Hugging Face's logo

Weni
/
ZeroShot-3.3.18-Mistral-7b-runpod-test-Multilanguage-3.2.0

Safetensors
Portuguese
English
Spanish
trl
SFT
zeroshot
Model card Files Community
Mel-Iza0 commited on
Commit
d02af00
•
1 Parent(s): 433c312

Upload folder using huggingface_hub

Browse files
Files changed (41) hide show
  1. checkpoint-100/adapter_config.json +2 -2
  2. checkpoint-100/adapter_model.safetensors +1 -1
  3. checkpoint-100/optimizer.pt +1 -1
  4. checkpoint-100/rng_state_0.pth +2 -2
  5. checkpoint-100/rng_state_1.pth +2 -2
  6. checkpoint-100/scheduler.pt +1 -1
  7. checkpoint-100/trainer_state.json +73 -73
  8. checkpoint-100/training_args.bin +1 -1
  9. checkpoint-30/adapter_config.json +2 -2
  10. checkpoint-30/adapter_model.safetensors +1 -1
  11. checkpoint-30/optimizer.pt +1 -1
  12. checkpoint-30/rng_state_0.pth +2 -2
  13. checkpoint-30/rng_state_1.pth +2 -2
  14. checkpoint-30/rng_state_2.pth +1 -1
  15. checkpoint-30/rng_state_3.pth +1 -1
  16. checkpoint-30/trainer_state.json +21 -21
  17. checkpoint-30/training_args.bin +1 -1
  18. checkpoint-70/adapter_config.json +2 -2
  19. checkpoint-70/adapter_model.safetensors +1 -1
  20. checkpoint-70/optimizer.pt +1 -1
  21. checkpoint-70/rng_state_0.pth +2 -2
  22. checkpoint-70/rng_state_1.pth +2 -2
  23. checkpoint-70/scheduler.pt +1 -1
  24. checkpoint-70/trainer_state.json +50 -50
  25. checkpoint-70/training_args.bin +1 -1
  26. checkpoint-80/adapter_config.json +2 -2
  27. checkpoint-80/adapter_model.safetensors +1 -1
  28. checkpoint-80/optimizer.pt +1 -1
  29. checkpoint-80/rng_state_0.pth +2 -2
  30. checkpoint-80/rng_state_1.pth +2 -2
  31. checkpoint-80/scheduler.pt +1 -1
  32. checkpoint-80/trainer_state.json +59 -59
  33. checkpoint-80/training_args.bin +1 -1
  34. checkpoint-90/adapter_config.json +2 -2
  35. checkpoint-90/adapter_model.safetensors +1 -1
  36. checkpoint-90/optimizer.pt +1 -1
  37. checkpoint-90/rng_state_0.pth +2 -2
  38. checkpoint-90/rng_state_1.pth +2 -2
  39. checkpoint-90/scheduler.pt +1 -1
  40. checkpoint-90/trainer_state.json +64 -64
  41. checkpoint-90/training_args.bin +1 -1
checkpoint-100/adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
23
- "k_proj",
24
  "v_proj",
25
- "o_proj"
 
