Training in progress, step 440, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 36981072
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13cce9bb0e996351713314d3f8524bbdb49c9ae98bbc5d38e88243b1df518f08
|
| 3 |
size 36981072
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 19859524
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb676f03323a195f09bb3fb1a56057c8e39c831ecf332633428cccdf75a1efdb
|
| 3 |
size 19859524
|
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:e99cd4efdc8ae360618f7c31cdaf24ebec835e26008dd443222ea8c2ad794ce3
|
| 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 sha256:3aa6d4cd0a1d119d88746df8d17b061da99249879d9cb64d05543ac4d112a2c5
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -2317,6 +2317,784 @@
|
|
| 2317 |
"learning_rate": 1.4965269896332885e-05,
|
| 2318 |
"loss": 0.6016,
|
| 2319 |
"step": 330
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2320 |
}
|
| 2321 |
],
|
| 2322 |
"logging_steps": 1,
|
|
@@ -2331,12 +3109,12 @@
|
|
| 2331 |
"should_evaluate": false,
|
| 2332 |
"should_log": false,
|
| 2333 |
"should_save": true,
|
| 2334 |
-
"should_training_stop":
|
| 2335 |
},
|
| 2336 |
"attributes": {}
|
| 2337 |
}
|
| 2338 |
},
|
| 2339 |
-
"total_flos":
|
| 2340 |
"train_batch_size": 4,
|
| 2341 |
"trial_name": null,
|
| 2342 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0017055144968732,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 440,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 2317 |
"learning_rate": 1.4965269896332885e-05,
|
| 2318 |
"loss": 0.6016,
|
| 2319 |
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7527003979533826,
|
| 2323 |
+
"grad_norm": 3.569581985473633,
|
| 2324 |
+
"learning_rate": 1.4708553057981355e-05,
|
| 2325 |
+
"loss": 0.4036,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.7549744172825469,
|
| 2330 |
+
"grad_norm": 3.5513603687286377,
|
| 2331 |
+
"learning_rate": 1.4453676944196476e-05,
|
| 2332 |
+
"loss": 0.3642,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7572484366117112,
|
| 2337 |
+
"grad_norm": 3.9503798484802246,
|
| 2338 |
+
"learning_rate": 1.4200654848757994e-05,
|
| 2339 |
+
"loss": 0.3798,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7595224559408755,
|
| 2344 |
+
"grad_norm": 3.2657759189605713,
|
| 2345 |
+
"learning_rate": 1.3949499968744206e-05,
|
| 2346 |
+
"loss": 0.3445,
|
| 2347 |
+
"step": 334
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.7617964752700398,
|
| 2351 |
+
"grad_norm": 1.5095345973968506,
|
| 2352 |
+
"learning_rate": 1.3700225403843469e-05,
|
| 2353 |
+
"loss": 0.1506,
|
| 2354 |
+
"step": 335
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.7640704945992041,
|
| 2358 |
+
"grad_norm": 3.1756370067596436,
|
| 2359 |
+
"learning_rate": 1.3452844155671052e-05,
|
| 2360 |
+
"loss": 0.3452,
|
| 2361 |
+
"step": 336
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.7663445139283684,
|
| 2365 |
+
"grad_norm": 2.36387038230896,
|
| 2366 |
+
"learning_rate": 1.3207369127090985e-05,
|
| 2367 |
+
"loss": 0.2018,
|
| 2368 |
+
"step": 337
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.7686185332575327,
|
| 2372 |
+
"grad_norm": 1.7574656009674072,
|
| 2373 |
+
"learning_rate": 1.296381312154305e-05,
|
| 2374 |
+
"loss": 0.1288,
|
| 2375 |
+
"step": 338
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.770892552586697,
