Upload folder using huggingface_hub
Browse files- checkpoint-2436/config.json +41 -0
- checkpoint-2436/generation_config.json +7 -0
- checkpoint-2436/model.safetensors +3 -0
- checkpoint-2436/optimizer.pt +3 -0
- checkpoint-2436/rng_state.pth +3 -0
- checkpoint-2436/scheduler.pt +3 -0
- checkpoint-2436/trainer_state.json +474 -0
- checkpoint-2436/training_args.bin +3 -0
- checkpoint-2842/config.json +41 -0
- checkpoint-2842/generation_config.json +7 -0
- checkpoint-2842/model.safetensors +3 -0
- checkpoint-2842/optimizer.pt +3 -0
- checkpoint-2842/rng_state.pth +3 -0
- checkpoint-2842/scheduler.pt +3 -0
- checkpoint-2842/trainer_state.json +546 -0
- checkpoint-2842/training_args.bin +3 -0
- checkpoint-3045/config.json +41 -0
- checkpoint-3045/generation_config.json +7 -0
- checkpoint-3045/model.safetensors +3 -0
- checkpoint-3045/optimizer.pt +3 -0
- checkpoint-3045/rng_state.pth +3 -0
- checkpoint-3045/scheduler.pt +3 -0
- checkpoint-3045/trainer_state.json +582 -0
- checkpoint-3045/training_args.bin +3 -0
checkpoint-2436/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ai-forever/rugpt3small_based_on_gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 2048,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 12,
|
| 26 |
+
"n_positions": 2048,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"reorder_and_upcast_attn": false,
|
| 29 |
+
"resid_pdrop": 0.1,
|
| 30 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 31 |
+
"scale_attn_weights": true,
|
| 32 |
+
"summary_activation": null,
|
| 33 |
+
"summary_first_dropout": 0.1,
|
| 34 |
+
"summary_proj_to_labels": true,
|
| 35 |
+
"summary_type": "cls_index",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.48.3",
|
| 39 |
+
"use_cache": true,
|
| 40 |
+
"vocab_size": 50264
|
| 41 |
+
}
|
checkpoint-2436/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.3"
|
| 7 |
+
}
|
checkpoint-2436/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa4f68ef82234dcec67d2093eb642f04208c0f17a3ee5478daf2e30df8eb83d2
|
| 3 |
+
size 500941440
|
checkpoint-2436/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a8ae197c56357756193df201987ddfb97fabcae78a99b060c2df40efe1909af
|
| 3 |
+
size 1389114
|
checkpoint-2436/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:817951fee2eb40d47fa8cb26992bc5cf4ccc62d0f032d75e3aa49b3142f2184a
|
| 3 |
+
size 14244
|
checkpoint-2436/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be1ac34f0f3f8a970a6b9152cdcffd1d7a6eceb6bc68f995045893573d52ca85
|
| 3 |
+
size 1064
|
checkpoint-2436/trainer_state.json
ADDED
|
@@ -0,0 +1,474 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.091330528259277,
|
| 3 |
+
"best_model_checkpoint": "aristotle_new_layer_plain/checkpoint-2436",
|
| 4 |
+
"epoch": 12.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2436,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.24721878862793573,
|
| 13 |
+
"grad_norm": 4.709590435028076,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 10.2189,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.49443757725587145,
|
| 20 |
+
"grad_norm": 3.6981208324432373,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 9.1727,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7416563658838071,
|
| 27 |
+
"grad_norm": 3.9341259002685547,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 8.6877,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9888751545117429,
|
| 34 |
+
"grad_norm": 3.341215133666992,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 8.1544,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 7.7336225509643555,
|
| 42 |
+
"eval_runtime": 14.2669,
|
| 43 |
+
"eval_samples_per_second": 50.396,
|
| 44 |
+
"eval_steps_per_second": 6.308,
|
| 45 |
+
"step": 203
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.2323856613102595,
|
| 49 |
+
"grad_norm": 4.8560404777526855,
|
| 50 |
+
"learning_rate": 2.5e-05,
|
| 51 |
+
"loss": 7.5653,
|
| 52 |
+
"step": 250
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.4796044499381953,
|
| 56 |
+
"grad_norm": 3.1774024963378906,
|
| 57 |
+
"learning_rate": 3e-05,
|
| 58 |
+
"loss": 7.1093,
|
| 59 |
+
"step": 300
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.726823238566131,
|
| 63 |
+
"grad_norm": 4.33836030960083,
|
| 64 |
+
"learning_rate": 3.5e-05,
|
| 65 |
+
"loss": 6.7529,
|
| 66 |
+
"step": 350
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.9740420271940669,
|
| 70 |
+
"grad_norm": 2.5972180366516113,
|
| 71 |
+
"learning_rate": 4e-05,
|
| 72 |
+
"loss": 6.536,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.0,
|
| 77 |
+
"eval_loss": 6.426527500152588,
|
| 78 |
+
"eval_runtime": 14.2681,
|
| 79 |
+
"eval_samples_per_second": 50.392,
|
| 80 |
+
"eval_steps_per_second": 6.308,
|
| 81 |
+
"step": 406
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.2175525339925835,
|
| 85 |
+
"grad_norm": 3.43729567527771,
|
| 86 |
+
"learning_rate": 4.5e-05,
|
| 87 |
+
"loss": 6.3559,
|
| 88 |
+
"step": 450
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.464771322620519,
|
| 92 |
+
"grad_norm": 3.318251848220825,
|
| 93 |
+
"learning_rate": 5e-05,
|
| 94 |
+
"loss": 6.3251,
|
| 95 |
+
"step": 500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.711990111248455,
|
| 99 |
+
"grad_norm": 3.502115488052368,
|
| 100 |
+
"learning_rate": 4.9550359712230215e-05,
|
| 101 |
+
"loss": 6.1368,
|
| 102 |
+
"step": 550
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.9592088998763906,
|
| 106 |
+
"grad_norm": 3.497938394546509,
|
| 107 |
+
"learning_rate": 4.9100719424460435e-05,
|
| 108 |
+
"loss": 6.0775,
|
| 109 |
+
"step": 600
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 3.0,
|
| 113 |
+
"eval_loss": 6.036961555480957,
|
| 114 |
+
"eval_runtime": 14.2475,
|
| 115 |
+
"eval_samples_per_second": 50.465,
|
| 116 |
+
"eval_steps_per_second": 6.317,
|
| 117 |
+
"step": 609
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 3.202719406674907,
|
| 121 |
+
"grad_norm": 3.4355390071868896,
|
| 122 |
+
"learning_rate": 4.865107913669065e-05,
|
| 123 |
+
"loss": 5.8684,
|
| 124 |
+
"step": 650
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.449938195302843,
|
| 128 |
+
"grad_norm": 3.9220526218414307,
|
| 129 |
+
"learning_rate": 4.820143884892087e-05,
|
| 130 |
+
"loss": 5.8101,
|
| 131 |
+
"step": 700
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.6971569839307787,
|
| 135 |
+
"grad_norm": 3.782421827316284,
|
| 136 |
+
"learning_rate": 4.775179856115108e-05,
|
| 137 |
+
"loss": 5.7784,
|
| 138 |
+
"step": 750
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.9443757725587143,
|
| 142 |
+
"grad_norm": 3.5181567668914795,
|
| 143 |
+
"learning_rate": 4.7302158273381294e-05,
|
| 144 |
+
"loss": 5.7181,
|
| 145 |
+
"step": 800
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 4.0,
|
| 149 |
+
"eval_loss": 5.772126197814941,
|
| 150 |
+
"eval_runtime": 14.2949,
|
| 151 |
+
"eval_samples_per_second": 50.298,
|
| 152 |
+
"eval_steps_per_second": 6.296,
