Training in progress, step 8000, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -30,15 +30,15 @@
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
-
"
|
| 34 |
"in_proj",
|
|
|
|
|
|
|
|
|
|
| 35 |
"v_proj",
|
| 36 |
-
"out_proj",
|
| 37 |
-
"w2",
|
| 38 |
-
"w3",
|
| 39 |
"o_proj",
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
],
|
| 43 |
"target_parameters": null,
|
| 44 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 30 |
"rank_pattern": {},
|
| 31 |
"revision": null,
|
| 32 |
"target_modules": [
|
| 33 |
+
"w2",
|
| 34 |
"in_proj",
|
| 35 |
+
"q_proj",
|
| 36 |
+
"w1",
|
| 37 |
+
"k_proj",
|
| 38 |
"v_proj",
|
|
|
|
|
|
|
|
|
|
| 39 |
"o_proj",
|
| 40 |
+
"out_proj",
|
| 41 |
+
"w3"
|
| 42 |
],
|
| 43 |
"target_parameters": null,
|
| 44 |
"task_type": "CAUSAL_LM",
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 88891320
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32fdfa6d607ad598f4e8c531b1c07f733bf092edd93ea70312d65954b67c932c
|
| 3 |
size 88891320
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 177891851
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ca217cfdd9cf2cdae8efbfb4b8a543c3b3f455219a04f51eefd52551294fd90
|
| 3 |
size 177891851
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0be01244c3b2c763a45f990c45e6e145dc2010d1bf6211bd81b647c8e30c773e
|
| 3 |
size 14917
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86fac0e325ca2829d3f73d2ce18f2897fd117304f66721d087b399fca35f0658
|
| 3 |
size 14917
|
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:c6a70ee5567b8ca150b1299cf6ed0ac265ce3735a78460200573731c39896f60
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "./lfm25-kodcode-lora/checkpoint-
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 250,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -7838,6 +7838,528 @@
|
|
| 7838 |
"eval_samples_per_second": 5.968,
|
| 7839 |
"eval_steps_per_second": 1.493,
|
| 7840 |
"step": 7500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7841 |
}
|
| 7842 |
],
|
| 7843 |
"logging_steps": 10,
|
|
@@ -7857,7 +8379,7 @@
|
|
| 7857 |
"attributes": {}
|
| 7858 |
}
|
| 7859 |
},
|
| 7860 |
-
"total_flos": 6.
|
| 7861 |
"train_batch_size": 2,
|
| 7862 |
"trial_name": null,
|
| 7863 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 8000,
|
| 3 |
+
"best_metric": 0.2922365963459015,
|
| 4 |
+
"best_model_checkpoint": "./lfm25-kodcode-lora/checkpoint-8000",
|
| 5 |
+
"epoch": 0.6060720846985739,
|
| 6 |
"eval_steps": 250,
|
| 7 |
+
"global_step": 8000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 7838 |
"eval_samples_per_second": 5.968,
|
| 7839 |
"eval_steps_per_second": 1.493,
|
| 7840 |
"step": 7500
|
| 7841 |
+
},
|
| 7842 |
+
{
|
| 7843 |
+
"entropy": 0.3109708122909069,
|
| 7844 |
+
"epoch": 0.5689501695107861,
|
| 7845 |
+
"grad_norm": 0.4822841286659241,
|
