Instructions to use kiritan/iteboshi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use kiritan/iteboshi with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="kiritan/iteboshi")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("kiritan/iteboshi") model = AutoModelForSpeechSeq2Seq.from_pretrained("kiritan/iteboshi", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Training in progress, step 16000, checkpoint
Browse files- last-checkpoint/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt +3 -0
- last-checkpoint/global_step16000/mp_rank_00_model_states.pt +3 -0
- last-checkpoint/latest +1 -1
- last-checkpoint/model.safetensors +1 -1
- last-checkpoint/rng_state.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +585 -5
last-checkpoint/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67fa9861fa0df19ded5cc14f32aecd88316ab0ac9c839e184924551e49d53c92
|
| 3 |
+
size 761059696
|
last-checkpoint/global_step16000/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:351bf70ccb2db16c8c7c3031418826a8a84c73fcf446d779eff047ba26127504
|
| 3 |
+
size 129965712
|
last-checkpoint/latest
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
|
|
|
|
| 1 |
+
global_step16000
|
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 181508256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a370fd951ebb9e606924b2b0e0b39e8cbeab364a48c2293a4da2fe84cca15ea3
|
| 3 |
size 181508256
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf47e6f3fc3df3ec2ec720d756ada5f3fe86dd4b309a3d2c50d42b22bc6fd7f8
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99418ddc40a829db03db6cc2c954e7b03b65fc0f7c9d78bead52fa43cebbd4fe
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 96.
|
| 3 |
-
"best_model_checkpoint": "./iteboshi_temp/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -4067,6 +4067,586 @@
|
|
| 4067 |
"eval_steps_per_second": 3.199,
|
| 4068 |
"eval_wer": 96.67138142385667,
|
| 4069 |
"step": 14000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4070 |
}
|
| 4071 |
],
|
| 4072 |
"logging_steps": 25,
|
|
@@ -4086,7 +4666,7 @@
|
|
| 4086 |
"attributes": {}
|
| 4087 |
}
|
| 4088 |
},
|
| 4089 |
-
"total_flos": 2.
|
| 4090 |
"train_batch_size": 4,
|
| 4091 |
"trial_name": null,
|
| 4092 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 96.5959453088166,
|
| 3 |
+
"best_model_checkpoint": "./iteboshi_temp/checkpoint-16000",
|
| 4 |
+
"epoch": 17.621145374449338,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 16000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 4067 |
"eval_steps_per_second": 3.199,
|
| 4068 |
"eval_wer": 96.67138142385667,
|
| 4069 |
"step": 14000
|
| 4070 |
+
},
|
| 4071 |
+
{
|
| 4072 |
+
"epoch": 15.44603524229075,
|
| 4073 |
+
"grad_norm": 0.14174892008304596,
