explorer_comparison / trainer_state.json
Evanwu50020's picture
Upload folder using huggingface_hub
5c148fb verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9978568366909558,
"eval_steps": 500,
"global_step": 291,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"clip_ratio": 0.0,
"completion_length": 90.30859375,
"epoch": 0.0034290612944706386,
"grad_norm": 4.169865651189505,
"kl": 0.0,
"learning_rate": 9.965635738831615e-07,
"loss": 0.0,
"reward": 0.984375,
"reward_std": 0.4253819230943918,
"rewards/accuracy_reward_action": 0.15625,
"rewards/accuracy_reward_coord": 0.01953125,
"rewards/format_reward": 0.80859375,
"step": 1
},
{
"clip_ratio": 0.0,
"completion_length": 90.2734375,
"epoch": 0.006858122588941277,
"grad_norm": 15.354884464504476,
"kl": 0.0011739730834960938,
"learning_rate": 9.93127147766323e-07,
"loss": 0.0,
"reward": 1.1484375,
"reward_std": 0.47851087525486946,
"rewards/accuracy_reward_action": 0.20703125,
"rewards/accuracy_reward_coord": 0.03125,
"rewards/format_reward": 0.91015625,
"step": 2
},
{
"clip_ratio": 0.0,
"completion_length": 103.06640625,
"epoch": 0.010287183883411917,
"grad_norm": 3.579290314608233,
"kl": 0.0019245147705078125,
"learning_rate": 9.896907216494845e-07,
"loss": 0.0001,
"reward": 1.27734375,
"reward_std": 0.5746662765741348,
"rewards/accuracy_reward_action": 0.33203125,
"rewards/accuracy_reward_coord": 0.09375,
"rewards/format_reward": 0.8515625,
"step": 3
},
{
"clip_ratio": 0.0,
"completion_length": 98.6875,
"epoch": 0.013716245177882554,
"grad_norm": 7.574944877248717,
"kl": 0.006870269775390625,
"learning_rate": 9.86254295532646e-07,
"loss": 0.0003,
"reward": 1.41796875,
"reward_std": 0.5559290274977684,
"rewards/accuracy_reward_action": 0.41015625,
"rewards/accuracy_reward_coord": 0.109375,
"rewards/format_reward": 0.8984375,
"step": 4
},
{
"clip_ratio": 0.0,
"completion_length": 98.12109375,
"epoch": 0.017145306472353194,
"grad_norm": 6.512027175960988,
"kl": 0.008758544921875,
"learning_rate": 9.828178694158075e-07,
"loss": 0.0004,
"reward": 1.4296875,
"reward_std": 0.5759975053369999,
"rewards/accuracy_reward_action": 0.49609375,
"rewards/accuracy_reward_coord": 0.03125,
"rewards/format_reward": 0.90234375,
"step": 5
},
{
"clip_ratio": 0.0,
"completion_length": 116.57421875,
"epoch": 0.020574367766823833,
"grad_norm": 8.442076588732956,
"kl": 0.014019012451171875,
"learning_rate": 9.79381443298969e-07,
"loss": 0.0006,
"reward": 1.6796875,
"reward_std": 0.6102640256285667,
"rewards/accuracy_reward_action": 0.64453125,
"rewards/accuracy_reward_coord": 0.11328125,
"rewards/format_reward": 0.921875,
"step": 6
},
{
"clip_ratio": 0.0,
"completion_length": 107.53515625,
"epoch": 0.02400342906129447,
"grad_norm": 4.77040444096485,
"kl": 0.017120361328125,
"learning_rate": 9.759450171821305e-07,
"loss": 0.0007,
"reward": 1.74609375,
"reward_std": 0.4844810925424099,
"rewards/accuracy_reward_action": 0.7265625,
"rewards/accuracy_reward_coord": 0.0546875,
"rewards/format_reward": 0.96484375,
"step": 7
},
{
"clip_ratio": 0.0,
"completion_length": 109.5390625,
"epoch": 0.02743249035576511,
"grad_norm": 3.7032651890067525,
"kl": 0.01229095458984375,
"learning_rate": 9.72508591065292e-07,
"loss": 0.0005,
"reward": 1.69921875,
"reward_std": 0.47686367109417915,
"rewards/accuracy_reward_action": 0.69140625,
"rewards/accuracy_reward_coord": 0.09375,
"rewards/format_reward": 0.9140625,
"step": 8
},
{
"clip_ratio": 0.0,
"completion_length": 111.12890625,
"epoch": 0.030861551650235748,
"grad_norm": 5.007916582021746,
"kl": 0.01473236083984375,
"learning_rate": 9.690721649484535e-07,
"loss": 0.0006,
"reward": 1.6875,
"reward_std": 0.5727768018841743,
"rewards/accuracy_reward_action": 0.65625,
"rewards/accuracy_reward_coord": 0.08203125,
"rewards/format_reward": 0.94921875,
"step": 9
},
{
"clip_ratio": 0.0,
"completion_length": 117.96875,
"epoch": 0.03429061294470639,
"grad_norm": 4.167812341811835,
"kl": 0.0190887451171875,
"learning_rate": 9.65635738831615e-07,
"loss": 0.0008,
"reward": 1.7109375,
"reward_std": 0.45130953192710876,
"rewards/accuracy_reward_action": 0.69921875,
"rewards/accuracy_reward_coord": 0.06640625,
"rewards/format_reward": 0.9453125,
"step": 10
},
{
"clip_ratio": 0.0,
"completion_length": 117.6875,
"epoch": 0.03771967423917703,
"grad_norm": 2.6525737109173715,
"kl": 0.01401519775390625,
"learning_rate": 9.621993127147767e-07,
"loss": 0.0006,
"reward": 1.6953125,
"reward_std": 0.4206106849014759,
"rewards/accuracy_reward_action": 0.671875,
"rewards/accuracy_reward_coord": 0.08984375,
"rewards/format_reward": 0.93359375,
"step": 11
},
{
"clip_ratio": 0.0,
"completion_length": 113.2734375,
"epoch": 0.041148735533647667,
"grad_norm": 6.690879246272424,
"kl": 0.01448822021484375,
"learning_rate": 9.587628865979382e-07,
"loss": 0.0006,
"reward": 1.78515625,
"reward_std": 0.36769258975982666,
"rewards/accuracy_reward_action": 0.71484375,
"rewards/accuracy_reward_coord": 0.09375,
"rewards/format_reward": 0.9765625,
"step": 12
},
{
"clip_ratio": 0.0,
"completion_length": 117.9375,
"epoch": 0.044577796828118306,
"grad_norm": 1.8641990568589721,
"kl": 0.01544952392578125,
"learning_rate": 9.553264604810997e-07,
"loss": 0.0006,
"reward": 1.9140625,
"reward_std": 0.29037280660122633,
"rewards/accuracy_reward_action": 0.8046875,
"rewards/accuracy_reward_coord": 0.12109375,
"rewards/format_reward": 0.98828125,
"step": 13
},
{
"clip_ratio": 0.0,
"completion_length": 111.015625,
"epoch": 0.04800685812258894,
"grad_norm": 4.660763921206706,
"kl": 0.01983642578125,
"learning_rate": 9.518900343642611e-07,
"loss": 0.0008,
"reward": 1.8515625,
"reward_std": 0.37489059288054705,
"rewards/accuracy_reward_action": 0.76171875,
"rewards/accuracy_reward_coord": 0.1015625,
"rewards/format_reward": 0.98828125,
"step": 14
},
{
"clip_ratio": 0.0,
"completion_length": 111.33203125,
"epoch": 0.05143591941705958,
"grad_norm": 4.101279763347818,
"kl": 0.01727294921875,
"learning_rate": 9.484536082474226e-07,
"loss": 0.0007,
"reward": 1.86328125,
"reward_std": 0.3477381020784378,
"rewards/accuracy_reward_action": 0.76171875,
"rewards/accuracy_reward_coord": 0.11328125,
"rewards/format_reward": 0.98828125,
"step": 15
},
{
"clip_ratio": 0.0,
"completion_length": 108.69140625,
"epoch": 0.05486498071153022,
"grad_norm": 4.209451477714062,
"kl": 0.02355194091796875,
"learning_rate": 9.450171821305841e-07,
"loss": 0.0009,
"reward": 2.06640625,
"reward_std": 0.3058856502175331,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.19921875,
"rewards/format_reward": 0.98046875,
"step": 16
},
{
"clip_ratio": 0.0,
"completion_length": 106.98828125,
"epoch": 0.05829404200600086,
"grad_norm": 3.315964426251405,
"kl": 0.023223876953125,
"learning_rate": 9.415807560137456e-07,
"loss": 0.0009,
"reward": 1.8828125,
"reward_std": 0.22431878745555878,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.0703125,
"rewards/format_reward": 0.984375,
"step": 17
},
{
"clip_ratio": 0.0,
"completion_length": 107.31640625,
"epoch": 0.061723103300471496,
"grad_norm": 5.469921639585009,
"kl": 0.02007293701171875,
"learning_rate": 9.381443298969071e-07,
"loss": 0.0008,
"reward": 1.9296875,
"reward_std": 0.26369384955614805,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.0859375,
"rewards/format_reward": 0.9921875,
"step": 18
},
{
"clip_ratio": 0.0,
"completion_length": 114.91796875,
"epoch": 0.06515216459494214,
"grad_norm": 5.66633473155319,
"kl": 0.02239227294921875,
"learning_rate": 9.347079037800687e-07,
"loss": 0.0009,
"reward": 1.86328125,
"reward_std": 0.3953590150922537,
"rewards/accuracy_reward_action": 0.76171875,
"rewards/accuracy_reward_coord": 0.1171875,
"rewards/format_reward": 0.984375,
"step": 19
},
{
"clip_ratio": 0.0,
"completion_length": 104.84375,
"epoch": 0.06858122588941278,
"grad_norm": 3.371967005233721,
"kl": 0.02564239501953125,
"learning_rate": 9.312714776632303e-07,
"loss": 0.001,
"reward": 1.90625,
"reward_std": 0.1727309264242649,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.03515625,
"rewards/format_reward": 0.99609375,
"step": 20
},
{
"clip_ratio": 0.0,
"completion_length": 102.23046875,
"epoch": 0.07201028718388341,
"grad_norm": 9.311589365168965,
"kl": 0.021026611328125,
"learning_rate": 9.278350515463918e-07,
"loss": 0.0008,
"reward": 1.94140625,
"reward_std": 0.301584979519248,
"rewards/accuracy_reward_action": 0.8046875,
"rewards/accuracy_reward_coord": 0.140625,
"rewards/format_reward": 0.99609375,
"step": 21
},
{
"clip_ratio": 0.0,
"completion_length": 103.15625,
"epoch": 0.07543934847835405,
"grad_norm": 2.4741963314612794,
"kl": 0.0182037353515625,
"learning_rate": 9.243986254295533e-07,
"loss": 0.0007,
"reward": 2.06640625,
"reward_std": 0.20661969017237425,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.1640625,
"rewards/format_reward": 0.99609375,
"step": 22
},
{
"clip_ratio": 0.0,
"completion_length": 105.8046875,
"epoch": 0.0788684097728247,
"grad_norm": 1.8074831491618741,
"kl": 0.02187347412109375,
"learning_rate": 9.209621993127147e-07,
"loss": 0.0009,
"reward": 1.9921875,
"reward_std": 0.22818244062364101,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.0859375,
"rewards/format_reward": 1.0,
"step": 23
},
{
"clip_ratio": 0.0,
"completion_length": 106.1640625,
"epoch": 0.08229747106729533,
"grad_norm": 2.116378010233233,
"kl": 0.0333404541015625,
"learning_rate": 9.175257731958762e-07,
"loss": 0.0013,
"reward": 1.9140625,
"reward_std": 0.27086862176656723,
"rewards/accuracy_reward_action": 0.78515625,
"rewards/accuracy_reward_coord": 0.13671875,
"rewards/format_reward": 0.9921875,
"step": 24
},
{
"clip_ratio": 0.0,
"completion_length": 108.45703125,
"epoch": 0.08572653236176597,
"grad_norm": 2.017007786819895,
"kl": 0.02716064453125,
"learning_rate": 9.140893470790378e-07,
"loss": 0.0011,
"reward": 1.875,
"reward_std": 0.38331808149814606,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.0859375,
"rewards/format_reward": 0.98046875,
"step": 25
},
{
"clip_ratio": 0.0,
"completion_length": 103.2734375,
"epoch": 0.08915559365623661,
"grad_norm": 3.1854233612302796,
"kl": 0.025115966796875,
"learning_rate": 9.106529209621993e-07,
"loss": 0.001,
"reward": 1.9140625,
"reward_std": 0.23374123964458704,
"rewards/accuracy_reward_action": 0.83984375,
"rewards/accuracy_reward_coord": 0.078125,
"rewards/format_reward": 0.99609375,
"step": 26
},
{
"clip_ratio": 0.0,
"completion_length": 104.72265625,
"epoch": 0.09258465495070724,
"grad_norm": 3.25367470492351,
"kl": 0.0232696533203125,
"learning_rate": 9.072164948453608e-07,
"loss": 0.0009,
"reward": 1.98828125,
"reward_std": 0.15656260214745998,
"rewards/accuracy_reward_action": 0.9296875,
"rewards/accuracy_reward_coord": 0.06640625,
"rewards/format_reward": 0.9921875,
"step": 27
},
{
"clip_ratio": 0.0,
"completion_length": 103.7890625,
"epoch": 0.09601371624517788,
"grad_norm": 3.8415142050104167,
"kl": 0.0357666015625,
"learning_rate": 9.037800687285223e-07,
"loss": 0.0014,
"reward": 1.8984375,
"reward_std": 0.2581741688773036,
"rewards/accuracy_reward_action": 0.83984375,
"rewards/accuracy_reward_coord": 0.06640625,
"rewards/format_reward": 0.9921875,
"step": 28
},
{
"clip_ratio": 0.0,
"completion_length": 98.82421875,
"epoch": 0.09944277753964852,
"grad_norm": 2.516551663652597,
"kl": 0.0266876220703125,
"learning_rate": 9.003436426116838e-07,
"loss": 0.0011,
"reward": 1.98828125,
"reward_std": 0.20948081836104393,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.078125,
"rewards/format_reward": 0.99609375,
"step": 29
},
{
"clip_ratio": 0.0,
"completion_length": 99.3828125,
"epoch": 0.10287183883411916,
"grad_norm": 2.095156678702029,
"kl": 0.03021240234375,
"learning_rate": 8.969072164948454e-07,
"loss": 0.0012,
"reward": 2.0234375,
"reward_std": 0.16610248386859894,
"rewards/accuracy_reward_action": 0.91015625,
"rewards/accuracy_reward_coord": 0.11328125,
"rewards/format_reward": 1.0,
"step": 30
},
{
"clip_ratio": 0.0,
"completion_length": 100.80859375,
"epoch": 0.1063009001285898,
"grad_norm": 5.27355668616815,
"kl": 0.0283203125,
"learning_rate": 8.934707903780069e-07,
"loss": 0.0011,
"reward": 1.96484375,
"reward_std": 0.2747951643541455,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.16015625,
"rewards/format_reward": 0.9921875,
"step": 31
},
{
"clip_ratio": 0.0,
"completion_length": 103.84375,
"epoch": 0.10972996142306043,
"grad_norm": 6.545819641462799,
"kl": 0.0323638916015625,
"learning_rate": 8.900343642611683e-07,
"loss": 0.0013,
"reward": 1.91796875,
"reward_std": 0.2592557203024626,
"rewards/accuracy_reward_action": 0.79296875,
"rewards/accuracy_reward_coord": 0.125,
"rewards/format_reward": 1.0,
"step": 32
},
{
"clip_ratio": 0.0,
"completion_length": 99.0859375,
"epoch": 0.11315902271753107,
"grad_norm": 2.345212742782235,
"kl": 0.034271240234375,
"learning_rate": 8.865979381443298e-07,
"loss": 0.0014,
"reward": 2.05078125,
"reward_std": 0.2973444387316704,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.140625,
"rewards/format_reward": 0.99609375,
"step": 33
},
{
"clip_ratio": 0.0,
"completion_length": 103.234375,
"epoch": 0.11658808401200171,
"grad_norm": 9.480377541811404,
"kl": 0.0315399169921875,
"learning_rate": 8.831615120274913e-07,
"loss": 0.0013,
"reward": 1.91015625,
"reward_std": 0.2874891981482506,
"rewards/accuracy_reward_action": 0.8046875,
"rewards/accuracy_reward_coord": 0.109375,
"rewards/format_reward": 0.99609375,
"step": 34
},
{
"clip_ratio": 0.0,
"completion_length": 104.14453125,
"epoch": 0.12001714530647235,
"grad_norm": 4.093247592407727,