| 7846 |
+
"learning_rate": 4.132161835697846e-05,
|
| 7847 |
+
"loss": 0.3667874336242676,
|
| 7848 |
+
"mean_token_accuracy": 0.9005939975380898,
|
| 7849 |
+
"num_tokens": 114968.0,
|
| 7850 |
+
"step": 7510
|
| 7851 |
+
},
|
| 7852 |
+
{
|
| 7853 |
+
"entropy": 0.3276409078389406,
|
| 7854 |
+
"epoch": 0.5697077596166594,
|
| 7855 |
+
"grad_norm": 0.4089864492416382,
|
| 7856 |
+
"learning_rate": 4.120082652019689e-05,
|
| 7857 |
+
"loss": 0.38182897567749025,
|
| 7858 |
+
"mean_token_accuracy": 0.8978955864906311,
|
| 7859 |
+
"num_tokens": 224970.0,
|
| 7860 |
+
"step": 7520
|
| 7861 |
+
},
|
| 7862 |
+
{
|
| 7863 |
+
"entropy": 0.2703943993896246,
|
| 7864 |
+
"epoch": 0.5704653497225326,
|
| 7865 |
+
"grad_norm": 0.39988771080970764,
|
| 7866 |
+
"learning_rate": 4.108008765586738e-05,
|
| 7867 |
+
"loss": 0.3071948528289795,
|
| 7868 |
+
"mean_token_accuracy": 0.9152684897184372,
|
| 7869 |
+
"num_tokens": 324147.0,
|
| 7870 |
+
"step": 7530
|
| 7871 |
+
},
|
| 7872 |
+
{
|
| 7873 |
+
"entropy": 0.2900085491128266,
|
| 7874 |
+
"epoch": 0.5712229398284059,
|
| 7875 |
+
"grad_norm": 0.40979936718940735,
|
| 7876 |
+
"learning_rate": 4.095940249085749e-05,
|
| 7877 |
+
"loss": 0.3266597270965576,
|
| 7878 |
+
"mean_token_accuracy": 0.9085231080651284,
|
| 7879 |
+
"num_tokens": 421752.0,
|
| 7880 |
+
"step": 7540
|
| 7881 |
+
},
|
| 7882 |
+
{
|
| 7883 |
+
"entropy": 0.2737187416292727,
|
| 7884 |
+
"epoch": 0.5719805299342791,
|
| 7885 |
+
"grad_norm": 0.4078211784362793,
|
| 7886 |
+
"learning_rate": 4.0838771751711505e-05,
|
| 7887 |
+
"loss": 0.3361893892288208,
|
| 7888 |
+
"mean_token_accuracy": 0.9113953113555908,
|
| 7889 |
+
"num_tokens": 517042.0,
|
| 7890 |
+
"step": 7550
|
| 7891 |
+
},
|
| 7892 |
+
{
|
| 7893 |
+
"entropy": 0.31109195090830327,
|
| 7894 |
+
"epoch": 0.5727381200401522,
|
| 7895 |
+
"grad_norm": 0.406087189912796,
|
| 7896 |
+
"learning_rate": 4.0718196164646036e-05,
|
| 7897 |
+
"loss": 0.389078950881958,
|
| 7898 |
+
"mean_token_accuracy": 0.8990072026848793,
|
| 7899 |
+
"num_tokens": 621106.0,
|
| 7900 |
+
"step": 7560
|
| 7901 |
+
},
|
| 7902 |
+
{
|
| 7903 |
+
"entropy": 0.2715472083538771,
|
| 7904 |
+
"epoch": 0.5734957101460255,
|
| 7905 |
+
"grad_norm": 0.48067522048950195,
|
| 7906 |
+
"learning_rate": 4.059767645554568e-05,
|
| 7907 |
+
"loss": 0.3081014633178711,
|
| 7908 |
+
"mean_token_accuracy": 0.914455634355545,
|
| 7909 |
+
"num_tokens": 720233.0,
|
| 7910 |
+
"step": 7570
|
| 7911 |
+
},
|
| 7912 |
+
{
|
| 7913 |
+
"entropy": 0.32704932652413843,
|
| 7914 |
+
"epoch": 0.5742533002518987,
|
| 7915 |
+
"grad_norm": 0.5467900633811951,
|
| 7916 |
+
"learning_rate": 4.047721334995865e-05,
|
| 7917 |
+
"loss": 0.376554012298584,
|
| 7918 |
+
"mean_token_accuracy": 0.8963387101888657,
|
| 7919 |
+
"num_tokens": 818271.0,
|
| 7920 |
+
"step": 7580
