Upload folder using huggingface_hub
Browse files- checkpoints-v3.0-discrete-conditional/checkpoint-7936/model.safetensors +3 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/optimizer.pt +3 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/rng_state.pth +3 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/scaler.pt +3 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/scheduler.pt +3 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/trainer_state.json +995 -0
- checkpoints-v3.0-discrete-conditional/checkpoint-7936/training_args.bin +3 -0
checkpoints-v3.0-discrete-conditional/checkpoint-7936/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e887f1b6e879d16aeeb945ac33cb90023f09063ee9794f96a646dfddd37a7cad
|
| 3 |
+
size 24272320
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:128a1fddbb7b77e2d5768ecdbfd209a56dc26dc9c1c229df5be3fb2e35853375
|
| 3 |
+
size 519947
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0b4c40d315b7b8d9f17bcfcf4edb1280a4a07504377c1268d285373fb8811ca
|
| 3 |
+
size 14645
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1dd6dea7a37283e26a2e3ea2e85617aaaf31092e6ed791a6d2f45d466c879233
|
| 3 |
+
size 1383
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e1898551e4594b5be5f763f4c93d36a51e8cefc331809d6bf24c41b291641ab
|
| 3 |
+
size 1465
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/trainer_state.json
ADDED
|
@@ -0,0 +1,995 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.9190123456790125,
|
| 6 |
+
"eval_steps": 256,
|
| 7 |
+
"global_step": 7936,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.06320987654320988,
|
| 14 |
+
"grad_norm": 0.23172101378440857,
|
| 15 |
+
"learning_rate": 0.000248046875,
|
| 16 |
+
"loss": 0.7997208833694458,
|
| 17 |
+
"step": 128
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.12641975308641976,
|
| 21 |
+
"grad_norm": 0.21966563165187836,
|
| 22 |
+
"learning_rate": 0.000498046875,
|
| 23 |
+
"loss": 0.7981692552566528,
|
| 24 |
+
"step": 256
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.12641975308641976,
|
| 28 |
+
"eval_cos_loss": 0.2296824600638413,
|
| 29 |
+
"eval_loss": 0.7970464796554751,
|
| 30 |
+
"eval_mse_loss": 0.7740782310323018,
|
| 31 |
+
"step": 256
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.12641975308641976,
|
| 35 |
+
"eval_cos_loss": 0.2296824600638413,
|
| 36 |
+
"eval_loss": 0.7970464796554751,
|
| 37 |
+
"eval_mse_loss": 0.7740782310323018,
|
| 38 |
+
"eval_runtime": 3.972,
|
| 39 |
+
"eval_samples_per_second": 659.115,
|
| 40 |
+
"eval_steps_per_second": 10.322,
|
| 41 |
+
"step": 256
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.18962962962962962,
|
| 45 |
+
"grad_norm": 0.2943345010280609,
|
| 46 |
+
"learning_rate": 0.000748046875,
|
| 47 |
+
"loss": 0.7972801327705383,
|
| 48 |
+
"step": 384
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.2528395061728395,
|
| 52 |
+
"grad_norm": 0.2626311182975769,
|
| 53 |
+
"learning_rate": 0.000998046875,
|
| 54 |
+
"loss": 0.7905202507972717,
|
| 55 |
+
"step": 512
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.2528395061728395,
|
| 59 |
+
"eval_cos_loss": 0.22880334061820332,
|
| 60 |
+
"eval_loss": 0.7939348089985732,
|
| 61 |
+
"eval_mse_loss": 0.7710544714113561,
|
| 62 |
+
"step": 512
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.2528395061728395,
|
| 66 |
+
"eval_cos_loss": 0.22880334061820332,
|
| 67 |
+
"eval_loss": 0.7939348089985732,
|
| 68 |
+
"eval_mse_loss": 0.7710544714113561,
|
| 69 |
+
"eval_runtime": 3.8814,
|
| 70 |
+
"eval_samples_per_second": 674.492,
|
| 71 |
+
"eval_steps_per_second": 10.563,
|
| 72 |
+
"step": 512
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3160493827160494,
|
| 76 |
+
"grad_norm": 0.22872641682624817,
|
| 77 |
+
"learning_rate": 0.0009993089770195807,
|
| 78 |
+
"loss": 0.8048543334007263,
|
| 79 |
+
"step": 640
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.37925925925925924,
|
| 83 |
+
"grad_norm": 0.25428205728530884,
|
| 84 |
+
"learning_rate": 0.0009972160460972733,
|
| 85 |
+
"loss": 0.797626793384552,
|
| 86 |
+
"step": 768
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.37925925925925924,
|
| 90 |
+
"eval_cos_loss": 0.23112982111733135,
|
| 91 |
+
"eval_loss": 0.8032825109435291,
|
| 92 |
+
"eval_mse_loss": 0.7801695291588946,
|
| 93 |
+
"step": 768
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.37925925925925924,
|
| 97 |
+
"eval_cos_loss": 0.23112982111733135,
|
| 98 |
+
"eval_loss": 0.8032825109435291,