|
| 153 |
+
"step": 812
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.187886279357231,
|
| 157 |
+
"grad_norm": 3.6087594032287598,
|
| 158 |
+
"learning_rate": 4.685251798561151e-05,
|
| 159 |
+
"loss": 5.5154,
|
| 160 |
+
"step": 850
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.435105067985167,
|
| 164 |
+
"grad_norm": 3.8448667526245117,
|
| 165 |
+
"learning_rate": 4.640287769784173e-05,
|
| 166 |
+
"loss": 5.4664,
|
| 167 |
+
"step": 900
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.6823238566131025,
|
| 171 |
+
"grad_norm": 3.594693660736084,
|
| 172 |
+
"learning_rate": 4.595323741007194e-05,
|
| 173 |
+
"loss": 5.4121,
|
| 174 |
+
"step": 950
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.929542645241038,
|
| 178 |
+
"grad_norm": 3.6225693225860596,
|
| 179 |
+
"learning_rate": 4.550359712230216e-05,
|
| 180 |
+
"loss": 5.3158,
|
| 181 |
+
"step": 1000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 5.0,
|
| 185 |
+
"eval_loss": 5.521895885467529,
|
| 186 |
+
"eval_runtime": 14.2777,
|
| 187 |
+
"eval_samples_per_second": 50.358,
|
| 188 |
+
"eval_steps_per_second": 6.304,
|
| 189 |
+
"step": 1015
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 5.173053152039555,
|
| 193 |
+
"grad_norm": 4.245815277099609,
|
| 194 |
+
"learning_rate": 4.505395683453237e-05,
|
| 195 |
+
"loss": 5.1744,
|
| 196 |
+
"step": 1050
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 5.420271940667491,
|
| 200 |
+
"grad_norm": 4.306251525878906,
|
| 201 |
+
"learning_rate": 4.460431654676259e-05,
|
| 202 |
+
"loss": 5.1245,
|
| 203 |
+
"step": 1100
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.667490729295427,
|
| 207 |
+
"grad_norm": 3.7834959030151367,
|
| 208 |
+
"learning_rate": 4.4154676258992806e-05,
|
| 209 |
+
"loss": 5.0729,
|
| 210 |
+
"step": 1150
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.914709517923362,
|
| 214 |
+
"grad_norm": 4.298359394073486,
|
| 215 |
+
"learning_rate": 4.3705035971223026e-05,
|
| 216 |
+
"loss": 5.0558,
|
| 217 |
+
"step": 1200
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.0,
|
| 221 |
+
"eval_loss": 5.389599323272705,
|
| 222 |
+
"eval_runtime": 14.2956,
|
| 223 |
+
"eval_samples_per_second": 50.295,
|
| 224 |
+
"eval_steps_per_second": 6.296,
|
| 225 |
+
"step": 1218
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.158220024721879,
|
| 229 |
+
"grad_norm": 4.416134357452393,
|
| 230 |
+
"learning_rate": 4.325539568345324e-05,
|
| 231 |
+
"loss": 4.839,
|
| 232 |
+
"step": 1250
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 6.405438813349814,
|
| 236 |
+
"grad_norm": 4.565963268280029,
|
| 237 |
+
"learning_rate": 4.280575539568346e-05,
|
| 238 |
+
"loss": 4.8297,
|
| 239 |
+
"step": 1300
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.652657601977751,
|
| 243 |
+
"grad_norm": 4.854921817779541,
|
| 244 |
+
"learning_rate": 4.235611510791367e-05,
|
| 245 |
+
"loss": 4.8175,
|
| 246 |
+
"step": 1350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.899876390605686,
|
| 250 |
+
"grad_norm": 4.982056617736816,
|
| 251 |
+
"learning_rate": 4.1906474820143885e-05,
|
| 252 |
+
"loss": 4.8081,
|
| 253 |
+
"step": 1400
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 7.0,
|
| 257 |
+
"eval_loss": 5.254246711730957,
|
| 258 |
+
"eval_runtime": 14.2665,
|
| 259 |
+
"eval_samples_per_second": 50.398,
|
| 260 |
+
"eval_steps_per_second": 6.308,
|
| 261 |
+
"step": 1421
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.143386897404203,
|
| 265 |
+
"grad_norm": 4.195478439331055,
|
| 266 |
+
"learning_rate": 4.14568345323741e-05,
|
| 267 |
+
"loss": 4.6322,
|
| 268 |
+
"step": 1450
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.3906056860321385,
|
| 272 |
+
"grad_norm": 4.963181972503662,
|
| 273 |
+
"learning_rate": 4.100719424460432e-05,
|
| 274 |
+
"loss": 4.547,
|
| 275 |
+
"step": 1500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.637824474660074,
|
| 279 |
+
"grad_norm": 5.290962219238281,
|
| 280 |
+
"learning_rate": 4.055755395683453e-05,
|
| 281 |
+
"loss": 4.5553,
|
| 282 |
+
"step": 1550
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.88504326328801,
|
| 286 |
+
"grad_norm": 5.0038838386535645,
|
| 287 |
+
"learning_rate": 4.010791366906475e-05,
|
| 288 |
+
"loss": 4.5651,
|
| 289 |
+
"step": 1600
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"eval_loss": 5.183382987976074,
|
| 294 |
+
"eval_runtime": 14.2977,
|
| 295 |
+
"eval_samples_per_second": 50.288,
|
| 296 |
+
"eval_steps_per_second": 6.295,
|
| 297 |
+
"step": 1624
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.128553770086526,
|
| 301 |
+
"grad_norm": 5.3380446434021,
|
| 302 |
+
"learning_rate": 3.965827338129496e-05,
|
| 303 |
+
"loss": 4.3966,
|
| 304 |
+
"step": 1650
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 8.375772558714463,
|
| 308 |
+
"grad_norm": 5.339470863342285,
|
| 309 |
+
"learning_rate": 3.920863309352518e-05,
|
| 310 |
+
"loss": 4.3068,
|
| 311 |
+
"step": 1700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 8.622991347342397,
|
| 315 |
+
"grad_norm": 4.9476189613342285,
|
| 316 |
+
"learning_rate": 3.8758992805755396e-05,
|
| 317 |
+
"loss": 4.3249,
|
| 318 |
+
"step": 1750
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.870210135970334,
|
| 322 |
+
"grad_norm": 5.430028915405273,
|
| 323 |
+
"learning_rate": 3.8309352517985616e-05,
|
| 324 |
+
"loss": 4.3407,
|
| 325 |
+
"step": 1800
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 9.0,
|
| 329 |
+
"eval_loss": 5.13620138168335,
|
| 330 |
+
"eval_runtime": 14.2616,
|
| 331 |
+
"eval_samples_per_second": 50.415,
|
| 332 |
+
"eval_steps_per_second": 6.311,
|
| 333 |
+
"step": 1827
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 9.11372064276885,
|
| 337 |
+
"grad_norm": 5.2561259269714355,
|
| 338 |
+
"learning_rate": 3.785971223021583e-05,
|
| 339 |
+
"loss": 4.1746,
|
| 340 |
+
"step": 1850
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 9.360939431396787,
|
| 344 |
+
"grad_norm": 5.811314105987549,
|
| 345 |
+
"learning_rate": 3.741007194244605e-05,
|
| 346 |
+
"loss": 4.1324,
|
| 347 |
+
"step": 1900
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 9.608158220024722,
|
| 351 |
+
"grad_norm": 5.552155017852783,
|
| 352 |
+
"learning_rate": 3.696043165467626e-05,
|
| 353 |
+
"loss": 4.1058,
|
| 354 |
+
"step": 1950
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.855377008652658,
|
| 358 |
+
"grad_norm": 6.073920726776123,
|
| 359 |
+
"learning_rate": 3.6510791366906475e-05,
|
| 360 |
+
"loss": 4.0436,
|
| 361 |
+
"step": 2000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 10.0,
|
| 365 |
+
"eval_loss": 5.104895114898682,
|
| 366 |
+
"eval_runtime": 14.2556,
|
| 367 |
+
"eval_samples_per_second": 50.436,
|
| 368 |
+
"eval_steps_per_second": 6.313,
|
| 369 |
+
"step": 2030
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 10.098887515451175,
|
| 373 |
+
"grad_norm": 5.994938373565674,
