jbomdev commited on
Commit
13b5604
·
1 Parent(s): ac44f37

Upload folder using huggingface_hub

Browse files
sft_logs/SAMPLES_2026-04-25_14-39-31.txt ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ ======================================================================
3
+ Step 0 (lr=4.00e-07)
4
+ Prompt: What is the capital of France?
5
+ Reply: It is the year 2033. The National Geographic Society in the U.S. has launched a new website to raise awareness about the importance of the environment and to inspire the next generation of global citizens. The online resource is called
6
+ “The Ecologist,” a new interactive website that will provide students and the general public with the opportunity to learn about how humans have affected the planet.
7
+ The website will provide students with the opportunity to study a wide variety of information related to the environment and the effects of
8
+
9
+ ======================================================================
10
+ Step 100 (lr=1.72e-05)
11
+ Prompt: What is the capital of France?
12
+ Reply: The capital of France is Paris.
13
+
14
+ ======================================================================
15
+ Step 200 (lr=4.19e-06)
16
+ Prompt: What is the capital of France?
17
+ Reply: What is the name of the country?
sft_logs/TRAINING_LOG_2026-04-25_14-39-31.csv ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ step,loss,val_loss,norm,lr,tokens_per_second,step_time
2
+ 0,2.596765,2.688483,4.1420,4.0000e-07,7729.9,33.9131
3
+ 1,2.620967,,4.1043,8.0000e-07,39142.5,6.6972
4
+ 2,2.622568,,4.6021,1.2000e-06,39188.8,6.6893
5
+ 3,2.649046,,4.2183,1.6000e-06,39151.6,6.6956
6
+ 4,2.582411,,3.4601,2.0000e-06,39152.7,6.6954
7
+ 5,2.599589,,3.0777,2.4000e-06,39136.0,6.6983
8
+ 6,2.568483,,3.1862,2.8000e-06,39119.9,6.7010
9
+ 7,2.503999,,3.0085,3.2000e-06,39103.5,6.7038
10
+ 8,2.486288,,3.0015,3.6000e-06,39070.2,6.7096
11
+ 9,2.364417,,2.5862,4.0000e-06,39082.3,6.7075
12
+ 10,2.382519,,2.1403,4.4000e-06,39085.0,6.7070
13
+ 11,2.424735,,2.0419,4.8000e-06,39105.3,6.7035
14
+ 12,2.338599,,1.9496,5.2000e-06,39096.9,6.7050
15
+ 13,2.332317,,1.8980,5.6000e-06,39078.1,6.7082
16
+ 14,2.334962,,2.6628,6.0000e-06,39022.1,6.7178
17
+ 15,2.308158,,1.9815,6.4000e-06,38981.9,6.7248
18
+ 16,2.302701,,2.0932,6.8000e-06,39027.4,6.7169
19
+ 17,2.263134,,1.7424,7.2000e-06,39016.2,6.7188
20