|
| 7921 |
+
},
|
| 7922 |
+
{
|
| 7923 |
+
"entropy": 0.3027558699250221,
|
| 7924 |
+
"epoch": 0.575010890357772,
|
| 7925 |
+
"grad_norm": 0.5470854640007019,
|
| 7926 |
+
"learning_rate": 4.035680757309238e-05,
|
| 7927 |
+
"loss": 0.34917449951171875,
|
| 7928 |
+
"mean_token_accuracy": 0.9045404583215714,
|
| 7929 |
+
"num_tokens": 918405.0,
|
| 7930 |
+
"step": 7590
|
| 7931 |
+
},
|
| 7932 |
+
{
|
| 7933 |
+
"entropy": 0.29923666920512915,
|
| 7934 |
+
"epoch": 0.5757684804636451,
|
| 7935 |
+
"grad_norm": 0.5230236053466797,
|
| 7936 |
+
"learning_rate": 4.023645984980919e-05,
|
| 7937 |
+
"loss": 0.34145679473876955,
|
| 7938 |
+
"mean_token_accuracy": 0.9045537799596787,
|
| 7939 |
+
"num_tokens": 1018239.0,
|
| 7940 |
+
"step": 7600
|
| 7941 |
+
},
|
| 7942 |
+
{
|
| 7943 |
+
"entropy": 0.31514527965337036,
|
| 7944 |
+
"epoch": 0.5765260705695183,
|
| 7945 |
+
"grad_norm": 0.3059269189834595,
|
| 7946 |
+
"learning_rate": 4.0116170904621896e-05,
|
| 7947 |
+
"loss": 0.37950544357299804,
|
| 7948 |
+
"mean_token_accuracy": 0.8975080713629723,
|
| 7949 |
+
"num_tokens": 1116705.0,
|
| 7950 |
+
"step": 7610
|
| 7951 |
+
},
|
| 7952 |
+
{
|
| 7953 |
+
"entropy": 0.26913794968277216,
|
| 7954 |
+
"epoch": 0.5772836606753916,
|
| 7955 |
+
"grad_norm": 0.5136588215827942,
|
| 7956 |
+
"learning_rate": 3.999594146168947e-05,
|
| 7957 |
+
"loss": 0.3233908176422119,
|
| 7958 |
+
"mean_token_accuracy": 0.9152696698904037,
|
| 7959 |
+
"num_tokens": 1213766.0,
|
| 7960 |
+
"step": 7620
|
| 7961 |
+
},
|
| 7962 |
+
{
|
| 7963 |
+
"entropy": 0.33589745350182054,
|
| 7964 |
+
"epoch": 0.5780412507812648,
|
| 7965 |
+
"grad_norm": 0.4772613048553467,
|
| 7966 |
+
"learning_rate": 3.987577224481266e-05,
|
| 7967 |
+
"loss": 0.3794508934020996,
|
| 7968 |
+
"mean_token_accuracy": 0.8917440369725227,
|
| 7969 |
+
"num_tokens": 1321188.0,
|
| 7970 |
+
"step": 7630
|
| 7971 |
+
},
|
| 7972 |
+
{
|
| 7973 |
+
"entropy": 0.2889779403805733,
|
| 7974 |
+
"epoch": 0.5787988408871381,
|
| 7975 |
+
"grad_norm": 0.36705854535102844,
|
| 7976 |
+
"learning_rate": 3.975566397742965e-05,
|
| 7977 |
+
"loss": 0.3197050094604492,
|
| 7978 |
+
"mean_token_accuracy": 0.9126046910881996,
|
| 7979 |
+
"num_tokens": 1417615.0,
|
| 7980 |
+
"step": 7640
|
| 7981 |
+
},
|
| 7982 |
+
{
|
| 7983 |
+
"entropy": 0.3254348419606686,
|
| 7984 |
+
"epoch": 0.5795564309930112,
|
| 7985 |
+
"grad_norm": 0.3900688886642456,
|
| 7986 |
+
"learning_rate": 3.963561738261173e-05,
|
| 7987 |
+
"loss": 0.37628955841064454,
|
| 7988 |
+
"mean_token_accuracy": 0.8971853539347648,
|
| 7989 |
+
"num_tokens": 1515666.0,
|
| 7990 |
+
"step": 7650
|
| 7991 |
+
},
|
| 7992 |
+
{
|
| 7993 |
+
"entropy": 0.29808914177119733,
|
| 7994 |
+
"epoch": 0.5803140210988844,
|
| 7995 |
+
"grad_norm": 0.4222153127193451,
|
| 7996 |
+
