Hugging Face's logo

PRATYUSH-BHARDWAJ
/
Cortex_A_0.5

Transformers
Safetensors
qwen3.5
unsloth
qat
edge
sft
int8-int4
Model card Files
xet
 Community
PRATYUSH-BHARDWAJ commited on
Commit
82cd661
·
verified ·
1 Parent(s): e554442

metrics @ step 24

Browse files
Files changed (1) hide show
  1. runs/metrics.jsonl +3 -0
runs/metrics.jsonl CHANGED
@@ -4,3 +4,6 @@
4
  {"type": "train", "step": 10, "epoch": 0.1739, "loss": 2.455470191107856, "train_ppl": 11.651910885969071, "mtp_loss": null, "grad_norm": 22.719650268554688, "lr": 7.5e-05, "tokens_per_sec": null, "gpu_mem_gb": 2.94, "elapsed_h": 0.027, "budget_h": 0.5}
5
  {"type": "train", "step": 12, "epoch": 0.2087, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.031, "budget_h": 0.5}
6
  {"type": "train", "step": 16, "epoch": 0.2783, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.037, "budget_h": 0.5}
 
 
 
 
4
  {"type": "train", "step": 10, "epoch": 0.1739, "loss": 2.455470191107856, "train_ppl": 11.651910885969071, "mtp_loss": null, "grad_norm": 22.719650268554688, "lr": 7.5e-05, "tokens_per_sec": null, "gpu_mem_gb": 2.94, "elapsed_h": 0.027, "budget_h": 0.5}
5
  {"type": "train", "step": 12, "epoch": 0.2087, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.031, "budget_h": 0.5}
6
  {"type": "train", "step": 16, "epoch": 0.2783, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.037, "budget_h": 0.5}
7
+ {"type": "train", "step": 20, "epoch": 0.3478, "loss": 0.5043737411499023, "train_ppl": 1.6559481434813406, "mtp_loss": null, "grad_norm": 3.528010606765747, "lr": 0.0001425726650926814, "tokens_per_sec": null, "gpu_mem_gb": 3.7, "elapsed_h": 0.047, "budget_h": 0.5}
8
+ {"type": "train", "step": 20, "epoch": 0.3478, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.047, "budget_h": 0.5}
9
+ {"type": "train", "step": 24, "epoch": 0.4174, "loss": null, "train_ppl": null, "mtp_loss": null, "grad_norm": null, "lr": null, "tokens_per_sec": null, "gpu_mem_gb": 2.79, "elapsed_h": 0.053, "budget_h": 0.5}