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
 
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
 
23
  "v_proj",
24
+ "o_proj",
25
+ "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
checkpoint-100/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f4a9f21cd881464354a5d422dcae331c3c4db388c18576c0ecb9a93d6bce5fb5
3
  size 436242776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8307659fd0862cf0805b8206dbc2159ba622b3301d90d9634a43e58ab49be69d
3
  size 436242776
checkpoint-100/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3cdc22fb86973bff2cc05ff2b6c9d57e33f82eaed8c614ce25ca8e457bf79192
3
  size 872568314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7ba7a8726d78998756dd55f553ecc0a377148f2698744401894deb794dff765
3
  size 872568314
checkpoint-100/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:94c603473dc4b52ea920ce532750d7fb82f8a6b1014d2d3bfeecb430ff7916e7
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23673dcfe8f26201d1b64968cfb8e7aeafa0dd44591f91dcef9addc46b95c804
3
+ size 15024
checkpoint-100/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2fb0b5af4969d319c137ef120d1a81b3c1905b6f9244439e660950fd2eabdcd0
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6469a17730051cc187688ed7abb5fd4c8dde61701d8ada08c6a999a9463ac217
3
+ size 15024
checkpoint-100/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3d2ae6e01a1808af3d88a28dc34a871dccf737b3b52505dba8e2c66818fbd018
3
  size 1000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad10f9bdd16d4aa7ed3e572ac6474b4271d2e92b701684b91f54abd90b8638ef
3
  size 1000
checkpoint-100/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.3921997845172882,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-100",
4
- "epoch": 0.49566294919454773,
5
  "eval_steps": 10,
6
  "global_step": 100,
7
  "is_hyper_param_search": false,
@@ -9,118 +9,118 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.05,
13
- "eval_loss": 0.6367942690849304,
14
- "eval_runtime": 180.8957,
15
- "eval_samples_per_second": 15.849,
16
- "eval_steps_per_second": 0.995,
17
  "step": 10
18
  },
19
  {
20
- "epoch": 0.1,
21
- "grad_norm": 0.34346863627433777,
22
- "learning_rate": 0.00019510565162951537,
23
- "loss": 0.9774,
24
  "step": 20
25
  },
26
  {
27
- "epoch": 0.1,
28
- "eval_loss": 0.48089343309402466,
29
- "eval_runtime": 180.8882,
30
- "eval_samples_per_second": 15.85,
31
- "eval_steps_per_second": 0.995,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.15,
36
- "eval_loss": 0.44581571221351624,
37
- "eval_runtime": 180.8305,
38
- "eval_samples_per_second": 15.855,
39
- "eval_steps_per_second": 0.995,
40
  "step": 30
41
  },
42
  {
43
- "epoch": 0.2,
44
- "grad_norm": 0.22729116678237915,
45
- "learning_rate": 0.0001529919264233205,
46
- "loss": 0.4459,
47
  "step": 40
48
  },
49
  {
50
- "epoch": 0.2,
51
- "eval_loss": 0.4221758544445038,
52
- "eval_runtime": 180.8406,
53
- "eval_samples_per_second": 15.854,
54
- "eval_steps_per_second": 0.995,
55
  "step": 40
56
  },
57
  {
58
- "epoch": 0.25,
59
- "eval_loss": 0.40922558307647705,
60
- "eval_runtime": 181.1791,
61
- "eval_samples_per_second": 15.824,
62
- "eval_steps_per_second": 0.993,
63
  "step": 50
64
  },
65
  {
66
- "epoch": 0.3,
67
- "grad_norm": 0.27356910705566406,
68
- "learning_rate": 8.608268990399349e-05,
69
- "loss": 0.4132,
70
  "step": 60
71
  },
72
  {
73
- "epoch": 0.3,
74
- "eval_loss": 0.4156472384929657,
75
- "eval_runtime": 180.7116,
76
- "eval_samples_per_second": 15.865,
77
- "eval_steps_per_second": 0.996,
78
  "step": 60
79
  },
80
  {
81
- "epoch": 0.35,
82
- "eval_loss": 0.39848339557647705,
83
- "eval_runtime": 180.7856,
84
- "eval_samples_per_second": 15.859,
85
- "eval_steps_per_second": 0.996,
86
  "step": 70
87
  },
88
  {
89
- "epoch": 0.4,
90
- "grad_norm": 0.1912572681903839,
91
- "learning_rate": 2.5685517452260567e-05,
92
- "loss": 0.4004,
93
  "step": 80
94
  },
95
  {
96
- "epoch": 0.4,
97
- "eval_loss": 0.3947867155075073,