"learning_rate": 3.951563318305883e-05,
|
| 7997 |
+
"loss": 0.33919458389282225,
|
| 7998 |
+
"mean_token_accuracy": 0.9053486153483391,
|
| 7999 |
+
"num_tokens": 1616529.0,
|
| 8000 |
+
"step": 7660
|
| 8001 |
+
},
|
| 8002 |
+
{
|
| 8003 |
+
"entropy": 0.30099980514496566,
|
| 8004 |
+
"epoch": 0.5810716112047577,
|
| 8005 |
+
"grad_norm": 0.4373079538345337,
|
| 8006 |
+
"learning_rate": 3.939571210109532e-05,
|
| 8007 |
+
"loss": 0.3660151958465576,
|
| 8008 |
+
"mean_token_accuracy": 0.9015607729554176,
|
| 8009 |
+
"num_tokens": 1711473.0,
|
| 8010 |
+
"step": 7670
|
| 8011 |
+
},
|
| 8012 |
+
{
|
| 8013 |
+
"entropy": 0.25706873182207346,
|
| 8014 |
+
"epoch": 0.5818292013106309,
|
| 8015 |
+
"grad_norm": 0.49894335865974426,
|
| 8016 |
+
"learning_rate": 3.9275854858665564e-05,
|
| 8017 |
+
"loss": 0.3213352680206299,
|
| 8018 |
+
"mean_token_accuracy": 0.9181353032588959,
|
| 8019 |
+
"num_tokens": 1794864.0,
|
| 8020 |
+
"step": 7680
|
| 8021 |
+
},
|
| 8022 |
+
{
|
| 8023 |
+
"entropy": 0.29280857387930154,
|
| 8024 |
+
"epoch": 0.582586791416504,
|
| 8025 |
+
"grad_norm": 0.35843634605407715,
|
| 8026 |
+
"learning_rate": 3.915606217732962e-05,
|
| 8027 |
+
"loss": 0.32027444839477537,
|
| 8028 |
+
"mean_token_accuracy": 0.9076616898179054,
|
| 8029 |
+
"num_tokens": 1889475.0,
|
| 8030 |
+
"step": 7690
|
| 8031 |
+
},
|
| 8032 |
+
{
|
| 8033 |
+
"entropy": 0.2827717169187963,
|
| 8034 |
+
"epoch": 0.5833443815223773,
|
| 8035 |
+
"grad_norm": 0.41261976957321167,
|
| 8036 |
+
"learning_rate": 3.903633477825886e-05,
|
| 8037 |
+
"loss": 0.33794167041778567,
|
| 8038 |
+
"mean_token_accuracy": 0.9098512396216393,
|
| 8039 |
+
"num_tokens": 1992584.0,
|
| 8040 |
+
"step": 7700
|
| 8041 |
+
},
|
| 8042 |
+
{
|
| 8043 |
+
"entropy": 0.33711165077984334,
|
| 8044 |
+
"epoch": 0.5841019716282505,
|
| 8045 |
+
"grad_norm": 0.44373470544815063,
|
| 8046 |
+
"learning_rate": 3.891667338223167e-05,
|
| 8047 |
+
"loss": 0.381363582611084,
|
| 8048 |
+
"mean_token_accuracy": 0.8914385855197906,
|
| 8049 |
+
"num_tokens": 2109147.0,
|
| 8050 |
+
"step": 7710
|
| 8051 |
+
},
|
| 8052 |
+
{
|
| 8053 |
+
"entropy": 0.2898610128089786,
|
| 8054 |
+
"epoch": 0.5848595617341238,
|
| 8055 |
+
"grad_norm": 0.47294947504997253,
|
| 8056 |
+
"learning_rate": 3.879707870962904e-05,
|
| 8057 |
+
"loss": 0.37704427242279054,
|
| 8058 |
+
"mean_token_accuracy": 0.9094848409295082,
|
| 8059 |
+
"num_tokens": 2206523.0,
|
| 8060 |
+
"step": 7720
|
| 8061 |
+
},
|
| 8062 |
+
{
|
| 8063 |
+
"entropy": 0.3302390962839127,
|
| 8064 |
+
"epoch": 0.585617151839997,
|
| 8065 |
+
"grad_norm": 0.4247268736362457,
|
| 8066 |
+
"learning_rate": 3.867755148043032e-05,
|
| 8067 |
+
"loss": 0.36937775611877444,
|
| 8068 |
+
"mean_token_accuracy": 0.8964511677622795,
|
| 8069 |
+
"num_tokens": 2313042.0,
|
| 8070 |
+