|
| 4074 |
+
"learning_rate": 6.128205128205129e-06,
|
| 4075 |
+
"loss": 0.0061,
|
| 4076 |
+
"step": 14025
|
| 4077 |
+
},
|
| 4078 |
+
{
|
| 4079 |
+
"epoch": 15.473568281938325,
|
| 4080 |
+
"grad_norm": 0.05173831433057785,
|
| 4081 |
+
"learning_rate": 6.102564102564104e-06,
|
| 4082 |
+
"loss": 0.0051,
|
| 4083 |
+
"step": 14050
|
| 4084 |
+
},
|
| 4085 |
+
{
|
| 4086 |
+
"epoch": 15.501101321585903,
|
| 4087 |
+
"grad_norm": 0.10093101114034653,
|
| 4088 |
+
"learning_rate": 6.076923076923077e-06,
|
| 4089 |
+
"loss": 0.0061,
|
| 4090 |
+
"step": 14075
|
| 4091 |
+
},
|
| 4092 |
+
{
|
| 4093 |
+
"epoch": 15.52863436123348,
|
| 4094 |
+
"grad_norm": 0.03614628687500954,
|
| 4095 |
+
"learning_rate": 6.051282051282051e-06,
|
| 4096 |
+
"loss": 0.0052,
|
| 4097 |
+
"step": 14100
|
| 4098 |
+
},
|
| 4099 |
+
{
|
| 4100 |
+
"epoch": 15.556167400881057,
|
| 4101 |
+
"grad_norm": 0.13150542974472046,
|
| 4102 |
+
"learning_rate": 6.025641025641026e-06,
|
| 4103 |
+
"loss": 0.0055,
|
| 4104 |
+
"step": 14125
|
| 4105 |
+
},
|
| 4106 |
+
{
|
| 4107 |
+
"epoch": 15.583700440528634,
|
| 4108 |
+
"grad_norm": 0.13639768958091736,
|
| 4109 |
+
"learning_rate": 6e-06,
|
| 4110 |
+
"loss": 0.0058,
|
| 4111 |
+
"step": 14150
|
| 4112 |
+
},
|
| 4113 |
+
{
|
| 4114 |
+
"epoch": 15.611233480176212,
|
| 4115 |
+
"grad_norm": 0.04362841695547104,
|
| 4116 |
+
"learning_rate": 5.974358974358975e-06,
|
| 4117 |
+
"loss": 0.0064,
|
| 4118 |
+
"step": 14175
|
| 4119 |
+
},
|
| 4120 |
+
{
|
| 4121 |
+
"epoch": 15.638766519823788,
|
| 4122 |
+
"grad_norm": 0.054506704211235046,
|
| 4123 |
+
"learning_rate": 5.948717948717949e-06,
|
| 4124 |
+
"loss": 0.0059,
|
| 4125 |
+
"step": 14200
|
| 4126 |
+
},
|
| 4127 |
+
{
|
| 4128 |
+
"epoch": 15.666299559471366,
|
| 4129 |
+
"grad_norm": 0.059661999344825745,
|
| 4130 |
+
"learning_rate": 5.923076923076924e-06,
|
| 4131 |
+
"loss": 0.0055,
|
| 4132 |
+
"step": 14225
|
| 4133 |
+
},
|
| 4134 |
+
{
|
| 4135 |
+
"epoch": 15.693832599118943,
|
| 4136 |
+
"grad_norm": 0.06359941512346268,
|
| 4137 |
+
"learning_rate": 5.897435897435898e-06,
|
| 4138 |
+
"loss": 0.0063,
|
| 4139 |
+
"step": 14250
|
| 4140 |
+
},
|
| 4141 |
+
{
|
| 4142 |
+
"epoch": 15.72136563876652,
|
| 4143 |
+
"grad_norm": 0.05899536609649658,
|
| 4144 |
+
"learning_rate": 5.871794871794873e-06,
|
| 4145 |
+
"loss": 0.0062,
|
| 4146 |
+
"step": 14275
|
| 4147 |
+
},
|
| 4148 |
+
{
|
| 4149 |
+
"epoch": 15.748898678414097,
|
| 4150 |
+
"grad_norm": 0.04062044993042946,
|
| 4151 |
+
"learning_rate": 5.846153846153847e-06,
|
| 4152 |
+
"loss": 0.006,
|
| 4153 |
+
"step": 14300
|
| 4154 |
+
},
|
| 4155 |
+
{
|
| 4156 |
+
"epoch": 15.776431718061675,
|
| 4157 |
+
"grad_norm": 0.07000496983528137,
