Commit
•
d126247
1
Parent(s):
4088083
Upload with huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitignore +1 -0
- checkpoint-106496/config.json +32 -0
- checkpoint-106496/optimizer.pt +3 -0
- checkpoint-106496/pytorch_model.bin +3 -0
- checkpoint-106496/rng_state.pth +3 -0
- checkpoint-106496/scaler.pt +3 -0
- checkpoint-106496/scheduler.pt +3 -0
- checkpoint-106496/trainer_state.json +744 -0
- checkpoint-106496/training_args.bin +3 -0
- checkpoint-114688/config.json +32 -0
- checkpoint-114688/optimizer.pt +3 -0
- checkpoint-114688/pytorch_model.bin +3 -0
- checkpoint-114688/rng_state.pth +3 -0
- checkpoint-114688/scaler.pt +3 -0
- checkpoint-114688/scheduler.pt +3 -0
- checkpoint-114688/trainer_state.json +800 -0
- checkpoint-114688/training_args.bin +3 -0
- checkpoint-81920/config.json +32 -0
- checkpoint-81920/optimizer.pt +3 -0
- checkpoint-81920/pytorch_model.bin +3 -0
- checkpoint-81920/rng_state.pth +3 -0
- checkpoint-81920/scaler.pt +3 -0
- checkpoint-81920/scheduler.pt +3 -0
- checkpoint-81920/trainer_state.json +576 -0
- checkpoint-81920/training_args.bin +3 -0
- checkpoint-90112/config.json +32 -0
- checkpoint-90112/optimizer.pt +3 -0
- checkpoint-90112/pytorch_model.bin +3 -0
- checkpoint-90112/rng_state.pth +3 -0
- checkpoint-90112/scaler.pt +3 -0
- checkpoint-90112/scheduler.pt +3 -0
- checkpoint-90112/trainer_state.json +632 -0
- checkpoint-90112/training_args.bin +3 -0
- checkpoint-98304/config.json +32 -0
- checkpoint-98304/optimizer.pt +3 -0
- checkpoint-98304/pytorch_model.bin +3 -0
- checkpoint-98304/rng_state.pth +3 -0
- checkpoint-98304/scaler.pt +3 -0
- checkpoint-98304/scheduler.pt +3 -0
- checkpoint-98304/trainer_state.json +688 -0
- checkpoint-98304/training_args.bin +3 -0
- config.json +32 -0
- manual_upload/manual_upload/manual_upload/training_args.json +1 -1
- manual_upload/manual_upload/training_args.json +1 -1
- manual_upload/training_args.json +1 -1
- pytorch_model.bin +3 -0
- special_tokens_map.json +3 -0
- tokenizer.json +357 -0
- tokenizer_config.json +4 -0
- trainer_state.json +15 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
checkpoint-106496/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
checkpoint-106496/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8953e89f02666525bd9a498c7c03aaeec5b1e62af8b1eb3461cb6932dff6e5b0
|
| 3 |
+
size 211432837
|
checkpoint-106496/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5562c193530a7b41f4b1ea4f3c8f66a7c3ca0885b673e4653afbb4e4e42a466a
|
| 3 |
+
size 139279005
|
checkpoint-106496/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04127a62b9953a815a2b8b0e80ea0aed36762c9a4402334a3728a971ad89d02a
|
| 3 |
+
size 15597
|
checkpoint-106496/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6757561c1e760e6da06aa4557681c0a6a9b0e2ee848768a273634675848aa501
|
| 3 |
+
size 557
|
checkpoint-106496/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbe670698963718269271b891ad22eb717988aca0bb6ff0290b216e356014cd1
|
| 3 |
+
size 627
|
checkpoint-106496/trainer_state.json
ADDED
|
@@ -0,0 +1,744 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.183056792121948,
|
| 5 |
+
"global_step": 106496,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 0.00020480000000000002,
|
| 13 |
+
"loss": 2.1719,
|
| 14 |
+
"step": 2048
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.0833024978637695,
|
| 19 |
+
"eval_runtime": 15.7887,
|
| 20 |
+
"eval_samples_per_second": 132.5,
|
| 21 |
+
"eval_steps_per_second": 8.297,
|
| 22 |
+
"step": 2048
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 0.00040960000000000004,
|
| 27 |
+
"loss": 0.9468,
|
| 28 |
+
"step": 4096
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 0.8000818490982056,
|
| 33 |
+
"eval_runtime": 17.2632,
|
| 34 |
+
"eval_samples_per_second": 121.183,
|
| 35 |
+
"eval_steps_per_second": 7.588,
|
| 36 |
+
"step": 4096
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.41,
|
| 40 |
+
"learning_rate": 0.0004998749142723946,
|
| 41 |
+
"loss": 0.7542,
|
| 42 |
+
"step": 6144
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41,
|
| 46 |
+
"eval_loss": 0.7439925074577332,
|
| 47 |
+
"eval_runtime": 17.4134,
|
| 48 |
+
"eval_samples_per_second": 120.138,
|
| 49 |
+
"eval_steps_per_second": 7.523,
|
| 50 |
+
"step": 6144
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.55,
|
| 54 |
+
"learning_rate": 0.0004990273340312486,
|
| 55 |
+
"loss": 0.6756,
|
| 56 |
+
"step": 8192
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"eval_loss": 0.7018134593963623,
|
| 61 |
+
"eval_runtime": 15.8601,
|
| 62 |
+
"eval_samples_per_second": 131.904,
|
| 63 |
+
"eval_steps_per_second": 8.26,
|
| 64 |
+
"step": 8192
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.69,
|
| 68 |
+
"learning_rate": 0.0004973820371438889,
|
| 69 |
+
"loss": 0.631,
|
| 70 |
+
"step": 10240
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.69,
|
| 74 |
+
"eval_loss": 0.684688925743103,
|
| 75 |
+
"eval_runtime": 18.0314,
|
| 76 |
+
"eval_samples_per_second": 116.02,
|
| 77 |
+
"eval_steps_per_second": 7.265,
|
| 78 |
+
"step": 10240
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.83,
|
| 82 |
+
"learning_rate": 0.0004949442940386407,
|
| 83 |
+
"loss": 0.5989,
|
| 84 |
+
"step": 12288
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.83,
|
| 88 |
+
"eval_loss": 0.6822534203529358,
|
| 89 |
+
"eval_runtime": 17.4182,
|
| 90 |
+
"eval_samples_per_second": 120.104,
|
| 91 |
+
"eval_steps_per_second": 7.521,
|
| 92 |
+
"step": 12288
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.97,
|
| 96 |
+
"learning_rate": 0.0004917201492451735,
|
| 97 |
+
"loss": 0.5776,
|
| 98 |
+
"step": 14336
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"eval_loss": 0.6674544215202332,
|
| 103 |
+
"eval_runtime": 17.7487,
|
| 104 |
+
"eval_samples_per_second": 117.868,
|
| 105 |
+
"eval_steps_per_second": 7.381,
|
| 106 |
+
"step": 14336
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.11,
|
| 110 |
+
"learning_rate": 0.0004877230785006913,
|
| 111 |
+
"loss": 0.5525,
|
| 112 |
+
"step": 16384
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.11,
|
| 116 |
+
"eval_loss": 0.6640163660049438,
|
| 117 |
+
"eval_runtime": 15.8951,
|
| 118 |
+
"eval_samples_per_second": 131.613,
|
| 119 |
+
"eval_steps_per_second": 8.242,
|
| 120 |
+
"step": 16384
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 0.0004829670105577666,
|
| 125 |
+
"loss": 0.5373,
|
| 126 |
+
"step": 18432
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.24,
|
| 130 |
+
"eval_loss": 0.6663170456886292,
|
| 131 |
+
"eval_runtime": 17.5,
|
| 132 |
+
"eval_samples_per_second": 119.543,
|
| 133 |
+
"eval_steps_per_second": 7.486,
|
| 134 |
+
"step": 18432
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.38,
|
| 138 |
+
"learning_rate": 0.0004774596641323791,
|
| 139 |
+
"loss": 0.5261,
|
| 140 |
+
"step": 20480
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"eval_loss": 0.6518951654434204,
|
| 145 |
+
"eval_runtime": 17.1326,
|
| 146 |
+
"eval_samples_per_second": 122.107,
|
| 147 |
+
"eval_steps_per_second": 7.646,
|
| 148 |
+
"step": 20480
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.52,
|
| 152 |
+
"learning_rate": 0.0004712261976082475,
|
| 153 |
+
"loss": 0.5115,
|
| 154 |
+
"step": 22528
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.52,
|
| 158 |
+
"eval_loss": 0.6658991575241089,
|
| 159 |
+
"eval_runtime": 17.6874,
|
| 160 |
+
"eval_samples_per_second": 118.277,
|
| 161 |
+
"eval_steps_per_second": 7.406,
|
| 162 |
+
"step": 22528
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.66,
|
| 166 |
+
"learning_rate": 0.00046428407064289515,
|
| 167 |
+
"loss": 0.5033,
|
| 168 |
+
"step": 24576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.66,
|
| 172 |
+
"eval_loss": 0.6602604389190674,
|
| 173 |
+
"eval_runtime": 15.8845,
|
| 174 |
+
"eval_samples_per_second": 131.701,
|
| 175 |
+
"eval_steps_per_second": 8.247,
|
| 176 |
+
"step": 24576
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8,
|
| 180 |
+
"learning_rate": 0.00045665163060732317,
|
| 181 |
+
"loss": 0.4925,
|
| 182 |
+
"step": 26624
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.8,
|
| 186 |
+
"eval_loss": 0.6551440954208374,
|
| 187 |
+
"eval_runtime": 17.7902,
|
| 188 |
+
"eval_samples_per_second": 117.593,
|
| 189 |
+
"eval_steps_per_second": 7.364,
|
| 190 |
+
"step": 26624
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.93,
|
| 194 |
+
"learning_rate": 0.000448360778288479,
|
| 195 |
+
"loss": 0.4868,
|
| 196 |
+
"step": 28672
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.93,
|
| 200 |
+
"eval_loss": 0.6580803394317627,
|
| 201 |
+
"eval_runtime": 15.7841,
|
| 202 |
+
"eval_samples_per_second": 132.538,
|
| 203 |
+
"eval_steps_per_second": 8.299,
|
| 204 |
+
"step": 28672
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.00043942999964293453,
|
| 209 |
+
"loss": 0.4741,
|
| 210 |
+
"step": 30720
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.07,
|
| 214 |
+
"eval_loss": 0.661230742931366,
|
| 215 |
+
"eval_runtime": 18.015,
|
| 216 |
+
"eval_samples_per_second": 116.126,
|
| 217 |
+
"eval_steps_per_second": 7.272,
|
| 218 |
+
"step": 30720
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.21,
|
| 222 |
+
"learning_rate": 0.0004298966220346151,
|
| 223 |
+
"loss": 0.4632,
|
| 224 |
+
"step": 32768
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.21,
|
| 228 |
+
"eval_loss": 0.6618030071258545,
|
| 229 |
+
"eval_runtime": 17.9073,
|
| 230 |
+
"eval_samples_per_second": 116.824,
|
| 231 |
+
"eval_steps_per_second": 7.315,
|
| 232 |
+
"step": 32768
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.35,
|
| 236 |
+
"learning_rate": 0.00041978697624050446,
|
| 237 |
+
"loss": 0.4584,
|
| 238 |
+
"step": 34816
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.35,
|
| 242 |
+
"eval_loss": 0.6579039096832275,
|
| 243 |
+
"eval_runtime": 17.8241,
|
| 244 |
+
"eval_samples_per_second": 117.369,
|
| 245 |
+
"eval_steps_per_second": 7.35,
|
| 246 |
+
"step": 34816
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.49,
|
| 250 |
+
"learning_rate": 0.0004091334467888659,
|
| 251 |
+
"loss": 0.451,
|
| 252 |
+
"step": 36864
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"eval_loss": 0.6627541780471802,
|
| 257 |
+
"eval_runtime": 17.6822,
|
| 258 |
+
"eval_samples_per_second": 118.311,
|
| 259 |
+
"eval_steps_per_second": 7.409,
|
| 260 |
+
"step": 36864
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.62,
|
| 264 |
+
"learning_rate": 0.00039796458815002033,
|
| 265 |
+
"loss": 0.445,
|
| 266 |
+
"step": 38912
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.62,
|
| 270 |
+
"eval_loss": 0.656867265701294,
|
| 271 |
+
"eval_runtime": 15.8659,
|
| 272 |
+
"eval_samples_per_second": 131.855,
|
| 273 |
+
"eval_steps_per_second": 8.257,
|
| 274 |
+
"step": 38912
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.76,
|
| 278 |
+
"learning_rate": 0.0003863212870708643,
|
| 279 |
+
"loss": 0.4404,
|
| 280 |
+
"step": 40960
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.76,
|
| 284 |
+
"eval_loss": 0.6653844118118286,
|
| 285 |
+
"eval_runtime": 17.9192,
|
| 286 |
+
"eval_samples_per_second": 116.747,
|
| 287 |
+
"eval_steps_per_second": 7.311,
|
| 288 |
+
"step": 40960
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.9,
|
| 292 |
+
"learning_rate": 0.00037424687637876156,
|
| 293 |
+
"loss": 0.4334,
|
| 294 |
+
"step": 43008
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.9,
|
| 298 |
+
"eval_loss": 0.6553301811218262,
|
| 299 |
+
"eval_runtime": 15.9304,
|
| 300 |
+
"eval_samples_per_second": 131.321,
|
| 301 |
+
"eval_steps_per_second": 8.223,
|
| 302 |
+
"step": 43008
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 0.00036178064571954134,
|
| 307 |
+
"loss": 0.4278,
|
| 308 |
+
"step": 45056
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.04,
|
| 312 |
+
"eval_loss": 0.6766741871833801,
|
| 313 |
+
"eval_runtime": 16.9834,
|
| 314 |
+
"eval_samples_per_second": 123.179,
|
| 315 |
+
"eval_steps_per_second": 7.713,
|
| 316 |
+
"step": 45056
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.18,
|
| 320 |
+
"learning_rate": 0.0003489439971353363,
|
| 321 |
+
"loss": 0.4152,
|
| 322 |
+
"step": 47104
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.18,
|
| 326 |
+
"eval_loss": 0.6645193696022034,
|
| 327 |
+
"eval_runtime": 15.816,
|
| 328 |
+
"eval_samples_per_second": 132.271,
|
| 329 |
+
"eval_steps_per_second": 8.283,
|
| 330 |
+
"step": 47104
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.32,
|
| 334 |
+
"learning_rate": 0.0003357965820476752,
|
| 335 |
+
"loss": 0.411,
|
| 336 |
+
"step": 49152
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.32,
|
| 340 |
+
"eval_loss": 0.6748972535133362,
|
| 341 |
+
"eval_runtime": 17.7507,
|
| 342 |
+
"eval_samples_per_second": 117.855,
|
| 343 |
+
"eval_steps_per_second": 7.38,
|
| 344 |
+
"step": 49152
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.45,
|
| 348 |
+
"learning_rate": 0.0003223743322236833,
|
| 349 |
+
"loss": 0.4085,
|
| 350 |
+
"step": 51200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.45,
|
| 354 |
+
"eval_loss": 0.6703893542289734,
|
| 355 |
+
"eval_runtime": 17.341,
|
| 356 |
+
"eval_samples_per_second": 120.639,
|
| 357 |
+
"eval_steps_per_second": 7.554,
|
| 358 |
+
"step": 51200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.59,
|
| 362 |
+
"learning_rate": 0.0003087269633577651,
|
| 363 |
+
"loss": 0.4037,
|
| 364 |
+
"step": 53248
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.59,
|
| 368 |
+
"eval_loss": 0.6561577916145325,
|
| 369 |
+
"eval_runtime": 15.9295,
|
| 370 |
+
"eval_samples_per_second": 131.329,
|
| 371 |
+
"eval_steps_per_second": 8.224,
|
| 372 |
+
"step": 53248
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.73,
|
| 376 |
+
"learning_rate": 0.0002948780545870247,
|
| 377 |
+
"loss": 0.3975,
|
| 378 |
+
"step": 55296
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.73,
|
| 382 |
+
"eval_loss": 0.6698916554450989,
|
| 383 |
+
"eval_runtime": 17.9008,
|
| 384 |
+
"eval_samples_per_second": 116.866,
|
| 385 |
+
"eval_steps_per_second": 7.318,
|
| 386 |
+
"step": 55296
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.87,
|
| 390 |
+
"learning_rate": 0.0002808921064161573,
|
| 391 |
+
"loss": 0.3917,
|
| 392 |
+
"step": 57344
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.87,
|
| 396 |
+
"eval_loss": 0.7029281854629517,
|
| 397 |
+
"eval_runtime": 17.8146,
|
| 398 |
+
"eval_samples_per_second": 117.432,
|
| 399 |
+
"eval_steps_per_second": 7.354,
|
| 400 |
+
"step": 57344
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.01,
|
| 404 |
+
"learning_rate": 0.00026680720064442787,
|
| 405 |
+
"loss": 0.3876,
|
| 406 |
+
"step": 59392
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.01,
|
| 410 |
+
"eval_loss": 0.6837398409843445,
|
| 411 |
+
"eval_runtime": 17.4354,
|
| 412 |
+
"eval_samples_per_second": 119.986,
|
| 413 |
+
"eval_steps_per_second": 7.513,
|
| 414 |
+
"step": 59392
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.14,
|
| 418 |
+
"learning_rate": 0.00025266845586830784,
|
| 419 |
+
"loss": 0.3751,
|
| 420 |
+
"step": 61440
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.14,
|
| 424 |
+
"eval_loss": 0.690500795841217,