"step": 7730
|
| 8071 |
+
},
|
| 8072 |
+
{
|
| 8073 |
+
"entropy": 0.29674111772328615,
|
| 8074 |
+
"epoch": 0.5863747419458701,
|
| 8075 |
+
"grad_norm": 0.38928377628326416,
|
| 8076 |
+
"learning_rate": 3.855809241420883e-05,
|
| 8077 |
+
"loss": 0.35187516212463377,
|
| 8078 |
+
"mean_token_accuracy": 0.907732842862606,
|
| 8079 |
+
"num_tokens": 2412624.0,
|
| 8080 |
+
"step": 7740
|
| 8081 |
+
},
|
| 8082 |
+
{
|
| 8083 |
+
"entropy": 0.26236414816230536,
|
| 8084 |
+
"epoch": 0.5871323320517434,
|
| 8085 |
+
"grad_norm": 0.44227948784828186,
|
| 8086 |
+
"learning_rate": 3.843870223012753e-05,
|
| 8087 |
+
"loss": 0.3066854000091553,
|
| 8088 |
+
"mean_token_accuracy": 0.918500104546547,
|
| 8089 |
+
"num_tokens": 2502901.0,
|
| 8090 |
+
"step": 7750
|
| 8091 |
+
},
|
| 8092 |
+
{
|
| 8093 |
+
"epoch": 0.5871323320517434,
|
| 8094 |
+
"eval_entropy": 0.29141705292050313,
|
| 8095 |
+
"eval_loss": 0.29264143109321594,
|
| 8096 |
+
"eval_mean_token_accuracy": 0.9052422897199567,
|
| 8097 |
+
"eval_num_tokens": 2502901.0,
|
| 8098 |
+
"eval_runtime": 366.9036,
|
| 8099 |
+
"eval_samples_per_second": 5.816,
|
| 8100 |
+
"eval_steps_per_second": 1.455,
|
| 8101 |
+
"step": 7750
|
| 8102 |
+
},
|
| 8103 |
+
{
|
| 8104 |
+
"entropy": 0.29118088809773324,
|
| 8105 |
+
"epoch": 0.5878899221576166,
|
| 8106 |
+
"grad_norm": 0.40471747517585754,
|
| 8107 |
+
"learning_rate": 3.831938164693469e-05,
|
| 8108 |
+
"loss": 0.3376481056213379,
|
| 8109 |
+
"mean_token_accuracy": 0.9038581416010857,
|
| 8110 |
+
"num_tokens": 2603072.0,
|
| 8111 |
+
"step": 7760
|
| 8112 |
+
},
|
| 8113 |
+
{
|
| 8114 |
+
"entropy": 0.3052817082032561,
|
| 8115 |
+
"epoch": 0.5886475122634899,
|
| 8116 |
+
"grad_norm": 0.5422359704971313,
|
| 8117 |
+
"learning_rate": 3.8200131382959586e-05,
|
| 8118 |
+
"loss": 0.34176883697509763,
|
| 8119 |
+
"mean_token_accuracy": 0.9007858261466026,
|
| 8120 |
+
"num_tokens": 2705340.0,
|
| 8121 |
+
"step": 7770
|
| 8122 |
+
},
|
| 8123 |
+
{
|
| 8124 |
+
"entropy": 0.2885507680475712,
|
| 8125 |
+
"epoch": 0.589405102369363,
|
| 8126 |
+
"grad_norm": 0.5075405240058899,
|
| 8127 |
+
"learning_rate": 3.808095215610816e-05,
|
| 8128 |
+
"loss": 0.3256500720977783,
|
| 8129 |
+
"mean_token_accuracy": 0.9076829209923745,
|
| 8130 |
+
"num_tokens": 2801725.0,
|
| 8131 |
+
"step": 7780
|
| 8132 |
+
},
|
| 8133 |
+
{
|
| 8134 |
+
"entropy": 0.3176051995716989,
|
| 8135 |
+
"epoch": 0.5901626924752362,
|
| 8136 |
+
"grad_norm": 0.36726605892181396,
|
| 8137 |
+
"learning_rate": 3.7961844683858684e-05,
|
| 8138 |
+
"loss": 0.36698002815246583,
|
| 8139 |
+
"mean_token_accuracy": 0.8985909134149551,
|
| 8140 |
+
"num_tokens": 2901989.0,
|
| 8141 |
+
"step": 7790
|
| 8142 |
+
},
|
| 8143 |
+
{
|
| 8144 |
+
"entropy": 0.3000574687495828,
|
| 8145 |
+
"epoch": 0.5909202825811095,
|
| 8146 |
