Upload 7 files
Browse files- config.json +35 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +775 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "malteos/gpt2-uk",
|
| 3 |
+
"activation_function": "gelu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_ctx": 1024,
|
| 16 |
+
"n_embd": 768,
|
| 17 |
+
"n_head": 12,
|
| 18 |
+
"n_inner": 3072,
|
| 19 |
+
"n_layer": 12,
|
| 20 |
+
"n_positions": 1024,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 31 |
+
"torch_dtype": "float32",
|
| 32 |
+
"transformers_version": "4.46.3",
|
| 33 |
+
"use_cache": true,
|
| 34 |
+
"vocab_size": 50304
|
| 35 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.46.3"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3781e402cb71f2643688eb3ddce2c6af9e4602086a22a9c2d298d4037a12573
|
| 3 |
+
size 497918592
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4747f4f8ab80488c29c9f988f359cfffdd4870a2176fb60e772409e8204234ea
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d455d2629f3992f6a52a2542a294f11f8fe2ae76bbeca6e2da72cf1d8c5bf138
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,775 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 10686,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02807411566535654,
|
| 13 |
+
"grad_norm": 8.254485130310059,
|
| 14 |
+
"learning_rate": 4.9550814149354296e-05,
|
| 15 |
+
"loss": 5.1202,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05614823133071308,
|
| 20 |
+
"grad_norm": 14.410130500793457,
|
| 21 |
+
"learning_rate": 4.908291222159835e-05,
|
| 22 |
+
"loss": 4.827,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08422234699606962,
|
| 27 |
+
"grad_norm": 12.091243743896484,
|
| 28 |
+
"learning_rate": 4.861501029384241e-05,
|
| 29 |
+
"loss": 4.8012,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.11229646266142616,
|
| 34 |
+
"grad_norm": 9.615291595458984,
|
| 35 |
+
"learning_rate": 4.814710836608647e-05,
|
| 36 |
+
"loss": 4.7478,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1403705783267827,
|
| 41 |
+
"grad_norm": 7.660737991333008,
|
| 42 |
+
"learning_rate": 4.767920643833052e-05,
|
| 43 |
+
"loss": 4.7,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.16844469399213924,
|
| 48 |
+
"grad_norm": 8.466590881347656,
|
| 49 |
+
"learning_rate": 4.7211304510574585e-05,
|
| 50 |
+
"loss": 4.7237,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1965188096574958,
|
| 55 |
+
"grad_norm": 9.133238792419434,
|
| 56 |
+
"learning_rate": 4.674340258281865e-05,
|
| 57 |
+
"loss": 4.6992,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.22459292532285233,
|
| 62 |
+
"grad_norm": 11.180377006530762,
|
| 63 |
+
"learning_rate": 4.62755006550627e-05,
|
| 64 |
+
"loss": 4.6911,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.25266704098820886,
|
| 69 |
+
"grad_norm": 9.834083557128906,
|
| 70 |
+
"learning_rate": 4.5807598727306764e-05,
|
| 71 |
+
"loss": 4.6565,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2807411566535654,
|
| 76 |
+
"grad_norm": 8.723132133483887,
|
| 77 |
+
"learning_rate": 4.533969679955082e-05,
|
| 78 |
+
"loss": 4.6333,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.308815272318922,
|
| 83 |
+
"grad_norm": 12.12197494506836,
|
| 84 |
+
"learning_rate": 4.4871794871794874e-05,
|
| 85 |
+
"loss": 4.5906,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3368893879842785,
|
| 90 |
+
"grad_norm": 10.753365516662598,
|
| 91 |
+
"learning_rate": 4.440389294403893e-05,
|
| 92 |
+
"loss": 4.5926,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.36496350364963503,
|
| 97 |
+
"grad_norm": 8.904462814331055,
|
| 98 |
+
"learning_rate": 4.393599101628299e-05,
|
| 99 |
+
"loss": 4.5735,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.3930376193149916,
|
| 104 |
+
"grad_norm": 7.0933098793029785,