|
| 425 |
+
"eval_runtime": 17.8964,
|
| 426 |
+
"eval_samples_per_second": 116.895,
|
| 427 |
+
"eval_steps_per_second": 7.32,
|
| 428 |
+
"step": 61440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.28,
|
| 432 |
+
"learning_rate": 0.00023851425721450398,
|
| 433 |
+
"loss": 0.3719,
|
| 434 |
+
"step": 63488
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.28,
|
| 438 |
+
"eval_loss": 0.6938452124595642,
|
| 439 |
+
"eval_runtime": 15.7962,
|
| 440 |
+
"eval_samples_per_second": 132.437,
|
| 441 |
+
"eval_steps_per_second": 8.293,
|
| 442 |
+
"step": 63488
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.42,
|
| 446 |
+
"learning_rate": 0.0002244106409269819,
|
| 447 |
+
"loss": 0.3644,
|
| 448 |
+
"step": 65536
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.42,
|
| 452 |
+
"eval_loss": 0.6987010836601257,
|
| 453 |
+
"eval_runtime": 15.9556,
|
| 454 |
+
"eval_samples_per_second": 131.114,
|
| 455 |
+
"eval_steps_per_second": 8.21,
|
| 456 |
+
"step": 65536
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.56,
|
| 460 |
+
"learning_rate": 0.00021037526400320187,
|
| 461 |
+
"loss": 0.357,
|
| 462 |
+
"step": 67584
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.56,
|
| 466 |
+
"eval_loss": 0.6973423957824707,
|
| 467 |
+
"eval_runtime": 15.8273,
|
| 468 |
+
"eval_samples_per_second": 132.177,
|
| 469 |
+
"eval_steps_per_second": 8.277,
|
| 470 |
+
"step": 67584
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7,
|
| 474 |
+
"learning_rate": 0.0001964736950807942,
|
| 475 |
+
"loss": 0.357,
|
| 476 |
+
"step": 69632
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.7,
|
| 480 |
+
"eval_loss": 0.6949540972709656,
|
| 481 |
+
"eval_runtime": 16.1065,
|
| 482 |
+
"eval_samples_per_second": 129.885,
|
| 483 |
+
"eval_steps_per_second": 8.133,
|
| 484 |
+
"step": 69632
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.83,
|
| 488 |
+
"learning_rate": 0.00018274358855873096,
|
| 489 |
+
"loss": 0.3502,
|
| 490 |
+
"step": 71680
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.83,
|
| 494 |
+
"eval_loss": 0.6959769129753113,
|
| 495 |
+
"eval_runtime": 16.2001,
|
| 496 |
+
"eval_samples_per_second": 129.135,
|
| 497 |
+
"eval_steps_per_second": 8.086,
|
| 498 |
+
"step": 71680
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.97,
|
| 502 |
+
"learning_rate": 0.00016922892649452222,
|
| 503 |
+
"loss": 0.3458,
|
| 504 |
+
"step": 73728
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.97,
|
| 508 |
+
"eval_loss": 0.7017838358879089,
|
| 509 |
+
"eval_runtime": 18.0467,
|
| 510 |
+
"eval_samples_per_second": 115.922,
|
| 511 |
+
"eval_steps_per_second": 7.259,
|
| 512 |
+
"step": 73728
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.11,
|
| 516 |
+
"learning_rate": 0.00015597300080605504,
|
| 517 |
+
"loss": 0.3302,
|
| 518 |
+
"step": 75776
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.11,
|
| 522 |
+
"eval_loss": 0.7075567841529846,
|
| 523 |
+
"eval_runtime": 16.7136,
|
| 524 |
+
"eval_samples_per_second": 125.168,
|
| 525 |
+
"eval_steps_per_second": 7.838,
|
| 526 |
+
"step": 75776
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.25,
|
| 530 |
+
"learning_rate": 0.0001430182745933093,
|
| 531 |
+
"loss": 0.3278,
|
| 532 |
+
"step": 77824
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.25,
|
| 536 |
+
"eval_loss": 0.7178707718849182,
|
| 537 |
+
"eval_runtime": 15.7714,
|
| 538 |
+
"eval_samples_per_second": 132.645,
|
| 539 |
+
"eval_steps_per_second": 8.306,
|
| 540 |
+
"step": 77824
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.39,
|
| 544 |
+
"learning_rate": 0.00013040017526934073,
|
| 545 |
+
"loss": 0.3224,
|
| 546 |
+
"step": 79872
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 5.39,
|
| 550 |
+
"eval_loss": 0.725006639957428,
|
| 551 |
+
"eval_runtime": 15.8618,
|
| 552 |
+
"eval_samples_per_second": 131.889,
|
| 553 |
+
"eval_steps_per_second": 8.259,
|
| 554 |
+
"step": 79872
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.53,
|
| 558 |
+
"learning_rate": 0.00011817144183980649,
|
| 559 |
+
"loss": 0.3181,
|
| 560 |
+
"step": 81920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.53,
|
| 564 |
+
"eval_loss": 0.7247176766395569,
|
| 565 |
+
"eval_runtime": 17.8123,
|
| 566 |
+
"eval_samples_per_second": 117.447,
|
| 567 |
+
"eval_steps_per_second": 7.354,
|
| 568 |
+
"step": 81920
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 5.66,
|
| 572 |
+
"learning_rate": 0.00010636499874117036,
|
| 573 |
+
"loss": 0.3154,
|
| 574 |
+
"step": 83968
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 5.66,
|
| 578 |
+
"eval_loss": 0.7248900532722473,
|
| 579 |
+
"eval_runtime": 17.751,
|
| 580 |
+
"eval_samples_per_second": 117.852,
|
| 581 |
+
"eval_steps_per_second": 7.38,
|
| 582 |
+
"step": 83968
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 5.8,
|
| 586 |
+
"learning_rate": 9.501866590283475e-05,
|
| 587 |
+
"loss": 0.3105,
|
| 588 |
+
"step": 86016
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.8,
|
| 592 |
+
"eval_loss": 0.7352888584136963,
|
| 593 |
+
"eval_runtime": 18.0338,
|
| 594 |
+
"eval_samples_per_second": 116.004,
|
| 595 |
+
"eval_steps_per_second": 7.264,
|
| 596 |
+
"step": 86016
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.94,
|
| 600 |
+
"learning_rate": 8.416361604489855e-05,
|
| 601 |
+
"loss": 0.3035,
|
| 602 |
+
"step": 88064
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.94,
|
| 606 |
+
"eval_loss": 0.7256708741188049,
|
| 607 |
+
"eval_runtime": 15.94,
|
| 608 |
+
"eval_samples_per_second": 131.242,
|
| 609 |
+
"eval_steps_per_second": 8.218,
|
| 610 |
+
"step": 88064
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 6.08,
|
| 614 |
+
"learning_rate": 7.384521927589935e-05,
|
| 615 |
+
"loss": 0.294,
|
| 616 |
+
"step": 90112
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.08,
|
| 620 |
+
"eval_loss": 0.7387065291404724,
|
| 621 |
+
"eval_runtime": 16.8562,
|
| 622 |
+
"eval_samples_per_second": 124.109,
|
| 623 |
+
"eval_steps_per_second": 7.772,
|
| 624 |
+
"step": 90112
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.22,
|
| 628 |
+
"learning_rate": 6.409110434142392e-05,
|
| 629 |
+
"loss": 0.2866,
|
| 630 |
+
"step": 92160
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.22,
|
| 634 |
+
"eval_loss": 0.7492235898971558,
|
| 635 |
+
"eval_runtime": 17.2872,
|
| 636 |
+
"eval_samples_per_second": 121.015,
|
| 637 |
+
"eval_steps_per_second": 7.578,
|
| 638 |
+
"step": 92160
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 6.35,
|
| 642 |
+
"learning_rate": 5.492819313147518e-05,
|
| 643 |
+
"loss": 0.2836,
|
| 644 |
+
"step": 94208
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 6.35,
|
| 648 |
+
"eval_loss": 0.7440558671951294,
|
| 649 |
+
"eval_runtime": 17.5257,
|
| 650 |
+
"eval_samples_per_second": 119.367,
|
| 651 |
+
"eval_steps_per_second": 7.475,
|
| 652 |
+
"step": 94208
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 6.49,
|
| 656 |
+
"learning_rate": 4.6394783238561305e-05,
|
| 657 |
+
"loss": 0.2811,
|
| 658 |
+
"step": 96256
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 6.49,
|
| 662 |
+
"eval_loss": 0.7577848434448242,
|
| 663 |
+
"eval_runtime": 15.8536,
|
| 664 |
+
"eval_samples_per_second": 131.957,
|
| 665 |
+
"eval_steps_per_second": 8.263,
|
| 666 |
+
"step": 96256
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 6.63,
|
| 670 |
+
"learning_rate": 3.851358797621554e-05,
|
| 671 |
+
"loss": 0.2774,
|
| 672 |
+
"step": 98304
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.63,
|
| 676 |
+
"eval_loss": 0.7483424544334412,
|
| 677 |
+
"eval_runtime": 15.8485,
|
| 678 |
+
"eval_samples_per_second": 132.0,
|
| 679 |
+
"eval_steps_per_second": 8.266,
|
| 680 |
+
"step": 98304
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 6.77,
|
| 684 |
+
"learning_rate": 3.130985341100834e-05,
|
| 685 |
+
"loss": 0.2736,
|
| 686 |
+
"step": 100352
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 6.77,
|
| 690 |
+
"eval_loss": 0.7481484413146973,
|
| 691 |
+
"eval_runtime": 17.9168,
|
| 692 |
+
"eval_samples_per_second": 116.762,
|
| 693 |
+
"eval_steps_per_second": 7.312,
|
| 694 |
+
"step": 100352
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 6.91,
|
| 698 |
+
"learning_rate": 2.4803653223119228e-05,
|
| 699 |
+
"loss": 0.27,
|
| 700 |
+
"step": 102400
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 6.91,
|
| 704 |
+
"eval_loss": 0.7659121751785278,
|
| 705 |
+
"eval_runtime": 15.8798,
|
| 706 |
+
"eval_samples_per_second": 131.739,
|
| 707 |
+
"eval_steps_per_second": 8.249,
|
| 708 |
+
"step": 102400
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 7.04,
|
| 712 |
+
"learning_rate": 1.902218093192909e-05,
|
| 713 |
+
"loss": 0.2636,
|
| 714 |
+
"step": 104448
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 7.04,
|
| 718 |
+
"eval_loss": 0.7734333276748657,
|
| 719 |
+
"eval_runtime": 17.8497,
|
| 720 |
+
"eval_samples_per_second": 117.201,
|
| 721 |
+
"eval_steps_per_second": 7.339,
|
| 722 |
+
"step": 104448
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 7.18,
|
| 726 |
+
"learning_rate": 1.398060674025281e-05,
|
| 727 |
+
"loss": 0.2583,
|
| 728 |
+
"step": 106496
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 7.18,
|
| 732 |
+
"eval_loss": 0.7774596810340881,
|
| 733 |
+
"eval_runtime": 17.9006,
|
| 734 |
+
"eval_samples_per_second": 116.868,
|
| 735 |
+
"eval_steps_per_second": 7.318,
|
| 736 |
+
"step": 106496
|
| 737 |
+
}
|
| 738 |
+
],
|
| 739 |
+
"max_steps": 118608,
|
| 740 |
+
"num_train_epochs": 8,
|
| 741 |
+
"total_flos": 4.366669386207068e+17,
|
| 742 |
+
"trial_name": null,
|
| 743 |
+
"trial_params": null
|
| 744 |
+
}
|
checkpoint-106496/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0200139492e444a9d322a4f90a96e6dde09c7a882f05b816c2345dade5ea0f98
|
| 3 |
+
size 3515
|
checkpoint-114688/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
checkpoint-114688/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b68c6edb60414cfe637e2ec074a5f4acec75f5337e31730978b158998228eac5
|
| 3 |
+
size 211432837
|
checkpoint-114688/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6849dc98fcc9e4181c102fa0bb6dd1d70c54edae9d1a8440b604c6f98ba117b4
|
| 3 |
+
size 139279005
|
checkpoint-114688/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc866ce242256f98027b6016930d35e9d45293516285c23fdbadea7b58591d8a
|
| 3 |
+
size 15597
|
checkpoint-114688/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9778e4932801d6fcefc7836eddd9fd5635b262a07b6eb7a75700d5c35643bb19
|
| 3 |
+
size 557
|
checkpoint-114688/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fa9e591dbfa043fe5b9503f81bacab889f557e7055edb15e3594fae6f28701b
|
| 3 |
+
size 627
|
checkpoint-114688/trainer_state.json
ADDED
|
@@ -0,0 +1,800 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.735599622285175,
|
| 5 |
+
"global_step": 114688,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 0.00020480000000000002,
|
| 13 |
+
"loss": 2.1719,
|
| 14 |
+
"step": 2048
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.0833024978637695,
|
| 19 |
+
"eval_runtime": 15.7887,
|
| 20 |
+
"eval_samples_per_second": 132.5,
|
| 21 |
+
"eval_steps_per_second": 8.297,
|
| 22 |
+
"step": 2048
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 0.00040960000000000004,
|
| 27 |
+
"loss": 0.9468,
|
| 28 |
+
"step": 4096
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 0.8000818490982056,
|
| 33 |
+
"eval_runtime": 17.2632,
|
| 34 |
+
"eval_samples_per_second": 121.183,
|
| 35 |
+
"eval_steps_per_second": 7.588,
|
| 36 |
+
"step": 4096
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.41,
|
| 40 |
+
"learning_rate": 0.0004998749142723946,
|
| 41 |
+
"loss": 0.7542,
|
| 42 |
+
"step": 6144
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41,
|
| 46 |
+
"eval_loss": 0.7439925074577332,
|
| 47 |
+
"eval_runtime": 17.4134,
|
| 48 |
+
"eval_samples_per_second": 120.138,
|
| 49 |
+
"eval_steps_per_second": 7.523,
|
| 50 |
+
"step": 6144
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.55,
|
| 54 |
+
"learning_rate": 0.0004990273340312486,
|
| 55 |
+
"loss": 0.6756,
|
| 56 |
+
"step": 8192
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"eval_loss": 0.7018134593963623,
|
| 61 |
+
"eval_runtime": 15.8601,
|
| 62 |
+
"eval_samples_per_second": 131.904,
|
| 63 |
+
"eval_steps_per_second": 8.26,
|
| 64 |
+
"step": 8192
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.69,
|
| 68 |
+
"learning_rate": 0.0004973820371438889,
|
| 69 |
+
"loss": 0.631,
|
| 70 |
+
"step": 10240
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.69,
|
| 74 |
+
"eval_loss": 0.684688925743103,
|
| 75 |
+
"eval_runtime": 18.0314,
|
| 76 |
+
"eval_samples_per_second": 116.02,
|
| 77 |
+
"eval_steps_per_second": 7.265,
|
| 78 |
+
"step": 10240
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.83,
|
| 82 |
+
"learning_rate": 0.0004949442940386407,
|
| 83 |
+
"loss": 0.5989,
|
| 84 |
+
"step": 12288
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.83,
|
| 88 |
+
"eval_loss": 0.6822534203529358,
|
| 89 |
+
"eval_runtime": 17.4182,
|
| 90 |
+
"eval_samples_per_second": 120.104,
|
| 91 |
+
"eval_steps_per_second": 7.521,
|
| 92 |
+
"step": 12288
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.97,
|
| 96 |
+
"learning_rate": 0.0004917201492451735,
|
| 97 |
+
"loss": 0.5776,
|
| 98 |
+
"step": 14336
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"eval_loss": 0.6674544215202332,
|
| 103 |
+
"eval_runtime": 17.7487,
|
| 104 |
+
"eval_samples_per_second": 117.868,
|
| 105 |
+
"eval_steps_per_second": 7.381,
|
| 106 |
+
"step": 14336
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.11,
|
| 110 |
+
"learning_rate": 0.0004877230785006913,
|
| 111 |
+
"loss": 0.5525,
|
| 112 |
+
"step": 16384
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.11,
|
| 116 |
+
"eval_loss": 0.6640163660049438,
|
| 117 |
+
"eval_runtime": 15.8951,
|
| 118 |
+
"eval_samples_per_second": 131.613,
|
| 119 |
+
"eval_steps_per_second": 8.242,
|
| 120 |
+
"step": 16384
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 0.0004829670105577666,
|
| 125 |
+
"loss": 0.5373,
|
| 126 |
+
"step": 18432
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.24,
|
| 130 |
+
"eval_loss": 0.6663170456886292,
|
| 131 |
+
"eval_runtime": 17.5,
|
| 132 |
+
"eval_samples_per_second": 119.543,
|
| 133 |
+
"eval_steps_per_second": 7.486,
|
| 134 |
+
"step": 18432
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.38,
|
| 138 |
+
"learning_rate": 0.0004774596641323791,
|
| 139 |
+
"loss": 0.5261,
|
| 140 |
+
"step": 20480
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"eval_loss": 0.6518951654434204,
|
| 145 |
+
"eval_runtime": 17.1326,
|
| 146 |
+
"eval_samples_per_second": 122.107,
|
| 147 |
+
"eval_steps_per_second": 7.646,
|
| 148 |
+
"step": 20480
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.52,
|
| 152 |
+
"learning_rate": 0.0004712261976082475,
|
| 153 |
+
"loss": 0.5115,
|
| 154 |
+
"step": 22528
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.52,
|
| 158 |
+
"eval_loss": 0.6658991575241089,
|
| 159 |
+
"eval_runtime": 17.6874,
|
| 160 |
+
"eval_samples_per_second": 118.277,
|
| 161 |
+
"eval_steps_per_second": 7.406,
|
| 162 |
+
"step": 22528