+
"grad_norm": 0.5310767889022827,
|
| 8147 |
+
"learning_rate": 3.784280968325748e-05,
|
| 8148 |
+
"loss": 0.3372817039489746,
|
| 8149 |
+
"mean_token_accuracy": 0.9080954909324646,
|
| 8150 |
+
"num_tokens": 2998881.0,
|
| 8151 |
+
"step": 7800
|
| 8152 |
+
},
|
| 8153 |
+
{
|
| 8154 |
+
"entropy": 0.2850060695782304,
|
| 8155 |
+
"epoch": 0.5916778726869827,
|
| 8156 |
+
"grad_norm": 0.39137741923332214,
|
| 8157 |
+
"learning_rate": 3.7723847870914575e-05,
|
| 8158 |
+
"loss": 0.34608912467956543,
|
| 8159 |
+
"mean_token_accuracy": 0.9057006880640983,
|
| 8160 |
+
"num_tokens": 3099067.0,
|
| 8161 |
+
"step": 7810
|
| 8162 |
+
},
|
| 8163 |
+
{
|
| 8164 |
+
"entropy": 0.28733792770653965,
|
| 8165 |
+
"epoch": 0.592435462792856,
|
| 8166 |
+
"grad_norm": 0.3380966782569885,
|
| 8167 |
+
"learning_rate": 3.760495996299935e-05,
|
| 8168 |
+
"loss": 0.32074360847473143,
|
| 8169 |
+
"mean_token_accuracy": 0.907772421836853,
|
| 8170 |
+
"num_tokens": 3199784.0,
|
| 8171 |
+
"step": 7820
|
| 8172 |
+
},
|
| 8173 |
+
{
|
| 8174 |
+
"entropy": 0.3030220426619053,
|
| 8175 |
+
"epoch": 0.5931930528987291,
|
| 8176 |
+
"grad_norm": 0.5010019540786743,
|
| 8177 |
+
"learning_rate": 3.748614667523632e-05,
|
| 8178 |
+
"loss": 0.3495840072631836,
|
| 8179 |
+
"mean_token_accuracy": 0.9055353298783302,
|
| 8180 |
+
"num_tokens": 3301790.0,
|
| 8181 |
+
"step": 7830
|
| 8182 |
+
},
|
| 8183 |
+
{
|
| 8184 |
+
"entropy": 0.25674657318741084,
|
| 8185 |
+
"epoch": 0.5939506430046023,
|
| 8186 |
+
"grad_norm": 0.5346329212188721,
|
| 8187 |
+
"learning_rate": 3.736740872290076e-05,
|
| 8188 |
+
"loss": 0.3107120513916016,
|
| 8189 |
+
"mean_token_accuracy": 0.9172234773635864,
|
| 8190 |
+
"num_tokens": 3391263.0,
|
| 8191 |
+
"step": 7840
|
| 8192 |
+
},
|
| 8193 |
+
{
|
| 8194 |
+
"entropy": 0.3097769618034363,
|
| 8195 |
+
"epoch": 0.5947082331104756,
|
| 8196 |
+
"grad_norm": 0.5312048196792603,
|
| 8197 |
+
"learning_rate": 3.7248746820814405e-05,
|
| 8198 |
+
"loss": 0.3648669719696045,
|
| 8199 |
+
"mean_token_accuracy": 0.898509131371975,
|
| 8200 |
+
"num_tokens": 3493351.0,
|
| 8201 |
+
"step": 7850
|
| 8202 |
+
},
|
| 8203 |
+
{
|
| 8204 |
+
"entropy": 0.3084961650893092,
|
| 8205 |
+
"epoch": 0.5954658232163488,
|
| 8206 |
+
"grad_norm": 0.38531121611595154,
|
| 8207 |
+
"learning_rate": 3.7130161683341146e-05,
|
| 8208 |
+
"loss": 0.36528897285461426,
|
| 8209 |
+
"mean_token_accuracy": 0.9024834498763085,
|
| 8210 |
+
"num_tokens": 3596548.0,
|
| 8211 |
+
"step": 7860
|
| 8212 |
+
},
|
| 8213 |
+
{
|
| 8214 |
+
"entropy": 0.28346777241677046,
|
| 8215 |
+
"epoch": 0.5962234133222221,
|
| 8216 |
+
"grad_norm": 0.39585480093955994,
|
| 8217 |
+
"learning_rate": 3.701165402438277e-05,
|
| 8218 |
+
"loss": 0.3357419967651367,
|
| 8219 |
+
"mean_token_accuracy": 0.9120162844657898,
|
