Training in progress, step 3000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 536223056
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6c549f8867ec3aa46fdb16d23e60b8f0ae222fa21f2d19da894e88c1f3b09c3
|
| 3 |
size 536223056
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1072594443
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53021cc365fb35689cf7935de3e1b4f7d09b54591f9f81b2f7c83736f1ee6045
|
| 3 |
size 1072594443
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:852ff1feb145f352899b6aa5117c88c8890d68604ca6bf2baf1e72eb1508c72e
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -2258,6 +2258,456 @@
|
|
| 2258 |
"mean_token_accuracy": 0.7839685261249543,
|
| 2259 |
"num_tokens": 2765568.0,
|
| 2260 |
"step": 2500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2261 |
}
|
| 2262 |
],
|
| 2263 |
"logging_steps": 10,
|
|
@@ -2277,7 +2727,7 @@
|
|
| 2277 |
"attributes": {}
|
| 2278 |
}
|
| 2279 |
},
|
| 2280 |
-
"total_flos":
|
| 2281 |
"train_batch_size": 8,
|
| 2282 |
"trial_name": null,
|
| 2283 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.6044731009470079,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 3000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 2258 |
"mean_token_accuracy": 0.7839685261249543,
|
| 2259 |
"num_tokens": 2765568.0,
|
| 2260 |
"step": 2500
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 0.5057424944589965,
|
| 2264 |
+
"grad_norm": 10.5,
|
| 2265 |
+
"learning_rate": 1.6629726643831016e-05,
|
| 2266 |
+
"loss": 0.9455,
|
| 2267 |
+
"mean_token_accuracy": 0.7747004866600037,
|
| 2268 |
+
"num_tokens": 2777172.0,
|
| 2269 |
+
"step": 2510
|
| 2270 |
+
},
|
| 2271 |
+
{
|
| 2272 |
+
"epoch": 0.5077574047954866,
|
| 2273 |
+
"grad_norm": 10.6875,
|
| 2274 |
+
"learning_rate": 1.6616293908254418e-05,
|
| 2275 |
+
"loss": 0.9188,
|
| 2276 |
+
"mean_token_accuracy": 0.7760649502277375,
|
| 2277 |
+
"num_tokens": 2789318.0,
|
| 2278 |
+
"step": 2520
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.5097723151319766,
|
| 2282 |
+
"grad_norm": 12.625,
|
| 2283 |
+
"learning_rate": 1.6602861172677817e-05,
|
| 2284 |
+
"loss": 0.9055,
|
| 2285 |
+
"mean_token_accuracy": 0.7822152853012085,
|
| 2286 |
+
"num_tokens": 2800307.0,
|
| 2287 |
+
"step": 2530
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 0.5117872254684667,
|
| 2291 |
+
"grad_norm": 12.9375,
|
| 2292 |
+
"learning_rate": 1.6589428437101216e-05,
|
| 2293 |
+
"loss": 0.9045,
|
| 2294 |
+
"mean_token_accuracy": 0.7808859765529632,
|
| 2295 |
+
"num_tokens": 2811228.0,
|
| 2296 |
+
"step": 2540
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"epoch": 0.5138021358049567,
|
| 2300 |
+
"grad_norm": 11.25,
|
| 2301 |
+
"learning_rate": 1.6575995701524618e-05,
|
| 2302 |
+
"loss": 0.8518,
|
| 2303 |
+
"mean_token_accuracy": 0.7890688478946686,
|
| 2304 |
+
