初始化项目,由ModelHub XC社区提供模型
Model: CyanMonkey/Petal-50M Source: Original Platform
This commit is contained in:
150
train_log.jsonl
Normal file
150
train_log.jsonl
Normal file
@@ -0,0 +1,150 @@
|
||||
{"event": "run_start", "step": 8800, "max_steps": 11652, "model_params_M": 49.691, "eff_batch": 1024, "amp_dtype": "torch.bfloat16", "attn_backend": "chunked", "started_at": "2026-07-24T18:29:26.130247"}
|
||||
{"step": 8820, "epoch": 0.757, "loss": 1.83717, "ppl": 6.279, "lr": 7.0758e-05, "lr_sched": "cosine", "grad_norm": 0.2893, "tokens": 9239408640, "tok_s": 267842.9, "elapsed_s": 131.4, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8840, "epoch": 0.759, "loss": 1.88987, "ppl": 6.619, "lr": 7.0215e-05, "lr_sched": "cosine", "grad_norm": 0.2819, "tokens": 9260359680, "tok_s": 269322.3, "elapsed_s": 209.1, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8860, "epoch": 0.76, "loss": 1.92752, "ppl": 6.872, "lr": 6.9675e-05, "lr_sched": "cosine", "grad_norm": 0.2926, "tokens": 9281310720, "tok_s": 269209.2, "elapsed_s": 286.8, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8880, "epoch": 0.762, "loss": 1.82917, "ppl": 6.229, "lr": 6.9138e-05, "lr_sched": "cosine", "grad_norm": 0.3123, "tokens": 9302261760, "tok_s": 269646.1, "elapsed_s": 364.5, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8900, "epoch": 0.764, "loss": 1.86472, "ppl": 6.454, "lr": 6.8604e-05, "lr_sched": "cosine", "grad_norm": 0.2714, "tokens": 9323212800, "tok_s": 268452.9, "elapsed_s": 442.3, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8920, "epoch": 0.766, "loss": 1.90822, "ppl": 6.741, "lr": 6.8073e-05, "lr_sched": "cosine", "grad_norm": 0.2723, "tokens": 9344163840, "tok_s": 268963.8, "elapsed_s": 520.1, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8940, "epoch": 0.767, "loss": 1.91169, "ppl": 6.765, "lr": 6.7545e-05, "lr_sched": "cosine", "grad_norm": 0.2922, "tokens": 9365114880, "tok_s": 268677.9, "elapsed_s": 598.0, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8960, "epoch": 0.769, "loss": 1.80572, "ppl": 6.084, "lr": 6.702e-05, "lr_sched": "cosine", "grad_norm": 0.2728, "tokens": 9386065920, "tok_s": 268837.0, "elapsed_s": 675.6, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 8980, "epoch": 0.771, "loss": 1.88449, "ppl": 6.583, "lr": 6.6499e-05, "lr_sched": "cosine", "grad_norm": 0.2686, "tokens": 9407016960, "tok_s": 270643.9, "elapsed_s": 753.3, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9000, "epoch": 0.772, "loss": 1.85663, "ppl": 6.402, "lr": 6.598e-05, "lr_sched": "cosine", "grad_norm": 0.3064, "tokens": 9427968000, "tok_s": 269440.4, "elapsed_s": 831.0, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9020, "epoch": 0.774, "loss": 1.86381, "ppl": 6.448, "lr": 6.5465e-05, "lr_sched": "cosine", "grad_norm": 0.2849, "tokens": 9448919040, "tok_s": 269911.5, "elapsed_s": 908.8, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9040, "epoch": 0.776, "loss": 1.84236, "ppl": 6.311, "lr": 6.4952e-05, "lr_sched": "cosine", "grad_norm": 0.2711, "tokens": 9469870080, "tok_s": 269519.0, "elapsed_s": 986.7, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9060, "epoch": 0.778, "loss": 1.90693, "ppl": 6.732, "lr": 6.4443e-05, "lr_sched": "cosine", "grad_norm": 0.3108, "tokens": 9490821120, "tok_s": 268620.4, "elapsed_s": 1064.6, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9080, "epoch": 0.779, "loss": 1.84251, "ppl": 6.312, "lr": 6.3937e-05, "lr_sched": "cosine", "grad_norm": 0.3007, "tokens": 9511772160, "tok_s": 268382.2, "elapsed_s": 1142.5, "vram_gb": 2.78, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9100, "epoch": 0.781, "loss": 1.86656, "ppl": 6.466, "lr": 6.3435e-05, "lr_sched": "cosine", "grad_norm": 0.2822, "tokens": 9532723200, "tok_s": 269944.3, "elapsed_s": 1220.5, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9120, "epoch": 0.783, "loss": 1.83974, "ppl": 6.295, "lr": 6.2935e-05, "lr_sched": "cosine", "grad_norm": 0.2854, "tokens": 9553674240, "tok_s": 269876.0, "elapsed_s": 1298.2, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9140, "epoch": 0.784, "loss": 1.83865, "ppl": 6.288, "lr": 6.2439e-05, "lr_sched": "cosine", "grad_norm": 0.2851, "tokens": 9574625280, "tok_s": 268713.0, "elapsed_s": 