|
| 99 |
+
"eval_mse_loss": 0.7801695291588946,
|
| 100 |
+
"eval_runtime": 3.0002,
|
| 101 |
+
"eval_samples_per_second": 872.605,
|
| 102 |
+
"eval_steps_per_second": 13.666,
|
| 103 |
+
"step": 768
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.44246913580246916,
|
| 107 |
+
"grad_norm": 0.28274211287498474,
|
| 108 |
+
"learning_rate": 0.0009937270408736224,
|
| 109 |
+
"loss": 0.793694794178009,
|
| 110 |
+
"step": 896
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.505679012345679,
|
| 114 |
+
"grad_norm": 0.24109429121017456,
|
| 115 |
+
"learning_rate": 0.0009888517577149526,
|
| 116 |
+
"loss": 0.7927389740943909,
|
| 117 |
+
"step": 1024
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.505679012345679,
|
| 121 |
+
"eval_cos_loss": 0.22725565804214012,
|
| 122 |
+
"eval_loss": 0.7899522345240523,
|
| 123 |
+
"eval_mse_loss": 0.7672266640314241,
|
| 124 |
+
"step": 1024
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.505679012345679,
|
| 128 |
+
"eval_cos_loss": 0.22725565804214012,
|
| 129 |
+
"eval_loss": 0.7899522345240523,
|
| 130 |
+
"eval_mse_loss": 0.7672266640314241,
|
| 131 |
+
"eval_runtime": 2.9714,
|
| 132 |
+
"eval_samples_per_second": 881.052,
|
| 133 |
+
"eval_steps_per_second": 13.798,
|
| 134 |
+
"step": 1024
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.5688888888888889,
|
| 138 |
+
"grad_norm": 0.20911817252635956,
|
| 139 |
+
"learning_rate": 0.0009826038853539248,
|
| 140 |
+
"loss": 0.7938523292541504,
|
| 141 |
+
"step": 1152
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.6320987654320988,
|
| 145 |
+
"grad_norm": 0.4241364896297455,
|
| 146 |
+
"learning_rate": 0.0009750009664545572,
|
| 147 |
+
"loss": 0.7987790107727051,
|
| 148 |
+
"step": 1280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.6320987654320988,
|
| 152 |
+
"eval_cos_loss": 0.22850282671974925,
|
| 153 |
+
"eval_loss": 0.7918573341718534,
|
| 154 |
+
"eval_mse_loss": 0.7690070562246369,
|
| 155 |
+
"step": 1280
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.6320987654320988,
|
| 159 |
+
"eval_cos_loss": 0.22850282671974925,
|
| 160 |
+
"eval_loss": 0.7918573341718534,
|
| 161 |
+
"eval_mse_loss": 0.7690070562246369,
|
| 162 |
+
"eval_runtime": 2.9523,
|
| 163 |
+
"eval_samples_per_second": 886.781,
|
| 164 |
+
"eval_steps_per_second": 13.888,
|
| 165 |
+
"step": 1280
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.6953086419753086,
|
| 169 |
+
"grad_norm": 0.1920246183872223,
|
| 170 |
+
"learning_rate": 0.0009660643483562486,
|
| 171 |
+
"loss": 0.7907214760780334,
|
| 172 |
+
"step": 1408
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.7585185185185185,
|
| 176 |
+
"grad_norm": 0.2991013526916504,
|
| 177 |
+
"learning_rate": 0.0009558191231351013,
|
| 178 |
+
"loss": 0.795460045337677,
|
| 179 |
+
"step": 1536
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.7585185185185185,
|
| 183 |
+
"eval_cos_loss": 0.22320823516787552,
|
| 184 |
+
"eval_loss": 0.780292726144558,
|
| 185 |
+
"eval_mse_loss": 0.7579719002653913,
|
| 186 |
+
"step": 1536
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.7585185185185185,
|
| 190 |
+
"eval_cos_loss": 0.22320823516787552,
|
| 191 |
+
"eval_loss": 0.780292726144558,
|
| 192 |
+
"eval_mse_loss": 0.7579719002653913,
|
| 193 |
+
"eval_runtime": 2.8726,
|
| 194 |
+
"eval_samples_per_second": 911.377,
|
| 195 |
+
"eval_steps_per_second": 14.273,
|
| 196 |
+
"step": 1536
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.8217283950617283,
|
| 200 |
+
"grad_norm": 0.2981093227863312,
|
| 201 |
+
"learning_rate": 0.0009442940571508399,
|
| 202 |
+
"loss": 0.7922792434692383,
|
| 203 |
+
"step": 1664
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.8849382716049383,
|
| 207 |
+
"grad_norm": 0.3410004675388336,
|
| 208 |
+
"learning_rate": 0.0009315215102771411,
|
| 209 |
+
"loss": 0.7970292568206787,
|
| 210 |
+
"step": 1792
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.8849382716049383,
|
| 214 |
+
"eval_cos_loss": 0.2277334343369414,
|
| 215 |
+
"eval_loss": 0.7910735476307753,
|
| 216 |
+
"eval_mse_loss": 0.7683002018347019,
|
| 217 |
+
"step": 1792
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.8849382716049383,
|
| 221 |
+
"eval_cos_loss": 0.2277334343369414,
|
| 222 |
+
"eval_loss": 0.7910735476307753,
|
| 223 |
+
"eval_mse_loss": 0.7683002018347019,
|
| 224 |
+
"eval_runtime": 2.8417,
|
| 225 |
+
"eval_samples_per_second": 921.294,
|
| 226 |
+