"num_tokens": 2821512.0,
|
| 2305 |
+
"step": 2550
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.5158170461414467,
|
| 2309 |
+
"grad_norm": 13.5,
|
| 2310 |
+
"learning_rate": 1.6562562965948017e-05,
|
| 2311 |
+
"loss": 0.9224,
|
| 2312 |
+
"mean_token_accuracy": 0.7703654527664184,
|
| 2313 |
+
"num_tokens": 2832524.0,
|
| 2314 |
+
"step": 2560
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 0.5178319564779368,
|
| 2318 |
+
"grad_norm": 11.1875,
|
| 2319 |
+
"learning_rate": 1.6549130230371416e-05,
|
| 2320 |
+
"loss": 0.9554,
|
| 2321 |
+
"mean_token_accuracy": 0.7688623070716858,
|
| 2322 |
+
"num_tokens": 2844366.0,
|
| 2323 |
+
"step": 2570
|
| 2324 |
+
},
|
| 2325 |
+
{
|
| 2326 |
+
"epoch": 0.5198468668144267,
|
| 2327 |
+
"grad_norm": 16.25,
|
| 2328 |
+
"learning_rate": 1.6535697494794815e-05,
|
| 2329 |
+
"loss": 0.8999,
|
| 2330 |
+
"mean_token_accuracy": 0.7806227684020997,
|
| 2331 |
+
"num_tokens": 2855210.0,
|
| 2332 |
+
"step": 2580
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 0.5218617771509168,
|
| 2336 |
+
"grad_norm": 12.25,
|
| 2337 |
+
"learning_rate": 1.6522264759218217e-05,
|
| 2338 |
+
"loss": 0.951,
|
| 2339 |
+
"mean_token_accuracy": 0.7751711547374726,
|
| 2340 |
+
"num_tokens": 2865914.0,
|
| 2341 |
+
"step": 2590
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.5238766874874068,
|
| 2345 |
+
"grad_norm": 13.375,
|
| 2346 |
+
"learning_rate": 1.6508832023641616e-05,
|
| 2347 |
+
"loss": 0.951,
|
| 2348 |
+
"mean_token_accuracy": 0.7700467705726624,
|
| 2349 |
+
"num_tokens": 2875344.0,
|
| 2350 |
+
"step": 2600
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 0.5258915978238968,
|
| 2354 |
+
"grad_norm": 11.25,
|
| 2355 |
+
"learning_rate": 1.6495399288065014e-05,
|
| 2356 |
+
"loss": 0.9551,
|
| 2357 |
+
"mean_token_accuracy": 0.7730932533740997,
|
| 2358 |
+
"num_tokens": 2886246.0,
|
| 2359 |
+
"step": 2610
|
| 2360 |
+
},
|
| 2361 |
+
{
|
| 2362 |
+
"epoch": 0.5279065081603869,
|
| 2363 |
+
"grad_norm": 11.25,
|
| 2364 |
+
"learning_rate": 1.6481966552488417e-05,
|
| 2365 |
+
"loss": 0.8656,
|
| 2366 |
+
"mean_token_accuracy": 0.7870718777179718,
|
| 2367 |
+
"num_tokens": 2896640.0,
|
| 2368 |
+
"step": 2620
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.5299214184968769,
|
| 2372 |
+
"grad_norm": 13.5,
|
| 2373 |
+
"learning_rate": 1.6468533816911816e-05,
|
| 2374 |
+
"loss": 0.8316,
|
| 2375 |
+
"mean_token_accuracy": 0.790769350528717,
|
| 2376 |
+
"num_tokens": 2907645.0,
|
| 2377 |
+
"step": 2630
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 0.531936328833367,
|
| 2381 |
+
"grad_norm": 10.5,
|
| 2382 |
+
"learning_rate": 1.6455101081335218e-05,
|
| 2383 |
+
"loss": 0.8114,
|
| 2384 |
+
"mean_token_accuracy": 0.7977836310863495,
|
| 2385 |
+
"num_tokens": 2918502.0,
|
| 2386 |
+
"step": 2640
