|  |  | Adding `safetensors` variant of this model (#1) | versae, SFconvertbot | Jul 19, 2025 | 
|  |  | Automatic correction of README.md metadata. Contact website@huggingface.co for any question | elishowk | 
|  |  | Update README.md | Pablogps | 
|  |  | Update README.md | Pablogps | 
|  |  | Create README.md | Pablogps | 
|  |  | Step... (249001/250000 | Loss: 1.678702473640442, Acc: 0.660828173160553):  76%|██████████████████▏     | 189999/250000 [97:45:10<29:36:08,  1.78s/it] | versae | 
|  |  | Step... (238001/250000 | Loss: 1.6818329095840454, Acc: 0.660420298576355):  71%|████████████████▍      | 178271/250000 [91:38:03<37:33:10,  1.88s/it] | versae | 
|  |  | Step... (232001/250000 | Loss: 1.6789664030075073, Acc: 0.6606228351593018):  69%|███████████████▏      | 172256/250000 [88:28:48<44:13:31,  2.05s/it] | versae | 
|  |  | Step... (232001/250000 | Loss: 1.6789664030075073, Acc: 0.6606228351593018):  69%|███████████████▏      | 172256/250000 [88:28:48<44:13:31,  2.05s/it] | versae | 
|  |  | Step... (219001/250000 | Loss: 1.6840890645980835, Acc: 0.6597731709480286):  64%|██████████████        | 159238/250000 [81:41:37<54:11:57,  2.15s/it] | versae | 
|  |  | Step... (214001/250000 | Loss: 1.6834079027175903, Acc: 0.6600053906440735):  62%|█████████████▋        | 154922/250000 [79:19:10<50:28:22,  1.91s/it] | versae | 
|  |  | Step... (212001/250000 | Loss: 1.6889899969100952, Acc: 0.6588694453239441):  61%|█████████████▍        | 152455/250000 [78:01:35<56:19:35,  2.08s/it] | versae | 
|  |  | Step... (208001/250000 | Loss: 1.6939743757247925, Acc: 0.6590701937675476):  59%|█████████████         | 148622/250000 [76:01:19<63:54:00,  2.27s/it] | versae | 
|  |  | Step... (200001/250000 | Loss: 1.6965291500091553, Acc: 0.6575480699539185):  56%|████████████▎         | 140093/250000 [71:33:45<52:41:15,  1.73s/it] | versae | 
|  |  | Model at 191k steps, acc 0.6562. Dataset stats | versae | 
|  |  | Step... (190001/250000 | Loss: 1.705819845199585, Acc: 0.6563901901245117):  52%|████████████           | 130628/250000 [66:37:26<59:10:21,  1.78s/it] | versae | 
|  |  | Dump 187k steps, acc 0.6574 | versae |