98
- "eval_runtime": 180.6659,
99
- "eval_samples_per_second": 15.869,
100
- "eval_steps_per_second": 0.996,
101
  "step": 80
102
  },
103
  {
104
- "epoch": 0.45,
105
- "eval_loss": 0.39267557859420776,
106
- "eval_runtime": 180.6936,
107
- "eval_samples_per_second": 15.867,
108
- "eval_steps_per_second": 0.996,
109
  "step": 90
110
  },
111
  {
112
- "epoch": 0.5,
113
- "grad_norm": 0.1869862824678421,
114
- "learning_rate": 6.09172980904238e-08,
115
- "loss": 0.3929,
116
  "step": 100
117
  },
118
  {
119
- "epoch": 0.5,
120
- "eval_loss": 0.3921997845172882,
121
- "eval_runtime": 180.1721,
122
- "eval_samples_per_second": 15.913,
123
- "eval_steps_per_second": 0.999,
124
  "step": 100
125
  }
126
  ],
@@ -129,7 +129,7 @@
129
  "num_input_tokens_seen": 0,
130
  "num_train_epochs": 1,
131
  "save_steps": 10,
132
- "total_flos": 4.521353175336223e+17,
133
  "train_batch_size": 16,
134
  "trial_name": null,
135
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.3796847462654114,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-100",
4
+ "epoch": 0.9900990099009901,
5
  "eval_steps": 10,
6
  "global_step": 100,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.1,
13
+ "eval_loss": 0.6981944441795349,
14
+ "eval_runtime": 94.6462,
15
+ "eval_samples_per_second": 30.292,
16
+ "eval_steps_per_second": 0.951,
17
  "step": 10
18
  },
19
  {
20
+ "epoch": 0.2,
21
+ "grad_norm": 0.2654201090335846,
22
+ "learning_rate": 0.0001961261695938319,
23
+ "loss": 1.0016,
24
  "step": 20
25
  },
26
  {
27
+ "epoch": 0.2,
28
+ "eval_loss": 0.467332661151886,
29
+ "eval_runtime": 94.5373,
30
+ "eval_samples_per_second": 30.327,
31
+ "eval_steps_per_second": 0.952,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.3,
36
+ "eval_loss": 0.434685617685318,
37
+ "eval_runtime": 94.5899,
38
+ "eval_samples_per_second": 30.31,
39
+ "eval_steps_per_second": 0.951,
40
  "step": 30
41
  },
42
  {
43
+ "epoch": 0.4,
44
+ "grad_norm": 1.3763986825942993,
45
+ "learning_rate": 0.0001559192903470747,
46
+ "loss": 0.4357,
47
  "step": 40
48
  },
49
  {
50
+ "epoch": 0.4,
51
+ "eval_loss": 0.4082697033882141,
52
+ "eval_runtime": 94.551,
53
+ "eval_samples_per_second": 30.322,
54
+ "eval_steps_per_second": 0.952,
55
  "step": 40
56
  },
57
  {
58
+ "epoch": 0.5,
59
+ "eval_loss": 0.40642818808555603,
60
+ "eval_runtime": 94.5831,
61
+ "eval_samples_per_second": 30.312,
62
+ "eval_steps_per_second": 0.952,
63
  "step": 50
64
  },
65
  {
66
+ "epoch": 0.59,
67
+ "grad_norm": 0.16300354897975922,
68
+ "learning_rate": 8.954715367323468e-05,
69
+ "loss": 0.4012,
70
  "step": 60
71
  },
72
  {
73
+ "epoch": 0.59,
74
+ "eval_loss": 0.39164847135543823,
75
+ "eval_runtime": 94.6074,
76
+ "eval_samples_per_second": 30.304,
77
+ "eval_steps_per_second": 0.951,
78
  "step": 60
79
  },
80
  {
81
+ "epoch": 0.69,
82
+ "eval_loss": 0.3856147527694702,
83
+ "eval_runtime": 94.5883,
84
+ "eval_samples_per_second": 30.31,
85
+ "eval_steps_per_second": 0.951,
86
  "step": 70
87
  },
88
  {
89
+ "epoch": 0.79,
90
+ "grad_norm": 0.14875428378582,
91
+ "learning_rate": 2.8066019966134904e-05,
92
+ "loss": 0.3846,
93
  "step": 80
94
  },
95
  {
96
+ "epoch": 0.79,
97
+ "eval_loss": 0.38140159845352173,
98
+ "eval_runtime": 94.5138,
99
+ "eval_samples_per_second": 30.334,
100
+ "eval_steps_per_second": 0.952,
101
  "step": 80
102
  },
103
  {
104
+ "epoch": 0.89,
105
+ "eval_loss": 0.3798506557941437,
106
+ "eval_runtime": 94.6456,
107
+ "eval_samples_per_second": 30.292,
108
+ "eval_steps_per_second": 0.951,
109
  "step": 90
110
  },
111
  {
112
+ "epoch": 0.99,
113
+ "grad_norm": 3.9488003253936768,
114
+ "learning_rate": 2.4359497401758024e-07,
115
+ "loss": 0.3775,
116
  "step": 100
117
  },
118
  {
119
+ "epoch": 0.99,
120
+ "eval_loss": 0.3796847462654114,
121
+ "eval_runtime": 94.5333,
122
+ "eval_samples_per_second": 30.328,
123
+ "eval_steps_per_second": 0.952,
124
  "step": 100
125
  }
126
  ],
 