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 0.533951239169857,
|
| 2390 |
+
"grad_norm": 13.0,
|
| 2391 |
+
"learning_rate": 1.6441668345758613e-05,
|
| 2392 |
+
"loss": 0.7938,
|
| 2393 |
+
"mean_token_accuracy": 0.8016635835170746,
|
| 2394 |
+
"num_tokens": 2929723.0,
|
| 2395 |
+
"step": 2650
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 0.535966149506347,
|
| 2399 |
+
"grad_norm": 8.8125,
|
| 2400 |
+
"learning_rate": 1.6428235610182015e-05,
|
| 2401 |
+
"loss": 0.8881,
|
| 2402 |
+
"mean_token_accuracy": 0.7800805151462555,
|
| 2403 |
+
"num_tokens": 2941026.0,
|
| 2404 |
+
"step": 2660
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.537981059842837,
|
| 2408 |
+
"grad_norm": 10.75,
|
| 2409 |
+
"learning_rate": 1.6414802874605414e-05,
|
| 2410 |
+
"loss": 0.98,
|
| 2411 |
+
"mean_token_accuracy": 0.7662722408771515,
|
| 2412 |
+
"num_tokens": 2952300.0,
|
| 2413 |
+
"step": 2670
|
| 2414 |
+
},
|
| 2415 |
+
{
|
| 2416 |
+
"epoch": 0.539995970179327,
|
| 2417 |
+
"grad_norm": 11.1875,
|
| 2418 |
+
"learning_rate": 1.6401370139028813e-05,
|
| 2419 |
+
"loss": 0.9393,
|
| 2420 |
+
"mean_token_accuracy": 0.7773295342922211,
|
| 2421 |
+
"num_tokens": 2963688.0,
|
| 2422 |
+
"step": 2680
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 0.542010880515817,
|
| 2426 |
+
"grad_norm": 10.5,
|
| 2427 |
+
"learning_rate": 1.6387937403452215e-05,
|
| 2428 |
+
"loss": 0.8579,
|
| 2429 |
+
"mean_token_accuracy": 0.7857769846916198,
|
| 2430 |
+
"num_tokens": 2975798.0,
|
| 2431 |
+
"step": 2690
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.5440257908523071,
|
| 2435 |
+
"grad_norm": 11.5625,
|
| 2436 |
+
"learning_rate": 1.6374504667875614e-05,
|
| 2437 |
+
"loss": 0.8915,
|
| 2438 |
+
"mean_token_accuracy": 0.7862484276294708,
|
| 2439 |
+
"num_tokens": 2986958.0,
|
| 2440 |
+
"step": 2700
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 0.5460407011887971,
|
| 2444 |
+
"grad_norm": 14.0625,
|
| 2445 |
+
"learning_rate": 1.6361071932299013e-05,
|
| 2446 |
+
"loss": 0.944,
|
| 2447 |
+
"mean_token_accuracy": 0.774361002445221,
|
| 2448 |
+
"num_tokens": 2998027.0,
|
| 2449 |
+
"step": 2710
|
| 2450 |
+
},
|
| 2451 |
+
{
|
| 2452 |
+
"epoch": 0.5480556115252871,
|
| 2453 |
+
"grad_norm": 9.8125,
|
| 2454 |
+
"learning_rate": 1.6347639196722412e-05,
|
| 2455 |
+
"loss": 0.8616,
|
| 2456 |
+
"mean_token_accuracy": 0.7841140806674958,
|
| 2457 |
+
"num_tokens": 3008990.0,
|
| 2458 |
+
"step": 2720
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 0.5500705218617772,
|
| 2462 |
+
"grad_norm": 11.625,
|
| 2463 |
+
"learning_rate": 1.6334206461145814e-05,
|
| 2464 |
+
"loss": 0.995,
|
| 2465 |
+
"mean_token_accuracy": 0.764478224515915,
|
| 2466 |
+
"num_tokens": 3020180.0,
|
| 2467 |
+
"step": 2730
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.5520854321982672,
|
| 2471 |
+