|
| 374 |
+
"learning_rate": 3.606115107913669e-05,
|
| 375 |
+
"loss": 3.99,
|
| 376 |
+
"step": 2050
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 10.34610630407911,
|
| 380 |
+
"grad_norm": 6.414961814880371,
|
| 381 |
+
"learning_rate": 3.561151079136691e-05,
|
| 382 |
+
"loss": 3.9013,
|
| 383 |
+
"step": 2100
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 10.593325092707046,
|
| 387 |
+
"grad_norm": 6.1248459815979,
|
| 388 |
+
"learning_rate": 3.516187050359712e-05,
|
| 389 |
+
"loss": 3.8884,
|
| 390 |
+
"step": 2150
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.840543881334982,
|
| 394 |
+
"grad_norm": 5.360867500305176,
|
| 395 |
+
"learning_rate": 3.471223021582734e-05,
|
| 396 |
+
"loss": 3.877,
|
| 397 |
+
"step": 2200
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 11.0,
|
| 401 |
+
"eval_loss": 5.103781700134277,
|
| 402 |
+
"eval_runtime": 14.2705,
|
| 403 |
+
"eval_samples_per_second": 50.384,
|
| 404 |
+
"eval_steps_per_second": 6.307,
|
| 405 |
+
"step": 2233
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 11.084054388133499,
|
| 409 |
+
"grad_norm": 5.840531349182129,
|
| 410 |
+
"learning_rate": 3.4262589928057554e-05,
|
| 411 |
+
"loss": 3.8216,
|
| 412 |
+
"step": 2250
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 11.331273176761433,
|
| 416 |
+
"grad_norm": 7.407821178436279,
|
| 417 |
+
"learning_rate": 3.3812949640287773e-05,
|
| 418 |
+
"loss": 3.6379,
|
| 419 |
+
"step": 2300
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 11.57849196538937,
|
| 423 |
+
"grad_norm": 7.770689487457275,
|
| 424 |
+
"learning_rate": 3.3363309352517986e-05,
|
| 425 |
+
"loss": 3.7063,
|
| 426 |
+
"step": 2350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.825710754017305,
|
| 430 |
+
"grad_norm": 6.17808198928833,
|
| 431 |
+
"learning_rate": 3.2913669064748206e-05,
|
| 432 |
+
"loss": 3.7008,
|
| 433 |
+
"step": 2400
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 12.0,
|
| 437 |
+
"eval_loss": 5.091330528259277,
|
| 438 |
+
"eval_runtime": 14.3396,
|
| 439 |
+
"eval_samples_per_second": 50.141,
|
| 440 |
+
"eval_steps_per_second": 6.276,
|
| 441 |
+
"step": 2436
|
| 442 |
+
}
|
| 443 |
+
],
|
| 444 |
+
"logging_steps": 50,
|
| 445 |
+
"max_steps": 6060,
|
| 446 |
+
"num_input_tokens_seen": 0,
|
| 447 |
+
"num_train_epochs": 30,
|
| 448 |
+
"save_steps": 500,
|
| 449 |
+
"stateful_callbacks": {
|
| 450 |
+
"EarlyStoppingCallback": {
|
| 451 |
+
"args": {
|
| 452 |
+
"early_stopping_patience": 3,
|
| 453 |
+
"early_stopping_threshold": 0.0
|
| 454 |
+
},
|
| 455 |
+
"attributes": {
|
| 456 |
+
"early_stopping_patience_counter": 0
|
| 457 |
+
}
|
| 458 |
+
},
|
| 459 |
+
"TrainerControl": {
|
| 460 |
+
"args": {
|
| 461 |
+
"should_epoch_stop": false,
|
| 462 |
+
"should_evaluate": false,
|
| 463 |
+
"should_log": false,
|
| 464 |
+
"should_save": true,
|
| 465 |
+
"should_training_stop": false
|
| 466 |
+
},
|
| 467 |
+
"attributes": {}
|
| 468 |
+
}
|
| 469 |
+
},
|
| 470 |
+
"total_flos": 2.0280442355712e+16,
|
| 471 |
+
"train_batch_size": 8,
|
| 472 |
+
"trial_name": null,
|
| 473 |
+
"trial_params": null
|
| 474 |
+
}
|
checkpoint-2436/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28ec087a0e6786ef2e0bcf02310b767379b6de459d7cb254c0c6ae4e881ba0e1
|
| 3 |
+
size 5304
|
checkpoint-2842/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ai-forever/rugpt3small_based_on_gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 2048,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 12,
|
| 26 |
+
"n_positions": 2048,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"reorder_and_upcast_attn": false,
|
| 29 |
+
"resid_pdrop": 0.1,
|
| 30 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 31 |
+
"scale_attn_weights": true,
|
| 32 |
+
"summary_activation": null,
|
| 33 |
+
"summary_first_dropout": 0.1,
|
| 34 |
+
"summary_proj_to_labels": true,
|
| 35 |
+
"summary_type": "cls_index",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.48.3",
|
| 39 |
+
"use_cache": true,
|
| 40 |
+
"vocab_size": 50264
|
| 41 |
+
}
|
checkpoint-2842/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.3"
|
| 7 |
+
}
|
checkpoint-2842/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:706dd31f608d02938920be68d84bb0433225705f634457946541a30c69fb4908
|
| 3 |
+
size 500941440
|
checkpoint-2842/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:509b4f4bc35322abd963fd780f5a4f07543d3a44ad818d5f1eadac5fbcaa80cd
|
| 3 |
+
size 1389114
|
checkpoint-2842/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a0e136fffb6d352ed6840c10d527267eee09f0ce8460de2ac56fb9d95167217
|
| 3 |
+
size 14244
|
checkpoint-2842/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88a2612a056def0df3ae3c082fb47fd5b4944d59eecf3efb96824e8304ff100e
|
| 3 |
+
size 1064
|
checkpoint-2842/trainer_state.json
ADDED
|
@@ -0,0 +1,546 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.091330528259277,
|
| 3 |
+
"best_model_checkpoint": "aristotle_new_layer_plain/checkpoint-2436",
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2842,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.24721878862793573,
|
| 13 |
+
"grad_norm": 4.709590435028076,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 10.2189,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.49443757725587145,
|
| 20 |
+
"grad_norm": 3.6981208324432373,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 9.1727,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7416563658838071,
|
| 27 |
+
"grad_norm": 3.9341259002685547,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 8.6877,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9888751545117429,
|
| 34 |
+
"grad_norm": 3.341215133666992,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 8.1544,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 7.7336225509643555,
|
| 42 |
+
"eval_runtime": 14.2669,
|
| 43 |
+
"eval_samples_per_second": 50.396,
|
| 44 |
+
"eval_steps_per_second": 6.308,
|
| 45 |
+
"step": 203
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.2323856613102595,
|
| 49 |
+
"grad_norm": 4.8560404777526855,
|
| 50 |
+
"learning_rate": 2.5e-05,
|
| 51 |
+
"loss": 7.5653,
|
| 52 |
+
"step": 250
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.4796044499381953,
|
| 56 |
+
"grad_norm": 3.1774024963378906,
|
| 57 |
+
"learning_rate": 3e-05,
|
| 58 |
+
"loss": 7.1093,
|
| 59 |
+
"step": 300
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.726823238566131,
|
| 63 |
+
"grad_norm": 4.33836030960083,
|
| 64 |
+
"learning_rate": 3.5e-05,
|
| 65 |
+
"loss": 6.7529,
|
| 66 |
+
"step": 350
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.9740420271940669,
|
| 70 |
+
"grad_norm": 2.5972180366516113,
|
| 71 |
+
"learning_rate": 4e-05,
|
| 72 |
+
"loss": 6.536,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.0,
|
| 77 |
+
"eval_loss": 6.426527500152588,
|
| 78 |
+
"eval_runtime": 14.2681,
|
| 79 |
+
"eval_samples_per_second": 50.392,
