Upload folder using huggingface_hub
Browse files- checkpoint-19500/config.json +30 -0
- checkpoint-19500/generation_config.json +6 -0
- checkpoint-19500/model.safetensors +3 -0
- checkpoint-19500/optimizer.pt +3 -0
- checkpoint-19500/rng_state.pth +3 -0
- checkpoint-19500/scaler.pt +3 -0
- checkpoint-19500/scheduler.pt +3 -0
- checkpoint-19500/trainer_state.json +619 -0
- checkpoint-19500/training_args.bin +3 -0
- checkpoint-19626/config.json +30 -0
- checkpoint-19626/generation_config.json +6 -0
- checkpoint-19626/model.safetensors +3 -0
- checkpoint-19626/optimizer.pt +3 -0
- checkpoint-19626/rng_state.pth +3 -0
- checkpoint-19626/scaler.pt +3 -0
- checkpoint-19626/scheduler.pt +3 -0
- checkpoint-19626/trainer_state.json +619 -0
- checkpoint-19626/training_args.bin +3 -0
- model.safetensors +1 -1
checkpoint-19500/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151643,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 32768,
|
| 15 |
+
"max_window_layers": 28,
|
| 16 |
+
"model_type": "qwen3",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 28,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rope_theta": 1000000,
|
| 23 |
+
"sliding_window": null,
|
| 24 |
+
"tie_word_embeddings": true,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.51.3",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"use_sliding_window": false,
|
| 29 |
+
"vocab_size": 151936
|
| 30 |
+
}
|
checkpoint-19500/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.51.3"
|
| 6 |
+
}
|
checkpoint-19500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6441cb0667c36efa86887e26994cc8ac604bbafe519c19f30e8df2ed2abbf12
|
| 3 |
+
size 2384234968
|
checkpoint-19500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0cc47f0c7962c11d067f64fa08fbe0a76c4600b783eb293af3ee1bd68993eb04
|
| 3 |
+
size 4768662910
|
checkpoint-19500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d2633b30d6f8bc95e53ac59f43880033523b63ea88408f92c1df3391a619329
|
| 3 |
+
size 14244
|
checkpoint-19500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:357ff841c76a8b9310fd6757f4692ac3f538d732af72abd40e90756f4ce91787
|
| 3 |
+
size 988
|
checkpoint-19500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db6cfc17935b028e31316653ef7c35172a8c2b4b65daf6ffa6510da393169fae
|
| 3 |
+
size 1064
|
checkpoint-19500/trainer_state.json
ADDED
|
@@ -0,0 +1,619 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.9807398349128706,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 19500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07642922653622745,
|
| 14 |
+
"grad_norm": 4.5165534019470215,
|
| 15 |
+
"learning_rate": 4.8738917762152245e-05,
|
| 16 |
+
"loss": 0.89,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.07642922653622745,
|
| 21 |
+
"eval_loss": 0.8767776489257812,
|
| 22 |
+
"eval_runtime": 159.519,
|
| 23 |
+
"eval_samples_per_second": 36.46,
|
| 24 |
+
"eval_steps_per_second": 4.557,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.1528584530724549,
|
| 29 |
+
"grad_norm": 3.430340528488159,
|
| 30 |
+
"learning_rate": 4.746509731988179e-05,
|
| 31 |
+
"loss": 0.8656,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.1528584530724549,
|
| 36 |
+
"eval_loss": 0.8811877965927124,
|
| 37 |
+
"eval_runtime": 157.6806,
|
| 38 |
+
"eval_samples_per_second": 36.885,
|
| 39 |
+
"eval_steps_per_second": 4.611,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.22928767960868235,
|
| 44 |
+
"grad_norm": 3.6226563453674316,
|
| 45 |
+
"learning_rate": 4.619127687761133e-05,
|
| 46 |
+
"loss": 0.8897,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.22928767960868235,
|
| 51 |
+
"eval_loss": 0.8792645931243896,
|
| 52 |
+
"eval_runtime": 157.5757,
|
| 53 |
+
"eval_samples_per_second": 36.909,
|
| 54 |
+
"eval_steps_per_second": 4.614,
|
| 55 |
+
"step": 1500
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.3057169061449098,
|
| 59 |
+
"grad_norm": 2.8311519622802734,
|
| 60 |
+
"learning_rate": 4.4917456435340875e-05,
|
| 61 |
+
"loss": 0.8921,
|
| 62 |
+
"step": 2000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.3057169061449098,
|
| 66 |
+
"eval_loss": 0.8744860291481018,
|
| 67 |
+
"eval_runtime": 157.645,
|
| 68 |
+
"eval_samples_per_second": 36.893,
|
| 69 |
+
"eval_steps_per_second": 4.612,
|
| 70 |
+
"step": 2000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.3821461326811373,
|
| 74 |
+
"grad_norm": 2.6471974849700928,
|
| 75 |
+
"learning_rate": 4.364363599307042e-05,
|
| 76 |
+
"loss": 0.8826,
|
| 77 |
+
"step": 2500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.3821461326811373,
|
| 81 |
+
"eval_loss": 0.8666115403175354,
|
| 82 |
+
"eval_runtime": 157.6483,
|
| 83 |
+
"eval_samples_per_second": 36.892,
|
| 84 |
+
"eval_steps_per_second": 4.612,
|
| 85 |
+
"step": 2500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.4585753592173647,
|
| 89 |
+
"grad_norm": 3.0229456424713135,
|
| 90 |
+
"learning_rate": 4.236981555079996e-05,
|
| 91 |
+
"loss": 0.8613,
|
| 92 |
+
"step": 3000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.4585753592173647,
|
| 96 |
+
"eval_loss": 0.8605388402938843,
|
| 97 |
+
"eval_runtime": 157.698,
|
| 98 |
+
"eval_samples_per_second": 36.881,
|
| 99 |
+
"eval_steps_per_second": 4.61,
|
| 100 |
+
"step": 3000
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5350045857535922,
|
| 104 |
+
"grad_norm": 3.2055766582489014,
|
| 105 |
+
"learning_rate": 4.1095995108529505e-05,
|
| 106 |
+
"loss": 0.8648,
|
| 107 |
+
"step": 3500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5350045857535922,