"grad_norm": 16.375,
|
| 2472 |
+
"learning_rate": 1.6320773725569213e-05,
|
| 2473 |
+
"loss": 0.9769,
|
| 2474 |
+
"mean_token_accuracy": 0.762674230337143,
|
| 2475 |
+
"num_tokens": 3031468.0,
|
| 2476 |
+
"step": 2740
|
| 2477 |
+
},
|
| 2478 |
+
{
|
| 2479 |
+
"epoch": 0.5541003425347572,
|
| 2480 |
+
"grad_norm": 11.0625,
|
| 2481 |
+
"learning_rate": 1.6307340989992612e-05,
|
| 2482 |
+
"loss": 0.9019,
|
| 2483 |
+
"mean_token_accuracy": 0.7809956490993499,
|
| 2484 |
+
"num_tokens": 3043604.0,
|
| 2485 |
+
"step": 2750
|
| 2486 |
+
},
|
| 2487 |
+
{
|
| 2488 |
+
"epoch": 0.5561152528712472,
|
| 2489 |
+
"grad_norm": 10.3125,
|
| 2490 |
+
"learning_rate": 1.6293908254416014e-05,
|
| 2491 |
+
"loss": 0.8849,
|
| 2492 |
+
"mean_token_accuracy": 0.7833445549011231,
|
| 2493 |
+
"num_tokens": 3054038.0,
|
| 2494 |
+
"step": 2760
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.5581301632077372,
|
| 2498 |
+
"grad_norm": 13.3125,
|
| 2499 |
+
"learning_rate": 1.6280475518839413e-05,
|
| 2500 |
+
"loss": 0.9858,
|
| 2501 |
+
"mean_token_accuracy": 0.7633112788200378,
|
| 2502 |
+
"num_tokens": 3064515.0,
|
| 2503 |
+
"step": 2770
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 0.5601450735442273,
|
| 2507 |
+
"grad_norm": 12.3125,
|
| 2508 |
+
"learning_rate": 1.626704278326281e-05,
|
| 2509 |
+
"loss": 0.8388,
|
| 2510 |
+
"mean_token_accuracy": 0.7929341971874238,
|
| 2511 |
+
"num_tokens": 3075661.0,
|
| 2512 |
+
"step": 2780
|
| 2513 |
+
},
|
| 2514 |
+
{
|
| 2515 |
+
"epoch": 0.5621599838807173,
|
| 2516 |
+
"grad_norm": 11.6875,
|
| 2517 |
+
"learning_rate": 1.6253610047686214e-05,
|
| 2518 |
+
"loss": 0.9444,
|
| 2519 |
+
"mean_token_accuracy": 0.7728216648101807,
|
| 2520 |
+
"num_tokens": 3087402.0,
|
| 2521 |
+
"step": 2790
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 0.5641748942172073,
|
| 2525 |
+
"grad_norm": 15.3125,
|
| 2526 |
+
"learning_rate": 1.6240177312109613e-05,
|
| 2527 |
+
"loss": 0.9326,
|
| 2528 |
+
"mean_token_accuracy": 0.7675645828247071,
|
| 2529 |
+
"num_tokens": 3097807.0,
|
| 2530 |
+
"step": 2800
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.5661898045536974,
|
| 2534 |
+
"grad_norm": 11.6875,
|
| 2535 |
+
"learning_rate": 1.6226744576533015e-05,
|
| 2536 |
+
"loss": 0.8888,
|
| 2537 |
+
"mean_token_accuracy": 0.7739431619644165,
|
| 2538 |
+
"num_tokens": 3109256.0,
|
| 2539 |
+
"step": 2810
|
| 2540 |
+
},
|
| 2541 |
+
{
|
| 2542 |
+
"epoch": 0.5682047148901874,
|
| 2543 |
+
"grad_norm": 9.9375,
|
| 2544 |
+
"learning_rate": 1.621331184095641e-05,
|
| 2545 |
+
"loss": 0.8275,
|
| 2546 |
+
"mean_token_accuracy": 0.796435010433197,
|
| 2547 |
+
"num_tokens": 3119274.0,
|
| 2548 |
+
"step": 2820
|
| 2549 |
+
},
|
| 2550 |
+
{
|
| 2551 |
+
"epoch": 0.5702196252266775,
|
| 2552 |
+
"grad_norm": 11.8125,
|
| 2553 |
+
"learning_rate": 1.6199879105379813e-05,