"eval_steps_per_second": 14.428,
|
| 227 |
+
"step": 1792
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.9481481481481482,
|
| 231 |
+
"grad_norm": 0.21089138090610504,
|
| 232 |
+
"learning_rate": 0.0009175373450421618,
|
| 233 |
+
"loss": 0.791707456111908,
|
| 234 |
+
"step": 1920
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.011358024691358,
|
| 238 |
+
"grad_norm": 0.28280070424079895,
|
| 239 |
+
"learning_rate": 0.0009023808259343743,
|
| 240 |
+
"loss": 0.7900460958480835,
|
| 241 |
+
"step": 2048
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.011358024691358,
|
| 245 |
+
"eval_cos_loss": 0.22489730068823185,
|
| 246 |
+
"eval_loss": 0.7806248490403338,
|
| 247 |
+
"eval_mse_loss": 0.7581351210431355,
|
| 248 |
+
"step": 2048
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.011358024691358,
|
| 252 |
+
"eval_cos_loss": 0.22489730068823185,
|
| 253 |
+
"eval_loss": 0.7806248490403338,
|
| 254 |
+
"eval_mse_loss": 0.7581351210431355,
|
| 255 |
+
"eval_runtime": 2.9074,
|
| 256 |
+
"eval_samples_per_second": 900.46,
|
| 257 |
+
"eval_steps_per_second": 14.102,
|
| 258 |
+
"step": 2048
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.074567901234568,
|
| 262 |
+
"grad_norm": 0.2784762382507324,
|
| 263 |
+
"learning_rate": 0.0008860945091564399,
|
| 264 |
+
"loss": 0.79193115234375,
|
| 265 |
+
"step": 2176
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.1377777777777778,
|
| 269 |
+
"grad_norm": 0.1766098588705063,
|
| 270 |
+
"learning_rate": 0.0008687241231366662,
|
| 271 |
+
"loss": 0.7897006273269653,
|
| 272 |
+
"step": 2304
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.1377777777777778,
|
| 276 |
+
"eval_cos_loss": 0.2275779985073136,
|
| 277 |
+
"eval_loss": 0.7958266517011131,
|
| 278 |
+
"eval_mse_loss": 0.7730688496333796,
|
| 279 |
+
"step": 2304
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 1.1377777777777778,
|
| 283 |
+
"eval_cos_loss": 0.2275779985073136,
|
| 284 |
+
"eval_loss": 0.7958266517011131,
|
| 285 |
+
"eval_mse_loss": 0.7730688496333796,
|
| 286 |
+
"eval_runtime": 2.8954,
|
| 287 |
+
"eval_samples_per_second": 904.193,
|
| 288 |
+
"eval_steps_per_second": 14.16,
|
| 289 |
+
"step": 2304
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.2009876543209876,
|
| 293 |
+
"grad_norm": 0.19487471878528595,
|
| 294 |
+
"learning_rate": 0.0008503184401335448,
|
| 295 |
+
"loss": 0.7908310294151306,
|
| 296 |
+
"step": 2432
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.2641975308641975,
|
| 300 |
+
"grad_norm": 0.2605104446411133,
|
| 301 |
+
"learning_rate": 0.0008309291392938795,
|
| 302 |
+
"loss": 0.7922152280807495,
|
| 303 |
+
"step": 2560
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.2641975308641975,
|
| 307 |
+
"eval_cos_loss": 0.22808007823257911,
|
| 308 |
+
"eval_loss": 0.7948527496035506,
|
| 309 |
+
"eval_mse_loss": 0.7720447400721108,
|
| 310 |
+
"step": 2560
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.2641975308641975,
|
| 314 |
+
"eval_cos_loss": 0.22808007823257911,
|
| 315 |
+
"eval_loss": 0.7948527496035506,
|
| 316 |
+
"eval_mse_loss": 0.7720447400721108,
|
| 317 |
+
"eval_runtime": 3.0225,
|
| 318 |
+
"eval_samples_per_second": 866.174,
|
| 319 |
+
"eval_steps_per_second": 13.565,
|
| 320 |
+
"step": 2560
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 1.3274074074074074,
|
| 324 |
+
"grad_norm": 0.18549929559230804,
|
| 325 |
+
"learning_rate": 0.0008106106615490032,
|
| 326 |
+
"loss": 0.7880870699882507,
|
| 327 |
+
"step": 2688
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.3906172839506172,
|
| 331 |
+
"grad_norm": 0.19430062174797058,
|
| 332 |
+
"learning_rate": 0.0007894200567565075,
|
| 333 |
+
"loss": 0.7790378332138062,
|
| 334 |
+
"step": 2816
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.3906172839506172,
|
| 338 |
+
"eval_cos_loss": 0.22629769473541073,
|
| 339 |
+
"eval_loss": 0.7880184243365032,
|
| 340 |
+
"eval_mse_loss": 0.765388655953291,
|
| 341 |
+
"step": 2816
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.3906172839506172,
|
| 345 |
+
"eval_cos_loss": 0.22629769473541073,
|
| 346 |
+
"eval_loss": 0.7880184243365032,
|
| 347 |
+
"eval_mse_loss": 0.765388655953291,
|
| 348 |
+
"eval_runtime": 2.9958,
|
| 349 |
+
"eval_samples_per_second": 873.889,
|
| 350 |
+
"eval_steps_per_second": 13.686,
|
| 351 |
+
"step": 2816
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.453827160493827,
|
| 355 |
+
"grad_norm": 0.193603977560997,
|