|
| 105 |
+
"learning_rate": 4.346808908852705e-05,
|
| 106 |
+
"loss": 4.6032,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4211117349803481,
|
| 111 |
+
"grad_norm": 6.25530481338501,
|
| 112 |
+
"learning_rate": 4.30001871607711e-05,
|
| 113 |
+
"loss": 4.6286,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.44918585064570465,
|
| 118 |
+
"grad_norm": 9.821544647216797,
|
| 119 |
+
"learning_rate": 4.2532285233015164e-05,
|
| 120 |
+
"loss": 4.5785,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4772599663110612,
|
| 125 |
+
"grad_norm": 8.794018745422363,
|
| 126 |
+
"learning_rate": 4.206438330525922e-05,
|
| 127 |
+
"loss": 4.5527,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5053340819764177,
|
| 132 |
+
"grad_norm": 9.12629222869873,
|
| 133 |
+
"learning_rate": 4.1596481377503274e-05,
|
| 134 |
+
"loss": 4.5306,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5334081976417743,
|
| 139 |
+
"grad_norm": 8.344796180725098,
|
| 140 |
+
"learning_rate": 4.1128579449747336e-05,
|
| 141 |
+
"loss": 4.4843,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.5614823133071308,
|
| 146 |
+
"grad_norm": 11.365053176879883,
|
| 147 |
+
"learning_rate": 4.066067752199139e-05,
|
| 148 |
+
"loss": 4.4841,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5895564289724874,
|
| 153 |
+
"grad_norm": 8.378218650817871,
|
| 154 |
+
"learning_rate": 4.0192775594235447e-05,
|
| 155 |
+
"loss": 4.5196,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.617630544637844,
|
| 160 |
+
"grad_norm": 11.260799407958984,
|
| 161 |
+
"learning_rate": 3.972487366647951e-05,
|
| 162 |
+
"loss": 4.4967,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.6457046603032004,
|
| 167 |
+
"grad_norm": 11.992345809936523,
|
| 168 |
+
"learning_rate": 3.9256971738723564e-05,
|
| 169 |
+
"loss": 4.509,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.673778775968557,
|
| 174 |
+
"grad_norm": 15.889375686645508,
|
| 175 |
+
"learning_rate": 3.878906981096762e-05,
|
| 176 |
+
"loss": 4.5183,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.7018528916339135,
|
| 181 |
+
"grad_norm": 6.526124954223633,
|
| 182 |
+
"learning_rate": 3.832116788321168e-05,
|
| 183 |
+
"loss": 4.4316,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.7299270072992701,
|
| 188 |
+
"grad_norm": 8.866271018981934,
|
| 189 |
+
"learning_rate": 3.7853265955455736e-05,
|
| 190 |
+
"loss": 4.4426,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.7580011229646266,
|
| 195 |
+
"grad_norm": 8.329658508300781,
|
| 196 |
+
"learning_rate": 3.73853640276998e-05,
|
| 197 |
+
"loss": 4.445,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.7860752386299832,
|
| 202 |
+
"grad_norm": 8.942349433898926,
|
| 203 |
+
"learning_rate": 3.691746209994385e-05,
|
| 204 |
+
"loss": 4.4743,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.8141493542953397,
|
| 209 |
+
"grad_norm": 7.263224124908447,
|
| 210 |
+
"learning_rate": 3.6449560172187915e-05,
|
| 211 |
+
"loss": 4.4459,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.8422234699606962,
|
| 216 |
+
"grad_norm": 8.767597198486328,
|
| 217 |
+
"learning_rate": 3.598165824443197e-05,
|
| 218 |
+
"loss": 4.3944,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.8702975856260527,
|
| 223 |
+
"grad_norm": 7.760046482086182,
|
| 224 |
+
"learning_rate": 3.5513756316676025e-05,
|
| 225 |
+
"loss": 4.4683,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.8983717012914093,
|
| 230 |
+
"grad_norm": 13.036920547485352,
|
| 231 |
+
"learning_rate": 3.504585438892009e-05,
|
| 232 |
+
"loss": 4.3891,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.9264458169567659,
|
| 237 |
+
"grad_norm": 10.132473945617676,
|
| 238 |
+
"learning_rate": 3.457795246116414e-05,
|
| 239 |
+
