qingy2024 commited on
Commit
02e4114
·
verified ·
1 Parent(s): 1ad2b3e

Upload checkpoint 80

Browse files
README.md CHANGED
@@ -2,7 +2,7 @@
2
  base_model: smirki/UIGEN-T1.1-Qwen-7B
3
  library_name: peft
4
  ---
5
- # UIGEN T1.2 7B GRPO (Step 70 Checkpoint)
6
 
7
  > [!NOTE]
8
  > Training in progress...
@@ -38,10 +38,10 @@ library_name: peft
38
  </head>
39
  <body>
40
  <div style="width: 100%; background-color: #e0e0e0; border-radius: 25px; overflow: hidden; margin: 20px 0;">
41
- <div style="height: 30px; width: 8.75%; background-color: #76c7c0; text-align: center; line-height: 30px; color: white; border-radius: 25px 0 0 25px;">
42
  <!-- 3.75% -->
43
  </div>
44
  </div>
45
- <p style="font-family: Arial, sans-serif; font-size: 16px;">Progress: 70 out of 800 steps</p>
46
  </body>
47
  </html>
 
2
  base_model: smirki/UIGEN-T1.1-Qwen-7B
3
  library_name: peft
4
  ---
5
+ # UIGEN T1.2 7B GRPO (Step 80 Checkpoint)
6
 
7
  > [!NOTE]
8
  > Training in progress...
 
38
  </head>
39
  <body>
40
  <div style="width: 100%; background-color: #e0e0e0; border-radius: 25px; overflow: hidden; margin: 20px 0;">
41
+ <div style="height: 30px; width: 10.00%; background-color: #76c7c0; text-align: center; line-height: 30px; color: white; border-radius: 25px 0 0 25px;">
42
  <!-- 3.75% -->
43
  </div>
44
  </div>
45
+ <p style="font-family: Arial, sans-serif; font-size: 16px;">Progress: 80 out of 800 steps</p>
46
  </body>
47
  </html>
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4c0210cbc8d69bf25339cac394d3680f47d38d585c2ce17a32d2bfef602918bd
3
  size 645975704
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b34be13cac85e531353b24196023550ab69ab9ba827e465e2e74f06459028725
3
  size 645975704
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3f85628e71cb2d4229e4b57d295bb1a66923ce07a4d9175917dd03cdfb32aca6
3
- size 328476596
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82b9f616fc05fde11af892fe8956cf06deffc715c16ea8015f6b5629c1e837ad
3
+ size 329488130
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3465453f4144a1c4d83b9a44d75c55191887d72d3020646067f4968599b018c3
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c114f275457185ef8169b0bcff932246b377e10c90f21d4ea57c026f91bbe37
3
  size 14244
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4ba5595e258bb29405e8dfe44a6b9d3642581b375601de156cacb52bc67895c9
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43d39ba2d24859063abc80e3bfc36da935184acfbf4c9aaca181156dbf29ff4b
3
  size 1064
trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 0.36222509702457956,
5
  "eval_steps": 500,
6
- "global_step": 70,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -2107,6 +2107,306 @@
2107
  "rewards/think_tag_reward_func": 0.018333333544433117,
2108
  "rewards/visual_hierarchy_reward_func": 0.008970170747488737,
2109
  "step": 70
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2110
  }
2111
  ],
2112
  "logging_steps": 1,
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 0.4139715394566624,
5
  "eval_steps": 500,
6
+ "global_step": 80,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
2107
  "rewards/think_tag_reward_func": 0.018333333544433117,