|
| 111 |
+
"eval_loss": 0.8576663732528687,
|
| 112 |
+
"eval_runtime": 157.7251,
|
| 113 |
+
"eval_samples_per_second": 36.874,
|
| 114 |
+
"eval_steps_per_second": 4.609,
|
| 115 |
+
"step": 3500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.6114338122898196,
|
| 119 |
+
"grad_norm": 2.2706174850463867,
|
| 120 |
+
"learning_rate": 3.982217466625904e-05,
|
| 121 |
+
"loss": 0.8607,
|
| 122 |
+
"step": 4000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.6114338122898196,
|
| 126 |
+
"eval_loss": 0.8507756590843201,
|
| 127 |
+
"eval_runtime": 157.6341,
|
| 128 |
+
"eval_samples_per_second": 36.896,
|
| 129 |
+
"eval_steps_per_second": 4.612,
|
| 130 |
+
"step": 4000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.687863038826047,
|
| 134 |
+
"grad_norm": 3.0524044036865234,
|
| 135 |
+
"learning_rate": 3.8548354223988585e-05,
|
| 136 |
+
"loss": 0.863,
|
| 137 |
+
"step": 4500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.687863038826047,
|
| 141 |
+
"eval_loss": 0.8432514667510986,
|
| 142 |
+
"eval_runtime": 157.6289,
|
| 143 |
+
"eval_samples_per_second": 36.897,
|
| 144 |
+
"eval_steps_per_second": 4.612,
|
| 145 |
+
"step": 4500
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.7642922653622746,
|
| 149 |
+
"grad_norm": 2.707669258117676,
|
| 150 |
+
"learning_rate": 3.727453378171813e-05,
|
| 151 |
+
"loss": 0.8444,
|
| 152 |
+
"step": 5000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.7642922653622746,
|
| 156 |
+
"eval_loss": 0.8389096856117249,
|
| 157 |
+
"eval_runtime": 157.6926,
|
| 158 |
+
"eval_samples_per_second": 36.882,
|
| 159 |
+
"eval_steps_per_second": 4.61,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.840721491898502,
|
| 164 |
+
"grad_norm": 3.0052075386047363,
|
| 165 |
+
"learning_rate": 3.600071333944767e-05,
|
| 166 |
+
"loss": 0.871,
|
| 167 |
+
"step": 5500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.840721491898502,
|
| 171 |
+
"eval_loss": 0.8305906057357788,
|
| 172 |
+
"eval_runtime": 157.7772,
|
| 173 |
+
"eval_samples_per_second": 36.862,
|
| 174 |
+
"eval_steps_per_second": 4.608,
|
| 175 |
+
"step": 5500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.9171507184347294,
|
| 179 |
+
"grad_norm": 1.7623426914215088,
|
| 180 |
+
"learning_rate": 3.4726892897177216e-05,
|
| 181 |
+
"loss": 0.8328,
|
| 182 |
+
"step": 6000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.9171507184347294,
|
| 186 |
+
"eval_loss": 0.8280592560768127,
|
| 187 |
+
"eval_runtime": 157.7986,
|
| 188 |
+
"eval_samples_per_second": 36.857,
|
| 189 |
+
"eval_steps_per_second": 4.607,
|
| 190 |
+
"step": 6000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.9935799449709569,
|
| 194 |
+
"grad_norm": 2.850409746170044,
|
| 195 |
+
"learning_rate": 3.345307245490676e-05,
|
| 196 |
+
"loss": 0.835,
|
| 197 |
+
"step": 6500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.9935799449709569,
|
| 201 |
+
"eval_loss": 0.8225808143615723,
|
| 202 |
+
"eval_runtime": 157.7059,
|
| 203 |
+
"eval_samples_per_second": 36.879,
|
| 204 |
+
"eval_steps_per_second": 4.61,
|
| 205 |
+
"step": 6500
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.0700091715071844,
|
| 209 |
+
"grad_norm": 2.08107590675354,
|
| 210 |
+
"learning_rate": 3.21792520126363e-05,
|
| 211 |
+
"loss": 0.5759,
|
| 212 |
+
"step": 7000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0700091715071844,
|
| 216 |
+
"eval_loss": 0.8543522357940674,
|
| 217 |
+
"eval_runtime": 157.8069,
|
| 218 |
+
"eval_samples_per_second": 36.855,
|
| 219 |
+
"eval_steps_per_second": 4.607,
|
| 220 |
+
"step": 7000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.1464383980434119,
|
| 224 |
+
"grad_norm": 2.4801783561706543,
|
| 225 |
+
"learning_rate": 3.0905431570365846e-05,
|
| 226 |
+
"loss": 0.5493,
|
| 227 |
+
"step": 7500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.1464383980434119,
|
| 231 |
+
"eval_loss": 0.8509367108345032,
|
| 232 |
+
"eval_runtime": 157.6691,
|
| 233 |
+
"eval_samples_per_second": 36.887,
|
| 234 |
+
"eval_steps_per_second": 4.611,
|
| 235 |
+
"step": 7500
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.2228676245796393,
|
| 239 |
+
"grad_norm": 2.688427686691284,
|
| 240 |
+
"learning_rate": 2.963161112809539e-05,
|
| 241 |
+
"loss": 0.5516,
|
| 242 |
+
"step": 8000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.2228676245796393,
|
| 246 |
+
"eval_loss": 0.8434808254241943,
|
| 247 |
+
"eval_runtime": 157.6951,
|
| 248 |
+
"eval_samples_per_second": 36.881,
|
| 249 |
+
"eval_steps_per_second": 4.61,
|
| 250 |
+
"step": 8000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.2992968511158667,
|
| 254 |
+
"grad_norm": 2.8583438396453857,
|
| 255 |
+
"learning_rate": 2.8357790685824926e-05,
|
| 256 |
+
"loss": 0.5608,
|
| 257 |
+
"step": 8500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.2992968511158667,
|
| 261 |
+
"eval_loss": 0.8415189981460571,
|
| 262 |
+
"eval_runtime": 157.7043,
|
| 263 |
+
"eval_samples_per_second": 36.879,
|
| 264 |
+
"eval_steps_per_second": 4.61,
|
| 265 |
+
"step": 8500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.375726077652094,
|
| 269 |
+
"grad_norm": 2.8310320377349854,
|
| 270 |
+
"learning_rate": 2.708397024355447e-05,
|
| 271 |
+
"loss": 0.5468,
|
| 272 |
+
"step": 9000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.375726077652094,
|
| 276 |
+
"eval_loss": 0.8396986126899719,
|
| 277 |
+
"eval_runtime": 157.7062,
|
| 278 |
+
