r3 test results

.gitattributes

@@ -70,3 +70,31 @@ llama-factory/data/alpaca_mgtv_p1_en.json filter=lfs diff=lfs merge=lfs -text
 llama-factory/data/alpaca_mgtv_p2_en.json filter=lfs diff=lfs merge=lfs -text
 datasets/mgtv/test_b.csv filter=lfs diff=lfs merge=lfs -text
 results/test_b-results.csv filter=lfs diff=lfs merge=lfs -text
+results/test_b-results_r3.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p1_en_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p2_r2.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_internlm_4bit_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_m3.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_p2_full_r2.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_p2_r2_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p2_r3.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p1_r3_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p2_r2_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-glm-4-9b_p2_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_p2_full_r3.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_p2_r3_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/glm-4-9b_lora_sft_bf16-p2.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p1_r2.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p1_r2_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p2_en_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_internlm_4bit.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-glm-4-9b_p1_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p1_r4_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-llama3_p2_r3_full_metrics.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p1_r3.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p1_r4.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p2_en.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_internlm_nv4090.csv filter=lfs diff=lfs merge=lfs -text
+results/glm-4-9b_lora_sft_bf16-p1.csv filter=lfs diff=lfs merge=lfs -text
+results/llama3-8b_lora_sft_bf16-p1_en.csv filter=lfs diff=lfs merge=lfs -text
+results/mgtv-results_internlm_best.csv filter=lfs diff=lfs merge=lfs -text

results/mgtv-glm-4-9b_p1_full_metrics.csv

@@ -1,6 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,THUDM/glm-4-9b-chat-1m,0.581,0.7030063507392827,0.581,0.6169152550060966
-1,THUDM/glm-4-9b-chat-1m_checkpoint-175,0.465,0.4626983424145981,0.47806716929403703,0.45282323637311034
-2,THUDM/glm-4-9b-chat-1m_checkpoint-350,0.579,0.677205249618572,0.579,0.607768567422897
-3,THUDM/glm-4-9b-chat-1m_checkpoint-525,0.6053333333333333,0.7220227211816735,0.6053333333333333,0.6379065537566594
-4,THUDM/glm-4-9b-chat-1m_checkpoint-700,0.593,0.7202870096146446,0.593,0.631178867641037
+version https://git-lfs.github.com/spec/v1
+oid sha256:edbcb992464c534c513f306139bd3bdb0cf69c073442c50826ed1a603ec97780
+size 514

results/mgtv-glm-4-9b_p2_full_metrics.csv

@@ -1,6 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,THUDM/glm-4-9b-chat-1m,0.395,0.667648445898918,0.395,0.4583896948511759
-1,THUDM/glm-4-9b-chat-1m_checkpoint-175,0.5946666666666667,0.7056249544340455,0.5946666666666667,0.631524021431916
-2,THUDM/glm-4-9b-chat-1m_checkpoint-350,0.549,0.7006542472262678,0.549,0.595639891731556
-3,THUDM/glm-4-9b-chat-1m_checkpoint-525,0.5986666666666667,0.7150511133873729,0.5986666666666667,0.6253567774596996
-4,THUDM/glm-4-9b-chat-1m_checkpoint-700,0.5843333333333334,0.730089967300272,0.5843333333333334,0.6195784049291421
+version https://git-lfs.github.com/spec/v1
+oid sha256:79058f1ae722d3f17125edc5d6c80f4828ec9fd52aba4c5faa7509d5af0cb2a5
+size 550

results/mgtv-llama3_p1_en_full_metrics.csv

@@ -1,5 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0.0,meta-llama/Meta-Llama-3-8B-Instruct_torch.bfloat16_lf,0.139,0.5741012854071863,0.13899999999999998,0.18649813199749535
-0.3333333333333333,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-117_torch.bfloat16_lf,0.6486666666666666,0.6525934632970077,0.6486666666666666,0.6312721163517108
-0.6666666666666666,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-234_torch.bfloat16_lf,0.561,0.6897096276142071,0.561,0.6083393704375663
-1.0,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-351_torch.bfloat16_lf,0.621,0.686842945161901,0.621,0.6417441253605001
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9c4ee77b8880e0d9fcd73173f0aa87a14df015a44af6b74fd9abe1822c3b8a4
+size 588