129
  "num_input_tokens_seen": 0,
130
  "num_train_epochs": 1,
131
  "save_steps": 10,
132
+ "total_flos": 9.048313688175411e+17,
133
  "train_batch_size": 16,
134
  "trial_name": null,
135
  "trial_params": null
checkpoint-100/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
3
  size 5176
checkpoint-30/adapter_config.json CHANGED
@@ -19,9 +19,9 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "q_proj",
23
- "k_proj",
24
  "v_proj",
 
 
25
  "o_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
 
22
  "v_proj",
23
+ "k_proj",
24
+ "q_proj",
25
  "o_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
checkpoint-30/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:24867322dfb4671521880b7952dfecf869c9ddf50ba8c8d0ab60442d46c3b6e7
3
  size 436242776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2492817cb114918e710f190656fdb97ff34ac6dde3bb65dcf275cedb6b03114f
3
  size 436242776
checkpoint-30/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6a095acb983c64222b3980c352b38e58ba4bd6e2f0b8e559eddea78a65277790
3
  size 872568314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:512f441cf6c5cc3424b3c70b05cf0de8f665d6ecff376c200b484e26d83afdb7
3
  size 872568314
checkpoint-30/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2c2a616f0c7a81cc42f0138e0036f3bc925663524713430ec02128feb7cc22f7
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b25b39cb0bb904eb749e12c5e06512371c04c82fb8f08f7382fac691210c1fc
3
+ size 15024
checkpoint-30/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3547aee58cf2280e2d966f891fba83ef181a792754cf251e62b59ae644dd2253
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:528debf43ac8e68f3972cc6eedc188d70690e0a9bfc2feeeff97e757be99c6b8
3
+ size 15024
checkpoint-30/rng_state_2.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dac99c37629c1d6e9b1b5b878c74c68d63aee51552b081f36bd9f3d65433ef80
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a173fc66329564d6c1030057eed0dcef01dbb013745312b30f88098bc55c0f4
3
  size 15024
checkpoint-30/rng_state_3.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1011841c7668479718ff7793f783bbec4a8142be22d159d6cb4be0ab72b68bb6
3
  size 15024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edb07a30cd8f91db33e04840408f09052e226b29a4f31ec4f5fb945308a1a98e
3
  size 15024
checkpoint-30/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.44581571221351624,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-30",
4
- "epoch": 0.14869888475836432,
5
  "eval_steps": 10,
6
  "global_step": 30,
7
  "is_hyper_param_search": false,
@@ -9,34 +9,34 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.05,
13
- "eval_loss": 0.6367942690849304,
14
- "eval_runtime": 180.8957,
15
- "eval_samples_per_second": 15.849,
16
- "eval_steps_per_second": 0.995,
17
  "step": 10
18
  },
19
  {
20
- "epoch": 0.1,
21
- "grad_norm": 0.34346863627433777,
22
  "learning_rate": 0.00019510565162951537,
23
- "loss": 0.9774,
24
  "step": 20
25
  },
26
  {
27
- "epoch": 0.1,
28
- "eval_loss": 0.48089343309402466,
29
- "eval_runtime": 180.8882,
30
- "eval_samples_per_second": 15.85,
31
- "eval_steps_per_second": 0.995,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.15,
36
- "eval_loss": 0.44581571221351624,
37
- "eval_runtime": 180.8305,
38
- "eval_samples_per_second": 15.855,
39
- "eval_steps_per_second": 0.995,
40
  "step": 30
41
  }
42
  ],
@@ -45,7 +45,7 @@
45
  "num_input_tokens_seen": 0,
46
  "num_train_epochs": 1,
47
  "save_steps": 10,
48
- "total_flos": 1.3497349192024064e+17,
49
  "train_batch_size": 16,
50
  "trial_name": null,
51
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.4341636002063751,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-30",
4
+ "epoch": 0.297029702970297,
5
  "eval_steps": 10,
6
  "global_step": 30,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.1,
13
+ "eval_loss": 0.6292536854743958,
14
+ "eval_runtime": 93.2576,
15
+ "eval_samples_per_second": 30.743,
16
+ "eval_steps_per_second": 0.965,
17
  "step": 10
18
  },
19
  {
20
+ "epoch": 0.2,
21
+ "grad_norm": 0.24918170273303986,
22
  "learning_rate": 0.00019510565162951537,
23
+ "loss": 0.9604,
24
  "step": 20
25
  },
26
  {
27
+ "epoch": 0.2,
28
+ "eval_loss": 0.46683359146118164,
29
+ "eval_runtime": 93.2154,
30
+ "eval_samples_per_second": 30.757,
31
+ "eval_steps_per_second": 0.966,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.3,
36
+ "eval_loss": 0.4341636002063751,
37
+ "eval_runtime": 93.1765,
38
+ "eval_samples_per_second": 30.77,
39
+ "eval_steps_per_second": 0.966,
40
  "step": 30
41
  }
42
  ],
 
45
  "num_input_tokens_seen": 0,
46
  "num_train_epochs": 1,
47
  "save_steps": 10,
48
+ "total_flos": 2.711425824766034e+17,
49
  "train_batch_size": 16,
50
  "trial_name": null,
51
  "trial_params": null
checkpoint-30/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daf4ca86cd3cb3688a488e999435cb5aec8a4b0702902c7de81df751b2180e05
3
  size 5176
checkpoint-70/adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
23
- "k_proj",
24
  "v_proj",
25
- "o_proj"
 