"eval_samples_per_second": 36.879,
|
| 279 |
+
"eval_steps_per_second": 4.61,
|
| 280 |
+
"step": 9000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.4521553041883215,
|
| 284 |
+
"grad_norm": 3.0906243324279785,
|
| 285 |
+
"learning_rate": 2.5810149801284013e-05,
|
| 286 |
+
"loss": 0.5499,
|
| 287 |
+
"step": 9500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.4521553041883215,
|
| 291 |
+
"eval_loss": 0.8367328643798828,
|
| 292 |
+
"eval_runtime": 157.7916,
|
| 293 |
+
"eval_samples_per_second": 36.859,
|
| 294 |
+
"eval_steps_per_second": 4.607,
|
| 295 |
+
"step": 9500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.5285845307245491,
|
| 299 |
+
"grad_norm": 3.706326723098755,
|
| 300 |
+
"learning_rate": 2.4536329359013556e-05,
|
| 301 |
+
"loss": 0.5503,
|
| 302 |
+
"step": 10000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.5285845307245491,
|
| 306 |
+
"eval_loss": 0.8307807445526123,
|
| 307 |
+
"eval_runtime": 157.6389,
|
| 308 |
+
"eval_samples_per_second": 36.894,
|
| 309 |
+
"eval_steps_per_second": 4.612,
|
| 310 |
+
"step": 10000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.6050137572607766,
|
| 314 |
+
"grad_norm": 2.6108150482177734,
|
| 315 |
+
"learning_rate": 2.3262508916743096e-05,
|
| 316 |
+
"loss": 0.5388,
|
| 317 |
+
"step": 10500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.6050137572607766,
|
| 321 |
+
"eval_loss": 0.8295947313308716,
|
| 322 |
+
"eval_runtime": 157.6638,
|
| 323 |
+
"eval_samples_per_second": 36.889,
|
| 324 |
+
"eval_steps_per_second": 4.611,
|
| 325 |
+
"step": 10500
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.681442983797004,
|
| 329 |
+
"grad_norm": 1.6078243255615234,
|
| 330 |
+
"learning_rate": 2.1991236115357182e-05,
|
| 331 |
+
"loss": 0.5473,
|
| 332 |
+
"step": 11000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.681442983797004,
|
| 336 |
+
"eval_loss": 0.8229663372039795,
|
| 337 |
+
"eval_runtime": 157.6846,
|
| 338 |
+
"eval_samples_per_second": 36.884,
|
| 339 |
+
"eval_steps_per_second": 4.61,
|
| 340 |
+
"step": 11000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.7578722103332314,
|
| 344 |
+
"grad_norm": 3.049797773361206,
|
| 345 |
+
"learning_rate": 2.0717415673086722e-05,
|
| 346 |
+
"loss": 0.5496,
|
| 347 |
+
"step": 11500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.7578722103332314,
|
| 351 |
+
"eval_loss": 0.8267400860786438,
|
| 352 |
+
"eval_runtime": 157.7336,
|
| 353 |
+
"eval_samples_per_second": 36.872,
|
| 354 |
+
"eval_steps_per_second": 4.609,
|
| 355 |
+
"step": 11500
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.8343014368694588,
|
| 359 |
+
"grad_norm": 2.292538642883301,
|
| 360 |
+
"learning_rate": 1.9443595230816262e-05,
|
| 361 |
+
"loss": 0.5448,
|
| 362 |
+
"step": 12000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.8343014368694588,
|
| 366 |
+
"eval_loss": 0.8191345930099487,
|
| 367 |
+
"eval_runtime": 158.4293,
|
| 368 |
+
"eval_samples_per_second": 36.71,
|
| 369 |
+
"eval_steps_per_second": 4.589,
|
| 370 |
+
"step": 12000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.9107306634056864,
|
| 374 |
+
"grad_norm": 2.1699585914611816,
|
| 375 |
+
"learning_rate": 1.8169774788545806e-05,
|
| 376 |
+
"loss": 0.5419,
|
| 377 |
+
"step": 12500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.9107306634056864,
|
| 381 |
+
"eval_loss": 0.8131210803985596,
|
| 382 |
+
"eval_runtime": 157.6629,
|
| 383 |
+
"eval_samples_per_second": 36.889,
|
| 384 |
+
"eval_steps_per_second": 4.611,
|
| 385 |
+
"step": 12500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.9871598899419138,
|
| 389 |
+
"grad_norm": 3.1323328018188477,
|
| 390 |
+
"learning_rate": 1.689595434627535e-05,
|
| 391 |
+
"loss": 0.5369,
|
| 392 |
+
"step": 13000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.9871598899419138,
|
| 396 |
+
"eval_loss": 0.8066145777702332,
|
| 397 |
+
"eval_runtime": 157.7959,
|
| 398 |
+
"eval_samples_per_second": 36.858,
|
| 399 |
+
"eval_steps_per_second": 4.607,
|
| 400 |
+
"step": 13000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.0635891164781412,
|
| 404 |
+
"grad_norm": 3.656402826309204,
|
| 405 |
+
"learning_rate": 1.562468154488943e-05,
|
| 406 |
+
"loss": 0.3304,
|
| 407 |
+
"step": 13500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.0635891164781412,
|
| 411 |
+
"eval_loss": 0.9408266544342041,
|
| 412 |
+
"eval_runtime": 157.7674,
|
| 413 |
+
"eval_samples_per_second": 36.864,
|
| 414 |
+
"eval_steps_per_second": 4.608,
|
| 415 |
+
"step": 13500
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.140018343014369,
|
| 419 |
+
"grad_norm": 2.4427192211151123,
|
| 420 |
+
"learning_rate": 1.4350861102618977e-05,
|
| 421 |
+
"loss": 0.2759,
|
| 422 |
+
"step": 14000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.140018343014369,
|
| 426 |
+
"eval_loss": 0.942986011505127,
|
| 427 |
+
"eval_runtime": 157.7087,
|
| 428 |
+
"eval_samples_per_second": 36.878,
|
| 429 |
+
"eval_steps_per_second": 4.61,
|
| 430 |
+
"step": 14000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.216447569550596,
|
| 434 |
+
"grad_norm": 1.6041910648345947,
|
| 435 |
+
"learning_rate": 1.3077040660348518e-05,
|
| 436 |
+
"loss": 0.2873,
|
| 437 |
+
"step": 14500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 2.216447569550596,
|
| 441 |
+
"eval_loss": 0.9449612498283386,
|
| 442 |
+
"eval_runtime": 157.767,
|
| 443 |
+
"eval_samples_per_second": 36.865,
|
| 444 |
+
"eval_steps_per_second": 4.608,
|
| 445 |
+