"loss": 4.3914,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.9545199326221224,
|
| 244 |
+
"grad_norm": 10.985709190368652,
|
| 245 |
+
"learning_rate": 3.41100505334082e-05,
|
| 246 |
+
"loss": 4.3861,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.982594048287479,
|
| 251 |
+
"grad_norm": 12.080757141113281,
|
| 252 |
+
"learning_rate": 3.364214860565226e-05,
|
| 253 |
+
"loss": 4.4272,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.0106681639528354,
|
| 258 |
+
"grad_norm": 10.2216157913208,
|
| 259 |
+
"learning_rate": 3.3174246677896315e-05,
|
| 260 |
+
"loss": 4.1802,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.038742279618192,
|
| 265 |
+
"grad_norm": 7.825625896453857,
|
| 266 |
+
"learning_rate": 3.270634475014037e-05,
|
| 267 |
+
"loss": 3.7209,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.0668163952835485,
|
| 272 |
+
"grad_norm": 9.543134689331055,
|
| 273 |
+
"learning_rate": 3.223844282238443e-05,
|
| 274 |
+
"loss": 3.7451,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.094890510948905,
|
| 279 |
+
"grad_norm": 6.093784809112549,
|
| 280 |
+
"learning_rate": 3.177054089462849e-05,
|
| 281 |
+
"loss": 3.7287,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.1229646266142617,
|
| 286 |
+
"grad_norm": 9.133221626281738,
|
| 287 |
+
"learning_rate": 3.130263896687254e-05,
|
| 288 |
+
"loss": 3.7896,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.1510387422796182,
|
| 293 |
+
"grad_norm": 8.328789710998535,
|
| 294 |
+
"learning_rate": 3.0834737039116604e-05,
|
| 295 |
+
"loss": 3.7729,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.1791128579449748,
|
| 300 |
+
"grad_norm": 12.665905952453613,
|
| 301 |
+
"learning_rate": 3.036683511136066e-05,
|
| 302 |
+
"loss": 3.768,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.2071869736103313,
|
| 307 |
+
"grad_norm": 9.492897987365723,
|
| 308 |
+
"learning_rate": 2.9898933183604715e-05,
|
| 309 |
+
"loss": 3.7355,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.235261089275688,
|
| 314 |
+
"grad_norm": 10.17910385131836,
|
| 315 |
+
"learning_rate": 2.9431031255848773e-05,
|
| 316 |
+
"loss": 3.8411,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.2633352049410442,
|
| 321 |
+
"grad_norm": 7.939426898956299,
|
| 322 |
+
"learning_rate": 2.896312932809283e-05,
|
| 323 |
+
"loss": 3.7935,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.2914093206064008,
|
| 328 |
+
"grad_norm": 8.300692558288574,
|
| 329 |
+
"learning_rate": 2.8495227400336887e-05,
|
| 330 |
+
"loss": 3.7495,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.3194834362717573,
|
| 335 |
+
"grad_norm": 11.419922828674316,
|
| 336 |
+
"learning_rate": 2.8027325472580952e-05,
|
| 337 |
+
"loss": 3.8478,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.347557551937114,
|
| 342 |
+
"grad_norm": 12.745059967041016,
|
| 343 |
+
"learning_rate": 2.7559423544825007e-05,
|
| 344 |
+
"loss": 3.8148,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.3756316676024705,
|
| 349 |
+
"grad_norm": 12.494312286376953,
|
| 350 |
+
"learning_rate": 2.7091521617069066e-05,
|
| 351 |
+
"loss": 3.7696,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.403705783267827,
|
| 356 |
+
"grad_norm": 7.439785480499268,
|
| 357 |
+
"learning_rate": 2.6623619689313124e-05,
|
| 358 |
+
"loss": 3.7559,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.4317798989331836,
|
| 363 |
+
"grad_norm": 12.009989738464355,
|
| 364 |
+
"learning_rate": 2.615571776155718e-05,
|
| 365 |
+
"loss": 3.7759,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.4598540145985401,
|
| 370 |
+
"grad_norm": 10.613336563110352,
|
| 371 |
+
"learning_rate": 2.5687815833801238e-05,
|
| 372 |
+
"loss": 3.7577,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.4879281302638967,