| 356 |
+
"learning_rate": 0.0007674168235166747,
|
| 357 |
+
"loss": 0.7834969758987427,
|
| 358 |
+
"step": 2944
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.5170370370370372,
|
| 362 |
+
"grad_norm": 0.25885820388793945,
|
| 363 |
+
"learning_rate": 0.000744662742113375,
|
| 364 |
+
"loss": 0.7839168906211853,
|
| 365 |
+
"step": 3072
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.5170370370370372,
|
| 369 |
+
"eval_cos_loss": 0.22938175281373466,
|
| 370 |
+
"eval_loss": 0.7993329240054619,
|
| 371 |
+
"eval_mse_loss": 0.7763947539213227,
|
| 372 |
+
"step": 3072
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.5170370370370372,
|
| 376 |
+
"eval_cos_loss": 0.22938175281373466,
|
| 377 |
+
"eval_loss": 0.7993329240054619,
|
| 378 |
+
"eval_mse_loss": 0.7763947539213227,
|
| 379 |
+
"eval_runtime": 3.8739,
|
| 380 |
+
"eval_samples_per_second": 675.802,
|
| 381 |
+
"eval_steps_per_second": 10.584,
|
| 382 |
+
"step": 3072
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.5802469135802468,
|
| 386 |
+
"grad_norm": 0.31186336278915405,
|
| 387 |
+
"learning_rate": 0.0007212217010484917,
|
| 388 |
+
"loss": 0.7857620716094971,
|
| 389 |
+
"step": 3200
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.643456790123457,
|
| 393 |
+
"grad_norm": 0.2609359323978424,
|
| 394 |
+
"learning_rate": 0.0006971595176569332,
|
| 395 |
+
"loss": 0.7916860580444336,
|
| 396 |
+
"step": 3328
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.643456790123457,
|
| 400 |
+
"eval_cos_loss": 0.22667416457722828,
|
| 401 |
+
"eval_loss": 0.7874437686873645,
|
| 402 |
+
"eval_mse_loss": 0.7647763519752316,
|
| 403 |
+
"step": 3328
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.643456790123457,
|
| 407 |
+
"eval_cos_loss": 0.22667416457722828,
|
| 408 |
+
"eval_loss": 0.7874437686873645,
|
| 409 |
+
"eval_mse_loss": 0.7647763519752316,
|
| 410 |
+
"eval_runtime": 3.9867,
|
| 411 |
+
"eval_samples_per_second": 656.685,
|
| 412 |
+
"eval_steps_per_second": 10.284,
|
| 413 |
+
"step": 3328
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.7066666666666666,
|
| 417 |
+
"grad_norm": 0.2544788718223572,
|
| 418 |
+
"learning_rate": 0.0006725437533058971,
|
| 419 |
+
"loss": 0.7838412523269653,
|
| 420 |
+
"step": 3456
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.7698765432098766,
|
| 424 |
+
"grad_norm": 0.14100751280784607,
|
| 425 |
+
"learning_rate": 0.0006474435236972767,
|
| 426 |
+
"loss": 0.7874654531478882,
|
| 427 |
+
"step": 3584
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.7698765432098766,
|
| 431 |
+
"eval_cos_loss": 0.22428179041641513,
|
| 432 |
+
"eval_loss": 0.7828184845970898,
|
| 433 |
+
"eval_mse_loss": 0.760390307845139,
|
| 434 |
+
"step": 3584
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.7698765432098766,
|
| 438 |
+
"eval_cos_loss": 0.22428179041641513,
|
| 439 |
+
"eval_loss": 0.7828184845970898,
|
| 440 |
+
"eval_mse_loss": 0.760390307845139,
|
| 441 |
+
"eval_runtime": 3.4413,
|
| 442 |
+
"eval_samples_per_second": 760.764,
|
| 443 |
+
"eval_steps_per_second": 11.914,
|
| 444 |
+
"step": 3584
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.8330864197530863,
|
| 448 |
+
"grad_norm": 0.19871363043785095,
|
| 449 |
+
"learning_rate": 0.0006219293048058301,
|
| 450 |
+
"loss": 0.7843989729881287,
|
| 451 |
+
"step": 3712
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 1.8962962962962964,
|
| 455 |
+
"grad_norm": 0.1531391590833664,
|
| 456 |
+
"learning_rate": 0.0005960727349980042,
|
| 457 |
+
"loss": 0.7820008993148804,
|
| 458 |
+
"step": 3840
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.8962962962962964,
|
| 462 |
+
"eval_cos_loss": 0.2254106565946486,
|
| 463 |
+
"eval_loss": 0.7854831189644046,
|
| 464 |
+
"eval_mse_loss": 0.7629420510152491,
|
| 465 |
+
"step": 3840
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.8962962962962964,
|
| 469 |
+
"eval_cos_loss": 0.2254106565946486,
|
| 470 |
+
"eval_loss": 0.7854831189644046,
|
| 471 |
+
"eval_mse_loss": 0.7629420510152491,
|
| 472 |
+
"eval_runtime": 3.0265,
|
| 473 |
+
"eval_samples_per_second": 865.028,
|
| 474 |
+
"eval_steps_per_second": 13.547,
|
| 475 |
+
"step": 3840
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 1.959506172839506,
|
| 479 |
+
"grad_norm": 0.30385342240333557,
|
| 480 |
+
"learning_rate": 0.0005699464138870106,
|
| 481 |
+
"loss": 0.7758321762084961,
|
| 482 |
+
"step": 3968
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.022716049382716,