|
| 80 |
+
"eval_steps_per_second": 6.308,
|
| 81 |
+
"step": 406
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.2175525339925835,
|
| 85 |
+
"grad_norm": 3.43729567527771,
|
| 86 |
+
"learning_rate": 4.5e-05,
|
| 87 |
+
"loss": 6.3559,
|
| 88 |
+
"step": 450
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.464771322620519,
|
| 92 |
+
"grad_norm": 3.318251848220825,
|
| 93 |
+
"learning_rate": 5e-05,
|
| 94 |
+
"loss": 6.3251,
|
| 95 |
+
"step": 500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.711990111248455,
|
| 99 |
+
"grad_norm": 3.502115488052368,
|
| 100 |
+
"learning_rate": 4.9550359712230215e-05,
|
| 101 |
+
"loss": 6.1368,
|
| 102 |
+
"step": 550
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.9592088998763906,
|
| 106 |
+
"grad_norm": 3.497938394546509,
|
| 107 |
+
"learning_rate": 4.9100719424460435e-05,
|
| 108 |
+
"loss": 6.0775,
|
| 109 |
+
"step": 600
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 3.0,
|
| 113 |
+
"eval_loss": 6.036961555480957,
|
| 114 |
+
"eval_runtime": 14.2475,
|
| 115 |
+
"eval_samples_per_second": 50.465,
|
| 116 |
+
"eval_steps_per_second": 6.317,
|
| 117 |
+
"step": 609
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 3.202719406674907,
|
| 121 |
+
"grad_norm": 3.4355390071868896,
|
| 122 |
+
"learning_rate": 4.865107913669065e-05,
|
| 123 |
+
"loss": 5.8684,
|
| 124 |
+
"step": 650
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.449938195302843,
|
| 128 |
+
"grad_norm": 3.9220526218414307,
|
| 129 |
+
"learning_rate": 4.820143884892087e-05,
|
| 130 |
+
"loss": 5.8101,
|
| 131 |
+
"step": 700
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.6971569839307787,
|
| 135 |
+
"grad_norm": 3.782421827316284,
|
| 136 |
+
"learning_rate": 4.775179856115108e-05,
|
| 137 |
+
"loss": 5.7784,
|
| 138 |
+
"step": 750
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.9443757725587143,
|
| 142 |
+
"grad_norm": 3.5181567668914795,
|
| 143 |
+
"learning_rate": 4.7302158273381294e-05,
|
| 144 |
+
"loss": 5.7181,
|
| 145 |
+
"step": 800
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 4.0,
|
| 149 |
+
"eval_loss": 5.772126197814941,
|
| 150 |
+
"eval_runtime": 14.2949,
|
| 151 |
+
"eval_samples_per_second": 50.298,
|
| 152 |
+
"eval_steps_per_second": 6.296,
|
| 153 |
+
"step": 812
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.187886279357231,
|
| 157 |
+
"grad_norm": 3.6087594032287598,
|
| 158 |
+
"learning_rate": 4.685251798561151e-05,
|
| 159 |
+
"loss": 5.5154,
|
| 160 |
+
"step": 850
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.435105067985167,
|
| 164 |
+
"grad_norm": 3.8448667526245117,
|
| 165 |
+
"learning_rate": 4.640287769784173e-05,
|
| 166 |
+
"loss": 5.4664,
|
| 167 |
+
"step": 900
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.6823238566131025,
|
| 171 |
+
"grad_norm": 3.594693660736084,
|
| 172 |
+
"learning_rate": 4.595323741007194e-05,
|
| 173 |
+
"loss": 5.4121,
|
| 174 |
+
"step": 950
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.929542645241038,
|
| 178 |
+
"grad_norm": 3.6225693225860596,
|
| 179 |
+
"learning_rate": 4.550359712230216e-05,
|
| 180 |
+
"loss": 5.3158,
|
| 181 |
+
"step": 1000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 5.0,
|
| 185 |
+
"eval_loss": 5.521895885467529,
|
| 186 |
+
"eval_runtime": 14.2777,
|
| 187 |
+
"eval_samples_per_second": 50.358,
|
| 188 |
+
"eval_steps_per_second": 6.304,
|
| 189 |
+
"step": 1015
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 5.173053152039555,
|
| 193 |
+
"grad_norm": 4.245815277099609,
|
| 194 |
+
"learning_rate": 4.505395683453237e-05,
|
| 195 |
+
"loss": 5.1744,
|
| 196 |
+
"step": 1050
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 5.420271940667491,
|
| 200 |
+
"grad_norm": 4.306251525878906,
|
| 201 |
+
"learning_rate": 4.460431654676259e-05,
|
| 202 |
+
"loss": 5.1245,
|
| 203 |
+
"step": 1100
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.667490729295427,
|
| 207 |
+
"grad_norm": 3.7834959030151367,
|
| 208 |
+
"learning_rate": 4.4154676258992806e-05,
|
| 209 |
+
"loss": 5.0729,
|
| 210 |
+
"step": 1150
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.914709517923362,
|
| 214 |
+
"grad_norm": 4.298359394073486,
|
| 215 |
+
"learning_rate": 4.3705035971223026e-05,
|
| 216 |
+
"loss": 5.0558,
|
| 217 |
+
"step": 1200
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.0,
|
| 221 |
+
"eval_loss": 5.389599323272705,
|
| 222 |
+
"eval_runtime": 14.2956,
|
| 223 |
+
"eval_samples_per_second": 50.295,
|
| 224 |
+
"eval_steps_per_second": 6.296,
|
| 225 |
+
"step": 1218
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.158220024721879,
|
| 229 |
+
"grad_norm": 4.416134357452393,
|
| 230 |
+
"learning_rate": 4.325539568345324e-05,
|
| 231 |
+
"loss": 4.839,
|
| 232 |
+
"step": 1250
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 6.405438813349814,
|
| 236 |
+
"grad_norm": 4.565963268280029,
|
| 237 |
+
"learning_rate": 4.280575539568346e-05,
|
| 238 |
+
"loss": 4.8297,
|
| 239 |
+
"step": 1300
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.652657601977751,
|
| 243 |
+
"grad_norm": 4.854921817779541,
|
| 244 |
+
"learning_rate": 4.235611510791367e-05,
|
| 245 |
+
"loss": 4.8175,
|
| 246 |
+
"step": 1350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.899876390605686,
|
| 250 |
+
"grad_norm": 4.982056617736816,
|
| 251 |
+
"learning_rate": 4.1906474820143885e-05,
|
| 252 |
+
"loss": 4.8081,
|
| 253 |
+
"step": 1400
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 7.0,
|
| 257 |
+
"eval_loss": 5.254246711730957,
|
| 258 |
+
"eval_runtime": 14.2665,
|
| 259 |
+
"eval_samples_per_second": 50.398,
|
| 260 |
+
"eval_steps_per_second": 6.308,
|
| 261 |
+
"step": 1421
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.143386897404203,
|
| 265 |
+
"grad_norm": 4.195478439331055,
|
| 266 |
+
"learning_rate": 4.14568345323741e-05,
|
| 267 |
+
"loss": 4.6322,
|
| 268 |
+
"step": 1450
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.3906056860321385,
|
| 272 |
+
"grad_norm": 4.963181972503662,
|
| 273 |
+
"learning_rate": 4.100719424460432e-05,
|
| 274 |
+
"loss": 4.547,
|
| 275 |
+
"step": 1500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.637824474660074,
|
| 279 |
+
"grad_norm": 5.290962219238281,
|
| 280 |
+
"learning_rate": 4.055755395683453e-05,
|
| 281 |
+
"loss": 4.5553,
|
| 282 |
+
"step": 1550
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.88504326328801,
|
| 286 |
+
"grad_norm": 5.0038838386535645,
|
| 287 |
+
"learning_rate": 4.010791366906475e-05,
|
| 288 |
+
"loss": 4.5651,
|
| 289 |
+
"step": 1600
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"eval_loss": 5.183382987976074,
|
| 294 |
+
"eval_runtime": 14.2977,
|
| 295 |
+
"eval_samples_per_second": 50.288,
|
| 296 |
+
"eval_steps_per_second": 6.295,
|
| 297 |
+
"step": 1624
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.128553770086526,
|
| 301 |
+
"grad_norm": 5.3380446434021,
|
| 302 |