|
| 2379 |
+
"grad_norm": 3.010063409805298,
|
| 2380 |
+
"learning_rate": 1.2722188842374966e-05,
|
| 2381 |
+
"loss": 0.2819,
|
| 2382 |
+
"step": 339
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.7731665719158612,
|
| 2386 |
+
"grad_norm": 1.5432311296463013,
|
| 2387 |
+
"learning_rate": 1.2482508892179884e-05,
|
| 2388 |
+
"loss": 0.0887,
|
| 2389 |
+
"step": 340
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.7754405912450256,
|
| 2393 |
+
"grad_norm": 2.749730110168457,
|
| 2394 |
+
"learning_rate": 1.2244785772138972e-05,
|
| 2395 |
+
"loss": 0.1669,
|
| 2396 |
+
"step": 341
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.7777146105741899,
|
| 2400 |
+
"grad_norm": 4.618091583251953,
|
| 2401 |
+
"learning_rate": 1.2009031881369431e-05,
|
| 2402 |
+
"loss": 0.227,
|
| 2403 |
+
"step": 342
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.7799886299033542,
|
| 2407 |
+
"grad_norm": 1.5743058919906616,
|
| 2408 |
+
"learning_rate": 1.177525951627781e-05,
|
| 2409 |
+
"loss": 0.1983,
|
| 2410 |
+
"step": 343
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.7822626492325184,
|
| 2414 |
+
"grad_norm": 1.480060338973999,
|
| 2415 |
+
"learning_rate": 1.1543480869918555e-05,
|
| 2416 |
+
"loss": 0.0891,
|
| 2417 |
+
"step": 344
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.7845366685616828,
|
| 2421 |
+
"grad_norm": 1.6388338804244995,
|
| 2422 |
+
"learning_rate": 1.1313708031358183e-05,
|
| 2423 |
+
"loss": 0.0913,
|
| 2424 |
+
"step": 345
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.7868106878908471,
|
| 2428 |
+
"grad_norm": 0.9587397575378418,
|
| 2429 |
+
"learning_rate": 1.1085952985044634e-05,
|
| 2430 |
+
"loss": 0.041,
|
| 2431 |
+
"step": 346
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.7890847072200113,
|
| 2435 |
+
"grad_norm": 1.376420497894287,
|
| 2436 |
+
"learning_rate": 1.0860227610182222e-05,
|
| 2437 |
+
"loss": 0.071,
|
| 2438 |
+
"step": 347
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.7913587265491757,
|
| 2442 |
+
"grad_norm": 1.479590892791748,
|
| 2443 |
+
"learning_rate": 1.0636543680112044e-05,
|
| 2444 |
+
"loss": 0.0905,
|
| 2445 |
+
"step": 348
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.79363274587834,
|
| 2449 |
+
"grad_norm": 1.486275553703308,
|
| 2450 |
+
"learning_rate": 1.04149128616979e-05,
|
| 2451 |
+
"loss": 0.1012,
|
| 2452 |
+
"step": 349
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.7959067652075043,
|
| 2456 |
+
"grad_norm": 0.9402182698249817,
|
| 2457 |
+
"learning_rate": 1.0195346714717813e-05,
|
| 2458 |
+
"loss": 0.1013,
|
| 2459 |
+
"step": 350
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.7981807845366685,
|
| 2463 |
+
"grad_norm": 4.111584186553955,
|
| 2464 |
+
"learning_rate": 9.977856691261057e-06,
|
| 2465 |
+
"loss": 0.6617,
|
| 2466 |
+
"step": 351
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8004548038658329,
|
| 2470 |
+
"grad_norm": 4.183617115020752,
|
| 2471 |
+
"learning_rate": 9.762454135130828e-06,
|
| 2472 |
+
"loss": 0.5664,
|
| 2473 |
+
"step": 352
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8027288231949972,
|
| 2477 |
+
"grad_norm": 3.7775447368621826,
|
| 2478 |
+
"learning_rate": 9.549150281252633e-06,
|
| 2479 |
+
"loss": 0.4659,
|
| 2480 |
+
"step": 353
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8050028425241614,