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.66,
|
| 166 |
+
"learning_rate": 0.00046428407064289515,
|
| 167 |
+
"loss": 0.5033,
|
| 168 |
+
"step": 24576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.66,
|
| 172 |
+
"eval_loss": 0.6602604389190674,
|
| 173 |
+
"eval_runtime": 15.8845,
|
| 174 |
+
"eval_samples_per_second": 131.701,
|
| 175 |
+
"eval_steps_per_second": 8.247,
|
| 176 |
+
"step": 24576
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8,
|
| 180 |
+
"learning_rate": 0.00045665163060732317,
|
| 181 |
+
"loss": 0.4925,
|
| 182 |
+
"step": 26624
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.8,
|
| 186 |
+
"eval_loss": 0.6551440954208374,
|
| 187 |
+
"eval_runtime": 17.7902,
|
| 188 |
+
"eval_samples_per_second": 117.593,
|
| 189 |
+
"eval_steps_per_second": 7.364,
|
| 190 |
+
"step": 26624
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.93,
|
| 194 |
+
"learning_rate": 0.000448360778288479,
|
| 195 |
+
"loss": 0.4868,
|
| 196 |
+
"step": 28672
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.93,
|
| 200 |
+
"eval_loss": 0.6580803394317627,
|
| 201 |
+
"eval_runtime": 15.7841,
|
| 202 |
+
"eval_samples_per_second": 132.538,
|
| 203 |
+
"eval_steps_per_second": 8.299,
|
| 204 |
+
"step": 28672
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.00043942999964293453,
|
| 209 |
+
"loss": 0.4741,
|
| 210 |
+
"step": 30720
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.07,
|
| 214 |
+
"eval_loss": 0.661230742931366,
|
| 215 |
+
"eval_runtime": 18.015,
|
| 216 |
+
"eval_samples_per_second": 116.126,
|
| 217 |
+
"eval_steps_per_second": 7.272,
|
| 218 |
+
"step": 30720
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.21,
|
| 222 |
+
"learning_rate": 0.0004298966220346151,
|
| 223 |
+
"loss": 0.4632,
|
| 224 |
+
"step": 32768
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.21,
|
| 228 |
+
"eval_loss": 0.6618030071258545,
|
| 229 |
+
"eval_runtime": 17.9073,
|
| 230 |
+
"eval_samples_per_second": 116.824,
|
| 231 |
+
"eval_steps_per_second": 7.315,
|
| 232 |
+
"step": 32768
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.35,
|
| 236 |
+
"learning_rate": 0.00041978697624050446,
|
| 237 |
+
"loss": 0.4584,
|
| 238 |
+
"step": 34816
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.35,
|
| 242 |
+
"eval_loss": 0.6579039096832275,
|
| 243 |
+
"eval_runtime": 17.8241,
|
| 244 |
+
"eval_samples_per_second": 117.369,
|
| 245 |
+
"eval_steps_per_second": 7.35,
|
| 246 |
+
"step": 34816
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.49,
|
| 250 |
+
"learning_rate": 0.0004091334467888659,
|
| 251 |
+
"loss": 0.451,
|
| 252 |
+
"step": 36864
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"eval_loss": 0.6627541780471802,
|
| 257 |
+
"eval_runtime": 17.6822,
|
| 258 |
+
"eval_samples_per_second": 118.311,
|
| 259 |
+
"eval_steps_per_second": 7.409,
|
| 260 |
+
"step": 36864
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.62,
|
| 264 |
+
"learning_rate": 0.00039796458815002033,
|
| 265 |
+
"loss": 0.445,
|
| 266 |
+
"step": 38912
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.62,
|
| 270 |
+
"eval_loss": 0.656867265701294,
|
| 271 |
+
"eval_runtime": 15.8659,
|
| 272 |
+
"eval_samples_per_second": 131.855,
|
| 273 |
+
"eval_steps_per_second": 8.257,
|
| 274 |
+
"step": 38912
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.76,
|
| 278 |
+
"learning_rate": 0.0003863212870708643,
|
| 279 |
+
"loss": 0.4404,
|
| 280 |
+
"step": 40960
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.76,
|
| 284 |
+
"eval_loss": 0.6653844118118286,
|
| 285 |
+
"eval_runtime": 17.9192,
|
| 286 |
+
"eval_samples_per_second": 116.747,
|
| 287 |
+
"eval_steps_per_second": 7.311,
|
| 288 |
+
"step": 40960
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.9,
|
| 292 |
+
"learning_rate": 0.00037424687637876156,
|
| 293 |
+
"loss": 0.4334,
|
| 294 |
+
"step": 43008
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.9,
|
| 298 |
+
"eval_loss": 0.6553301811218262,
|
| 299 |
+
"eval_runtime": 15.9304,
|
| 300 |
+
"eval_samples_per_second": 131.321,
|
| 301 |
+
"eval_steps_per_second": 8.223,
|
| 302 |
+
"step": 43008
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 0.00036178064571954134,
|
| 307 |
+
"loss": 0.4278,
|
| 308 |
+
"step": 45056
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.04,
|
| 312 |
+
"eval_loss": 0.6766741871833801,
|
| 313 |
+
"eval_runtime": 16.9834,
|
| 314 |
+
"eval_samples_per_second": 123.179,
|
| 315 |
+
"eval_steps_per_second": 7.713,
|
| 316 |
+
"step": 45056
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.18,
|
| 320 |
+
"learning_rate": 0.0003489439971353363,
|
| 321 |
+
"loss": 0.4152,
|
| 322 |
+
"step": 47104
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.18,
|
| 326 |
+
"eval_loss": 0.6645193696022034,
|
| 327 |
+
"eval_runtime": 15.816,
|
| 328 |
+
"eval_samples_per_second": 132.271,
|
| 329 |
+
"eval_steps_per_second": 8.283,
|
| 330 |
+
"step": 47104
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.32,
|
| 334 |
+
"learning_rate": 0.0003357965820476752,
|
| 335 |
+
"loss": 0.411,
|
| 336 |
+
"step": 49152
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.32,
|
| 340 |
+
"eval_loss": 0.6748972535133362,
|
| 341 |
+
"eval_runtime": 17.7507,
|
| 342 |
+
"eval_samples_per_second": 117.855,
|
| 343 |
+
"eval_steps_per_second": 7.38,
|
| 344 |
+
"step": 49152
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.45,
|
| 348 |
+
"learning_rate": 0.0003223743322236833,
|
| 349 |
+
"loss": 0.4085,
|
| 350 |
+
"step": 51200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.45,
|
| 354 |
+
"eval_loss": 0.6703893542289734,
|
| 355 |
+
"eval_runtime": 17.341,
|
| 356 |
+
"eval_samples_per_second": 120.639,
|
| 357 |
+
"eval_steps_per_second": 7.554,
|
| 358 |
+
"step": 51200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.59,
|
| 362 |
+
"learning_rate": 0.0003087269633577651,
|
| 363 |
+
"loss": 0.4037,
|
| 364 |
+
"step": 53248
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.59,
|
| 368 |
+
"eval_loss": 0.6561577916145325,
|
| 369 |
+
"eval_runtime": 15.9295,
|
| 370 |
+
"eval_samples_per_second": 131.329,
|
| 371 |
+
"eval_steps_per_second": 8.224,
|
| 372 |
+
"step": 53248
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.73,
|
| 376 |
+
"learning_rate": 0.0002948780545870247,
|
| 377 |
+
"loss": 0.3975,
|
| 378 |
+
"step": 55296
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.73,
|
| 382 |
+
"eval_loss": 0.6698916554450989,
|
| 383 |
+
"eval_runtime": 17.9008,
|
| 384 |
+
"eval_samples_per_second": 116.866,
|
| 385 |
+
"eval_steps_per_second": 7.318,
|
| 386 |
+
"step": 55296
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.87,
|
| 390 |
+
"learning_rate": 0.0002808921064161573,
|
| 391 |
+
"loss": 0.3917,
|
| 392 |
+
"step": 57344
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.87,
|
| 396 |
+
"eval_loss": 0.7029281854629517,
|
| 397 |
+
"eval_runtime": 17.8146,
|
| 398 |
+
"eval_samples_per_second": 117.432,
|
| 399 |
+
"eval_steps_per_second": 7.354,
|
| 400 |
+
"step": 57344
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.01,
|
| 404 |
+
"learning_rate": 0.00026680720064442787,
|
| 405 |
+
"loss": 0.3876,
|
| 406 |
+
"step": 59392
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.01,
|
| 410 |
+
"eval_loss": 0.6837398409843445,
|
| 411 |
+
"eval_runtime": 17.4354,
|
| 412 |
+
"eval_samples_per_second": 119.986,
|
| 413 |
+
"eval_steps_per_second": 7.513,
|
| 414 |
+
"step": 59392
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.14,
|
| 418 |
+
"learning_rate": 0.00025266845586830784,
|
| 419 |
+
"loss": 0.3751,
|
| 420 |
+
"step": 61440
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.14,
|
| 424 |
+
"eval_loss": 0.690500795841217,
|
| 425 |
+
"eval_runtime": 17.8964,
|
| 426 |
+
"eval_samples_per_second": 116.895,
|
| 427 |
+
"eval_steps_per_second": 7.32,
|
| 428 |
+
"step": 61440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.28,
|
| 432 |
+
"learning_rate": 0.00023851425721450398,
|
| 433 |
+
"loss": 0.3719,
|
| 434 |
+
"step": 63488
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.28,
|
| 438 |
+
"eval_loss": 0.6938452124595642,
|
| 439 |
+
"eval_runtime": 15.7962,
|
| 440 |
+
"eval_samples_per_second": 132.437,
|
| 441 |
+
"eval_steps_per_second": 8.293,
|
| 442 |
+
"step": 63488
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.42,
|
| 446 |
+
"learning_rate": 0.0002244106409269819,
|
| 447 |
+
"loss": 0.3644,
|
| 448 |
+
"step": 65536
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.42,
|
| 452 |
+
"eval_loss": 0.6987010836601257,
|
| 453 |
+
"eval_runtime": 15.9556,
|
| 454 |
+
"eval_samples_per_second": 131.114,
|
| 455 |
+
"eval_steps_per_second": 8.21,
|
| 456 |
+
"step": 65536
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.56,
|
| 460 |
+
"learning_rate": 0.00021037526400320187,
|
| 461 |
+
"loss": 0.357,
|
| 462 |
+
"step": 67584
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.56,
|
| 466 |
+
"eval_loss": 0.6973423957824707,
|
| 467 |
+
"eval_runtime": 15.8273,
|
| 468 |
+
"eval_samples_per_second": 132.177,
|
| 469 |
+
"eval_steps_per_second": 8.277,
|
| 470 |
+
"step": 67584
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7,
|
| 474 |
+
"learning_rate": 0.0001964736950807942,
|
| 475 |
+
"loss": 0.357,
|
| 476 |
+
"step": 69632
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.7,
|
| 480 |
+
"eval_loss": 0.6949540972709656,
|
| 481 |
+
"eval_runtime": 16.1065,
|
| 482 |
+
"eval_samples_per_second": 129.885,
|
| 483 |
+
"eval_steps_per_second": 8.133,
|
| 484 |
+
"step": 69632
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.83,
|
| 488 |
+
"learning_rate": 0.00018274358855873096,
|
| 489 |
+
"loss": 0.3502,
|
| 490 |
+
"step": 71680
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.83,
|
| 494 |
+
"eval_loss": 0.6959769129753113,
|
| 495 |
+
"eval_runtime": 16.2001,
|
| 496 |
+
"eval_samples_per_second": 129.135,
|
| 497 |
+
"eval_steps_per_second": 8.086,
|
| 498 |
+
"step": 71680
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.97,
|
| 502 |
+
"learning_rate": 0.00016922892649452222,
|
| 503 |
+
"loss": 0.3458,
|
| 504 |
+
"step": 73728
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.97,
|
| 508 |
+
"eval_loss": 0.7017838358879089,
|
| 509 |
+
"eval_runtime": 18.0467,
|
| 510 |
+
"eval_samples_per_second": 115.922,
|
| 511 |
+
"eval_steps_per_second": 7.259,
|
| 512 |
+
"step": 73728
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.11,
|
| 516 |
+
"learning_rate": 0.00015597300080605504,
|
| 517 |
+
"loss": 0.3302,
|
| 518 |
+
"step": 75776
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.11,
|
| 522 |
+
"eval_loss": 0.7075567841529846,
|
| 523 |
+
"eval_runtime": 16.7136,
|
| 524 |
+
"eval_samples_per_second": 125.168,
|
| 525 |
+
"eval_steps_per_second": 7.838,
|
| 526 |
+
"step": 75776
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.25,
|
| 530 |
+
"learning_rate": 0.0001430182745933093,
|
| 531 |
+
"loss": 0.3278,
|
| 532 |
+
"step": 77824
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.25,
|
| 536 |
+
"eval_loss": 0.7178707718849182,
|
| 537 |
+
"eval_runtime": 15.7714,
|
| 538 |
+
"eval_samples_per_second": 132.645,
|
| 539 |
+
"eval_steps_per_second": 8.306,
|
| 540 |
+
"step": 77824
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.39,
|
| 544 |
+
"learning_rate": 0.00013040017526934073,
|
| 545 |
+
"loss": 0.3224,
|
| 546 |
+
"step": 79872
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 5.39,
|
| 550 |
+
"eval_loss": 0.725006639957428,
|
| 551 |
+
"eval_runtime": 15.8618,
|
| 552 |
+
"eval_samples_per_second": 131.889,
|
| 553 |
+
"eval_steps_per_second": 8.259,
|
| 554 |
+
"step": 79872
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.53,
|
| 558 |
+
"learning_rate": 0.00011817144183980649,
|
| 559 |
+
"loss": 0.3181,
|
| 560 |
+
"step": 81920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.53,
|
| 564 |
+
"eval_loss": 0.7247176766395569,
|
| 565 |
+
"eval_runtime": 17.8123,
|
| 566 |
+
"eval_samples_per_second": 117.447,
|
| 567 |
+
"eval_steps_per_second": 7.354,
|
| 568 |
+
"step": 81920
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 5.66,
|
| 572 |
+
"learning_rate": 0.00010636499874117036,
|
| 573 |
+
"loss": 0.3154,
|
| 574 |
+
"step": 83968
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 5.66,
|
| 578 |
+
"eval_loss": 0.7248900532722473,
|
| 579 |
+
"eval_runtime": 17.751,
|
| 580 |
+
"eval_samples_per_second": 117.852,
|
| 581 |
+
"eval_steps_per_second": 7.38,
|
| 582 |
+
"step": 83968
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 5.8,
|
| 586 |
+
"learning_rate": 9.501866590283475e-05,
|
| 587 |
+
"loss": 0.3105,
|
| 588 |
+
"step": 86016
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.8,
|
| 592 |
+
"eval_loss": 0.7352888584136963,
|
| 593 |
+
"eval_runtime": 18.0338,
|
| 594 |
+
"eval_samples_per_second": 116.004,
|
| 595 |
+
"eval_steps_per_second": 7.264,
|
| 596 |
+
"step": 86016
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.94,
|
| 600 |
+
"learning_rate": 8.416361604489855e-05,
|
| 601 |
+
"loss": 0.3035,
|
| 602 |
+
"step": 88064
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.94,
|
| 606 |
+
"eval_loss": 0.7256708741188049,
|
| 607 |
+
"eval_runtime": 15.94,
|
| 608 |
+
"eval_samples_per_second": 131.242,
|
| 609 |
+
"eval_steps_per_second": 8.218,
|
| 610 |
+
"step": 88064
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 6.08,
|
| 614 |
+
"learning_rate": 7.384521927589935e-05,
|
| 615 |
+
"loss": 0.294,
|
| 616 |
+
"step": 90112
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.08,
|
| 620 |
+
"eval_loss": 0.7387065291404724,
|
| 621 |
+
"eval_runtime": 16.8562,
|
| 622 |
+
"eval_samples_per_second": 124.109,
|
| 623 |
+
"eval_steps_per_second": 7.772,
|
| 624 |
+
"step": 90112
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.22,
|
| 628 |
+
"learning_rate": 6.409110434142392e-05,
|
| 629 |
+
"loss": 0.2866,
|
| 630 |
+
"step": 92160
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.22,
|
| 634 |
+
"eval_loss": 0.7492235898971558,
|
| 635 |
+
"eval_runtime": 17.2872,
|
| 636 |
+
"eval_samples_per_second": 121.015,
|
| 637 |
+
"eval_steps_per_second": 7.578,
|
| 638 |
+
"step": 92160
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 6.35,
|
| 642 |
+
"learning_rate": 5.492819313147518e-05,
|
| 643 |
+
"loss": 0.2836,
|
| 644 |
+
"step": 94208
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 6.35,
|
| 648 |
+
"eval_loss": 0.7440558671951294,
|
| 649 |
+
"eval_runtime": 17.5257,
|
| 650 |
+
"eval_samples_per_second": 119.367,
|
| 651 |
+
"eval_steps_per_second": 7.475,
|
| 652 |
+
"step": 94208
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 6.49,
|
| 656 |
+
"learning_rate": 4.6394783238561305e-05,
|
| 657 |
+
"loss": 0.2811,
|
| 658 |
+
"step": 96256
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 6.49,
|
| 662 |
+
"eval_loss": 0.7577848434448242,
|
| 663 |
+
"eval_runtime": 15.8536,
|
| 664 |
+
"eval_samples_per_second": 131.957,
|
| 665 |
+
"eval_steps_per_second": 8.263,
|
| 666 |
+
"step": 96256
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 6.63,
|
| 670 |
+
"learning_rate": 3.851358797621554e-05,
|
| 671 |
+
"loss": 0.2774,
|
| 672 |
+
"step": 98304
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.63,