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
 
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
 
23
  "v_proj",
24
+ "o_proj",
25
+ "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
checkpoint-70/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cfb912af6b1796046b92cd31a140d407273ff753ac7f3269fc3c5f3185cb84e1
3
  size 436242776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3729ade0578702d582ba4201d2056d525c6b98445fc9f287499aa8cc4a04ad0f
3
  size 436242776
checkpoint-70/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:631ee3323927506f442194f35db41e1ecdfb11bea1ce0e91876dfd94f7df30c1
3
  size 872568314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f41106fec66c8da66955717a5c0e57087dbb2ce1e9c2786539e9d14f13dba082
3
  size 872568314
checkpoint-70/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b7a0a43eb1da4e2bdacd5b9613bc246aef81888090db42765afb6f369c67c4c1
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f23d434a5adf472af32d0d067bd6c7196c5fada3cf7bbec69afdf226737ae9c
3
+ size 15024
checkpoint-70/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:90cb909835d393126c788ffdccbb0e802fe522bcd647fc16615a348d6c39cfd4
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8cd675fa29ceb7104e041bb8b6f813a24babb71dc2874373344d885230ca229
3
+ size 15024
checkpoint-70/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:74915e0aef978c84c56eee8bed06ba18d446da4dd1125eda8fbb469b255811be
3
  size 1000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a45b3b845e6140514d22eb89d3d82950a7efd6b6e0c772b6f3e23d8289d822fd
3
  size 1000
checkpoint-70/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.39848339557647705,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-70",
4
- "epoch": 0.3469640644361834,
5
  "eval_steps": 10,
6
  "global_step": 70,
7
  "is_hyper_param_search": false,
@@ -9,80 +9,80 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.05,
13
- "eval_loss": 0.6367942690849304,
14
- "eval_runtime": 180.8957,
15
- "eval_samples_per_second": 15.849,
16
- "eval_steps_per_second": 0.995,
17
  "step": 10
18
  },
19
  {
20
- "epoch": 0.1,
21
- "grad_norm": 0.34346863627433777,
22
- "learning_rate": 0.00019510565162951537,
23
- "loss": 0.9774,
24
  "step": 20
25
  },
26
  {
27
- "epoch": 0.1,
28
- "eval_loss": 0.48089343309402466,
29
- "eval_runtime": 180.8882,
30
- "eval_samples_per_second": 15.85,
31
- "eval_steps_per_second": 0.995,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.15,
36
- "eval_loss": 0.44581571221351624,
37
- "eval_runtime": 180.8305,
38
- "eval_samples_per_second": 15.855,
39
- "eval_steps_per_second": 0.995,
40
  "step": 30
41
  },
42
  {
43
- "epoch": 0.2,
44
- "grad_norm": 0.22729116678237915,
45
- "learning_rate": 0.0001529919264233205,
46
- "loss": 0.4459,
47
  "step": 40
48
  },
49
  {
50
- "epoch": 0.2,
51
- "eval_loss": 0.4221758544445038,
52
- "eval_runtime": 180.8406,
53
- "eval_samples_per_second": 15.854,
54
- "eval_steps_per_second": 0.995,
55
  "step": 40
56
  },
57
  {
58
- "epoch": 0.25,
59
- "eval_loss": 0.40922558307647705,
60
- "eval_runtime": 181.1791,
61
- "eval_samples_per_second": 15.824,
62
- "eval_steps_per_second": 0.993,
63
  "step": 50
64
  },
65
  {
66
- "epoch": 0.3,
67
- "grad_norm": 0.27356910705566406,
68
- "learning_rate": 8.608268990399349e-05,
69
- "loss": 0.4132,
70
  "step": 60
71
  },
72
  {
73
- "epoch": 0.3,
74
- "eval_loss": 0.4156472384929657,
75
- "eval_runtime": 180.7116,
76
- "eval_samples_per_second": 15.865,
77
- "eval_steps_per_second": 0.996,
78
  "step": 60
79
  },
80
  {
81
- "epoch": 0.35,
82
- "eval_loss": 0.39848339557647705,
83
- "eval_runtime": 180.7856,
84
- "eval_samples_per_second": 15.859,
85
- "eval_steps_per_second": 0.996,
86
  "step": 70
87
  }
88
  ],
@@ -91,7 +91,7 @@
91
  "num_input_tokens_seen": 0,
92
  "num_train_epochs": 1,
93
  "save_steps": 10,
94
- "total_flos": 3.162393138828411e+17,
95
  "train_batch_size": 16,
96
  "trial_name": null,
97
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.3856147527694702,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-70",
4
+ "epoch": 0.693069306930693,
5
  "eval_steps": 10,
6
  "global_step": 70,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.1,
13
+ "eval_loss": 0.6981944441795349,
14
+ "eval_runtime": 94.6462,
15
+ "eval_samples_per_second": 30.292,
16
+ "eval_steps_per_second": 0.951,
17
  "step": 10
18
  },
19
  {
20
+ "epoch": 0.2,
21
+ "grad_norm": 0.2654201090335846,
22
+ "learning_rate": 0.0001961261695938319,
23
+ "loss": 1.0016,
24
  "step": 20
25
  },
26
  {
27
+ "epoch": 0.2,
28
+ "eval_loss": 0.467332661151886,
29
+ "eval_runtime": 94.5373,
30
+ "eval_samples_per_second": 30.327,
31
+ "eval_steps_per_second": 0.952,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.3,
36
+ "eval_loss": 0.434685617685318,
37
+ "eval_runtime": 94.5899,
38
+ "eval_samples_per_second": 30.31,
39
+ "eval_steps_per_second": 0.951,
40
  "step": 30
41
  },
42
  {
43
+ "epoch": 0.4,
44
+ "grad_norm": 1.3763986825942993,
45
+ "learning_rate": 0.0001559192903470747,
46
+ "loss": 0.4357,
47
  "step": 40
48
  },
49
  {
50
+ "epoch": 0.4,
51
+ "eval_loss": 0.4082697033882141,
52
+ "eval_runtime": 94.551,
53
+ "eval_samples_per_second": 30.322,
54
+ "eval_steps_per_second": 0.952,
55
  "step": 40
56
  },
57
  {
58
+ "epoch": 0.5,
59
+ "eval_loss": 0.40642818808555603,
60
+ "eval_runtime": 94.5831,
61
+ "eval_samples_per_second": 30.312,
62
+ "eval_steps_per_second": 0.952,
63
  "step": 50
64
  },
65
  {
66
+ "epoch": 0.59,
67
+ "grad_norm": 0.16300354897975922,
68
+ "learning_rate": 8.954715367323468e-05,
69
+ "loss": 0.4012,
70
  "step": 60
71
  },
72
  {
73
+ "epoch": 0.59,
74
+ "eval_loss": 0.39164847135543823,
75
+ "eval_runtime": 94.6074,
76
+ "eval_samples_per_second": 30.304,
77
+ "eval_steps_per_second": 0.951,
78
  "step": 60
79
  },
80
  {
81
+ "epoch": 0.69,
82
+ "eval_loss": 0.3856147527694702,
83
+ "eval_runtime": 94.5883,
84
+ "eval_samples_per_second": 30.31,
85
+ "eval_steps_per_second": 0.951,
86
  "step": 70
87
  }
88
  ],
 