1376.0, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9160, "epoch": 0.786, "loss": 1.86074, "ppl": 6.428, "lr": 6.1946e-05, "lr_sched": "cosine", "grad_norm": 0.2714, "tokens": 9595576320, "tok_s": 269246.5, "elapsed_s": 1453.9, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9180, "epoch": 0.788, "loss": 1.82733, "ppl": 6.217, "lr": 6.1456e-05, "lr_sched": "cosine", "grad_norm": 0.2863, "tokens": 9616527360, "tok_s": 267633.3, "elapsed_s": 1531.7, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9200, "epoch": 0.79, "loss": 1.80697, "ppl": 6.092, "lr": 6.097e-05, "lr_sched": "cosine", "grad_norm": 0.2805, "tokens": 9637478400, "tok_s": 269846.8, "elapsed_s": 1609.5, "vram_gb": 4.92, "ram_pct": 4.0, "disk_free_gb": 9223371971.07, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9200, "epoch": 0.79, "val_loss": 1.91013, "val_ppl": 6.754, "best_val_ppl": 6.754, "is_best": true}
|
||||
{"step": 9220, "epoch": 0.791, "loss": 1.84067, "ppl": 6.301, "lr": 6.0487e-05, "lr_sched": "cosine", "grad_norm": 0.2881, "tokens": 9658429440, "tok_s": 269275.0, "elapsed_s": 2040.7, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9240, "epoch": 0.793, "loss": 1.80503, "ppl": 6.08, "lr": 6.0007e-05, "lr_sched": "cosine", "grad_norm": 0.2881, "tokens": 9679380480, "tok_s": 269726.3, "elapsed_s": 2118.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9260, "epoch": 0.795, "loss": 1.85043, "ppl": 6.363, "lr": 5.9531e-05, "lr_sched": "cosine", "grad_norm": 0.3165, "tokens": 9700331520, "tok_s": 269143.4, "elapsed_s": 2196.2, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9280, "epoch": 0.796, "loss": 1.76735, "ppl": 5.855, "lr": 5.9058e-05, "lr_sched": "cosine", "grad_norm": 0.2952, "tokens": 9721282560, "tok_s": 270518.8, "elapsed_s": 2274.0, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9300, "epoch": 0.798, "loss": 1.84474, "ppl": 6.326, "lr": 5.8588e-05, "lr_sched": "cosine", "grad_norm": 0.2775, "tokens": 9742233600, "tok_s": 268534.3, "elapsed_s": 2351.8, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9320, "epoch": 0.8, "loss": 1.80603, "ppl": 6.086, "lr": 5.8121e-05, "lr_sched": "cosine", "grad_norm": 0.2781, "tokens": 9763184640, "tok_s": 269600.7, "elapsed_s": 2429.6, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9340, "epoch": 0.802, "loss": 1.88347, "ppl": 6.576, "lr": 5.7659e-05, "lr_sched": "cosine", "grad_norm": 0.2817, "tokens": 9784135680, "tok_s": 270046.1, "elapsed_s": 2507.3, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9360, "epoch": 0.803, "loss": 1.82086, "ppl": 6.177, "lr": 5.7199e-05, "lr_sched": "cosine", "grad_norm": 0.2807, "tokens": 9805086720, "tok_s": 268721.7, "elapsed_s": 2585.2, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9380, "epoch": 0.805, "loss": 1.82677, "ppl": 6.214, "lr": 5.6743e-05, "lr_sched": "cosine", "grad_norm": 0.2882, "tokens": 9826037760, "tok_s": 269369.6, "elapsed_s": 2662.9, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9400, "epoch": 0.807, "loss": 1.85959, "ppl": 6.421, "lr": 5.629e-05, "lr_sched": "cosine", "grad_norm": 0.333, "tokens": 9846988800, "tok_s": 270002.4, "elapsed_s": 2740.7, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9420, "epoch": 0.808, "loss": 1.785, "ppl": 5.96, "lr": 5.5841e-05, "lr_sched": "cosine", "grad_norm": 0.2814, "tokens": 9867939840, "tok_s": 268378.1, "elapsed_s": 2818.7, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9440, "epoch": 0.81, "loss": 1.84568, "ppl": 6.332, "lr": 5.5395e-05, "lr_sched": "cosine", "grad_norm": 0.2887, "tokens": 9888890880, "tok_s": 270474.4, "elapsed_s": 2896.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9460, "epoch": 0.812, "loss": 1.86376, "ppl": 6.448, "lr": 5.4953e-05, "lr_sched": "cosine", "grad_norm": 0.3094, "tokens": 9909841920, "tok_s": 269359.7, "elapsed_s": 2974.3, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9480, "epoch": 0.814, "loss": 1.8521, "ppl": 6.373, "lr": 5.4514e-05, "lr_sched": "cosine", "grad_norm": 0.2873, "tokens": 9930792960, "tok_s": 269679.7, "elapsed_s": 3052.1, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9500, "epoch": 0.815, "loss": 1.76625, "ppl": 5.849, "lr": 5.4079e-05, "lr_sched": "cosine", "grad_norm": 0.3056, "tokens": 9951744000, "tok_s": 268422.4, "elapsed_s": 3129.9, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9520, "epoch": 0.817, "loss": 1.89656, "ppl": 6.663, "lr": 5.3647e-05, "lr_sched": "cosine", "grad_norm": 0.3015, "tokens": 9972695040, "tok_s": 269291.3, "elapsed_s": 3207.7, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9540, "epoch": 0.819, "loss": 1.80714, "ppl": 6.093, "lr": 5.3219e-05, "lr_sched": "cosine", "grad_norm": 0.2886, "tokens": 9993646080, "tok_s": 269399.2, "elapsed_s": 3285.