|
| 4158 |
+
"learning_rate": 5.820512820512822e-06,
|
| 4159 |
+
"loss": 0.0057,
|
| 4160 |
+
"step": 14325
|
| 4161 |
+
},
|
| 4162 |
+
{
|
| 4163 |
+
"epoch": 15.80396475770925,
|
| 4164 |
+
"grad_norm": 0.14449462294578552,
|
| 4165 |
+
"learning_rate": 5.794871794871796e-06,
|
| 4166 |
+
"loss": 0.0061,
|
| 4167 |
+
"step": 14350
|
| 4168 |
+
},
|
| 4169 |
+
{
|
| 4170 |
+
"epoch": 15.831497797356828,
|
| 4171 |
+
"grad_norm": 0.03734823316335678,
|
| 4172 |
+
"learning_rate": 5.769230769230769e-06,
|
| 4173 |
+
"loss": 0.0055,
|
| 4174 |
+
"step": 14375
|
| 4175 |
+
},
|
| 4176 |
+
{
|
| 4177 |
+
"epoch": 15.859030837004406,
|
| 4178 |
+
"grad_norm": 0.08209270238876343,
|
| 4179 |
+
"learning_rate": 5.743589743589743e-06,
|
| 4180 |
+
"loss": 0.0059,
|
| 4181 |
+
"step": 14400
|
| 4182 |
+
},
|
| 4183 |
+
{
|
| 4184 |
+
"epoch": 15.886563876651982,
|
| 4185 |
+
"grad_norm": 0.11653455346822739,
|
| 4186 |
+
"learning_rate": 5.717948717948718e-06,
|
| 4187 |
+
"loss": 0.0059,
|
| 4188 |
+
"step": 14425
|
| 4189 |
+
},
|
| 4190 |
+
{
|
| 4191 |
+
"epoch": 15.91409691629956,
|
| 4192 |
+
"grad_norm": 0.09956145286560059,
|
| 4193 |
+
"learning_rate": 5.692307692307692e-06,
|
| 4194 |
+
"loss": 0.006,
|
| 4195 |
+
"step": 14450
|
| 4196 |
+
},
|
| 4197 |
+
{
|
| 4198 |
+
"epoch": 15.941629955947137,
|
| 4199 |
+
"grad_norm": 0.06071498617529869,
|
| 4200 |
+
"learning_rate": 5.666666666666667e-06,
|
| 4201 |
+
"loss": 0.0056,
|
| 4202 |
+
"step": 14475
|
| 4203 |
+
},
|
| 4204 |
+
{
|
| 4205 |
+
"epoch": 15.969162995594713,
|
| 4206 |
+
"grad_norm": 0.05018926039338112,
|
| 4207 |
+
"learning_rate": 5.641025641025641e-06,
|
| 4208 |
+
"loss": 0.0054,
|
| 4209 |
+
"step": 14500
|
| 4210 |
+
},
|
| 4211 |
+
{
|
| 4212 |
+
"epoch": 15.996696035242291,
|
| 4213 |
+
"grad_norm": 0.04555997997522354,
|
| 4214 |
+
"learning_rate": 5.615384615384616e-06,
|
| 4215 |
+
"loss": 0.0056,
|
| 4216 |
+
"step": 14525
|
| 4217 |
+
},
|
| 4218 |
+
{
|
| 4219 |
+
"epoch": 16.024229074889867,
|
| 4220 |
+
"grad_norm": 0.03836214169859886,
|
| 4221 |
+
"learning_rate": 5.58974358974359e-06,
|
| 4222 |
+
"loss": 0.0048,
|
| 4223 |
+
"step": 14550
|
| 4224 |
+
},
|
| 4225 |
+
{
|
| 4226 |
+
"epoch": 16.051762114537446,
|
| 4227 |
+
"grad_norm": 0.03030628338456154,
|
| 4228 |
+
"learning_rate": 5.564102564102565e-06,
|
| 4229 |
+
"loss": 0.0043,
|
| 4230 |
+
"step": 14575
|
| 4231 |
+
},
|
| 4232 |
+
{
|
| 4233 |
+
"epoch": 16.079295154185022,
|
| 4234 |
+
"grad_norm": 0.027087492868304253,
|
| 4235 |
+
"learning_rate": 5.538461538461539e-06,
|
| 4236 |
+
"loss": 0.0041,
|
| 4237 |
+
"step": 14600
|
| 4238 |
+
},
|
| 4239 |
+
{
|
| 4240 |
+
"epoch": 16.106828193832598,
|
| 4241 |
+
"grad_norm": 0.12359996885061264,