"step": 14500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.2928767960868237,
|
| 449 |
+
"grad_norm": 3.233290433883667,
|
| 450 |
+
"learning_rate": 1.180322021807806e-05,
|
| 451 |
+
"loss": 0.2818,
|
| 452 |
+
"step": 15000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.2928767960868237,
|
| 456 |
+
"eval_loss": 0.9387638568878174,
|
| 457 |
+
"eval_runtime": 157.7297,
|
| 458 |
+
"eval_samples_per_second": 36.873,
|
| 459 |
+
"eval_steps_per_second": 4.609,
|
| 460 |
+
"step": 15000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.369306022623051,
|
| 464 |
+
"grad_norm": 3.9653565883636475,
|
| 465 |
+
"learning_rate": 1.0529399775807602e-05,
|
| 466 |
+
"loss": 0.2795,
|
| 467 |
+
"step": 15500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.369306022623051,
|
| 471 |
+
"eval_loss": 0.9435889720916748,
|
| 472 |
+
"eval_runtime": 157.7236,
|
| 473 |
+
"eval_samples_per_second": 36.875,
|
| 474 |
+
"eval_steps_per_second": 4.609,
|
| 475 |
+
"step": 15500
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 2.4457352491592785,
|
| 479 |
+
"grad_norm": 2.453057289123535,
|
| 480 |
+
"learning_rate": 9.255579333537145e-06,
|
| 481 |
+
"loss": 0.2801,
|
| 482 |
+
"step": 16000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.4457352491592785,
|
| 486 |
+
"eval_loss": 0.9410313963890076,
|
| 487 |
+
"eval_runtime": 157.7031,
|
| 488 |
+
"eval_samples_per_second": 36.879,
|
| 489 |
+
"eval_steps_per_second": 4.61,
|
| 490 |
+
"step": 16000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.522164475695506,
|
| 494 |
+
"grad_norm": 2.9924492835998535,
|
| 495 |
+
"learning_rate": 7.981758891266687e-06,
|
| 496 |
+
"loss": 0.2788,
|
| 497 |
+
"step": 16500
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.522164475695506,
|
| 501 |
+
"eval_loss": 0.9427609443664551,
|
| 502 |
+
"eval_runtime": 157.6635,
|
| 503 |
+
"eval_samples_per_second": 36.889,
|
| 504 |
+
"eval_steps_per_second": 4.611,
|
| 505 |
+
"step": 16500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 2.5985937022317334,
|
| 509 |
+
"grad_norm": 2.626593828201294,
|
| 510 |
+
"learning_rate": 6.7079384489962305e-06,
|
| 511 |
+
"loss": 0.2752,
|
| 512 |
+
"step": 17000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.5985937022317334,
|
| 516 |
+
"eval_loss": 0.9421259164810181,
|
| 517 |
+
"eval_runtime": 157.689,
|
| 518 |
+
"eval_samples_per_second": 36.883,
|
| 519 |
+
"eval_steps_per_second": 4.61,
|
| 520 |
+
"step": 17000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.675022928767961,
|
| 524 |
+
"grad_norm": 2.623121500015259,
|
| 525 |
+
"learning_rate": 5.436665647610313e-06,
|
| 526 |
+
"loss": 0.2695,
|
| 527 |
+
"step": 17500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.675022928767961,
|
| 531 |
+
"eval_loss": 0.9395164251327515,
|
| 532 |
+
"eval_runtime": 157.7665,
|
| 533 |
+
"eval_samples_per_second": 36.865,
|
| 534 |
+
"eval_steps_per_second": 4.608,
|
| 535 |
+
"step": 17500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.751452155304188,
|
| 539 |
+
"grad_norm": 2.4113175868988037,
|
| 540 |
+
"learning_rate": 4.1653928462243965e-06,
|
| 541 |
+
"loss": 0.2697,
|
| 542 |
+
"step": 18000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.751452155304188,
|
| 546 |
+
"eval_loss": 0.9405816197395325,
|
| 547 |
+
"eval_runtime": 157.6703,
|
| 548 |
+
"eval_samples_per_second": 36.887,
|
| 549 |
+
"eval_steps_per_second": 4.611,
|
| 550 |
+
"step": 18000
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.827881381840416,
|
| 554 |
+
"grad_norm": 3.3730709552764893,
|
| 555 |
+
"learning_rate": 2.8915724039539386e-06,
|
| 556 |
+
"loss": 0.2769,
|
| 557 |
+
"step": 18500
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.827881381840416,
|
| 561 |
+
"eval_loss": 0.9389672875404358,
|
| 562 |
+
"eval_runtime": 157.689,
|
| 563 |
+
"eval_samples_per_second": 36.883,
|
| 564 |
+
"eval_steps_per_second": 4.61,
|
| 565 |
+
"step": 18500
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 2.904310608376643,
|
| 569 |
+
"grad_norm": 3.464594841003418,
|
| 570 |
+
"learning_rate": 1.6177519616834812e-06,
|
| 571 |
+
"loss": 0.271,
|
| 572 |
+
"step": 19000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 2.904310608376643,
|
| 576 |
+
"eval_loss": 0.9393123984336853,
|
| 577 |
+
"eval_runtime": 157.8608,
|
| 578 |
+
"eval_samples_per_second": 36.843,
|
| 579 |
+
"eval_steps_per_second": 4.605,
|
| 580 |
+
"step": 19000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.9807398349128706,
|
| 584 |
+
"grad_norm": 2.9936461448669434,
|
| 585 |
+
"learning_rate": 3.439315194130236e-07,
|
| 586 |
+
"loss": 0.2584,
|
| 587 |
+
"step": 19500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.9807398349128706,
|
| 591 |
+
"eval_loss": 0.9363918900489807,
|
| 592 |
+
"eval_runtime": 157.724,
|
| 593 |
+
"eval_samples_per_second": 36.875,
|
| 594 |
+
"eval_steps_per_second": 4.609,
|
| 595 |
+
"step": 19500
|
| 596 |
+
}
|
| 597 |
+
],
|
| 598 |
+
"logging_steps": 500,
|
| 599 |
+
"max_steps": 19626,
|
| 600 |
+
"num_input_tokens_seen": 0,
|
| 601 |
+
"num_train_epochs": 3,
|
| 602 |
+
"save_steps": 500,
|
| 603 |
+
"stateful_callbacks": {
|
| 604 |
+
"TrainerControl": {
|
| 605 |
+
"args": {
|
| 606 |
+
"should_epoch_stop": false,
|
| 607 |
+
"should_evaluate": false,
|
| 608 |
+
"should_log": false,
|
| 609 |
+
"should_save": true,
|
| 610 |
+
"should_training_stop": false
|
| 611 |
+
},
|
| 612 |
+
"attributes": {}
|
| 613 |
+
}
|
| 614 |
+
},
|
| 615 |
+
"total_flos": 2.11086099873792e+17,