2108
  "rewards/visual_hierarchy_reward_func": 0.008970170747488737,
2109
  "step": 70
2110
+ },
2111
+ {
2112
+ "completion_length": 4422.25,
2113
+ "epoch": 0.36739974126778785,
2114
+ "grad_norm": 0.1776587814092636,
2115
+ "kl": 0.0004568991789710708,
2116
+ "learning_rate": 2.9583333333333335e-06,
2117
+ "loss": 0.0,
2118
+ "reward": 0.05333333648741245,
2119
+ "reward_std": 0.02051416039466858,
2120
+ "rewards/accessibility_reward_func": 0.0,
2121
+ "rewards/animation_reward_func": 0.0,
2122
+ "rewards/color_theory_reward_func": 0.0,
2123
+ "rewards/design_system_reward_func": 0.0,
2124
+ "rewards/element_spacing_reward_func": 0.0,
2125
+ "rewards/enhanced_html_validity_reward_func": 0.029999999329447746,
2126
+ "rewards/framer_aesthetic_bonus_reward_func": 0.0,
2127
+ "rewards/general_responsiveness_reward_func": 0.0,
2128
+ "rewards/hover_reward_func": 0.0,
2129
+ "rewards/image_size_reward_func": 0.0,
2130
+ "rewards/interactivity_reward_func": 0.0,
2131
+ "rewards/link_validity_reward_func": 0.0,
2132
+ "rewards/lorem_picsum_reward_func": 0.0,
2133
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2134
+ "rewards/section_size_reward_func": 0.0,
2135
+ "rewards/tailwind_cdn_reward_func": 0.0,
2136
+ "rewards/text_contrast_reward_func": 0.0,
2137
+ "rewards/think_tag_reward_func": 0.023333332501351833,
2138
+ "rewards/visual_hierarchy_reward_func": 0.0,
2139
+ "step": 71
2140
+ },
2141
+ {
2142
+ "completion_length": 3789.0001220703125,
2143
+ "epoch": 0.37257438551099614,
2144
+ "grad_norm": 0.1108006164431572,
2145
+ "kl": 0.0005179334693821147,
2146
+ "learning_rate": 3e-06,
2147
+ "loss": 0.0,
2148
+ "reward": 0.65419393312186,
2149
+ "reward_std": 1.1815121872350574,
2150
+ "rewards/accessibility_reward_func": 0.04608333297073841,
2151
+ "rewards/animation_reward_func": 0.04000000096857548,
2152
+ "rewards/color_theory_reward_func": 0.03441666718572378,
2153
+ "rewards/design_system_reward_func": 0.1666666716337204,
2154
+ "rewards/element_spacing_reward_func": 0.025832883082330227,
2155
+ "rewards/enhanced_html_validity_reward_func": 0.04008333431556821,
2156
+ "rewards/framer_aesthetic_bonus_reward_func": 0.10500000417232513,
2157
+ "rewards/general_responsiveness_reward_func": 0.00900000031106174,
2158
+ "rewards/hover_reward_func": 0.030000001192092896,
2159
+ "rewards/image_size_reward_func": 0.003000000026077032,
2160
+ "rewards/interactivity_reward_func": 0.10000000894069672,
2161
+ "rewards/link_validity_reward_func": 0.005874999915249646,
2162
+ "rewards/lorem_picsum_reward_func": 0.0005416666681412607,
2163
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2164
+ "rewards/section_size_reward_func": 0.0,
2165
+ "rewards/tailwind_cdn_reward_func": 0.011666666716337204,
2166
+ "rewards/text_contrast_reward_func": 0.006666666711680591,
2167
+ "rewards/think_tag_reward_func": 0.02000000001862645,
2168
+ "rewards/visual_hierarchy_reward_func": 0.00936096953228116,
2169
+ "step": 72
2170
+ },
2171
+ {
2172
+ "completion_length": 4307.541687011719,
2173
+ "epoch": 0.3777490297542044,
2174
+ "grad_norm": 0.08446227759122849,
2175
+ "kl": 0.0004331339296186343,
2176
+ "learning_rate": 3.0416666666666666e-06,
