Training in progress, step 2000, checkpoint

checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4517f955c40e9c14e76a877694d9ba8d1556e85e07793c7b485c52d3b0f4b166
+oid sha256:3bc3fffaeba9b08d3acdff0076fc3c2415f89adbe719d95c5493b80a2b98c42c
 size 1711848436

checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bff352223bf5c6d1804085624846a2e25ff1f9551a1479fd0c84d52f264b7ea8
+oid sha256:9e0f452cff2198fed4da7b97d08e6128513ee3a96e304e76803fd1a1ae175f7c
 size 3424043887

checkpoint-2000/trainer_state.json

@@ -1,5 +1,5 @@
 {
- "best_metric": 1.1584910154342651,
+ "best_metric": 1.159011960029602,
  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l17-l/checkpoint-2000",
  "epoch": 0.21331058020477817,
  "eval_steps": 500,
@@ -10,62 +10,62 @@
  "log_history": [
  {
  "epoch": 0.05332764505119454,
- "grad_norm": 247.07174682617188,
+ "grad_norm": 246.1711883544922,
  "learning_rate": 9.822241183162683e-08,
  "loss": 0.4515,
  "step": 500
  },
  {
  "epoch": 0.05332764505119454,
- "eval_loss": 1.38526451587677,
- "eval_runtime": 63.6956,
- "eval_samples_per_second": 15.496,
- "eval_steps_per_second": 1.947,
+ "eval_loss": 1.3850886821746826,
+ "eval_runtime": 63.8439,
+ "eval_samples_per_second": 15.46,
+ "eval_steps_per_second": 1.942,
  "step": 500
  },
  {
  "epoch": 0.10665529010238908,
- "grad_norm": 72.11767578125,
+ "grad_norm": 81.48503875732422,
  "learning_rate": 9.64448236632537e-08,
- "loss": 0.414,
+ "loss": 0.4148,
  "step": 1000
  },
  {
  "epoch": 0.10665529010238908,
- "eval_loss": 1.284197211265564,
- "eval_runtime": 62.7418,
- "eval_samples_per_second": 15.731,
- "eval_steps_per_second": 1.976,
+ "eval_loss": 1.284144401550293,
+ "eval_runtime": 63.853,
+ "eval_samples_per_second": 15.457,
+ "eval_steps_per_second": 1.942,
  "step": 1000
  },
  {
  "epoch": 0.1599829351535836,
- "grad_norm": 812.2269287109375,
+ "grad_norm": 718.367919921875,
  "learning_rate": 9.466723549488054e-08,
- "loss": 0.3279,
+ "loss": 0.3281,
  "step": 1500
  },
  {
  "epoch": 0.1599829351535836,
- "eval_loss": 1.2100127935409546,
- "eval_runtime": 62.6215,
- "eval_samples_per_second": 15.761,
- "eval_steps_per_second": 1.98,
+ "eval_loss": 1.2112621068954468,
+ "eval_runtime": 64.127,
+ "eval_samples_per_second": 15.391,
+ "eval_steps_per_second": 1.934,
  "step": 1500
  },
  {
  "epoch": 0.21331058020477817,
- "grad_norm": 4.459943771362305,
+ "grad_norm": 4.5923566818237305,
  "learning_rate": 9.288964732650739e-08,
  "loss": 0.2912,
  "step": 2000
  },
  {
  "epoch": 0.21331058020477817,
- "eval_loss": 1.1584910154342651,
- "eval_runtime": 64.0453,
- "eval_samples_per_second": 15.411,
- "eval_steps_per_second": 1.936,
+ "eval_loss": 1.159011960029602,
+ "eval_runtime": 63.7062,
+ "eval_samples_per_second": 15.493,
+ "eval_steps_per_second": 1.946,
  "step": 2000
  }
  ],

checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b52c451dc1fefed1a64b0a23a69d323685c58d3023f0abb7cb18bc82e1be68c7
+oid sha256:52c8db40c27633302368150be011282e1650a78d29cdfc5e366df6966e8e6c70
 size 4847