|
| 2484 |
+
"grad_norm": 3.4957637786865234,
|
| 2485 |
+
"learning_rate": 9.337956255088237e-06,
|
| 2486 |
+
"loss": 0.344,
|
| 2487 |
+
"step": 354
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8072768618533257,
|
| 2491 |
+
"grad_norm": 3.0935065746307373,
|
| 2492 |
+
"learning_rate": 9.12888307205541e-06,
|
| 2493 |
+
"loss": 0.3546,
|
| 2494 |
+
"step": 355
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8095508811824901,
|
| 2498 |
+
"grad_norm": 4.095314025878906,
|
| 2499 |
+
"learning_rate": 8.921941636953435e-06,
|
| 2500 |
+
"loss": 0.4984,
|
| 2501 |
+
"step": 356
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.8118249005116543,
|
| 2505 |
+
"grad_norm": 3.417351007461548,
|
| 2506 |
+
"learning_rate": 8.717142743394236e-06,
|
| 2507 |
+
"loss": 0.3108,
|
| 2508 |
+
"step": 357
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8140989198408186,
|
| 2512 |
+
"grad_norm": 2.5139520168304443,
|
| 2513 |
+
"learning_rate": 8.514497073239491e-06,
|
| 2514 |
+
"loss": 0.3556,
|
| 2515 |
+
"step": 358
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.816372939169983,
|
| 2519 |
+
"grad_norm": 4.973466873168945,
|
| 2520 |
+
"learning_rate": 8.3140151960435e-06,
|
| 2521 |
+
"loss": 0.5424,
|
| 2522 |
+
"step": 359
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.8186469584991473,
|
| 2526 |
+
"grad_norm": 4.196943759918213,
|
| 2527 |
+
"learning_rate": 8.115707568501768e-06,
|
| 2528 |
+
"loss": 0.4552,
|
| 2529 |
+
"step": 360
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.8209209778283115,
|
| 2533 |
+
"grad_norm": 4.142265796661377,
|
| 2534 |
+
"learning_rate": 7.919584533905777e-06,
|
| 2535 |
+
"loss": 0.3666,
|
| 2536 |
+
"step": 361
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.8231949971574758,
|
| 2540 |
+
"grad_norm": 3.281536102294922,
|
| 2541 |
+
"learning_rate": 7.725656321603413e-06,
|
| 2542 |
+
"loss": 0.3572,
|
| 2543 |
+
"step": 362
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8254690164866402,
|
| 2547 |
+
"grad_norm": 5.057455539703369,
|
| 2548 |
+
"learning_rate": 7.533933046465419e-06,
|
| 2549 |
+
"loss": 0.5856,
|
| 2550 |
+
"step": 363
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.8277430358158044,
|
| 2554 |
+
"grad_norm": 5.18012809753418,
|
| 2555 |
+
"learning_rate": 7.344424708357867e-06,
|
| 2556 |
+
"loss": 0.4198,
|
| 2557 |
+
"step": 364
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8300170551449687,
|
| 2561 |
+
"grad_norm": 2.427621841430664,
|
| 2562 |
+
"learning_rate": 7.157141191620548e-06,
|
| 2563 |
+
"loss": 0.1815,
|
| 2564 |
+
"step": 365
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.832291074474133,
|
| 2568 |
+
"grad_norm": 0.5157700777053833,
|
| 2569 |
+
"learning_rate": 6.972092264551438e-06,
|
| 2570 |
+
"loss": 0.0464,
|
| 2571 |
+
"step": 366
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.8345650938032974,
|
| 2575 |
+
"grad_norm": 0.9035636782646179,
|
| 2576 |
+
"learning_rate": 6.789287578897252e-06,
|
| 2577 |
+
"loss": 0.0749,
|
| 2578 |
+
"step": 367
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8368391131324616,
|
| 2582 |
+
"grad_norm": 0.434658020734787,
|
| 2583 |
+
"learning_rate": 6.6087366693499295e-06,
|
| 2584 |
+
"loss": 0.0233,
|
| 2585 |
+
"step": 368
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8391131324616259,