+ 18,2.264545,,1.6132,7.6000e-06,39019.7,6.7182
21
+ 19,2.341298,,1.4786,8.0000e-06,39024.4,6.7174
22
+ 20,2.192963,,1.7024,8.4000e-06,39012.6,6.7195
23
+ 21,2.205914,,1.2219,8.8000e-06,39017.2,6.7187
24
+ 22,2.213036,,1.0610,9.2000e-06,39016.4,6.7188
25
+ 23,2.232377,,1.1446,9.6000e-06,39022.9,6.7177
26
+ 24,2.186729,,1.0527,1.0000e-05,38973.5,6.7262
27
+ 25,2.152267,,0.8365,1.0400e-05,39019.0,6.7184
28
+ 26,2.118651,,0.8740,1.0800e-05,39008.5,6.7202
29
+ 27,2.149728,,0.8858,1.1200e-05,39006.5,6.7205
30
+ 28,2.109647,,0.7811,1.1600e-05,39018.8,6.7184
31
+ 29,2.083347,,0.7307,1.2000e-05,39006.8,6.7205
32
+ 30,2.161695,,0.7611,1.2400e-05,39014.3,6.7192
33
+ 31,2.125426,,0.7389,1.2800e-05,39008.7,6.7201
34
+ 32,2.125073,,0.6667,1.3200e-05,39021.7,6.7179
35
+ 33,2.082039,,0.6093,1.3600e-05,38993.1,6.7228
36
+ 34,2.113923,,0.5902,1.4000e-05,39026.9,6.7170
37
+ 35,2.052234,,0.6843,1.4400e-05,39008.3,6.7202
38
+ 36,2.100197,,0.6837,1.4800e-05,39009.9,6.7199
39
+ 37,1.973511,,0.6460,1.5200e-05,39033.5,6.7159
40
+ 38,1.966371,,0.6668,1.5600e-05,39025.4,6.7173
41
+ 39,1.998548,,0.6791,1.6000e-05,39014.7,6.7191
42
+ 40,2.024856,,0.7034,1.6400e-05,39008.7,6.7201
43
+ 41,2.037716,,0.6234,1.6800e-05,39020.0,6.7182
44
+ 42,1.972760,,0.6661,1.7200e-05,39013.6,6.7193
45
+ 43,2.051978,,0.7578,1.7600e-05,39017.5,6.7186
46
+ 44,1.915081,,0.5545,1.8000e-05,36931.1,7.0982
47
+ 45,1.883206,,0.5461,1.8400e-05,39011.8,6.7196
48
+ 46,2.073134,,0.6401,1.8800e-05,39009.5,6.7200
49
+ 47,1.990191,,0.5656,1.9200e-05,39013.8,6.7193
50
+ 48,1.960026,,0.5309,1.9600e-05,38998.8,6.7218
51
+ 49,1.988840,,0.5712,2.0000e-05,39004.4,6.7209
52
+ 50,1.923164,1.966426,0.5325,2.0000e-05,29727.7,8.8182
53
+ 51,2.046320,,0.5886,1.9999e-05,39025.4,6.7173
54
+ 52,1.965524,,0.5407,1.9995e-05,39011.6,6.7196
55
+ 53,1.943951,,0.4980,1.9989e-05,38977.3,6.7255
56
+ 54,2.004841,,0.5091,1.9981e-05,39025.3,6.7173
57
+ 55,1.863690,,0.4829,1.9971e-05,39038.5,6.7150
58
+ 56,1.866938,,0.5044,1.9958e-05,39021.8,6.7179
59
+ 57,1.867185,,0.5021,1.9942e-05,39006.4,6.7205
60
+ 58,1.860586,,0.4843,1.9925e-05,38985.7,6.7241
61
+ 59,1.880746,,0.4633,1.9905e-05,38988.6,6.7236
62
+ 60,1.954420,,0.4892,1.9882e-05,39030.4,6.7164
63
+ 61,1.956734,,0.5103,1.9858e-05,39020.2,6.7182
64
+ 62,1.877093,,0.4178,1.9831e-05,38995.0,6.7225
65
+ 63,1.955152,,0.5853,1.9801e-05,39030.8,6.7163
66