"kl": 0.032470703125,
"learning_rate": 8.797250859106528e-07,
"loss": 0.0013,
"reward": 1.96875,
"reward_std": 0.2139760795980692,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.07421875,
"rewards/format_reward": 1.0,
"step": 35
},
{
"clip_ratio": 0.0,
"completion_length": 102.47265625,
"epoch": 0.12344620660094299,
"grad_norm": 4.414899743605823,
"kl": 0.0321807861328125,
"learning_rate": 8.762886597938144e-07,
"loss": 0.0013,
"reward": 1.99609375,
"reward_std": 0.10718034580349922,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.07421875,
"rewards/format_reward": 1.0,
"step": 36
},
{
"clip_ratio": 0.0,
"completion_length": 99.171875,
"epoch": 0.12687526789541362,
"grad_norm": 1.9240548034063036,
"kl": 0.0298919677734375,
"learning_rate": 8.728522336769759e-07,
"loss": 0.0012,
"reward": 1.90625,
"reward_std": 0.19055858626961708,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.078125,
"rewards/format_reward": 1.0,
"step": 37
},
{
"clip_ratio": 0.0,
"completion_length": 101.30859375,
"epoch": 0.13030432918988427,
"grad_norm": 1.7182163505314758,
"kl": 0.0316925048828125,
"learning_rate": 8.694158075601374e-07,
"loss": 0.0013,
"reward": 1.9921875,
"reward_std": 0.19306486751884222,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.09375,
"rewards/format_reward": 0.99609375,
"step": 38
},
{
"clip_ratio": 0.0,
"completion_length": 102.19921875,
"epoch": 0.1337333904843549,
"grad_norm": 5.580302117235681,
"kl": 0.0377655029296875,
"learning_rate": 8.659793814432989e-07,
"loss": 0.0015,
"reward": 1.96484375,
"reward_std": 0.1960927415639162,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.1015625,
"rewards/format_reward": 0.99609375,
"step": 39
},
{
"clip_ratio": 0.0,
"completion_length": 99.640625,
"epoch": 0.13716245177882555,
"grad_norm": 5.347881422440897,
"kl": 0.03704833984375,
"learning_rate": 8.625429553264604e-07,
"loss": 0.0015,
"reward": 1.953125,
"reward_std": 0.20940626692026854,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.12890625,
"rewards/format_reward": 1.0,
"step": 40
},
{
"clip_ratio": 0.0,
"completion_length": 105.08203125,
"epoch": 0.14059151307329618,
"grad_norm": 1.7711400045751344,
"kl": 0.03179931640625,
"learning_rate": 8.591065292096219e-07,
"loss": 0.0013,
"reward": 1.93359375,
"reward_std": 0.16333417687565088,
"rewards/accuracy_reward_action": 0.85546875,
"rewards/accuracy_reward_coord": 0.078125,
"rewards/format_reward": 1.0,
"step": 41
},
{
"clip_ratio": 0.0,
"completion_length": 98.72265625,
"epoch": 0.14402057436776683,
"grad_norm": 7.596003099794815,
"kl": 0.038818359375,
"learning_rate": 8.556701030927834e-07,
"loss": 0.0016,
"reward": 2.015625,
"reward_std": 0.2686883732676506,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.1328125,
"rewards/format_reward": 1.0,
"step": 42
},
{
"clip_ratio": 0.0,
"completion_length": 101.046875,
"epoch": 0.14744963566223745,
"grad_norm": 2.1589949902852568,
"kl": 0.0390167236328125,
"learning_rate": 8.52233676975945e-07,
"loss": 0.0016,
"reward": 1.8984375,
"reward_std": 0.19987381342798471,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.08984375,
"rewards/format_reward": 0.99609375,
"step": 43
},
{
"clip_ratio": 0.0,
"completion_length": 102.0859375,
"epoch": 0.1508786969567081,
"grad_norm": 2.985222575414514,
"kl": 0.03912353515625,
"learning_rate": 8.487972508591065e-07,
"loss": 0.0016,
"reward": 1.95703125,
"reward_std": 0.26305136550217867,
"rewards/accuracy_reward_action": 0.80078125,
"rewards/accuracy_reward_coord": 0.16015625,
"rewards/format_reward": 0.99609375,
"step": 44
},
{
"clip_ratio": 0.0,
"completion_length": 101.421875,
"epoch": 0.15430775825117873,
"grad_norm": 1.3069548395660002,
"kl": 0.043212890625,
"learning_rate": 8.45360824742268e-07,
"loss": 0.0017,
"reward": 1.94921875,
"reward_std": 0.20725996419787407,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.1328125,
"rewards/format_reward": 0.9921875,
"step": 45
},
{
"clip_ratio": 0.0,
"completion_length": 99.7109375,
"epoch": 0.1577368195456494,
"grad_norm": 2.351651859555171,
"kl": 0.0411376953125,
"learning_rate": 8.419243986254296e-07,
"loss": 0.0016,
"reward": 2.0234375,
"reward_std": 0.23960923682898283,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.1328125,
"rewards/format_reward": 1.0,
"step": 46
},
{
"clip_ratio": 0.0,
"completion_length": 102.859375,
"epoch": 0.16116588084012,
"grad_norm": 3.6702022844794633,
"kl": 0.03863525390625,
"learning_rate": 8.384879725085911e-07,
"loss": 0.0015,
"reward": 1.98828125,
"reward_std": 0.3140581510961056,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.1796875,
"rewards/format_reward": 1.0,
"step": 47
},
{
"clip_ratio": 0.0,
"completion_length": 100.94921875,
"epoch": 0.16459494213459067,
"grad_norm": 2.2650752826812113,
"kl": 0.036224365234375,
"learning_rate": 8.350515463917526e-07,
"loss": 0.0015,
"reward": 2.0,
"reward_std": 0.21676901914179325,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.1328125,
"rewards/format_reward": 1.0,
"step": 48
},
{
"clip_ratio": 0.0,
"completion_length": 99.484375,
"epoch": 0.1680240034290613,
"grad_norm": 1.6014441912105555,
"kl": 0.0355682373046875,
"learning_rate": 8.316151202749141e-07,
"loss": 0.0014,
"reward": 1.96484375,
"reward_std": 0.2216318678110838,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.14453125,
"rewards/format_reward": 1.0,
"step": 49
},
{
"clip_ratio": 0.0,
"completion_length": 98.46484375,
"epoch": 0.17145306472353194,
"grad_norm": 6.83244910904526,
"kl": 0.04339599609375,
"learning_rate": 8.281786941580756e-07,
"loss": 0.0017,
"reward": 1.90625,
"reward_std": 0.22567125875502825,
"rewards/accuracy_reward_action": 0.78515625,
"rewards/accuracy_reward_coord": 0.12109375,
"rewards/format_reward": 1.0,
"step": 50
},
{
"clip_ratio": 0.0,
"completion_length": 98.20703125,
"epoch": 0.17488212601800257,
"grad_norm": 1.3227718672397777,
"kl": 0.0413818359375,
"learning_rate": 8.24742268041237e-07,
"loss": 0.0017,
"reward": 2.16796875,
"reward_std": 0.15281605254858732,
"rewards/accuracy_reward_action": 0.984375,
"rewards/accuracy_reward_coord": 0.18359375,
"rewards/format_reward": 1.0,
"step": 51
},
{
"clip_ratio": 0.0,
"completion_length": 99.94921875,
"epoch": 0.17831118731247322,
"grad_norm": 2.2204178372642938,
"kl": 0.0389251708984375,
"learning_rate": 8.213058419243986e-07,
"loss": 0.0016,
"reward": 2.02734375,
"reward_std": 0.18083926383405924,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.17578125,
"rewards/format_reward": 1.0,
"step": 52
},
{
"clip_ratio": 0.0,
"completion_length": 105.4453125,
"epoch": 0.18174024860694385,
"grad_norm": 3.119541528344858,
"kl": 0.0362396240234375,
"learning_rate": 8.178694158075601e-07,
"loss": 0.0014,
"reward": 1.9921875,
"reward_std": 0.2792065069079399,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.1796875,
"rewards/format_reward": 1.0,
"step": 53
},
{
"clip_ratio": 0.0,
"completion_length": 107.23828125,
"epoch": 0.18516930990141447,
"grad_norm": 6.184155571429652,
"kl": 0.0556182861328125,
"learning_rate": 8.144329896907216e-07,
"loss": 0.0022,
"reward": 2.06640625,
"reward_std": 0.25091784074902534,
"rewards/accuracy_reward_action": 0.93359375,
"rewards/accuracy_reward_coord": 0.13671875,
"rewards/format_reward": 0.99609375,
"step": 54
},
{
"clip_ratio": 0.0,
"completion_length": 103.27734375,
"epoch": 0.18859837119588513,
"grad_norm": 2.2553293815115985,
"kl": 0.04754638671875,
"learning_rate": 8.109965635738831e-07,
"loss": 0.0019,
"reward": 2.109375,
"reward_std": 0.22331714257597923,
"rewards/accuracy_reward_action": 0.9296875,
"rewards/accuracy_reward_coord": 0.18359375,
"rewards/format_reward": 0.99609375,
"step": 55
},
{
"clip_ratio": 0.0,
"completion_length": 103.9296875,
"epoch": 0.19202743249035575,
"grad_norm": 5.306929144656568,
"kl": 0.0445556640625,
"learning_rate": 8.075601374570446e-07,
"loss": 0.0018,
"reward": 2.0546875,
"reward_std": 0.33029378205537796,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.2265625,
"rewards/format_reward": 0.9921875,
"step": 56
},
{
"clip_ratio": 0.0,
"completion_length": 104.375,
"epoch": 0.1954564937848264,
"grad_norm": 3.3360995991006313,
"kl": 0.037750244140625,
"learning_rate": 8.041237113402062e-07,
"loss": 0.0015,
"reward": 1.9921875,
"reward_std": 0.18702251464128494,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.15625,
"rewards/format_reward": 1.0,
"step": 57
},
{
"clip_ratio": 0.0,
"completion_length": 103.2578125,
"epoch": 0.19888555507929703,
"grad_norm": 2.5746550275702855,
"kl": 0.046600341796875,
"learning_rate": 8.006872852233677e-07,
"loss": 0.0019,
"reward": 2.06640625,
"reward_std": 0.27793182618916035,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.23828125,
"rewards/format_reward": 1.0,
"step": 58
},
{
"clip_ratio": 0.0,
"completion_length": 106.4765625,
"epoch": 0.2023146163737677,
"grad_norm": 7.922133753917514,
"kl": 0.046875,
"learning_rate": 7.972508591065292e-07,
"loss": 0.0019,
"reward": 2.14453125,
"reward_std": 0.3949808795005083,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 1.0,
"step": 59
},
{
"clip_ratio": 0.0,
"completion_length": 107.3046875,
"epoch": 0.2057436776682383,
"grad_norm": 4.668578467258757,
"kl": 0.051544189453125,
"learning_rate": 7.938144329896906e-07,
"loss": 0.0021,
"reward": 2.04296875,
"reward_std": 0.24435585644096136,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.1484375,
"rewards/format_reward": 1.0,
"step": 60
},
{
"clip_ratio": 0.0,
"completion_length": 103.453125,
"epoch": 0.20917273896270897,
"grad_norm": 2.1768199938478032,
"kl": 0.046600341796875,
"learning_rate": 7.903780068728521e-07,
"loss": 0.0019,
"reward": 2.02734375,
"reward_std": 0.23643438424915075,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.15625,
"rewards/format_reward": 0.99609375,
"step": 61
},
{
"clip_ratio": 0.0,
"completion_length": 103.3515625,
"epoch": 0.2126018002571796,
"grad_norm": 2.580776755790538,
"kl": 0.06036376953125,
"learning_rate": 7.869415807560136e-07,
"loss": 0.0024,
"reward": 2.2109375,
"reward_std": 0.2773527940735221,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.3046875,
"rewards/format_reward": 1.0,
"step": 62
},
{
"clip_ratio": 0.0,
"completion_length": 104.28515625,
"epoch": 0.21603086155165024,
"grad_norm": 3.6655554460146433,
"kl": 0.0533447265625,
"learning_rate": 7.835051546391752e-07,
"loss": 0.0021,
"reward": 2.046875,
"reward_std": 0.21200658101588488,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.1640625,
"rewards/format_reward": 0.99609375,
"step": 63
},
{
"clip_ratio": 0.0,
"completion_length": 104.75390625,
"epoch": 0.21945992284612087,
"grad_norm": 2.313170128101062,
"kl": 0.060028076171875,
"learning_rate": 7.800687285223368e-07,
"loss": 0.0024,
"reward": 2.1953125,
"reward_std": 0.14374582655727863,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.2578125,
"rewards/format_reward": 1.0,
"step": 64
},
{
"clip_ratio": 0.0,
"completion_length": 101.44140625,
"epoch": 0.22288898414059152,
"grad_norm": 2.3617237358151515,
"kl": 0.086151123046875,
"learning_rate": 7.766323024054983e-07,
"loss": 0.0034,
"reward": 2.16796875,
"reward_std": 0.23817015439271927,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.2890625,
"rewards/format_reward": 1.0,
"step": 65
},
{
"clip_ratio": 0.0,
"completion_length": 106.0078125,
"epoch": 0.22631804543506215,
"grad_norm": 3.3699940809288886,
"kl": 0.05230712890625,
"learning_rate": 7.731958762886598e-07,
"loss": 0.0021,
"reward": 2.1953125,
"reward_std": 0.2761722579598427,
"rewards/accuracy_reward_action": 0.91796875,
"rewards/accuracy_reward_coord": 0.28125,
"rewards/format_reward": 0.99609375,
"step": 66
},
{
"clip_ratio": 0.0,
"completion_length": 105.72265625,
"epoch": 0.2297471067295328,
"grad_norm": 45.77607256234026,
"kl": 0.054962158203125,
"learning_rate": 7.697594501718213e-07,
"loss": 0.0022,
"reward": 2.125,
"reward_std": 0.24921870045363903,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.234375,
"rewards/format_reward": 0.99609375,
"step": 67
},
{
"clip_ratio": 0.0,
"completion_length": 106.109375,
"epoch": 0.23317616802400343,
"grad_norm": 2.503895542298818,
"kl": 0.053070068359375,
"learning_rate": 7.663230240549829e-07,
"loss": 0.0021,
"reward": 2.1953125,
"reward_std": 0.18460038863122463,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.2890625,
"rewards/format_reward": 1.0,
"step": 68
},
{
"clip_ratio": 0.0,
"completion_length": 105.36328125,
"epoch": 0.23660522931847408,
"grad_norm": 2.9062050779497084,
"kl": 0.074798583984375,
"learning_rate": 7.628865979381443e-07,
"loss": 0.003,
"reward": 2.22265625,
"reward_std": 0.2973867813125253,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 1.0,
"step": 69
},
{
"clip_ratio": 0.0,
"completion_length": 105.3671875,
"epoch": 0.2400342906129447,
"grad_norm": 5.380735077015675,
"kl": 0.056243896484375,
"learning_rate": 7.594501718213058e-07,
"loss": 0.0023,
"reward": 2.12890625,
"reward_std": 0.4472039174288511,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.32421875,
"rewards/format_reward": 0.99609375,
"step": 70
},
{
"clip_ratio": 0.0,
"completion_length": 107.67578125,
"epoch": 0.24346335190741533,
"grad_norm": 3.9243843409226655,
"kl": 0.046630859375,
"learning_rate": 7.560137457044673e-07,
"loss": 0.0019,
"reward": 1.95703125,
"reward_std": 0.27867609076201916,
"rewards/accuracy_reward_action": 0.7734375,
"rewards/accuracy_reward_coord": 0.18359375,
"rewards/format_reward": 1.0,
"step": 71
},
{
"clip_ratio": 0.0,
"completion_length": 108.04296875,
"epoch": 0.24689241320188599,