+
"learning_rate": 3.965827338129496e-05,
|
| 303 |
+
"loss": 4.3966,
|
| 304 |
+
"step": 1650
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 8.375772558714463,
|
| 308 |
+
"grad_norm": 5.339470863342285,
|
| 309 |
+
"learning_rate": 3.920863309352518e-05,
|
| 310 |
+
"loss": 4.3068,
|
| 311 |
+
"step": 1700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 8.622991347342397,
|
| 315 |
+
"grad_norm": 4.9476189613342285,
|
| 316 |
+
"learning_rate": 3.8758992805755396e-05,
|
| 317 |
+
"loss": 4.3249,
|
| 318 |
+
"step": 1750
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.870210135970334,
|
| 322 |
+
"grad_norm": 5.430028915405273,
|
| 323 |
+
"learning_rate": 3.8309352517985616e-05,
|
| 324 |
+
"loss": 4.3407,
|
| 325 |
+
"step": 1800
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 9.0,
|
| 329 |
+
"eval_loss": 5.13620138168335,
|
| 330 |
+
"eval_runtime": 14.2616,
|
| 331 |
+
"eval_samples_per_second": 50.415,
|
| 332 |
+
"eval_steps_per_second": 6.311,
|
| 333 |
+
"step": 1827
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 9.11372064276885,
|
| 337 |
+
"grad_norm": 5.2561259269714355,
|
| 338 |
+
"learning_rate": 3.785971223021583e-05,
|
| 339 |
+
"loss": 4.1746,
|
| 340 |
+
"step": 1850
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 9.360939431396787,
|
| 344 |
+
"grad_norm": 5.811314105987549,
|
| 345 |
+
"learning_rate": 3.741007194244605e-05,
|
| 346 |
+
"loss": 4.1324,
|
| 347 |
+
"step": 1900
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 9.608158220024722,
|
| 351 |
+
"grad_norm": 5.552155017852783,
|
| 352 |
+
"learning_rate": 3.696043165467626e-05,
|
| 353 |
+
"loss": 4.1058,
|
| 354 |
+
"step": 1950
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.855377008652658,
|
| 358 |
+
"grad_norm": 6.073920726776123,
|
| 359 |
+
"learning_rate": 3.6510791366906475e-05,
|
| 360 |
+
"loss": 4.0436,
|
| 361 |
+
"step": 2000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 10.0,
|
| 365 |
+
"eval_loss": 5.104895114898682,
|
| 366 |
+
"eval_runtime": 14.2556,
|
| 367 |
+
"eval_samples_per_second": 50.436,
|
| 368 |
+
"eval_steps_per_second": 6.313,
|
| 369 |
+
"step": 2030
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 10.098887515451175,
|
| 373 |
+
"grad_norm": 5.994938373565674,
|
| 374 |
+
"learning_rate": 3.606115107913669e-05,
|
| 375 |
+
"loss": 3.99,
|
| 376 |
+
"step": 2050
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 10.34610630407911,
|
| 380 |
+
"grad_norm": 6.414961814880371,
|
| 381 |
+
"learning_rate": 3.561151079136691e-05,
|
| 382 |
+
"loss": 3.9013,
|
| 383 |
+
"step": 2100
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 10.593325092707046,
|
| 387 |
+
"grad_norm": 6.1248459815979,
|
| 388 |
+
"learning_rate": 3.516187050359712e-05,
|
| 389 |
+
"loss": 3.8884,
|
| 390 |
+
"step": 2150
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.840543881334982,
|
| 394 |
+
"grad_norm": 5.360867500305176,
|
| 395 |
+
"learning_rate": 3.471223021582734e-05,
|
| 396 |
+
"loss": 3.877,
|
| 397 |
+
"step": 2200
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 11.0,
|
| 401 |
+
"eval_loss": 5.103781700134277,
|
| 402 |
+
"eval_runtime": 14.2705,
|
| 403 |
+
"eval_samples_per_second": 50.384,
|
| 404 |
+
"eval_steps_per_second": 6.307,
|
| 405 |
+
"step": 2233
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 11.084054388133499,
|
| 409 |
+
"grad_norm": 5.840531349182129,
|
| 410 |
+
"learning_rate": 3.4262589928057554e-05,
|
| 411 |
+
"loss": 3.8216,
|
| 412 |
+
"step": 2250
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 11.331273176761433,
|
| 416 |
+
"grad_norm": 7.407821178436279,
|
| 417 |
+
"learning_rate": 3.3812949640287773e-05,
|
| 418 |
+
"loss": 3.6379,
|
| 419 |
+
"step": 2300
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 11.57849196538937,
|
| 423 |
+
"grad_norm": 7.770689487457275,
|
| 424 |
+
"learning_rate": 3.3363309352517986e-05,
|
| 425 |
+
"loss": 3.7063,
|
| 426 |
+
"step": 2350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.825710754017305,
|
| 430 |
+
"grad_norm": 6.17808198928833,
|
| 431 |
+
"learning_rate": 3.2913669064748206e-05,
|
| 432 |
+
"loss": 3.7008,
|
| 433 |
+
"step": 2400
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 12.0,
|
| 437 |
+
"eval_loss": 5.091330528259277,
|
| 438 |
+
"eval_runtime": 14.3396,
|
| 439 |
+
"eval_samples_per_second": 50.141,
|
| 440 |
+
"eval_steps_per_second": 6.276,
|
| 441 |
+
"step": 2436
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 12.069221260815821,
|
| 445 |
+
"grad_norm": 6.636974334716797,
|
| 446 |
+
"learning_rate": 3.246402877697842e-05,
|
| 447 |
+
"loss": 3.6062,
|
| 448 |
+
"step": 2450
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 12.316440049443758,
|
| 452 |
+
"grad_norm": 6.129552364349365,
|
| 453 |
+
"learning_rate": 3.201438848920863e-05,
|
| 454 |
+
"loss": 3.4693,
|
| 455 |
+
"step": 2500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 12.563658838071694,
|
| 459 |
+
"grad_norm": 7.458967208862305,
|
| 460 |
+
"learning_rate": 3.1564748201438845e-05,
|
| 461 |
+
"loss": 3.507,
|
| 462 |
+
"step": 2550
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 12.810877626699629,
|
| 466 |
+
"grad_norm": 6.7472243309021,
|
| 467 |
+
"learning_rate": 3.1115107913669065e-05,
|
| 468 |
+
"loss": 3.5042,
|
| 469 |
+
"step": 2600
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 13.0,
|
| 473 |
+
"eval_loss": 5.1072187423706055,
|
| 474 |
+
"eval_runtime": 14.225,
|
| 475 |
+
"eval_samples_per_second": 50.545,
|
| 476 |
+
"eval_steps_per_second": 6.327,
|
| 477 |
+
"step": 2639
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 13.054388133498145,
|
| 481 |
+
"grad_norm": 6.99562406539917,
|
| 482 |
+
"learning_rate": 3.066546762589928e-05,
|
| 483 |
+
"loss": 3.4307,
|
| 484 |
+
"step": 2650
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 13.301606922126082,
|
| 488 |
+
"grad_norm": 7.893692493438721,
|
| 489 |
+
"learning_rate": 3.0215827338129498e-05,
|
| 490 |
+
"loss": 3.2852,
|
| 491 |
+
"step": 2700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 13.548825710754016,
|
| 495 |
+
"grad_norm": 7.662129878997803,
|
| 496 |
+
"learning_rate": 2.976618705035971e-05,
|
| 497 |
+
"loss": 3.3116,
|
| 498 |
+
"step": 2750
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 13.796044499381953,
|
| 502 |
+
"grad_norm": 7.612554550170898,
|
| 503 |
+
"learning_rate": 2.931654676258993e-05,
|
| 504 |
+
"loss": 3.3243,
|
| 505 |
+
"step": 2800
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 14.0,
|
| 509 |
+
"eval_loss": 5.115802764892578,
|
| 510 |
+
"eval_runtime": 14.2589,
|
| 511 |
+
"eval_samples_per_second": 50.425,
|
| 512 |
+
"eval_steps_per_second": 6.312,
|
| 513 |
+
"step": 2842
|
| 514 |
+
}
|
| 515 |
+
],
|
| 516 |
+
"logging_steps": 50,
|
| 517 |
+
"max_steps": 6060,
|
| 518 |
+
"num_input_tokens_seen": 0,
|
| 519 |
+
"num_train_epochs": 30,
|
| 520 |
+
"save_steps": 500,
|
| 521 |