|
| 676 |
+
"eval_loss": 0.7483424544334412,
|
| 677 |
+
"eval_runtime": 15.8485,
|
| 678 |
+
"eval_samples_per_second": 132.0,
|
| 679 |
+
"eval_steps_per_second": 8.266,
|
| 680 |
+
"step": 98304
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 6.77,
|
| 684 |
+
"learning_rate": 3.130985341100834e-05,
|
| 685 |
+
"loss": 0.2736,
|
| 686 |
+
"step": 100352
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 6.77,
|
| 690 |
+
"eval_loss": 0.7481484413146973,
|
| 691 |
+
"eval_runtime": 17.9168,
|
| 692 |
+
"eval_samples_per_second": 116.762,
|
| 693 |
+
"eval_steps_per_second": 7.312,
|
| 694 |
+
"step": 100352
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 6.91,
|
| 698 |
+
"learning_rate": 2.4803653223119228e-05,
|
| 699 |
+
"loss": 0.27,
|
| 700 |
+
"step": 102400
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 6.91,
|
| 704 |
+
"eval_loss": 0.7659121751785278,
|
| 705 |
+
"eval_runtime": 15.8798,
|
| 706 |
+
"eval_samples_per_second": 131.739,
|
| 707 |
+
"eval_steps_per_second": 8.249,
|
| 708 |
+
"step": 102400
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 7.04,
|
| 712 |
+
"learning_rate": 1.902218093192909e-05,
|
| 713 |
+
"loss": 0.2636,
|
| 714 |
+
"step": 104448
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 7.04,
|
| 718 |
+
"eval_loss": 0.7734333276748657,
|
| 719 |
+
"eval_runtime": 17.8497,
|
| 720 |
+
"eval_samples_per_second": 117.201,
|
| 721 |
+
"eval_steps_per_second": 7.339,
|
| 722 |
+
"step": 104448
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 7.18,
|
| 726 |
+
"learning_rate": 1.398060674025281e-05,
|
| 727 |
+
"loss": 0.2583,
|
| 728 |
+
"step": 106496
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 7.18,
|
| 732 |
+
"eval_loss": 0.7774596810340881,
|
| 733 |
+
"eval_runtime": 17.9006,
|
| 734 |
+
"eval_samples_per_second": 116.868,
|
| 735 |
+
"eval_steps_per_second": 7.318,
|
| 736 |
+
"step": 106496
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 7.32,
|
| 740 |
+
"learning_rate": 9.695080472251094e-06,
|
| 741 |
+
"loss": 0.2577,
|
| 742 |
+
"step": 108544
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 7.32,
|
| 746 |
+
"eval_loss": 0.7828282117843628,
|
| 747 |
+
"eval_runtime": 15.9506,
|
| 748 |
+
"eval_samples_per_second": 131.155,
|
| 749 |
+
"eval_steps_per_second": 8.213,
|
| 750 |
+
"step": 108544
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 7.46,
|
| 754 |
+
"learning_rate": 6.177802621565725e-06,
|
| 755 |
+
"loss": 0.2553,
|
| 756 |
+
"step": 110592
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 7.46,
|
| 760 |
+
"eval_loss": 0.7818995714187622,
|
| 761 |
+
"eval_runtime": 15.975,
|
| 762 |
+
"eval_samples_per_second": 130.955,
|
| 763 |
+
"eval_steps_per_second": 8.2,
|
| 764 |
+
"step": 110592
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 7.6,
|
| 768 |
+
"learning_rate": 3.4434741135661028e-06,
|
| 769 |
+
"loss": 0.2545,
|
| 770 |
+
"step": 112640
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 7.6,
|
| 774 |
+
"eval_loss": 0.7817807197570801,
|
| 775 |
+
"eval_runtime": 15.9873,
|
| 776 |
+
"eval_samples_per_second": 130.854,
|
| 777 |
+
"eval_steps_per_second": 8.194,
|
| 778 |
+
"step": 112640
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 7.74,
|
| 782 |
+
"learning_rate": 1.498947438756143e-06,
|
| 783 |
+
"loss": 0.2536,
|
| 784 |
+
"step": 114688
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 7.74,
|
| 788 |
+
"eval_loss": 0.7830276489257812,
|
| 789 |
+
"eval_runtime": 17.6639,
|
| 790 |
+
"eval_samples_per_second": 118.434,
|
| 791 |
+
"eval_steps_per_second": 7.416,
|
| 792 |
+
"step": 114688
|
| 793 |
+
}
|
| 794 |
+
],
|
| 795 |
+
"max_steps": 118608,
|
| 796 |
+
"num_train_epochs": 8,
|
| 797 |
+
"total_flos": 4.702934009537618e+17,
|
| 798 |
+
"trial_name": null,
|
| 799 |
+
"trial_params": null
|
| 800 |
+
}
|
checkpoint-114688/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0200139492e444a9d322a4f90a96e6dde09c7a882f05b816c2345dade5ea0f98
|
| 3 |
+
size 3515
|
checkpoint-81920/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
checkpoint-81920/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aea33963d1ff00b69c9b95a75bb66e6037e0f7327b98d60140a4bd4462b7b6ea
|
| 3 |
+
size 211432837
|
checkpoint-81920/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a1f35d58c7e8e917620a1f737aa8a38a873486b8360ff6a31809734fc09e9d1
|
| 3 |
+
size 139279005
|
checkpoint-81920/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b15af6bf46f127b839408de433ccac01f207f8dc309c7de790456737ff556c2
|
| 3 |
+
size 15597
|
checkpoint-81920/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83456236364b987a3fb6423d38cc8c9760be6421738dbf4def0fd71bff91206d
|
| 3 |
+
size 557
|
checkpoint-81920/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8da89aecb8a313e489684dbdef383d10e3af94945ff708bb72b8626d826e9512
|
| 3 |
+
size 627
|
checkpoint-81920/trainer_state.json
ADDED
|
@@ -0,0 +1,576 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.525428301632267,
|
| 5 |
+
"global_step": 81920,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 0.00020480000000000002,
|
| 13 |
+
"loss": 2.1719,
|
| 14 |
+
"step": 2048
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.0833024978637695,
|
| 19 |
+
"eval_runtime": 15.7887,
|
| 20 |
+
"eval_samples_per_second": 132.5,
|
| 21 |
+
"eval_steps_per_second": 8.297,
|
| 22 |
+
"step": 2048
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 0.00040960000000000004,
|
| 27 |
+
"loss": 0.9468,
|
| 28 |
+
"step": 4096
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 0.8000818490982056,
|
| 33 |
+
"eval_runtime": 17.2632,
|
| 34 |
+
"eval_samples_per_second": 121.183,
|
| 35 |
+
"eval_steps_per_second": 7.588,
|
| 36 |
+
"step": 4096
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.41,
|
| 40 |
+
"learning_rate": 0.0004998749142723946,
|
| 41 |
+
"loss": 0.7542,
|
| 42 |
+
"step": 6144
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41,
|
| 46 |
+
"eval_loss": 0.7439925074577332,
|
| 47 |
+
"eval_runtime": 17.4134,
|
| 48 |
+
"eval_samples_per_second": 120.138,
|
| 49 |
+
"eval_steps_per_second": 7.523,
|
| 50 |
+
"step": 6144
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.55,
|
| 54 |
+
"learning_rate": 0.0004990273340312486,
|
| 55 |
+
"loss": 0.6756,
|
| 56 |
+
"step": 8192
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"eval_loss": 0.7018134593963623,
|
| 61 |
+
"eval_runtime": 15.8601,
|
| 62 |
+
"eval_samples_per_second": 131.904,
|
| 63 |
+
"eval_steps_per_second": 8.26,
|
| 64 |
+
"step": 8192
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.69,
|
| 68 |
+
"learning_rate": 0.0004973820371438889,
|
| 69 |
+
"loss": 0.631,
|
| 70 |
+
"step": 10240
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.69,
|
| 74 |
+
"eval_loss": 0.684688925743103,
|
| 75 |
+
"eval_runtime": 18.0314,
|
| 76 |
+
"eval_samples_per_second": 116.02,
|
| 77 |
+
"eval_steps_per_second": 7.265,
|
| 78 |
+
"step": 10240
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.83,
|
| 82 |
+
"learning_rate": 0.0004949442940386407,
|
| 83 |
+
"loss": 0.5989,
|
| 84 |
+
"step": 12288
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.83,
|
| 88 |
+
"eval_loss": 0.6822534203529358,
|
| 89 |
+
"eval_runtime": 17.4182,
|
| 90 |
+
"eval_samples_per_second": 120.104,
|
| 91 |
+
"eval_steps_per_second": 7.521,
|
| 92 |
+
"step": 12288
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.97,
|
| 96 |
+
"learning_rate": 0.0004917201492451735,
|
| 97 |
+
"loss": 0.5776,
|
| 98 |
+
"step": 14336
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"eval_loss": 0.6674544215202332,
|
| 103 |
+
"eval_runtime": 17.7487,
|
| 104 |
+
"eval_samples_per_second": 117.868,
|
| 105 |
+
"eval_steps_per_second": 7.381,
|
| 106 |
+
"step": 14336
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.11,
|
| 110 |
+
"learning_rate": 0.0004877230785006913,
|
| 111 |
+
"loss": 0.5525,
|
| 112 |
+
"step": 16384
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.11,
|
| 116 |
+
"eval_loss": 0.6640163660049438,
|
| 117 |
+
"eval_runtime": 15.8951,
|
| 118 |
+
"eval_samples_per_second": 131.613,
|
| 119 |
+
"eval_steps_per_second": 8.242,
|
| 120 |
+
"step": 16384
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 0.0004829670105577666,
|
| 125 |
+
"loss": 0.5373,
|
| 126 |
+
"step": 18432
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.24,
|
| 130 |
+
"eval_loss": 0.6663170456886292,
|
| 131 |
+
"eval_runtime": 17.5,
|
| 132 |
+
"eval_samples_per_second": 119.543,
|
| 133 |
+
"eval_steps_per_second": 7.486,
|
| 134 |
+
"step": 18432
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.38,
|
| 138 |
+
"learning_rate": 0.0004774596641323791,
|
| 139 |
+
"loss": 0.5261,
|
| 140 |
+
"step": 20480
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"eval_loss": 0.6518951654434204,
|
| 145 |
+
"eval_runtime": 17.1326,
|
| 146 |
+
"eval_samples_per_second": 122.107,
|
| 147 |
+
"eval_steps_per_second": 7.646,
|
| 148 |
+
"step": 20480
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.52,
|
| 152 |
+
"learning_rate": 0.0004712261976082475,
|
| 153 |
+
"loss": 0.5115,
|
| 154 |
+
"step": 22528
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.52,
|
| 158 |
+
"eval_loss": 0.6658991575241089,
|
| 159 |
+
"eval_runtime": 17.6874,
|
| 160 |
+
"eval_samples_per_second": 118.277,
|
| 161 |
+
"eval_steps_per_second": 7.406,
|
| 162 |
+
"step": 22528
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.66,
|
| 166 |
+
"learning_rate": 0.00046428407064289515,
|
| 167 |
+
"loss": 0.5033,
|
| 168 |
+
"step": 24576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.66,
|
| 172 |
+
"eval_loss": 0.6602604389190674,
|
| 173 |
+
"eval_runtime": 15.8845,
|
| 174 |
+
"eval_samples_per_second": 131.701,
|
| 175 |
+
"eval_steps_per_second": 8.247,
|
| 176 |
+
"step": 24576
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8,
|
| 180 |
+
"learning_rate": 0.00045665163060732317,
|
| 181 |
+
"loss": 0.4925,
|
| 182 |
+
"step": 26624
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.8,
|
| 186 |
+
"eval_loss": 0.6551440954208374,
|
| 187 |
+
"eval_runtime": 17.7902,
|
| 188 |
+
"eval_samples_per_second": 117.593,
|
| 189 |
+
"eval_steps_per_second": 7.364,
|
| 190 |
+
"step": 26624
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.93,
|
| 194 |
+
"learning_rate": 0.000448360778288479,
|
| 195 |
+
"loss": 0.4868,
|
| 196 |
+
"step": 28672
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.93,
|
| 200 |
+
"eval_loss": 0.6580803394317627,
|
| 201 |
+
"eval_runtime": 15.7841,
|
| 202 |
+
"eval_samples_per_second": 132.538,
|
| 203 |
+
"eval_steps_per_second": 8.299,
|
| 204 |
+
"step": 28672
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.00043942999964293453,
|
| 209 |
+
"loss": 0.4741,
|
| 210 |
+
"step": 30720
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.07,
|
| 214 |
+
"eval_loss": 0.661230742931366,
|
| 215 |
+
"eval_runtime": 18.015,
|
| 216 |
+
"eval_samples_per_second": 116.126,
|
| 217 |
+
"eval_steps_per_second": 7.272,
|
| 218 |
+
"step": 30720
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.21,
|
| 222 |
+
"learning_rate": 0.0004298966220346151,
|
| 223 |
+
"loss": 0.4632,
|
| 224 |
+
"step": 32768
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.21,
|
| 228 |
+
"eval_loss": 0.6618030071258545,
|
| 229 |
+
"eval_runtime": 17.9073,
|
| 230 |
+
"eval_samples_per_second": 116.824,
|
| 231 |
+
"eval_steps_per_second": 7.315,
|
| 232 |
+
"step": 32768
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.35,
|
| 236 |
+
"learning_rate": 0.00041978697624050446,
|
| 237 |
+
"loss": 0.4584,
|
| 238 |
+
"step": 34816
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.35,
|
| 242 |
+
"eval_loss": 0.6579039096832275,
|
| 243 |
+
"eval_runtime": 17.8241,
|
| 244 |
+
"eval_samples_per_second": 117.369,
|
| 245 |
+
"eval_steps_per_second": 7.35,
|
| 246 |
+
"step": 34816
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.49,
|
| 250 |
+
"learning_rate": 0.0004091334467888659,
|
| 251 |
+
"loss": 0.451,
|
| 252 |
+
"step": 36864
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"eval_loss": 0.6627541780471802,
|
| 257 |
+
"eval_runtime": 17.6822,
|
| 258 |
+
"eval_samples_per_second": 118.311,
|
| 259 |
+
"eval_steps_per_second": 7.409,
|
| 260 |
+
"step": 36864
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.62,
|
| 264 |
+
"learning_rate": 0.00039796458815002033,
|
| 265 |
+
"loss": 0.445,
|
| 266 |
+
"step": 38912
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.62,
|
| 270 |
+
"eval_loss": 0.656867265701294,
|
| 271 |
+
"eval_runtime": 15.8659,
|
| 272 |
+
"eval_samples_per_second": 131.855,
|
| 273 |
+
"eval_steps_per_second": 8.257,
|
| 274 |
+
"step": 38912
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.76,
|
| 278 |
+
"learning_rate": 0.0003863212870708643,
|
| 279 |
+
"loss": 0.4404,
|
| 280 |
+
"step": 40960
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.76,
|
| 284 |
+
"eval_loss": 0.6653844118118286,
|
| 285 |
+
"eval_runtime": 17.9192,
|
| 286 |
+
"eval_samples_per_second": 116.747,
|
| 287 |
+
"eval_steps_per_second": 7.311,
|
| 288 |
+
"step": 40960
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.9,
|
| 292 |
+
"learning_rate": 0.00037424687637876156,
|
| 293 |
+
"loss": 0.4334,
|
| 294 |
+
"step": 43008
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.9,
|
| 298 |
+
"eval_loss": 0.6553301811218262,
|
| 299 |
+
"eval_runtime": 15.9304,
|
| 300 |
+
"eval_samples_per_second": 131.321,
|
| 301 |
+
"eval_steps_per_second": 8.223,
|
| 302 |
+
"step": 43008
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 0.00036178064571954134,
|
| 307 |
+
"loss": 0.4278,
|
| 308 |
+
"step": 45056
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.04,
|
| 312 |
+
"eval_loss": 0.6766741871833801,
|
| 313 |
+
"eval_runtime": 16.9834,
|
| 314 |
+
"eval_samples_per_second": 123.179,
|
| 315 |
+
"eval_steps_per_second": 7.713,
|
| 316 |
+
"step": 45056
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.18,
|
| 320 |
+
"learning_rate": 0.0003489439971353363,
|
| 321 |
+
"loss": 0.4152,
|
| 322 |
+
"step": 47104
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.18,
|
| 326 |
+
"eval_loss": 0.6645193696022034,
|
| 327 |
+
"eval_runtime": 15.816,
|
| 328 |
+
"eval_samples_per_second": 132.271,
|
| 329 |
+
"eval_steps_per_second": 8.283,
|
| 330 |
+
"step": 47104
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.32,
|
| 334 |
+
"learning_rate": 0.0003357965820476752,
|
| 335 |
+
"loss": 0.411,
|
| 336 |
+
"step": 49152
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.32,
|
| 340 |
+
"eval_loss": 0.6748972535133362,
|
| 341 |
+
"eval_runtime": 17.7507,
|
| 342 |
+
"eval_samples_per_second": 117.855,
|
| 343 |
+
"eval_steps_per_second": 7.38,
|
| 344 |
+
"step": 49152
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.45,
|
| 348 |
+
"learning_rate": 0.0003223743322236833,
|
| 349 |
+
"loss": 0.4085,
|
| 350 |
+
"step": 51200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.45,
|
| 354 |
+
"eval_loss": 0.6703893542289734,
|
| 355 |
+
"eval_runtime": 17.341,
|
| 356 |
+
"eval_samples_per_second": 120.639,
|
| 357 |
+
"eval_steps_per_second": 7.554,
|
| 358 |
+
"step": 51200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.59,
|