"grad_norm": 3.4045750384697007,
"kl": 0.0482177734375,
"learning_rate": 7.525773195876288e-07,
"loss": 0.0019,
"reward": 2.16796875,
"reward_std": 0.3632250167429447,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 1.0,
"step": 72
},
{
"clip_ratio": 0.0,
"completion_length": 105.5078125,
"epoch": 0.2503214744963566,
"grad_norm": 4.797188528974318,
"kl": 0.0504150390625,
"learning_rate": 7.491408934707904e-07,
"loss": 0.002,
"reward": 2.265625,
"reward_std": 0.3736470714211464,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.390625,
"rewards/format_reward": 0.99609375,
"step": 73
},
{
"clip_ratio": 0.0,
"completion_length": 104.95703125,
"epoch": 0.25375053579082724,
"grad_norm": 2.0498266773727294,
"kl": 0.044097900390625,
"learning_rate": 7.457044673539519e-07,
"loss": 0.0018,
"reward": 2.09765625,
"reward_std": 0.32716665603220463,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.25390625,
"rewards/format_reward": 1.0,
"step": 74
},
{
"clip_ratio": 0.0,
"completion_length": 106.2734375,
"epoch": 0.2571795970852979,
"grad_norm": 5.395875774985416,
"kl": 0.052581787109375,
"learning_rate": 7.422680412371134e-07,
"loss": 0.0021,
"reward": 2.00390625,
"reward_std": 0.34148280695080757,
"rewards/accuracy_reward_action": 0.8046875,
"rewards/accuracy_reward_coord": 0.21875,
"rewards/format_reward": 0.98046875,
"step": 75
},
{
"clip_ratio": 0.0,
"completion_length": 107.4921875,
"epoch": 0.26060865837976854,
"grad_norm": 2.5491567686917365,
"kl": 0.05126953125,
"learning_rate": 7.388316151202749e-07,
"loss": 0.0021,
"reward": 2.0625,
"reward_std": 0.2948285508900881,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.2109375,
"rewards/format_reward": 1.0,
"step": 76
},
{
"clip_ratio": 0.0,
"completion_length": 107.29296875,
"epoch": 0.26403771967423917,
"grad_norm": 3.801863021542061,
"kl": 0.05523681640625,
"learning_rate": 7.353951890034364e-07,
"loss": 0.0022,
"reward": 2.26171875,
"reward_std": 0.310369904153049,
"rewards/accuracy_reward_action": 0.9296875,
"rewards/accuracy_reward_coord": 0.33203125,
"rewards/format_reward": 1.0,
"step": 77
},
{
"clip_ratio": 0.0,
"completion_length": 101.5390625,
"epoch": 0.2674667809687098,
"grad_norm": 1.7906784993530962,
"kl": 0.057159423828125,
"learning_rate": 7.319587628865978e-07,
"loss": 0.0023,
"reward": 2.09375,
"reward_std": 0.25855094753205776,
"rewards/accuracy_reward_action": 0.83984375,
"rewards/accuracy_reward_coord": 0.25390625,
"rewards/format_reward": 1.0,
"step": 78
},
{
"clip_ratio": 0.0,
"completion_length": 102.703125,
"epoch": 0.2708958422631805,
"grad_norm": 3.2482720424647993,
"kl": 0.06103515625,
"learning_rate": 7.285223367697594e-07,
"loss": 0.0024,
"reward": 2.15625,
"reward_std": 0.2630355451256037,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.2890625,
"rewards/format_reward": 1.0,
"step": 79
},
{
"clip_ratio": 0.0,
"completion_length": 105.47265625,
"epoch": 0.2743249035576511,
"grad_norm": 6.643515027367905,
"kl": 0.075408935546875,
"learning_rate": 7.250859106529209e-07,
"loss": 0.003,
"reward": 2.25390625,
"reward_std": 0.2786760898306966,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.33984375,
"rewards/format_reward": 1.0,
"step": 80
},
{
"clip_ratio": 0.0,
"completion_length": 101.171875,
"epoch": 0.2777539648521217,
"grad_norm": 2.4853316682654345,
"kl": 0.058929443359375,
"learning_rate": 7.216494845360824e-07,
"loss": 0.0024,
"reward": 2.04296875,
"reward_std": 0.3351527014747262,
"rewards/accuracy_reward_action": 0.7734375,
"rewards/accuracy_reward_coord": 0.2734375,
"rewards/format_reward": 0.99609375,
"step": 81
},
{
"clip_ratio": 0.0,
"completion_length": 104.640625,
"epoch": 0.28118302614659235,
"grad_norm": 4.577244916480412,
"kl": 0.06158447265625,
"learning_rate": 7.182130584192439e-07,
"loss": 0.0025,
"reward": 2.0625,
"reward_std": 0.2432457758113742,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.25390625,
"rewards/format_reward": 1.0,
"step": 82
},
{
"clip_ratio": 0.0,
"completion_length": 108.34765625,
"epoch": 0.28461208744106303,
"grad_norm": 4.178949531512928,
"kl": 0.054229736328125,
"learning_rate": 7.147766323024054e-07,
"loss": 0.0022,
"reward": 2.11328125,
"reward_std": 0.3627553675323725,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.27734375,
"rewards/format_reward": 0.9921875,
"step": 83
},
{
"clip_ratio": 0.0,
"completion_length": 105.85546875,
"epoch": 0.28804114873553366,
"grad_norm": 1.9207012352711366,
"kl": 0.0654296875,
"learning_rate": 7.11340206185567e-07,
"loss": 0.0026,
"reward": 2.1484375,
"reward_std": 0.29073167592287064,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.296875,
"rewards/format_reward": 0.9921875,
"step": 84
},
{
"clip_ratio": 0.0,
"completion_length": 110.2421875,
"epoch": 0.2914702100300043,
"grad_norm": 3.819937446906779,
"kl": 0.096038818359375,
"learning_rate": 7.079037800687286e-07,
"loss": 0.0038,
"reward": 2.09375,
"reward_std": 0.3331120489165187,
"rewards/accuracy_reward_action": 0.79296875,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.9921875,
"step": 85
},
{
"clip_ratio": 0.0,
"completion_length": 104.3046875,
"epoch": 0.2948992713244749,
"grad_norm": 2.4595638989820072,
"kl": 0.07666015625,
"learning_rate": 7.044673539518901e-07,
"loss": 0.0031,
"reward": 2.16796875,
"reward_std": 0.2625892860814929,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 0.99609375,
"step": 86
},
{
"clip_ratio": 0.0,
"completion_length": 103.01171875,
"epoch": 0.2983283326189456,
"grad_norm": 2.8999588777720158,
"kl": 0.066436767578125,
"learning_rate": 7.010309278350515e-07,
"loss": 0.0027,
"reward": 2.07421875,
"reward_std": 0.257931400090456,
"rewards/accuracy_reward_action": 0.91796875,
"rewards/accuracy_reward_coord": 0.1640625,
"rewards/format_reward": 0.9921875,
"step": 87
},
{
"clip_ratio": 0.0,
"completion_length": 103.02734375,
"epoch": 0.3017573939134162,
"grad_norm": 2.4489745976786357,
"kl": 0.064361572265625,
"learning_rate": 6.97594501718213e-07,
"loss": 0.0026,
"reward": 2.19140625,
"reward_std": 0.276412351988256,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 1.0,
"step": 88
},
{
"clip_ratio": 0.0,
"completion_length": 106.01953125,
"epoch": 0.30518645520788684,
"grad_norm": 3.049686535143818,
"kl": 0.059539794921875,
"learning_rate": 6.941580756013746e-07,
"loss": 0.0024,
"reward": 2.140625,
"reward_std": 0.253778631798923,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.29296875,
"rewards/format_reward": 0.99609375,
"step": 89
},
{
"clip_ratio": 0.0,
"completion_length": 106.2421875,
"epoch": 0.30861551650235747,
"grad_norm": 4.904039350613208,
"kl": 0.05462646484375,
"learning_rate": 6.907216494845361e-07,
"loss": 0.0022,
"reward": 2.13671875,
"reward_std": 0.32303678803145885,
"rewards/accuracy_reward_action": 0.87109375,
"rewards/accuracy_reward_coord": 0.265625,
"rewards/format_reward": 1.0,
"step": 90
},
{
"clip_ratio": 0.0,
"completion_length": 103.69140625,
"epoch": 0.3120445777968281,
"grad_norm": 2.9357481693358882,
"kl": 0.055145263671875,
"learning_rate": 6.872852233676976e-07,
"loss": 0.0022,
"reward": 2.21875,
"reward_std": 0.2909819334745407,
"rewards/accuracy_reward_action": 0.84765625,
"rewards/accuracy_reward_coord": 0.37890625,
"rewards/format_reward": 0.9921875,
"step": 91
},
{
"clip_ratio": 0.0,
"completion_length": 111.52734375,
"epoch": 0.3154736390912988,
"grad_norm": 2.5466287089726616,
"kl": 0.058929443359375,
"learning_rate": 6.838487972508591e-07,
"loss": 0.0024,
"reward": 2.24609375,
"reward_std": 0.23528317362070084,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 0.9921875,
"step": 92
},
{
"clip_ratio": 0.0,
"completion_length": 104.81640625,
"epoch": 0.3189027003857694,
"grad_norm": 2.6122685492613713,
"kl": 0.062408447265625,
"learning_rate": 6.804123711340206e-07,
"loss": 0.0025,
"reward": 2.14453125,
"reward_std": 0.317325497046113,
"rewards/accuracy_reward_action": 0.91015625,
"rewards/accuracy_reward_coord": 0.23828125,
"rewards/format_reward": 0.99609375,
"step": 93
},
{
"clip_ratio": 0.0,
"completion_length": 106.265625,
"epoch": 0.32233176168024,
"grad_norm": 2.1487987963276676,
"kl": 0.0521392822265625,
"learning_rate": 6.769759450171821e-07,
"loss": 0.0021,
"reward": 2.2421875,
"reward_std": 0.31470869667828083,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 0.99609375,
"step": 94
},
{
"clip_ratio": 0.0,
"completion_length": 113.640625,
"epoch": 0.32576082297471065,
"grad_norm": 2.2669655916615024,
"kl": 0.053009033203125,
"learning_rate": 6.735395189003437e-07,
"loss": 0.0021,
"reward": 2.16796875,
"reward_std": 0.2151595028117299,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.25390625,
"rewards/format_reward": 0.98828125,
"step": 95
},
{
"clip_ratio": 0.0,
"completion_length": 111.5859375,
"epoch": 0.32918988426918133,
"grad_norm": 2.167196384717679,
"kl": 0.054656982421875,
"learning_rate": 6.701030927835051e-07,
"loss": 0.0022,
"reward": 2.25,
"reward_std": 0.29653357807546854,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 0.99609375,
"step": 96
},
{
"clip_ratio": 0.0,
"completion_length": 117.77734375,
"epoch": 0.33261894556365196,
"grad_norm": 3.6750107049628564,
"kl": 0.05615234375,
"learning_rate": 6.666666666666666e-07,
"loss": 0.0022,
"reward": 2.125,
"reward_std": 0.26538573764264584,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 0.98828125,
"step": 97
},
{
"clip_ratio": 0.0,
"completion_length": 112.62109375,
"epoch": 0.3360480068581226,
"grad_norm": 22.80143467477284,
"kl": 0.05291748046875,
"learning_rate": 6.632302405498281e-07,
"loss": 0.0021,
"reward": 2.16796875,
"reward_std": 0.24659765977412462,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.25,
"rewards/format_reward": 0.99609375,
"step": 98
},
{
"clip_ratio": 0.0,
"completion_length": 109.6015625,
"epoch": 0.3394770681525932,
"grad_norm": 2.8406924020432385,
"kl": 0.055755615234375,
"learning_rate": 6.597938144329896e-07,
"loss": 0.0022,
"reward": 2.33203125,
"reward_std": 0.22755195572972298,
"rewards/accuracy_reward_action": 0.94921875,
"rewards/accuracy_reward_coord": 0.38671875,
"rewards/format_reward": 0.99609375,
"step": 99
},
{
"clip_ratio": 0.0,
"completion_length": 107.59765625,
"epoch": 0.3429061294470639,
"grad_norm": 2.4016747335432047,
"kl": 0.056793212890625,
"learning_rate": 6.563573883161512e-07,
"loss": 0.0023,
"reward": 2.16015625,
"reward_std": 0.29211112204939127,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.29296875,
"rewards/format_reward": 0.98828125,
"step": 100
},
{
"clip_ratio": 0.0,
"completion_length": 112.04296875,
"epoch": 0.3463351907415345,
"grad_norm": 3.724814765953206,
"kl": 0.0584716796875,
"learning_rate": 6.529209621993127e-07,
"loss": 0.0023,
"reward": 2.2265625,
"reward_std": 0.2973606949672103,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.34765625,
"rewards/format_reward": 0.98828125,
"step": 101
},
{
"clip_ratio": 0.0,
"completion_length": 106.7734375,
"epoch": 0.34976425203600514,
"grad_norm": 2.903396563380394,
"kl": 0.05841064453125,
"learning_rate": 6.494845360824742e-07,
"loss": 0.0023,
"reward": 2.1484375,
"reward_std": 0.30724026821553707,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.99609375,
"step": 102
},
{
"clip_ratio": 0.0,
"completion_length": 107.01171875,
"epoch": 0.35319331333047577,
"grad_norm": 4.389088874211473,
"kl": 0.062530517578125,
"learning_rate": 6.460481099656357e-07,
"loss": 0.0025,
"reward": 2.140625,
"reward_std": 0.2465912913903594,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.23828125,
"rewards/format_reward": 0.99609375,
"step": 103
},
{
"clip_ratio": 0.0,
"completion_length": 112.8203125,
"epoch": 0.35662237462494645,
"grad_norm": 7.379865047803019,
"kl": 0.06072998046875,
"learning_rate": 6.426116838487972e-07,
"loss": 0.0024,
"reward": 2.265625,
"reward_std": 0.25317383371293545,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 0.98828125,
"step": 104
},
{
"clip_ratio": 0.0,
"completion_length": 107.44140625,
"epoch": 0.3600514359194171,
"grad_norm": 10.504549459335141,
"kl": 0.052886962890625,
"learning_rate": 6.391752577319586e-07,
"loss": 0.0021,
"reward": 2.33984375,
"reward_std": 0.28095037396997213,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.40625,
"rewards/format_reward": 0.99609375,
"step": 105
},
{
"clip_ratio": 0.0,
"completion_length": 109.91796875,
"epoch": 0.3634804972138877,
"grad_norm": 5.045451831718068,
"kl": 0.059295654296875,
"learning_rate": 6.357388316151202e-07,
"loss": 0.0024,
"reward": 2.1875,
"reward_std": 0.3055070545524359,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 0.99609375,
"step": 106
},
{
"clip_ratio": 0.0,
"completion_length": 108.6640625,
"epoch": 0.3669095585083583,
"grad_norm": 4.063268332407979,
"kl": 0.054443359375,
"learning_rate": 6.323024054982817e-07,
"loss": 0.0022,
"reward": 2.24609375,
"reward_std": 0.2154239807277918,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 1.0,
"step": 107
},
{
"clip_ratio": 0.0,
"completion_length": 103.9609375,
"epoch": 0.37033861980282895,
"grad_norm": 3.041280291099776,
"kl": 0.0679931640625,
"learning_rate": 6.288659793814433e-07,
"loss": 0.0027,
"reward": 2.12890625,
"reward_std": 0.2933124750852585,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.99609375,
"step": 108
},
{
"clip_ratio": 0.0,
"completion_length": 108.65234375,
"epoch": 0.37376768109729963,
"grad_norm": 2.45268780571316,
"kl": 0.0626220703125,
"learning_rate": 6.254295532646048e-07,
"loss": 0.0025,
"reward": 2.2421875,