|
| 616 |
+
"train_batch_size": 2,
|
| 617 |
+
"trial_name": null,
|
| 618 |
+
"trial_params": null
|
| 619 |
+
}
|
checkpoint-19500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1847242637dff66547b119c57e8cfeaf293fec9ad4dd17713f23bb7846bba98d
|
| 3 |
+
size 5240
|
checkpoint-19626/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151643,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 32768,
|
| 15 |
+
"max_window_layers": 28,
|
| 16 |
+
"model_type": "qwen3",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 28,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rope_theta": 1000000,
|
| 23 |
+
"sliding_window": null,
|
| 24 |
+
"tie_word_embeddings": true,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.51.3",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"use_sliding_window": false,
|
| 29 |
+
"vocab_size": 151936
|
| 30 |
+
}
|
checkpoint-19626/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.51.3"
|
| 6 |
+
}
|
checkpoint-19626/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:221b87618292959e3626a872c0881b82b65e408b3b772ca863840da63636cfa7
|
| 3 |
+
size 2384234968
|
checkpoint-19626/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9fd858b3a0d13c57158816a95955f71b965ed779f30438ccb99ed3e54531cd07
|
| 3 |
+
size 4768662910
|
checkpoint-19626/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d2633b30d6f8bc95e53ac59f43880033523b63ea88408f92c1df3391a619329
|
| 3 |
+
size 14244
|
checkpoint-19626/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db2586e965071efd3a640555799509c9c7b0e0ad7b6e513007361266359886b2
|
| 3 |
+
size 988
|
checkpoint-19626/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ea91cd0fa2b2456602fba879b4cd74e2b83e776ba391db299c617a1e441b75a
|
| 3 |
+
size 1064
|
checkpoint-19626/trainer_state.json
ADDED
|
@@ -0,0 +1,619 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 19626,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.07642922653622745,
|
| 14 |
+
"grad_norm": 4.5165534019470215,
|
| 15 |
+
"learning_rate": 4.8738917762152245e-05,
|
| 16 |
+
"loss": 0.89,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.07642922653622745,
|
| 21 |
+
"eval_loss": 0.8767776489257812,
|
| 22 |
+
"eval_runtime": 159.519,
|
| 23 |
+
"eval_samples_per_second": 36.46,
|
| 24 |
+
"eval_steps_per_second": 4.557,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.1528584530724549,
|
| 29 |
+
"grad_norm": 3.430340528488159,
|
| 30 |
+
"learning_rate": 4.746509731988179e-05,
|
| 31 |
+
"loss": 0.8656,
|
| 32 |
+
"step": 1000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.1528584530724549,
|
| 36 |
+
"eval_loss": 0.8811877965927124,
|
| 37 |
+
"eval_runtime": 157.6806,
|
| 38 |
+
"eval_samples_per_second": 36.885,
|
| 39 |
+
"eval_steps_per_second": 4.611,
|
| 40 |
+
"step": 1000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.22928767960868235,
|
| 44 |
+
"grad_norm": 3.6226563453674316,
|
| 45 |
+
"learning_rate": 4.619127687761133e-05,
|
| 46 |
+
"loss": 0.8897,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.22928767960868235,
|
| 51 |
+
"eval_loss": 0.8792645931243896,
|
| 52 |
+
"eval_runtime": 157.5757,
|
| 53 |
+
"eval_samples_per_second": 36.909,
|
| 54 |
+
"eval_steps_per_second": 4.614,
|
| 55 |
+
"step": 1500
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.3057169061449098,
|
| 59 |
+
"grad_norm": 2.8311519622802734,
|
| 60 |
+
"learning_rate": 4.4917456435340875e-05,
|
| 61 |
+
"loss": 0.8921,
|
| 62 |
+
"step": 2000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.3057169061449098,
|
| 66 |
+
"eval_loss": 0.8744860291481018,
|
| 67 |
+
"eval_runtime": 157.645,
|
| 68 |
+
"eval_samples_per_second": 36.893,
|
| 69 |
+
"eval_steps_per_second": 4.612,
|
| 70 |
+
"step": 2000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.3821461326811373,
|
| 74 |
+
"grad_norm": 2.6471974849700928,
|
| 75 |
+
"learning_rate": 4.364363599307042e-05,
|
| 76 |
+
"loss": 0.8826,
|
| 77 |
+
"step": 2500
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.3821461326811373,
|
| 81 |
+
"eval_loss": 0.8666115403175354,
|
| 82 |
+
"eval_runtime": 157.6483,
|
| 83 |
+
"eval_samples_per_second": 36.892,
|
| 84 |
+
"eval_steps_per_second": 4.612,
|
| 85 |
+
"step": 2500
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.4585753592173647,
|
| 89 |
+
"grad_norm": 3.0229456424713135,
|
| 90 |
+
"learning_rate": 4.236981555079996e-05,
|
| 91 |
+
"loss": 0.8613,
|
| 92 |
+
"step": 3000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.4585753592173647,
|
| 96 |
+
"eval_loss": 0.8605388402938843,
|
| 97 |
+
"eval_runtime": 157.698,
|
| 98 |
+
"eval_samples_per_second": 36.881,
|
| 99 |
+
"eval_steps_per_second": 4.61,
|
| 100 |
+
"step": 3000
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.5350045857535922,
|
| 104 |
+
"grad_norm": 3.2055766582489014,
|
| 105 |
+
"learning_rate": 4.1095995108529505e-05,
|
| 106 |
+
"loss": 0.8648,
|
| 107 |
+
"step": 3500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5350045857535922,
|
| 111 |
+
"eval_loss": 0.8576663732528687,
|
| 112 |
+
"eval_runtime": 157.7251,
|
| 113 |
+
"eval_samples_per_second": 36.874,
|
| 114 |
+
"eval_steps_per_second": 4.609,
|
| 115 |
+
"step": 3500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.6114338122898196,
|
| 119 |
+
"grad_norm": 2.2706174850463867,
|
| 120 |
+
"learning_rate": 3.982217466625904e-05,
|
| 121 |
+
"loss": 0.8607,
|
| 122 |
+
"step": 4000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.6114338122898196,
|
| 126 |