|
| 2589 |
+
"grad_norm": 0.6567199230194092,
|
| 2590 |
+
"learning_rate": 6.430448953049434e-06,
|
| 2591 |
+
"loss": 0.017,
|
| 2592 |
+
"step": 369
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8413871517907903,
|
| 2596 |
+
"grad_norm": 0.537125289440155,
|
| 2597 |
+
"learning_rate": 6.2544337290925185e-06,
|
| 2598 |
+
"loss": 0.0145,
|
| 2599 |
+
"step": 370
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.8436611711199545,
|
| 2603 |
+
"grad_norm": 0.5989580750465393,
|
| 2604 |
+
"learning_rate": 6.080700178047688e-06,
|
| 2605 |
+
"loss": 0.0196,
|
| 2606 |
+
"step": 371
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8459351904491188,
|
| 2610 |
+
"grad_norm": 0.7875169515609741,
|
| 2611 |
+
"learning_rate": 5.909257361476405e-06,
|
| 2612 |
+
"loss": 0.0282,
|
| 2613 |
+
"step": 372
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8482092097782831,
|
| 2617 |
+
"grad_norm": 0.6391351819038391,
|
| 2618 |
+
"learning_rate": 5.740114221460424e-06,
|
| 2619 |
+
"loss": 0.0167,
|
| 2620 |
+
"step": 373
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8504832291074474,
|
| 2624 |
+
"grad_norm": 0.4413357377052307,
|
| 2625 |
+
"learning_rate": 5.573279580135438e-06,
|
| 2626 |
+
"loss": 0.0425,
|
| 2627 |
+
"step": 374
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8527572484366117,
|
| 2631 |
+
"grad_norm": 0.4121001958847046,
|
| 2632 |
+
"learning_rate": 5.408762139230888e-06,
|
| 2633 |
+
"loss": 0.0165,
|
| 2634 |
+
"step": 375
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.855031267765776,
|
| 2638 |
+
"grad_norm": 0.6832776665687561,
|
| 2639 |
+
"learning_rate": 5.246570479616103e-06,
|
| 2640 |
+
"loss": 0.0272,
|
| 2641 |
+
"step": 376
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8573052870949404,
|
| 2645 |
+
"grad_norm": 3.090912342071533,
|
| 2646 |
+
"learning_rate": 5.086713060852788e-06,
|
| 2647 |
+
"loss": 0.176,
|
| 2648 |
+
"step": 377
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.8595793064241046,
|
| 2652 |
+
"grad_norm": 2.9185822010040283,
|
| 2653 |
+
"learning_rate": 4.929198220753722e-06,
|
| 2654 |
+
"loss": 0.2692,
|
| 2655 |
+
"step": 378
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8618533257532689,
|
| 2659 |
+
"grad_norm": 5.094744682312012,
|
| 2660 |
+
"learning_rate": 4.774034174947922e-06,
|
| 2661 |
+
"loss": 0.4768,
|
| 2662 |
+
"step": 379
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.8641273450824332,
|
| 2666 |
+
"grad_norm": 4.020772933959961,
|
| 2667 |
+
"learning_rate": 4.621229016452156e-06,
|
| 2668 |
+
"loss": 0.332,
|
| 2669 |
+
"step": 380
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.8664013644115975,
|
| 2673 |
+
"grad_norm": 4.474671840667725,
|
| 2674 |
+
"learning_rate": 4.4707907152487405e-06,
|
| 2675 |
+
"loss": 0.3265,
|
| 2676 |
+
"step": 381
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.8686753837407618,
|
| 2680 |
+
"grad_norm": 3.18890643119812,
|
| 2681 |
+
"learning_rate": 4.322727117869951e-06,
|
| 2682 |
+
"loss": 0.2424,
|
| 2683 |
+
"step": 382
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.8709494030699261,
|
| 2687 |
+
"grad_norm": 4.199413299560547,
|
| 2688 |
+
"learning_rate": 4.1770459469887005e-06,
|
| 2689 |
+
"loss": 0.2787,
|
| 2690 |
+
"step": 383
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.8732234223990903,