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9560, "epoch": 0.82, "loss": 1.83155, "ppl": 6.244, "lr": 5.2794e-05, "lr_sched": "cosine", "grad_norm": 0.2981, "tokens": 10014597120, "tok_s": 269282.1, "elapsed_s": 3363.4, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9580, "epoch": 0.822, "loss": 1.87333, "ppl": 6.51, "lr": 5.2373e-05, "lr_sched": "cosine", "grad_norm": 0.3048, "tokens": 10035548160, "tok_s": 269561.1, "elapsed_s": 3441.2, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9600, "epoch": 0.824, "loss": 1.83243, "ppl": 6.249, "lr": 5.1955e-05, "lr_sched": "cosine", "grad_norm": 0.2648, "tokens": 10056499200, "tok_s": 269895.9, "elapsed_s": 3519.0, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371969.45, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9600, "epoch": 0.824, "val_loss": 1.905, "val_ppl": 6.719, "best_val_ppl": 6.719, "is_best": true}
|
||||
{"step": 9620, "epoch": 0.826, "loss": 1.85277, "ppl": 6.377, "lr": 5.1541e-05, "lr_sched": "cosine", "grad_norm": 0.3153, "tokens": 10077450240, "tok_s": 270132.4, "elapsed_s": 3915.2, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9640, "epoch": 0.827, "loss": 1.86648, "ppl": 6.466, "lr": 5.1131e-05, "lr_sched": "cosine", "grad_norm": 0.2929, "tokens": 10098401280, "tok_s": 269661.8, "elapsed_s": 3993.0, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9660, "epoch": 0.829, "loss": 1.85813, "ppl": 6.412, "lr": 5.0724e-05, "lr_sched": "cosine", "grad_norm": 0.2854, "tokens": 10119352320, "tok_s": 268408.3, "elapsed_s": 4070.8, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9680, "epoch": 0.831, "loss": 1.83478, "ppl": 6.264, "lr": 5.0321e-05, "lr_sched": "cosine", "grad_norm": 0.2679, "tokens": 10140303360, "tok_s": 270136.8, "elapsed_s": 4148.6, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9700, "epoch": 0.832, "loss": 1.91241, "ppl": 6.769, "lr": 4.9921e-05, "lr_sched": "cosine", "grad_norm": 0.3052, "tokens": 10161254400, "tok_s": 268227.7, "elapsed_s": 4226.4, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9720, "epoch": 0.834, "loss": 1.88976, "ppl": 6.618, "lr": 4.9526e-05, "lr_sched": "cosine", "grad_norm": 0.2979, "tokens": 10182205440, "tok_s": 270479.4, "elapsed_s": 4304.1, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9740, "epoch": 0.836, "loss": 1.85868, "ppl": 6.415, "lr": 4.9133e-05, "lr_sched": "cosine", "grad_norm": 0.2899, "tokens": 10203156480, "tok_s": 269697.5, "elapsed_s": 4381.9, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9760, "epoch": 0.838, "loss": 1.88569, "ppl": 6.591, "lr": 4.8745e-05, "lr_sched": "cosine", "grad_norm": 0.2767, "tokens": 10224107520, "tok_s": 269154.8, "elapsed_s": 4459.7, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9780, "epoch": 0.839, "loss": 1.83973, "ppl": 6.295, "lr": 4.836e-05, "lr_sched": "cosine", "grad_norm": 0.3132, "tokens": 10245058560, "tok_s": 268768.6, "elapsed_s": 4537.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9800, "epoch": 0.841, "loss": 1.84633, "ppl": 6.337, "lr": 4.7979e-05, "lr_sched": "cosine", "grad_norm": 0.2694, "tokens": 10266009600, "tok_s": 269045.9, "elapsed_s": 4615.3, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9820, "epoch": 0.843, "loss": 1.85392, "ppl": 6.385, "lr": 4.7602e-05, "lr_sched": "cosine", "grad_norm": 0.2859, "tokens": 10286960640, "tok_s": 269634.0, "elapsed_s": 4693.1, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9840, "epoch": 0.844, "loss": 1.83725, "ppl": 6.279, "lr": 4.7228e-05, "lr_sched": "cosine", "grad_norm": 0.2653, "tokens": 10307911680, "tok_s": 269518.0, "elapsed_s": 4770.8, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9860, "epoch": 0.846, "loss": 1.81998, "ppl": 6.172, "lr": 4.6858e-05, "lr_sched": "cosine", "grad_norm": 0.3287, "tokens": 10328862720, "tok_s": 268917.3, "elapsed_s": 4848.