|
| 377 |
+
"grad_norm": 10.656264305114746,
|
| 378 |
+
"learning_rate": 2.5219913906045293e-05,
|
| 379 |
+
"loss": 3.7515,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.5160022459292533,
|
| 384 |
+
"grad_norm": 9.189164161682129,
|
| 385 |
+
"learning_rate": 2.4752011978289352e-05,
|
| 386 |
+
"loss": 3.8009,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.5440763615946098,
|
| 391 |
+
"grad_norm": 9.947083473205566,
|
| 392 |
+
"learning_rate": 2.428411005053341e-05,
|
| 393 |
+
"loss": 3.8175,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.5721504772599664,
|
| 398 |
+
"grad_norm": 14.514716148376465,
|
| 399 |
+
"learning_rate": 2.3816208122777466e-05,
|
| 400 |
+
"loss": 3.7141,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.600224592925323,
|
| 405 |
+
"grad_norm": 8.071717262268066,
|
| 406 |
+
"learning_rate": 2.3348306195021524e-05,
|
| 407 |
+
"loss": 3.7252,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.6282987085906795,
|
| 412 |
+
"grad_norm": 11.72280216217041,
|
| 413 |
+
"learning_rate": 2.2880404267265583e-05,
|
| 414 |
+
"loss": 3.7949,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.656372824256036,
|
| 419 |
+
"grad_norm": 9.227740287780762,
|
| 420 |
+
"learning_rate": 2.2412502339509638e-05,
|
| 421 |
+
"loss": 3.733,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.6844469399213926,
|
| 426 |
+
"grad_norm": 13.090754508972168,
|
| 427 |
+
"learning_rate": 2.1944600411753697e-05,
|
| 428 |
+
"loss": 3.8362,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.7125210555867492,
|
| 433 |
+
"grad_norm": 12.168136596679688,
|
| 434 |
+
"learning_rate": 2.1476698483997755e-05,
|
| 435 |
+
"loss": 3.7261,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.7405951712521057,
|
| 440 |
+
"grad_norm": 8.332971572875977,
|
| 441 |
+
"learning_rate": 2.1008796556241814e-05,
|
| 442 |
+
"loss": 3.8215,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.768669286917462,
|
| 447 |
+
"grad_norm": 8.5094633102417,
|
| 448 |
+
"learning_rate": 2.0540894628485872e-05,
|
| 449 |
+
"loss": 3.804,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.7967434025828186,
|
| 454 |
+
"grad_norm": 11.575583457946777,
|
| 455 |
+
"learning_rate": 2.0072992700729927e-05,
|
| 456 |
+
"loss": 3.7369,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.8248175182481752,
|
| 461 |
+
"grad_norm": 12.597991943359375,
|
| 462 |
+
"learning_rate": 1.9605090772973986e-05,
|
| 463 |
+
"loss": 3.7377,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.8528916339135317,
|
| 468 |
+
"grad_norm": 9.923151016235352,
|
| 469 |
+
"learning_rate": 1.9137188845218044e-05,
|
| 470 |
+
"loss": 3.7808,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.8809657495788883,
|
| 475 |
+
"grad_norm": 10.516582489013672,
|
| 476 |
+
"learning_rate": 1.86692869174621e-05,
|
| 477 |
+
"loss": 3.7584,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.9090398652442448,
|
| 482 |
+
"grad_norm": 8.432146072387695,
|
| 483 |
+
"learning_rate": 1.8201384989706158e-05,
|
| 484 |
+
"loss": 3.7837,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.9371139809096012,
|
| 489 |
+
"grad_norm": 12.533796310424805,
|
| 490 |
+
"learning_rate": 1.7733483061950217e-05,
|
| 491 |
+
"loss": 3.6908,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.9651880965749577,
|
| 496 |
+
"grad_norm": 11.924694061279297,
|
| 497 |
+
"learning_rate": 1.7265581134194272e-05,
|
| 498 |
+
"loss": 3.7846,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.9932622122403143,
|
| 503 |
+
"grad_norm": 10.164834976196289,
|
| 504 |
+
"learning_rate": 1.6797679206438334e-05,
|
| 505 |
+
"loss": 3.766,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.021336327905671,
|
| 510 |
+
"grad_norm": 9.073919296264648,
|
| 511 |
+
