||
|
"loss": 0.25966610908508303,
|
||||||
|
"step": 660
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.7916666666666665,
|
||||||
|
"grad_norm": 3.238943099975586,
|
||||||
|
"learning_rate": 7.603025030471001e-08,
|
||||||
|
"loss": 0.27193603515625,
|
||||||
|
"step": 670
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.8333333333333335,
|
||||||
|
"grad_norm": 3.4344398975372314,
|
||||||
|
"learning_rate": 4.922628254493994e-08,
|
||||||
|
"loss": 0.27923831939697263,
|
||||||
|
"step": 680
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.875,
|
||||||
|
"grad_norm": 3.205024480819702,
|
||||||
|
"learning_rate": 2.81815906426533e-08,
|
||||||
|
"loss": 0.26351704597473147,
|
||||||
|
"step": 690
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.9166666666666665,
|
||||||
|
"grad_norm": 3.3555197715759277,
|
||||||
|
"learning_rate": 1.2945629259917547e-08,
|
||||||
|
"loss": 0.25306267738342286,
|
||||||
|
"step": 700
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.9583333333333335,
|
||||||
|
"grad_norm": 3.810171365737915,
|
||||||
|
"learning_rate": 3.5542026425233746e-09,
|
||||||
|
"loss": 0.26765923500061034,
|
||||||
|
"step": 710
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 3.0,
|
||||||
|
"grad_norm": 3.211577892303467,
|
||||||
|
"learning_rate": 2.938048062617327e-11,
|
||||||
|
"loss": 0.26591691970825193,
|
||||||
|
"step": 720
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 3.0,
|
||||||
|
"step": 720,
|
||||||
|
"total_flos": 3.143410755085271e+17,
|
||||||
|
"train_loss": 0.2539728307061725,
|
||||||
|
"train_runtime": 24192.1432,
|
||||||
|
"train_samples_per_second": 0.237,
|
||||||
|
"train_steps_per_second": 0.03
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 10,
|
||||||
|
"max_steps": 720,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 3,
|
||||||
|
"save_steps": 200,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": true
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 3.143410755085271e+17,
|
||||||
|
"train_batch_size": 1,
|
||||||
|
"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 sha256:9777cd03d8594b153207002c0e704d228de181f0657a62357a36baacf8e5a030
|
||||||
|
size 5841
|
||||||
BIN
training_loss.png
Normal file
BIN
training_loss.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 39 KiB |
Reference in New Issue
Block a user