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9880, "epoch": 0.848, "loss": 1.83675, "ppl": 6.276, "lr": 4.6492e-05, "lr_sched": "cosine", "grad_norm": 0.2987, "tokens": 10349813760, "tok_s": 268454.0, "elapsed_s": 4926.3, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9900, "epoch": 0.85, "loss": 1.84696, "ppl": 6.34, "lr": 4.6129e-05, "lr_sched": "cosine", "grad_norm": 0.2693, "tokens": 10370764800, "tok_s": 268964.2, "elapsed_s": 5004.1, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9920, "epoch": 0.851, "loss": 1.86191, "ppl": 6.436, "lr": 4.5771e-05, "lr_sched": "cosine", "grad_norm": 0.2919, "tokens": 10391715840, "tok_s": 270257.6, "elapsed_s": 5081.8, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9940, "epoch": 0.853, "loss": 1.85315, "ppl": 6.38, "lr": 4.5416e-05, "lr_sched": "cosine", "grad_norm": 0.3028, "tokens": 10412666880, "tok_s": 270312.6, "elapsed_s": 5159.5, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9960, "epoch": 0.855, "loss": 1.8162, "ppl": 6.148, "lr": 4.5065e-05, "lr_sched": "cosine", "grad_norm": 0.2705, "tokens": 10433617920, "tok_s": 268821.7, "elapsed_s": 5237.9, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 9980, "epoch": 0.857, "loss": 1.85228, "ppl": 6.374, "lr": 4.4718e-05, "lr_sched": "cosine", "grad_norm": 0.302, "tokens": 10454568960, "tok_s": 268223.6, "elapsed_s": 5316.1, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10000, "epoch": 0.858, "loss": 1.90574, "ppl": 6.724, "lr": 4.4374e-05, "lr_sched": "cosine", "grad_norm": 0.2711, "tokens": 10475520000, "tok_s": 268494.0, "elapsed_s": 5394.3, "vram_gb": 2.78, "ram_pct": 4.1, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10000, "epoch": 0.858, "val_loss": 1.89983, "val_ppl": 6.685, "best_val_ppl": 6.685, "is_best": true}
|
||||
{"step": 10020, "epoch": 0.86, "loss": 1.84914, "ppl": 6.354, "lr": 4.4035e-05, "lr_sched": "cosine", "grad_norm": 0.2677, "tokens": 10496471040, "tok_s": 268508.1, "elapsed_s": 5791.5, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10040, "epoch": 0.862, "loss": 1.84401, "ppl": 6.322, "lr": 4.3699e-05, "lr_sched": "cosine", "grad_norm": 0.2547, "tokens": 10517422080, "tok_s": 270180.0, "elapsed_s": 5869.2, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10060, "epoch": 0.863, "loss": 1.86883, "ppl": 6.481, "lr": 4.3367e-05, "lr_sched": "cosine", "grad_norm": 0.2798, "tokens": 10538373120, "tok_s": 270413.4, "elapsed_s": 5946.9, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10080, "epoch": 0.865, "loss": 1.84381, "ppl": 6.321, "lr": 4.3039e-05, "lr_sched": "cosine", "grad_norm": 0.3107, "tokens": 10559324160, "tok_s": 268577.4, "elapsed_s": 6024.7, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10100, "epoch": 0.867, "loss": 1.99061, "ppl": 7.32, "lr": 4.2715e-05, "lr_sched": "cosine", "grad_norm": 0.3802, "tokens": 10580275200, "tok_s": 269402.6, "elapsed_s": 6102.6, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10120, "epoch": 0.869, "loss": 1.75485, "ppl": 5.783, "lr": 4.2394e-05, "lr_sched": "cosine", "grad_norm": 0.2806, "tokens": 10601226240, "tok_s": 267510.9, "elapsed_s": 6180.5, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10140, "epoch": 0.87, "loss": 1.99572, "ppl": 7.357, "lr": 4.2078e-05, "lr_sched": "cosine", "grad_norm": 0.8364, "tokens": 10622177280, "tok_s": 268305.2, "elapsed_s": 6258.7, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10160, "epoch": 0.872, "loss": 1.86967, "ppl": 6.486, "lr": 4.1765e-05, "lr_sched": "cosine", "grad_norm": 0.3092, "tokens": 10643128320, "tok_s": 268552.8, "elapsed_s": 6336.6, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10180, "epoch": 0.874, "loss": 1.83294, "ppl": 6.252, "lr": 4.1457e-05, "lr_sched": "cosine", "grad_norm": 0.2595, "tokens": 10664079360, "tok_s": 269278.4, "elapsed_s": 6414.5, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10200, "epoch": 0.875, "loss": 1.85806, "ppl": 6.411, "lr": 4.1152e-05, "lr_sched": "cosine", "grad_norm": 0.3023, "tokens": 10685030400, "tok_s": 267794.3, "elapsed_s": 6492.5, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10220, "epoch": 0.877, "loss": 1.80467, "ppl": 6.078, "lr": 4.0851e-05, "lr_sched": "cosine", "grad_norm": 0.293, "tokens": 10705981440, "tok_s": 