results/mgtv-llama3_p1_r2_full_metrics.csv

@@ -1,6 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,shenzhi-wang/Llama3-8B-Chinese-Chat_torch.bfloat16,0.7836666666666666,0.7667122897184859,0.7929173693086004,0.7679400621793133
-1,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-175_torch.bfloat16,0.7706666666666667,0.8072750943858197,0.7706666666666667,0.7835719791561528
-2,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-350_torch.bfloat16,0.724,0.8118050163437011,0.724,0.7562266825513707
-3,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-525_torch.bfloat16,0.6756666666666666,0.7811762160181578,0.6756666666666666,0.7108457483297581
-4,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-700_torch.bfloat16,0.6496666666666666,0.779896556141616,0.6496666666666666,0.6931844557591907
+version https://git-lfs.github.com/spec/v1
+oid sha256:9428131509ba10f54ee8112191a5908d3a053a47c86b21223a2c75fb723cfc84
+size 719

results/mgtv-llama3_p1_r3_full_metrics.csv

@@ -1,7 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0.0,hfl/llama-3-chinese-8b-instruct-v3_torch.bfloat16_lf,0.4563333333333333,0.6744501777722286,0.4563333333333333,0.5301217655617138
-0.2,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-35_torch.bfloat16_lf,0.6406666666666667,0.7652410757316818,0.6406666666666667,0.686506602559313
-0.4,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-70_torch.bfloat16_lf,0.7223333333333334,0.761495200766395,0.7223333333333334,0.7296692772947554
-0.6,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-105_torch.bfloat16_lf,0.6256666666666667,0.7694288569564225,0.6256666666666667,0.6747421032154606
-0.8,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-140_torch.bfloat16_lf,0.7173333333333334,0.7746925589996599,0.7173333333333334,0.7391047683054164
-1.0,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-175_torch.bfloat16_lf,0.688,0.7678475683985698,0.688,0.7181965107150645
+version https://git-lfs.github.com/spec/v1
+oid sha256:68e47f1be3ae5ce000d707003c6f3460cce7d475c413d17a86790557a304dabe
+size 884

results/mgtv-llama3_p1_r4_full_metrics.csv

@@ -1,12 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0.0,shenzhi-wang/Llama3-8B-Chinese-Chat_torch.bfloat16_lf,0.7806666666666666,0.7573618757244,0.7898819561551433,0.7652422962253286
-0.2,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-35_torch.bfloat16_lf,0.7406666666666667,0.7968252989238709,0.7406666666666667,0.7531148920145138
-0.4,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-70_torch.bfloat16_lf,0.778,0.8030955576782131,0.778,0.7815185410268678
-0.6,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-105_torch.bfloat16_lf,0.739,0.8025788156962886,0.739,0.7612977570469681
-0.8,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-140_torch.bfloat16_lf,0.7496666666666667,0.7967576318450474,0.7496666666666667,0.7547381069942627
-1.0,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-175_torch.bfloat16_lf,0.7226666666666667,0.8078695709123398,0.7226666666666667,0.7538027484740696
-1.2,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-210_torch.bfloat16_lf,0.6993333333333334,0.8053423632532914,0.6993333333333334,0.7388494823688349
-1.4,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-245_torch.bfloat16_lf,0.7306666666666667,0.7976913307876342,0.7306666666666667,0.752534867135234
-1.6,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-280_torch.bfloat16_lf,0.7123333333333334,0.8017497457465886,0.7123333333333334,0.7451979461870872
-1.8,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-315_torch.bfloat16_lf,0.699,0.8094773954029667,0.699,0.7399804367113129
-2.0,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-350_torch.bfloat16_lf,0.721,0.8097769853787679,0.721,0.7542245417773473
+version https://git-lfs.github.com/spec/v1
+oid sha256:594ce71ff1b42530a9090b68f1dc8ea61c68e3090db0cd6aa4c6c5ed0ddcff8c
+size 1555