+ 64,1.875380,,0.5358,1.9770e-05,39027.8,6.7169
67
+ 65,2.061920,,0.5237,1.9736e-05,39015.0,6.7191
68
+ 66,1.883819,,0.4832,1.9700e-05,39006.5,6.7205
69
+ 67,1.876885,,0.3960,1.9661e-05,39004.0,6.7210
70
+ 68,1.801681,,0.4935,1.9620e-05,38980.1,6.7251
71
+ 69,1.949412,,0.5039,1.9577e-05,39013.5,6.7193
72
+ 70,1.916114,,0.4766,1.9532e-05,38997.5,6.7221
73
+ 71,1.888946,,0.4902,1.9485e-05,38993.6,6.7227
74
+ 72,1.864144,,0.4113,1.9435e-05,39000.7,6.7215
75
+ 73,1.846538,,0.4585,1.9383e-05,39017.5,6.7186
76
+ 74,1.899733,,0.4662,1.9329e-05,39031.0,6.7163
77
+ 75,1.904039,,0.4197,1.9272e-05,39025.5,6.7173
78
+ 76,1.857437,,0.5062,1.9214e-05,39024.9,6.7174
79
+ 77,1.910785,,0.4380,1.9153e-05,38981.9,6.7248
80
+ 78,1.871540,,0.4569,1.9091e-05,39015.3,6.7190
81
+ 79,1.886530,,0.4135,1.9026e-05,39002.3,6.7213
82
+ 80,1.880064,,0.4178,1.8959e-05,38998.8,6.7218
83
+ 81,1.864492,,0.5314,1.8890e-05,39026.1,6.7171
84
+ 82,1.922211,,0.4776,1.8818e-05,39002.1,6.7213
85
+ 83,1.851485,,0.4298,1.8745e-05,39014.7,6.7191
86
+ 84,1.879654,,0.4370,1.8670e-05,39004.5,6.7209
87
+ 85,1.918141,,0.5154,1.8593e-05,38781.5,6.7595
88
+ 86,1.870528,,0.4195,1.8513e-05,38882.0,6.7420
89
+ 87,1.872387,,0.4156,1.8432e-05,38691.3,6.7753
90
+ 88,1.898265,,0.4327,1.8349e-05,38596.6,6.7919
91
+ 89,1.863208,,0.4024,1.8264e-05,38535.0,6.8028
92
+ 90,1.883613,,0.4290,1.8177e-05,38592.8,6.7926
93
+ 91,1.892363,,0.4131,1.8088e-05,38558.1,6.7987
94
+ 92,1.904886,,0.4122,1.7997e-05,38593.2,6.7925
95
+ 93,1.874003,,0.3997,1.7905e-05,38592.4,6.7926
96
+ 94,1.887361,,0.4356,1.7810e-05,38718.2,6.7706
97
+ 95,1.877295,,0.3673,1.7714e-05,38512.2,6.8068
98
+ 96,1.956879,,0.4321,1.7616e-05,38651.5,6.7823
99
+ 97,1.867290,,0.4220,1.7517e-05,38744.9,6.7659
100
+ 98,1.859444,,0.3980,1.7415e-05,38614.2,6.7888
101
+ 99,1.901427,,0.4261,1.7312e-05,38643.4,6.7837
102
+ 100,1.870644,1.856444,0.3911,1.7208e-05,28971.4,9.0484
103
+ 101,1.862615,,0.4373,1.7101e-05,38964.7,6.7277
104
+ 102,1.834544,,0.4101,1.6993e-05,38965.3,6.7276
105
+ 103,1.848305,,0.3790,1.6884e-05,38912.8,6.7367
106
+ 104,1.915386,,0.4089,1.6773e-05,38912.2,6.7368
107
+ 105,1.812944,,0.3796,1.6660e-05,38912.4,6.7368
108
+ 106,1.832860,,0.4108,1.6546e-05,38934.4,6.7330
109
+ 107,1.780070,,0.3791,1.6431e-05,38937.5,6.7324
110
+ 108,1.830826,,0.3744,1.6314e-05,38968.9,6.7270
111
+ 109,1.892211,,0.4215,1.6195e-05,38947.8,6.7306
112