| 8220 |
+
"num_tokens": 3695360.0,
|
| 8221 |
+
"step": 7870
|
| 8222 |
+
},
|
| 8223 |
+
{
|
| 8224 |
+
"entropy": 0.3283775428310037,
|
| 8225 |
+
"epoch": 0.5969810034280952,
|
| 8226 |
+
"grad_norm": 0.2965605854988098,
|
| 8227 |
+
"learning_rate": 3.6893224557374595e-05,
|
| 8228 |
+
"loss": 0.37510778903961184,
|
| 8229 |
+
"mean_token_accuracy": 0.8940520346164703,
|
| 8230 |
+
"num_tokens": 3795712.0,
|
| 8231 |
+
"step": 7880
|
| 8232 |
+
},
|
| 8233 |
+
{
|
| 8234 |
+
"entropy": 0.303850819543004,
|
| 8235 |
+
"epoch": 0.5977385935339684,
|
| 8236 |
+
"grad_norm": 0.5088040828704834,
|
| 8237 |
+
"learning_rate": 3.677487399528124e-05,
|
| 8238 |
+
"loss": 0.32466897964477537,
|
| 8239 |
+
"mean_token_accuracy": 0.9058476537466049,
|
| 8240 |
+
"num_tokens": 3891781.0,
|
| 8241 |
+
"step": 7890
|
| 8242 |
+
},
|
| 8243 |
+
{
|
| 8244 |
+
"entropy": 0.28571315221488475,
|
| 8245 |
+
"epoch": 0.5984961836398417,
|
| 8246 |
+
"grad_norm": 0.39025136828422546,
|
| 8247 |
+
"learning_rate": 3.6656603050592294e-05,
|
| 8248 |
+
"loss": 0.3613213062286377,
|
| 8249 |
+
"mean_token_accuracy": 0.9081512659788131,
|
| 8250 |
+
"num_tokens": 3989798.0,
|
| 8251 |
+
"step": 7900
|
| 8252 |
+
},
|
| 8253 |
+
{
|
| 8254 |
+
"entropy": 0.2969894910231233,
|
| 8255 |
+
"epoch": 0.5992537737457149,
|
| 8256 |
+
"grad_norm": 0.5206224322319031,
|
| 8257 |
+
"learning_rate": 3.6538412435318036e-05,
|
| 8258 |
+
"loss": 0.35520272254943847,
|
| 8259 |
+
"mean_token_accuracy": 0.903459957242012,
|
| 8260 |
+
"num_tokens": 4092198.0,
|
| 8261 |
+
"step": 7910
|
| 8262 |
+
},
|
| 8263 |
+
{
|
| 8264 |
+
"entropy": 0.261045260168612,
|
| 8265 |
+
"epoch": 0.6000113638515882,
|
| 8266 |
+
"grad_norm": 0.4172385334968567,
|
| 8267 |
+
"learning_rate": 3.642030286098513e-05,
|
| 8268 |
+
"loss": 0.3094768047332764,
|
| 8269 |
+
"mean_token_accuracy": 0.9193493768572807,
|
| 8270 |
+
"num_tokens": 4192896.0,
|
| 8271 |
+
"step": 7920
|
| 8272 |
+
},
|
| 8273 |
+
{
|
| 8274 |
+
"entropy": 0.3263862133026123,
|
| 8275 |
+
"epoch": 0.6007689539574613,
|
| 8276 |
+
"grad_norm": 0.4942246079444885,
|
| 8277 |
+
"learning_rate": 3.63022750386324e-05,
|
| 8278 |
+
"loss": 0.3887051582336426,
|
| 8279 |
+
"mean_token_accuracy": 0.8954097390174866,
|
| 8280 |
+
"num_tokens": 4286686.0,
|
| 8281 |
+
"step": 7930
|
| 8282 |
+
},
|
| 8283 |
+
{
|
| 8284 |
+
"entropy": 0.2514767138287425,
|
| 8285 |
+
"epoch": 0.6015265440633345,
|
| 8286 |
+
"grad_norm": 0.48457324504852295,
|
| 8287 |
+
"learning_rate": 3.618432967880647e-05,
|
| 8288 |
+
"loss": 0.27435855865478515,
|
| 8289 |
+
"mean_token_accuracy": 0.9226398333907128,
|
| 8290 |
+
"num_tokens": 4370302.0,
|
| 8291 |
+
"step": 7940
|
| 8292 |
+
},
|
| 8293 |
+
{
|
| 8294 |
+
"entropy": 0.3096071735024452,
|
| 8295 |
+
"epoch": 0.6022841341692078,
|
| 8296 |
+