results/mgtv-llama3_p2_en_full_metrics.csv

@@ -1,5 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0.0,meta-llama/Meta-Llama-3-8B-Instruct_torch.bfloat16_lf,0.15466666666666667,0.5218515852800517,0.15466666666666667,0.17611762474007195
-0.3333333333333333,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-117_torch.bfloat16_lf,0.6203333333333333,0.663582082981778,0.6203333333333333,0.6363626392286635
-0.6666666666666666,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-234_torch.bfloat16_lf,0.5613333333333334,0.7000506187405509,0.5613333333333334,0.6113039056178092
-1.0,meta-llama/Meta-Llama-3-8B-Instruct/checkpoint-351_torch.bfloat16_lf,0.6203333333333333,0.6819200833733873,0.6203333333333333,0.6405153767205392
+version https://git-lfs.github.com/spec/v1
+oid sha256:71abffb961ae9cbb3f32ea8643907e5163c437e2a811ea0838489b7efc37279e
+size 654

results/mgtv-llama3_p2_r2_full_metrics.csv

@@ -1,6 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,shenzhi-wang/Llama3-8B-Chinese-Chat_torch.bfloat16,0.73,0.7709739363586101,0.73,0.7462914191370829
-1,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-175_torch.bfloat16,0.718,0.8113087212796575,0.718,0.75010620119596
-2,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-350_torch.bfloat16,0.7273333333333334,0.8025119533060184,0.7273333333333334,0.7549822040428019
-3,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-525_torch.bfloat16,0.6883333333333334,0.7816168321128566,0.6883333333333334,0.716763388345211
-4,shenzhi-wang/Llama3-8B-Chinese-Chat/checkpoint-700_torch.bfloat16,0.6406666666666667,0.7636301286950402,0.6406666666666667,0.6807929233309169
+version https://git-lfs.github.com/spec/v1
+oid sha256:69fd56c4f7811e57eb987afa789dc9d4722a8b4ffc1717a9a5cb67e071d2287e
+size 689

results/mgtv-llama3_p2_r3_full_metrics.csv

@@ -1,7 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0.0,hfl/llama-3-chinese-8b-instruct-v3_torch.bfloat16_lf,0.25066666666666665,0.6852419041932336,0.25066666666666665,0.32636449818329016
-0.2,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-35_torch.bfloat16_lf,0.7283333333333334,0.7722393813259697,0.7283333333333334,0.7426450360790026
-0.4,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-70_torch.bfloat16_lf,0.741,0.7868300593752113,0.741,0.7514058688729928
-0.6,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-105_torch.bfloat16_lf,0.6223333333333333,0.7771706776754249,0.6223333333333333,0.6762790454549326
-0.8,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-140_torch.bfloat16_lf,0.7,0.7767966010489314,0.7,0.7298480873851099
-1.0,hfl/llama-3-chinese-8b-instruct-v3/checkpoint-175_torch.bfloat16_lf,0.697,0.78712001874989,0.697,0.7309586130328194
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc571905d1b7c2b850054ac77ef34d3b9c61bae429b05aa505406eab0d0c86d1
+size 831

results/mgtv-results_internlm_4bit_metrics.csv

@@ -1,17 +1,3 @@
-repetition penalty,model,accuracy,precision,recall,f1
-1.00,internlm/internlm2_5-7b-chat-1m_4bit_rp1.0,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.01,internlm/internlm2_5-7b-chat-1m_4bit_rp1.01,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.02,internlm/internlm2_5-7b-chat-1m_4bit_rp1.02,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.03,internlm/internlm2_5-7b-chat-1m_4bit_rp1.03,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.04,internlm/internlm2_5-7b-chat-1m_4bit_rp1.04,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.05,internlm/internlm2_5-7b-chat-1m_4bit_rp1.05,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.06,internlm/internlm2_5-7b-chat-1m_4bit_rp1.06,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.07,internlm/internlm2_5-7b-chat-1m_4bit_rp1.07,0.783,0.8089894370996324,0.783,0.7918676604452453
-1.08,internlm/internlm2_5-7b-chat-1m_4bit_rp1.08,0.7796666666666666,0.8092240536393919,0.7796666666666666,0.790029078222227
-1.09,internlm/internlm2_5-7b-chat-1m_4bit_rp1.09,0.694,0.8069396574279459,0.694,0.7320540780553862
-1.10,internlm/internlm2_5-7b-chat-1m_4bit_rp1.1,0.532,0.7907326858338991,0.532,0.5650084362647749
-1.11,internlm/internlm2_5-7b-chat-1m_4bit_rp1.11,0.4703333333333333,0.8029879177982981,0.4703333333333333,0.4846135307750191
-1.12,internlm/internlm2_5-7b-chat-1m_4bit_rp1.12,0.438,0.854597269027913,0.438,0.46295579268753667
-1.13,internlm/internlm2_5-7b-chat-1m_4bit_rp1.1300000000000001,0.4076666666666667,0.8777389419821567,0.4076666666666667,0.4483520542323376
-1.14,internlm/internlm2_5-7b-chat-1m_4bit_rp1.1400000000000001,0.35933333333333334,0.8851092633830993,0.35933333333333334,0.4204411983471381
-1.15,internlm/internlm2_5-7b-chat-1m_4bit_rp1.1500000000000001,0.301,0.8493737317963316,0.301,0.3786855007398624
+version https://git-lfs.github.com/spec/v1
+oid sha256:db680ebf24b92e29c61022fc0b848d8432ce69f8915ae878c8ed580c5f34ea0c
+size 1783