|
| 4242 |
+
"learning_rate": 5.512820512820514e-06,
|
| 4243 |
+
"loss": 0.0042,
|
| 4244 |
+
"step": 14625
|
| 4245 |
+
},
|
| 4246 |
+
{
|
| 4247 |
+
"epoch": 16.134361233480178,
|
| 4248 |
+
"grad_norm": 0.026834385469555855,
|
| 4249 |
+
"learning_rate": 5.487179487179488e-06,
|
| 4250 |
+
"loss": 0.0049,
|
| 4251 |
+
"step": 14650
|
| 4252 |
+
},
|
| 4253 |
+
{
|
| 4254 |
+
"epoch": 16.161894273127754,
|
| 4255 |
+
"grad_norm": 0.09593966603279114,
|
| 4256 |
+
"learning_rate": 5.461538461538461e-06,
|
| 4257 |
+
"loss": 0.0045,
|
| 4258 |
+
"step": 14675
|
| 4259 |
+
},
|
| 4260 |
+
{
|
| 4261 |
+
"epoch": 16.18942731277533,
|
| 4262 |
+
"grad_norm": 0.03235497698187828,
|
| 4263 |
+
"learning_rate": 5.435897435897436e-06,
|
| 4264 |
+
"loss": 0.0043,
|
| 4265 |
+
"step": 14700
|
| 4266 |
+
},
|
| 4267 |
+
{
|
| 4268 |
+
"epoch": 16.21696035242291,
|
| 4269 |
+
"grad_norm": 0.032602131366729736,
|
| 4270 |
+
"learning_rate": 5.41025641025641e-06,
|
| 4271 |
+
"loss": 0.0055,
|
| 4272 |
+
"step": 14725
|
| 4273 |
+
},
|
| 4274 |
+
{
|
| 4275 |
+
"epoch": 16.244493392070485,
|
| 4276 |
+
"grad_norm": 0.0448690690100193,
|
| 4277 |
+
"learning_rate": 5.384615384615385e-06,
|
| 4278 |
+
"loss": 0.0042,
|
| 4279 |
+
"step": 14750
|
| 4280 |
+
},
|
| 4281 |
+
{
|
| 4282 |
+
"epoch": 16.27202643171806,
|
| 4283 |
+
"grad_norm": 0.03582735359668732,
|
| 4284 |
+
"learning_rate": 5.358974358974359e-06,
|
| 4285 |
+
"loss": 0.0045,
|
| 4286 |
+
"step": 14775
|
| 4287 |
+
},
|
| 4288 |
+
{
|
| 4289 |
+
"epoch": 16.29955947136564,
|
| 4290 |
+
"grad_norm": 0.039651673287153244,
|
| 4291 |
+
"learning_rate": 5.333333333333334e-06,
|
| 4292 |
+
"loss": 0.0045,
|
| 4293 |
+
"step": 14800
|
| 4294 |
+
},
|
| 4295 |
+
{
|
| 4296 |
+
"epoch": 16.327092511013216,
|
| 4297 |
+
"grad_norm": 0.029694747179746628,
|
| 4298 |
+
"learning_rate": 5.307692307692308e-06,
|
| 4299 |
+
"loss": 0.0041,
|
| 4300 |
+
"step": 14825
|
| 4301 |
+
},
|
| 4302 |
+
{
|
| 4303 |
+
"epoch": 16.354625550660792,
|
| 4304 |
+
"grad_norm": 0.027761176228523254,
|
| 4305 |
+
"learning_rate": 5.282051282051283e-06,
|
| 4306 |
+
"loss": 0.0045,
|
| 4307 |
+
"step": 14850
|
| 4308 |
+
},
|
| 4309 |
+
{
|
| 4310 |
+
"epoch": 16.38215859030837,
|
| 4311 |
+
"grad_norm": 0.038026902824640274,
|
| 4312 |
+
"learning_rate": 5.256410256410257e-06,
|
| 4313 |
+
"loss": 0.005,
|
| 4314 |
+
"step": 14875
|
| 4315 |
+
},
|
| 4316 |
+
{
|
| 4317 |
+
"epoch": 16.409691629955947,
|
| 4318 |
+
"grad_norm": 0.04538935795426369,
|
| 4319 |
+
"learning_rate": 5.230769230769232e-06,
|
| 4320 |
+
"loss": 0.0046,
|
| 4321 |
+
"step": 14900
|
| 4322 |
+
},
|
| 4323 |
+
{
|
| 4324 |
+
"epoch": 16.437224669603523,
|
| 4325 |
+
"grad_norm": 0.0366770476102829,