"reward_std": 0.21857689507305622,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.3671875,
"rewards/format_reward": 1.0,
"step": 109
},
{
"clip_ratio": 0.0,
"completion_length": 106.7265625,
"epoch": 0.37719674239177026,
"grad_norm": 6.836912158823584,
"kl": 0.056427001953125,
"learning_rate": 6.219931271477663e-07,
"loss": 0.0023,
"reward": 2.26171875,
"reward_std": 0.20179079100489616,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 0.99609375,
"step": 110
},
{
"clip_ratio": 0.0,
"completion_length": 106.28125,
"epoch": 0.3806258036862409,
"grad_norm": 7.071556300462292,
"kl": 0.0528564453125,
"learning_rate": 6.185567010309279e-07,
"loss": 0.0021,
"reward": 2.30078125,
"reward_std": 0.3173297345638275,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.46875,
"rewards/format_reward": 0.99609375,
"step": 111
},
{
"clip_ratio": 0.0,
"completion_length": 109.140625,
"epoch": 0.3840548649807115,
"grad_norm": 3.1256736042468343,
"kl": 0.051055908203125,
"learning_rate": 6.151202749140894e-07,
"loss": 0.002,
"reward": 2.08984375,
"reward_std": 0.16978827212005854,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.1875,
"rewards/format_reward": 1.0,
"step": 112
},
{
"clip_ratio": 0.0,
"completion_length": 101.12109375,
"epoch": 0.3874839262751822,
"grad_norm": 4.067305837595256,
"kl": 0.07073974609375,
"learning_rate": 6.116838487972509e-07,
"loss": 0.0028,
"reward": 2.265625,
"reward_std": 0.32294214703142643,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.4375,
"rewards/format_reward": 1.0,
"step": 113
},
{
"clip_ratio": 0.0,
"completion_length": 104.82421875,
"epoch": 0.3909129875696528,
"grad_norm": 2.0554187661584153,
"kl": 0.06512451171875,
"learning_rate": 6.082474226804123e-07,
"loss": 0.0026,
"reward": 2.15625,
"reward_std": 0.28534636087715626,
"rewards/accuracy_reward_action": 0.81640625,
"rewards/accuracy_reward_coord": 0.34375,
"rewards/format_reward": 0.99609375,
"step": 114
},
{
"clip_ratio": 0.0,
"completion_length": 109.33203125,
"epoch": 0.39434204886412344,
"grad_norm": 3.3342749231581004,
"kl": 0.0616455078125,
"learning_rate": 6.048109965635738e-07,
"loss": 0.0025,
"reward": 2.23828125,
"reward_std": 0.32312591560184956,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 1.0,
"step": 115
},
{
"clip_ratio": 0.0,
"completion_length": 106.9296875,
"epoch": 0.39777111015859407,
"grad_norm": 3.9108229943071153,
"kl": 0.05938720703125,
"learning_rate": 6.013745704467354e-07,
"loss": 0.0024,
"reward": 2.2265625,
"reward_std": 0.33316784631460905,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 1.0,
"step": 116
},
{
"clip_ratio": 0.0,
"completion_length": 103.77734375,
"epoch": 0.40120017145306475,
"grad_norm": 4.755539569606583,
"kl": 0.056365966796875,
"learning_rate": 5.979381443298969e-07,
"loss": 0.0023,
"reward": 2.26171875,
"reward_std": 0.31653857976198196,
"rewards/accuracy_reward_action": 0.84765625,
"rewards/accuracy_reward_coord": 0.421875,
"rewards/format_reward": 0.9921875,
"step": 117
},
{
"clip_ratio": 0.0,
"completion_length": 104.7421875,
"epoch": 0.4046292327475354,
"grad_norm": 4.463216231770725,
"kl": 0.07086181640625,
"learning_rate": 5.945017182130584e-07,
"loss": 0.0028,
"reward": 2.15625,
"reward_std": 0.24183008819818497,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.2734375,
"rewards/format_reward": 1.0,
"step": 118
},
{
"clip_ratio": 0.0,
"completion_length": 103.5390625,
"epoch": 0.408058294042006,
"grad_norm": 2.924845087516111,
"kl": 0.063323974609375,
"learning_rate": 5.910652920962199e-07,
"loss": 0.0025,
"reward": 2.10546875,
"reward_std": 0.2516478532925248,
"rewards/accuracy_reward_action": 0.83203125,
"rewards/accuracy_reward_coord": 0.27734375,
"rewards/format_reward": 0.99609375,
"step": 119
},
{
"clip_ratio": 0.0,
"completion_length": 102.78125,
"epoch": 0.4114873553364766,
"grad_norm": 1.789363566352834,
"kl": 0.067291259765625,
"learning_rate": 5.876288659793814e-07,
"loss": 0.0027,
"reward": 2.078125,
"reward_std": 0.28578172344714403,
"rewards/accuracy_reward_action": 0.796875,
"rewards/accuracy_reward_coord": 0.28125,
"rewards/format_reward": 1.0,
"step": 120
},
{
"clip_ratio": 0.0,
"completion_length": 98.8984375,
"epoch": 0.4149164166309473,
"grad_norm": 3.6455708988858415,
"kl": 0.067718505859375,
"learning_rate": 5.841924398625429e-07,
"loss": 0.0027,
"reward": 2.4140625,
"reward_std": 0.24929404351860285,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.51171875,
"rewards/format_reward": 1.0,
"step": 121
},
{
"clip_ratio": 0.0,
"completion_length": 108.5,
"epoch": 0.41834547792541793,
"grad_norm": 1.7899103075940643,
"kl": 0.063751220703125,
"learning_rate": 5.807560137457045e-07,
"loss": 0.0025,
"reward": 2.15625,
"reward_std": 0.22952503617852926,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.3046875,
"rewards/format_reward": 1.0,
"step": 122
},
{
"clip_ratio": 0.0,
"completion_length": 105.60546875,
"epoch": 0.42177453921988856,
"grad_norm": 3.293296128480086,
"kl": 0.060089111328125,
"learning_rate": 5.773195876288659e-07,
"loss": 0.0024,
"reward": 2.26171875,
"reward_std": 0.25149053148925304,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 1.0,
"step": 123
},
{
"clip_ratio": 0.0,
"completion_length": 101.36328125,
"epoch": 0.4252036005143592,
"grad_norm": 2.9208531097004053,
"kl": 0.05560302734375,
"learning_rate": 5.738831615120274e-07,
"loss": 0.0022,
"reward": 2.1875,
"reward_std": 0.24435340333729982,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 0.99609375,
"step": 124
},
{
"clip_ratio": 0.0,
"completion_length": 106.1484375,
"epoch": 0.4286326618088298,
"grad_norm": 6.435409084870086,
"kl": 0.0673828125,
"learning_rate": 5.704467353951889e-07,
"loss": 0.0027,
"reward": 2.17578125,
"reward_std": 0.3276154175400734,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.3046875,
"rewards/format_reward": 0.99609375,
"step": 125
},
{
"clip_ratio": 0.0,
"completion_length": 103.1953125,
"epoch": 0.4320617231033005,
"grad_norm": 1.7152252351092034,
"kl": 0.05810546875,
"learning_rate": 5.670103092783504e-07,
"loss": 0.0023,
"reward": 2.08203125,
"reward_std": 0.2330364715307951,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.2734375,
"rewards/format_reward": 1.0,
"step": 126
},
{
"clip_ratio": 0.0,
"completion_length": 100.0625,
"epoch": 0.4354907843977711,
"grad_norm": 2.6261301563581956,
"kl": 0.07513427734375,
"learning_rate": 5.63573883161512e-07,
"loss": 0.003,
"reward": 2.4765625,
"reward_std": 0.2715753559023142,
"rewards/accuracy_reward_action": 0.95703125,
"rewards/accuracy_reward_coord": 0.51953125,
"rewards/format_reward": 1.0,
"step": 127
},
{
"clip_ratio": 0.0,
"completion_length": 104.765625,
"epoch": 0.43891984569224174,
"grad_norm": 2.903714927682652,
"kl": 0.06536865234375,
"learning_rate": 5.601374570446735e-07,
"loss": 0.0026,
"reward": 2.32421875,
"reward_std": 0.2046777717769146,
"rewards/accuracy_reward_action": 0.9609375,
"rewards/accuracy_reward_coord": 0.3671875,
"rewards/format_reward": 0.99609375,
"step": 128
},
{
"clip_ratio": 0.0,
"completion_length": 100.91015625,
"epoch": 0.44234890698671236,
"grad_norm": 3.011327246311379,
"kl": 0.0595703125,
"learning_rate": 5.56701030927835e-07,
"loss": 0.0024,
"reward": 2.41796875,
"reward_std": 0.2730232598260045,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.5078125,
"rewards/format_reward": 0.99609375,
"step": 129
},
{
"clip_ratio": 0.0,
"completion_length": 107.16796875,
"epoch": 0.44577796828118305,
"grad_norm": 2.0761484784905995,
"kl": 0.0486907958984375,
"learning_rate": 5.532646048109966e-07,
"loss": 0.0019,
"reward": 2.25,
"reward_std": 0.26358178071677685,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.38671875,
"rewards/format_reward": 1.0,
"step": 130
},
{
"clip_ratio": 0.0,
"completion_length": 103.5546875,
"epoch": 0.44920702957565367,
"grad_norm": 5.491907296283607,
"kl": 0.066558837890625,
"learning_rate": 5.498281786941581e-07,
"loss": 0.0027,
"reward": 2.2421875,
"reward_std": 0.3427243307232857,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 0.9921875,
"step": 131
},
{
"clip_ratio": 0.0,
"completion_length": 105.234375,
"epoch": 0.4526360908701243,
"grad_norm": 2.6841999435159627,
"kl": 0.06378173828125,
"learning_rate": 5.463917525773195e-07,
"loss": 0.0026,
"reward": 2.140625,
"reward_std": 0.18373434152454138,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 0.99609375,
"step": 132
},
{
"clip_ratio": 0.0,
"completion_length": 100.66796875,
"epoch": 0.4560651521645949,
"grad_norm": 3.417110980260384,
"kl": 0.060943603515625,
"learning_rate": 5.429553264604811e-07,
"loss": 0.0024,
"reward": 2.296875,
"reward_std": 0.24462180212140083,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 1.0,
"step": 133
},
{
"clip_ratio": 0.0,
"completion_length": 104.265625,
"epoch": 0.4594942134590656,
"grad_norm": 5.788338178961461,
"kl": 0.066986083984375,
"learning_rate": 5.395189003436426e-07,
"loss": 0.0027,
"reward": 2.22265625,
"reward_std": 0.20572047866880894,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 1.0,
"step": 134
},
{
"clip_ratio": 0.0,
"completion_length": 101.81640625,
"epoch": 0.46292327475353623,
"grad_norm": 2.963316780716916,
"kl": 0.063934326171875,
"learning_rate": 5.360824742268041e-07,
"loss": 0.0026,
"reward": 2.19140625,
"reward_std": 0.32010049652308226,
"rewards/accuracy_reward_action": 0.8984375,
"rewards/accuracy_reward_coord": 0.296875,
"rewards/format_reward": 0.99609375,
"step": 135
},
{
"clip_ratio": 0.0,
"completion_length": 103.42578125,
"epoch": 0.46635233604800685,
"grad_norm": 6.143892291035493,
"kl": 0.0615234375,
"learning_rate": 5.326460481099656e-07,
"loss": 0.0025,
"reward": 2.0859375,
"reward_std": 0.31697416584938765,
"rewards/accuracy_reward_action": 0.72265625,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 1.0,
"step": 136
},
{
"clip_ratio": 0.0,
"completion_length": 107.37109375,
"epoch": 0.4697813973424775,
"grad_norm": 6.9359445377938815,
"kl": 0.064300537109375,
"learning_rate": 5.292096219931271e-07,
"loss": 0.0026,
"reward": 2.2578125,
"reward_std": 0.33061617612838745,
"rewards/accuracy_reward_action": 0.9453125,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 1.0,
"step": 137
},
{
"clip_ratio": 0.0,
"completion_length": 104.6796875,
"epoch": 0.47321045863694816,
"grad_norm": 2.588210038877604,
"kl": 0.050445556640625,
"learning_rate": 5.257731958762887e-07,
"loss": 0.002,
"reward": 2.25390625,
"reward_std": 0.2738156979903579,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.37890625,
"rewards/format_reward": 0.99609375,
"step": 138
},
{
"clip_ratio": 0.0,
"completion_length": 100.6171875,
"epoch": 0.4766395199314188,
"grad_norm": 1.8678875446907652,
"kl": 0.06463623046875,
"learning_rate": 5.223367697594502e-07,
"loss": 0.0026,
"reward": 2.32421875,
"reward_std": 0.2723741699010134,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.46484375,
"rewards/format_reward": 1.0,
"step": 139
},
{
"clip_ratio": 0.0,
"completion_length": 103.08203125,
"epoch": 0.4800685812258894,
"grad_norm": 3.9056747727411705,
"kl": 0.05560302734375,
"learning_rate": 5.189003436426117e-07,
"loss": 0.0022,
"reward": 2.3671875,
"reward_std": 0.29187846928834915,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.43359375,
"rewards/format_reward": 0.99609375,
"step": 140
},
{
"clip_ratio": 0.0,
"completion_length": 110.125,
"epoch": 0.48349764252036004,
"grad_norm": 2.073984326464115,
"kl": 0.06390380859375,
"learning_rate": 5.154639175257731e-07,
"loss": 0.0026,
"reward": 2.08984375,
"reward_std": 0.25432016886770725,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.24609375,
"rewards/format_reward": 1.0,
"step": 141
},
{
"clip_ratio": 0.0,
"completion_length": 104.83984375,
"epoch": 0.48692670381483066,
"grad_norm": 2.149401200913545,
"kl": 0.056243896484375,
"learning_rate": 5.120274914089346e-07,
"loss": 0.0022,
"reward": 2.203125,
"reward_std": 0.35073984786868095,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 1.0,
"step": 142
},
{
"clip_ratio": 0.0,
"completion_length": 107.6875,
"epoch": 0.49035576510930134,
"grad_norm": 12.159410637716215,
"kl": 0.0594482421875,
"learning_rate": 5.085910652920962e-07,
"loss": 0.0024,
"reward": 2.06640625,
"reward_std": 0.22434448823332787,
"rewards/accuracy_reward_action": 0.76171875,
"rewards/accuracy_reward_coord": 0.3046875,
"rewards/format_reward": 1.0,
"step": 143
},
{
"clip_ratio": 0.0,
"completion_length": 108.7265625,
"epoch": 0.49378482640377197,
"grad_norm": 2.9357482808195083,
"kl": 0.060821533203125,
"learning_rate": 5.051546391752577e-07,
"loss": 0.0024,
"reward": 2.171875,
"reward_std": 0.25714581832289696,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 0.99609375,
"step": 144
},
{
"clip_ratio": 0.0,
"completion_length": 100.07421875,
"epoch": 0.4972138876982426,
"grad_norm": 1.6679561448700784,
"kl": 0.062225341796875,
"learning_rate": 5.017182130584192e-07,
"loss": 0.0025,
"reward": 2.29296875,
"reward_std": 0.2424728237092495,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.421875,
"rewards/format_reward": 0.99609375,
"step": 145
},
{
"clip_ratio": 0.0,
"completion_length": 106.25,
"epoch": 0.5006429489927132,
"grad_norm": 2.811523326110659,
"kl": 0.06048583984375,
"learning_rate": 4.982817869415807e-07,
"loss": 0.0024,
"reward": 2.15625,
"reward_std": 0.21675075124949217,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.26171875,
"rewards/format_reward": 0.9921875,
"step": 146
},
{
"clip_ratio": 0.0,
"completion_length": 102.671875,
"epoch": 0.5040720102871838,
"grad_norm": 9.755349918032064,