+ 110,1.845418,,0.4388,1.6076e-05,38907.5,6.7376
113
+ 111,1.776545,,0.4177,1.5955e-05,37100.1,7.0659
114
+ 112,1.820671,,0.4390,1.5832e-05,38958.6,6.7288
115
+ 113,1.901331,,0.4080,1.5709e-05,38994.2,6.7226
116
+ 114,1.831425,,0.4097,1.5584e-05,39003.9,6.7210
117
+ 115,1.799106,,0.3578,1.5458e-05,38998.9,6.7218
118
+ 116,1.822684,,0.3822,1.5331e-05,39032.5,6.7160
119
+ 117,1.905982,,0.4072,1.5202e-05,38997.3,6.7221
120
+ 118,1.828314,,0.4098,1.5073e-05,38982.6,6.7246
121
+ 119,1.869104,,0.3817,1.4942e-05,38991.6,6.7231
122
+ 120,1.834974,,0.3978,1.4811e-05,38979.1,6.7252
123
+ 121,1.820533,,0.3960,1.4678e-05,38982.8,6.7246
124
+ 122,1.926475,,0.4061,1.4545e-05,39014.8,6.7191
125
+ 123,1.823885,,0.4370,1.4410e-05,38965.2,6.7276
126
+ 124,1.827898,,0.3842,1.4275e-05,38952.8,6.7298
127
+ 125,1.872620,,0.3841,1.4139e-05,38993.7,6.7227
128
+ 126,1.863147,,0.3858,1.4002e-05,39009.9,6.7199
129
+ 127,1.816786,,0.3629,1.3864e-05,39018.2,6.7185
130
+ 128,1.816913,,0.4232,1.3726e-05,39011.9,6.7196
131
+ 129,1.838953,,0.3707,1.3586e-05,38975.0,6.7259
132
+ 130,1.823679,,0.3922,1.3446e-05,38976.1,6.7258
133
+ 131,1.831664,,0.4038,1.3306e-05,38975.5,6.7259
134
+ 132,1.810620,,0.3936,1.3165e-05,38988.7,6.7236
135
+ 133,1.836939,,0.3878,1.3023e-05,38956.8,6.7291
136
+ 134,1.839403,,0.3659,1.2881e-05,39013.1,6.7194
137
+ 135,1.826810,,0.3715,1.2738e-05,39001.7,6.7214
138
+ 136,1.798390,,0.3834,1.2595e-05,39032.0,6.7161
139
+ 137,1.874390,,0.3713,1.2451e-05,38991.9,6.7230
140
+ 138,1.850365,,0.3744,1.2307e-05,38981.4,6.7248
141
+ 139,1.754206,,0.4089,1.2163e-05,39011.8,6.7196
142
+ 140,1.777730,,0.3807,1.2018e-05,38983.6,6.7245
143
+ 141,1.790778,,0.3373,1.1873e-05,38978.0,6.7254
144
+ 142,1.817963,,0.3775,1.1728e-05,38974.1,6.7261
145
+ 143,1.815875,,0.3859,1.1583e-05,39003.3,6.7211
146
+ 144,1.875113,,0.3822,1.1437e-05,39013.6,6.7193
147
+ 145,1.856843,,0.3565,1.1291e-05,38992.5,6.7229
148
+ 146,1.802585,,0.3656,1.1146e-05,38999.1,6.7218
149
+ 147,1.806073,,0.3553,1.1000e-05,39002.6,6.7212
150
+ 148,1.777791,,0.3807,1.0854e-05,38959.3,6.7287
151
+ 149,1.777414,,0.3823,1.0709e-05,38994.6,6.7226
152
+ 150,1.846710,1.820311,0.3679,1.0563e-05,29725.2,8.8189
153
+ 151,1.861467,,0.3361,1.0417e-05,38973.2,6.7263
154
+ 152,1.820384,,0.4160,1.0272e-05,38964.6,6.7277
155
+ 153,1.809566,,0.3973,1.0127e-05,39008.7,6.7201
156
+ 154,1.861523,,0.3846,9.9820e-06,38978.7,6.7253