|
| 4326 |
+
"learning_rate": 5.205128205128206e-06,
|
| 4327 |
+
"loss": 0.0044,
|
| 4328 |
+
"step": 14925
|
| 4329 |
+
},
|
| 4330 |
+
{
|
| 4331 |
+
"epoch": 16.464757709251103,
|
| 4332 |
+
"grad_norm": 0.04620107263326645,
|
| 4333 |
+
"learning_rate": 5.179487179487181e-06,
|
| 4334 |
+
"loss": 0.0049,
|
| 4335 |
+
"step": 14950
|
| 4336 |
+
},
|
| 4337 |
+
{
|
| 4338 |
+
"epoch": 16.49229074889868,
|
| 4339 |
+
"grad_norm": 0.026445144787430763,
|
| 4340 |
+
"learning_rate": 5.1538461538461534e-06,
|
| 4341 |
+
"loss": 0.0048,
|
| 4342 |
+
"step": 14975
|
| 4343 |
+
},
|
| 4344 |
+
{
|
| 4345 |
+
"epoch": 16.519823788546255,
|
| 4346 |
+
"grad_norm": 0.36674734950065613,
|
| 4347 |
+
"learning_rate": 5.128205128205128e-06,
|
| 4348 |
+
"loss": 0.0044,
|
| 4349 |
+
"step": 15000
|
| 4350 |
+
},
|
| 4351 |
+
{
|
| 4352 |
+
"epoch": 16.519823788546255,
|
| 4353 |
+
"eval_cer": 57.25986725207204,
|
| 4354 |
+
"eval_loss": 1.008760690689087,
|
| 4355 |
+
"eval_runtime": 842.7849,
|
| 4356 |
+
"eval_samples_per_second": 12.555,
|
| 4357 |
+
"eval_steps_per_second": 3.14,
|
| 4358 |
+
"eval_wer": 96.68081093823669,
|
| 4359 |
+
"step": 15000
|
| 4360 |
+
},
|
| 4361 |
+
{
|
| 4362 |
+
"epoch": 16.547356828193834,
|
| 4363 |
+
"grad_norm": 0.10816098749637604,
|
| 4364 |
+
"learning_rate": 5.1025641025641024e-06,
|
| 4365 |
+
"loss": 0.0046,
|
| 4366 |
+
"step": 15025
|
| 4367 |
+
},
|
| 4368 |
+
{
|
| 4369 |
+
"epoch": 16.57488986784141,
|
| 4370 |
+
"grad_norm": 0.047246623784303665,
|
| 4371 |
+
"learning_rate": 5.076923076923077e-06,
|
| 4372 |
+
"loss": 0.0042,
|
| 4373 |
+
"step": 15050
|
| 4374 |
+
},
|
| 4375 |
+
{
|
| 4376 |
+
"epoch": 16.602422907488986,
|
| 4377 |
+
"grad_norm": 0.029092887416481972,
|
| 4378 |
+
"learning_rate": 5.051282051282051e-06,
|
| 4379 |
+
"loss": 0.0047,
|
| 4380 |
+
"step": 15075
|
| 4381 |
+
},
|
| 4382 |
+
{
|
| 4383 |
+
"epoch": 16.629955947136565,
|
| 4384 |
+
"grad_norm": 0.041775159537792206,
|
| 4385 |
+
"learning_rate": 5.025641025641026e-06,
|
| 4386 |
+
"loss": 0.0039,
|
| 4387 |
+
"step": 15100
|
| 4388 |
+
},
|
| 4389 |
+
{
|
| 4390 |
+
"epoch": 16.65748898678414,
|
| 4391 |
+
"grad_norm": 0.04027463123202324,
|
| 4392 |
+
"learning_rate": 5e-06,
|
| 4393 |
+
"loss": 0.0046,
|
| 4394 |
+
"step": 15125
|
| 4395 |
+
},
|
| 4396 |
+
{
|
| 4397 |
+
"epoch": 16.685022026431717,
|
| 4398 |
+
"grad_norm": 0.025997977703809738,
|
| 4399 |
+
"learning_rate": 4.974358974358975e-06,
|
| 4400 |
+
"loss": 0.0039,
|
| 4401 |
+
"step": 15150
|
| 4402 |
+
},
|
| 4403 |
+
{
|
| 4404 |
+
"epoch": 16.712555066079297,
|
| 4405 |
+
"grad_norm": 0.038961417973041534,
|
| 4406 |
+
"learning_rate": 4.948717948717949e-06,
|
| 4407 |
+
"loss": 0.0043,
|