"kl": 0.059417724609375,
"learning_rate": 4.948453608247422e-07,
"loss": 0.0024,
"reward": 2.36328125,
"reward_std": 0.2519870037212968,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.48046875,
"rewards/format_reward": 0.9921875,
"step": 147
},
{
"clip_ratio": 0.0,
"completion_length": 100.1953125,
"epoch": 0.5075010715816545,
"grad_norm": 3.260159720176464,
"kl": 0.0811767578125,
"learning_rate": 4.914089347079037e-07,
"loss": 0.0032,
"reward": 2.30859375,
"reward_std": 0.32169321179389954,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.43359375,
"rewards/format_reward": 0.9921875,
"step": 148
},
{
"clip_ratio": 0.0,
"completion_length": 103.390625,
"epoch": 0.5109301328761252,
"grad_norm": 3.6764074554385893,
"kl": 0.059173583984375,
"learning_rate": 4.879725085910652e-07,
"loss": 0.0024,
"reward": 2.13671875,
"reward_std": 0.20608290564268827,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.21484375,
"rewards/format_reward": 1.0,
"step": 149
},
{
"clip_ratio": 0.0,
"completion_length": 107.28515625,
"epoch": 0.5143591941705958,
"grad_norm": 2.4013259367233135,
"kl": 0.060577392578125,
"learning_rate": 4.845360824742267e-07,
"loss": 0.0024,
"reward": 2.16796875,
"reward_std": 0.2955062361434102,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.27734375,
"rewards/format_reward": 0.99609375,
"step": 150
},
{
"clip_ratio": 0.0,
"completion_length": 104.3984375,
"epoch": 0.5177882554650665,
"grad_norm": 20.068068245198482,
"kl": 0.0634765625,
"learning_rate": 4.810996563573884e-07,
"loss": 0.0025,
"reward": 2.1484375,
"reward_std": 0.19920963142067194,
"rewards/accuracy_reward_action": 0.80859375,
"rewards/accuracy_reward_coord": 0.33984375,
"rewards/format_reward": 1.0,
"step": 151
},
{
"clip_ratio": 0.0,
"completion_length": 102.203125,
"epoch": 0.5212173167595371,
"grad_norm": 1.9706184774795734,
"kl": 0.058929443359375,
"learning_rate": 4.776632302405499e-07,
"loss": 0.0024,
"reward": 2.359375,
"reward_std": 0.29422858636826277,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.4765625,
"rewards/format_reward": 0.99609375,
"step": 152
},
{
"clip_ratio": 0.0,
"completion_length": 103.28515625,
"epoch": 0.5246463780540077,
"grad_norm": 2.0806228304348635,
"kl": 0.0726318359375,
"learning_rate": 4.742268041237113e-07,
"loss": 0.0029,
"reward": 2.296875,
"reward_std": 0.23042033333331347,
"rewards/accuracy_reward_action": 0.9296875,
"rewards/accuracy_reward_coord": 0.37109375,
"rewards/format_reward": 0.99609375,
"step": 153
},
{
"clip_ratio": 0.0,
"completion_length": 106.9765625,
"epoch": 0.5280754393484783,
"grad_norm": 4.3345166524028595,
"kl": 0.07025146484375,
"learning_rate": 4.707903780068728e-07,
"loss": 0.0028,
"reward": 2.16015625,
"reward_std": 0.35637666657567024,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.265625,
"rewards/format_reward": 0.98828125,
"step": 154
},
{
"clip_ratio": 0.0,
"completion_length": 103.31640625,
"epoch": 0.531504500642949,
"grad_norm": 7.707571860540067,
"kl": 0.061309814453125,
"learning_rate": 4.6735395189003437e-07,
"loss": 0.0025,
"reward": 2.23828125,
"reward_std": 0.18736864533275366,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 1.0,
"step": 155
},
{
"clip_ratio": 0.0,
"completion_length": 102.5859375,
"epoch": 0.5349335619374196,
"grad_norm": 2.1418912033584734,
"kl": 0.0621337890625,
"learning_rate": 4.639175257731959e-07,
"loss": 0.0025,
"reward": 2.17578125,
"reward_std": 0.22305511962622404,
"rewards/accuracy_reward_action": 0.83984375,
"rewards/accuracy_reward_coord": 0.33984375,
"rewards/format_reward": 0.99609375,
"step": 156
},
{
"clip_ratio": 0.0,
"completion_length": 108.3203125,
"epoch": 0.5383626232318903,
"grad_norm": 2.42747826627425,
"kl": 0.072479248046875,
"learning_rate": 4.6048109965635733e-07,
"loss": 0.0029,
"reward": 2.26953125,
"reward_std": 0.3015656163915992,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 0.99609375,
"step": 157
},
{
"clip_ratio": 0.0,
"completion_length": 112.7578125,
"epoch": 0.541791684526361,
"grad_norm": 2.1517223333043876,
"kl": 0.0494842529296875,
"learning_rate": 4.570446735395189e-07,
"loss": 0.002,
"reward": 2.28125,
"reward_std": 0.3337950073182583,
"rewards/accuracy_reward_action": 0.91015625,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 0.99609375,
"step": 158
},
{
"clip_ratio": 0.0,
"completion_length": 101.859375,
"epoch": 0.5452207458208316,
"grad_norm": 2.5541435290813603,
"kl": 0.055023193359375,
"learning_rate": 4.536082474226804e-07,
"loss": 0.0022,
"reward": 2.26171875,
"reward_std": 0.15250477846711874,
"rewards/accuracy_reward_action": 0.9296875,
"rewards/accuracy_reward_coord": 0.3359375,
"rewards/format_reward": 0.99609375,
"step": 159
},
{
"clip_ratio": 0.0,
"completion_length": 105.9609375,
"epoch": 0.5486498071153022,
"grad_norm": 4.51678749386207,
"kl": 0.0596923828125,
"learning_rate": 4.501718213058419e-07,
"loss": 0.0024,
"reward": 2.21875,
"reward_std": 0.31441445648670197,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 0.99609375,
"step": 160
},
{
"clip_ratio": 0.0,
"completion_length": 106.73046875,
"epoch": 0.5520788684097728,
"grad_norm": 3.0458061345691503,
"kl": 0.0517120361328125,
"learning_rate": 4.4673539518900345e-07,
"loss": 0.0021,
"reward": 2.1640625,
"reward_std": 0.2575791822746396,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 1.0,
"step": 161
},
{
"clip_ratio": 0.0,
"completion_length": 106.2265625,
"epoch": 0.5555079297042435,
"grad_norm": 5.855833740137747,
"kl": 0.066070556640625,
"learning_rate": 4.432989690721649e-07,
"loss": 0.0026,
"reward": 2.21875,
"reward_std": 0.40313977375626564,
"rewards/accuracy_reward_action": 0.83203125,
"rewards/accuracy_reward_coord": 0.390625,
"rewards/format_reward": 0.99609375,
"step": 162
},
{
"clip_ratio": 0.0,
"completion_length": 105.11328125,
"epoch": 0.5589369909987141,
"grad_norm": 2.9004827339181425,
"kl": 0.065338134765625,
"learning_rate": 4.398625429553264e-07,
"loss": 0.0026,
"reward": 2.36328125,
"reward_std": 0.18846283666789532,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.4609375,
"rewards/format_reward": 1.0,
"step": 163
},
{
"clip_ratio": 0.0,
"completion_length": 103.9375,
"epoch": 0.5623660522931847,
"grad_norm": 2.127861485761365,
"kl": 0.0596923828125,
"learning_rate": 4.3642611683848796e-07,
"loss": 0.0024,
"reward": 2.1171875,
"reward_std": 0.1504594897851348,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.296875,
"rewards/format_reward": 1.0,
"step": 164
},
{
"clip_ratio": 0.0,
"completion_length": 103.36328125,
"epoch": 0.5657951135876553,
"grad_norm": 3.131594231201492,
"kl": 0.05096435546875,
"learning_rate": 4.3298969072164947e-07,
"loss": 0.002,
"reward": 1.99609375,
"reward_std": 0.22178402822464705,
"rewards/accuracy_reward_action": 0.77734375,
"rewards/accuracy_reward_coord": 0.22265625,
"rewards/format_reward": 0.99609375,
"step": 165
},
{
"clip_ratio": 0.0,
"completion_length": 103.11328125,
"epoch": 0.5692241748821261,
"grad_norm": 2.6116928502512744,
"kl": 0.05963134765625,
"learning_rate": 4.2955326460481097e-07,
"loss": 0.0024,
"reward": 2.3515625,
"reward_std": 0.21527426224201918,
"rewards/accuracy_reward_action": 0.96484375,
"rewards/accuracy_reward_coord": 0.38671875,
"rewards/format_reward": 1.0,
"step": 166
},
{
"clip_ratio": 0.0,
"completion_length": 103.90625,
"epoch": 0.5726532361765967,
"grad_norm": 2.8869141614124287,
"kl": 0.055877685546875,
"learning_rate": 4.261168384879725e-07,
"loss": 0.0022,
"reward": 2.0,
"reward_std": 0.24724038317799568,
"rewards/accuracy_reward_action": 0.78515625,
"rewards/accuracy_reward_coord": 0.21875,
"rewards/format_reward": 0.99609375,
"step": 167
},
{
"clip_ratio": 0.0,
"completion_length": 101.80078125,
"epoch": 0.5760822974710673,
"grad_norm": 2.649000966370819,
"kl": 0.05657958984375,
"learning_rate": 4.22680412371134e-07,
"loss": 0.0023,
"reward": 2.1875,
"reward_std": 0.23751860670745373,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.3046875,
"rewards/format_reward": 0.99609375,
"step": 168
},
{
"clip_ratio": 0.0,
"completion_length": 104.24609375,
"epoch": 0.5795113587655379,
"grad_norm": 3.4059941551394037,
"kl": 0.0653076171875,
"learning_rate": 4.1924398625429554e-07,
"loss": 0.0026,
"reward": 2.171875,
"reward_std": 0.14085884485393763,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.29296875,
"rewards/format_reward": 1.0,
"step": 169
},
{
"clip_ratio": 0.0,
"completion_length": 107.3203125,
"epoch": 0.5829404200600086,
"grad_norm": 1.824844597127134,
"kl": 0.05792236328125,
"learning_rate": 4.1580756013745704e-07,
"loss": 0.0023,
"reward": 2.09765625,
"reward_std": 0.17149819899350405,
"rewards/accuracy_reward_action": 0.7578125,
"rewards/accuracy_reward_coord": 0.34765625,
"rewards/format_reward": 0.9921875,
"step": 170
},
{
"clip_ratio": 0.0,
"completion_length": 104.3984375,
"epoch": 0.5863694813544792,
"grad_norm": 2.187167981523429,
"kl": 0.0566253662109375,
"learning_rate": 4.123711340206185e-07,
"loss": 0.0023,
"reward": 2.203125,
"reward_std": 0.303788922727108,
"rewards/accuracy_reward_action": 0.76171875,
"rewards/accuracy_reward_coord": 0.44140625,
"rewards/format_reward": 1.0,
"step": 171
},
{
"clip_ratio": 0.0,
"completion_length": 100.2890625,
"epoch": 0.5897985426489498,
"grad_norm": 4.038396242056258,
"kl": 0.0596923828125,
"learning_rate": 4.0893470790378005e-07,
"loss": 0.0024,
"reward": 2.296875,
"reward_std": 0.2917390838265419,
"rewards/accuracy_reward_action": 0.93359375,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 1.0,
"step": 172
},
{
"clip_ratio": 0.0,
"completion_length": 103.3671875,
"epoch": 0.5932276039434204,
"grad_norm": 17.79728215989008,
"kl": 0.062255859375,
"learning_rate": 4.0549828178694155e-07,
"loss": 0.0025,
"reward": 2.265625,
"reward_std": 0.2565556038171053,
"rewards/accuracy_reward_action": 0.95703125,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 1.0,
"step": 173
},
{
"clip_ratio": 0.0,
"completion_length": 98.75,
"epoch": 0.5966566652378912,
"grad_norm": 3.1203510157697134,
"kl": 0.062957763671875,
"learning_rate": 4.020618556701031e-07,
"loss": 0.0025,
"reward": 2.00390625,
"reward_std": 0.20635129511356354,
"rewards/accuracy_reward_action": 0.76953125,
"rewards/accuracy_reward_coord": 0.234375,
"rewards/format_reward": 1.0,
"step": 174
},
{
"clip_ratio": 0.0,
"completion_length": 96.6328125,
"epoch": 0.6000857265323618,
"grad_norm": 3.6965883400356403,
"kl": 0.069488525390625,
"learning_rate": 3.986254295532646e-07,
"loss": 0.0028,
"reward": 2.21875,
"reward_std": 0.24308231100440025,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.390625,
"rewards/format_reward": 1.0,
"step": 175
},
{
"clip_ratio": 0.0,
"completion_length": 105.3203125,
"epoch": 0.6035147878268324,
"grad_norm": 4.26935936224658,
"kl": 0.0628662109375,
"learning_rate": 3.9518900343642607e-07,
"loss": 0.0025,
"reward": 2.23828125,
"reward_std": 0.2842397876083851,
"rewards/accuracy_reward_action": 0.90234375,
"rewards/accuracy_reward_coord": 0.3359375,
"rewards/format_reward": 1.0,
"step": 176
},
{
"clip_ratio": 0.0,
"completion_length": 107.75390625,
"epoch": 0.6069438491213031,
"grad_norm": 2.4715106694078286,
"kl": 0.0633544921875,
"learning_rate": 3.917525773195876e-07,
"loss": 0.0025,
"reward": 2.21875,
"reward_std": 0.25477858539670706,
"rewards/accuracy_reward_action": 0.87109375,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 0.99609375,
"step": 177
},
{
"clip_ratio": 0.0,
"completion_length": 108.7734375,
"epoch": 0.6103729104157737,
"grad_norm": 3.8819744356958306,
"kl": 0.064788818359375,
"learning_rate": 3.8831615120274913e-07,
"loss": 0.0026,
"reward": 2.3515625,
"reward_std": 0.25167326908558607,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.4296875,
"rewards/format_reward": 1.0,
"step": 178
},
{
"clip_ratio": 0.0,
"completion_length": 105.6875,
"epoch": 0.6138019717102443,
"grad_norm": 2.0131016192778795,
"kl": 0.0460205078125,
"learning_rate": 3.8487972508591063e-07,
"loss": 0.0018,
"reward": 2.1796875,
"reward_std": 0.3403683602809906,
"rewards/accuracy_reward_action": 0.80078125,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 0.99609375,
"step": 179
},
{
"clip_ratio": 0.0,
"completion_length": 100.91015625,
"epoch": 0.6172310330047149,
"grad_norm": 2.40863521885105,
"kl": 0.063385009765625,
"learning_rate": 3.8144329896907214e-07,
"loss": 0.0025,
"reward": 2.28125,
"reward_std": 0.230240011587739,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.40234375,
"rewards/format_reward": 0.99609375,
"step": 180
},
{
"clip_ratio": 0.0,
"completion_length": 104.58203125,
"epoch": 0.6206600942991856,
"grad_norm": 3.1847190717179026,
"kl": 0.056396484375,
"learning_rate": 3.7800687285223364e-07,
"loss": 0.0023,
"reward": 2.1328125,
"reward_std": 0.3664143308997154,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 0.99609375,
"step": 181
},
{
"clip_ratio": 0.0,
"completion_length": 105.5078125,
"epoch": 0.6240891555936562,
"grad_norm": 3.3211334385699303,
"kl": 0.05389404296875,
"learning_rate": 3.745704467353952e-07,
"loss": 0.0022,
"reward": 2.25,
"reward_std": 0.2401421070098877,
"rewards/accuracy_reward_action": 0.83984375,
"rewards/accuracy_reward_coord": 0.41015625,
"rewards/format_reward": 1.0,
"step": 182
},
{
"clip_ratio": 0.0,
"completion_length": 103.125,
"epoch": 0.6275182168881269,
"grad_norm": 3.704018941036333,
"kl": 0.065277099609375,
"learning_rate": 3.711340206185567e-07,
"loss": 0.0026,
"reward": 2.1875,
"reward_std": 0.1892604036256671,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 1.0,