2177
+ "loss": 0.0,
2178
+ "reward": 0.9242126233875751,
2179
+ "reward_std": 1.4178663492202759,
2180
+ "rewards/accessibility_reward_func": 0.07408333290368319,
2181
+ "rewards/animation_reward_func": 0.06666666641831398,
2182
+ "rewards/color_theory_reward_func": 0.06650000065565109,
2183
+ "rewards/design_system_reward_func": 0.2083333358168602,
2184
+ "rewards/element_spacing_reward_func": 0.03397295484319329,
2185
+ "rewards/enhanced_html_validity_reward_func": 0.04183333506807685,
2186
+ "rewards/framer_aesthetic_bonus_reward_func": 0.17500001192092896,
2187
+ "rewards/general_responsiveness_reward_func": 0.011000000406056643,
2188
+ "rewards/hover_reward_func": 0.03750000149011612,
2189
+ "rewards/image_size_reward_func": 0.012624999973922968,
2190
+ "rewards/interactivity_reward_func": 0.1250000074505806,
2191
+ "rewards/link_validity_reward_func": 0.009749999968335032,
2192
+ "rewards/lorem_picsum_reward_func": 0.0009583333157934248,
2193
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2194
+ "rewards/section_size_reward_func": 0.0,
2195
+ "rewards/tailwind_cdn_reward_func": 0.01458333432674408,
2196
+ "rewards/text_contrast_reward_func": 0.0025000000605359674,
2197
+ "rewards/think_tag_reward_func": 0.03166666626930237,
2198
+ "rewards/visual_hierarchy_reward_func": 0.012239584000781178,
2199
+ "step": 73
2200
+ },
2201
+ {
2202
+ "completion_length": 3586.416748046875,
2203
+ "epoch": 0.3829236739974127,
2204
+ "grad_norm": 2.2337851524353027,
2205
+ "kl": 0.0028644533740589395,
2206
+ "learning_rate": 3.0833333333333336e-06,
2207
+ "loss": 0.0001,
2208
+ "reward": 0.6753452653065324,
2209
+ "reward_std": 0.9863682854920626,
2210
+ "rewards/accessibility_reward_func": 0.032083336263895035,
2211
+ "rewards/animation_reward_func": 0.0533333346247673,
2212
+ "rewards/color_theory_reward_func": 0.04025000333786011,
2213
+ "rewards/design_system_reward_func": 0.1666666716337204,
2214
+ "rewards/element_spacing_reward_func": 0.02678571455180645,
2215
+ "rewards/enhanced_html_validity_reward_func": 0.03816666826605797,
2216
+ "rewards/framer_aesthetic_bonus_reward_func": 0.1250000074505806,
2217
+ "rewards/general_responsiveness_reward_func": 0.009000000078231096,
2218
+ "rewards/hover_reward_func": 0.030000001192092896,
2219
+ "rewards/image_size_reward_func": 0.007499999832361937,
2220
+ "rewards/interactivity_reward_func": 0.10000000894069672,
2221
+ "rewards/link_validity_reward_func": 0.006999999983236194,
2222
+ "rewards/lorem_picsum_reward_func": 0.0005833333416376263,
2223
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2224
+ "rewards/section_size_reward_func": 0.0,
2225
+ "rewards/tailwind_cdn_reward_func": 0.011666666716337204,
2226
+ "rewards/text_contrast_reward_func": 0.0008333333535119891,
2227
+ "rewards/think_tag_reward_func": 0.018333333544433117,
2228
+ "rewards/visual_hierarchy_reward_func": 0.00814285734668374,
2229
+ "step": 74
2230
+ },
2231
+ {
2232
+ "completion_length": 3698.791748046875,
2233
+ "epoch": 0.38809831824062097,
2234
+ "grad_norm": 0.15234656631946564,
2235
+ "kl": 0.00046756669325986877,
2236
+ "learning_rate": 3.125e-06,
2237
+ "loss": 0.0,