| 4408 |
+
"step": 15175
|
| 4409 |
+
},
|
| 4410 |
+
{
|
| 4411 |
+
"epoch": 16.740088105726873,
|
| 4412 |
+
"grad_norm": 0.06082303076982498,
|
| 4413 |
+
"learning_rate": 4.923076923076924e-06,
|
| 4414 |
+
"loss": 0.0048,
|
| 4415 |
+
"step": 15200
|
| 4416 |
+
},
|
| 4417 |
+
{
|
| 4418 |
+
"epoch": 16.76762114537445,
|
| 4419 |
+
"grad_norm": 0.027074238285422325,
|
| 4420 |
+
"learning_rate": 4.8974358974358975e-06,
|
| 4421 |
+
"loss": 0.0046,
|
| 4422 |
+
"step": 15225
|
| 4423 |
+
},
|
| 4424 |
+
{
|
| 4425 |
+
"epoch": 16.795154185022028,
|
| 4426 |
+
"grad_norm": 0.03542567789554596,
|
| 4427 |
+
"learning_rate": 4.871794871794872e-06,
|
| 4428 |
+
"loss": 0.0045,
|
| 4429 |
+
"step": 15250
|
| 4430 |
+
},
|
| 4431 |
+
{
|
| 4432 |
+
"epoch": 16.822687224669604,
|
| 4433 |
+
"grad_norm": 0.04304325580596924,
|
| 4434 |
+
"learning_rate": 4.8461538461538465e-06,
|
| 4435 |
+
"loss": 0.0044,
|
| 4436 |
+
"step": 15275
|
| 4437 |
+
},
|
| 4438 |
+
{
|
| 4439 |
+
"epoch": 16.85022026431718,
|
| 4440 |
+
"grad_norm": 0.08053645491600037,
|
| 4441 |
+
"learning_rate": 4.820512820512821e-06,
|
| 4442 |
+
"loss": 0.0045,
|
| 4443 |
+
"step": 15300
|
| 4444 |
+
},
|
| 4445 |
+
{
|
| 4446 |
+
"epoch": 16.87775330396476,
|
| 4447 |
+
"grad_norm": 0.0946226641535759,
|
| 4448 |
+
"learning_rate": 4.7948717948717955e-06,
|
| 4449 |
+
"loss": 0.0064,
|
| 4450 |
+
"step": 15325
|
| 4451 |
+
},
|
| 4452 |
+
{
|
| 4453 |
+
"epoch": 16.905286343612335,
|
| 4454 |
+
"grad_norm": 0.03868434950709343,
|
| 4455 |
+
"learning_rate": 4.76923076923077e-06,
|
| 4456 |
+
"loss": 0.0042,
|
| 4457 |
+
"step": 15350
|
| 4458 |
+
},
|
| 4459 |
+
{
|
| 4460 |
+
"epoch": 16.93281938325991,
|
| 4461 |
+
"grad_norm": 0.07756806910037994,
|
| 4462 |
+
"learning_rate": 4.743589743589744e-06,
|
| 4463 |
+
"loss": 0.0041,
|
| 4464 |
+
"step": 15375
|
| 4465 |
+
},
|
| 4466 |
+
{
|
| 4467 |
+
"epoch": 16.96035242290749,
|
| 4468 |
+
"grad_norm": 0.03353738412261009,
|
| 4469 |
+
"learning_rate": 4.717948717948718e-06,
|
| 4470 |
+
"loss": 0.0048,
|
| 4471 |
+
"step": 15400
|
| 4472 |
+
},
|
| 4473 |
+
{
|
| 4474 |
+
"epoch": 16.987885462555067,
|
| 4475 |
+
"grad_norm": 0.12302552163600922,
|
| 4476 |
+
"learning_rate": 4.692307692307693e-06,
|
| 4477 |
+
"loss": 0.0048,
|
| 4478 |
+
"step": 15425
|
| 4479 |
+
},
|
| 4480 |
+
{
|
| 4481 |
+
"epoch": 17.015418502202643,
|
| 4482 |
+
"grad_norm": 0.040505509823560715,
|
| 4483 |
+
"learning_rate": 4.666666666666667e-06,
|
| 4484 |
+
"loss": 0.004,
|
| 4485 |
+
"step": 15450
|
| 4486 |
+
},
|
| 4487 |
+
{
|
| 4488 |
+
"epoch": 17.042951541850222,
|
| 4489 |
+
"grad_norm": 0.0250072181224823,
|
| 4490 |
+
"learning_rate": 4.641025641025642e-06,
|
| 4491 |
+
"loss": 0.0048,