"step": 183
},
{
"clip_ratio": 0.0,
"completion_length": 104.8125,
"epoch": 0.6309472781825975,
"grad_norm": 5.766622845742471,
"kl": 0.063873291015625,
"learning_rate": 3.676975945017182e-07,
"loss": 0.0026,
"reward": 2.37109375,
"reward_std": 0.26052681915462017,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.44921875,
"rewards/format_reward": 0.99609375,
"step": 184
},
{
"clip_ratio": 0.0,
"completion_length": 106.64453125,
"epoch": 0.6343763394770682,
"grad_norm": 12.693794819646222,
"kl": 0.05938720703125,
"learning_rate": 3.642611683848797e-07,
"loss": 0.0024,
"reward": 2.37890625,
"reward_std": 0.28175163455307484,
"rewards/accuracy_reward_action": 0.9453125,
"rewards/accuracy_reward_coord": 0.4375,
"rewards/format_reward": 0.99609375,
"step": 185
},
{
"clip_ratio": 0.0,
"completion_length": 104.68359375,
"epoch": 0.6378054007715388,
"grad_norm": 7.235548008451321,
"kl": 0.060546875,
"learning_rate": 3.608247422680412e-07,
"loss": 0.0024,
"reward": 2.1171875,
"reward_std": 0.28209348395466805,
"rewards/accuracy_reward_action": 0.796875,
"rewards/accuracy_reward_coord": 0.3203125,
"rewards/format_reward": 1.0,
"step": 186
},
{
"clip_ratio": 0.0,
"completion_length": 100.96484375,
"epoch": 0.6412344620660094,
"grad_norm": 7.687215435921219,
"kl": 0.065704345703125,
"learning_rate": 3.573883161512027e-07,
"loss": 0.0026,
"reward": 2.1484375,
"reward_std": 0.1814600545912981,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.24609375,
"rewards/format_reward": 0.99609375,
"step": 187
},
{
"clip_ratio": 0.0,
"completion_length": 106.16015625,
"epoch": 0.64466352336048,
"grad_norm": 3.825184151511829,
"kl": 0.058258056640625,
"learning_rate": 3.539518900343643e-07,
"loss": 0.0023,
"reward": 2.3984375,
"reward_std": 0.2644512262195349,
"rewards/accuracy_reward_action": 0.96875,
"rewards/accuracy_reward_coord": 0.4296875,
"rewards/format_reward": 1.0,
"step": 188
},
{
"clip_ratio": 0.0,
"completion_length": 105.3828125,
"epoch": 0.6480925846549507,
"grad_norm": 6.231341414590355,
"kl": 0.06439208984375,
"learning_rate": 3.5051546391752573e-07,
"loss": 0.0026,
"reward": 2.3125,
"reward_std": 0.19818973448127508,
"rewards/accuracy_reward_action": 0.953125,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 0.99609375,
"step": 189
},
{
"clip_ratio": 0.0,
"completion_length": 108.7421875,
"epoch": 0.6515216459494213,
"grad_norm": 2.6776596064402503,
"kl": 0.064849853515625,
"learning_rate": 3.470790378006873e-07,
"loss": 0.0026,
"reward": 2.265625,
"reward_std": 0.2254919307306409,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.33984375,
"rewards/format_reward": 1.0,
"step": 190
},
{
"clip_ratio": 0.0,
"completion_length": 103.80078125,
"epoch": 0.654950707243892,
"grad_norm": 2.4403487522323797,
"kl": 0.060577392578125,
"learning_rate": 3.436426116838488e-07,
"loss": 0.0024,
"reward": 2.35546875,
"reward_std": 0.25711186695843935,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.43359375,
"rewards/format_reward": 0.99609375,
"step": 191
},
{
"clip_ratio": 0.0,
"completion_length": 105.50390625,
"epoch": 0.6583797685383627,
"grad_norm": 2.726606434781571,
"kl": 0.057037353515625,
"learning_rate": 3.402061855670103e-07,
"loss": 0.0023,
"reward": 2.10546875,
"reward_std": 0.2689245492219925,
"rewards/accuracy_reward_action": 0.75,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 0.99609375,
"step": 192
},
{
"clip_ratio": 0.0,
"completion_length": 103.4375,
"epoch": 0.6618088298328333,
"grad_norm": 2.2014192461202593,
"kl": 0.059173583984375,
"learning_rate": 3.3676975945017185e-07,
"loss": 0.0024,
"reward": 2.16015625,
"reward_std": 0.1743851751089096,
"rewards/accuracy_reward_action": 0.7734375,
"rewards/accuracy_reward_coord": 0.38671875,
"rewards/format_reward": 1.0,
"step": 193
},
{
"clip_ratio": 0.0,
"completion_length": 104.65234375,
"epoch": 0.6652378911273039,
"grad_norm": 2.6193872685904025,
"kl": 0.057861328125,
"learning_rate": 3.333333333333333e-07,
"loss": 0.0023,
"reward": 2.37109375,
"reward_std": 0.3137936666607857,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.484375,
"rewards/format_reward": 0.9921875,
"step": 194
},
{
"clip_ratio": 0.0,
"completion_length": 99.796875,
"epoch": 0.6686669524217745,
"grad_norm": 2.685086268578767,
"kl": 0.066864013671875,
"learning_rate": 3.298969072164948e-07,
"loss": 0.0027,
"reward": 2.23046875,
"reward_std": 0.2271191542968154,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 1.0,
"step": 195
},
{
"clip_ratio": 0.0,
"completion_length": 107.04296875,
"epoch": 0.6720960137162452,
"grad_norm": 3.271710525721659,
"kl": 0.06494140625,
"learning_rate": 3.2646048109965636e-07,
"loss": 0.0026,
"reward": 2.0625,
"reward_std": 0.32577752880752087,
"rewards/accuracy_reward_action": 0.828125,
"rewards/accuracy_reward_coord": 0.23828125,
"rewards/format_reward": 0.99609375,
"step": 196
},
{
"clip_ratio": 0.0,
"completion_length": 113.484375,
"epoch": 0.6755250750107158,
"grad_norm": 3.36795886600386,
"kl": 0.0592803955078125,
"learning_rate": 3.2302405498281787e-07,
"loss": 0.0024,
"reward": 2.3046875,
"reward_std": 0.21713144704699516,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 1.0,
"step": 197
},
{
"clip_ratio": 0.0,
"completion_length": 104.31640625,
"epoch": 0.6789541363051864,
"grad_norm": 4.155750547891662,
"kl": 0.05615234375,
"learning_rate": 3.195876288659793e-07,
"loss": 0.0022,
"reward": 2.3984375,
"reward_std": 0.317256017588079,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.5078125,
"rewards/format_reward": 1.0,
"step": 198
},
{
"clip_ratio": 0.0,
"completion_length": 106.3515625,
"epoch": 0.682383197599657,
"grad_norm": 2.2382157235308853,
"kl": 0.05859375,
"learning_rate": 3.161512027491409e-07,
"loss": 0.0023,
"reward": 2.1484375,
"reward_std": 0.26314294897019863,
"rewards/accuracy_reward_action": 0.7890625,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 0.99609375,
"step": 199
},
{
"clip_ratio": 0.0,
"completion_length": 102.2734375,
"epoch": 0.6858122588941278,
"grad_norm": 2.254961340187696,
"kl": 0.05914306640625,
"learning_rate": 3.127147766323024e-07,
"loss": 0.0024,
"reward": 2.24609375,
"reward_std": 0.16557206399738789,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 0.9921875,
"step": 200
},
{
"clip_ratio": 0.0,
"completion_length": 107.82421875,
"epoch": 0.6892413201885984,
"grad_norm": 2.7434719866808837,
"kl": 0.05810546875,
"learning_rate": 3.0927835051546394e-07,
"loss": 0.0023,
"reward": 2.234375,
"reward_std": 0.22423463128507137,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 1.0,
"step": 201
},
{
"clip_ratio": 0.0,
"completion_length": 110.7578125,
"epoch": 0.692670381483069,
"grad_norm": 2.382655775119527,
"kl": 0.059539794921875,
"learning_rate": 3.0584192439862544e-07,
"loss": 0.0024,
"reward": 2.3125,
"reward_std": 0.34181051328778267,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.390625,
"rewards/format_reward": 1.0,
"step": 202
},
{
"clip_ratio": 0.0,
"completion_length": 107.80859375,
"epoch": 0.6960994427775397,
"grad_norm": 11.014895178134898,
"kl": 0.065826416015625,
"learning_rate": 3.024054982817869e-07,
"loss": 0.0026,
"reward": 2.2265625,
"reward_std": 0.19931845646351576,
"rewards/accuracy_reward_action": 0.89453125,
"rewards/accuracy_reward_coord": 0.33984375,
"rewards/format_reward": 0.9921875,
"step": 203
},
{
"clip_ratio": 0.0,
"completion_length": 111.80859375,
"epoch": 0.6995285040720103,
"grad_norm": 5.45051374488145,
"kl": 0.0550994873046875,
"learning_rate": 2.9896907216494845e-07,
"loss": 0.0022,
"reward": 2.17578125,
"reward_std": 0.2756984196603298,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.98046875,
"step": 204
},
{
"clip_ratio": 0.0,
"completion_length": 109.12890625,
"epoch": 0.7029575653664809,
"grad_norm": 2.5795644690128547,
"kl": 0.0765380859375,
"learning_rate": 2.9553264604810995e-07,
"loss": 0.0031,
"reward": 2.3359375,
"reward_std": 0.2355452049523592,
"rewards/accuracy_reward_action": 0.94140625,
"rewards/accuracy_reward_coord": 0.39453125,
"rewards/format_reward": 1.0,
"step": 205
},
{
"clip_ratio": 0.0,
"completion_length": 102.05859375,
"epoch": 0.7063866266609515,
"grad_norm": 1.9623795208382564,
"kl": 0.0682373046875,
"learning_rate": 2.9209621993127146e-07,
"loss": 0.0027,
"reward": 2.41796875,
"reward_std": 0.22526716440916061,
"rewards/accuracy_reward_action": 0.94140625,
"rewards/accuracy_reward_coord": 0.48046875,
"rewards/format_reward": 0.99609375,
"step": 206
},
{
"clip_ratio": 0.0,
"completion_length": 109.40625,
"epoch": 0.7098156879554222,
"grad_norm": 1.3987661010367967,
"kl": 0.05712890625,
"learning_rate": 2.8865979381443296e-07,
"loss": 0.0023,
"reward": 2.203125,
"reward_std": 0.18596168421208858,
"rewards/accuracy_reward_action": 0.85546875,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 0.9921875,
"step": 207
},
{
"clip_ratio": 0.0,
"completion_length": 108.296875,
"epoch": 0.7132447492498929,
"grad_norm": 2.5454111949134366,
"kl": 0.0555877685546875,
"learning_rate": 2.8522336769759446e-07,
"loss": 0.0022,
"reward": 2.2109375,
"reward_std": 0.2109950603917241,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.36328125,
"rewards/format_reward": 0.99609375,
"step": 208
},
{
"clip_ratio": 0.0,
"completion_length": 108.93359375,
"epoch": 0.7166738105443635,
"grad_norm": 4.680608047371802,
"kl": 0.061920166015625,
"learning_rate": 2.81786941580756e-07,
"loss": 0.0025,
"reward": 2.12109375,
"reward_std": 0.19135739468038082,
"rewards/accuracy_reward_action": 0.7734375,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 0.99609375,
"step": 209
},
{
"clip_ratio": 0.0,
"completion_length": 110.9453125,
"epoch": 0.7201028718388341,
"grad_norm": 3.8695417245183217,
"kl": 0.05340576171875,
"learning_rate": 2.783505154639175e-07,
"loss": 0.0021,
"reward": 2.26953125,
"reward_std": 0.3266449635848403,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 1.0,
"step": 210
},
{
"clip_ratio": 0.0,
"completion_length": 108.078125,
"epoch": 0.7235319331333048,
"grad_norm": 7.186477052295602,
"kl": 0.0621337890625,
"learning_rate": 2.7491408934707903e-07,
"loss": 0.0025,
"reward": 2.18359375,
"reward_std": 0.2528016325086355,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.3203125,
"rewards/format_reward": 1.0,
"step": 211
},
{
"clip_ratio": 0.0,
"completion_length": 102.28125,
"epoch": 0.7269609944277754,
"grad_norm": 2.0631486565417867,
"kl": 0.060333251953125,
"learning_rate": 2.7147766323024053e-07,
"loss": 0.0024,
"reward": 2.26171875,
"reward_std": 0.21227251924574375,
"rewards/accuracy_reward_action": 0.83203125,
"rewards/accuracy_reward_coord": 0.4296875,
"rewards/format_reward": 1.0,
"step": 212
},
{
"clip_ratio": 0.0,
"completion_length": 107.47265625,
"epoch": 0.730390055722246,
"grad_norm": 2.3982126246518214,
"kl": 0.059661865234375,
"learning_rate": 2.6804123711340204e-07,
"loss": 0.0024,
"reward": 2.375,
"reward_std": 0.22252471186220646,
"rewards/accuracy_reward_action": 0.98828125,
"rewards/accuracy_reward_coord": 0.38671875,
"rewards/format_reward": 1.0,
"step": 213
},
{
"clip_ratio": 0.0,
"completion_length": 109.74609375,
"epoch": 0.7338191170167166,
"grad_norm": 7.053341803393176,
"kl": 0.0509033203125,
"learning_rate": 2.6460481099656354e-07,
"loss": 0.002,
"reward": 2.1875,
"reward_std": 0.2063190806657076,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.32421875,
"rewards/format_reward": 1.0,
"step": 214
},
{
"clip_ratio": 0.0,
"completion_length": 102.19140625,
"epoch": 0.7372481783111873,
"grad_norm": 6.251749314545945,
"kl": 0.06671142578125,
"learning_rate": 2.611683848797251e-07,
"loss": 0.0027,
"reward": 2.203125,
"reward_std": 0.18767161015421152,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 1.0,
"step": 215
},
{
"clip_ratio": 0.0,
"completion_length": 100.71484375,
"epoch": 0.7406772396056579,
"grad_norm": 1.5875978023666912,
"kl": 0.066802978515625,
"learning_rate": 2.5773195876288655e-07,
"loss": 0.0027,
"reward": 2.40625,
"reward_std": 0.2597773177549243,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.55078125,
"rewards/format_reward": 0.99609375,
"step": 216
},
{
"clip_ratio": 0.0,
"completion_length": 109.0859375,
"epoch": 0.7441063009001286,
"grad_norm": 2.200432153622085,
"kl": 0.058349609375,
"learning_rate": 2.542955326460481e-07,
"loss": 0.0023,
"reward": 2.21484375,
"reward_std": 0.17895132210105658,
"rewards/accuracy_reward_action": 0.8984375,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 1.0,
"step": 217
},
{
"clip_ratio": 0.0,
"completion_length": 106.2890625,
"epoch": 0.7475353621945993,
"grad_norm": 1.8948662873298148,
"kl": 0.05731201171875,
"learning_rate": 2.508591065292096e-07,
"loss": 0.0023,
"reward": 2.02734375,
"reward_std": 0.2757891062647104,
"rewards/accuracy_reward_action": 0.77734375,
"rewards/accuracy_reward_coord": 0.2578125,
"rewards/format_reward": 0.9921875,
"step": 218
},
{
"clip_ratio": 0.0,
"completion_length": 106.51953125,
"epoch": 0.7509644234890699,
"grad_norm": 3.4206291883450235,
"kl": 0.06536865234375,
"learning_rate": 2.474226804123711e-07,
"loss": 0.0026,
"reward": 2.24609375,
"reward_std": 0.245152210816741,
"rewards/accuracy_reward_action": 0.9375,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 0.99609375,
"step": 219
},
{
"clip_ratio": 0.0,
"completion_length": 104.96875,
"epoch": 0.7543934847835405,
"grad_norm": 2.1536521583783585,
"kl": 0.0999755859375,
"learning_rate": 2.439862542955326e-07,
"loss": 0.004,
"reward": 2.30859375,
