Commit
•
d761575
1
Parent(s):
b65f702
Training in progress, step 37000
Browse files- adapter_model.bin +1 -1
- {checkpoint-33000 → checkpoint-36000/adapter_model}/README.md +0 -0
- {checkpoint-33000 → checkpoint-36000/adapter_model}/adapter_config.json +0 -0
- {checkpoint-33000 → checkpoint-36000/adapter_model}/adapter_model.bin +1 -1
- {checkpoint-33000/adapter_model → checkpoint-37000}/README.md +0 -0
- {checkpoint-33000/adapter_model → checkpoint-37000}/adapter_config.json +0 -0
- {checkpoint-33000/adapter_model → checkpoint-37000}/adapter_model.bin +1 -1
- {checkpoint-33000 → checkpoint-37000}/optimizer.pt +1 -1
- {checkpoint-33000 → checkpoint-37000}/rng_state.pth +1 -1
- {checkpoint-33000 → checkpoint-37000}/scheduler.pt +1 -1
- {checkpoint-33000 → checkpoint-37000}/trainer_state.json +277 -5
- {checkpoint-33000 → checkpoint-37000}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd9546c32a7c69f32e36c612415bdd464231e08b43267c0c48a31c7f40c0d408
|
| 3 |
size 500897101
|
{checkpoint-33000 → checkpoint-36000/adapter_model}/README.md
RENAMED
|
File without changes
|
{checkpoint-33000 → checkpoint-36000/adapter_model}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-33000 → checkpoint-36000/adapter_model}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23c2d32c288feead4318a69d3ff414e535f97637bea0a9ae28d2ce95cd3de348
|
| 3 |
size 500897101
|
{checkpoint-33000/adapter_model → checkpoint-37000}/README.md
RENAMED
|
File without changes
|
{checkpoint-33000/adapter_model → checkpoint-37000}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-33000/adapter_model → checkpoint-37000}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 500897101
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd9546c32a7c69f32e36c612415bdd464231e08b43267c0c48a31c7f40c0d408
|
| 3 |
size 500897101
|
{checkpoint-33000 → checkpoint-37000}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1001723453
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66752d28f3b7ca12e1b483bbb5bd48fa49998d34e026c908100c9cd122ff144f
|
| 3 |
size 1001723453
|
{checkpoint-33000 → checkpoint-37000}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14575
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12a1e84087c196e7e618617b884516dfa94b6d7a7e668c1d280c6b4c8239f845
|
| 3 |
size 14575
|
{checkpoint-33000 → checkpoint-37000}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d7046cc5cdc134adf08950434c6180a09920d2a73dfbc0c6da7d0543135abf2
|
| 3 |
size 627
|
{checkpoint-33000 → checkpoint-37000}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./qlora-out/checkpoint-
|
| 4 |
-
"epoch": 1.
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -2250,11 +2250,283 @@
|
|
| 2250 |
"eval_samples_per_second": 0.428,
|
| 2251 |
"eval_steps_per_second": 0.428,
|
| 2252 |
"step": 33000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2253 |
}
|
| 2254 |
],
|
| 2255 |
"max_steps": 80463,
|
| 2256 |
"num_train_epochs": 3,
|
| 2257 |
-
"total_flos":
|
| 2258 |
"trial_name": null,
|
| 2259 |
"trial_params": null
|
| 2260 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.5555862188339233,
|
| 3 |
+
"best_model_checkpoint": "./qlora-out/checkpoint-37000",
|
| 4 |
+
"epoch": 1.3795160508556727,
|
| 5 |
+
"global_step": 37000,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 2250 |
"eval_samples_per_second": 0.428,
|
| 2251 |
"eval_steps_per_second": 0.428,
|
| 2252 |
"step": 33000
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 1.23,
|
| 2256 |
+
"learning_rate": 0.0001275078855425007,
|
| 2257 |
+
"loss": 0.4971,
|
| 2258 |
+
"step": 33100
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 1.24,
|
| 2262 |
+
"learning_rate": 0.0001271322532054803,
|
| 2263 |
+
"loss": 0.4977,
|
| 2264 |
+
"step": 33200
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.24,
|
| 2268 |
+
"learning_rate": 0.0001267562071541254,