|
| 2694 |
+
"grad_norm": 2.4939513206481934,
|
| 2695 |
+
"learning_rate": 4.033754801015732e-06,
|
| 2696 |
+
"loss": 0.1588,
|
| 2697 |
+
"step": 384
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.8754974417282547,
|
| 2701 |
+
"grad_norm": 3.3816030025482178,
|
| 2702 |
+
"learning_rate": 3.892861153703342e-06,
|
| 2703 |
+
"loss": 0.2011,
|
| 2704 |
+
"step": 385
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.877771461057419,
|
| 2708 |
+
"grad_norm": 2.142763376235962,
|
| 2709 |
+
"learning_rate": 3.7543723537555585e-06,
|
| 2710 |
+
"loss": 0.1491,
|
| 2711 |
+
"step": 386
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.8800454803865833,
|
| 2715 |
+
"grad_norm": 2.4426450729370117,
|
| 2716 |
+
"learning_rate": 3.6182956244448117e-06,
|
| 2717 |
+
"loss": 0.1654,
|
| 2718 |
+
"step": 387
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.8823194997157476,
|
| 2722 |
+
"grad_norm": 1.4727312326431274,
|
| 2723 |
+
"learning_rate": 3.4846380632352458e-06,
|
| 2724 |
+
"loss": 0.0619,
|
| 2725 |
+
"step": 388
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.8845935190449119,
|
| 2729 |
+
"grad_norm": 1.7243990898132324,
|
| 2730 |
+
"learning_rate": 3.35340664141246e-06,
|
| 2731 |
+
"loss": 0.0671,
|
| 2732 |
+
"step": 389
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.8868675383740762,
|
| 2736 |
+
"grad_norm": 1.6698848009109497,
|
| 2737 |
+
"learning_rate": 3.2246082037199532e-06,
|
| 2738 |
+
"loss": 0.0581,
|
| 2739 |
+
"step": 390
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.8891415577032404,
|
| 2743 |
+
"grad_norm": 2.760230302810669,
|
| 2744 |
+
"learning_rate": 3.0982494680021177e-06,
|
| 2745 |
+
"loss": 0.1919,
|
| 2746 |
+
"step": 391
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.8914155770324048,
|
| 2750 |
+
"grad_norm": 4.160801410675049,
|
| 2751 |
+
"learning_rate": 2.9743370248538017e-06,
|
| 2752 |
+
"loss": 0.3573,
|
| 2753 |
+
"step": 392
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.8936895963615691,
|
| 2757 |
+
"grad_norm": 1.2745391130447388,
|
| 2758 |
+
"learning_rate": 2.8528773372766216e-06,
|
| 2759 |
+
"loss": 0.0565,
|
| 2760 |
+
"step": 393
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.8959636156907334,
|
| 2764 |
+
"grad_norm": 1.1233683824539185,
|
| 2765 |
+
"learning_rate": 2.7338767403418287e-06,
|
| 2766 |
+
"loss": 0.0371,
|
| 2767 |
+
"step": 394
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.8982376350198976,
|
| 2771 |
+
"grad_norm": 1.4613378047943115,
|
| 2772 |
+
"learning_rate": 2.6173414408598827e-06,
|
| 2773 |
+
"loss": 0.0638,
|
| 2774 |
+
"step": 395
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.900511654349062,
|
| 2778 |
+
"grad_norm": 1.0064411163330078,
|
| 2779 |
+
"learning_rate": 2.503277517056729e-06,
|
| 2780 |
+
"loss": 0.0226,
|
| 2781 |
+
"step": 396
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.9027856736782263,
|
| 2785 |
+
"grad_norm": 1.1050655841827393,
|
| 2786 |
+
"learning_rate": 2.3916909182567782e-06,
|
| 2787 |
+
"loss": 0.0482,
|
| 2788 |
+
"step": 397
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.9050596930073905,
|
| 2792 |
+
"grad_norm": 1.100659966468811,
|
| 2793 |
+
"learning_rate": 2.282587464572594e-06,
|
| 2794 |
+
"loss": 0.0531,
|
| 2795 |
+
"step": 398
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9073337123365549,