91
  "num_input_tokens_seen": 0,
92
  "num_train_epochs": 1,
93
  "save_steps": 10,
94
+ "total_flos": 6.30628785351295e+17,
95
  "train_batch_size": 16,
96
  "trial_name": null,
97
  "trial_params": null
checkpoint-70/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
3
  size 5176
checkpoint-80/adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
23
- "k_proj",
24
  "v_proj",
25
- "o_proj"
 
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
 
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
 
23
  "v_proj",
24
+ "o_proj",
25
+ "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
checkpoint-80/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b527b969a7485a4a5babdabe6555b5c383c05ebc9f26e37428a33b13c8b5da9d
3
  size 436242776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24e21a859be9d207f149ffc0b46dfcc320ea1ee34732d41bf90a24c1b253b907
3
  size 436242776
checkpoint-80/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b1eec7cb54f027bab8ca6c3381b6ffff8c752180f8c5c51218b412ed94f8b847
3
  size 872568314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f70660925ff51a9f94e304f49828ef04a284f17d3a7bc460079ba9ce1a8e117e
3
  size 872568314
checkpoint-80/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0a801669d16f26e581b00baf98c1e7c4a03b79841796ecdf09044e710d489912
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:93f8281a861fa369d43b6beb0ae6c67fde2ae747fd1b83daa3e6a92279b33244
3
+ size 15024
checkpoint-80/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:832aea46110ff9ec827428f726a2265f68e7159fe5184ac4697ecf55087199d1
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:090bdbfa5ddd772617ed624bd84911da0c7bf8de602f4e422d57d8f9b2a59994
3
+ size 15024
checkpoint-80/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2b71e69a6ae8462bc71f9d5c4bc5cfca3ea5f220e53bbe18bfaf849b8d7997dd
3
  size 1000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0af4379a58cfae44f0dbe3cf08fcd1bc98068c05bfa6733ba98cd23f1ebaae09
3
  size 1000
checkpoint-80/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.3947867155075073,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-80",
4
- "epoch": 0.3965303593556382,
5
  "eval_steps": 10,
6
  "global_step": 80,
7
  "is_hyper_param_search": false,
@@ -9,95 +9,95 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.05,
13
- "eval_loss": 0.6367942690849304,
14
- "eval_runtime": 180.8957,
15
- "eval_samples_per_second": 15.849,
16
- "eval_steps_per_second": 0.995,
17
  "step": 10
18
  },
19
  {
20
- "epoch": 0.1,
21
- "grad_norm": 0.34346863627433777,
22
- "learning_rate": 0.00019510565162951537,
23
- "loss": 0.9774,
24
  "step": 20
25
  },
26
  {
27
- "epoch": 0.1,
28
- "eval_loss": 0.48089343309402466,
29
- "eval_runtime": 180.8882,
30
- "eval_samples_per_second": 15.85,
31
- "eval_steps_per_second": 0.995,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.15,
36
- "eval_loss": 0.44581571221351624,
37
- "eval_runtime": 180.8305,
38
- "eval_samples_per_second": 15.855,
39
- "eval_steps_per_second": 0.995,
40
  "step": 30
41
  },
42
  {
43
- "epoch": 0.2,
44
- "grad_norm": 0.22729116678237915,
45
- "learning_rate": 0.0001529919264233205,
46
- "loss": 0.4459,
47
  "step": 40
48
  },
49
  {
50
- "epoch": 0.2,
51
- "eval_loss": 0.4221758544445038,
52
- "eval_runtime": 180.8406,
53
- "eval_samples_per_second": 15.854,
54
- "eval_steps_per_second": 0.995,
55
  "step": 40
56
  },
57
  {
58
- "epoch": 0.25,
59
- "eval_loss": 0.40922558307647705,
60
- "eval_runtime": 181.1791,
61
- "eval_samples_per_second": 15.824,
62
- "eval_steps_per_second": 0.993,
63
  "step": 50
64
  },
65
  {
66
- "epoch": 0.3,
67
- "grad_norm": 0.27356910705566406,
68
- "learning_rate": 8.608268990399349e-05,
69
- "loss": 0.4132,
70
  "step": 60
71
  },
72
  {
73
- "epoch": 0.3,
74