266895.5, "elapsed_s": 6570.5, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10240, "epoch": 0.879, "loss": 1.78058, "ppl": 5.933, "lr": 4.0555e-05, "lr_sched": "cosine", "grad_norm": 0.2668, "tokens": 10726932480, "tok_s": 266388.7, "elapsed_s": 6648.8, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10260, "epoch": 0.881, "loss": 1.79627, "ppl": 6.027, "lr": 4.0262e-05, "lr_sched": "cosine", "grad_norm": 0.2867, "tokens": 10747883520, "tok_s": 267577.8, "elapsed_s": 6727.3, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10280, "epoch": 0.882, "loss": 1.8656, "ppl": 6.46, "lr": 3.9973e-05, "lr_sched": "cosine", "grad_norm": 0.3036, "tokens": 10768834560, "tok_s": 267173.3, "elapsed_s": 6805.6, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10300, "epoch": 0.884, "loss": 1.78567, "ppl": 5.964, "lr": 3.9688e-05, "lr_sched": "cosine", "grad_norm": 0.29, "tokens": 10789785600, "tok_s": 269065.3, "elapsed_s": 6883.9, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10320, "epoch": 0.886, "loss": 1.80172, "ppl": 6.06, "lr": 3.9407e-05, "lr_sched": "cosine", "grad_norm": 0.2759, "tokens": 10810736640, "tok_s": 268775.9, "elapsed_s": 6961.9, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10340, "epoch": 0.887, "loss": 1.89481, "ppl": 6.651, "lr": 3.913e-05, "lr_sched": "cosine", "grad_norm": 0.2805, "tokens": 10831687680, "tok_s": 269707.8, "elapsed_s": 7039.9, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10360, "epoch": 0.889, "loss": 1.86798, "ppl": 6.475, "lr": 3.8857e-05, "lr_sched": "cosine", "grad_norm": 0.2885, "tokens": 10852638720, "tok_s": 266906.9, "elapsed_s": 7117.9, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10380, "epoch": 0.891, "loss": 1.80931, "ppl": 6.106, "lr": 3.8588e-05, "lr_sched": "cosine", "grad_norm": 0.2764, "tokens": 10873589760, "tok_s": 268031.8, "elapsed_s": 7195.8, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10400, "epoch": 0.893, "loss": 1.86338, "ppl": 6.446, "lr": 3.8323e-05, "lr_sched": "cosine", "grad_norm": 0.2763, "tokens": 10894540800, "tok_s": 269515.4, "elapsed_s": 7273.6, "vram_gb": 2.78, "ram_pct": 4.2, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10400, "epoch": 0.893, "val_loss": 1.896, "val_ppl": 6.659, "best_val_ppl": 6.659, "is_best": true}
|
||||
{"step": 10420, "epoch": 0.894, "loss": 1.83668, "ppl": 6.276, "lr": 3.8062e-05, "lr_sched": "cosine", "grad_norm": 0.2941, "tokens": 10915491840, "tok_s": 269255.4, "elapsed_s": 7673.1, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10440, "epoch": 0.896, "loss": 1.87926, "ppl": 6.549, "lr": 3.7805e-05, "lr_sched": "cosine", "grad_norm": 0.2699, "tokens": 10936442880, "tok_s": 268953.8, "elapsed_s": 7751.2, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10460, "epoch": 0.898, "loss": 1.83896, "ppl": 6.29, "lr": 3.7552e-05, "lr_sched": "cosine", "grad_norm": 0.2742, "tokens": 10957393920, "tok_s": 267855.8, "elapsed_s": 7829.3, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10480, "epoch": 0.899, "loss": 1.87086, "ppl": 6.494, "lr": 3.7304e-05, "lr_sched": "cosine", "grad_norm": 0.2886, "tokens": 10978344960, "tok_s": 268423.4, "elapsed_s": 7907.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10500, "epoch": 0.901, "loss": 1.8841, "ppl": 6.58, "lr": 3.7059e-05, "lr_sched": "cosine", "grad_norm": 0.2775, "tokens": 10999296000, "tok_s": 266778.4, "elapsed_s": 7985.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10520, "epoch": 0.903, "loss": 1.9385, "ppl": 6.948, "lr": 3.6818e-05, "lr_sched": "cosine", "grad_norm": 0.4248, "tokens": 11020247040, "tok_s": 268818.6, "elapsed_s": 8063.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10540, "epoch": 0.905, "loss": 1.82055, "ppl": 6.175, "lr": 3.6582e-05, "lr_sched": "cosine", "grad_norm": 0.3095, "tokens": 11041198080, "tok_s": 266669.7, "elapsed_s": 8141.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10560, "epoch": 0.906, "loss": 1.87304, "ppl": 6.508, "lr": 3.6349e-05, "lr_sched": "cosine", "grad_norm": 0.3212, "tokens": 11062149120, "tok_s": 268009.9, "elapsed_s": 8219.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10580, "epoch": 0.908, "loss": 1.8236, "ppl": 6.194, "lr": 3.6121e-05, "lr_sched": "cosine", "grad_norm": 0.3018, "tokens": 11083100160, "tok_s": 269219.3, "elapsed_s": 8297.