初始化项目,由ModelHub XC社区提供模型

Model: Bykot/rugpt3small_minhash_dedup_08
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-01 06:21:22 +08:00
commit f53754e11d
11 changed files with 250686 additions and 0 deletions

View File

@@ -0,0 +1,2 @@
experiment,method,source_docs,cleaned_docs,removed_docs,removed_percent,created_at_utc
minhash_dedup_08,minhash_lsh,198000,197775,225,0.11363636363636363,2026-05-12T01:13:07.790224+00:00
1 experiment method source_docs cleaned_docs removed_docs removed_percent created_at_utc
2 minhash_dedup_08 minhash_lsh 198000 197775 225 0.11363636363636363 2026-05-12T01:13:07.790224+00:00

View File

@@ -0,0 +1,11 @@
{
"experiment": "minhash_dedup_08",
"method": "minhash_lsh",
"base_dataset_repo": "Bykot/c4_ru_200k_split",
"source_train_docs": 198000,
"cleaned_docs": 197775,
"removed_docs": 225,
"removed_percent": 0.11363636363636363,
"eval_docs": 2000,
"created_at_utc": "2026-05-12T01:13:09.279538+00:00"
}

View File

@@ -0,0 +1,2 @@
experiment,method,model_name,base_dataset_repo,train_docs,removed_docs,eval_docs,eval_loss_before,perplexity_before,eval_loss_after,perplexity_after,train_runtime,train_samples_per_second,created_at_utc
minhash_dedup_08,minhash_lsh,ai-forever/rugpt3small_based_on_gpt2,Bykot/c4_ru_200k_split,197775,225,2000,3.8492491245269775,46.95779053735424,3.0526859760284424,21.172135967712943,19308.2323,10.243,2026-05-12T06:41:45.751863+00:00
1 experiment method model_name base_dataset_repo train_docs removed_docs eval_docs eval_loss_before perplexity_before eval_loss_after perplexity_after train_runtime train_samples_per_second created_at_utc
2 minhash_dedup_08 minhash_lsh ai-forever/rugpt3small_based_on_gpt2 Bykot/c4_ru_200k_split 197775 225 2000 3.8492491245269775 46.95779053735424 3.0526859760284424 21.172135967712943 19308.2323 10.243 2026-05-12T06:41:45.751863+00:00

View File

@@ -0,0 +1,16 @@
{
"experiment": "minhash_dedup_08",
"method": "minhash_lsh",
"model_name": "ai-forever/rugpt3small_based_on_gpt2",
"base_dataset_repo": "Bykot/c4_ru_200k_split",
"train_docs": 197775,
"removed_docs": 225,
"eval_docs": 2000,
"eval_loss_before": 3.8492491245269775,
"perplexity_before": 46.95779053735424,
"eval_loss_after": 3.0526859760284424,
"perplexity_after": 21.172135967712943,
"train_runtime": 19308.2323,
"train_samples_per_second": 10.243,
"created_at_utc": "2026-05-12T06:41:45.751863+00:00"
}