|
| 486 |
+
"grad_norm": 0.3034447133541107,
|
| 487 |
+
"learning_rate": 0.0005436236984889356,
|
| 488 |
+
"loss": 0.7818131446838379,
|
| 489 |
+
"step": 4096
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.022716049382716,
|
| 493 |
+
"eval_cos_loss": 0.22541342421275815,
|
| 494 |
+
"eval_loss": 0.7809779222418622,
|
| 495 |
+
"eval_mse_loss": 0.758436579529832,
|
| 496 |
+
"step": 4096
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.022716049382716,
|
| 500 |
+
"eval_cos_loss": 0.22541342421275815,
|
| 501 |
+
"eval_loss": 0.7809779222418622,
|
| 502 |
+
"eval_mse_loss": 0.758436579529832,
|
| 503 |
+
"eval_runtime": 3.01,
|
| 504 |
+
"eval_samples_per_second": 869.76,
|
| 505 |
+
"eval_steps_per_second": 13.621,
|
| 506 |
+
"step": 4096
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.0859259259259257,
|
| 510 |
+
"grad_norm": 0.1902398020029068,
|
| 511 |
+
"learning_rate": 0.0005171784972522235,
|
| 512 |
+
"loss": 0.7848505973815918,
|
| 513 |
+
"step": 4224
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 2.149135802469136,
|
| 517 |
+
"grad_norm": 0.18901203572750092,
|
| 518 |
+
"learning_rate": 0.0004906850625388614,
|
| 519 |
+
"loss": 0.7853664755821228,
|
| 520 |
+
"step": 4352
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.149135802469136,
|
| 524 |
+
"eval_cos_loss": 0.2236430724946464,
|
| 525 |
+
"eval_loss": 0.7772586345672607,
|
| 526 |
+
"eval_mse_loss": 0.7548943234653007,
|
| 527 |
+
"step": 4352
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.149135802469136,
|
| 531 |
+
"eval_cos_loss": 0.2236430724946464,
|
| 532 |
+
"eval_loss": 0.7772586345672607,
|
| 533 |
+
"eval_mse_loss": 0.7548943234653007,
|
| 534 |
+
"eval_runtime": 2.898,
|
| 535 |
+
"eval_samples_per_second": 903.388,
|
| 536 |
+
"eval_steps_per_second": 14.148,
|
| 537 |
+
"step": 4352
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 2.212345679012346,
|
| 541 |
+
"grad_norm": 0.24396322667598724,
|
| 542 |
+
"learning_rate": 0.0004642177821399269,
|
| 543 |
+
"loss": 0.773828387260437,
|
| 544 |
+
"step": 4480
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 2.2755555555555556,
|
| 548 |
+
"grad_norm": 0.2060992270708084,
|
| 549 |
+
"learning_rate": 0.00043785097041088293,
|
| 550 |
+
"loss": 0.7780256271362305,
|
| 551 |
+
"step": 4608
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 2.2755555555555556,
|
| 555 |
+
"eval_cos_loss": 0.22296342435406474,
|
| 556 |
+
"eval_loss": 0.7785176024204348,
|
| 557 |
+
"eval_mse_loss": 0.7562212595125524,
|
| 558 |
+
"step": 4608
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 2.2755555555555556,
|
| 562 |
+
"eval_cos_loss": 0.22296342435406474,
|
| 563 |
+
"eval_loss": 0.7785176024204348,
|
| 564 |
+
"eval_mse_loss": 0.7562212595125524,
|
| 565 |
+
"eval_runtime": 2.9308,
|
| 566 |
+
"eval_samples_per_second": 893.271,
|
| 567 |
+
"eval_steps_per_second": 13.989,
|
| 568 |
+
"step": 4608
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.338765432098765,
|
| 572 |
+
"grad_norm": 0.3069784641265869,
|
| 573 |
+
"learning_rate": 0.00041165865961306135,
|
| 574 |
+
"loss": 0.7798830270767212,
|
| 575 |
+
"step": 4736
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.4019753086419753,
|
| 579 |
+
"grad_norm": 0.2529975175857544,
|
| 580 |
+
"learning_rate": 0.000385714392047208,
|
| 581 |
+
"loss": 0.7820331454277039,
|
| 582 |
+
"step": 4864
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.4019753086419753,
|
| 586 |
+
"eval_cos_loss": 0.22622913849063037,
|
| 587 |
+
"eval_loss": 0.7879578732862705,
|
| 588 |
+
"eval_mse_loss": 0.7653349579834356,
|
| 589 |
+
"step": 4864
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 2.4019753086419753,
|
| 593 |
+
"eval_cos_loss": 0.22622913849063037,
|
| 594 |
+
"eval_loss": 0.7879578732862705,
|
| 595 |
+
"eval_mse_loss": 0.7653349579834356,
|
| 596 |
+
"eval_runtime": 2.8737,
|
| 597 |
+
"eval_samples_per_second": 911.028,
|
| 598 |
+
"eval_steps_per_second": 14.267,
|
| 599 |
+
"step": 4864
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.4651851851851854,
|
| 603 |
+
"grad_norm": 0.1531030684709549,
|
| 604 |
+
"learning_rate": 0.00036009101356272814,
|
| 605 |
+
"loss": 0.7813113331794739,
|
| 606 |
+
"step": 4992
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.528395061728395,
|
| 610 |
+
"grad_norm": 0.28510212898254395,
|
| 611 |
+
"learning_rate": 0.00033486046902241664,
|
| 612 |
+
"loss": 0.7822959423065186,