|
| 4492 |
+
"step": 15475
|
| 4493 |
+
},
|
| 4494 |
+
{
|
| 4495 |
+
"epoch": 17.070484581497798,
|
| 4496 |
+
"grad_norm": 0.027064405381679535,
|
| 4497 |
+
"learning_rate": 4.615384615384616e-06,
|
| 4498 |
+
"loss": 0.0029,
|
| 4499 |
+
"step": 15500
|
| 4500 |
+
},
|
| 4501 |
+
{
|
| 4502 |
+
"epoch": 17.098017621145374,
|
| 4503 |
+
"grad_norm": 0.021600468084216118,
|
| 4504 |
+
"learning_rate": 4.58974358974359e-06,
|
| 4505 |
+
"loss": 0.0033,
|
| 4506 |
+
"step": 15525
|
| 4507 |
+
},
|
| 4508 |
+
{
|
| 4509 |
+
"epoch": 17.125550660792953,
|
| 4510 |
+
"grad_norm": 0.024508068338036537,
|
| 4511 |
+
"learning_rate": 4.564102564102564e-06,
|
| 4512 |
+
"loss": 0.0033,
|
| 4513 |
+
"step": 15550
|
| 4514 |
+
},
|
| 4515 |
+
{
|
| 4516 |
+
"epoch": 17.15308370044053,
|
| 4517 |
+
"grad_norm": 0.05471380800008774,
|
| 4518 |
+
"learning_rate": 4.538461538461539e-06,
|
| 4519 |
+
"loss": 0.0042,
|
| 4520 |
+
"step": 15575
|
| 4521 |
+
},
|
| 4522 |
+
{
|
| 4523 |
+
"epoch": 17.180616740088105,
|
| 4524 |
+
"grad_norm": 0.019899729639291763,
|
| 4525 |
+
"learning_rate": 4.512820512820513e-06,
|
| 4526 |
+
"loss": 0.003,
|
| 4527 |
+
"step": 15600
|
| 4528 |
+
},
|
| 4529 |
+
{
|
| 4530 |
+
"epoch": 17.208149779735685,
|
| 4531 |
+
"grad_norm": 0.03066374734044075,
|
| 4532 |
+
"learning_rate": 4.487179487179488e-06,
|
| 4533 |
+
"loss": 0.0038,
|
| 4534 |
+
"step": 15625
|
| 4535 |
+
},
|
| 4536 |
+
{
|
| 4537 |
+
"epoch": 17.23568281938326,
|
| 4538 |
+
"grad_norm": 0.043680623173713684,
|
| 4539 |
+
"learning_rate": 4.461538461538462e-06,
|
| 4540 |
+
"loss": 0.0041,
|
| 4541 |
+
"step": 15650
|
| 4542 |
+
},
|
| 4543 |
+
{
|
| 4544 |
+
"epoch": 17.263215859030836,
|
| 4545 |
+
"grad_norm": 0.03351521119475365,
|
| 4546 |
+
"learning_rate": 4.435897435897436e-06,
|
| 4547 |
+
"loss": 0.0039,
|
| 4548 |
+
"step": 15675
|
| 4549 |
+
},
|
| 4550 |
+
{
|
| 4551 |
+
"epoch": 17.290748898678412,
|
| 4552 |
+
"grad_norm": 0.18343234062194824,
|
| 4553 |
+
"learning_rate": 4.4102564102564104e-06,
|
| 4554 |
+
"loss": 0.0032,
|
| 4555 |
+
"step": 15700
|
| 4556 |
+
},
|
| 4557 |
+
{
|
| 4558 |
+
"epoch": 17.318281938325992,
|
| 4559 |
+
"grad_norm": 0.10387376695871353,
|
| 4560 |
+
"learning_rate": 4.384615384615385e-06,
|
| 4561 |
+
"loss": 0.0033,
|
| 4562 |
+
"step": 15725
|
| 4563 |
+
},
|
| 4564 |
+
{
|
| 4565 |
+
"epoch": 17.345814977973568,
|
| 4566 |
+
"grad_norm": 0.02336566336452961,
|
| 4567 |
+
"learning_rate": 4.358974358974359e-06,
|
| 4568 |
+
"loss": 0.0042,
|
| 4569 |
+
"step": 15750
|
| 4570 |
+
},
|
| 4571 |
+
{
|
| 4572 |
+
"epoch": 17.373348017621144,
|
| 4573 |
+
"grad_norm": 0.022074054926633835,
|
| 4574 |
+
"learning_rate": 4.333333333333334e-06,
|
| 4575 |
+