| 362 |
+
"learning_rate": 0.0003087269633577651,
|
| 363 |
+
"loss": 0.4037,
|
| 364 |
+
"step": 53248
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.59,
|
| 368 |
+
"eval_loss": 0.6561577916145325,
|
| 369 |
+
"eval_runtime": 15.9295,
|
| 370 |
+
"eval_samples_per_second": 131.329,
|
| 371 |
+
"eval_steps_per_second": 8.224,
|
| 372 |
+
"step": 53248
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.73,
|
| 376 |
+
"learning_rate": 0.0002948780545870247,
|
| 377 |
+
"loss": 0.3975,
|
| 378 |
+
"step": 55296
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.73,
|
| 382 |
+
"eval_loss": 0.6698916554450989,
|
| 383 |
+
"eval_runtime": 17.9008,
|
| 384 |
+
"eval_samples_per_second": 116.866,
|
| 385 |
+
"eval_steps_per_second": 7.318,
|
| 386 |
+
"step": 55296
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.87,
|
| 390 |
+
"learning_rate": 0.0002808921064161573,
|
| 391 |
+
"loss": 0.3917,
|
| 392 |
+
"step": 57344
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.87,
|
| 396 |
+
"eval_loss": 0.7029281854629517,
|
| 397 |
+
"eval_runtime": 17.8146,
|
| 398 |
+
"eval_samples_per_second": 117.432,
|
| 399 |
+
"eval_steps_per_second": 7.354,
|
| 400 |
+
"step": 57344
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.01,
|
| 404 |
+
"learning_rate": 0.00026680720064442787,
|
| 405 |
+
"loss": 0.3876,
|
| 406 |
+
"step": 59392
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.01,
|
| 410 |
+
"eval_loss": 0.6837398409843445,
|
| 411 |
+
"eval_runtime": 17.4354,
|
| 412 |
+
"eval_samples_per_second": 119.986,
|
| 413 |
+
"eval_steps_per_second": 7.513,
|
| 414 |
+
"step": 59392
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.14,
|
| 418 |
+
"learning_rate": 0.00025266845586830784,
|
| 419 |
+
"loss": 0.3751,
|
| 420 |
+
"step": 61440
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.14,
|
| 424 |
+
"eval_loss": 0.690500795841217,
|
| 425 |
+
"eval_runtime": 17.8964,
|
| 426 |
+
"eval_samples_per_second": 116.895,
|
| 427 |
+
"eval_steps_per_second": 7.32,
|
| 428 |
+
"step": 61440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.28,
|
| 432 |
+
"learning_rate": 0.00023851425721450398,
|
| 433 |
+
"loss": 0.3719,
|
| 434 |
+
"step": 63488
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.28,
|
| 438 |
+
"eval_loss": 0.6938452124595642,
|
| 439 |
+
"eval_runtime": 15.7962,
|
| 440 |
+
"eval_samples_per_second": 132.437,
|
| 441 |
+
"eval_steps_per_second": 8.293,
|
| 442 |
+
"step": 63488
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.42,
|
| 446 |
+
"learning_rate": 0.0002244106409269819,
|
| 447 |
+
"loss": 0.3644,
|
| 448 |
+
"step": 65536
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.42,
|
| 452 |
+
"eval_loss": 0.6987010836601257,
|
| 453 |
+
"eval_runtime": 15.9556,
|
| 454 |
+
"eval_samples_per_second": 131.114,
|
| 455 |
+
"eval_steps_per_second": 8.21,
|
| 456 |
+
"step": 65536
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.56,
|
| 460 |
+
"learning_rate": 0.00021037526400320187,
|
| 461 |
+
"loss": 0.357,
|
| 462 |
+
"step": 67584
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.56,
|
| 466 |
+
"eval_loss": 0.6973423957824707,
|
| 467 |
+
"eval_runtime": 15.8273,
|
| 468 |
+
"eval_samples_per_second": 132.177,
|
| 469 |
+
"eval_steps_per_second": 8.277,
|
| 470 |
+
"step": 67584
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7,
|
| 474 |
+
"learning_rate": 0.0001964736950807942,
|
| 475 |
+
"loss": 0.357,
|
| 476 |
+
"step": 69632
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.7,
|
| 480 |
+
"eval_loss": 0.6949540972709656,
|
| 481 |
+
"eval_runtime": 16.1065,
|
| 482 |
+
"eval_samples_per_second": 129.885,
|
| 483 |
+
"eval_steps_per_second": 8.133,
|
| 484 |
+
"step": 69632
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.83,
|
| 488 |
+
"learning_rate": 0.00018274358855873096,
|
| 489 |
+
"loss": 0.3502,
|
| 490 |
+
"step": 71680
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.83,
|
| 494 |
+
"eval_loss": 0.6959769129753113,
|
| 495 |
+
"eval_runtime": 16.2001,
|
| 496 |
+
"eval_samples_per_second": 129.135,
|
| 497 |
+
"eval_steps_per_second": 8.086,
|
| 498 |
+
"step": 71680
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.97,
|
| 502 |
+
"learning_rate": 0.00016922892649452222,
|
| 503 |
+
"loss": 0.3458,
|
| 504 |
+
"step": 73728
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.97,
|
| 508 |
+
"eval_loss": 0.7017838358879089,
|
| 509 |
+
"eval_runtime": 18.0467,
|
| 510 |
+
"eval_samples_per_second": 115.922,
|
| 511 |
+
"eval_steps_per_second": 7.259,
|
| 512 |
+
"step": 73728
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.11,
|
| 516 |
+
"learning_rate": 0.00015597300080605504,
|
| 517 |
+
"loss": 0.3302,
|
| 518 |
+
"step": 75776
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.11,
|
| 522 |
+
"eval_loss": 0.7075567841529846,
|
| 523 |
+
"eval_runtime": 16.7136,
|
| 524 |
+
"eval_samples_per_second": 125.168,
|
| 525 |
+
"eval_steps_per_second": 7.838,
|
| 526 |
+
"step": 75776
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.25,
|
| 530 |
+
"learning_rate": 0.0001430182745933093,
|
| 531 |
+
"loss": 0.3278,
|
| 532 |
+
"step": 77824
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.25,
|
| 536 |
+
"eval_loss": 0.7178707718849182,
|
| 537 |
+
"eval_runtime": 15.7714,
|
| 538 |
+
"eval_samples_per_second": 132.645,
|
| 539 |
+
"eval_steps_per_second": 8.306,
|
| 540 |
+
"step": 77824
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.39,
|
| 544 |
+
"learning_rate": 0.00013040017526934073,
|
| 545 |
+
"loss": 0.3224,
|
| 546 |
+
"step": 79872
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 5.39,
|
| 550 |
+
"eval_loss": 0.725006639957428,
|
| 551 |
+
"eval_runtime": 15.8618,
|
| 552 |
+
"eval_samples_per_second": 131.889,
|
| 553 |
+
"eval_steps_per_second": 8.259,
|
| 554 |
+
"step": 79872
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.53,
|
| 558 |
+
"learning_rate": 0.00011817144183980649,
|
| 559 |
+
"loss": 0.3181,
|
| 560 |
+
"step": 81920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.53,
|
| 564 |
+
"eval_loss": 0.7247176766395569,
|
| 565 |
+
"eval_runtime": 17.8123,
|
| 566 |
+
"eval_samples_per_second": 117.447,
|
| 567 |
+
"eval_steps_per_second": 7.354,
|
| 568 |
+
"step": 81920
|
| 569 |
+
}
|
| 570 |
+
],
|
| 571 |
+
"max_steps": 118608,
|
| 572 |
+
"num_train_epochs": 8,
|
| 573 |
+
"total_flos": 3.358824123721482e+17,
|
| 574 |
+
"trial_name": null,
|
| 575 |
+
"trial_params": null
|
| 576 |
+
}
|
checkpoint-81920/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0200139492e444a9d322a4f90a96e6dde09c7a882f05b816c2345dade5ea0f98
|
| 3 |
+
size 3515
|
checkpoint-90112/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
checkpoint-90112/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9348484461aa0efc6ef43d1cd9f68f1780cc6f0a8d16df3dba1a94e0a24334c9
|
| 3 |
+
size 211432837
|
checkpoint-90112/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13bca6d5041775e0ef37e300c12fa2ef95343fea9202def32ad383199bfdc35b
|
| 3 |
+
size 139279005
|
checkpoint-90112/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:957f5daa73c63386d7677bd24415e97f9a46fa12239f90b764d0b1e53483b6a4
|
| 3 |
+
size 15597
|
checkpoint-90112/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65677617c481c4dea5774baaa3893a9d30edaa0722420de801d77f9d5102f186
|
| 3 |
+
size 557
|
checkpoint-90112/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d58edf844986f37be7d4bc128a885c7e910ca1227f1e618aed7827dd6a504ce
|
| 3 |
+
size 627
|
checkpoint-90112/trainer_state.json
ADDED
|
@@ -0,0 +1,632 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.077971131795494,
|
| 5 |
+
"global_step": 90112,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 0.00020480000000000002,
|
| 13 |
+
"loss": 2.1719,
|
| 14 |
+
"step": 2048
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.0833024978637695,
|
| 19 |
+
"eval_runtime": 15.7887,
|
| 20 |
+
"eval_samples_per_second": 132.5,
|
| 21 |
+
"eval_steps_per_second": 8.297,
|
| 22 |
+
"step": 2048
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 0.00040960000000000004,
|
| 27 |
+
"loss": 0.9468,
|
| 28 |
+
"step": 4096
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 0.8000818490982056,
|
| 33 |
+
"eval_runtime": 17.2632,
|
| 34 |
+
"eval_samples_per_second": 121.183,
|
| 35 |
+
"eval_steps_per_second": 7.588,
|
| 36 |
+
"step": 4096
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.41,
|
| 40 |
+
"learning_rate": 0.0004998749142723946,
|
| 41 |
+
"loss": 0.7542,
|
| 42 |
+
"step": 6144
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41,
|
| 46 |
+
"eval_loss": 0.7439925074577332,
|
| 47 |
+
"eval_runtime": 17.4134,
|
| 48 |
+
"eval_samples_per_second": 120.138,
|
| 49 |
+
"eval_steps_per_second": 7.523,
|
| 50 |
+
"step": 6144
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.55,
|
| 54 |
+
"learning_rate": 0.0004990273340312486,
|
| 55 |
+
"loss": 0.6756,
|
| 56 |
+
"step": 8192
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"eval_loss": 0.7018134593963623,
|
| 61 |
+
"eval_runtime": 15.8601,
|
| 62 |
+
"eval_samples_per_second": 131.904,
|
| 63 |
+
"eval_steps_per_second": 8.26,
|
| 64 |
+
"step": 8192
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.69,
|
| 68 |
+
"learning_rate": 0.0004973820371438889,
|
| 69 |
+
"loss": 0.631,
|
| 70 |
+
"step": 10240
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.69,
|
| 74 |
+
"eval_loss": 0.684688925743103,
|
| 75 |
+
"eval_runtime": 18.0314,
|
| 76 |
+
"eval_samples_per_second": 116.02,
|
| 77 |
+
"eval_steps_per_second": 7.265,
|
| 78 |
+
"step": 10240
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.83,
|
| 82 |
+
"learning_rate": 0.0004949442940386407,
|
| 83 |
+
"loss": 0.5989,
|
| 84 |
+
"step": 12288
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.83,
|
| 88 |
+
"eval_loss": 0.6822534203529358,
|
| 89 |
+
"eval_runtime": 17.4182,
|
| 90 |
+
"eval_samples_per_second": 120.104,
|
| 91 |
+
"eval_steps_per_second": 7.521,
|
| 92 |
+
"step": 12288
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.97,
|
| 96 |
+
"learning_rate": 0.0004917201492451735,
|
| 97 |
+
"loss": 0.5776,
|
| 98 |
+
"step": 14336
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"eval_loss": 0.6674544215202332,
|
| 103 |
+
"eval_runtime": 17.7487,
|
| 104 |
+
"eval_samples_per_second": 117.868,
|
| 105 |
+
"eval_steps_per_second": 7.381,
|
| 106 |
+
"step": 14336
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.11,
|
| 110 |
+
"learning_rate": 0.0004877230785006913,
|
| 111 |
+
"loss": 0.5525,
|
| 112 |
+
"step": 16384
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.11,
|
| 116 |
+
"eval_loss": 0.6640163660049438,
|
| 117 |
+
"eval_runtime": 15.8951,
|
| 118 |
+
"eval_samples_per_second": 131.613,
|
| 119 |
+
"eval_steps_per_second": 8.242,
|
| 120 |
+
"step": 16384
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 0.0004829670105577666,
|
| 125 |
+
"loss": 0.5373,
|
| 126 |
+
"step": 18432
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.24,
|
| 130 |
+
"eval_loss": 0.6663170456886292,
|
| 131 |
+
"eval_runtime": 17.5,
|
| 132 |
+
"eval_samples_per_second": 119.543,
|
| 133 |
+
"eval_steps_per_second": 7.486,
|
| 134 |
+
"step": 18432
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.38,
|
| 138 |
+
"learning_rate": 0.0004774596641323791,
|
| 139 |
+
"loss": 0.5261,
|
| 140 |
+
"step": 20480
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"eval_loss": 0.6518951654434204,
|
| 145 |
+
"eval_runtime": 17.1326,
|
| 146 |
+
"eval_samples_per_second": 122.107,
|
| 147 |
+
"eval_steps_per_second": 7.646,
|
| 148 |
+
"step": 20480
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.52,
|
| 152 |
+
"learning_rate": 0.0004712261976082475,
|
| 153 |
+
"loss": 0.5115,
|
| 154 |
+
"step": 22528
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.52,
|
| 158 |
+
"eval_loss": 0.6658991575241089,
|
| 159 |
+
"eval_runtime": 17.6874,
|
| 160 |
+
"eval_samples_per_second": 118.277,
|
| 161 |
+
"eval_steps_per_second": 7.406,
|
| 162 |
+
"step": 22528
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.66,
|
| 166 |
+
"learning_rate": 0.00046428407064289515,
|
| 167 |
+
"loss": 0.5033,
|
| 168 |
+
"step": 24576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.66,
|
| 172 |
+
"eval_loss": 0.6602604389190674,
|
| 173 |
+
"eval_runtime": 15.8845,
|
| 174 |
+
"eval_samples_per_second": 131.701,
|
| 175 |
+
"eval_steps_per_second": 8.247,
|
| 176 |
+
"step": 24576
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8,
|
| 180 |
+
"learning_rate": 0.00045665163060732317,
|
| 181 |
+
"loss": 0.4925,
|
| 182 |
+
"step": 26624
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.8,
|
| 186 |
+
"eval_loss": 0.6551440954208374,
|
| 187 |
+
"eval_runtime": 17.7902,
|
| 188 |
+
"eval_samples_per_second": 117.593,
|
| 189 |
+
"eval_steps_per_second": 7.364,
|
| 190 |
+
"step": 26624
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.93,
|
| 194 |
+
"learning_rate": 0.000448360778288479,
|
| 195 |
+
"loss": 0.4868,
|
| 196 |
+
"step": 28672
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.93,
|
| 200 |
+
"eval_loss": 0.6580803394317627,
|
| 201 |
+
"eval_runtime": 15.7841,
|
| 202 |
+
"eval_samples_per_second": 132.538,
|
| 203 |
+
"eval_steps_per_second": 8.299,
|
| 204 |
+
"step": 28672
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.00043942999964293453,
|
| 209 |
+
"loss": 0.4741,
|
| 210 |
+
"step": 30720
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.07,
|
| 214 |
+
"eval_loss": 0.661230742931366,
|
| 215 |
+
"eval_runtime": 18.015,
|
| 216 |
+
"eval_samples_per_second": 116.126,
|
| 217 |
+
"eval_steps_per_second": 7.272,
|
| 218 |
+
"step": 30720
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.21,
|
| 222 |
+
"learning_rate": 0.0004298966220346151,
|
| 223 |
+
"loss": 0.4632,
|
| 224 |
+
"step": 32768
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.21,
|
| 228 |
+
"eval_loss": 0.6618030071258545,
|
| 229 |
+
"eval_runtime": 17.9073,
|
| 230 |
+
"eval_samples_per_second": 116.824,
|
| 231 |
+
"eval_steps_per_second": 7.315,
|
| 232 |
+
"step": 32768
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.35,
|
| 236 |
+
"learning_rate": 0.00041978697624050446,
|
| 237 |
+
"loss": 0.4584,
|
| 238 |
+
"step": 34816
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.35,
|
| 242 |
+
"eval_loss": 0.6579039096832275,
|
| 243 |
+
"eval_runtime": 17.8241,
|
| 244 |
+
"eval_samples_per_second": 117.369,
|
| 245 |
+
"eval_steps_per_second": 7.35,
|
| 246 |
+
"step": 34816
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.49,
|
| 250 |
+
"learning_rate": 0.0004091334467888659,
|
| 251 |
+
"loss": 0.451,
|
| 252 |
+
"step": 36864
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"eval_loss": 0.6627541780471802,
|
| 257 |
+
"eval_runtime": 17.6822,
|
| 258 |
+
"eval_samples_per_second": 118.311,
|
| 259 |
+
"eval_steps_per_second": 7.409,
|
| 260 |
+
"step": 36864
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.62,
|
| 264 |
+
"learning_rate": 0.00039796458815002033,
|
| 265 |
+
"loss": 0.445,
|
| 266 |
+
"step": 38912
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.62,
|
| 270 |
+
"eval_loss": 0.656867265701294,
|
| 271 |
+
"eval_runtime": 15.8659,
|
| 272 |
+
"eval_samples_per_second": 131.855,