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10600, "epoch": 0.91, "loss": 1.74339, "ppl": 5.717, "lr": 3.5896e-05, "lr_sched": "cosine", "grad_norm": 0.3303, "tokens": 11104051200, "tok_s": 269249.3, "elapsed_s": 8375.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10620, "epoch": 0.911, "loss": 1.84348, "ppl": 6.318, "lr": 3.5676e-05, "lr_sched": "cosine", "grad_norm": 0.3023, "tokens": 11125002240, "tok_s": 268201.1, "elapsed_s": 8453.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10640, "epoch": 0.913, "loss": 1.83174, "ppl": 6.245, "lr": 3.546e-05, "lr_sched": "cosine", "grad_norm": 0.3136, "tokens": 11145953280, "tok_s": 268062.0, "elapsed_s": 8531.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10660, "epoch": 0.915, "loss": 1.86049, "ppl": 6.427, "lr": 3.5248e-05, "lr_sched": "cosine", "grad_norm": 0.2733, "tokens": 11166904320, "tok_s": 267553.6, "elapsed_s": 8609.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10680, "epoch": 0.917, "loss": 1.81944, "ppl": 6.168, "lr": 3.504e-05, "lr_sched": "cosine", "grad_norm": 0.3243, "tokens": 11187855360, "tok_s": 269257.8, "elapsed_s": 8687.8, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10700, "epoch": 0.918, "loss": 1.84637, "ppl": 6.337, "lr": 3.4836e-05, "lr_sched": "cosine", "grad_norm": 0.3338, "tokens": 11208806400, "tok_s": 268388.6, "elapsed_s": 8765.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10720, "epoch": 0.92, "loss": 1.82114, "ppl": 6.179, "lr": 3.4636e-05, "lr_sched": "cosine", "grad_norm": 0.3199, "tokens": 11229757440, "tok_s": 268854.1, "elapsed_s": 8844.0, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10740, "epoch": 0.922, "loss": 1.78886, "ppl": 5.983, "lr": 3.4441e-05, "lr_sched": "cosine", "grad_norm": 0.3535, "tokens": 11250708480, "tok_s": 269901.7, "elapsed_s": 8921.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10760, "epoch": 0.923, "loss": 1.86321, "ppl": 6.444, "lr": 3.4249e-05, "lr_sched": "cosine", "grad_norm": 0.2679, "tokens": 11271659520, "tok_s": 269196.4, "elapsed_s": 8999.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10780, "epoch": 0.925, "loss": 1.92557, "ppl": 6.859, "lr": 3.4062e-05, "lr_sched": "cosine", "grad_norm": 0.2609, "tokens": 11292610560, "tok_s": 268816.8, "elapsed_s": 9077.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10800, "epoch": 0.927, "loss": 1.88716, "ppl": 6.601, "lr": 3.3879e-05, "lr_sched": "cosine", "grad_norm": 0.2489, "tokens": 11313561600, "tok_s": 268131.0, "elapsed_s": 9155.8, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.85, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10800, "epoch": 0.927, "val_loss": 1.89445, "val_ppl": 6.649, "best_val_ppl": 6.649, "is_best": true}
|
||||
{"step": 10820, "epoch": 0.929, "loss": 1.8496, "ppl": 6.357, "lr": 3.37e-05, "lr_sched": "cosine", "grad_norm": 0.2849, "tokens": 11334512640, "tok_s": 269642.8, "elapsed_s": 9554.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10840, "epoch": 0.93, "loss": 1.90154, "ppl": 6.696, "lr": 3.3525e-05, "lr_sched": "cosine", "grad_norm": 0.2654, "tokens": 11355463680, "tok_s": 267937.2, "elapsed_s": 9632.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10860, "epoch": 0.932, "loss": 1.84298, "ppl": 6.315, "lr": 3.3355e-05, "lr_sched": "cosine", "grad_norm": 0.2703, "tokens": 11376414720, "tok_s": 268045.3, "elapsed_s": 9710.3, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10880, "epoch": 0.934, "loss": 1.88474, "ppl": 6.585, "lr": 3.3188e-05, "lr_sched": "cosine", "grad_norm": 0.2822, "tokens": 11397365760, "tok_s": 268891.2, "elapsed_s": 9788.0, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10900, "epoch": 0.935, "loss": 1.86963, "ppl": 6.486, "lr": 3.3026e-05, "lr_sched": "cosine", "grad_norm": 0.2737, "tokens": 11418316800, "tok_s": 269174.0, "elapsed_s": 9865.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10920, "epoch": 0.937, "loss": 1.86235, "ppl": 6.439, "lr": 3.2868e-05, "lr_sched": "cosine", "grad_norm": 0.2685, "tokens": 11439267840, "tok_s": 268914.4, "elapsed_s": 9943.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10940, "epoch": 0.939, "loss": 1.82836, "ppl": 6.224, "lr": 3.2714e-05, "lr_sched": "cosine", "grad_norm": 0.2662, "tokens": 11460218880, "tok_s": 268764.1, "elapsed_s": 10021.