|
| 613 |
+
"step": 5120
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.528395061728395,
|
| 617 |
+
"eval_cos_loss": 0.22616689961131026,
|
| 618 |
+
"eval_loss": 0.7882982463371463,
|
| 619 |
+
"eval_mse_loss": 0.7656815560852609,
|
| 620 |
+
"step": 5120
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.528395061728395,
|
| 624 |
+
"eval_cos_loss": 0.22616689961131026,
|
| 625 |
+
"eval_loss": 0.7882982463371463,
|
| 626 |
+
"eval_mse_loss": 0.7656815560852609,
|
| 627 |
+
"eval_runtime": 3.4221,
|
| 628 |
+
"eval_samples_per_second": 765.035,
|
| 629 |
+
"eval_steps_per_second": 11.981,
|
| 630 |
+
"step": 5120
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.591604938271605,
|
| 634 |
+
"grad_norm": 0.3471938371658325,
|
| 635 |
+
"learning_rate": 0.00031009360029696107,
|
| 636 |
+
"loss": 0.7768149971961975,
|
| 637 |
+
"step": 5248
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 2.6548148148148147,
|
| 641 |
+
"grad_norm": 0.30711060762405396,
|
| 642 |
+
"learning_rate": 0.00028585994735640823,
|
| 643 |
+
"loss": 0.7821690440177917,
|
| 644 |
+
"step": 5376
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.6548148148148147,
|
| 648 |
+
"eval_cos_loss": 0.22484926334241542,
|
| 649 |
+
"eval_loss": 0.7831642933008147,
|
| 650 |
+
"eval_mse_loss": 0.7606793656581785,
|
| 651 |
+
"step": 5376
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 2.6548148148148147,
|
| 655 |
+
"eval_cos_loss": 0.22484926334241542,
|
| 656 |
+
"eval_loss": 0.7831642933008147,
|
| 657 |
+
"eval_mse_loss": 0.7606793656581785,
|
| 658 |
+
"eval_runtime": 3.9797,
|
| 659 |
+
"eval_samples_per_second": 657.845,
|
| 660 |
+
"eval_steps_per_second": 10.302,
|
| 661 |
+
"step": 5376
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.718024691358025,
|
| 665 |
+
"grad_norm": 0.3152957558631897,
|
| 666 |
+
"learning_rate": 0.0002622275530170825,
|
| 667 |
+
"loss": 0.780761182308197,
|
| 668 |
+
"step": 5504
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.7812345679012345,
|
| 672 |
+
"grad_norm": 0.20758290588855743,
|
| 673 |
+
"learning_rate": 0.0002392627718921852,
|
| 674 |
+
"loss": 0.7799718976020813,
|
| 675 |
+
"step": 5632
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.7812345679012345,
|
| 679 |
+
"eval_cos_loss": 0.22278807185045102,
|
| 680 |
+
"eval_loss": 0.7812715899653551,
|
| 681 |
+
"eval_mse_loss": 0.758992785360755,
|
| 682 |
+
"step": 5632
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.7812345679012345,
|
| 686 |
+
"eval_cos_loss": 0.22278807185045102,
|
| 687 |
+
"eval_loss": 0.7812715899653551,
|
| 688 |
+
"eval_mse_loss": 0.758992785360755,
|
| 689 |
+
"eval_runtime": 3.9125,
|
| 690 |
+
"eval_samples_per_second": 669.129,
|
| 691 |
+
"eval_steps_per_second": 10.479,
|
| 692 |
+
"step": 5632
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.8444444444444446,
|
| 696 |
+
"grad_norm": 0.12054615467786789,
|
| 697 |
+
"learning_rate": 0.00021703008408250187,
|
| 698 |
+
"loss": 0.7806150913238525,
|
| 699 |
+
"step": 5760
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.907654320987654,
|
| 703 |
+
"grad_norm": 0.2542758584022522,
|
| 704 |
+
"learning_rate": 0.00019559191413033017,
|
| 705 |
+
"loss": 0.7816909551620483,
|
| 706 |
+
"step": 5888
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.907654320987654,
|
| 710 |
+
"eval_cos_loss": 0.2219865827298746,
|
| 711 |
+
"eval_loss": 0.7693439911051494,
|
| 712 |
+
"eval_mse_loss": 0.7471453373025103,
|
| 713 |
+
"step": 5888
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.907654320987654,
|
| 717 |
+
"eval_cos_loss": 0.2219865827298746,
|
| 718 |
+
"eval_loss": 0.7693439911051494,
|
| 719 |
+
"eval_mse_loss": 0.7471453373025103,
|
| 720 |
+
"eval_runtime": 3.1964,
|
| 721 |
+
"eval_samples_per_second": 819.045,
|
| 722 |
+
"eval_steps_per_second": 12.827,
|
| 723 |
+
"step": 5888
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.9708641975308643,
|
| 727 |
+
"grad_norm": 0.18909312784671783,
|
| 728 |
+
"learning_rate": 0.00017500845574497032,
|
| 729 |
+
"loss": 0.7817902565002441,
|
| 730 |
+
"step": 6016
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 3.034074074074074,
|
| 734 |
+
"grad_norm": 0.22536472976207733,
|
| 735 |
+
"learning_rate": 0.00015533750279190617,
|
| 736 |
+
"loss": 0.7782471179962158,
|
| 737 |
+
"step": 6144
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 3.034074074074074,
|
| 741 |
+
"eval_cos_loss": 0.2228206588727672,
|
| 742 |
+