2238
+ "reward": 0.8749917726963758,
2239
+ "reward_std": 1.0643024062737823,
2240
+ "rewards/accessibility_reward_func": 0.060083333402872086,
2241
+ "rewards/animation_reward_func": 0.06666666641831398,
2242
+ "rewards/color_theory_reward_func": 0.07466666959226131,
2243
+ "rewards/design_system_reward_func": 0.2083333358168602,
2244
+ "rewards/element_spacing_reward_func": 0.036458333022892475,
2245
+ "rewards/enhanced_html_validity_reward_func": 0.040750001557171345,
2246
+ "rewards/framer_aesthetic_bonus_reward_func": 0.14750000089406967,
2247
+ "rewards/general_responsiveness_reward_func": 0.011000000406056643,
2248
+ "rewards/hover_reward_func": 0.03750000149011612,
2249
+ "rewards/image_size_reward_func": 0.013375000096857548,
2250
+ "rewards/interactivity_reward_func": 0.1250000074505806,
2251
+ "rewards/link_validity_reward_func": 0.006500000599771738,
2252
+ "rewards/lorem_picsum_reward_func": 0.001041666662786156,
2253
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2254
+ "rewards/section_size_reward_func": 0.0,
2255
+ "rewards/tailwind_cdn_reward_func": 0.008750000037252903,
2256
+ "rewards/text_contrast_reward_func": 0.0025000000605359674,
2257
+ "rewards/think_tag_reward_func": 0.021666666958481073,
2258
+ "rewards/visual_hierarchy_reward_func": 0.013200081419199705,
2259
+ "step": 75
2260
+ },
2261
+ {
2262
+ "completion_length": 3371.3751220703125,
2263
+ "epoch": 0.39327296248382926,
2264
+ "grad_norm": 0.1175752505660057,
2265
+ "kl": 0.00052155191951897,
2266
+ "learning_rate": 3.1666666666666667e-06,
2267
+ "loss": 0.0,
2268
+ "reward": 1.0607985109090805,
2269
+ "reward_std": 1.8764821887016296,
2270
+ "rewards/accessibility_reward_func": 0.058333336375653744,
2271
+ "rewards/animation_reward_func": 0.08000000193715096,
2272
+ "rewards/color_theory_reward_func": 0.08050000108778477,
2273
+ "rewards/design_system_reward_func": 0.2500000074505806,
2274
+ "rewards/element_spacing_reward_func": 0.04074477916583419,
2275
+ "rewards/enhanced_html_validity_reward_func": 0.044250001199543476,
2276
+ "rewards/framer_aesthetic_bonus_reward_func": 0.2175000086426735,
2277
+ "rewards/general_responsiveness_reward_func": 0.013000000268220901,
2278
+ "rewards/hover_reward_func": 0.04500000178813934,
2279
+ "rewards/image_size_reward_func": 0.011750000645406544,
2280
+ "rewards/interactivity_reward_func": 0.15000001341104507,
2281
+ "rewards/link_validity_reward_func": 0.012375000165775418,
2282
+ "rewards/lorem_picsum_reward_func": 0.0010416667064419016,
2283
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2284
+ "rewards/section_size_reward_func": 0.0,
2285
+ "rewards/tailwind_cdn_reward_func": 0.017500000074505806,
2286
+ "rewards/text_contrast_reward_func": 0.004166666651144624,
2287
+ "rewards/think_tag_reward_func": 0.019999999552965164,
2288
+ "rewards/visual_hierarchy_reward_func": 0.01463697268627584,
2289
+ "step": 76
2290
+ },
2291
+ {
2292
+ "completion_length": 3947.750213623047,
2293
+ "epoch": 0.3984476067270375,
2294
+ "grad_norm": 0.1153072938323021,
2295
+ "kl": 0.0004762320822919719,
2296
+ "learning_rate": 3.2083333333333337e-06,
2297
+ "loss": 0.0,
2298
+ "reward": 1.0708599789068103,