|
| 2799 |
+
"grad_norm": 0.8102996945381165,
|
| 2800 |
+
"learning_rate": 2.175972846601343e-06,
|
| 2801 |
+
"loss": 0.0399,
|
| 2802 |
+
"step": 399
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.9096077316657192,
|
| 2806 |
+
"grad_norm": 2.706613779067993,
|
| 2807 |
+
"learning_rate": 2.0718526251279346e-06,
|
| 2808 |
+
"loss": 0.2256,
|
| 2809 |
+
"step": 400
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.9118817509948834,
|
| 2813 |
+
"grad_norm": 4.959632396697998,
|
| 2814 |
+
"learning_rate": 1.9702322308350674e-06,
|
| 2815 |
+
"loss": 0.6795,
|
| 2816 |
+
"step": 401
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.9141557703240477,
|
| 2820 |
+
"grad_norm": 4.623620510101318,
|
| 2821 |
+
"learning_rate": 1.8711169640198977e-06,
|
| 2822 |
+
"loss": 0.3954,
|
| 2823 |
+
"step": 402
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.9164297896532121,
|
| 2827 |
+
"grad_norm": 3.527909517288208,
|
| 2828 |
+
"learning_rate": 1.774511994317629e-06,
|
| 2829 |
+
"loss": 0.4934,
|
| 2830 |
+
"step": 403
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.9187038089823764,
|
| 2834 |
+
"grad_norm": 4.126560688018799,
|
| 2835 |
+
"learning_rate": 1.6804223604318825e-06,
|
| 2836 |
+
"loss": 0.2839,
|
| 2837 |
+
"step": 404
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.9209778283115406,
|
| 2841 |
+
"grad_norm": 2.926298141479492,
|
| 2842 |
+
"learning_rate": 1.5888529698718346e-06,
|
| 2843 |
+
"loss": 0.2572,
|
| 2844 |
+
"step": 405
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.9232518476407049,
|
| 2848 |
+
"grad_norm": 4.207761764526367,
|
| 2849 |
+
"learning_rate": 1.4998085986963283e-06,
|
| 2850 |
+
"loss": 0.4188,
|
| 2851 |
+
"step": 406
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.9255258669698693,
|
| 2855 |
+
"grad_norm": 4.310345649719238,
|
| 2856 |
+
"learning_rate": 1.413293891264722e-06,
|
| 2857 |
+
"loss": 0.3598,
|
| 2858 |
+
"step": 407
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9277998862990335,
|
| 2862 |
+
"grad_norm": 2.979948043823242,
|
| 2863 |
+
"learning_rate": 1.3293133599946329e-06,
|
| 2864 |
+
"loss": 0.2512,
|
| 2865 |
+
"step": 408
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9300739056281978,
|
| 2869 |
+
"grad_norm": 2.9833507537841797,
|
| 2870 |
+
"learning_rate": 1.2478713851266088e-06,
|
| 2871 |
+
"loss": 0.305,
|
| 2872 |
+
"step": 409
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9323479249573622,
|
| 2876 |
+
"grad_norm": 3.2704060077667236,
|
| 2877 |
+
"learning_rate": 1.1689722144956671e-06,
|
| 2878 |
+
"loss": 0.431,
|
| 2879 |
+
"step": 410
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9346219442865265,
|
| 2883 |
+
"grad_norm": 3.7926714420318604,
|
| 2884 |
+
"learning_rate": 1.0926199633097157e-06,
|
| 2885 |
+
"loss": 0.3448,
|
| 2886 |
+
"step": 411
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.9368959636156907,
|
| 2890 |
+
"grad_norm": 3.6115946769714355,
|
| 2891 |
+
"learning_rate": 1.0188186139349354e-06,
|
| 2892 |
+
"loss": 0.4043,
|
| 2893 |
+
"step": 412
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.939169982944855,
|
| 2897 |
+
"grad_norm": 1.762302041053772,
|
| 2898 |
+
"learning_rate": 9.475720156880419e-07,
|
| 2899 |
+
"loss": 0.099,
|
| 2900 |
+
"step": 413
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9414440022740194,