+
"stateful_callbacks": {
|
| 522 |
+
"EarlyStoppingCallback": {
|
| 523 |
+
"args": {
|
| 524 |
+
"early_stopping_patience": 3,
|
| 525 |
+
"early_stopping_threshold": 0.0
|
| 526 |
+
},
|
| 527 |
+
"attributes": {
|
| 528 |
+
"early_stopping_patience_counter": 2
|
| 529 |
+
}
|
| 530 |
+
},
|
| 531 |
+
"TrainerControl": {
|
| 532 |
+
"args": {
|
| 533 |
+
"should_epoch_stop": false,
|
| 534 |
+
"should_evaluate": false,
|
| 535 |
+
"should_log": false,
|
| 536 |
+
"should_save": true,
|
| 537 |
+
"should_training_stop": false
|
| 538 |
+
},
|
| 539 |
+
"attributes": {}
|
| 540 |
+
}
|
| 541 |
+
},
|
| 542 |
+
"total_flos": 2.3660516081664e+16,
|
| 543 |
+
"train_batch_size": 8,
|
| 544 |
+
"trial_name": null,
|
| 545 |
+
"trial_params": null
|
| 546 |
+
}
|
checkpoint-2842/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28ec087a0e6786ef2e0bcf02310b767379b6de459d7cb254c0c6ae4e881ba0e1
|
| 3 |
+
size 5304
|
checkpoint-3045/config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "ai-forever/rugpt3small_based_on_gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0"
|
| 14 |
+
},
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"LABEL_0": 0
|
| 18 |
+
},
|
| 19 |
+
"layer_norm_epsilon": 1e-05,
|
| 20 |
+
"model_type": "gpt2",
|
| 21 |
+
"n_ctx": 2048,
|
| 22 |
+
"n_embd": 768,
|
| 23 |
+
"n_head": 12,
|
| 24 |
+
"n_inner": null,
|
| 25 |
+
"n_layer": 12,
|
| 26 |
+
"n_positions": 2048,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"reorder_and_upcast_attn": false,
|
| 29 |
+
"resid_pdrop": 0.1,
|
| 30 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 31 |
+
"scale_attn_weights": true,
|
| 32 |
+
"summary_activation": null,
|
| 33 |
+
"summary_first_dropout": 0.1,
|
| 34 |
+
"summary_proj_to_labels": true,
|
| 35 |
+
"summary_type": "cls_index",
|
| 36 |
+
"summary_use_proj": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.48.3",
|
| 39 |
+
"use_cache": true,
|
| 40 |
+
"vocab_size": 50264
|
| 41 |
+
}
|
checkpoint-3045/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.3"
|
| 7 |
+
}
|
checkpoint-3045/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a77765d2854aa28043621c32dfe457dbd70e5ccd3c462b7da20e24e1321bedfe
|
| 3 |
+
size 500941440
|
checkpoint-3045/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90e0b26d5997481a5036e1cbf6a55f2d7133a8568f6cc47c3b00ef57de5118ec
|
| 3 |
+
size 1389114
|
checkpoint-3045/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a06712f07314b639891611e3372c71ae4f05e3f52fb233051f2b1da8dbc95c5c
|
| 3 |
+
size 14244
|
checkpoint-3045/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b973dad44b1d1fff3b9930c7bd03223dc77c72636ac760061bf865c110dc04db
|
| 3 |
+
size 1064
|
checkpoint-3045/trainer_state.json
ADDED
|
@@ -0,0 +1,582 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.091330528259277,
|
| 3 |
+
"best_model_checkpoint": "aristotle_new_layer_plain/checkpoint-2436",
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 3045,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.24721878862793573,
|
| 13 |
+
"grad_norm": 4.709590435028076,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 10.2189,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.49443757725587145,
|
| 20 |
+
"grad_norm": 3.6981208324432373,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 9.1727,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7416563658838071,
|
| 27 |
+
"grad_norm": 3.9341259002685547,
|
| 28 |
+
"learning_rate": 1.5e-05,
|
| 29 |
+
"loss": 8.6877,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9888751545117429,
|
| 34 |
+
"grad_norm": 3.341215133666992,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 8.1544,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_loss": 7.7336225509643555,
|
| 42 |
+
"eval_runtime": 14.2669,
|
| 43 |
+
"eval_samples_per_second": 50.396,
|
| 44 |
+
"eval_steps_per_second": 6.308,
|
| 45 |
+
"step": 203
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.2323856613102595,
|
| 49 |
+
"grad_norm": 4.8560404777526855,
|
| 50 |
+
"learning_rate": 2.5e-05,
|
| 51 |
+
"loss": 7.5653,
|
| 52 |
+
"step": 250
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.4796044499381953,
|
| 56 |
+
"grad_norm": 3.1774024963378906,
|
| 57 |
+
"learning_rate": 3e-05,
|
| 58 |
+
"loss": 7.1093,
|
| 59 |
+
"step": 300
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.726823238566131,
|
| 63 |
+
"grad_norm": 4.33836030960083,
|
| 64 |
+
"learning_rate": 3.5e-05,
|
| 65 |
+
"loss": 6.7529,
|
| 66 |
+
"step": 350
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.9740420271940669,
|
| 70 |
+
"grad_norm": 2.5972180366516113,
|
| 71 |
+
"learning_rate": 4e-05,
|
| 72 |
+
"loss": 6.536,
|
| 73 |
+
"step": 400
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.0,
|
| 77 |
+
"eval_loss": 6.426527500152588,
|
| 78 |
+
"eval_runtime": 14.2681,
|
| 79 |
+
"eval_samples_per_second": 50.392,
|
| 80 |
+
"eval_steps_per_second": 6.308,
|
| 81 |
+
"step": 406
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.2175525339925835,
|
| 85 |
+
"grad_norm": 3.43729567527771,
|
| 86 |
+
"learning_rate": 4.5e-05,
|
| 87 |
+
"loss": 6.3559,
|
| 88 |
+
"step": 450
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.464771322620519,
|
| 92 |
+
"grad_norm": 3.318251848220825,
|
| 93 |
+
"learning_rate": 5e-05,
|
| 94 |
+
"loss": 6.3251,
|
| 95 |
+
"step": 500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.711990111248455,
|
| 99 |
+
"grad_norm": 3.502115488052368,
|
| 100 |
+
"learning_rate": 4.9550359712230215e-05,
|
| 101 |
+
"loss": 6.1368,
|
| 102 |
+
"step": 550
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.9592088998763906,
|
| 106 |
+
"grad_norm": 3.497938394546509,
|
| 107 |
+
"learning_rate": 4.9100719424460435e-05,
|
| 108 |
+
"loss": 6.0775,
|
| 109 |
+
"step": 600
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 3.0,
|
| 113 |
+
"eval_loss": 6.036961555480957,
|
| 114 |
+
"eval_runtime": 14.2475,
|
| 115 |
+
"eval_samples_per_second": 50.465,
|
| 116 |
+
"eval_steps_per_second": 6.317,
|
| 117 |
+
"step": 609
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 3.202719406674907,
|
| 121 |
+
"grad_norm": 3.4355390071868896,
|
| 122 |
+
"learning_rate": 4.865107913669065e-05,
|
| 123 |
+
"loss": 5.8684,
|
| 124 |
+
"step": 650
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.449938195302843,
|
| 128 |
+
"grad_norm": 3.9220526218414307,
|
| 129 |
+
"learning_rate": 4.820143884892087e-05,
|
| 130 |
+
"loss": 5.8101,
|
| 131 |
+
"step": 700
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.6971569839307787,
|
| 135 |
+
"grad_norm": 3.782421827316284,
|
| 136 |
+
"learning_rate": 4.775179856115108e-05,
|
| 137 |
+
"loss": 5.7784,
|
| 138 |
+
"step": 750
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.9443757725587143,
|
| 142 |
+
"grad_norm": 3.5181567668914795,
|
| 143 |
+
"learning_rate": 4.7302158273381294e-05,
|
| 144 |
+
"loss": 5.7181,
|
| 145 |
+
"step": 800
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 4.0,