+
"eval_loss": 0.8507756590843201,
|
| 127 |
+
"eval_runtime": 157.6341,
|
| 128 |
+
"eval_samples_per_second": 36.896,
|
| 129 |
+
"eval_steps_per_second": 4.612,
|
| 130 |
+
"step": 4000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.687863038826047,
|
| 134 |
+
"grad_norm": 3.0524044036865234,
|
| 135 |
+
"learning_rate": 3.8548354223988585e-05,
|
| 136 |
+
"loss": 0.863,
|
| 137 |
+
"step": 4500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.687863038826047,
|
| 141 |
+
"eval_loss": 0.8432514667510986,
|
| 142 |
+
"eval_runtime": 157.6289,
|
| 143 |
+
"eval_samples_per_second": 36.897,
|
| 144 |
+
"eval_steps_per_second": 4.612,
|
| 145 |
+
"step": 4500
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.7642922653622746,
|
| 149 |
+
"grad_norm": 2.707669258117676,
|
| 150 |
+
"learning_rate": 3.727453378171813e-05,
|
| 151 |
+
"loss": 0.8444,
|
| 152 |
+
"step": 5000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.7642922653622746,
|
| 156 |
+
"eval_loss": 0.8389096856117249,
|
| 157 |
+
"eval_runtime": 157.6926,
|
| 158 |
+
"eval_samples_per_second": 36.882,
|
| 159 |
+
"eval_steps_per_second": 4.61,
|
| 160 |
+
"step": 5000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.840721491898502,
|
| 164 |
+
"grad_norm": 3.0052075386047363,
|
| 165 |
+
"learning_rate": 3.600071333944767e-05,
|
| 166 |
+
"loss": 0.871,
|
| 167 |
+
"step": 5500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.840721491898502,
|
| 171 |
+
"eval_loss": 0.8305906057357788,
|
| 172 |
+
"eval_runtime": 157.7772,
|
| 173 |
+
"eval_samples_per_second": 36.862,
|
| 174 |
+
"eval_steps_per_second": 4.608,
|
| 175 |
+
"step": 5500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.9171507184347294,
|
| 179 |
+
"grad_norm": 1.7623426914215088,
|
| 180 |
+
"learning_rate": 3.4726892897177216e-05,
|
| 181 |
+
"loss": 0.8328,
|
| 182 |
+
"step": 6000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.9171507184347294,
|
| 186 |
+
"eval_loss": 0.8280592560768127,
|
| 187 |
+
"eval_runtime": 157.7986,
|
| 188 |
+
"eval_samples_per_second": 36.857,
|
| 189 |
+
"eval_steps_per_second": 4.607,
|
| 190 |
+
"step": 6000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.9935799449709569,
|
| 194 |
+
"grad_norm": 2.850409746170044,
|
| 195 |
+
"learning_rate": 3.345307245490676e-05,
|
| 196 |
+
"loss": 0.835,
|
| 197 |
+
"step": 6500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.9935799449709569,
|
| 201 |
+
"eval_loss": 0.8225808143615723,
|
| 202 |
+
"eval_runtime": 157.7059,
|
| 203 |
+
"eval_samples_per_second": 36.879,
|
| 204 |
+
"eval_steps_per_second": 4.61,
|
| 205 |
+
"step": 6500
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.0700091715071844,
|
| 209 |
+
"grad_norm": 2.08107590675354,
|
| 210 |
+
"learning_rate": 3.21792520126363e-05,
|
| 211 |
+
"loss": 0.5759,
|
| 212 |
+
"step": 7000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0700091715071844,
|
| 216 |
+
"eval_loss": 0.8543522357940674,
|
| 217 |
+
"eval_runtime": 157.8069,
|
| 218 |
+
"eval_samples_per_second": 36.855,
|
| 219 |
+
"eval_steps_per_second": 4.607,
|
| 220 |
+
"step": 7000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.1464383980434119,
|
| 224 |
+
"grad_norm": 2.4801783561706543,
|
| 225 |
+
"learning_rate": 3.0905431570365846e-05,
|
| 226 |
+
"loss": 0.5493,
|
| 227 |
+
"step": 7500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.1464383980434119,
|
| 231 |
+
"eval_loss": 0.8509367108345032,
|
| 232 |
+
"eval_runtime": 157.6691,
|
| 233 |
+
"eval_samples_per_second": 36.887,
|
| 234 |
+
"eval_steps_per_second": 4.611,
|
| 235 |
+
"step": 7500
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 1.2228676245796393,
|
| 239 |
+
"grad_norm": 2.688427686691284,
|
| 240 |
+
"learning_rate": 2.963161112809539e-05,
|
| 241 |
+
"loss": 0.5516,
|
| 242 |
+
"step": 8000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 1.2228676245796393,
|
| 246 |
+
"eval_loss": 0.8434808254241943,
|
| 247 |
+
"eval_runtime": 157.6951,
|
| 248 |
+
"eval_samples_per_second": 36.881,
|
| 249 |
+
"eval_steps_per_second": 4.61,
|
| 250 |
+
"step": 8000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.2992968511158667,
|
| 254 |
+
"grad_norm": 2.8583438396453857,
|
| 255 |
+
"learning_rate": 2.8357790685824926e-05,
|
| 256 |
+
"loss": 0.5608,
|
| 257 |
+
"step": 8500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.2992968511158667,
|
| 261 |
+
"eval_loss": 0.8415189981460571,
|
| 262 |
+
"eval_runtime": 157.7043,
|
| 263 |
+
"eval_samples_per_second": 36.879,
|
| 264 |
+
"eval_steps_per_second": 4.61,
|
| 265 |
+
"step": 8500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.375726077652094,
|
| 269 |
+
"grad_norm": 2.8310320377349854,
|
| 270 |
+
"learning_rate": 2.708397024355447e-05,
|
| 271 |
+
"loss": 0.5468,
|
| 272 |
+
"step": 9000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.375726077652094,
|
| 276 |
+
"eval_loss": 0.8396986126899719,
|
| 277 |
+
"eval_runtime": 157.7062,
|
| 278 |
+
"eval_samples_per_second": 36.879,
|
| 279 |
+
"eval_steps_per_second": 4.61,
|
| 280 |
+
"step": 9000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.4521553041883215,
|
| 284 |
+
"grad_norm": 3.0906243324279785,
|
| 285 |
+
"learning_rate": 2.5810149801284013e-05,
|
| 286 |
+
"loss": 0.5499,
|
| 287 |
+
"step": 9500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.4521553041883215,
|
| 291 |
+
"eval_loss": 0.8367328643798828,
|
| 292 |
+
"eval_runtime": 157.7916,
|
| 293 |
+
"eval_samples_per_second": 36.859,