"loss": 0.0032,
|
| 4576 |
+
"step": 15775
|
| 4577 |
+
},
|
| 4578 |
+
{
|
| 4579 |
+
"epoch": 17.400881057268723,
|
| 4580 |
+
"grad_norm": 0.035601481795310974,
|
| 4581 |
+
"learning_rate": 4.307692307692308e-06,
|
| 4582 |
+
"loss": 0.003,
|
| 4583 |
+
"step": 15800
|
| 4584 |
+
},
|
| 4585 |
+
{
|
| 4586 |
+
"epoch": 17.4284140969163,
|
| 4587 |
+
"grad_norm": 0.028988130390644073,
|
| 4588 |
+
"learning_rate": 4.282051282051282e-06,
|
| 4589 |
+
"loss": 0.0034,
|
| 4590 |
+
"step": 15825
|
| 4591 |
+
},
|
| 4592 |
+
{
|
| 4593 |
+
"epoch": 17.455947136563875,
|
| 4594 |
+
"grad_norm": 0.037720389664173126,
|
| 4595 |
+
"learning_rate": 4.2564102564102566e-06,
|
| 4596 |
+
"loss": 0.0041,
|
| 4597 |
+
"step": 15850
|
| 4598 |
+
},
|
| 4599 |
+
{
|
| 4600 |
+
"epoch": 17.483480176211454,
|
| 4601 |
+
"grad_norm": 0.038582686334848404,
|
| 4602 |
+
"learning_rate": 4.230769230769231e-06,
|
| 4603 |
+
"loss": 0.0042,
|
| 4604 |
+
"step": 15875
|
| 4605 |
+
},
|
| 4606 |
+
{
|
| 4607 |
+
"epoch": 17.51101321585903,
|
| 4608 |
+
"grad_norm": 0.032543014734983444,
|
| 4609 |
+
"learning_rate": 4.2051282051282055e-06,
|
| 4610 |
+
"loss": 0.0046,
|
| 4611 |
+
"step": 15900
|
| 4612 |
+
},
|
| 4613 |
+
{
|
| 4614 |
+
"epoch": 17.538546255506606,
|
| 4615 |
+
"grad_norm": 0.033068541437387466,
|
| 4616 |
+
"learning_rate": 4.17948717948718e-06,
|
| 4617 |
+
"loss": 0.0037,
|
| 4618 |
+
"step": 15925
|
| 4619 |
+
},
|
| 4620 |
+
{
|
| 4621 |
+
"epoch": 17.566079295154186,
|
| 4622 |
+
"grad_norm": 0.0271660964936018,
|
| 4623 |
+
"learning_rate": 4.1538461538461545e-06,
|
| 4624 |
+
"loss": 0.0041,
|
| 4625 |
+
"step": 15950
|
| 4626 |
+
},
|
| 4627 |
+
{
|
| 4628 |
+
"epoch": 17.59361233480176,
|
| 4629 |
+
"grad_norm": 0.03645619750022888,
|
| 4630 |
+
"learning_rate": 4.128205128205128e-06,
|
| 4631 |
+
"loss": 0.0039,
|
| 4632 |
+
"step": 15975
|
| 4633 |
+
},
|
| 4634 |
+
{
|
| 4635 |
+
"epoch": 17.621145374449338,
|
| 4636 |
+
"grad_norm": 0.033829644322395325,
|
| 4637 |
+
"learning_rate": 4.102564102564103e-06,
|
| 4638 |
+
"loss": 0.0034,
|
| 4639 |
+
"step": 16000
|
| 4640 |
+
},
|
| 4641 |
+
{
|
| 4642 |
+
"epoch": 17.621145374449338,
|
| 4643 |
+
"eval_cer": 55.244001696605636,
|
| 4644 |
+
"eval_loss": 1.0241984128952026,
|
| 4645 |
+
"eval_runtime": 821.5561,
|
| 4646 |
+
"eval_samples_per_second": 12.879,
|
| 4647 |
+
"eval_steps_per_second": 3.221,
|
| 4648 |
+
"eval_wer": 96.5959453088166,
|
| 4649 |
+
"step": 16000
|
| 4650 |
}
|
| 4651 |
],
|
| 4652 |
"logging_steps": 25,
|
|
|
|
| 4666 |
"attributes": {}
|
| 4667 |
}
|
| 4668 |
},
|
| 4669 |
+
"total_flos": 2.701073184056148e+19,
|
| 4670 |
"train_batch_size": 4,
|
| 4671 |
"trial_name": null,
|
| 4672 |
"trial_params": null
|