|
| 2554 |
+
"loss": 0.9389,
|
| 2555 |
+
"mean_token_accuracy": 0.7698701798915863,
|
| 2556 |
+
"num_tokens": 3129740.0,
|
| 2557 |
+
"step": 2830
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.5722345355631674,
|
| 2561 |
+
"grad_norm": 10.5625,
|
| 2562 |
+
"learning_rate": 1.618644636980321e-05,
|
| 2563 |
+
"loss": 0.9461,
|
| 2564 |
+
"mean_token_accuracy": 0.7643969297409058,
|
| 2565 |
+
"num_tokens": 3141041.0,
|
| 2566 |
+
"step": 2840
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 0.5742494458996574,
|
| 2570 |
+
"grad_norm": 13.8125,
|
| 2571 |
+
"learning_rate": 1.617301363422661e-05,
|
| 2572 |
+
"loss": 1.0038,
|
| 2573 |
+
"mean_token_accuracy": 0.7637837052345275,
|
| 2574 |
+
"num_tokens": 3151060.0,
|
| 2575 |
+
"step": 2850
|
| 2576 |
+
},
|
| 2577 |
+
{
|
| 2578 |
+
"epoch": 0.5762643562361475,
|
| 2579 |
+
"grad_norm": 10.75,
|
| 2580 |
+
"learning_rate": 1.6159580898650012e-05,
|
| 2581 |
+
"loss": 0.862,
|
| 2582 |
+
"mean_token_accuracy": 0.785337769985199,
|
| 2583 |
+
"num_tokens": 3162459.0,
|
| 2584 |
+
"step": 2860
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 0.5782792665726375,
|
| 2588 |
+
"grad_norm": 10.5625,
|
| 2589 |
+
"learning_rate": 1.614614816307341e-05,
|
| 2590 |
+
"loss": 0.9322,
|
| 2591 |
+
"mean_token_accuracy": 0.7696834802627563,
|
| 2592 |
+
"num_tokens": 3172476.0,
|
| 2593 |
+
"step": 2870
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.5802941769091275,
|
| 2597 |
+
"grad_norm": 13.3125,
|
| 2598 |
+
"learning_rate": 1.613271542749681e-05,
|
| 2599 |
+
"loss": 0.9818,
|
| 2600 |
+
"mean_token_accuracy": 0.7669282913208008,
|
| 2601 |
+
"num_tokens": 3182571.0,
|
| 2602 |
+
"step": 2880
|
| 2603 |
+
},
|
| 2604 |
+
{
|
| 2605 |
+
"epoch": 0.5823090872456176,
|
| 2606 |
+
"grad_norm": 10.0,
|
| 2607 |
+
"learning_rate": 1.611928269192021e-05,
|
| 2608 |
+
"loss": 0.8449,
|
| 2609 |
+
"mean_token_accuracy": 0.7930525064468383,
|
| 2610 |
+
"num_tokens": 3194456.0,
|
| 2611 |
+
"step": 2890
|
| 2612 |
+
},
|
| 2613 |
+
{
|
| 2614 |
+
"epoch": 0.5843239975821076,
|
| 2615 |
+
"grad_norm": 10.4375,
|
| 2616 |
+
"learning_rate": 1.610584995634361e-05,
|
| 2617 |
+
"loss": 0.862,
|
| 2618 |
+
"mean_token_accuracy": 0.7849324703216553,
|
| 2619 |
+
"num_tokens": 3204377.0,
|
| 2620 |
+
"step": 2900
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.5863389079185977,
|
| 2624 |
+
"grad_norm": 11.875,
|
| 2625 |
+
"learning_rate": 1.609241722076701e-05,
|
| 2626 |
+
"loss": 0.8853,
|
| 2627 |
+
"mean_token_accuracy": 0.7843615412712097,
|
| 2628 |
+
"num_tokens": 3216082.0,
|
| 2629 |
+
"step": 2910
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 0.5883538182550877,
|
| 2633 |
+
"grad_norm": 9.375,
|
| 2634 |
+
"learning_rate": 1.607898448519041e-05,
|
| 2635 |
+
"loss": 0.7821,
|
| 2636 |
+