|
| 2904 |
+
"grad_norm": 0.5029296278953552,
|
| 2905 |
+
"learning_rate": 8.788838846355341e-07,
|
| 2906 |
+
"loss": 0.0218,
|
| 2907 |
+
"step": 414
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9437180216031836,
|
| 2911 |
+
"grad_norm": 0.4769535958766937,
|
| 2912 |
+
"learning_rate": 8.127578033998662e-07,
|
| 2913 |
+
"loss": 0.0152,
|
| 2914 |
+
"step": 415
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9459920409323479,
|
| 2918 |
+
"grad_norm": 0.8436768651008606,
|
| 2919 |
+
"learning_rate": 7.491972209725806e-07,
|
| 2920 |
+
"loss": 0.0319,
|
| 2921 |
+
"step": 416
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.9482660602615123,
|
| 2925 |
+
"grad_norm": 0.6843982934951782,
|
| 2926 |
+
"learning_rate": 6.88205452534435e-07,
|
| 2927 |
+
"loss": 0.038,
|
| 2928 |
+
"step": 417
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9505400795906765,
|
| 2932 |
+
"grad_norm": 0.6065624356269836,
|
| 2933 |
+
"learning_rate": 6.297856792824741e-07,
|
| 2934 |
+
"loss": 0.0266,
|
| 2935 |
+
"step": 418
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9528140989198408,
|
| 2939 |
+
"grad_norm": 1.784542202949524,
|
| 2940 |
+
"learning_rate": 5.739409482640956e-07,
|
| 2941 |
+
"loss": 0.041,
|
| 2942 |
+
"step": 419
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.9550881182490051,
|
| 2946 |
+
"grad_norm": 0.4885413646697998,
|
| 2947 |
+
"learning_rate": 5.206741722181386e-07,
|
| 2948 |
+
"loss": 0.0198,
|
| 2949 |
+
"step": 420
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9573621375781695,
|
| 2953 |
+
"grad_norm": 0.47082552313804626,
|
| 2954 |
+
"learning_rate": 4.699881294229602e-07,
|
| 2955 |
+
"loss": 0.0176,
|
| 2956 |
+
"step": 421
|
| 2957 |
+
},
|
| 2958 |
+
{
|
| 2959 |
+
"epoch": 0.9596361569073337,
|
| 2960 |
+
"grad_norm": 0.7649410963058472,
|
| 2961 |
+
"learning_rate": 4.2188546355153013e-07,
|
| 2962 |
+
"loss": 0.0433,
|
| 2963 |
+
"step": 422
|
| 2964 |
+
},
|
| 2965 |
+
{
|
| 2966 |
+
"epoch": 0.961910176236498,
|
| 2967 |
+
"grad_norm": 2.893561840057373,
|
| 2968 |
+
"learning_rate": 3.763686835335345e-07,
|
| 2969 |
+
"loss": 0.2227,
|
| 2970 |
+
"step": 423
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.9641841955656623,
|
| 2974 |
+
"grad_norm": 2.788997173309326,
|
| 2975 |
+
"learning_rate": 3.334401634245032e-07,
|
| 2976 |
+
"loss": 0.226,
|
| 2977 |
+
"step": 424
|
| 2978 |
+
},
|
| 2979 |
+
{
|
| 2980 |
+
"epoch": 0.9664582148948266,
|
| 2981 |
+
"grad_norm": 5.2662739753723145,
|
| 2982 |
+
"learning_rate": 2.9310214228202013e-07,
|
| 2983 |
+
"loss": 0.5212,
|
| 2984 |
+
"step": 425
|
| 2985 |
+
},
|
| 2986 |
+
{
|
| 2987 |
+
"epoch": 0.9687322342239909,
|
| 2988 |
+
"grad_norm": 2.4451019763946533,
|
| 2989 |
+
"learning_rate": 2.553567240489052e-07,
|
| 2990 |
+
"loss": 0.235,
|
| 2991 |
+
"step": 426
|
| 2992 |
+
},
|
| 2993 |
+
{
|
| 2994 |
+
"epoch": 0.9710062535531552,
|
| 2995 |
+
"grad_norm": 3.208662748336792,
|
| 2996 |
+
"learning_rate": 2.202058774434912e-07,
|
| 2997 |
+
"loss": 0.2381,
|
| 2998 |
+
"step": 427
|
| 2999 |
+
},
|
| 3000 |
+
{
|
| 3001 |
+
"epoch": 0.9732802728823196,
|
| 3002 |
+
"grad_norm": 3.0693438053131104,
|
| 3003 |
+
"learning_rate": 1.8765143585693922e-07,
|
| 3004 |
+
"loss": 0.2596,
|
| 3005 |
+
"step": 428
|
| 3006 |
+
},
|
| 3007 |
+
{
|
| 3008 |