2299
+ "reward_std": 1.495519052259624,
2300
+ "rewards/accessibility_reward_func": 0.07408333290368319,
2301
+ "rewards/animation_reward_func": 0.06333333440124989,
2302
+ "rewards/color_theory_reward_func": 0.08691666647791862,
2303
+ "rewards/design_system_reward_func": 0.2500000037252903,
2304
+ "rewards/element_spacing_reward_func": 0.04120370512828231,
2305
+ "rewards/enhanced_html_validity_reward_func": 0.044750000815838575,
2306
+ "rewards/framer_aesthetic_bonus_reward_func": 0.21249999850988388,
2307
+ "rewards/general_responsiveness_reward_func": 0.012000000453554094,
2308
+ "rewards/hover_reward_func": 0.04500000178813934,
2309
+ "rewards/image_size_reward_func": 0.013500000815838575,
2310
+ "rewards/interactivity_reward_func": 0.15000000968575478,
2311
+ "rewards/link_validity_reward_func": 0.011625000042840838,
2312
+ "rewards/lorem_picsum_reward_func": 0.0011249999952269718,
2313
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2314
+ "rewards/section_size_reward_func": 0.0,
2315
+ "rewards/tailwind_cdn_reward_func": 0.01750000100582838,
2316
+ "rewards/text_contrast_reward_func": 0.004166666651144624,
2317
+ "rewards/think_tag_reward_func": 0.028333331923931837,
2318
+ "rewards/visual_hierarchy_reward_func": 0.014822917059063911,
2319
+ "step": 77
2320
+ },
2321
+ {
2322
+ "completion_length": 3954.6251220703125,
2323
+ "epoch": 0.4036222509702458,
2324
+ "grad_norm": 0.1342078447341919,
2325
+ "kl": 0.0004849206525250338,
2326
+ "learning_rate": 3.2500000000000002e-06,
2327
+ "loss": 0.0,
2328
+ "reward": 0.7595390053465962,
2329
+ "reward_std": 1.111768708564341,
2330
+ "rewards/accessibility_reward_func": 0.056000005453825,
2331
+ "rewards/animation_reward_func": 0.0533333346247673,
2332
+ "rewards/color_theory_reward_func": 0.05949999950826168,
2333
+ "rewards/design_system_reward_func": 0.1666666716337204,
2334
+ "rewards/element_spacing_reward_func": 0.02711988426744938,
2335
+ "rewards/enhanced_html_validity_reward_func": 0.040166666731238365,
2336
+ "rewards/framer_aesthetic_bonus_reward_func": 0.14750000834465027,
2337
+ "rewards/general_responsiveness_reward_func": 0.010999999940395355,
2338
+ "rewards/hover_reward_func": 0.030000001192092896,
2339
+ "rewards/image_size_reward_func": 0.011375000700354576,
2340
+ "rewards/interactivity_reward_func": 0.10000000894069672,
2341
+ "rewards/link_validity_reward_func": 0.008750000037252903,
2342
+ "rewards/lorem_picsum_reward_func": 0.0008333333535119891,
2343
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2344
+ "rewards/section_size_reward_func": 0.0,
2345
+ "rewards/tailwind_cdn_reward_func": 0.008750000037252903,
2346
+ "rewards/text_contrast_reward_func": 0.005833333358168602,
2347
+ "rewards/think_tag_reward_func": 0.023333333432674408,
2348
+ "rewards/visual_hierarchy_reward_func": 0.0093774045817554,
2349
+ "step": 78
2350
+ },
2351
+ {
2352
+ "completion_length": 3193.2084350585938,
2353
+ "epoch": 0.4087968952134541,
2354
+ "grad_norm": 0.1169697716832161,
2355
+ "kl": 0.0005721483175875619,
2356
+ "learning_rate": 3.2916666666666668e-06,
2357
+ "loss": 0.0,
2358
+ "reward": 0.7780545204877853,
2359