- "eval_loss": 0.4156472384929657,
75
- "eval_runtime": 180.7116,
76
- "eval_samples_per_second": 15.865,
77
- "eval_steps_per_second": 0.996,
78
  "step": 60
79
  },
80
  {
81
- "epoch": 0.35,
82
- "eval_loss": 0.39848339557647705,
83
- "eval_runtime": 180.7856,
84
- "eval_samples_per_second": 15.859,
85
- "eval_steps_per_second": 0.996,
86
  "step": 70
87
  },
88
  {
89
- "epoch": 0.4,
90
- "grad_norm": 0.1912572681903839,
91
- "learning_rate": 2.5685517452260567e-05,
92
- "loss": 0.4004,
93
  "step": 80
94
  },
95
  {
96
- "epoch": 0.4,
97
- "eval_loss": 0.3947867155075073,
98
- "eval_runtime": 180.6659,
99
- "eval_samples_per_second": 15.869,
100
- "eval_steps_per_second": 0.996,
101
  "step": 80
102
  }
103
  ],
@@ -106,7 +106,7 @@
106
  "num_input_tokens_seen": 0,
107
  "num_train_epochs": 1,
108
  "save_steps": 10,
109
- "total_flos": 3.617727103190958e+17,
110
  "train_batch_size": 16,
111
  "trial_name": null,
112
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.38140159845352173,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-80",
4
+ "epoch": 0.7920792079207921,
5
  "eval_steps": 10,
6
  "global_step": 80,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.1,
13
+ "eval_loss": 0.6981944441795349,
14
+ "eval_runtime": 94.6462,
15
+ "eval_samples_per_second": 30.292,
16
+ "eval_steps_per_second": 0.951,
17
  "step": 10
18
  },
19
  {
20
+ "epoch": 0.2,
21
+ "grad_norm": 0.2654201090335846,
22
+ "learning_rate": 0.0001961261695938319,
23
+ "loss": 1.0016,
24
  "step": 20
25
  },
26
  {
27
+ "epoch": 0.2,
28
+ "eval_loss": 0.467332661151886,
29
+ "eval_runtime": 94.5373,
30
+ "eval_samples_per_second": 30.327,
31
+ "eval_steps_per_second": 0.952,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.3,
36
+ "eval_loss": 0.434685617685318,
37
+ "eval_runtime": 94.5899,
38
+ "eval_samples_per_second": 30.31,
39
+ "eval_steps_per_second": 0.951,
40
  "step": 30
41
  },
42
  {
43
+ "epoch": 0.4,
44
+ "grad_norm": 1.3763986825942993,
45
+ "learning_rate": 0.0001559192903470747,
46
+ "loss": 0.4357,
47
  "step": 40
48
  },
49
  {
50
+ "epoch": 0.4,
51
+ "eval_loss": 0.4082697033882141,
52
+ "eval_runtime": 94.551,
53
+ "eval_samples_per_second": 30.322,
54
+ "eval_steps_per_second": 0.952,
55
  "step": 40
56
  },
57
  {
58
+ "epoch": 0.5,
59
+ "eval_loss": 0.40642818808555603,
60
+ "eval_runtime": 94.5831,
61
+ "eval_samples_per_second": 30.312,
62
+ "eval_steps_per_second": 0.952,
63
  "step": 50
64
  },
65
  {
66
+ "epoch": 0.59,
67
+ "grad_norm": 0.16300354897975922,
68
+ "learning_rate": 8.954715367323468e-05,
69
+ "loss": 0.4012,
70
  "step": 60
71
  },
72
  {
73
+ "epoch": 0.59,
74
+ "eval_loss": 0.39164847135543823,
75
+ "eval_runtime": 94.6074,
76
+ "eval_samples_per_second": 30.304,
77
+ "eval_steps_per_second": 0.951,
78
  "step": 60
79
  },
80
  {
81
+ "epoch": 0.69,
82
+ "eval_loss": 0.3856147527694702,
83
+ "eval_runtime": 94.5883,
84
+ "eval_samples_per_second": 30.31,
85
+ "eval_steps_per_second": 0.951,
86
  "step": 70
87
  },
88
  {
89
+ "epoch": 0.79,
90
+ "grad_norm": 0.14875428378582,
91
+ "learning_rate": 2.8066019966134904e-05,
92
+ "loss": 0.3846,
93
  "step": 80
94
  },
95
  {
96
+ "epoch": 0.79,
97
+ "eval_loss": 0.38140159845352173,
98
+ "eval_runtime": 94.5138,
99
+ "eval_samples_per_second": 30.334,
100
+ "eval_steps_per_second": 0.952,
101
  "step": 80
102
  }
103
  ],
 
106
  "num_input_tokens_seen": 0,
107
  "num_train_epochs": 1,
108
  "save_steps": 10,
109
+ "total_flos": 7.216782437357978e+17,
110
  "train_batch_size": 16,
111
  "trial_name": null,
112
  "trial_params": null
checkpoint-80/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
3
  size 5176
checkpoint-90/adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
23
- "k_proj",
24
  "v_proj",
25
- "o_proj"
 