"grad_norm": 0.4468488097190857,
|
| 8297 |
+
"learning_rate": 3.6066467491557554e-05,
|
| 8298 |
+
"loss": 0.36656315326690675,
|
| 8299 |
+
"mean_token_accuracy": 0.9038876727223396,
|
| 8300 |
+
"num_tokens": 4471708.0,
|
| 8301 |
+
"step": 7950
|
| 8302 |
+
},
|
| 8303 |
+
{
|
| 8304 |
+
"entropy": 0.25050507206469774,
|
| 8305 |
+
"epoch": 0.603041724275081,
|
| 8306 |
+
"grad_norm": 0.4183157682418823,
|
| 8307 |
+
"learning_rate": 3.5948689186435143e-05,
|
| 8308 |
+
"loss": 0.2869309425354004,
|
| 8309 |
+
"mean_token_accuracy": 0.9210500806570053,
|
| 8310 |
+
"num_tokens": 4563613.0,
|
| 8311 |
+
"step": 7960
|
| 8312 |
+
},
|
| 8313 |
+
{
|
| 8314 |
+
"entropy": 0.28720583356916907,
|
| 8315 |
+
"epoch": 0.6037993143809541,
|
| 8316 |
+
"grad_norm": 0.45605581998825073,
|
| 8317 |
+
"learning_rate": 3.583099547248373e-05,
|
| 8318 |
+
"loss": 0.32700610160827637,
|
| 8319 |
+
"mean_token_accuracy": 0.9074159458279609,
|
| 8320 |
+
"num_tokens": 4658290.0,
|
| 8321 |
+
"step": 7970
|
| 8322 |
+
},
|
| 8323 |
+
{
|
| 8324 |
+
"entropy": 0.28136809784919026,
|
| 8325 |
+
"epoch": 0.6045569044868274,
|
| 8326 |
+
"grad_norm": 0.5523653030395508,
|
| 8327 |
+
"learning_rate": 3.5713387058238584e-05,
|
| 8328 |
+
"loss": 0.3111879825592041,
|
| 8329 |
+
"mean_token_accuracy": 0.909286692738533,
|
| 8330 |
+
"num_tokens": 4754393.0,
|
| 8331 |
+
"step": 7980
|
| 8332 |
+
},
|
| 8333 |
+
{
|
| 8334 |
+
"entropy": 0.3156980287283659,
|
| 8335 |
+
"epoch": 0.6053144945927006,
|
| 8336 |
+
"grad_norm": 0.4386664628982544,
|
| 8337 |
+
"learning_rate": 3.559586465172143e-05,
|
| 8338 |
+
"loss": 0.3415090322494507,
|
| 8339 |
+
"mean_token_accuracy": 0.9031947538256645,
|
| 8340 |
+
"num_tokens": 4866668.0,
|
| 8341 |
+
"step": 7990
|
| 8342 |
+
},
|
| 8343 |
+
{
|
| 8344 |
+
"entropy": 0.33502144310623405,
|
| 8345 |
+
"epoch": 0.6060720846985739,
|
| 8346 |
+
"grad_norm": 0.4574124217033386,
|
| 8347 |
+
"learning_rate": 3.547842896043624e-05,
|
| 8348 |
+
"loss": 0.3874842405319214,
|
| 8349 |
+
"mean_token_accuracy": 0.8882740035653114,
|
| 8350 |
+
"num_tokens": 4973743.0,
|
| 8351 |
+
"step": 8000
|
| 8352 |
+
},
|
| 8353 |
+
{
|
| 8354 |
+
"epoch": 0.6060720846985739,
|
| 8355 |
+
"eval_entropy": 0.2932357283567221,
|
| 8356 |
+
"eval_loss": 0.2922365963459015,
|
| 8357 |
+
"eval_mean_token_accuracy": 0.9054181920678428,
|
| 8358 |
+
"eval_num_tokens": 4973743.0,
|
| 8359 |
+
"eval_runtime": 365.9511,
|
| 8360 |
+
"eval_samples_per_second": 5.831,
|
| 8361 |
+
"eval_steps_per_second": 1.459,
|
| 8362 |
+
"step": 8000
|
| 8363 |
}
|
| 8364 |
],
|
| 8365 |
"logging_steps": 10,
|
|
|
|
| 8379 |
"attributes": {}
|
| 8380 |
}
|
| 8381 |
},
|
| 8382 |
+
"total_flos": 6.948618572943852e+17,
|
| 8383 |
"train_batch_size": 2,
|
| 8384 |
"trial_name": null,
|
| 8385 |
"trial_params": null
|