"reward_std": 0.18533831275999546,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.453125,
"rewards/format_reward": 0.99609375,
"step": 220
},
{
"clip_ratio": 0.0,
"completion_length": 103.06640625,
"epoch": 0.7578225460780111,
"grad_norm": 3.5470879643297595,
"kl": 0.06903076171875,
"learning_rate": 2.405498281786942e-07,
"loss": 0.0028,
"reward": 2.17578125,
"reward_std": 0.22653949353843927,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.26171875,
"rewards/format_reward": 1.0,
"step": 221
},
{
"clip_ratio": 0.0,
"completion_length": 104.58984375,
"epoch": 0.7612516073724818,
"grad_norm": 1.939359428181939,
"kl": 0.063720703125,
"learning_rate": 2.3711340206185566e-07,
"loss": 0.0025,
"reward": 2.375,
"reward_std": 0.2126298239454627,
"rewards/accuracy_reward_action": 0.97265625,
"rewards/accuracy_reward_coord": 0.40625,
"rewards/format_reward": 0.99609375,
"step": 222
},
{
"clip_ratio": 0.0,
"completion_length": 103.68359375,
"epoch": 0.7646806686669524,
"grad_norm": 2.675302553850158,
"kl": 0.058349609375,
"learning_rate": 2.3367697594501719e-07,
"loss": 0.0023,
"reward": 2.40234375,
"reward_std": 0.19962881412357092,
"rewards/accuracy_reward_action": 0.94140625,
"rewards/accuracy_reward_coord": 0.47265625,
"rewards/format_reward": 0.98828125,
"step": 223
},
{
"clip_ratio": 0.0,
"completion_length": 104.3671875,
"epoch": 0.768109729961423,
"grad_norm": 1.8156057304901947,
"kl": 0.06451416015625,
"learning_rate": 2.3024054982817866e-07,
"loss": 0.0026,
"reward": 2.21484375,
"reward_std": 0.2694152584299445,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.39453125,
"rewards/format_reward": 1.0,
"step": 224
},
{
"clip_ratio": 0.0,
"completion_length": 100.8828125,
"epoch": 0.7715387912558938,
"grad_norm": 2.292898497856771,
"kl": 0.06561279296875,
"learning_rate": 2.268041237113402e-07,
"loss": 0.0026,
"reward": 2.30078125,
"reward_std": 0.24273008853197098,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.4140625,
"rewards/format_reward": 0.99609375,
"step": 225
},
{
"clip_ratio": 0.0,
"completion_length": 99.65625,
"epoch": 0.7749678525503644,
"grad_norm": 2.4049932524471895,
"kl": 0.0560302734375,
"learning_rate": 2.2336769759450173e-07,
"loss": 0.0022,
"reward": 2.2265625,
"reward_std": 0.18312525562942028,
"rewards/accuracy_reward_action": 0.91796875,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 0.99609375,
"step": 226
},
{
"clip_ratio": 0.0,
"completion_length": 101.26953125,
"epoch": 0.778396913844835,
"grad_norm": 3.0510228384132336,
"kl": 0.067291259765625,
"learning_rate": 2.199312714776632e-07,
"loss": 0.0027,
"reward": 2.26953125,
"reward_std": 0.29204913042485714,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.3828125,
"rewards/format_reward": 1.0,
"step": 227
},
{
"clip_ratio": 0.0,
"completion_length": 105.13671875,
"epoch": 0.7818259751393056,
"grad_norm": 2.801134477925018,
"kl": 0.064300537109375,
"learning_rate": 2.1649484536082473e-07,
"loss": 0.0026,
"reward": 2.1875,
"reward_std": 0.2845958499237895,
"rewards/accuracy_reward_action": 0.87109375,
"rewards/accuracy_reward_coord": 0.3203125,
"rewards/format_reward": 0.99609375,
"step": 228
},
{
"clip_ratio": 0.0,
"completion_length": 98.6015625,
"epoch": 0.7852550364337763,
"grad_norm": 1.5238507681242188,
"kl": 0.061279296875,
"learning_rate": 2.1305841924398624e-07,
"loss": 0.0024,
"reward": 2.296875,
"reward_std": 0.19438526965677738,
"rewards/accuracy_reward_action": 0.94921875,
"rewards/accuracy_reward_coord": 0.34765625,
"rewards/format_reward": 1.0,
"step": 229
},
{
"clip_ratio": 0.0,
"completion_length": 98.84375,
"epoch": 0.7886840977282469,
"grad_norm": 3.2174810156336737,
"kl": 0.0709228515625,
"learning_rate": 2.0962199312714777e-07,
"loss": 0.0028,
"reward": 2.27734375,
"reward_std": 0.21200903225690126,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.39453125,
"rewards/format_reward": 0.99609375,
"step": 230
},
{
"clip_ratio": 0.0,
"completion_length": 99.6953125,
"epoch": 0.7921131590227175,
"grad_norm": 3.438022774954148,
"kl": 0.080291748046875,
"learning_rate": 2.0618556701030925e-07,
"loss": 0.0032,
"reward": 2.1953125,
"reward_std": 0.2185729807242751,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 1.0,
"step": 231
},
{
"clip_ratio": 0.0,
"completion_length": 98.8125,
"epoch": 0.7955422203171881,
"grad_norm": 4.55498504791608,
"kl": 0.055389404296875,
"learning_rate": 2.0274914089347078e-07,
"loss": 0.0022,
"reward": 2.09375,
"reward_std": 0.12073516938835382,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.28125,
"rewards/format_reward": 1.0,
"step": 232
},
{
"clip_ratio": 0.0,
"completion_length": 108.6953125,
"epoch": 0.7989712816116588,
"grad_norm": 9.204589267475633,
"kl": 0.059478759765625,
"learning_rate": 1.993127147766323e-07,
"loss": 0.0024,
"reward": 2.1015625,
"reward_std": 0.21872024424374104,
"rewards/accuracy_reward_action": 0.76953125,
"rewards/accuracy_reward_coord": 0.3359375,
"rewards/format_reward": 0.99609375,
"step": 233
},
{
"clip_ratio": 0.0,
"completion_length": 94.46484375,
"epoch": 0.8024003429061295,
"grad_norm": 3.1838861993641356,
"kl": 0.061279296875,
"learning_rate": 1.958762886597938e-07,
"loss": 0.0025,
"reward": 2.3828125,
"reward_std": 0.26823191065341234,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.515625,
"rewards/format_reward": 1.0,
"step": 234
},
{
"clip_ratio": 0.0,
"completion_length": 99.81640625,
"epoch": 0.8058294042006001,
"grad_norm": 5.022920175105425,
"kl": 0.080108642578125,
"learning_rate": 1.9243986254295532e-07,
"loss": 0.0032,
"reward": 2.19140625,
"reward_std": 0.2484198957681656,
"rewards/accuracy_reward_action": 0.91796875,
"rewards/accuracy_reward_coord": 0.2734375,
"rewards/format_reward": 1.0,
"step": 235
},
{
"clip_ratio": 0.0,
"completion_length": 96.625,
"epoch": 0.8092584654950707,
"grad_norm": 3.6452089439099224,
"kl": 0.069732666015625,
"learning_rate": 1.8900343642611682e-07,
"loss": 0.0028,
"reward": 2.3515625,
"reward_std": 0.23472690861672163,
"rewards/accuracy_reward_action": 0.953125,
"rewards/accuracy_reward_coord": 0.40234375,
"rewards/format_reward": 0.99609375,
"step": 236
},
{
"clip_ratio": 0.0,
"completion_length": 97.62109375,
"epoch": 0.8126875267895414,
"grad_norm": 2.7630581789606308,
"kl": 0.073883056640625,
"learning_rate": 1.8556701030927835e-07,
"loss": 0.003,
"reward": 2.1640625,
"reward_std": 0.200991983525455,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.3203125,
"rewards/format_reward": 1.0,
"step": 237
},
{
"clip_ratio": 0.0,
"completion_length": 93.6484375,
"epoch": 0.816116588084012,
"grad_norm": 3.4936750980868094,
"kl": 0.0870361328125,
"learning_rate": 1.8213058419243986e-07,
"loss": 0.0035,
"reward": 2.44921875,
"reward_std": 0.22026342526078224,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.54296875,
"rewards/format_reward": 1.0,
"step": 238
},
{
"clip_ratio": 0.0,
"completion_length": 96.6953125,
"epoch": 0.8195456493784826,
"grad_norm": 2.4875655561955297,
"kl": 0.07073974609375,
"learning_rate": 1.7869415807560136e-07,
"loss": 0.0028,
"reward": 2.48046875,
"reward_std": 0.26424654945731163,
"rewards/accuracy_reward_action": 0.94921875,
"rewards/accuracy_reward_coord": 0.53125,
"rewards/format_reward": 1.0,
"step": 239
},
{
"clip_ratio": 0.0,
"completion_length": 97.20703125,
"epoch": 0.8229747106729532,
"grad_norm": 1.7425268860067689,
"kl": 0.066192626953125,
"learning_rate": 1.7525773195876286e-07,
"loss": 0.0027,
"reward": 2.125,
"reward_std": 0.1960951965302229,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.30078125,
"rewards/format_reward": 1.0,
"step": 240
},
{
"clip_ratio": 0.0,
"completion_length": 101.17578125,
"epoch": 0.8264037719674239,
"grad_norm": 2.230888311358435,
"kl": 0.057891845703125,
"learning_rate": 1.718213058419244e-07,
"loss": 0.0023,
"reward": 2.1640625,
"reward_std": 0.23385854996740818,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 0.9921875,
"step": 241
},
{
"clip_ratio": 0.0,
"completion_length": 98.64453125,
"epoch": 0.8298328332618946,
"grad_norm": 2.799450768259519,
"kl": 0.073486328125,
"learning_rate": 1.6838487972508592e-07,
"loss": 0.0029,
"reward": 2.1640625,
"reward_std": 0.14768873061984777,
"rewards/accuracy_reward_action": 0.8828125,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 0.99609375,
"step": 242
},
{
"clip_ratio": 0.0,
"completion_length": 99.8828125,
"epoch": 0.8332618945563652,
"grad_norm": 2.4868444484547996,
"kl": 0.076263427734375,
"learning_rate": 1.649484536082474e-07,
"loss": 0.0031,
"reward": 2.26171875,
"reward_std": 0.2543821521103382,
"rewards/accuracy_reward_action": 0.8984375,
"rewards/accuracy_reward_coord": 0.3671875,
"rewards/format_reward": 0.99609375,
"step": 243
},
{
"clip_ratio": 0.0,
"completion_length": 98.5234375,
"epoch": 0.8366909558508359,
"grad_norm": 1.6923875186009838,
"kl": 0.0662841796875,
"learning_rate": 1.6151202749140893e-07,
"loss": 0.0026,
"reward": 2.234375,
"reward_std": 0.17942285630851984,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 1.0,
"step": 244
},
{
"clip_ratio": 0.0,
"completion_length": 100.28515625,
"epoch": 0.8401200171453065,
"grad_norm": 2.5018601650619496,
"kl": 0.0675048828125,
"learning_rate": 1.5807560137457044e-07,
"loss": 0.0027,
"reward": 2.19140625,
"reward_std": 0.18885111529380083,
"rewards/accuracy_reward_action": 0.88671875,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.99609375,
"step": 245
},
{
"clip_ratio": 0.0,
"completion_length": 99.0390625,
"epoch": 0.8435490784397771,
"grad_norm": 3.1069582033856453,
"kl": 0.061248779296875,
"learning_rate": 1.5463917525773197e-07,
"loss": 0.0025,
"reward": 2.18359375,
"reward_std": 0.274249623529613,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.37109375,
"rewards/format_reward": 1.0,
"step": 246
},
{
"clip_ratio": 0.0,
"completion_length": 96.8671875,
"epoch": 0.8469781397342477,
"grad_norm": 4.533280248014449,
"kl": 0.0634765625,
"learning_rate": 1.5120274914089345e-07,
"loss": 0.0025,
"reward": 2.1875,
"reward_std": 0.20642524771392345,
"rewards/accuracy_reward_action": 0.8203125,
"rewards/accuracy_reward_coord": 0.37109375,
"rewards/format_reward": 0.99609375,
"step": 247
},
{
"clip_ratio": 0.0,
"completion_length": 100.51171875,
"epoch": 0.8504072010287184,
"grad_norm": 4.186784641550275,
"kl": 0.068511962890625,
"learning_rate": 1.4776632302405498e-07,
"loss": 0.0027,
"reward": 2.1640625,
"reward_std": 0.29810838773846626,
"rewards/accuracy_reward_action": 0.82421875,
"rewards/accuracy_reward_coord": 0.34375,
"rewards/format_reward": 0.99609375,
"step": 248
},
{
"clip_ratio": 0.0,
"completion_length": 100.25,
"epoch": 0.853836262323189,
"grad_norm": 3.510648421329416,
"kl": 0.061370849609375,
"learning_rate": 1.4432989690721648e-07,
"loss": 0.0025,
"reward": 2.2734375,
"reward_std": 0.15110857971012592,
"rewards/accuracy_reward_action": 0.9140625,
"rewards/accuracy_reward_coord": 0.359375,
"rewards/format_reward": 1.0,
"step": 249
},
{
"clip_ratio": 0.0,
"completion_length": 101.171875,
"epoch": 0.8572653236176596,
"grad_norm": 2.033902237284848,
"kl": 0.07513427734375,
"learning_rate": 1.40893470790378e-07,
"loss": 0.003,
"reward": 2.33984375,
"reward_std": 0.2207325054332614,
"rewards/accuracy_reward_action": 0.8984375,
"rewards/accuracy_reward_coord": 0.44140625,
"rewards/format_reward": 1.0,
"step": 250
},
{
"clip_ratio": 0.0,
"completion_length": 100.89453125,
"epoch": 0.8606943849121304,
"grad_norm": 17.844554886648854,
"kl": 0.065521240234375,
"learning_rate": 1.3745704467353952e-07,
"loss": 0.0026,
"reward": 2.1328125,
"reward_std": 0.2432925645262003,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.296875,
"rewards/format_reward": 1.0,
"step": 251
},
{
"clip_ratio": 0.0,
"completion_length": 100.05078125,
"epoch": 0.864123446206601,
"grad_norm": 2.564885218079542,
"kl": 0.064483642578125,
"learning_rate": 1.3402061855670102e-07,
"loss": 0.0026,
"reward": 2.40625,
"reward_std": 0.23802954237908125,
"rewards/accuracy_reward_action": 0.921875,
"rewards/accuracy_reward_coord": 0.484375,
"rewards/format_reward": 1.0,
"step": 252
},
{
"clip_ratio": 0.0,
"completion_length": 100.4921875,
"epoch": 0.8675525075010716,
"grad_norm": 3.73127025667996,
"kl": 0.055328369140625,
"learning_rate": 1.3058419243986255e-07,
"loss": 0.0022,
"reward": 2.16015625,
"reward_std": 0.14309427700936794,
"rewards/accuracy_reward_action": 0.91015625,
"rewards/accuracy_reward_coord": 0.25,
"rewards/format_reward": 1.0,
"step": 253
},
{
"clip_ratio": 0.0,
"completion_length": 98.7890625,
"epoch": 0.8709815687955422,
"grad_norm": 3.118867721818899,
"kl": 0.072265625,
"learning_rate": 1.2714776632302405e-07,
"loss": 0.0029,
"reward": 2.2578125,
"reward_std": 0.2689540944993496,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.3984375,
"rewards/format_reward": 1.0,
"step": 254
},
{
"clip_ratio": 0.0,
"completion_length": 102.6640625,
"epoch": 0.8744106300900129,
"grad_norm": 11.341014381766739,
"kl": 0.060821533203125,
"learning_rate": 1.2371134020618556e-07,
"loss": 0.0024,
"reward": 2.15625,
"reward_std": 0.2020435081794858,
"rewards/accuracy_reward_action": 0.84765625,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 1.0,
"step": 255
},
{
"clip_ratio": 0.0,
"completion_length": 102.55859375,
"epoch": 0.8778396913844835,
"grad_norm": 2.3194171315590846,
"kl": 0.063568115234375,
"learning_rate": 1.202749140893471e-07,
"loss": 0.0025,
"reward": 2.12890625,
"reward_std": 0.22880323231220245,
"rewards/accuracy_reward_action": 0.81640625,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 0.99609375,