|
| 149 |
+
"eval_loss": 5.772126197814941,
|
| 150 |
+
"eval_runtime": 14.2949,
|
| 151 |
+
"eval_samples_per_second": 50.298,
|
| 152 |
+
"eval_steps_per_second": 6.296,
|
| 153 |
+
"step": 812
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 4.187886279357231,
|
| 157 |
+
"grad_norm": 3.6087594032287598,
|
| 158 |
+
"learning_rate": 4.685251798561151e-05,
|
| 159 |
+
"loss": 5.5154,
|
| 160 |
+
"step": 850
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 4.435105067985167,
|
| 164 |
+
"grad_norm": 3.8448667526245117,
|
| 165 |
+
"learning_rate": 4.640287769784173e-05,
|
| 166 |
+
"loss": 5.4664,
|
| 167 |
+
"step": 900
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.6823238566131025,
|
| 171 |
+
"grad_norm": 3.594693660736084,
|
| 172 |
+
"learning_rate": 4.595323741007194e-05,
|
| 173 |
+
"loss": 5.4121,
|
| 174 |
+
"step": 950
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.929542645241038,
|
| 178 |
+
"grad_norm": 3.6225693225860596,
|
| 179 |
+
"learning_rate": 4.550359712230216e-05,
|
| 180 |
+
"loss": 5.3158,
|
| 181 |
+
"step": 1000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 5.0,
|
| 185 |
+
"eval_loss": 5.521895885467529,
|
| 186 |
+
"eval_runtime": 14.2777,
|
| 187 |
+
"eval_samples_per_second": 50.358,
|
| 188 |
+
"eval_steps_per_second": 6.304,
|
| 189 |
+
"step": 1015
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 5.173053152039555,
|
| 193 |
+
"grad_norm": 4.245815277099609,
|
| 194 |
+
"learning_rate": 4.505395683453237e-05,
|
| 195 |
+
"loss": 5.1744,
|
| 196 |
+
"step": 1050
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 5.420271940667491,
|
| 200 |
+
"grad_norm": 4.306251525878906,
|
| 201 |
+
"learning_rate": 4.460431654676259e-05,
|
| 202 |
+
"loss": 5.1245,
|
| 203 |
+
"step": 1100
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.667490729295427,
|
| 207 |
+
"grad_norm": 3.7834959030151367,
|
| 208 |
+
"learning_rate": 4.4154676258992806e-05,
|
| 209 |
+
"loss": 5.0729,
|
| 210 |
+
"step": 1150
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.914709517923362,
|
| 214 |
+
"grad_norm": 4.298359394073486,
|
| 215 |
+
"learning_rate": 4.3705035971223026e-05,
|
| 216 |
+
"loss": 5.0558,
|
| 217 |
+
"step": 1200
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.0,
|
| 221 |
+
"eval_loss": 5.389599323272705,
|
| 222 |
+
"eval_runtime": 14.2956,
|
| 223 |
+
"eval_samples_per_second": 50.295,
|
| 224 |
+
"eval_steps_per_second": 6.296,
|
| 225 |
+
"step": 1218
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.158220024721879,
|
| 229 |
+
"grad_norm": 4.416134357452393,
|
| 230 |
+
"learning_rate": 4.325539568345324e-05,
|
| 231 |
+
"loss": 4.839,
|
| 232 |
+
"step": 1250
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 6.405438813349814,
|
| 236 |
+
"grad_norm": 4.565963268280029,
|
| 237 |
+
"learning_rate": 4.280575539568346e-05,
|
| 238 |
+
"loss": 4.8297,
|
| 239 |
+
"step": 1300
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.652657601977751,
|
| 243 |
+
"grad_norm": 4.854921817779541,
|
| 244 |
+
"learning_rate": 4.235611510791367e-05,
|
| 245 |
+
"loss": 4.8175,
|
| 246 |
+
"step": 1350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.899876390605686,
|
| 250 |
+
"grad_norm": 4.982056617736816,
|
| 251 |
+
"learning_rate": 4.1906474820143885e-05,
|
| 252 |
+
"loss": 4.8081,
|
| 253 |
+
"step": 1400
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 7.0,
|
| 257 |
+
"eval_loss": 5.254246711730957,
|
| 258 |
+
"eval_runtime": 14.2665,
|
| 259 |
+
"eval_samples_per_second": 50.398,
|
| 260 |
+
"eval_steps_per_second": 6.308,
|
| 261 |
+
"step": 1421
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.143386897404203,
|
| 265 |
+
"grad_norm": 4.195478439331055,
|
| 266 |
+
"learning_rate": 4.14568345323741e-05,
|
| 267 |
+
"loss": 4.6322,
|
| 268 |
+
"step": 1450
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.3906056860321385,
|
| 272 |
+
"grad_norm": 4.963181972503662,
|
| 273 |
+
"learning_rate": 4.100719424460432e-05,
|
| 274 |
+
"loss": 4.547,
|
| 275 |
+
"step": 1500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.637824474660074,
|
| 279 |
+
"grad_norm": 5.290962219238281,
|
| 280 |
+
"learning_rate": 4.055755395683453e-05,
|
| 281 |
+
"loss": 4.5553,
|
| 282 |
+
"step": 1550
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.88504326328801,
|
| 286 |
+
"grad_norm": 5.0038838386535645,
|
| 287 |
+
"learning_rate": 4.010791366906475e-05,
|
| 288 |
+
"loss": 4.5651,
|
| 289 |
+
"step": 1600
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"eval_loss": 5.183382987976074,
|
| 294 |
+
"eval_runtime": 14.2977,
|
| 295 |
+
"eval_samples_per_second": 50.288,
|
| 296 |
+
"eval_steps_per_second": 6.295,
|
| 297 |
+
"step": 1624
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.128553770086526,
|
| 301 |
+
"grad_norm": 5.3380446434021,
|
| 302 |
+
"learning_rate": 3.965827338129496e-05,
|
| 303 |
+
"loss": 4.3966,
|
| 304 |
+
"step": 1650
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 8.375772558714463,
|
| 308 |
+
"grad_norm": 5.339470863342285,
|
| 309 |
+
"learning_rate": 3.920863309352518e-05,
|
| 310 |
+
"loss": 4.3068,
|
| 311 |
+
"step": 1700
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 8.622991347342397,
|
| 315 |
+
"grad_norm": 4.9476189613342285,
|
| 316 |
+
"learning_rate": 3.8758992805755396e-05,
|
| 317 |
+
"loss": 4.3249,
|
| 318 |
+
"step": 1750
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.870210135970334,
|
| 322 |
+
"grad_norm": 5.430028915405273,
|
| 323 |
+
"learning_rate": 3.8309352517985616e-05,
|
| 324 |
+
"loss": 4.3407,
|
| 325 |
+
"step": 1800
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 9.0,
|
| 329 |
+
"eval_loss": 5.13620138168335,
|
| 330 |
+
"eval_runtime": 14.2616,
|
| 331 |
+
"eval_samples_per_second": 50.415,
|
| 332 |
+
"eval_steps_per_second": 6.311,
|
| 333 |
+
"step": 1827
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 9.11372064276885,
|
| 337 |
+
"grad_norm": 5.2561259269714355,
|
| 338 |
+
"learning_rate": 3.785971223021583e-05,
|
| 339 |
+
"loss": 4.1746,
|
| 340 |
+
"step": 1850
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 9.360939431396787,
|
| 344 |
+
"grad_norm": 5.811314105987549,
|
| 345 |
+
"learning_rate": 3.741007194244605e-05,
|
| 346 |
+
"loss": 4.1324,
|
| 347 |
+
"step": 1900
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 9.608158220024722,
|
| 351 |
+
"grad_norm": 5.552155017852783,
|
| 352 |
+
"learning_rate": 3.696043165467626e-05,
|
| 353 |
+
"loss": 4.1058,
|
| 354 |
+
"step": 1950
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.855377008652658,
|
| 358 |
+
"grad_norm": 6.073920726776123,
|
| 359 |
+
"learning_rate": 3.6510791366906475e-05,
|
| 360 |
+
"loss": 4.0436,
|
| 361 |
+
"step": 2000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 10.0,
|
| 365 |
+
"eval_loss": 5.104895114898682,
|
| 366 |
+
"eval_runtime": 14.2556,
|
| 367 |
+