157
+ 155,1.837350,,0.3871,9.8373e-06,38986.3,6.7240
158
+ 156,1.853784,,0.4010,9.6929e-06,38998.8,6.7218
159
+ 157,1.814328,,0.3867,9.5489e-06,38950.9,6.7301
160
+ 158,1.788189,,0.3722,9.4053e-06,39021.7,6.7179
161
+ 159,1.790600,,0.3489,9.2621e-06,38990.4,6.7233
162
+ 160,1.780846,,0.4129,9.1193e-06,38996.5,6.7222
163
+ 161,1.747178,,0.3736,8.9770e-06,39012.5,6.7195
164
+ 162,1.777217,,0.3558,8.8353e-06,38974.8,6.7260
165
+ 163,1.908285,,0.3761,8.6941e-06,38988.2,6.7237
166
+ 164,1.788122,,0.3826,8.5535e-06,38991.8,6.7231
167
+ 165,1.760198,,0.3459,8.4136e-06,39003.3,6.7211
168
+ 166,1.823772,,0.3430,8.2743e-06,39007.3,6.7204
169
+ 167,1.766874,,0.3635,8.1358e-06,38976.3,6.7257
170
+ 168,1.783306,,0.3817,7.9980e-06,38999.2,6.7218
171
+ 169,1.805502,,0.3881,7.8610e-06,39019.0,6.7184
172
+ 170,1.792593,,0.3904,7.7249e-06,38986.1,6.7240
173
+ 171,1.742153,,0.3716,7.5895e-06,39012.0,6.7196
174
+ 172,1.818473,,0.4058,7.4551e-06,38974.0,6.7261
175
+ 173,1.796199,,0.3688,7.3216e-06,39017.0,6.7187
176
+ 174,1.724125,,0.3697,7.1891e-06,38977.9,6.7254
177
+ 175,1.847812,,0.4051,7.0576e-06,38906.5,6.7378
178
+ 176,1.821590,,0.3970,6.9271e-06,38965.8,6.7275
179
+ 177,1.787026,,0.3765,6.7977e-06,39007.9,6.7203
180
+ 178,1.801762,,0.4069,6.6693e-06,37074.9,7.0707
181
+ 179,1.852575,,0.3788,6.5421e-06,39001.4,6.7214
182
+ 180,1.783015,,0.3471,6.4161e-06,38999.1,6.7218
183
+ 181,1.817899,,0.3853,6.2913e-06,39009.1,6.7201
184
+ 182,1.868378,,0.3730,6.1677e-06,38974.6,6.7260
185
+ 183,1.796606,,0.3396,6.0454e-06,39002.5,6.7212
186
+ 184,1.851884,,0.3736,5.9244e-06,38994.6,6.7226
187
+ 185,1.809291,,0.3828,5.8047e-06,38984.6,6.7243
188
+ 186,1.813265,,0.3741,5.6864e-06,38980.7,6.7250
189
+ 187,1.853295,,0.4112,5.5694e-06,39002.2,6.7213
190
+ 188,1.793295,,0.4066,5.4539e-06,39024.3,6.7175
191
+ 189,1.850681,,0.3704,5.3399e-06,38981.8,6.7248
192
+ 190,1.884629,,0.3912,5.2273e-06,39002.0,6.7213
193
+ 191,1.859340,,0.3733,5.1163e-06,39026.6,6.7171
194
+ 192,1.784193,,0.3582,5.0068e-06,38975.5,6.7259
195
+ 193,1.822049,,0.3393,4.8988e-06,38981.0,6.7249
196
+ 194,1.793281,,0.3697,4.7925e-06,38990.7,6.7232
197
+ 195,1.772500,,0.3402,4.6878e-06,38996.9,6.7222
198
+ 196,1.790982,,0.3864,4.5847e-06,38989.6,6.7234
199
+ 197,1.822788,,0.3798,4.4833e-06,39019.5,6.7183
200
+ 198,1.767981,,0.3633,4.3837e-06,38995.5,6.7224
201