|
| 294 |
+
"eval_steps_per_second": 4.607,
|
| 295 |
+
"step": 9500
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.5285845307245491,
|
| 299 |
+
"grad_norm": 3.706326723098755,
|
| 300 |
+
"learning_rate": 2.4536329359013556e-05,
|
| 301 |
+
"loss": 0.5503,
|
| 302 |
+
"step": 10000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.5285845307245491,
|
| 306 |
+
"eval_loss": 0.8307807445526123,
|
| 307 |
+
"eval_runtime": 157.6389,
|
| 308 |
+
"eval_samples_per_second": 36.894,
|
| 309 |
+
"eval_steps_per_second": 4.612,
|
| 310 |
+
"step": 10000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.6050137572607766,
|
| 314 |
+
"grad_norm": 2.6108150482177734,
|
| 315 |
+
"learning_rate": 2.3262508916743096e-05,
|
| 316 |
+
"loss": 0.5388,
|
| 317 |
+
"step": 10500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.6050137572607766,
|
| 321 |
+
"eval_loss": 0.8295947313308716,
|
| 322 |
+
"eval_runtime": 157.6638,
|
| 323 |
+
"eval_samples_per_second": 36.889,
|
| 324 |
+
"eval_steps_per_second": 4.611,
|
| 325 |
+
"step": 10500
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.681442983797004,
|
| 329 |
+
"grad_norm": 1.6078243255615234,
|
| 330 |
+
"learning_rate": 2.1991236115357182e-05,
|
| 331 |
+
"loss": 0.5473,
|
| 332 |
+
"step": 11000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.681442983797004,
|
| 336 |
+
"eval_loss": 0.8229663372039795,
|
| 337 |
+
"eval_runtime": 157.6846,
|
| 338 |
+
"eval_samples_per_second": 36.884,
|
| 339 |
+
"eval_steps_per_second": 4.61,
|
| 340 |
+
"step": 11000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.7578722103332314,
|
| 344 |
+
"grad_norm": 3.049797773361206,
|
| 345 |
+
"learning_rate": 2.0717415673086722e-05,
|
| 346 |
+
"loss": 0.5496,
|
| 347 |
+
"step": 11500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 1.7578722103332314,
|
| 351 |
+
"eval_loss": 0.8267400860786438,
|
| 352 |
+
"eval_runtime": 157.7336,
|
| 353 |
+
"eval_samples_per_second": 36.872,
|
| 354 |
+
"eval_steps_per_second": 4.609,
|
| 355 |
+
"step": 11500
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.8343014368694588,
|
| 359 |
+
"grad_norm": 2.292538642883301,
|
| 360 |
+
"learning_rate": 1.9443595230816262e-05,
|
| 361 |
+
"loss": 0.5448,
|
| 362 |
+
"step": 12000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 1.8343014368694588,
|
| 366 |
+
"eval_loss": 0.8191345930099487,
|
| 367 |
+
"eval_runtime": 158.4293,
|
| 368 |
+
"eval_samples_per_second": 36.71,
|
| 369 |
+
"eval_steps_per_second": 4.589,
|
| 370 |
+
"step": 12000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.9107306634056864,
|
| 374 |
+
"grad_norm": 2.1699585914611816,
|
| 375 |
+
"learning_rate": 1.8169774788545806e-05,
|
| 376 |
+
"loss": 0.5419,
|
| 377 |
+
"step": 12500
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.9107306634056864,
|
| 381 |
+
"eval_loss": 0.8131210803985596,
|
| 382 |
+
"eval_runtime": 157.6629,
|
| 383 |
+
"eval_samples_per_second": 36.889,
|
| 384 |
+
"eval_steps_per_second": 4.611,
|
| 385 |
+
"step": 12500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 1.9871598899419138,
|
| 389 |
+
"grad_norm": 3.1323328018188477,
|
| 390 |
+
"learning_rate": 1.689595434627535e-05,
|
| 391 |
+
"loss": 0.5369,
|
| 392 |
+
"step": 13000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.9871598899419138,
|
| 396 |
+
"eval_loss": 0.8066145777702332,
|
| 397 |
+
"eval_runtime": 157.7959,
|
| 398 |
+
"eval_samples_per_second": 36.858,
|
| 399 |
+
"eval_steps_per_second": 4.607,
|
| 400 |
+
"step": 13000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.0635891164781412,
|
| 404 |
+
"grad_norm": 3.656402826309204,
|
| 405 |
+
"learning_rate": 1.562468154488943e-05,
|
| 406 |
+
"loss": 0.3304,
|
| 407 |
+
"step": 13500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.0635891164781412,
|
| 411 |
+
"eval_loss": 0.9408266544342041,
|
| 412 |
+
"eval_runtime": 157.7674,
|
| 413 |
+
"eval_samples_per_second": 36.864,
|
| 414 |
+
"eval_steps_per_second": 4.608,
|
| 415 |
+
"step": 13500
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.140018343014369,
|
| 419 |
+
"grad_norm": 2.4427192211151123,
|
| 420 |
+
"learning_rate": 1.4350861102618977e-05,
|
| 421 |
+
"loss": 0.2759,
|
| 422 |
+
"step": 14000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.140018343014369,
|
| 426 |
+
"eval_loss": 0.942986011505127,
|
| 427 |
+
"eval_runtime": 157.7087,
|
| 428 |
+
"eval_samples_per_second": 36.878,
|
| 429 |
+
"eval_steps_per_second": 4.61,
|
| 430 |
+
"step": 14000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.216447569550596,
|
| 434 |
+
"grad_norm": 1.6041910648345947,
|
| 435 |
+
"learning_rate": 1.3077040660348518e-05,
|
| 436 |
+
"loss": 0.2873,
|
| 437 |
+
"step": 14500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 2.216447569550596,
|
| 441 |
+
"eval_loss": 0.9449612498283386,
|
| 442 |
+
"eval_runtime": 157.767,
|
| 443 |
+
"eval_samples_per_second": 36.865,
|
| 444 |
+
"eval_steps_per_second": 4.608,
|
| 445 |
+
"step": 14500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.2928767960868237,
|
| 449 |
+
"grad_norm": 3.233290433883667,
|
| 450 |
+
"learning_rate": 1.180322021807806e-05,
|
| 451 |
+
"loss": 0.2818,
|
| 452 |
+
"step": 15000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.2928767960868237,
|
| 456 |
+
"eval_loss": 0.9387638568878174,
|
| 457 |
+
"eval_runtime": 157.7297,
|
| 458 |
+
"eval_samples_per_second": 36.873,
|
| 459 |
+
"eval_steps_per_second": 4.609,
|
| 460 |
+