"mean_token_accuracy": 0.7997645199298858,
|
| 2637 |
+
"num_tokens": 3226582.0,
|
| 2638 |
+
"step": 2920
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 0.5903687285915776,
|
| 2642 |
+
"grad_norm": 9.1875,
|
| 2643 |
+
"learning_rate": 1.606555174961381e-05,
|
| 2644 |
+
"loss": 0.897,
|
| 2645 |
+
"mean_token_accuracy": 0.7760909557342529,
|
| 2646 |
+
"num_tokens": 3238387.0,
|
| 2647 |
+
"step": 2930
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"epoch": 0.5923836389280677,
|
| 2651 |
+
"grad_norm": 12.0,
|
| 2652 |
+
"learning_rate": 1.605211901403721e-05,
|
| 2653 |
+
"loss": 0.9027,
|
| 2654 |
+
"mean_token_accuracy": 0.78245330452919,
|
| 2655 |
+
"num_tokens": 3248670.0,
|
| 2656 |
+
"step": 2940
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.5943985492645577,
|
| 2660 |
+
"grad_norm": 10.0625,
|
| 2661 |
+
"learning_rate": 1.603868627846061e-05,
|
| 2662 |
+
"loss": 0.7964,
|
| 2663 |
+
"mean_token_accuracy": 0.8025750041007995,
|
| 2664 |
+
"num_tokens": 3259282.0,
|
| 2665 |
+
"step": 2950
|
| 2666 |
+
},
|
| 2667 |
+
{
|
| 2668 |
+
"epoch": 0.5964134596010477,
|
| 2669 |
+
"grad_norm": 10.6875,
|
| 2670 |
+
"learning_rate": 1.602525354288401e-05,
|
| 2671 |
+
"loss": 0.8159,
|
| 2672 |
+
"mean_token_accuracy": 0.7951205134391784,
|
| 2673 |
+
"num_tokens": 3270771.0,
|
| 2674 |
+
"step": 2960
|
| 2675 |
+
},
|
| 2676 |
+
{
|
| 2677 |
+
"epoch": 0.5984283699375378,
|
| 2678 |
+
"grad_norm": 12.75,
|
| 2679 |
+
"learning_rate": 1.601182080730741e-05,
|
| 2680 |
+
"loss": 1.0354,
|
| 2681 |
+
"mean_token_accuracy": 0.7645578503608703,
|
| 2682 |
+
"num_tokens": 3282321.0,
|
| 2683 |
+
"step": 2970
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.6004432802740278,
|
| 2687 |
+
"grad_norm": 14.6875,
|
| 2688 |
+
"learning_rate": 1.599838807173081e-05,
|
| 2689 |
+
"loss": 0.9655,
|
| 2690 |
+
"mean_token_accuracy": 0.7646917760372162,
|
| 2691 |
+
"num_tokens": 3294284.0,
|
| 2692 |
+
"step": 2980
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 0.6024581906105179,
|
| 2696 |
+
"grad_norm": 11.375,
|
| 2697 |
+
"learning_rate": 1.5984955336154207e-05,
|
| 2698 |
+
"loss": 0.9532,
|
| 2699 |
+
"mean_token_accuracy": 0.7740876019001007,
|
| 2700 |
+
"num_tokens": 3305259.0,
|
| 2701 |
+
"step": 2990
|
| 2702 |
+
},
|
| 2703 |
+
{
|
| 2704 |
+
"epoch": 0.6044731009470079,
|
| 2705 |
+
"grad_norm": 11.5625,
|
| 2706 |
+
"learning_rate": 1.597152260057761e-05,
|
| 2707 |
+
"loss": 0.8269,
|
| 2708 |
+
"mean_token_accuracy": 0.7911386549472809,
|
| 2709 |
+
"num_tokens": 3316348.0,
|
| 2710 |
+
"step": 3000
|
| 2711 |
}
|
| 2712 |
],
|
| 2713 |
"logging_steps": 10,
|
|
|
|
| 2727 |
"attributes": {}
|
| 2728 |
}
|
| 2729 |
},
|
| 2730 |
+
"total_flos": 4014132187054080.0,
|
| 2731 |
"train_batch_size": 8,
|
| 2732 |
"trial_name": null,
|
| 2733 |
"trial_params": null
|