+ "reward_std": 1.5366026759147644,
2360
+ "rewards/accessibility_reward_func": 0.035583335906267166,
2361
+ "rewards/animation_reward_func": 0.0533333346247673,
2362
+ "rewards/color_theory_reward_func": 0.0618333350867033,
2363
+ "rewards/design_system_reward_func": 0.2083333395421505,
2364
+ "rewards/element_spacing_reward_func": 0.036137820687144995,
2365
+ "rewards/enhanced_html_validity_reward_func": 0.04000000096857548,
2366
+ "rewards/framer_aesthetic_bonus_reward_func": 0.10250000096857548,
2367
+ "rewards/general_responsiveness_reward_func": 0.011000000406056643,
2368
+ "rewards/hover_reward_func": 0.03750000149011612,
2369
+ "rewards/image_size_reward_func": 0.00849999999627471,
2370
+ "rewards/interactivity_reward_func": 0.1250000111758709,
2371
+ "rewards/link_validity_reward_func": 0.00375000003259629,
2372
+ "rewards/lorem_picsum_reward_func": 0.0006250000151339918,
2373
+ "rewards/mobile_responsiveness_reward_func": 0.0,
2374
+ "rewards/section_size_reward_func": 0.0,
2375
+ "rewards/tailwind_cdn_reward_func": 0.014583333395421505,
2376
+ "rewards/text_contrast_reward_func": 0.004166666767559946,
2377
+ "rewards/think_tag_reward_func": 0.023333332501351833,
2378
+ "rewards/visual_hierarchy_reward_func": 0.011875000316649675,
2379
+ "step": 79
2380
+ },
2381
+ {
2382
+ "completion_length": 4120.166748046875,
2383
+ "epoch": 0.4139715394566624,
2384
+ "grad_norm": 0.15751726925373077,
2385
+ "kl": 0.0004891953431069851,
2386
+ "learning_rate": 3.3333333333333333e-06,
2387
+ "loss": 0.0,
2388
+ "reward": 0.3934128861874342,
2389
+ "reward_std": 0.8384984349831939,
2390
+ "rewards/accessibility_reward_func": 0.045500002801418304,
2391
+ "rewards/animation_reward_func": 0.028666667640209198,
2392
+ "rewards/color_theory_reward_func": 0.02916666679084301,
2393
+ "rewards/design_system_reward_func": 0.0833333358168602,
2394
+ "rewards/element_spacing_reward_func": 0.014204545877873898,
2395
+ "rewards/enhanced_html_validity_reward_func": 0.03366666566580534,
2396
+ "rewards/framer_aesthetic_bonus_reward_func": 0.03999999910593033,
2397
+ "rewards/general_responsiveness_reward_func": 0.006750000175088644,
2398
+ "rewards/hover_reward_func": 0.015000000596046448,
2399
+ "rewards/image_size_reward_func": 0.007500000298023224,
2400
+ "rewards/interactivity_reward_func": 0.05000000447034836,
2401
+ "rewards/link_validity_reward_func": 0.0016250000335276127,
2402
+ "rewards/lorem_picsum_reward_func": 0.00041666667675599456,
2403
+ "rewards/mobile_responsiveness_reward_func": 0.0013333334354683757,
2404
+ "rewards/section_size_reward_func": 0.0,
2405
+ "rewards/tailwind_cdn_reward_func": 0.002916666679084301,
2406
+ "rewards/text_contrast_reward_func": 0.0008333333535119891,
2407
+ "rewards/think_tag_reward_func": 0.026666666381061077,
2408
+ "rewards/visual_hierarchy_reward_func": 0.005833333590999246,
2409
+ "step": 80
2410
  }
2411
  ],
2412
  "logging_steps": 1,
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:17076a25afee4f8ac67e7c762d9e3723db466b1ba66737c2c8fdc3a6c26b7609
3
  size 5816
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f4cb26b8217ed93495a15fd02ff1b49edf2b5b1b9e9112ceb2ae91759f3bd84
3
  size 5816