|
| 2269 |
+
"loss": 0.499,
|
| 2270 |
+
"step": 33300
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 1.25,
|
| 2274 |
+
"learning_rate": 0.00012637975312241022,
|
| 2275 |
+
"loss": 0.5044,
|
| 2276 |
+
"step": 33400
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 1.25,
|
| 2280 |
+
"learning_rate": 0.00012600289685052996,
|
| 2281 |
+
"loss": 0.5019,
|
| 2282 |
+
"step": 33500
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 1.25,
|
| 2286 |
+
"learning_rate": 0.00012562564408481327,
|
| 2287 |
+
"loss": 0.5225,
|
| 2288 |
+
"step": 33600
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 1.26,
|
| 2292 |
+
"learning_rate": 0.00012524800057763438,
|
| 2293 |
+
"loss": 0.5503,
|
| 2294 |
+
"step": 33700
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 1.26,
|
| 2298 |
+
"learning_rate": 0.00012486997208732573,
|
| 2299 |
+
"loss": 0.5025,
|
| 2300 |
+
"step": 33800
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 1.26,
|
| 2304 |
+
"learning_rate": 0.0001244915643780899,
|
| 2305 |
+
"loss": 0.5187,
|
| 2306 |
+
"step": 33900
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.27,
|
| 2310 |
+
"learning_rate": 0.00012411278321991195,
|
| 2311 |
+
"loss": 0.5199,
|
| 2312 |
+
"step": 34000
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 1.27,
|
| 2316 |
+
"eval_loss": 0.5665221810340881,
|
| 2317 |
+
"eval_runtime": 1263.0264,
|
| 2318 |
+
"eval_samples_per_second": 0.429,
|
| 2319 |
+
"eval_steps_per_second": 0.429,
|
| 2320 |
+
"step": 34000
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.27,
|
| 2324 |
+
"learning_rate": 0.00012373363438847117,
|
| 2325 |
+
"loss": 0.5135,
|
| 2326 |
+
"step": 34100
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 1.28,
|
| 2330 |
+
"learning_rate": 0.00012335412366505324,
|
| 2331 |
+
"loss": 0.5065,
|
| 2332 |
+
"step": 34200
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 1.28,
|
| 2336 |
+
"learning_rate": 0.000122974256836462,
|
| 2337 |
+
"loss": 0.5223,
|
| 2338 |
+
"step": 34300
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 1.28,
|
| 2342 |
+
"learning_rate": 0.00012259403969493114,
|
| 2343 |
+
"loss": 0.4946,
|
| 2344 |
+
"step": 34400
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"epoch": 1.29,
|
| 2348 |
+
"learning_rate": 0.00012221347803803605,
|
| 2349 |
+
"loss": 0.5105,
|
| 2350 |
+
"step": 34500
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 1.29,
|
| 2354 |
+
"learning_rate": 0.00012183257766860514,
|
| 2355 |
+
"loss": 0.4812,
|
| 2356 |
+
"step": 34600
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 1.29,
|
| 2360 |
+
"learning_rate": 0.00012145134439463178,
|
| 2361 |
+
"loss": 0.4981,
|
| 2362 |
+
"step": 34700
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.3,
|
| 2366 |
+
"learning_rate": 0.0001210697840291852,
|
| 2367 |
+
"loss": 0.5038,
|
| 2368 |
+
"step": 34800
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 1.3,
|
| 2372 |
+
"learning_rate": 0.00012068790239032241,
|
| 2373 |
+
"loss": 0.5551,
|
| 2374 |
+
"step": 34900
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 1.3,
|
| 2378 |
+
"learning_rate": 0.00012030570530099902,
|
| 2379 |
+
"loss": 0.4964,
|
| 2380 |
+
"step": 35000
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 1.3,
|
| 2384 |
+
"eval_loss": 0.562954843044281,
|
| 2385 |
+
"eval_runtime": 1252.1434,
|
| 2386 |
+
"eval_samples_per_second": 0.433,
|
| 2387 |
+
"eval_steps_per_second": 0.433,
|
| 2388 |
+
"step": 35000
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 1.31,
|
| 2392 |
+
"learning_rate": 0.00011992319858898077,
|
| 2393 |
+
"loss": 0.4952,
|
| 2394 |
+
"step": 35100
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 1.31,
|
| 2398 |
+
"learning_rate": 0.0001195403880867545,
|
| 2399 |
+
"loss": 0.5157,
|
| 2400 |
+
"step": 35200
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 1.32,