results/mgtv-results_p2_r2_full_metrics.csv

@@ -1,6 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,internlm/internlm2_5-7b-chat-1m,0.766,0.7479690198649127,0.7875257025359835,0.7649220492304646
-1,internlm/internlm2_5-7b-chat-1m_checkpoint-175,0.812,0.8122861942516547,0.812,0.8102342544894316
-2,internlm/internlm2_5-7b-chat-1m_checkpoint-350,0.7653333333333333,0.8068892149662973,0.7653333333333333,0.7799982606366916
-3,internlm/internlm2_5-7b-chat-1m_checkpoint-525,0.7476666666666667,0.8120325497709814,0.7476666666666667,0.7731222076608317
-4,internlm/internlm2_5-7b-chat-1m_checkpoint-700,0.717,0.8046420022590015,0.717,0.7510339687376877
+version https://git-lfs.github.com/spec/v1
+oid sha256:c834c71931e94acfa3d482a1d21e7000697d8f2de43d11b17635e199132a518f
+size 586

results/mgtv-results_p2_r3_full_metrics.csv

@@ -1,12 +1,3 @@
-epoch,model,accuracy,precision,recall,f1
-0,internlm/internlm2_5-7b-chat-1m_torch.bfloat16_lf,0.5106666666666667,0.751987084146518,0.5166947723440135,0.5420573881946011
-1,internlm/internlm2_5-7b-chat-1m/checkpoint-35_torch.bfloat16_lf,0.7843333333333333,0.7977648302848388,0.7843333333333333,0.7864944570659659
-2,internlm/internlm2_5-7b-chat-1m/checkpoint-70_torch.bfloat16_lf,0.7836666666666666,0.7996977262947886,0.7836666666666666,0.7886881726841081
-3,internlm/internlm2_5-7b-chat-1m/checkpoint-105_torch.bfloat16_lf,0.7243333333333334,0.8171172705912051,0.7243333333333334,0.7565804830382912
-4,internlm/internlm2_5-7b-chat-1m/checkpoint-140_torch.bfloat16_lf,0.803,0.8031411888150441,0.803,0.8028064320197301
-5,internlm/internlm2_5-7b-chat-1m/checkpoint-175_torch.bfloat16_lf,0.7676666666666667,0.8108441731715863,0.7676666666666667,0.7843187816704813
-6,internlm/internlm2_5-7b-chat-1m/checkpoint-210_torch.bfloat16_lf,0.7736666666666666,0.8091671780923799,0.7736666666666666,0.7876874850235454
-7,internlm/internlm2_5-7b-chat-1m/checkpoint-245_torch.bfloat16_lf,0.7623333333333333,0.8062291602218205,0.7623333333333333,0.777669094563925
-8,internlm/internlm2_5-7b-chat-1m/checkpoint-280_torch.bfloat16_lf,0.7553333333333333,0.8086197936829652,0.7553333333333333,0.7755588811428297
-9,internlm/internlm2_5-7b-chat-1m/checkpoint-315_torch.bfloat16_lf,0.748,0.8171996792797457,0.748,0.773990849396903
-10,internlm/internlm2_5-7b-chat-1m/checkpoint-350_torch.bfloat16_lf,0.756,0.8126875394266148,0.756,0.7777812522863184
+version https://git-lfs.github.com/spec/v1
+oid sha256:939594a2230ec75abdcc3c3d330ce9441e3ccac2a447a21679febab651ef8512
+size 1517

results/test_b-results_r3.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e0b3b5ed431d12d417156d22a30cd19745ecfc13c2df34e974b9af33f1027b6
+size 23121608