"step": 15000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.369306022623051,
|
| 464 |
+
"grad_norm": 3.9653565883636475,
|
| 465 |
+
"learning_rate": 1.0529399775807602e-05,
|
| 466 |
+
"loss": 0.2795,
|
| 467 |
+
"step": 15500
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.369306022623051,
|
| 471 |
+
"eval_loss": 0.9435889720916748,
|
| 472 |
+
"eval_runtime": 157.7236,
|
| 473 |
+
"eval_samples_per_second": 36.875,
|
| 474 |
+
"eval_steps_per_second": 4.609,
|
| 475 |
+
"step": 15500
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 2.4457352491592785,
|
| 479 |
+
"grad_norm": 2.453057289123535,
|
| 480 |
+
"learning_rate": 9.255579333537145e-06,
|
| 481 |
+
"loss": 0.2801,
|
| 482 |
+
"step": 16000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.4457352491592785,
|
| 486 |
+
"eval_loss": 0.9410313963890076,
|
| 487 |
+
"eval_runtime": 157.7031,
|
| 488 |
+
"eval_samples_per_second": 36.879,
|
| 489 |
+
"eval_steps_per_second": 4.61,
|
| 490 |
+
"step": 16000
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 2.522164475695506,
|
| 494 |
+
"grad_norm": 2.9924492835998535,
|
| 495 |
+
"learning_rate": 7.981758891266687e-06,
|
| 496 |
+
"loss": 0.2788,
|
| 497 |
+
"step": 16500
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 2.522164475695506,
|
| 501 |
+
"eval_loss": 0.9427609443664551,
|
| 502 |
+
"eval_runtime": 157.6635,
|
| 503 |
+
"eval_samples_per_second": 36.889,
|
| 504 |
+
"eval_steps_per_second": 4.611,
|
| 505 |
+
"step": 16500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 2.5985937022317334,
|
| 509 |
+
"grad_norm": 2.626593828201294,
|
| 510 |
+
"learning_rate": 6.7079384489962305e-06,
|
| 511 |
+
"loss": 0.2752,
|
| 512 |
+
"step": 17000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.5985937022317334,
|
| 516 |
+
"eval_loss": 0.9421259164810181,
|
| 517 |
+
"eval_runtime": 157.689,
|
| 518 |
+
"eval_samples_per_second": 36.883,
|
| 519 |
+
"eval_steps_per_second": 4.61,
|
| 520 |
+
"step": 17000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.675022928767961,
|
| 524 |
+
"grad_norm": 2.623121500015259,
|
| 525 |
+
"learning_rate": 5.436665647610313e-06,
|
| 526 |
+
"loss": 0.2695,
|
| 527 |
+
"step": 17500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.675022928767961,
|
| 531 |
+
"eval_loss": 0.9395164251327515,
|
| 532 |
+
"eval_runtime": 157.7665,
|
| 533 |
+
"eval_samples_per_second": 36.865,
|
| 534 |
+
"eval_steps_per_second": 4.608,
|
| 535 |
+
"step": 17500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.751452155304188,
|
| 539 |
+
"grad_norm": 2.4113175868988037,
|
| 540 |
+
"learning_rate": 4.1653928462243965e-06,
|
| 541 |
+
"loss": 0.2697,
|
| 542 |
+
"step": 18000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.751452155304188,
|
| 546 |
+
"eval_loss": 0.9405816197395325,
|
| 547 |
+
"eval_runtime": 157.6703,
|
| 548 |
+
"eval_samples_per_second": 36.887,
|
| 549 |
+
"eval_steps_per_second": 4.611,
|
| 550 |
+
"step": 18000
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.827881381840416,
|
| 554 |
+
"grad_norm": 3.3730709552764893,
|
| 555 |
+
"learning_rate": 2.8915724039539386e-06,
|
| 556 |
+
"loss": 0.2769,
|
| 557 |
+
"step": 18500
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.827881381840416,
|
| 561 |
+
"eval_loss": 0.9389672875404358,
|
| 562 |
+
"eval_runtime": 157.689,
|
| 563 |
+
"eval_samples_per_second": 36.883,
|
| 564 |
+
"eval_steps_per_second": 4.61,
|
| 565 |
+
"step": 18500
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 2.904310608376643,
|
| 569 |
+
"grad_norm": 3.464594841003418,
|
| 570 |
+
"learning_rate": 1.6177519616834812e-06,
|
| 571 |
+
"loss": 0.271,
|
| 572 |
+
"step": 19000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 2.904310608376643,
|
| 576 |
+
"eval_loss": 0.9393123984336853,
|
| 577 |
+
"eval_runtime": 157.8608,
|
| 578 |
+
"eval_samples_per_second": 36.843,
|
| 579 |
+
"eval_steps_per_second": 4.605,
|
| 580 |
+
"step": 19000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.9807398349128706,
|
| 584 |
+
"grad_norm": 2.9936461448669434,
|
| 585 |
+
"learning_rate": 3.439315194130236e-07,
|
| 586 |
+
"loss": 0.2584,
|
| 587 |
+
"step": 19500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 2.9807398349128706,
|
| 591 |
+
"eval_loss": 0.9363918900489807,
|
| 592 |
+
"eval_runtime": 157.724,
|
| 593 |
+
"eval_samples_per_second": 36.875,
|
| 594 |
+
"eval_steps_per_second": 4.609,
|
| 595 |
+
"step": 19500
|
| 596 |
+
}
|
| 597 |
+
],
|
| 598 |
+
"logging_steps": 500,
|
| 599 |
+
"max_steps": 19626,
|
| 600 |
+
"num_input_tokens_seen": 0,
|
| 601 |
+
"num_train_epochs": 3,
|
| 602 |
+
"save_steps": 500,
|
| 603 |
+
"stateful_callbacks": {
|
| 604 |
+
"TrainerControl": {
|
| 605 |
+
"args": {
|
| 606 |
+
"should_epoch_stop": false,
|
| 607 |
+
"should_evaluate": false,
|
| 608 |
+
"should_log": false,
|
| 609 |
+
"should_save": true,
|
| 610 |
+
"should_training_stop": true
|
| 611 |
+
},
|
| 612 |
+
"attributes": {}
|
| 613 |
+
}
|
| 614 |
+
},
|
| 615 |
+
"total_flos": 2.1245004082682266e+17,
|
| 616 |
+
"train_batch_size": 2,
|
| 617 |
+
"trial_name": null,
|
| 618 |
+
"trial_params": null
|
| 619 |
+
}
|
checkpoint-19626/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1847242637dff66547b119c57e8cfeaf293fec9ad4dd17713f23bb7846bba98d
|
| 3 |
+
size 5240
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2384234968
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:221b87618292959e3626a872c0881b82b65e408b3b772ca863840da63636cfa7
|
| 3 |
size 2384234968
|