+ 199,1.811864,,0.3595,4.2858e-06,38993.0,6.7229
202
+ 200,1.801135,1.805712,0.3585,4.1896e-06,29314.1,8.9426
203
+ 201,1.805537,,0.3662,4.0952e-06,38930.2,6.7337
204
+ 202,1.821266,,0.3514,4.0026e-06,38959.2,6.7287
205
+ 203,1.860320,,0.3630,3.9119e-06,38911.1,6.7370
206
+ 204,1.856123,,0.3544,3.8230e-06,38980.3,6.7250
207
+ 205,1.801354,,0.3706,3.7360e-06,38956.1,6.7292
208
+ 206,1.852909,,0.3543,3.6509e-06,38955.2,6.7294
209
+ 207,1.836904,,0.3629,3.5678e-06,38909.2,6.7373
210
+ 208,1.825097,,0.3603,3.4865e-06,38945.4,6.7311
211
+ 209,1.828156,,0.3678,3.4073e-06,38940.1,6.7320
212
+ 210,1.821467,,0.3626,3.3300e-06,38924.2,6.7347
213
+ 211,1.799865,,0.3675,3.2548e-06,38923.8,6.7348
214
+ 212,1.820646,,0.3619,3.1816e-06,38929.2,6.7339
215
+ 213,1.751599,,0.3562,3.1104e-06,38932.1,6.7334
216
+ 214,1.776612,,0.3664,3.0413e-06,38943.3,6.7314
217
+ 215,1.829155,,0.3677,2.9743e-06,38966.6,6.7274
218
+ 216,1.810005,,0.3593,2.9094e-06,38942.1,6.7316
219
+ 217,1.807661,,0.3706,2.8467e-06,38944.9,6.7312
220
+ 218,1.749687,,0.3424,2.7860e-06,38945.1,6.7311
221
+ 219,1.816526,,0.3461,2.7275e-06,38910.2,6.7371
222
+ 220,1.805110,,0.3730,2.6712e-06,38922.0,6.7351
223
+ 221,1.808812,,0.3560,2.6171e-06,38944.5,6.7312
224
+ 222,1.815417,,0.3696,2.5651e-06,38927.3,6.7342
225
+ 223,1.749148,,0.3753,2.5154e-06,38944.1,6.7313
226
+ 224,1.807836,,0.3589,2.4679e-06,38967.7,6.7272
227
+ 225,1.825391,,0.3634,2.4227e-06,38957.0,6.7291
228
+ 226,1.868578,,0.3608,2.3796e-06,38925.4,6.7345
229
+ 227,1.818423,,0.3452,2.3389e-06,38940.9,6.7318
230
+ 228,1.742561,,0.3608,2.3004e-06,38895.9,6.7396
231
+ 229,1.766687,,0.3226,2.2642e-06,38932.5,6.7333
232
+ 230,1.854108,,0.3628,2.2303e-06,38920.6,6.7354
233
+ 231,1.839664,,0.3789,2.1987e-06,38955.2,6.7294
234
+ 232,1.833855,,0.3655,2.1694e-06,38927.6,6.7341
235
+ 233,1.785398,,0.3874,2.1424e-06,38914.6,6.7364
236
+ 234,1.794494,,0.3452,2.1177e-06,38923.4,6.7349
237
+ 235,1.827139,,0.3529,2.0954e-06,38941.2,6.7318
238
+ 236,1.760012,,0.3339,2.0754e-06,38930.4,6.7337
239
+ 237,1.721672,,0.3902,2.0578e-06,38899.5,6.7390
240
+ 238,1.743281,,0.3575,2.0424e-06,38944.2,6.7313
241
+ 239,1.855971,,0.3882,2.0295e-06,38958.4,6.7288
242
+ 240,1.830788,,0.3578,2.0189e-06,38940.2,6.7320
243
+ 241,1.794156,,0.3526,2.0106e-06,38958.8,6.7287
244
+ 242,1.808245,,0.3663,2.0047e-06,38947.5,6.7307
245
+ 243,1.829680,,0.3747,2.0012e-06,38886.5,6.7413