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10960, "epoch": 0.941, "loss": 1.82516, "ppl": 6.204, "lr": 3.2564e-05, "lr_sched": "cosine", "grad_norm": 0.2566, "tokens": 11481169920, "tok_s": 269449.4, "elapsed_s": 10099.3, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 10980, "epoch": 0.942, "loss": 1.78976, "ppl": 5.988, "lr": 3.2419e-05, "lr_sched": "cosine", "grad_norm": 0.2703, "tokens": 11502120960, "tok_s": 269021.7, "elapsed_s": 10177.1, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11000, "epoch": 0.944, "loss": 1.87938, "ppl": 6.549, "lr": 3.2278e-05, "lr_sched": "cosine", "grad_norm": 0.2862, "tokens": 11523072000, "tok_s": 269527.7, "elapsed_s": 10254.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11020, "epoch": 0.946, "loss": 1.86884, "ppl": 6.481, "lr": 3.2141e-05, "lr_sched": "cosine", "grad_norm": 0.2893, "tokens": 11544023040, "tok_s": 269423.7, "elapsed_s": 10332.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11040, "epoch": 0.947, "loss": 1.77045, "ppl": 5.874, "lr": 3.2008e-05, "lr_sched": "cosine", "grad_norm": 0.2762, "tokens": 11564974080, "tok_s": 268932.0, "elapsed_s": 10410.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11060, "epoch": 0.949, "loss": 1.8436, "ppl": 6.319, "lr": 3.1879e-05, "lr_sched": "cosine", "grad_norm": 0.2839, "tokens": 11585925120, "tok_s": 268462.4, "elapsed_s": 10488.3, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11080, "epoch": 0.951, "loss": 1.83254, "ppl": 6.25, "lr": 3.1755e-05, "lr_sched": "cosine", "grad_norm": 0.2593, "tokens": 11606876160, "tok_s": 269721.7, "elapsed_s": 10566.1, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11100, "epoch": 0.953, "loss": 1.81423, "ppl": 6.136, "lr": 3.1635e-05, "lr_sched": "cosine", "grad_norm": 0.2913, "tokens": 11627827200, "tok_s": 268992.3, "elapsed_s": 10643.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11120, "epoch": 0.954, "loss": 1.83394, "ppl": 6.259, "lr": 3.1519e-05, "lr_sched": "cosine", "grad_norm": 0.2901, "tokens": 11648778240, "tok_s": 269165.8, "elapsed_s": 10721.8, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11140, "epoch": 0.956, "loss": 1.84829, "ppl": 6.349, "lr": 3.1407e-05, "lr_sched": "cosine", "grad_norm": 0.2954, "tokens": 11669729280, "tok_s": 268702.6, "elapsed_s": 10799.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11160, "epoch": 0.958, "loss": 1.88456, "ppl": 6.583, "lr": 3.13e-05, "lr_sched": "cosine", "grad_norm": 0.2636, "tokens": 11690680320, "tok_s": 268959.7, "elapsed_s": 10877.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11180, "epoch": 0.959, "loss": 1.87637, "ppl": 6.53, "lr": 3.1197e-05, "lr_sched": "cosine", "grad_norm": 0.2698, "tokens": 11711631360, "tok_s": 269320.6, "elapsed_s": 10955.4, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11200, "epoch": 0.961, "loss": 1.88903, "ppl": 6.613, "lr": 3.1098e-05, "lr_sched": "cosine", "grad_norm": 0.263, "tokens": 11732582400, "tok_s": 269877.5, "elapsed_s": 11033.2, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11200, "epoch": 0.961, "val_loss": 1.89248, "val_ppl": 6.636, "best_val_ppl": 6.636, "is_best": true}
|
||||
{"step": 11220, "epoch": 0.963, "loss": 1.92929, "ppl": 6.885, "lr": 3.1003e-05, "lr_sched": "cosine", "grad_norm": 0.2958, "tokens": 11753533440, "tok_s": 268538.6, "elapsed_s": 11428.5, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11240, "epoch": 0.965, "loss": 1.89082, "ppl": 6.625, "lr": 3.0913e-05, "lr_sched": "cosine", "grad_norm": 0.2614, "tokens": 11774484480, "tok_s": 269335.0, "elapsed_s": 11506.2, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11260, "epoch": 0.966, "loss": 1.85848, "ppl": 6.414, "lr": 3.0826e-05, "lr_sched": "cosine", "grad_norm": 0.2983, "tokens": 11795435520, "tok_s": 268240.1, "elapsed_s": 11584.0, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11280, "epoch": 0.968, "loss": 1.84137, "ppl": 6.305, "lr": 3.0745e-05, "lr_sched": "cosine", "grad_norm": 0.2807, "tokens": 11816386560, "tok_s": 268500.9, "elapsed_s": 11661.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11300, "epoch": 0.97, "loss": 1.71302, "ppl": 5.546, "lr": 3.0667e-05, "lr_sched": "cosine", "grad_norm": 0.3021, "tokens": 11837337600, "tok_s": 268683.8, "elapsed_s": 11739.