"step": 256
},
{
"clip_ratio": 0.0,
"completion_length": 99.0859375,
"epoch": 0.8812687526789541,
"grad_norm": 4.2176091245947305,
"kl": 0.050262451171875,
"learning_rate": 1.1683848797250859e-07,
"loss": 0.002,
"reward": 2.08203125,
"reward_std": 0.2253336338326335,
"rewards/accuracy_reward_action": 0.7578125,
"rewards/accuracy_reward_coord": 0.328125,
"rewards/format_reward": 0.99609375,
"step": 257
},
{
"clip_ratio": 0.0,
"completion_length": 96.01953125,
"epoch": 0.8846978139734247,
"grad_norm": 2.207390288446601,
"kl": 0.06121826171875,
"learning_rate": 1.134020618556701e-07,
"loss": 0.0024,
"reward": 2.2734375,
"reward_std": 0.2755319904536009,
"rewards/accuracy_reward_action": 0.84375,
"rewards/accuracy_reward_coord": 0.43359375,
"rewards/format_reward": 0.99609375,
"step": 258
},
{
"clip_ratio": 0.0,
"completion_length": 98.47265625,
"epoch": 0.8881268752678955,
"grad_norm": 2.186357157969937,
"kl": 0.0671844482421875,
"learning_rate": 1.099656357388316e-07,
"loss": 0.0027,
"reward": 2.2578125,
"reward_std": 0.20067016035318375,
"rewards/accuracy_reward_action": 0.890625,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 0.9921875,
"step": 259
},
{
"clip_ratio": 0.0,
"completion_length": 103.87109375,
"epoch": 0.8915559365623661,
"grad_norm": 4.422356099392904,
"kl": 0.0606689453125,
"learning_rate": 1.0652920962199312e-07,
"loss": 0.0024,
"reward": 2.2109375,
"reward_std": 0.2908977773040533,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 1.0,
"step": 260
},
{
"clip_ratio": 0.0,
"completion_length": 98.69921875,
"epoch": 0.8949849978568367,
"grad_norm": 56.35961554033066,
"kl": 0.0672607421875,
"learning_rate": 1.0309278350515462e-07,
"loss": 0.0027,
"reward": 2.24609375,
"reward_std": 0.33419137727469206,
"rewards/accuracy_reward_action": 0.8359375,
"rewards/accuracy_reward_coord": 0.41015625,
"rewards/format_reward": 1.0,
"step": 261
},
{
"clip_ratio": 0.0,
"completion_length": 96.671875,
"epoch": 0.8984140591513073,
"grad_norm": 29.13880298224036,
"kl": 0.0675048828125,
"learning_rate": 9.965635738831615e-08,
"loss": 0.0027,
"reward": 2.484375,
"reward_std": 0.19108655117452145,
"rewards/accuracy_reward_action": 0.953125,
"rewards/accuracy_reward_coord": 0.53125,
"rewards/format_reward": 1.0,
"step": 262
},
{
"clip_ratio": 0.0,
"completion_length": 98.28125,
"epoch": 0.901843120445778,
"grad_norm": 2.920667247726801,
"kl": 0.06536865234375,
"learning_rate": 9.621993127147766e-08,
"loss": 0.0026,
"reward": 2.265625,
"reward_std": 0.23333880677819252,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.39453125,
"rewards/format_reward": 0.99609375,
"step": 263
},
{
"clip_ratio": 0.0,
"completion_length": 103.6953125,
"epoch": 0.9052721817402486,
"grad_norm": 3.260184200475953,
"kl": 0.05902099609375,
"learning_rate": 9.278350515463918e-08,
"loss": 0.0024,
"reward": 2.26953125,
"reward_std": 0.29997681826353073,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.41796875,
"rewards/format_reward": 1.0,
"step": 264
},
{
"clip_ratio": 0.0,
"completion_length": 97.39453125,
"epoch": 0.9087012430347192,
"grad_norm": 1.7797660454793223,
"kl": 0.06048583984375,
"learning_rate": 8.934707903780068e-08,
"loss": 0.0024,
"reward": 2.3671875,
"reward_std": 0.275061103515327,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.4453125,
"rewards/format_reward": 0.99609375,
"step": 265
},
{
"clip_ratio": 0.0,
"completion_length": 98.48828125,
"epoch": 0.9121303043291898,
"grad_norm": 9.315295958574032,
"kl": 0.058807373046875,
"learning_rate": 8.59106529209622e-08,
"loss": 0.0024,
"reward": 2.12890625,
"reward_std": 0.22346731834113598,
"rewards/accuracy_reward_action": 0.875,
"rewards/accuracy_reward_coord": 0.2578125,
"rewards/format_reward": 0.99609375,
"step": 266
},
{
"clip_ratio": 0.0,
"completion_length": 100.91796875,
"epoch": 0.9155593656236605,
"grad_norm": 1.7100721169087572,
"kl": 0.072113037109375,
"learning_rate": 8.24742268041237e-08,
"loss": 0.0029,
"reward": 2.265625,
"reward_std": 0.1885802736505866,
"rewards/accuracy_reward_action": 0.95703125,
"rewards/accuracy_reward_coord": 0.3125,
"rewards/format_reward": 0.99609375,
"step": 267
},
{
"clip_ratio": 0.0,
"completion_length": 100.3671875,
"epoch": 0.9189884269181312,
"grad_norm": 2.5827467208411803,
"kl": 0.07196044921875,
"learning_rate": 7.903780068728522e-08,
"loss": 0.0029,
"reward": 2.2734375,
"reward_std": 0.2126556746661663,
"rewards/accuracy_reward_action": 0.8671875,
"rewards/accuracy_reward_coord": 0.40625,
"rewards/format_reward": 1.0,
"step": 268
},
{
"clip_ratio": 0.0,
"completion_length": 104.4296875,
"epoch": 0.9224174882126018,
"grad_norm": 2.490432791412204,
"kl": 0.06195068359375,
"learning_rate": 7.560137457044672e-08,
"loss": 0.0025,
"reward": 2.22265625,
"reward_std": 0.276971060782671,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.3671875,
"rewards/format_reward": 0.99609375,
"step": 269
},
{
"clip_ratio": 0.0,
"completion_length": 102.1171875,
"epoch": 0.9258465495070725,
"grad_norm": 4.622032205390463,
"kl": 0.064910888671875,
"learning_rate": 7.216494845360824e-08,
"loss": 0.0026,
"reward": 2.33984375,
"reward_std": 0.18649209756404161,
"rewards/accuracy_reward_action": 0.96484375,
"rewards/accuracy_reward_coord": 0.375,
"rewards/format_reward": 1.0,
"step": 270
},
{
"clip_ratio": 0.0,
"completion_length": 98.1875,
"epoch": 0.9292756108015431,
"grad_norm": 8.463978759815769,
"kl": 0.067596435546875,
"learning_rate": 6.872852233676976e-08,
"loss": 0.0027,
"reward": 2.1484375,
"reward_std": 0.27947098668664694,
"rewards/accuracy_reward_action": 0.84765625,
"rewards/accuracy_reward_coord": 0.30078125,
"rewards/format_reward": 1.0,
"step": 271
},
{
"clip_ratio": 0.0,
"completion_length": 100.359375,
"epoch": 0.9327046720960137,
"grad_norm": 5.986716399294285,
"kl": 0.06573486328125,
"learning_rate": 6.529209621993128e-08,
"loss": 0.0026,
"reward": 2.24609375,
"reward_std": 0.22154166735708714,
"rewards/accuracy_reward_action": 0.92578125,
"rewards/accuracy_reward_coord": 0.3203125,
"rewards/format_reward": 1.0,
"step": 272
},
{
"clip_ratio": 0.0,
"completion_length": 97.65625,
"epoch": 0.9361337333904843,
"grad_norm": 12.162794110740386,
"kl": 0.06353759765625,
"learning_rate": 6.185567010309278e-08,
"loss": 0.0025,
"reward": 2.13671875,
"reward_std": 0.2439352972432971,
"rewards/accuracy_reward_action": 0.8515625,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 1.0,
"step": 273
},
{
"clip_ratio": 0.0,
"completion_length": 101.953125,
"epoch": 0.939562794684955,
"grad_norm": 2.06279310839585,
"kl": 0.062286376953125,
"learning_rate": 5.8419243986254297e-08,
"loss": 0.0025,
"reward": 2.140625,
"reward_std": 0.19306486658751965,
"rewards/accuracy_reward_action": 0.85546875,
"rewards/accuracy_reward_coord": 0.28515625,
"rewards/format_reward": 1.0,
"step": 274
},
{
"clip_ratio": 0.0,
"completion_length": 97.53125,
"epoch": 0.9429918559794256,
"grad_norm": 1.9068683121900685,
"kl": 0.0609130859375,
"learning_rate": 5.49828178694158e-08,
"loss": 0.0024,
"reward": 2.1640625,
"reward_std": 0.277761060744524,
"rewards/accuracy_reward_action": 0.8125,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 1.0,
"step": 275
},
{
"clip_ratio": 0.0,
"completion_length": 99.32421875,
"epoch": 0.9464209172738963,
"grad_norm": 3.0285990315232683,
"kl": 0.067657470703125,
"learning_rate": 5.154639175257731e-08,
"loss": 0.0027,
"reward": 2.328125,
"reward_std": 0.290500033646822,
"rewards/accuracy_reward_action": 0.91796875,
"rewards/accuracy_reward_coord": 0.4140625,
"rewards/format_reward": 0.99609375,
"step": 276
},
{
"clip_ratio": 0.0,
"completion_length": 99.52734375,
"epoch": 0.949849978568367,
"grad_norm": 1.5774729320269285,
"kl": 0.065521240234375,
"learning_rate": 4.810996563573883e-08,
"loss": 0.0026,
"reward": 2.1796875,
"reward_std": 0.14032597374171019,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.30859375,
"rewards/format_reward": 0.9921875,
"step": 277
},
{
"clip_ratio": 0.0,
"completion_length": 99.0625,
"epoch": 0.9532790398628376,
"grad_norm": 2.2857988006535033,
"kl": 0.075531005859375,
"learning_rate": 4.467353951890034e-08,
"loss": 0.003,
"reward": 2.4375,
"reward_std": 0.27437760680913925,
"rewards/accuracy_reward_action": 0.953125,
"rewards/accuracy_reward_coord": 0.48828125,
"rewards/format_reward": 0.99609375,
"step": 278
},
{
"clip_ratio": 0.0,
"completion_length": 99.14453125,
"epoch": 0.9567081011573082,
"grad_norm": 2.9913403629956012,
"kl": 0.064788818359375,
"learning_rate": 4.123711340206185e-08,
"loss": 0.0026,
"reward": 2.41015625,
"reward_std": 0.24962799157947302,
"rewards/accuracy_reward_action": 0.9453125,
"rewards/accuracy_reward_coord": 0.46484375,
"rewards/format_reward": 1.0,
"step": 279
},
{
"clip_ratio": 0.0,
"completion_length": 95.4921875,
"epoch": 0.9601371624517788,
"grad_norm": 15.074431320780395,
"kl": 0.067779541015625,
"learning_rate": 3.780068728522336e-08,
"loss": 0.0027,
"reward": 2.31640625,
"reward_std": 0.2494807243347168,
"rewards/accuracy_reward_action": 0.90625,
"rewards/accuracy_reward_coord": 0.41015625,
"rewards/format_reward": 1.0,
"step": 280
},
{
"clip_ratio": 0.0,
"completion_length": 102.80859375,
"epoch": 0.9635662237462495,
"grad_norm": 4.661496714661205,
"kl": 0.0673828125,
"learning_rate": 3.436426116838488e-08,
"loss": 0.0027,
"reward": 2.15625,
"reward_std": 0.2237510783597827,
"rewards/accuracy_reward_action": 0.84765625,
"rewards/accuracy_reward_coord": 0.31640625,
"rewards/format_reward": 0.9921875,
"step": 281
},
{
"clip_ratio": 0.0,
"completion_length": 97.8046875,
"epoch": 0.9669952850407201,
"grad_norm": 4.1061268232089025,
"kl": 0.063873291015625,
"learning_rate": 3.092783505154639e-08,
"loss": 0.0026,
"reward": 2.30078125,
"reward_std": 0.2961772596463561,
"rewards/accuracy_reward_action": 0.87109375,
"rewards/accuracy_reward_coord": 0.4296875,
"rewards/format_reward": 1.0,
"step": 282
},
{
"clip_ratio": 0.0,
"completion_length": 98.58984375,
"epoch": 0.9704243463351907,
"grad_norm": 4.3384427281781734,
"kl": 0.06597900390625,
"learning_rate": 2.74914089347079e-08,
"loss": 0.0026,
"reward": 2.4140625,
"reward_std": 0.1958307158201933,
"rewards/accuracy_reward_action": 0.97265625,
"rewards/accuracy_reward_coord": 0.44140625,
"rewards/format_reward": 1.0,
"step": 283
},
{
"clip_ratio": 0.0,
"completion_length": 98.4609375,
"epoch": 0.9738534076296613,
"grad_norm": 2.7479574458607425,
"kl": 0.072906494140625,
"learning_rate": 2.4054982817869415e-08,
"loss": 0.0029,
"reward": 2.34765625,
"reward_std": 0.3268229737877846,
"rewards/accuracy_reward_action": 0.94140625,
"rewards/accuracy_reward_coord": 0.41015625,
"rewards/format_reward": 0.99609375,
"step": 284
},
{
"clip_ratio": 0.0,
"completion_length": 98.21875,
"epoch": 0.9772824689241321,
"grad_norm": 37.19873096068833,
"kl": 0.06402587890625,
"learning_rate": 2.0618556701030925e-08,
"loss": 0.0026,
"reward": 2.21484375,
"reward_std": 0.2886574361473322,
"rewards/accuracy_reward_action": 0.86328125,
"rewards/accuracy_reward_coord": 0.35546875,
"rewards/format_reward": 0.99609375,
"step": 285
},
{
"clip_ratio": 0.0,
"completion_length": 95.6953125,
"epoch": 0.9807115302186027,
"grad_norm": 1.3343021057081734,
"kl": 0.073028564453125,
"learning_rate": 1.718213058419244e-08,
"loss": 0.0029,
"reward": 2.18359375,
"reward_std": 0.22028045170009136,
"rewards/accuracy_reward_action": 0.83203125,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 1.0,
"step": 286
},
{
"clip_ratio": 0.0,
"completion_length": 98.89453125,
"epoch": 0.9841405915130733,
"grad_norm": 1.9072408562440208,
"kl": 0.0650634765625,
"learning_rate": 1.374570446735395e-08,
"loss": 0.0026,
"reward": 2.078125,
"reward_std": 0.1972224973142147,
"rewards/accuracy_reward_action": 0.859375,
"rewards/accuracy_reward_coord": 0.22265625,
"rewards/format_reward": 0.99609375,
"step": 287
},
{
"clip_ratio": 0.0,
"completion_length": 92.73046875,
"epoch": 0.9875696528075439,
"grad_norm": 2.7207614693913573,
"kl": 0.0624542236328125,
"learning_rate": 1.0309278350515463e-08,
"loss": 0.0025,
"reward": 2.35546875,
"reward_std": 0.2721018549054861,
"rewards/accuracy_reward_action": 0.91015625,
"rewards/accuracy_reward_coord": 0.4453125,
"rewards/format_reward": 1.0,
"step": 288
},
{
"clip_ratio": 0.0,
"completion_length": 99.9453125,
"epoch": 0.9909987141020146,
"grad_norm": 21.507130433498247,
"kl": 0.074371337890625,
"learning_rate": 6.872852233676975e-09,
"loss": 0.003,
"reward": 2.23046875,
"reward_std": 0.3140362147241831,
"rewards/accuracy_reward_action": 0.87890625,
"rewards/accuracy_reward_coord": 0.3515625,
"rewards/format_reward": 1.0,
"step": 289
},
{
"clip_ratio": 0.0,
"completion_length": 96.6484375,
"epoch": 0.9944277753964852,
"grad_norm": 2.92422057668803,
"kl": 0.077880859375,
"learning_rate": 3.4364261168384875e-09,
"loss": 0.0031,
"reward": 2.32421875,
"reward_std": 0.27937449142336845,
"rewards/accuracy_reward_action": 0.9453125,
"rewards/accuracy_reward_coord": 0.37890625,
"rewards/format_reward": 1.0,
"step": 290
},
{
"clip_ratio": 0.0,
"completion_length": 99.09722232818604,
"epoch": 0.9978568366909558,
"grad_norm": 5.0178346633813415,
"kl": 0.074005126953125,
"learning_rate": 0.0,
"loss": 0.003,
"reward": 1.9513889104127884,
"reward_std": 0.13368949433788657,
"rewards/accuracy_reward_action": 0.8125000111758709,
"rewards/accuracy_reward_coord": 0.14583333441987634,
"rewards/format_reward": 0.993055559694767,
"step": 291
}
],
"logging_steps": 1.0,
"max_steps": 291,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}