"eval_loss": 0.7775220769207652,
|
| 743 |
+
"eval_mse_loss": 0.7552400100521925,
|
| 744 |
+
"step": 6144
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 3.034074074074074,
|
| 748 |
+
"eval_cos_loss": 0.2228206588727672,
|
| 749 |
+
"eval_loss": 0.7775220769207652,
|
| 750 |
+
"eval_mse_loss": 0.7552400100521925,
|
| 751 |
+
"eval_runtime": 2.9098,
|
| 752 |
+
"eval_samples_per_second": 899.731,
|
| 753 |
+
"eval_steps_per_second": 14.091,
|
| 754 |
+
"step": 6144
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.097283950617284,
|
| 758 |
+
"grad_norm": 0.1673809289932251,
|
| 759 |
+
"learning_rate": 0.000136634287020226,
|
| 760 |
+
"loss": 0.7807482481002808,
|
| 761 |
+
"step": 6272
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.1604938271604937,
|
| 765 |
+
"grad_norm": 0.3716048300266266,
|
| 766 |
+
"learning_rate": 0.00011895132298390743,
|
| 767 |
+
"loss": 0.7781609296798706,
|
| 768 |
+
"step": 6400
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.1604938271604937,
|
| 772 |
+
"eval_cos_loss": 0.2247363121771231,
|
| 773 |
+
"eval_loss": 0.7763603416884818,
|
| 774 |
+
"eval_mse_loss": 0.7538867098529164,
|
| 775 |
+
"step": 6400
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.1604938271604937,
|
| 779 |
+
"eval_cos_loss": 0.2247363121771231,
|
| 780 |
+
"eval_loss": 0.7763603416884818,
|
| 781 |
+
"eval_mse_loss": 0.7538867098529164,
|
| 782 |
+
"eval_runtime": 2.9575,
|
| 783 |
+
"eval_samples_per_second": 885.206,
|
| 784 |
+
"eval_steps_per_second": 13.863,
|
| 785 |
+
"step": 6400
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 3.2237037037037037,
|
| 789 |
+
"grad_norm": 0.23243093490600586,
|
| 790 |
+
"learning_rate": 0.00010233826059239426,
|
| 791 |
+
"loss": 0.7826894521713257,
|
| 792 |
+
"step": 6528
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 3.286913580246914,
|
| 796 |
+
"grad_norm": 0.20378775894641876,
|
| 797 |
+
"learning_rate": 8.68417457044704e-05,
|
| 798 |
+
"loss": 0.7802485823631287,
|
| 799 |
+
"step": 6656
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 3.286913580246914,
|
| 803 |
+
"eval_cos_loss": 0.22500916079777042,
|
| 804 |
+
"eval_loss": 0.7808182821041201,
|
| 805 |
+
"eval_mse_loss": 0.7583173658789658,
|
| 806 |
+
"step": 6656
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 3.286913580246914,
|
| 810 |
+
"eval_cos_loss": 0.22500916079777042,
|
| 811 |
+
"eval_loss": 0.7808182821041201,
|
| 812 |
+
"eval_mse_loss": 0.7583173658789658,
|
| 813 |
+
"eval_runtime": 2.8855,
|
| 814 |
+
"eval_samples_per_second": 907.28,
|
| 815 |
+
"eval_steps_per_second": 14.209,
|
| 816 |
+
"step": 6656
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.3501234567901235,
|
| 820 |
+
"grad_norm": 0.12901104986667633,
|
| 821 |
+
"learning_rate": 7.250528915685422e-05,
|
| 822 |
+
"loss": 0.7797313332557678,
|
| 823 |
+
"step": 6784
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 3.413333333333333,
|
| 827 |
+
"grad_norm": 0.1314738541841507,
|
| 828 |
+
"learning_rate": 5.93691445952505e-05,
|
| 829 |
+
"loss": 0.7752676606178284,
|
| 830 |
+
"step": 6912
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 3.413333333333333,
|
| 834 |
+
"eval_cos_loss": 0.22601125480198278,
|
| 835 |
+
"eval_loss": 0.7906920822655282,
|
| 836 |
+
"eval_mse_loss": 0.7680909546410165,
|
| 837 |
+
"step": 6912
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.413333333333333,
|
| 841 |
+
"eval_cos_loss": 0.22601125480198278,
|
| 842 |
+
"eval_loss": 0.7906920822655282,
|
| 843 |
+
"eval_mse_loss": 0.7680909546410165,
|
| 844 |
+
"eval_runtime": 2.9757,
|
| 845 |
+
"eval_samples_per_second": 879.801,
|
| 846 |
+
"eval_steps_per_second": 13.778,
|
| 847 |
+
"step": 6912
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 3.476543209876543,
|
| 851 |
+
"grad_norm": 0.48443037271499634,
|
| 852 |
+
"learning_rate": 4.7470195450886376e-05,
|
| 853 |
+
"loss": 0.7797003388404846,
|
| 854 |
+
"step": 7040
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 3.5397530864197533,
|
| 858 |
+
"grad_norm": 0.2399303913116455,
|
| 859 |
+
"learning_rate": 3.684185137987378e-05,
|
| 860 |
+
"loss": 0.7735744714736938,
|
| 861 |
+
"step": 7168
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 3.5397530864197533,
|
| 865 |
+
"eval_cos_loss": 0.22420256748432066,
|
| 866 |
+
"eval_loss": 0.784631618639318,
|
| 867 |
+
"eval_mse_loss": 0.7622113591287194,
|
| 868 |
+
"step": 7168
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 3.5397530864197533,