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11320, "epoch": 0.972, "loss": 1.84834, "ppl": 6.349, "lr": 3.0594e-05, "lr_sched": "cosine", "grad_norm": 0.3048, "tokens": 11858288640, "tok_s": 270240.1, "elapsed_s": 11817.3, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11340, "epoch": 0.973, "loss": 1.85074, "ppl": 6.365, "lr": 3.0524e-05, "lr_sched": "cosine", "grad_norm": 0.2703, "tokens": 11879239680, "tok_s": 269813.0, "elapsed_s": 11895.0, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11360, "epoch": 0.975, "loss": 1.39704, "ppl": 4.043, "lr": 3.046e-05, "lr_sched": "cosine", "grad_norm": 0.3029, "tokens": 11900190720, "tok_s": 268804.5, "elapsed_s": 11972.8, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11380, "epoch": 0.977, "loss": 1.87909, "ppl": 6.548, "lr": 3.0399e-05, "lr_sched": "cosine", "grad_norm": 0.3174, "tokens": 11921141760, "tok_s": 270384.7, "elapsed_s": 12050.4, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11400, "epoch": 0.978, "loss": 1.8049, "ppl": 6.079, "lr": 3.0343e-05, "lr_sched": "cosine", "grad_norm": 0.2804, "tokens": 11942092800, "tok_s": 270251.3, "elapsed_s": 12128.2, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11420, "epoch": 0.98, "loss": 1.78632, "ppl": 5.967, "lr": 3.0291e-05, "lr_sched": "cosine", "grad_norm": 0.2868, "tokens": 11963043840, "tok_s": 269264.0, "elapsed_s": 12205.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11440, "epoch": 0.982, "loss": 1.81081, "ppl": 6.115, "lr": 3.0243e-05, "lr_sched": "cosine", "grad_norm": 0.2888, "tokens": 11983994880, "tok_s": 269880.8, "elapsed_s": 12283.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11460, "epoch": 0.984, "loss": 1.86827, "ppl": 6.477, "lr": 3.0199e-05, "lr_sched": "cosine", "grad_norm": 0.3069, "tokens": 12004945920, "tok_s": 268846.8, "elapsed_s": 12361.4, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11480, "epoch": 0.985, "loss": 1.81303, "ppl": 6.129, "lr": 3.016e-05, "lr_sched": "cosine", "grad_norm": 0.3257, "tokens": 12025896960, "tok_s": 268923.7, "elapsed_s": 12439.2, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11500, "epoch": 0.987, "loss": 1.87356, "ppl": 6.511, "lr": 3.0125e-05, "lr_sched": "cosine", "grad_norm": 0.2923, "tokens": 12046848000, "tok_s": 269397.7, "elapsed_s": 12516.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11520, "epoch": 0.989, "loss": 1.81918, "ppl": 6.167, "lr": 3.0095e-05, "lr_sched": "cosine", "grad_norm": 0.2609, "tokens": 12067799040, "tok_s": 269716.9, "elapsed_s": 12594.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11540, "epoch": 0.99, "loss": 1.83873, "ppl": 6.289, "lr": 3.0068e-05, "lr_sched": "cosine", "grad_norm": 0.2673, "tokens": 12088750080, "tok_s": 268892.8, "elapsed_s": 12672.4, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11560, "epoch": 0.992, "loss": 1.8156, "ppl": 6.145, "lr": 3.0046e-05, "lr_sched": "cosine", "grad_norm": 0.3319, "tokens": 12109701120, "tok_s": 269900.1, "elapsed_s": 12750.1, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11580, "epoch": 0.994, "loss": 1.8363, "ppl": 6.273, "lr": 3.0029e-05, "lr_sched": "cosine", "grad_norm": 0.2887, "tokens": 12130652160, "tok_s": 269548.5, "elapsed_s": 12827.9, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11600, "epoch": 0.996, "loss": 1.82102, "ppl": 6.178, "lr": 3.0015e-05, "lr_sched": "cosine", "grad_norm": 0.2774, "tokens": 12151603200, "tok_s": 270289.1, "elapsed_s": 12905.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11600, "epoch": 0.996, "val_loss": 1.89176, "val_ppl": 6.631, "best_val_ppl": 6.631, "is_best": true}
|
||||
{"step": 11620, "epoch": 0.997, "loss": 1.84643, "ppl": 6.337, "lr": 3.0006e-05, "lr_sched": "cosine", "grad_norm": 0.3379, "tokens": 12172554240, "tok_s": 270729.8, "elapsed_s": 13300.7, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
{"step": 11640, "epoch": 0.999, "loss": 1.79849, "ppl": 6.041, "lr": 3.0001e-05, "lr_sched": "cosine", "grad_norm": 0.3078, "tokens": 12193505280, "tok_s": 268458.1, "elapsed_s": 13378.6, "vram_gb": 2.78, "ram_pct": 4.4, "disk_free_gb": 9223371968.84, "attn_backend": "chunked", "amp_dtype": "torch.bfloat16"}
|
||||
Reference in New Issue
Block a user