"learning_rate": 1.633445629795995e-05,
|
| 512 |
+
"loss": 3.3417,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.0494104435710274,
|
| 517 |
+
"grad_norm": 10.306299209594727,
|
| 518 |
+
"learning_rate": 1.5866554370204005e-05,
|
| 519 |
+
"loss": 3.2644,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.077484559236384,
|
| 524 |
+
"grad_norm": 11.006250381469727,
|
| 525 |
+
"learning_rate": 1.5398652442448064e-05,
|
| 526 |
+
"loss": 3.243,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.1055586749017405,
|
| 531 |
+
"grad_norm": 9.138731956481934,
|
| 532 |
+
"learning_rate": 1.4930750514692121e-05,
|
| 533 |
+
"loss": 3.2938,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.133632790567097,
|
| 538 |
+
"grad_norm": 9.330265998840332,
|
| 539 |
+
"learning_rate": 1.4462848586936178e-05,
|
| 540 |
+
"loss": 3.247,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.1617069062324537,
|
| 545 |
+
"grad_norm": 9.435338973999023,
|
| 546 |
+
"learning_rate": 1.3994946659180236e-05,
|
| 547 |
+
"loss": 3.2847,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.18978102189781,
|
| 552 |
+
"grad_norm": 9.446837425231934,
|
| 553 |
+
"learning_rate": 1.3527044731424293e-05,
|
| 554 |
+
"loss": 3.2522,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 2.2178551375631668,
|
| 559 |
+
"grad_norm": 11.070699691772461,
|
| 560 |
+
"learning_rate": 1.305914280366835e-05,
|
| 561 |
+
"loss": 3.2073,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.2459292532285233,
|
| 566 |
+
"grad_norm": 9.919466018676758,
|
| 567 |
+
"learning_rate": 1.259124087591241e-05,
|
| 568 |
+
"loss": 3.2816,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.27400336889388,
|
| 573 |
+
"grad_norm": 11.001142501831055,
|
| 574 |
+
"learning_rate": 1.2123338948156467e-05,
|
| 575 |
+
"loss": 3.2669,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.3020774845592364,
|
| 580 |
+
"grad_norm": 10.917780876159668,
|
| 581 |
+
"learning_rate": 1.1655437020400524e-05,
|
| 582 |
+
"loss": 3.2488,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.330151600224593,
|
| 587 |
+
"grad_norm": 10.114619255065918,
|
| 588 |
+
"learning_rate": 1.1187535092644583e-05,
|
| 589 |
+
"loss": 3.2317,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.3582257158899496,
|
| 594 |
+
"grad_norm": 8.730778694152832,
|
| 595 |
+
"learning_rate": 1.071963316488864e-05,
|
| 596 |
+
"loss": 3.274,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.386299831555306,
|
| 601 |
+
"grad_norm": 7.530703544616699,
|
| 602 |
+
"learning_rate": 1.0251731237132698e-05,
|
| 603 |
+
"loss": 3.2506,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.4143739472206627,
|
| 608 |
+
"grad_norm": 13.161232948303223,
|
| 609 |
+
"learning_rate": 9.783829309376755e-06,
|
| 610 |
+
"loss": 3.3056,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.4424480628860192,
|
| 615 |
+
"grad_norm": 7.7642598152160645,
|
| 616 |
+
"learning_rate": 9.315927381620813e-06,
|
| 617 |
+
"loss": 3.2745,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.470522178551376,
|
| 622 |
+
"grad_norm": 8.774062156677246,
|
| 623 |
+
"learning_rate": 8.84802545386487e-06,
|
| 624 |
+
"loss": 3.263,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.4985962942167323,
|
| 629 |
+
"grad_norm": 13.077786445617676,
|
| 630 |
+
"learning_rate": 8.380123526108929e-06,
|
| 631 |
+
"loss": 3.2418,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.5266704098820885,
|
| 636 |
+
"grad_norm": 20.591571807861328,
|
| 637 |
+
"learning_rate": 7.912221598352986e-06,
|
| 638 |
+
"loss": 3.3027,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.554744525547445,
|
| 643 |
+
"grad_norm": 10.915307998657227,
|
| 644 |
+
"learning_rate": 7.444319670597043e-06,
|
| 645 |
+