|
| 2404 |
+
"learning_rate": 0.00011915727963143922,
|
| 2405 |
+
"loss": 0.4973,
|
| 2406 |
+
"step": 35300
|
| 2407 |
+
},
|
| 2408 |
+
{
|
| 2409 |
+
"epoch": 1.32,
|
| 2410 |
+
"learning_rate": 0.00011877387906469721,
|
| 2411 |
+
"loss": 0.4884,
|
| 2412 |
+
"step": 35400
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 1.32,
|
| 2416 |
+
"learning_rate": 0.00011839019223264489,
|
| 2417 |
+
"loss": 0.5017,
|
| 2418 |
+
"step": 35500
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.33,
|
| 2422 |
+
"learning_rate": 0.00011800622498576363,
|
| 2423 |
+
"loss": 0.5157,
|
| 2424 |
+
"step": 35600
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 1.33,
|
| 2428 |
+
"learning_rate": 0.00011762198317881059,
|
| 2429 |
+
"loss": 0.4774,
|
| 2430 |
+
"step": 35700
|
| 2431 |
+
},
|
| 2432 |
+
{
|
| 2433 |
+
"epoch": 1.33,
|
| 2434 |
+
"learning_rate": 0.0001172374726707295,
|
| 2435 |
+
"loss": 0.4855,
|
| 2436 |
+
"step": 35800
|
| 2437 |
+
},
|
| 2438 |
+
{
|
| 2439 |
+
"epoch": 1.34,
|
| 2440 |
+
"learning_rate": 0.00011685269932456115,
|
| 2441 |
+
"loss": 0.5134,
|
| 2442 |
+
"step": 35900
|
| 2443 |
+
},
|
| 2444 |
+
{
|
| 2445 |
+
"epoch": 1.34,
|
| 2446 |
+
"learning_rate": 0.00011646766900735422,
|
| 2447 |
+
"loss": 0.5143,
|
| 2448 |
+
"step": 36000
|
| 2449 |
+
},
|
| 2450 |
+
{
|
| 2451 |
+
"epoch": 1.34,
|
| 2452 |
+
"eval_loss": 0.5594063997268677,
|
| 2453 |
+
"eval_runtime": 1270.0722,
|
| 2454 |
+
"eval_samples_per_second": 0.427,
|
| 2455 |
+
"eval_steps_per_second": 0.427,
|
| 2456 |
+
"step": 36000
|
| 2457 |
+
},
|
| 2458 |
+
{
|
| 2459 |
+
"epoch": 1.35,
|
| 2460 |
+
"learning_rate": 0.00011608238759007561,
|
| 2461 |
+
"loss": 0.5268,
|
| 2462 |
+
"step": 36100
|
| 2463 |
+
},
|
| 2464 |
+
{
|
| 2465 |
+
"epoch": 1.35,
|
| 2466 |
+
"learning_rate": 0.00011569686094752101,
|
| 2467 |
+
"loss": 0.5179,
|
| 2468 |
+
"step": 36200
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 1.35,
|
| 2472 |
+
"learning_rate": 0.00011531109495822545,
|
| 2473 |
+
"loss": 0.5236,
|
| 2474 |
+
"step": 36300
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.36,
|
| 2478 |
+
"learning_rate": 0.00011492509550437339,
|
| 2479 |
+
"loss": 0.5197,
|
| 2480 |
+
"step": 36400
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 1.36,
|
| 2484 |
+
"learning_rate": 0.0001145388684717092,
|
| 2485 |
+
"loss": 0.5109,
|
| 2486 |
+
"step": 36500
|
| 2487 |
+
},
|
| 2488 |
+
{
|
| 2489 |
+
"epoch": 1.36,
|
| 2490 |
+
"learning_rate": 0.00011415241974944744,
|
| 2491 |
+
"loss": 0.5126,
|
| 2492 |
+
"step": 36600
|
| 2493 |
+
},
|
| 2494 |
+
{
|
| 2495 |
+
"epoch": 1.37,
|
| 2496 |
+
"learning_rate": 0.00011376575523018296,
|
| 2497 |
+
"loss": 0.501,
|
| 2498 |
+
"step": 36700
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"epoch": 1.37,
|
| 2502 |
+
"learning_rate": 0.00011337888080980115,
|
| 2503 |
+
"loss": 0.4888,
|
| 2504 |
+
"step": 36800
|
| 2505 |
+
},
|
| 2506 |
+
{
|
| 2507 |
+
"epoch": 1.38,
|
| 2508 |
+
"learning_rate": 0.00011299180238738789,
|
| 2509 |
+
"loss": 0.5324,
|
| 2510 |
+
"step": 36900
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 1.38,
|
| 2514 |
+
"learning_rate": 0.00011260452586513981,
|
| 2515 |
+
"loss": 0.5053,
|
| 2516 |
+
"step": 37000
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.38,
|
| 2520 |
+
"eval_loss": 0.5555862188339233,
|
| 2521 |
+
"eval_runtime": 1324.5375,
|
| 2522 |
+
"eval_samples_per_second": 0.409,
|
| 2523 |
+
"eval_steps_per_second": 0.409,
|
| 2524 |
+
"step": 37000
|
| 2525 |
}
|
| 2526 |
],
|
| 2527 |
"max_steps": 80463,
|
| 2528 |
"num_train_epochs": 3,
|
| 2529 |
+
"total_flos": 1.0375131020284969e+19,
|
| 2530 |
"trial_name": null,
|
| 2531 |
"trial_params": null
|
| 2532 |
}
|
{checkpoint-33000 → checkpoint-37000}/training_args.bin
RENAMED
|
File without changes
|