|
| 273 |
+
"eval_steps_per_second": 8.257,
|
| 274 |
+
"step": 38912
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.76,
|
| 278 |
+
"learning_rate": 0.0003863212870708643,
|
| 279 |
+
"loss": 0.4404,
|
| 280 |
+
"step": 40960
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.76,
|
| 284 |
+
"eval_loss": 0.6653844118118286,
|
| 285 |
+
"eval_runtime": 17.9192,
|
| 286 |
+
"eval_samples_per_second": 116.747,
|
| 287 |
+
"eval_steps_per_second": 7.311,
|
| 288 |
+
"step": 40960
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.9,
|
| 292 |
+
"learning_rate": 0.00037424687637876156,
|
| 293 |
+
"loss": 0.4334,
|
| 294 |
+
"step": 43008
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.9,
|
| 298 |
+
"eval_loss": 0.6553301811218262,
|
| 299 |
+
"eval_runtime": 15.9304,
|
| 300 |
+
"eval_samples_per_second": 131.321,
|
| 301 |
+
"eval_steps_per_second": 8.223,
|
| 302 |
+
"step": 43008
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 0.00036178064571954134,
|
| 307 |
+
"loss": 0.4278,
|
| 308 |
+
"step": 45056
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.04,
|
| 312 |
+
"eval_loss": 0.6766741871833801,
|
| 313 |
+
"eval_runtime": 16.9834,
|
| 314 |
+
"eval_samples_per_second": 123.179,
|
| 315 |
+
"eval_steps_per_second": 7.713,
|
| 316 |
+
"step": 45056
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.18,
|
| 320 |
+
"learning_rate": 0.0003489439971353363,
|
| 321 |
+
"loss": 0.4152,
|
| 322 |
+
"step": 47104
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.18,
|
| 326 |
+
"eval_loss": 0.6645193696022034,
|
| 327 |
+
"eval_runtime": 15.816,
|
| 328 |
+
"eval_samples_per_second": 132.271,
|
| 329 |
+
"eval_steps_per_second": 8.283,
|
| 330 |
+
"step": 47104
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.32,
|
| 334 |
+
"learning_rate": 0.0003357965820476752,
|
| 335 |
+
"loss": 0.411,
|
| 336 |
+
"step": 49152
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.32,
|
| 340 |
+
"eval_loss": 0.6748972535133362,
|
| 341 |
+
"eval_runtime": 17.7507,
|
| 342 |
+
"eval_samples_per_second": 117.855,
|
| 343 |
+
"eval_steps_per_second": 7.38,
|
| 344 |
+
"step": 49152
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.45,
|
| 348 |
+
"learning_rate": 0.0003223743322236833,
|
| 349 |
+
"loss": 0.4085,
|
| 350 |
+
"step": 51200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.45,
|
| 354 |
+
"eval_loss": 0.6703893542289734,
|
| 355 |
+
"eval_runtime": 17.341,
|
| 356 |
+
"eval_samples_per_second": 120.639,
|
| 357 |
+
"eval_steps_per_second": 7.554,
|
| 358 |
+
"step": 51200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.59,
|
| 362 |
+
"learning_rate": 0.0003087269633577651,
|
| 363 |
+
"loss": 0.4037,
|
| 364 |
+
"step": 53248
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.59,
|
| 368 |
+
"eval_loss": 0.6561577916145325,
|
| 369 |
+
"eval_runtime": 15.9295,
|
| 370 |
+
"eval_samples_per_second": 131.329,
|
| 371 |
+
"eval_steps_per_second": 8.224,
|
| 372 |
+
"step": 53248
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.73,
|
| 376 |
+
"learning_rate": 0.0002948780545870247,
|
| 377 |
+
"loss": 0.3975,
|
| 378 |
+
"step": 55296
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.73,
|
| 382 |
+
"eval_loss": 0.6698916554450989,
|
| 383 |
+
"eval_runtime": 17.9008,
|
| 384 |
+
"eval_samples_per_second": 116.866,
|
| 385 |
+
"eval_steps_per_second": 7.318,
|
| 386 |
+
"step": 55296
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.87,
|
| 390 |
+
"learning_rate": 0.0002808921064161573,
|
| 391 |
+
"loss": 0.3917,
|
| 392 |
+
"step": 57344
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.87,
|
| 396 |
+
"eval_loss": 0.7029281854629517,
|
| 397 |
+
"eval_runtime": 17.8146,
|
| 398 |
+
"eval_samples_per_second": 117.432,
|
| 399 |
+
"eval_steps_per_second": 7.354,
|
| 400 |
+
"step": 57344
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.01,
|
| 404 |
+
"learning_rate": 0.00026680720064442787,
|
| 405 |
+
"loss": 0.3876,
|
| 406 |
+
"step": 59392
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.01,
|
| 410 |
+
"eval_loss": 0.6837398409843445,
|
| 411 |
+
"eval_runtime": 17.4354,
|
| 412 |
+
"eval_samples_per_second": 119.986,
|
| 413 |
+
"eval_steps_per_second": 7.513,
|
| 414 |
+
"step": 59392
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.14,
|
| 418 |
+
"learning_rate": 0.00025266845586830784,
|
| 419 |
+
"loss": 0.3751,
|
| 420 |
+
"step": 61440
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.14,
|
| 424 |
+
"eval_loss": 0.690500795841217,
|
| 425 |
+
"eval_runtime": 17.8964,
|
| 426 |
+
"eval_samples_per_second": 116.895,
|
| 427 |
+
"eval_steps_per_second": 7.32,
|
| 428 |
+
"step": 61440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.28,
|
| 432 |
+
"learning_rate": 0.00023851425721450398,
|
| 433 |
+
"loss": 0.3719,
|
| 434 |
+
"step": 63488
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.28,
|
| 438 |
+
"eval_loss": 0.6938452124595642,
|
| 439 |
+
"eval_runtime": 15.7962,
|
| 440 |
+
"eval_samples_per_second": 132.437,
|
| 441 |
+
"eval_steps_per_second": 8.293,
|
| 442 |
+
"step": 63488
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.42,
|
| 446 |
+
"learning_rate": 0.0002244106409269819,
|
| 447 |
+
"loss": 0.3644,
|
| 448 |
+
"step": 65536
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.42,
|
| 452 |
+
"eval_loss": 0.6987010836601257,
|
| 453 |
+
"eval_runtime": 15.9556,
|
| 454 |
+
"eval_samples_per_second": 131.114,
|
| 455 |
+
"eval_steps_per_second": 8.21,
|
| 456 |
+
"step": 65536
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.56,
|
| 460 |
+
"learning_rate": 0.00021037526400320187,
|
| 461 |
+
"loss": 0.357,
|
| 462 |
+
"step": 67584
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.56,
|
| 466 |
+
"eval_loss": 0.6973423957824707,
|
| 467 |
+
"eval_runtime": 15.8273,
|
| 468 |
+
"eval_samples_per_second": 132.177,
|
| 469 |
+
"eval_steps_per_second": 8.277,
|
| 470 |
+
"step": 67584
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7,
|
| 474 |
+
"learning_rate": 0.0001964736950807942,
|
| 475 |
+
"loss": 0.357,
|
| 476 |
+
"step": 69632
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.7,
|
| 480 |
+
"eval_loss": 0.6949540972709656,
|
| 481 |
+
"eval_runtime": 16.1065,
|
| 482 |
+
"eval_samples_per_second": 129.885,
|
| 483 |
+
"eval_steps_per_second": 8.133,
|
| 484 |
+
"step": 69632
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.83,
|
| 488 |
+
"learning_rate": 0.00018274358855873096,
|
| 489 |
+
"loss": 0.3502,
|
| 490 |
+
"step": 71680
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.83,
|
| 494 |
+
"eval_loss": 0.6959769129753113,
|
| 495 |
+
"eval_runtime": 16.2001,
|
| 496 |
+
"eval_samples_per_second": 129.135,
|
| 497 |
+
"eval_steps_per_second": 8.086,
|
| 498 |
+
"step": 71680
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.97,
|
| 502 |
+
"learning_rate": 0.00016922892649452222,
|
| 503 |
+
"loss": 0.3458,
|
| 504 |
+
"step": 73728
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.97,
|
| 508 |
+
"eval_loss": 0.7017838358879089,
|
| 509 |
+
"eval_runtime": 18.0467,
|
| 510 |
+
"eval_samples_per_second": 115.922,
|
| 511 |
+
"eval_steps_per_second": 7.259,
|
| 512 |
+
"step": 73728
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.11,
|
| 516 |
+
"learning_rate": 0.00015597300080605504,
|
| 517 |
+
"loss": 0.3302,
|
| 518 |
+
"step": 75776
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.11,
|
| 522 |
+
"eval_loss": 0.7075567841529846,
|
| 523 |
+
"eval_runtime": 16.7136,
|
| 524 |
+
"eval_samples_per_second": 125.168,
|
| 525 |
+
"eval_steps_per_second": 7.838,
|
| 526 |
+
"step": 75776
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.25,
|
| 530 |
+
"learning_rate": 0.0001430182745933093,
|
| 531 |
+
"loss": 0.3278,
|
| 532 |
+
"step": 77824
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.25,
|
| 536 |
+
"eval_loss": 0.7178707718849182,
|
| 537 |
+
"eval_runtime": 15.7714,
|
| 538 |
+
"eval_samples_per_second": 132.645,
|
| 539 |
+
"eval_steps_per_second": 8.306,
|
| 540 |
+
"step": 77824
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.39,
|
| 544 |
+
"learning_rate": 0.00013040017526934073,
|
| 545 |
+
"loss": 0.3224,
|
| 546 |
+
"step": 79872
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 5.39,
|
| 550 |
+
"eval_loss": 0.725006639957428,
|
| 551 |
+
"eval_runtime": 15.8618,
|
| 552 |
+
"eval_samples_per_second": 131.889,
|
| 553 |
+
"eval_steps_per_second": 8.259,
|
| 554 |
+
"step": 79872
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.53,
|
| 558 |
+
"learning_rate": 0.00011817144183980649,
|
| 559 |
+
"loss": 0.3181,
|
| 560 |
+
"step": 81920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.53,
|
| 564 |
+
"eval_loss": 0.7247176766395569,
|
| 565 |
+
"eval_runtime": 17.8123,
|
| 566 |
+
"eval_samples_per_second": 117.447,
|
| 567 |
+
"eval_steps_per_second": 7.354,
|
| 568 |
+
"step": 81920
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 5.66,
|
| 572 |
+
"learning_rate": 0.00010636499874117036,
|
| 573 |
+
"loss": 0.3154,
|
| 574 |
+
"step": 83968
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 5.66,
|
| 578 |
+
"eval_loss": 0.7248900532722473,
|
| 579 |
+
"eval_runtime": 17.751,
|
| 580 |
+
"eval_samples_per_second": 117.852,
|
| 581 |
+
"eval_steps_per_second": 7.38,
|
| 582 |
+
"step": 83968
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 5.8,
|
| 586 |
+
"learning_rate": 9.501866590283475e-05,
|
| 587 |
+
"loss": 0.3105,
|
| 588 |
+
"step": 86016
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.8,
|
| 592 |
+
"eval_loss": 0.7352888584136963,
|
| 593 |
+
"eval_runtime": 18.0338,
|
| 594 |
+
"eval_samples_per_second": 116.004,
|
| 595 |
+
"eval_steps_per_second": 7.264,
|
| 596 |
+
"step": 86016
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.94,
|
| 600 |
+
"learning_rate": 8.416361604489855e-05,
|
| 601 |
+
"loss": 0.3035,
|
| 602 |
+
"step": 88064
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.94,
|
| 606 |
+
"eval_loss": 0.7256708741188049,
|
| 607 |
+
"eval_runtime": 15.94,
|
| 608 |
+
"eval_samples_per_second": 131.242,
|
| 609 |
+
"eval_steps_per_second": 8.218,
|
| 610 |
+
"step": 88064
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 6.08,
|
| 614 |
+
"learning_rate": 7.384521927589935e-05,
|
| 615 |
+
"loss": 0.294,
|
| 616 |
+
"step": 90112
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.08,
|
| 620 |
+
"eval_loss": 0.7387065291404724,
|
| 621 |
+
"eval_runtime": 16.8562,
|
| 622 |
+
"eval_samples_per_second": 124.109,
|
| 623 |
+
"eval_steps_per_second": 7.772,
|
| 624 |
+
"step": 90112
|
| 625 |
+
}
|
| 626 |
+
],
|
| 627 |
+
"max_steps": 118608,
|
| 628 |
+
"num_train_epochs": 8,
|
| 629 |
+
"total_flos": 3.6946934765226394e+17,
|
| 630 |
+
"trial_name": null,
|
| 631 |
+
"trial_params": null
|
| 632 |
+
}
|
checkpoint-90112/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0200139492e444a9d322a4f90a96e6dde09c7a882f05b816c2345dade5ea0f98
|
| 3 |
+
size 3515
|
checkpoint-98304/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
checkpoint-98304/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d7734a595a50ad41bcd32219a078b91e1114f3997fc3e799e26f06248256806
|
| 3 |
+
size 211432837
|
checkpoint-98304/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a99311b788b49a1b31c7ea78dbd1c0a61dbe1fccd8921a0e214881612326a75
|
| 3 |
+
size 139279005
|
checkpoint-98304/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c784673775d4d2de7e801421c9e7e86119774c857f824cd3e81863a68f877c6
|
| 3 |
+
size 15597
|
checkpoint-98304/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78b807021032a80d50f2569a9d524c7cc698851ef9731d8e200116f32fef0109
|
| 3 |
+
size 557
|
checkpoint-98304/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2f16428ba1197dd50c363ca86ce195e15241b8dd827371e6482f2b04693f68a
|
| 3 |
+
size 627
|
checkpoint-98304/trainer_state.json
ADDED
|
@@ -0,0 +1,688 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.630513961958721,
|
| 5 |
+
"global_step": 98304,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 0.00020480000000000002,
|
| 13 |
+
"loss": 2.1719,
|
| 14 |
+
"step": 2048
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_loss": 1.0833024978637695,
|
| 19 |
+
"eval_runtime": 15.7887,
|
| 20 |
+
"eval_samples_per_second": 132.5,
|
| 21 |
+
"eval_steps_per_second": 8.297,
|
| 22 |
+
"step": 2048
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.28,
|
| 26 |
+
"learning_rate": 0.00040960000000000004,
|
| 27 |
+
"loss": 0.9468,
|
| 28 |
+
"step": 4096
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.28,
|
| 32 |
+
"eval_loss": 0.8000818490982056,
|
| 33 |
+
"eval_runtime": 17.2632,
|
| 34 |
+
"eval_samples_per_second": 121.183,
|
| 35 |
+
"eval_steps_per_second": 7.588,
|
| 36 |
+
"step": 4096
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.41,
|
| 40 |
+
"learning_rate": 0.0004998749142723946,
|
| 41 |
+
"loss": 0.7542,
|
| 42 |
+
"step": 6144
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.41,
|
| 46 |
+
"eval_loss": 0.7439925074577332,
|
| 47 |
+
"eval_runtime": 17.4134,
|
| 48 |
+
"eval_samples_per_second": 120.138,
|
| 49 |
+
"eval_steps_per_second": 7.523,
|
| 50 |
+
"step": 6144
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.55,
|
| 54 |
+
"learning_rate": 0.0004990273340312486,
|
| 55 |
+
"loss": 0.6756,
|
| 56 |
+
"step": 8192
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"eval_loss": 0.7018134593963623,
|
| 61 |
+
"eval_runtime": 15.8601,
|
| 62 |
+
"eval_samples_per_second": 131.904,
|
| 63 |
+
"eval_steps_per_second": 8.26,
|
| 64 |
+
"step": 8192
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.69,
|
| 68 |
+
"learning_rate": 0.0004973820371438889,
|
| 69 |
+
"loss": 0.631,
|
| 70 |
+
"step": 10240
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.69,
|
| 74 |
+
"eval_loss": 0.684688925743103,
|
| 75 |
+
"eval_runtime": 18.0314,
|
| 76 |
+
"eval_samples_per_second": 116.02,
|
| 77 |
+
"eval_steps_per_second": 7.265,
|
| 78 |
+
"step": 10240
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.83,
|
| 82 |
+
"learning_rate": 0.0004949442940386407,
|
| 83 |
+
"loss": 0.5989,
|
| 84 |
+
"step": 12288
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.83,
|
| 88 |
+
"eval_loss": 0.6822534203529358,
|
| 89 |
+
"eval_runtime": 17.4182,
|
| 90 |
+
"eval_samples_per_second": 120.104,
|
| 91 |
+
"eval_steps_per_second": 7.521,
|
| 92 |
+
"step": 12288
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.97,
|
| 96 |
+
"learning_rate": 0.0004917201492451735,
|
| 97 |
+
"loss": 0.5776,
|
| 98 |
+
"step": 14336
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"eval_loss": 0.6674544215202332,
|
| 103 |
+
"eval_runtime": 17.7487,
|
| 104 |
+
"eval_samples_per_second": 117.868,
|
| 105 |
+
"eval_steps_per_second": 7.381,
|
| 106 |
+
"step": 14336
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.11,
|
| 110 |
+
"learning_rate": 0.0004877230785006913,
|
| 111 |
+
"loss": 0.5525,
|
| 112 |
+
"step": 16384
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.11,
|
| 116 |
+
"eval_loss": 0.6640163660049438,
|
| 117 |
+
"eval_runtime": 15.8951,
|
| 118 |
+
"eval_samples_per_second": 131.613,
|
| 119 |
+
"eval_steps_per_second": 8.242,
|
| 120 |
+
"step": 16384
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.24,