+
"epoch": 0.9755542922114838,
|
| 3009 |
+
"grad_norm": 4.01469612121582,
|
| 3010 |
+
"learning_rate": 1.5769509725760966e-07,
|
| 3011 |
+
"loss": 0.2541,
|
| 3012 |
+
"step": 429
|
| 3013 |
+
},
|
| 3014 |
+
{
|
| 3015 |
+
"epoch": 0.9778283115406481,
|
| 3016 |
+
"grad_norm": 2.0251407623291016,
|
| 3017 |
+
"learning_rate": 1.3033842410251075e-07,
|
| 3018 |
+
"loss": 0.1398,
|
| 3019 |
+
"step": 430
|
| 3020 |
+
},
|
| 3021 |
+
{
|
| 3022 |
+
"epoch": 0.9801023308698124,
|
| 3023 |
+
"grad_norm": 1.4239128828048706,
|
| 3024 |
+
"learning_rate": 1.0558284325578038e-07,
|
| 3025 |
+
"loss": 0.0793,
|
| 3026 |
+
"step": 431
|
| 3027 |
+
},
|
| 3028 |
+
{
|
| 3029 |
+
"epoch": 0.9823763501989767,
|
| 3030 |
+
"grad_norm": 1.6692121028900146,
|
| 3031 |
+
"learning_rate": 8.342964591430136e-08,
|
| 3032 |
+
"loss": 0.1298,
|
| 3033 |
+
"step": 432
|
| 3034 |
+
},
|
| 3035 |
+
{
|
| 3036 |
+
"epoch": 0.984650369528141,
|
| 3037 |
+
"grad_norm": 1.8359993696212769,
|
| 3038 |
+
"learning_rate": 6.38799875403051e-08,
|
| 3039 |
+
"loss": 0.0778,
|
| 3040 |
+
"step": 433
|
| 3041 |
+
},
|
| 3042 |
+
{
|
| 3043 |
+
"epoch": 0.9869243888573053,
|
| 3044 |
+
"grad_norm": 1.6211203336715698,
|
| 3045 |
+
"learning_rate": 4.6934887801164396e-08,
|
| 3046 |
+
"loss": 0.148,
|
| 3047 |
+
"step": 434
|
| 3048 |
+
},
|
| 3049 |
+
{
|
| 3050 |
+
"epoch": 0.9891984081864695,
|
| 3051 |
+
"grad_norm": 0.7776852250099182,
|
| 3052 |
+
"learning_rate": 3.259523051615254e-08,
|
| 3053 |
+
"loss": 0.0437,
|
| 3054 |
+
"step": 435
|
| 3055 |
+
},
|
| 3056 |
+
{
|
| 3057 |
+
"epoch": 0.9914724275156339,
|
| 3058 |
+
"grad_norm": 0.6490265727043152,
|
| 3059 |
+
"learning_rate": 2.086176361038583e-08,
|
| 3060 |
+
"loss": 0.0175,
|
| 3061 |
+
"step": 436
|
| 3062 |
+
},
|
| 3063 |
+
{
|
| 3064 |
+
"epoch": 0.9937464468447982,
|
| 3065 |
+
"grad_norm": 1.8342341184616089,
|
| 3066 |
+
"learning_rate": 1.173509907579362e-08,
|
| 3067 |
+
"loss": 0.0702,
|
| 3068 |
+
"step": 437
|
| 3069 |
+
},
|
| 3070 |
+
{
|
| 3071 |
+
"epoch": 0.9960204661739624,
|
| 3072 |
+
"grad_norm": 1.4983391761779785,
|
| 3073 |
+
"learning_rate": 5.215712939210526e-09,
|
| 3074 |
+
"loss": 0.0609,
|
| 3075 |
+
"step": 438
|
| 3076 |
+
},
|
| 3077 |
+
{
|
| 3078 |
+
"epoch": 0.9982944855031268,
|
| 3079 |
+
"grad_norm": 0.8002070784568787,
|
| 3080 |
+
"learning_rate": 1.3039452375351868e-09,
|
| 3081 |
+
"loss": 0.0323,
|
| 3082 |
+
"step": 439
|
| 3083 |
+
},
|
| 3084 |
+
{
|
| 3085 |
+
"epoch": 0.9982944855031268,
|
| 3086 |
+
"eval_loss": 0.180589497089386,
|
| 3087 |
+
"eval_runtime": 13.3573,
|
| 3088 |
+
"eval_samples_per_second": 27.775,
|
| 3089 |
+
"eval_steps_per_second": 6.962,
|
| 3090 |
+
"step": 439
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"epoch": 1.0017055144968732,
|
| 3094 |
+
"grad_norm": 3.280466079711914,
|
| 3095 |
+
"learning_rate": 0.0,
|
| 3096 |
+
"loss": 0.4975,
|
| 3097 |
+
"step": 440
|
| 3098 |
}
|
| 3099 |
],
|
| 3100 |
"logging_steps": 1,
|
|
|
|
| 3109 |
"should_evaluate": false,
|
| 3110 |
"should_log": false,
|
| 3111 |
"should_save": true,
|
| 3112 |
+
"should_training_stop": true
|
| 3113 |
},
|
| 3114 |
"attributes": {}
|
| 3115 |
}
|
| 3116 |
},
|
| 3117 |
+
"total_flos": 9.712732842491904e+16,
|
| 3118 |
"train_batch_size": 4,
|
| 3119 |
"trial_name": null,
|
| 3120 |
"trial_params": null
|