|
| 872 |
+
"eval_cos_loss": 0.22420256748432066,
|
| 873 |
+
"eval_loss": 0.784631618639318,
|
| 874 |
+
"eval_mse_loss": 0.7622113591287194,
|
| 875 |
+
"eval_runtime": 2.8474,
|
| 876 |
+
"eval_samples_per_second": 919.424,
|
| 877 |
+
"eval_steps_per_second": 14.399,
|
| 878 |
+
"step": 7168
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 3.602962962962963,
|
| 882 |
+
"grad_norm": 0.2262357771396637,
|
| 883 |
+
"learning_rate": 2.751395445617594e-05,
|
| 884 |
+
"loss": 0.7803007364273071,
|
| 885 |
+
"step": 7296
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.6661728395061726,
|
| 889 |
+
"grad_norm": 0.12137070298194885,
|
| 890 |
+
"learning_rate": 1.9512695381567302e-05,
|
| 891 |
+
"loss": 0.776282548904419,
|
| 892 |
+
"step": 7424
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 3.6661728395061726,
|
| 896 |
+
"eval_cos_loss": 0.22227415696876804,
|
| 897 |
+
"eval_loss": 0.7707422797272845,
|
| 898 |
+
"eval_mse_loss": 0.7485148630491117,
|
| 899 |
+
"step": 7424
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 3.6661728395061726,
|
| 903 |
+
"eval_cos_loss": 0.22227415696876804,
|
| 904 |
+
"eval_loss": 0.7707422797272845,
|
| 905 |
+
"eval_mse_loss": 0.7485148630491117,
|
| 906 |
+
"eval_runtime": 2.937,
|
| 907 |
+
"eval_samples_per_second": 891.381,
|
| 908 |
+
"eval_steps_per_second": 13.96,
|
| 909 |
+
"step": 7424
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 3.7293827160493827,
|
| 913 |
+
"grad_norm": 0.18187913298606873,
|
| 914 |
+
"learning_rate": 1.2860539947850303e-05,
|
| 915 |
+
"loss": 0.780539870262146,
|
| 916 |
+
"step": 7552
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 3.7925925925925927,
|
| 920 |
+
"grad_norm": 0.17140205204486847,
|
| 921 |
+
"learning_rate": 7.5761659578078565e-06,
|
| 922 |
+
"loss": 0.7720261812210083,
|
| 923 |
+
"step": 7680
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 3.7925925925925927,
|
| 927 |
+
"eval_cos_loss": 0.22543459512838504,
|
| 928 |
+
"eval_loss": 0.7825955297888779,
|
| 929 |
+
"eval_mse_loss": 0.7600520703850723,
|
| 930 |
+
"step": 7680
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 3.7925925925925927,
|
| 934 |
+
"eval_cos_loss": 0.22543459512838504,
|
| 935 |
+
"eval_loss": 0.7825955297888779,
|
| 936 |
+
"eval_mse_loss": 0.7600520703850723,
|
| 937 |
+
"eval_runtime": 3.0241,
|
| 938 |
+
"eval_samples_per_second": 865.716,
|
| 939 |
+
"eval_steps_per_second": 13.558,
|
| 940 |
+
"step": 7680
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 3.8558024691358024,
|
| 944 |
+
"grad_norm": 0.2181410789489746,
|
| 945 |
+
"learning_rate": 3.674410782003812e-06,
|
| 946 |
+
"loss": 0.7785236835479736,
|
| 947 |
+
"step": 7808
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 3.9190123456790125,
|
| 951 |
+
"grad_norm": 0.1741209775209427,
|
| 952 |
+
"learning_rate": 1.1662296986795217e-06,
|
| 953 |
+
"loss": 0.7693132162094116,
|
| 954 |
+
"step": 7936
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 3.9190123456790125,
|
| 958 |
+
"eval_cos_loss": 0.22448242837336005,
|
| 959 |
+
"eval_loss": 0.7836573894430952,
|
| 960 |
+
"eval_mse_loss": 0.7612091462786604,
|
| 961 |
+
"step": 7936
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 3.9190123456790125,
|
| 965 |
+
"eval_cos_loss": 0.22448242837336005,
|
| 966 |
+
"eval_loss": 0.7836573894430952,
|
| 967 |
+
"eval_mse_loss": 0.7612091462786604,
|
| 968 |
+
"eval_runtime": 3.6021,
|
| 969 |
+
"eval_samples_per_second": 726.789,
|
| 970 |
+
"eval_steps_per_second": 11.382,
|
| 971 |
+
"step": 7936
|
| 972 |
+
}
|
| 973 |
+
],
|
| 974 |
+
"logging_steps": 128,
|
| 975 |
+
"max_steps": 8100,
|
| 976 |
+
"num_input_tokens_seen": 0,
|
| 977 |
+
"num_train_epochs": 4,
|
| 978 |
+
"save_steps": 256,
|
| 979 |
+
"stateful_callbacks": {
|
| 980 |
+
"TrainerControl": {
|
| 981 |
+
"args": {
|
| 982 |
+
"should_epoch_stop": false,
|
| 983 |
+
"should_evaluate": false,
|
| 984 |
+
"should_log": false,
|
| 985 |
+
"should_save": true,
|
| 986 |
+
"should_training_stop": false
|
| 987 |
+
},
|
| 988 |
+
"attributes": {}
|
| 989 |
+
}
|
| 990 |
+
},
|
| 991 |
+
"total_flos": 0.0,
|
| 992 |
+
"train_batch_size": 64,
|
| 993 |
+
"trial_name": null,
|
| 994 |
+
"trial_params": null
|
| 995 |
+
}
|
checkpoints-v3.0-discrete-conditional/checkpoint-7936/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6c0fa495a5924736d222e27760d060d662f12f7b3a2b9af86925d4d1f405f6c
|
| 3 |
+
size 5137
|