|
| 124 |
+
"learning_rate": 0.0004829670105577666,
|
| 125 |
+
"loss": 0.5373,
|
| 126 |
+
"step": 18432
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.24,
|
| 130 |
+
"eval_loss": 0.6663170456886292,
|
| 131 |
+
"eval_runtime": 17.5,
|
| 132 |
+
"eval_samples_per_second": 119.543,
|
| 133 |
+
"eval_steps_per_second": 7.486,
|
| 134 |
+
"step": 18432
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.38,
|
| 138 |
+
"learning_rate": 0.0004774596641323791,
|
| 139 |
+
"loss": 0.5261,
|
| 140 |
+
"step": 20480
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"eval_loss": 0.6518951654434204,
|
| 145 |
+
"eval_runtime": 17.1326,
|
| 146 |
+
"eval_samples_per_second": 122.107,
|
| 147 |
+
"eval_steps_per_second": 7.646,
|
| 148 |
+
"step": 20480
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.52,
|
| 152 |
+
"learning_rate": 0.0004712261976082475,
|
| 153 |
+
"loss": 0.5115,
|
| 154 |
+
"step": 22528
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.52,
|
| 158 |
+
"eval_loss": 0.6658991575241089,
|
| 159 |
+
"eval_runtime": 17.6874,
|
| 160 |
+
"eval_samples_per_second": 118.277,
|
| 161 |
+
"eval_steps_per_second": 7.406,
|
| 162 |
+
"step": 22528
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.66,
|
| 166 |
+
"learning_rate": 0.00046428407064289515,
|
| 167 |
+
"loss": 0.5033,
|
| 168 |
+
"step": 24576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.66,
|
| 172 |
+
"eval_loss": 0.6602604389190674,
|
| 173 |
+
"eval_runtime": 15.8845,
|
| 174 |
+
"eval_samples_per_second": 131.701,
|
| 175 |
+
"eval_steps_per_second": 8.247,
|
| 176 |
+
"step": 24576
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8,
|
| 180 |
+
"learning_rate": 0.00045665163060732317,
|
| 181 |
+
"loss": 0.4925,
|
| 182 |
+
"step": 26624
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.8,
|
| 186 |
+
"eval_loss": 0.6551440954208374,
|
| 187 |
+
"eval_runtime": 17.7902,
|
| 188 |
+
"eval_samples_per_second": 117.593,
|
| 189 |
+
"eval_steps_per_second": 7.364,
|
| 190 |
+
"step": 26624
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.93,
|
| 194 |
+
"learning_rate": 0.000448360778288479,
|
| 195 |
+
"loss": 0.4868,
|
| 196 |
+
"step": 28672
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.93,
|
| 200 |
+
"eval_loss": 0.6580803394317627,
|
| 201 |
+
"eval_runtime": 15.7841,
|
| 202 |
+
"eval_samples_per_second": 132.538,
|
| 203 |
+
"eval_steps_per_second": 8.299,
|
| 204 |
+
"step": 28672
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.07,
|
| 208 |
+
"learning_rate": 0.00043942999964293453,
|
| 209 |
+
"loss": 0.4741,
|
| 210 |
+
"step": 30720
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.07,
|
| 214 |
+
"eval_loss": 0.661230742931366,
|
| 215 |
+
"eval_runtime": 18.015,
|
| 216 |
+
"eval_samples_per_second": 116.126,
|
| 217 |
+
"eval_steps_per_second": 7.272,
|
| 218 |
+
"step": 30720
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.21,
|
| 222 |
+
"learning_rate": 0.0004298966220346151,
|
| 223 |
+
"loss": 0.4632,
|
| 224 |
+
"step": 32768
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.21,
|
| 228 |
+
"eval_loss": 0.6618030071258545,
|
| 229 |
+
"eval_runtime": 17.9073,
|
| 230 |
+
"eval_samples_per_second": 116.824,
|
| 231 |
+
"eval_steps_per_second": 7.315,
|
| 232 |
+
"step": 32768
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.35,
|
| 236 |
+
"learning_rate": 0.00041978697624050446,
|
| 237 |
+
"loss": 0.4584,
|
| 238 |
+
"step": 34816
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 2.35,
|
| 242 |
+
"eval_loss": 0.6579039096832275,
|
| 243 |
+
"eval_runtime": 17.8241,
|
| 244 |
+
"eval_samples_per_second": 117.369,
|
| 245 |
+
"eval_steps_per_second": 7.35,
|
| 246 |
+
"step": 34816
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.49,
|
| 250 |
+
"learning_rate": 0.0004091334467888659,
|
| 251 |
+
"loss": 0.451,
|
| 252 |
+
"step": 36864
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"eval_loss": 0.6627541780471802,
|
| 257 |
+
"eval_runtime": 17.6822,
|
| 258 |
+
"eval_samples_per_second": 118.311,
|
| 259 |
+
"eval_steps_per_second": 7.409,
|
| 260 |
+
"step": 36864
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.62,
|
| 264 |
+
"learning_rate": 0.00039796458815002033,
|
| 265 |
+
"loss": 0.445,
|
| 266 |
+
"step": 38912
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.62,
|
| 270 |
+
"eval_loss": 0.656867265701294,
|
| 271 |
+
"eval_runtime": 15.8659,
|
| 272 |
+
"eval_samples_per_second": 131.855,
|
| 273 |
+
"eval_steps_per_second": 8.257,
|
| 274 |
+
"step": 38912
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.76,
|
| 278 |
+
"learning_rate": 0.0003863212870708643,
|
| 279 |
+
"loss": 0.4404,
|
| 280 |
+
"step": 40960
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 2.76,
|
| 284 |
+
"eval_loss": 0.6653844118118286,
|
| 285 |
+
"eval_runtime": 17.9192,
|
| 286 |
+
"eval_samples_per_second": 116.747,
|
| 287 |
+
"eval_steps_per_second": 7.311,
|
| 288 |
+
"step": 40960
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.9,
|
| 292 |
+
"learning_rate": 0.00037424687637876156,
|
| 293 |
+
"loss": 0.4334,
|
| 294 |
+
"step": 43008
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.9,
|
| 298 |
+
"eval_loss": 0.6553301811218262,
|
| 299 |
+
"eval_runtime": 15.9304,
|
| 300 |
+
"eval_samples_per_second": 131.321,
|
| 301 |
+
"eval_steps_per_second": 8.223,
|
| 302 |
+
"step": 43008
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 0.00036178064571954134,
|
| 307 |
+
"loss": 0.4278,
|
| 308 |
+
"step": 45056
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.04,
|
| 312 |
+
"eval_loss": 0.6766741871833801,
|
| 313 |
+
"eval_runtime": 16.9834,
|
| 314 |
+
"eval_samples_per_second": 123.179,
|
| 315 |
+
"eval_steps_per_second": 7.713,
|
| 316 |
+
"step": 45056
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.18,
|
| 320 |
+
"learning_rate": 0.0003489439971353363,
|
| 321 |
+
"loss": 0.4152,
|
| 322 |
+
"step": 47104
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.18,
|
| 326 |
+
"eval_loss": 0.6645193696022034,
|
| 327 |
+
"eval_runtime": 15.816,
|
| 328 |
+
"eval_samples_per_second": 132.271,
|
| 329 |
+
"eval_steps_per_second": 8.283,
|
| 330 |
+
"step": 47104
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.32,
|
| 334 |
+
"learning_rate": 0.0003357965820476752,
|
| 335 |
+
"loss": 0.411,
|
| 336 |
+
"step": 49152
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 3.32,
|
| 340 |
+
"eval_loss": 0.6748972535133362,
|
| 341 |
+
"eval_runtime": 17.7507,
|
| 342 |
+
"eval_samples_per_second": 117.855,
|
| 343 |
+
"eval_steps_per_second": 7.38,
|
| 344 |
+
"step": 49152
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.45,
|
| 348 |
+
"learning_rate": 0.0003223743322236833,
|
| 349 |
+
"loss": 0.4085,
|
| 350 |
+
"step": 51200
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.45,
|
| 354 |
+
"eval_loss": 0.6703893542289734,
|
| 355 |
+
"eval_runtime": 17.341,
|
| 356 |
+
"eval_samples_per_second": 120.639,
|
| 357 |
+
"eval_steps_per_second": 7.554,
|
| 358 |
+
"step": 51200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.59,
|
| 362 |
+
"learning_rate": 0.0003087269633577651,
|
| 363 |
+
"loss": 0.4037,
|
| 364 |
+
"step": 53248
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 3.59,
|
| 368 |
+
"eval_loss": 0.6561577916145325,
|
| 369 |
+
"eval_runtime": 15.9295,
|
| 370 |
+
"eval_samples_per_second": 131.329,
|
| 371 |
+
"eval_steps_per_second": 8.224,
|
| 372 |
+
"step": 53248
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.73,
|
| 376 |
+
"learning_rate": 0.0002948780545870247,
|
| 377 |
+
"loss": 0.3975,
|
| 378 |
+
"step": 55296
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 3.73,
|
| 382 |
+
"eval_loss": 0.6698916554450989,
|
| 383 |
+
"eval_runtime": 17.9008,
|
| 384 |
+
"eval_samples_per_second": 116.866,
|
| 385 |
+
"eval_steps_per_second": 7.318,
|
| 386 |
+
"step": 55296
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.87,
|
| 390 |
+
"learning_rate": 0.0002808921064161573,
|
| 391 |
+
"loss": 0.3917,
|
| 392 |
+
"step": 57344
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.87,
|
| 396 |
+
"eval_loss": 0.7029281854629517,
|
| 397 |
+
"eval_runtime": 17.8146,
|
| 398 |
+
"eval_samples_per_second": 117.432,
|
| 399 |
+
"eval_steps_per_second": 7.354,
|
| 400 |
+
"step": 57344
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 4.01,
|
| 404 |
+
"learning_rate": 0.00026680720064442787,
|
| 405 |
+
"loss": 0.3876,
|
| 406 |
+
"step": 59392
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.01,
|
| 410 |
+
"eval_loss": 0.6837398409843445,
|
| 411 |
+
"eval_runtime": 17.4354,
|
| 412 |
+
"eval_samples_per_second": 119.986,
|
| 413 |
+
"eval_steps_per_second": 7.513,
|
| 414 |
+
"step": 59392
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 4.14,
|
| 418 |
+
"learning_rate": 0.00025266845586830784,
|
| 419 |
+
"loss": 0.3751,
|
| 420 |
+
"step": 61440
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.14,
|
| 424 |
+
"eval_loss": 0.690500795841217,
|
| 425 |
+
"eval_runtime": 17.8964,
|
| 426 |
+
"eval_samples_per_second": 116.895,
|
| 427 |
+
"eval_steps_per_second": 7.32,
|
| 428 |
+
"step": 61440
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.28,
|
| 432 |
+
"learning_rate": 0.00023851425721450398,
|
| 433 |
+
"loss": 0.3719,
|
| 434 |
+
"step": 63488
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.28,
|
| 438 |
+
"eval_loss": 0.6938452124595642,
|
| 439 |
+
"eval_runtime": 15.7962,
|
| 440 |
+
"eval_samples_per_second": 132.437,
|
| 441 |
+
"eval_steps_per_second": 8.293,
|
| 442 |
+
"step": 63488
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 4.42,
|
| 446 |
+
"learning_rate": 0.0002244106409269819,
|
| 447 |
+
"loss": 0.3644,
|
| 448 |
+
"step": 65536
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.42,
|
| 452 |
+
"eval_loss": 0.6987010836601257,
|
| 453 |
+
"eval_runtime": 15.9556,
|
| 454 |
+
"eval_samples_per_second": 131.114,
|
| 455 |
+
"eval_steps_per_second": 8.21,
|
| 456 |
+
"step": 65536
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 4.56,
|
| 460 |
+
"learning_rate": 0.00021037526400320187,
|
| 461 |
+
"loss": 0.357,
|
| 462 |
+
"step": 67584
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.56,
|
| 466 |
+
"eval_loss": 0.6973423957824707,
|
| 467 |
+
"eval_runtime": 15.8273,
|
| 468 |
+
"eval_samples_per_second": 132.177,
|
| 469 |
+
"eval_steps_per_second": 8.277,
|
| 470 |
+
"step": 67584
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.7,
|
| 474 |
+
"learning_rate": 0.0001964736950807942,
|
| 475 |
+
"loss": 0.357,
|
| 476 |
+
"step": 69632
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.7,
|
| 480 |
+
"eval_loss": 0.6949540972709656,
|
| 481 |
+
"eval_runtime": 16.1065,
|
| 482 |
+
"eval_samples_per_second": 129.885,
|
| 483 |
+
"eval_steps_per_second": 8.133,
|
| 484 |
+
"step": 69632
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 4.83,
|
| 488 |
+
"learning_rate": 0.00018274358855873096,
|
| 489 |
+
"loss": 0.3502,
|
| 490 |
+
"step": 71680
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 4.83,
|
| 494 |
+
"eval_loss": 0.6959769129753113,
|
| 495 |
+
"eval_runtime": 16.2001,
|
| 496 |
+
"eval_samples_per_second": 129.135,
|
| 497 |
+
"eval_steps_per_second": 8.086,
|
| 498 |
+
"step": 71680
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 4.97,
|
| 502 |
+
"learning_rate": 0.00016922892649452222,
|
| 503 |
+
"loss": 0.3458,
|
| 504 |
+
"step": 73728
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 4.97,
|
| 508 |
+
"eval_loss": 0.7017838358879089,
|
| 509 |
+
"eval_runtime": 18.0467,
|
| 510 |
+
"eval_samples_per_second": 115.922,
|
| 511 |
+
"eval_steps_per_second": 7.259,
|
| 512 |
+
"step": 73728
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.11,
|
| 516 |
+
"learning_rate": 0.00015597300080605504,
|
| 517 |
+
"loss": 0.3302,
|
| 518 |
+
"step": 75776
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.11,
|
| 522 |
+
"eval_loss": 0.7075567841529846,
|
| 523 |
+
"eval_runtime": 16.7136,
|
| 524 |
+
"eval_samples_per_second": 125.168,
|
| 525 |
+
"eval_steps_per_second": 7.838,
|
| 526 |
+
"step": 75776
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.25,
|
| 530 |
+
"learning_rate": 0.0001430182745933093,
|
| 531 |
+
"loss": 0.3278,
|
| 532 |
+
"step": 77824
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 5.25,
|
| 536 |
+
"eval_loss": 0.7178707718849182,
|
| 537 |
+
"eval_runtime": 15.7714,
|
| 538 |
+
"eval_samples_per_second": 132.645,
|
| 539 |
+
"eval_steps_per_second": 8.306,
|
| 540 |
+
"step": 77824
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.39,
|
| 544 |
+
"learning_rate": 0.00013040017526934073,
|
| 545 |
+
"loss": 0.3224,
|
| 546 |
+
"step": 79872
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 5.39,
|
| 550 |
+
"eval_loss": 0.725006639957428,
|
| 551 |
+
"eval_runtime": 15.8618,
|
| 552 |
+
"eval_samples_per_second": 131.889,
|
| 553 |
+
"eval_steps_per_second": 8.259,
|
| 554 |
+
"step": 79872
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.53,
|
| 558 |
+
"learning_rate": 0.00011817144183980649,
|
| 559 |
+
"loss": 0.3181,
|
| 560 |
+
"step": 81920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.53,
|
| 564 |
+
"eval_loss": 0.7247176766395569,
|
| 565 |
+
"eval_runtime": 17.8123,
|
| 566 |
+
"eval_samples_per_second": 117.447,
|
| 567 |
+
"eval_steps_per_second": 7.354,
|
| 568 |
+
"step": 81920
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 5.66,
|
| 572 |
+
"learning_rate": 0.00010636499874117036,
|
| 573 |
+
"loss": 0.3154,
|
| 574 |
+
"step": 83968
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 5.66,
|
| 578 |
+
"eval_loss": 0.7248900532722473,
|
| 579 |
+
"eval_runtime": 17.751,
|
| 580 |
+
"eval_samples_per_second": 117.852,
|
| 581 |
+
"eval_steps_per_second": 7.38,
|
| 582 |
+
"step": 83968
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 5.8,
|
| 586 |
+
"learning_rate": 9.501866590283475e-05,
|
| 587 |
+
"loss": 0.3105,
|
| 588 |
+
"step": 86016
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.8,
|
| 592 |
+
"eval_loss": 0.7352888584136963,
|
| 593 |
+
"eval_runtime": 18.0338,
|
| 594 |
+
"eval_samples_per_second": 116.004,
|
| 595 |
+
"eval_steps_per_second": 7.264,
|
| 596 |
+
"step": 86016
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 5.94,
|
| 600 |
+
"learning_rate": 8.416361604489855e-05,
|
| 601 |
+
"loss": 0.3035,
|
| 602 |
+
"step": 88064
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.94,
|
| 606 |
+
"eval_loss": 0.7256708741188049,
|
| 607 |
+
"eval_runtime": 15.94,
|
| 608 |
+
"eval_samples_per_second": 131.242,
|
| 609 |
+
"eval_steps_per_second": 8.218,
|
| 610 |
+
"step": 88064
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 6.08,
|
| 614 |
+
"learning_rate": 7.384521927589935e-05,
|
| 615 |
+
"loss": 0.294,
|
| 616 |
+
"step": 90112
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.08,
|
| 620 |
+
"eval_loss": 0.7387065291404724,
|
| 621 |
+
"eval_runtime": 16.8562,
|
| 622 |
+
"eval_samples_per_second": 124.109,
|
| 623 |
+
"eval_steps_per_second": 7.772,
|
| 624 |
+
"step": 90112
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 6.22,
|
| 628 |
+
"learning_rate": 6.409110434142392e-05,
|
| 629 |
+
"loss": 0.2866,
|
| 630 |
+
"step": 92160