"eval_samples_per_second": 50.436,
|
| 368 |
+
"eval_steps_per_second": 6.313,
|
| 369 |
+
"step": 2030
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 10.098887515451175,
|
| 373 |
+
"grad_norm": 5.994938373565674,
|
| 374 |
+
"learning_rate": 3.606115107913669e-05,
|
| 375 |
+
"loss": 3.99,
|
| 376 |
+
"step": 2050
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 10.34610630407911,
|
| 380 |
+
"grad_norm": 6.414961814880371,
|
| 381 |
+
"learning_rate": 3.561151079136691e-05,
|
| 382 |
+
"loss": 3.9013,
|
| 383 |
+
"step": 2100
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 10.593325092707046,
|
| 387 |
+
"grad_norm": 6.1248459815979,
|
| 388 |
+
"learning_rate": 3.516187050359712e-05,
|
| 389 |
+
"loss": 3.8884,
|
| 390 |
+
"step": 2150
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.840543881334982,
|
| 394 |
+
"grad_norm": 5.360867500305176,
|
| 395 |
+
"learning_rate": 3.471223021582734e-05,
|
| 396 |
+
"loss": 3.877,
|
| 397 |
+
"step": 2200
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 11.0,
|
| 401 |
+
"eval_loss": 5.103781700134277,
|
| 402 |
+
"eval_runtime": 14.2705,
|
| 403 |
+
"eval_samples_per_second": 50.384,
|
| 404 |
+
"eval_steps_per_second": 6.307,
|
| 405 |
+
"step": 2233
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 11.084054388133499,
|
| 409 |
+
"grad_norm": 5.840531349182129,
|
| 410 |
+
"learning_rate": 3.4262589928057554e-05,
|
| 411 |
+
"loss": 3.8216,
|
| 412 |
+
"step": 2250
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 11.331273176761433,
|
| 416 |
+
"grad_norm": 7.407821178436279,
|
| 417 |
+
"learning_rate": 3.3812949640287773e-05,
|
| 418 |
+
"loss": 3.6379,
|
| 419 |
+
"step": 2300
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 11.57849196538937,
|
| 423 |
+
"grad_norm": 7.770689487457275,
|
| 424 |
+
"learning_rate": 3.3363309352517986e-05,
|
| 425 |
+
"loss": 3.7063,
|
| 426 |
+
"step": 2350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.825710754017305,
|
| 430 |
+
"grad_norm": 6.17808198928833,
|
| 431 |
+
"learning_rate": 3.2913669064748206e-05,
|
| 432 |
+
"loss": 3.7008,
|
| 433 |
+
"step": 2400
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 12.0,
|
| 437 |
+
"eval_loss": 5.091330528259277,
|
| 438 |
+
"eval_runtime": 14.3396,
|
| 439 |
+
"eval_samples_per_second": 50.141,
|
| 440 |
+
"eval_steps_per_second": 6.276,
|
| 441 |
+
"step": 2436
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 12.069221260815821,
|
| 445 |
+
"grad_norm": 6.636974334716797,
|
| 446 |
+
"learning_rate": 3.246402877697842e-05,
|
| 447 |
+
"loss": 3.6062,
|
| 448 |
+
"step": 2450
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 12.316440049443758,
|
| 452 |
+
"grad_norm": 6.129552364349365,
|
| 453 |
+
"learning_rate": 3.201438848920863e-05,
|
| 454 |
+
"loss": 3.4693,
|
| 455 |
+
"step": 2500
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 12.563658838071694,
|
| 459 |
+
"grad_norm": 7.458967208862305,
|
| 460 |
+
"learning_rate": 3.1564748201438845e-05,
|
| 461 |
+
"loss": 3.507,
|
| 462 |
+
"step": 2550
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 12.810877626699629,
|
| 466 |
+
"grad_norm": 6.7472243309021,
|
| 467 |
+
"learning_rate": 3.1115107913669065e-05,
|
| 468 |
+
"loss": 3.5042,
|
| 469 |
+
"step": 2600
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 13.0,
|
| 473 |
+
"eval_loss": 5.1072187423706055,
|
| 474 |
+
"eval_runtime": 14.225,
|
| 475 |
+
"eval_samples_per_second": 50.545,
|
| 476 |
+
"eval_steps_per_second": 6.327,
|
| 477 |
+
"step": 2639
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 13.054388133498145,
|
| 481 |
+
"grad_norm": 6.99562406539917,
|
| 482 |
+
"learning_rate": 3.066546762589928e-05,
|
| 483 |
+
"loss": 3.4307,
|
| 484 |
+
"step": 2650
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 13.301606922126082,
|
| 488 |
+
"grad_norm": 7.893692493438721,
|
| 489 |
+
"learning_rate": 3.0215827338129498e-05,
|
| 490 |
+
"loss": 3.2852,
|
| 491 |
+
"step": 2700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 13.548825710754016,
|
| 495 |
+
"grad_norm": 7.662129878997803,
|
| 496 |
+
"learning_rate": 2.976618705035971e-05,
|
| 497 |
+
"loss": 3.3116,
|
| 498 |
+
"step": 2750
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 13.796044499381953,
|
| 502 |
+
"grad_norm": 7.612554550170898,
|
| 503 |
+
"learning_rate": 2.931654676258993e-05,
|
| 504 |
+
"loss": 3.3243,
|
| 505 |
+
"step": 2800
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 14.0,
|
| 509 |
+
"eval_loss": 5.115802764892578,
|
| 510 |
+
"eval_runtime": 14.2589,
|
| 511 |
+
"eval_samples_per_second": 50.425,
|
| 512 |
+
"eval_steps_per_second": 6.312,
|
| 513 |
+
"step": 2842
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 14.03955500618047,
|
| 517 |
+
"grad_norm": 6.775301456451416,
|
| 518 |
+
"learning_rate": 2.8866906474820144e-05,
|
| 519 |
+
"loss": 3.2555,
|
| 520 |
+
"step": 2850
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 14.286773794808406,
|
| 524 |
+
"grad_norm": 7.874576091766357,
|
| 525 |
+
"learning_rate": 2.841726618705036e-05,
|
| 526 |
+
"loss": 3.075,
|
| 527 |
+
"step": 2900
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 14.53399258343634,
|
| 531 |
+
"grad_norm": 7.965319633483887,
|
| 532 |
+
"learning_rate": 2.7967625899280573e-05,
|
| 533 |
+
"loss": 3.177,
|
| 534 |
+
"step": 2950
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 14.781211372064277,
|
| 538 |
+
"grad_norm": 7.535608291625977,
|
| 539 |
+
"learning_rate": 2.7517985611510793e-05,
|
| 540 |
+
"loss": 3.14,
|
| 541 |
+
"step": 3000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 15.0,
|
| 545 |
+
"eval_loss": 5.132053375244141,
|
| 546 |
+
"eval_runtime": 14.2378,
|
| 547 |
+
"eval_samples_per_second": 50.499,
|
| 548 |
+
"eval_steps_per_second": 6.321,
|
| 549 |
+
"step": 3045
|
| 550 |
+
}
|
| 551 |
+
],
|
| 552 |
+
"logging_steps": 50,
|
| 553 |
+
"max_steps": 6060,
|
| 554 |
+
"num_input_tokens_seen": 0,
|
| 555 |
+
"num_train_epochs": 30,
|
| 556 |
+
"save_steps": 500,
|
| 557 |
+
"stateful_callbacks": {
|
| 558 |
+
"EarlyStoppingCallback": {
|
| 559 |
+
"args": {
|
| 560 |
+
"early_stopping_patience": 3,
|
| 561 |
+
"early_stopping_threshold": 0.0
|
| 562 |
+
},
|
| 563 |
+
"attributes": {
|
| 564 |
+
"early_stopping_patience_counter": 3
|
| 565 |
+
}
|
| 566 |
+
},
|
| 567 |
+
"TrainerControl": {
|
| 568 |
+
"args": {
|
| 569 |
+
"should_epoch_stop": false,
|
| 570 |
+
"should_evaluate": false,
|
| 571 |
+
"should_log": false,
|
| 572 |
+
"should_save": true,
|
| 573 |
+
"should_training_stop": true
|
| 574 |
+
},
|
| 575 |
+
"attributes": {}
|
| 576 |
+
}
|
| 577 |
+
},
|
| 578 |
+
"total_flos": 2.535055294464e+16,
|
| 579 |
+
"train_batch_size": 8,
|
| 580 |
+
"trial_name": null,
|
| 581 |
+
"trial_params": null
|
| 582 |
+
}
|
checkpoint-3045/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28ec087a0e6786ef2e0bcf02310b767379b6de459d7cb254c0c6ae4e881ba0e1
|
| 3 |
+
size 5304
|