"loss": 3.2292,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.5828186412128016,
|
| 650 |
+
"grad_norm": 10.937311172485352,
|
| 651 |
+
"learning_rate": 6.976417742841101e-06,
|
| 652 |
+
"loss": 3.3281,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.610892756878158,
|
| 657 |
+
"grad_norm": 8.748698234558105,
|
| 658 |
+
"learning_rate": 6.508515815085159e-06,
|
| 659 |
+
"loss": 3.2823,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.6389668725435147,
|
| 664 |
+
"grad_norm": 8.032021522521973,
|
| 665 |
+
"learning_rate": 6.040613887329216e-06,
|
| 666 |
+
"loss": 3.2265,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.6670409882088713,
|
| 671 |
+
"grad_norm": 14.668137550354004,
|
| 672 |
+
"learning_rate": 5.5727119595732735e-06,
|
| 673 |
+
"loss": 3.2673,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.695115103874228,
|
| 678 |
+
"grad_norm": 9.834213256835938,
|
| 679 |
+
"learning_rate": 5.104810031817331e-06,
|
| 680 |
+
"loss": 3.2213,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.7231892195395844,
|
| 685 |
+
"grad_norm": 14.294054985046387,
|
| 686 |
+
"learning_rate": 4.636908104061389e-06,
|
| 687 |
+
"loss": 3.2487,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.751263335204941,
|
| 692 |
+
"grad_norm": 8.172629356384277,
|
| 693 |
+
"learning_rate": 4.169006176305447e-06,
|
| 694 |
+
"loss": 3.2598,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.7793374508702975,
|
| 699 |
+
"grad_norm": 18.083383560180664,
|
| 700 |
+
"learning_rate": 3.7011042485495043e-06,
|
| 701 |
+
"loss": 3.2655,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.807411566535654,
|
| 706 |
+
"grad_norm": 13.841841697692871,
|
| 707 |
+
"learning_rate": 3.2332023207935616e-06,
|
| 708 |
+
"loss": 3.2676,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.8354856822010106,
|
| 713 |
+
"grad_norm": 16.98509407043457,
|
| 714 |
+
"learning_rate": 2.7653003930376193e-06,
|
| 715 |
+
"loss": 3.2499,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.863559797866367,
|
| 720 |
+
"grad_norm": 10.365632057189941,
|
| 721 |
+
"learning_rate": 2.297398465281677e-06,
|
| 722 |
+
"loss": 3.2457,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.8916339135317237,
|
| 727 |
+
"grad_norm": 10.886728286743164,
|
| 728 |
+
"learning_rate": 1.8294965375257348e-06,
|
| 729 |
+
"loss": 3.2445,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.9197080291970803,
|
| 734 |
+
"grad_norm": 10.890334129333496,
|
| 735 |
+
"learning_rate": 1.3615946097697923e-06,
|
| 736 |
+
"loss": 3.211,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.947782144862437,
|
| 741 |
+
"grad_norm": 11.197315216064453,
|
| 742 |
+
"learning_rate": 8.936926820138499e-07,
|
| 743 |
+
"loss": 3.2483,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.9758562605277934,
|
| 748 |
+
"grad_norm": 9.767260551452637,
|
| 749 |
+
"learning_rate": 4.2579075425790754e-07,
|
| 750 |
+
"loss": 3.2594,
|
| 751 |
+
"step": 10600
|
| 752 |
+
}
|
| 753 |
+
],
|
| 754 |
+
"logging_steps": 100,
|
| 755 |
+
"max_steps": 10686,
|
| 756 |
+
"num_input_tokens_seen": 0,
|
| 757 |
+
"num_train_epochs": 3,
|
| 758 |
+
"save_steps": 500,
|
| 759 |
+
"stateful_callbacks": {
|
| 760 |
+
"TrainerControl": {
|
| 761 |
+
"args": {
|
| 762 |
+
"should_epoch_stop": false,
|
| 763 |
+
"should_evaluate": false,
|
| 764 |
+
"should_log": false,
|
| 765 |
+
"should_save": true,
|
| 766 |
+
"should_training_stop": true
|
| 767 |
+
},
|
| 768 |
+
"attributes": {}
|
| 769 |
+
}
|
| 770 |
+
},
|
| 771 |
+
"total_flos": 5583353462784000.0,
|
| 772 |
+
"train_batch_size": 8,
|
| 773 |
+
"trial_name": null,
|
| 774 |
+
"trial_params": null
|
| 775 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6245000f2e4ee1bf05e28f050a11efd7bd1170bac65d96dac5b44bd259a228a
|
| 3 |
+
size 5304
|