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.22,
|
| 634 |
+
"eval_loss": 0.7492235898971558,
|
| 635 |
+
"eval_runtime": 17.2872,
|
| 636 |
+
"eval_samples_per_second": 121.015,
|
| 637 |
+
"eval_steps_per_second": 7.578,
|
| 638 |
+
"step": 92160
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 6.35,
|
| 642 |
+
"learning_rate": 5.492819313147518e-05,
|
| 643 |
+
"loss": 0.2836,
|
| 644 |
+
"step": 94208
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 6.35,
|
| 648 |
+
"eval_loss": 0.7440558671951294,
|
| 649 |
+
"eval_runtime": 17.5257,
|
| 650 |
+
"eval_samples_per_second": 119.367,
|
| 651 |
+
"eval_steps_per_second": 7.475,
|
| 652 |
+
"step": 94208
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 6.49,
|
| 656 |
+
"learning_rate": 4.6394783238561305e-05,
|
| 657 |
+
"loss": 0.2811,
|
| 658 |
+
"step": 96256
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 6.49,
|
| 662 |
+
"eval_loss": 0.7577848434448242,
|
| 663 |
+
"eval_runtime": 15.8536,
|
| 664 |
+
"eval_samples_per_second": 131.957,
|
| 665 |
+
"eval_steps_per_second": 8.263,
|
| 666 |
+
"step": 96256
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 6.63,
|
| 670 |
+
"learning_rate": 3.851358797621554e-05,
|
| 671 |
+
"loss": 0.2774,
|
| 672 |
+
"step": 98304
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.63,
|
| 676 |
+
"eval_loss": 0.7483424544334412,
|
| 677 |
+
"eval_runtime": 15.8485,
|
| 678 |
+
"eval_samples_per_second": 132.0,
|
| 679 |
+
"eval_steps_per_second": 8.266,
|
| 680 |
+
"step": 98304
|
| 681 |
+
}
|
| 682 |
+
],
|
| 683 |
+
"max_steps": 118608,
|
| 684 |
+
"num_train_epochs": 8,
|
| 685 |
+
"total_flos": 4.030582068221829e+17,
|
| 686 |
+
"trial_name": null,
|
| 687 |
+
"trial_params": null
|
| 688 |
+
}
|
checkpoint-98304/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0200139492e444a9d322a4f90a96e6dde09c7a882f05b816c2345dade5ea0f98
|
| 3 |
+
size 3515
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 512,
|
| 14 |
+
"n_head": 8,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 8,
|
| 17 |
+
"n_positions": 2048,
|
| 18 |
+
"pad_token_id": 1,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.26.0.dev0",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 299
|
| 32 |
+
}
|
manual_upload/manual_upload/manual_upload/training_args.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"do_predict": false,
|
| 7 |
"evaluation_strategy": "steps",
|
| 8 |
"prediction_loss_only": false,
|
| 9 |
-
"per_device_train_batch_size":
|
| 10 |
"per_device_eval_batch_size": 8,
|
| 11 |
"per_gpu_train_batch_size": null,
|
| 12 |
"per_gpu_eval_batch_size": null,
|
|
|
|
| 6 |
"do_predict": false,
|
| 7 |
"evaluation_strategy": "steps",
|
| 8 |
"prediction_loss_only": false,
|
| 9 |
+
"per_device_train_batch_size": 10,
|
| 10 |
"per_device_eval_batch_size": 8,
|
| 11 |
"per_gpu_train_batch_size": null,
|
| 12 |
"per_gpu_eval_batch_size": null,
|
manual_upload/manual_upload/training_args.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"do_predict": false,
|
| 7 |
"evaluation_strategy": "steps",
|
| 8 |
"prediction_loss_only": false,
|
| 9 |
-
"per_device_train_batch_size":
|
| 10 |
"per_device_eval_batch_size": 8,
|
| 11 |
"per_gpu_train_batch_size": null,
|
| 12 |
"per_gpu_eval_batch_size": null,
|
|
|
|
| 6 |
"do_predict": false,
|
| 7 |
"evaluation_strategy": "steps",
|
| 8 |
"prediction_loss_only": false,
|
| 9 |
+
"per_device_train_batch_size": 8,
|
| 10 |
"per_device_eval_batch_size": 8,
|
| 11 |
"per_gpu_train_batch_size": null,
|
| 12 |
"per_gpu_eval_batch_size": null,
|
manual_upload/training_args.json
CHANGED
|
@@ -19,7 +19,7 @@
|
|
| 19 |
"adam_beta2": 0.999,
|
| 20 |
"adam_epsilon": 1e-08,
|
| 21 |
"max_grad_norm": 1.0,
|
| 22 |
-
"num_train_epochs":
|
| 23 |
"max_steps": -1,
|
| 24 |
"lr_scheduler_type": "cosine",
|
| 25 |
"warmup_ratio": 0.0,
|
|
|
|
| 19 |
"adam_beta2": 0.999,
|
| 20 |
"adam_epsilon": 1e-08,
|
| 21 |
"max_grad_norm": 1.0,
|
| 22 |
+
"num_train_epochs": 10,
|
| 23 |
"max_steps": -1,
|
| 24 |
"lr_scheduler_type": "cosine",
|
| 25 |
"warmup_ratio": 0.0,
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c428cc7de0185bbb550e80aa69a32f25f05fe29d68c409a71c41069f58cbae30
|
| 3 |
+
size 139279005
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"pad_token": "[PAD]"
|
| 3 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,357 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"version": "1.0",
|
| 3 |
+
"truncation": {
|
| 4 |
+
"direction": "Right",
|
| 5 |
+
"max_length": 2048,
|
| 6 |
+
"strategy": "LongestFirst",
|
| 7 |
+
"stride": 0
|
| 8 |
+
},
|
| 9 |
+
"padding": {
|
| 10 |
+
"strategy": "BatchLongest",
|
| 11 |
+
"direction": "Right",
|
| 12 |
+
"pad_to_multiple_of": null,
|
| 13 |
+
"pad_id": 1,
|
| 14 |
+
"pad_type_id": 0,
|
| 15 |
+
"pad_token": "[PAD]"
|
| 16 |
+
},
|
| 17 |
+
"added_tokens": [
|
| 18 |
+
{
|
| 19 |
+
"id": 0,
|
| 20 |
+
"content": "[UNK]",
|
| 21 |
+
"single_word": false,
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"id": 1,
|
| 29 |
+
"content": "[PAD]",
|
| 30 |
+
"single_word": false,
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"id": 2,
|
| 38 |
+
"content": "[MASK]",
|
| 39 |
+
"single_word": false,
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"normalized": false,
|
| 43 |
+
"special": true
|
| 44 |
+
}
|
| 45 |
+
],
|
| 46 |
+
"normalizer": null,
|
| 47 |
+
"pre_tokenizer": {
|
| 48 |
+
"type": "WhitespaceSplit"
|
| 49 |
+
},
|
| 50 |
+
"post_processor": null,
|
| 51 |
+
"decoder": null,
|
| 52 |
+
"model": {
|
| 53 |
+
"type": "WordLevel",
|
| 54 |
+
"vocab": {
|
| 55 |
+
"[UNK]": 0,
|
| 56 |
+
"[PAD]": 1,
|
| 57 |
+
"[MASK]": 2,
|
| 58 |
+
"TIME_DELTA=1": 3,
|
| 59 |
+
"TIME_DELTA=2": 4,
|
| 60 |
+
"BAR_END": 5,
|
| 61 |
+
"BAR_START": 6,
|
| 62 |
+
"NOTE_OFF=42": 7,
|
| 63 |
+
"NOTE_ON=42": 8,
|
| 64 |
+
"NOTE_OFF=36": 9,
|
| 65 |
+
"NOTE_ON=36": 10,
|
| 66 |
+
"TIME_DELTA=4": 11,
|
| 67 |
+
"NOTE_OFF=38": 12,
|
| 68 |
+
"NOTE_ON=38": 13,
|
| 69 |
+
"NOTE_OFF=54": 14,
|
| 70 |
+
"NOTE_ON=54": 15,
|
| 71 |
+
"NOTE_OFF=62": 16,
|
| 72 |
+
"NOTE_ON=62": 17,
|
| 73 |
+
"NOTE_OFF=64": 18,
|
| 74 |
+
"NOTE_ON=64": 19,
|
| 75 |
+
"TIME_DELTA=3": 20,
|
| 76 |
+
"NOTE_OFF=57": 21,
|
| 77 |
+
"NOTE_ON=57": 22,
|
| 78 |
+
"NOTE_OFF=69": 23,
|
| 79 |
+
"NOTE_ON=69": 24,
|
| 80 |
+
"TRACK_END": 25,
|
| 81 |
+
"TRACK_START": 26,
|
| 82 |
+
"NOTE_OFF=40": 27,
|
| 83 |
+
"NOTE_ON=40": 28,
|
| 84 |
+
"NOTE_OFF=60": 29,
|
| 85 |
+
"NOTE_ON=60": 30,
|
| 86 |
+
"NOTE_OFF=35": 31,
|
| 87 |
+
"NOTE_ON=35": 32,
|
| 88 |
+
"NOTE_OFF=59": 33,
|
| 89 |
+
"NOTE_ON=59": 34,
|
| 90 |
+
"NOTE_OFF=55": 35,
|
| 91 |
+
"NOTE_ON=55": 36,
|
| 92 |
+
"NOTE_OFF=46": 37,
|
| 93 |
+
"NOTE_ON=46": 38,
|
| 94 |
+
"NOTE_OFF=67": 39,
|
| 95 |
+
"NOTE_ON=67": 40,
|
| 96 |
+
"NOTE_OFF=70": 41,
|
| 97 |
+
"NOTE_ON=70": 42,
|
| 98 |
+
"NOTE_OFF=50": 43,
|
| 99 |
+
"NOTE_ON=50": 44,
|
| 100 |
+
"NOTE_OFF=44": 45,
|
| 101 |
+
"NOTE_ON=44": 46,
|
| 102 |
+
"NOTE_OFF=52": 47,
|
| 103 |
+
"NOTE_ON=52": 48,
|
| 104 |
+
"NOTE_OFF=61": 49,
|
| 105 |
+
"NOTE_ON=61": 50,
|
| 106 |
+
"NOTE_OFF=65": 51,
|
| 107 |
+
"NOTE_ON=65": 52,
|
| 108 |
+
"NOTE_OFF=63": 53,
|
| 109 |
+
"NOTE_ON=63": 54,
|
| 110 |
+
"NOTE_OFF=66": 55,
|
| 111 |
+
"NOTE_ON=66": 56,
|
| 112 |
+
"NOTE_OFF=45": 57,
|
| 113 |
+
"NOTE_ON=45": 58,
|
| 114 |
+
"NOTE_OFF=51": 59,
|
| 115 |
+
"NOTE_ON=51": 60,
|
| 116 |
+
"NOTE_OFF=43": 61,
|
| 117 |
+
"NOTE_ON=43": 62,
|
| 118 |
+
"NOTE_OFF=48": 63,
|
| 119 |
+
"NOTE_ON=48": 64,
|
| 120 |
+
"NOTE_OFF=58": 65,
|
| 121 |
+
"NOTE_ON=58": 66,
|
| 122 |
+
"NOTE_OFF=39": 67,
|
| 123 |
+
"NOTE_ON=39": 68,
|
| 124 |
+
"NOTE_OFF=53": 69,
|
| 125 |
+
"NOTE_ON=53": 70,
|
| 126 |
+
"NOTE_OFF=56": 71,
|
| 127 |
+
"NOTE_ON=56": 72,
|
| 128 |
+
"NOTE_OFF=47": 73,
|
| 129 |
+
"NOTE_ON=47": 74,
|
| 130 |
+
"NOTE_OFF=68": 75,
|
| 131 |
+
"NOTE_ON=68": 76,
|
| 132 |
+
"NOTE_OFF=49": 77,
|
| 133 |
+
"NOTE_ON=49": 78,
|
| 134 |
+
"NOTE_OFF=72": 79,
|
| 135 |
+
"NOTE_ON=72": 80,
|
| 136 |
+
"NOTE_OFF=71": 81,
|
| 137 |
+
"NOTE_ON=71": 82,
|
| 138 |
+
"NOTE_OFF=41": 83,
|
| 139 |
+
"NOTE_ON=41": 84,
|
| 140 |
+
"NOTE_OFF=74": 85,
|
| 141 |
+
"NOTE_ON=74": 86,
|
| 142 |
+
"NOTE_OFF=33": 87,
|
| 143 |
+
"NOTE_ON=33": 88,
|
| 144 |
+
"TIME_DELTA=6": 89,
|
| 145 |
+
"NOTE_OFF=82": 90,
|
| 146 |
+
"NOTE_ON=82": 91,
|
| 147 |
+
"TIME_DELTA=16": 92,
|
| 148 |
+
"TIME_DELTA=8": 93,
|
| 149 |
+
"NOTE_OFF=37": 94,
|
| 150 |
+
"NOTE_ON=37": 95,
|
| 151 |
+
"NOTE_OFF=31": 96,
|
| 152 |
+
"NOTE_ON=31": 97,
|
| 153 |
+
"NOTE_OFF=76": 98,
|
| 154 |
+
"NOTE_ON=76": 99,
|
| 155 |
+
"DENSITY=3": 100,
|
| 156 |
+
"NOTE_OFF=73": 101,
|
| 157 |
+
"NOTE_ON=73": 102,
|
| 158 |
+
"DENSITY=0": 103,
|
| 159 |
+
"NOTE_OFF=28": 104,
|
| 160 |
+
"NOTE_ON=28": 105,
|
| 161 |
+
"DENSITY=1": 106,
|
| 162 |
+
"DENSITY=2": 107,
|
| 163 |
+
"NOTE_OFF=34": 108,
|
| 164 |
+
"NOTE_ON=34": 109,
|
| 165 |
+
"INST=3": 110,
|
| 166 |
+
"NOTE_OFF=75": 111,
|
| 167 |
+
"NOTE_ON=75": 112,
|
| 168 |
+
"NOTE_OFF=77": 113,
|
| 169 |
+
"NOTE_ON=77": 114,
|
| 170 |
+
"PIECE_START": 115,
|
| 171 |
+
"NOTE_OFF=79": 116,
|
| 172 |
+
"NOTE_ON=79": 117,
|
| 173 |
+
"INST=DRUMS": 118,
|
| 174 |
+
"NOTE_OFF=32": 119,
|
| 175 |
+
"NOTE_ON=32": 120,
|
| 176 |
+
"NOTE_OFF=29": 121,
|
| 177 |
+
"NOTE_ON=29": 122,
|
| 178 |
+
"INST=4": 123,
|
| 179 |
+
"NOTE_OFF=81": 124,
|
| 180 |
+
"NOTE_ON=81": 125,
|
| 181 |
+
"TIME_DELTA=5": 126,
|
| 182 |
+
"NOTE_OFF=78": 127,
|
| 183 |
+
"NOTE_ON=78": 128,
|
| 184 |
+
"NOTE_OFF=30": 129,
|
| 185 |
+
"NOTE_ON=30": 130,
|
| 186 |
+
"NOTE_OFF=27": 131,
|
| 187 |
+
"NOTE_ON=27": 132,
|
| 188 |
+
"INST=6": 133,
|
| 189 |
+
"NOTE_OFF=80": 134,
|
| 190 |
+
"NOTE_ON=80": 135,
|
| 191 |
+
"TIME_DELTA=7": 136,
|
| 192 |
+
"NOTE_OFF=26": 137,
|
| 193 |
+
"NOTE_ON=26": 138,
|
| 194 |
+
"INST=0": 139,
|
| 195 |
+
"NOTE_OFF=83": 140,
|
| 196 |
+
"NOTE_ON=83": 141,
|
| 197 |
+
"TIME_DELTA=12": 142,
|
| 198 |
+
"TIME_DELTA=10": 143,
|
| 199 |
+
"NOTE_OFF=84": 144,
|
| 200 |
+
"NOTE_ON=84": 145,
|
| 201 |
+
"NOTE_OFF=86": 146,
|
| 202 |
+
"NOTE_ON=86": 147,
|
| 203 |
+
"INST=10": 148,
|
| 204 |
+
"NOTE_OFF=85": 149,
|
| 205 |
+
"NOTE_ON=85": 150,
|
| 206 |
+
"TIME_DELTA=14": 151,
|
| 207 |
+
"TIME_DELTA=15": 152,
|
| 208 |
+
"NOTE_OFF=88": 153,
|
| 209 |
+
"NOTE_ON=88": 154,
|
| 210 |
+
"INST=8": 155,
|
| 211 |
+
"INST=11": 156,
|
| 212 |
+
"NOTE_OFF=87": 157,
|
| 213 |
+
"NOTE_ON=87": 158,
|
| 214 |
+
"TIME_DELTA=9": 159,
|
| 215 |
+
"NOTE_OFF=24": 160,
|
| 216 |
+
"NOTE_ON=24": 161,
|
| 217 |
+
"INST=7": 162,
|
| 218 |
+
"NOTE_OFF=25": 163,
|
| 219 |
+
"NOTE_ON=25": 164,
|
| 220 |
+
"NOTE_OFF=89": 165,
|
| 221 |
+
"NOTE_ON=89": 166,
|
| 222 |
+
"NOTE_OFF=91": 167,
|
| 223 |
+
"NOTE_ON=91": 168,
|
| 224 |
+
"TIME_DELTA=11": 169,
|
| 225 |
+
"TIME_DELTA=13": 170,
|
| 226 |
+
"INST=2": 171,
|
| 227 |
+
"NOTE_OFF=93": 172,
|
| 228 |
+
"NOTE_ON=93": 173,
|
| 229 |
+
"NOTE_OFF=22": 174,
|
| 230 |
+
"NOTE_ON=22": 175,
|
| 231 |
+
"NOTE_OFF=23": 176,
|
| 232 |
+
"NOTE_ON=23": 177,
|
| 233 |
+
"NOTE_OFF=90": 178,
|
| 234 |
+
"NOTE_ON=90": 179,
|
| 235 |
+
"INST=9": 180,
|
| 236 |
+
"INST=5": 181,
|
| 237 |
+
"INST=1": 182,
|
| 238 |
+
"NOTE_OFF=94": 183,
|
| 239 |
+
"NOTE_ON=94": 184,
|
| 240 |
+
"INST=12": 185,
|
| 241 |
+
"INST=14": 186,
|
| 242 |
+
"NOTE_OFF=92": 187,
|
| 243 |
+
"NOTE_ON=92": 188,
|
| 244 |
+
"NOTE_OFF=96": 189,
|
| 245 |
+
"NOTE_ON=96": 190,
|
| 246 |
+
"NOTE_OFF=95": 191,
|
| 247 |
+
"NOTE_ON=95": 192,
|
| 248 |
+
"NOTE_OFF=98": 193,
|
| 249 |
+
"NOTE_ON=98": 194,
|
| 250 |
+
"INST=15": 195,
|
| 251 |
+
"NOTE_OFF=21": 196,
|
| 252 |
+
"NOTE_ON=21": 197,
|
| 253 |
+
"INST=13": 198,
|
| 254 |
+
"NOTE_OFF=19": 199,
|
| 255 |
+
"NOTE_ON=19": 200,
|
| 256 |
+
"NOTE_OFF=99": 201,
|
| 257 |
+
"NOTE_ON=99": 202,
|
| 258 |
+
"NOTE_OFF=97": 203,
|
| 259 |
+
"NOTE_ON=97": 204,
|
| 260 |
+
"NOTE_OFF=0": 205,
|
| 261 |
+
"NOTE_ON=0": 206,
|
| 262 |
+
"NOTE_OFF=100": 207,
|
| 263 |
+
"NOTE_ON=100": 208,
|
| 264 |
+
"NOTE_OFF=16": 209,
|
| 265 |
+
"NOTE_ON=16": 210,
|
| 266 |
+
"NOTE_OFF=13": 211,
|
| 267 |
+
"NOTE_ON=13": 212,
|
| 268 |
+
"NOTE_OFF=20": 213,
|
| 269 |
+
"NOTE_ON=20": 214,
|
| 270 |
+
"NOTE_OFF=105": 215,
|
| 271 |
+
"NOTE_ON=105": 216,
|
| 272 |
+
"NOTE_OFF=103": 217,
|
| 273 |
+
"NOTE_ON=103": 218,
|
| 274 |
+
"NOTE_OFF=101": 219,
|
| 275 |
+
"NOTE_ON=101": 220,
|
| 276 |
+
"NOTE_OFF=102": 221,
|
| 277 |
+
"NOTE_ON=102": 222,
|
| 278 |
+
"NOTE_OFF=17": 223,
|
| 279 |
+
"NOTE_ON=17": 224,
|
| 280 |
+
"NOTE_OFF=107": 225,
|
| 281 |
+
"NOTE_ON=107": 226,
|
| 282 |
+
"NOTE_OFF=108": 227,
|
| 283 |
+
"NOTE_ON=108": 228,
|
| 284 |
+
"NOTE_OFF=18": 229,
|
| 285 |
+
"NOTE_ON=18": 230,
|
| 286 |
+
"NOTE_OFF=126": 231,
|
| 287 |
+
"NOTE_ON=126": 232,
|
| 288 |
+
"NOTE_OFF=104": 233,
|
| 289 |
+
"NOTE_ON=104": 234,
|
| 290 |
+
"NOTE_OFF=8": 235,
|
| 291 |
+
"NOTE_ON=8": 236,
|
| 292 |
+
"NOTE_OFF=117": 237,
|
| 293 |
+
"NOTE_ON=117": 238,
|
| 294 |
+
"NOTE_OFF=106": 239,
|
| 295 |
+
"NOTE_ON=106": 240,
|
| 296 |
+
"NOTE_OFF=110": 241,
|
| 297 |
+
"NOTE_ON=110": 242,
|
| 298 |
+
"NOTE_OFF=112": 243,
|
| 299 |
+
"NOTE_ON=112": 244,
|
| 300 |
+
"NOTE_OFF=12": 245,
|
| 301 |
+
"NOTE_ON=12": 246,
|
| 302 |
+
"NOTE_OFF=9": 247,
|
| 303 |
+
"NOTE_ON=9": 248,
|
| 304 |
+
"NOTE_OFF=14": 249,
|
| 305 |
+
"NOTE_ON=14": 250,
|
| 306 |
+
"NOTE_OFF=113": 251,
|
| 307 |
+
"NOTE_ON=113": 252,
|
| 308 |
+
"NOTE_OFF=15": 253,
|
| 309 |
+
"NOTE_ON=15": 254,
|
| 310 |
+
"NOTE_OFF=125": 255,
|
| 311 |
+
"NOTE_ON=125": 256,
|
| 312 |
+
"NOTE_OFF=109": 257,
|
| 313 |
+
"NOTE_ON=109": 258,
|
| 314 |
+
"NOTE_OFF=115": 259,
|
| 315 |
+
"NOTE_ON=115": 260,
|
| 316 |
+
"NOTE_OFF=120": 261,
|
| 317 |
+
"NOTE_ON=120": 262,
|
| 318 |
+
"NOTE_OFF=119": 263,
|
| 319 |
+
"NOTE_ON=119": 264,
|
| 320 |
+
"NOTE_OFF=122": 265,
|
| 321 |
+
"NOTE_ON=122": 266,
|
| 322 |
+
"NOTE_OFF=124": 267,
|
| 323 |
+
"NOTE_OFF=127": 268,
|
| 324 |
+
"NOTE_ON=124": 269,
|
| 325 |
+
"NOTE_ON=127": 270,
|
| 326 |
+
"NOTE_OFF=11": 271,
|
| 327 |
+
"NOTE_ON=11": 272,
|
| 328 |
+
"NOTE_OFF=4": 273,
|
| 329 |
+
"NOTE_ON=4": 274,
|
| 330 |
+
"NOTE_OFF=10": 275,
|
| 331 |
+
"NOTE_ON=10": 276,
|
| 332 |
+
"NOTE_OFF=111": 277,
|
| 333 |
+
"NOTE_ON=111": 278,
|
| 334 |
+
"NOTE_OFF=5": 279,
|
| 335 |
+
"NOTE_OFF=6": 280,
|
| 336 |
+
"NOTE_ON=5": 281,
|
| 337 |
+
"NOTE_ON=6": 282,
|
| 338 |
+
"NOTE_OFF=1": 283,
|
| 339 |
+
"NOTE_ON=1": 284,
|
| 340 |
+
"NOTE_OFF=114": 285,
|
| 341 |
+
"NOTE_ON=114": 286,
|
| 342 |
+
"NOTE_OFF=2": 287,
|
| 343 |
+
"NOTE_ON=2": 288,
|
| 344 |
+
"NOTE_OFF=7": 289,
|
| 345 |
+
"NOTE_ON=7": 290,
|
| 346 |
+
"NOTE_OFF=3": 291,
|
| 347 |
+
"NOTE_ON=3": 292,
|
| 348 |
+
"NOTE_OFF=116": 293,
|
| 349 |
+
"NOTE_OFF=121": 294,
|
| 350 |
+
"NOTE_ON=116": 295,
|
| 351 |
+
"NOTE_ON=121": 296,
|
| 352 |
+
"NOTE_OFF=118": 297,
|
| 353 |
+
"NOTE_ON=118": 298
|
| 354 |
+
},
|
| 355 |
+
"unk_token": "[UNK]"
|
| 356 |
+
}
|
| 357 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 3 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 4 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": null,
|
| 5 |
+
"global_step": 0,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [],
|
| 10 |
+
"max_steps": 0,
|
| 11 |
+
"num_train_epochs": 0,
|
| 12 |
+
"total_flos": 0,
|
| 13 |
+
"trial_name": null,
|
| 14 |
+
"trial_params": null
|
| 15 |
+
}
|