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
 
20
  "revision": null,
21
  "target_modules": [
22
  "q_proj",
 
23
  "v_proj",
24
+ "o_proj",
25
+ "k_proj"
26
  ],
27
  "task_type": "CAUSAL_LM",
28
  "use_rslora": false
checkpoint-90/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:656b072afb9540faadac0b0741a7a1048ad82d161f456b99cbe9cb27de8904b5
3
  size 436242776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0c30fc1ce7eac8619bae4324c083d54fc3ab754951ccedd52fac47403a20743
3
  size 436242776
checkpoint-90/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b434dc7b4c8b42e026285b4cfd5bba70c7c605b762fc1f40a2497755ac2cc407
3
  size 872568314
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:486aa4e54416a204b7db0aeed6777c24c2ff3852690250be9c7cbf8a71f41c44
3
  size 872568314
checkpoint-90/rng_state_0.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d1ca7c38f60ccd88483b3a63102bc43b67e4b014bcfc0d88a6a512edcbc12712
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a254ba0cdafb651978de9121c20a71b149b425dfe94a899e5728c492162cdab
3
+ size 15024
checkpoint-90/rng_state_1.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2d92ee0aa8cf2e1100b47e0467d2fb7b0bafc09e2325a5eebad03c7f3d262f88
3
- size 14512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d1a2d1d643c84a57098344dddf622cb085a1400d5ef0b7fc0334642388ea389
3
+ size 15024
checkpoint-90/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0658f8849eb58961b159a204d5c11e639b04f4416306a9ce692d0fa29467f5f4
3
  size 1000
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05a9a94c03ab13518f6155b6d7ecc8a918e6d6e0a99e88f4f661c51a3bc91db3
3
  size 1000
checkpoint-90/trainer_state.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
- "best_metric": 0.39267557859420776,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-90",
4
- "epoch": 0.44609665427509293,
5
  "eval_steps": 10,
6
  "global_step": 90,
7
  "is_hyper_param_search": false,
@@ -9,103 +9,103 @@
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.05,
13
- "eval_loss": 0.6367942690849304,
14
- "eval_runtime": 180.8957,
15
- "eval_samples_per_second": 15.849,
16
- "eval_steps_per_second": 0.995,
17
  "step": 10
18
  },
19
  {
20
- "epoch": 0.1,
21
- "grad_norm": 0.34346863627433777,
22
- "learning_rate": 0.00019510565162951537,
23
- "loss": 0.9774,
24
  "step": 20
25
  },
26
  {
27
- "epoch": 0.1,
28
- "eval_loss": 0.48089343309402466,
29
- "eval_runtime": 180.8882,
30
- "eval_samples_per_second": 15.85,
31
- "eval_steps_per_second": 0.995,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.15,
36
- "eval_loss": 0.44581571221351624,
37
- "eval_runtime": 180.8305,
38
- "eval_samples_per_second": 15.855,
39
- "eval_steps_per_second": 0.995,
40
  "step": 30
41
  },
42
  {
43
- "epoch": 0.2,
44
- "grad_norm": 0.22729116678237915,
45
- "learning_rate": 0.0001529919264233205,
46
- "loss": 0.4459,
47
  "step": 40
48
  },
49
  {
50
- "epoch": 0.2,
51
- "eval_loss": 0.4221758544445038,
52
- "eval_runtime": 180.8406,
53
- "eval_samples_per_second": 15.854,
54
- "eval_steps_per_second": 0.995,
55
  "step": 40
56
  },
57
  {
58
- "epoch": 0.25,
59
- "eval_loss": 0.40922558307647705,
60
- "eval_runtime": 181.1791,
61
- "eval_samples_per_second": 15.824,
62
- "eval_steps_per_second": 0.993,
63
  "step": 50
64
  },
65
  {
66
- "epoch": 0.3,
67
- "grad_norm": 0.27356910705566406,
68
- "learning_rate": 8.608268990399349e-05,
69
- "loss": 0.4132,
70
  "step": 60
71
  },
72
  {
73
- "epoch": 0.3,
74
- "eval_loss": 0.4156472384929657,
75
- "eval_runtime": 180.7116,
76
- "eval_samples_per_second": 15.865,
77
- "eval_steps_per_second": 0.996,
78
  "step": 60
79
  },
80
  {
81
- "epoch": 0.35,
82
- "eval_loss": 0.39848339557647705,
83
- "eval_runtime": 180.7856,
84
- "eval_samples_per_second": 15.859,
85
- "eval_steps_per_second": 0.996,
86
  "step": 70
87
  },
88
  {
89
- "epoch": 0.4,
90
- "grad_norm": 0.1912572681903839,
91
- "learning_rate": 2.5685517452260567e-05,
92
- "loss": 0.4004,
93
  "step": 80
94
  },
95
  {
96
- "epoch": 0.4,
97
- "eval_loss": 0.3947867155075073,
98
- "eval_runtime": 180.6659,
99
- "eval_samples_per_second": 15.869,
100
- "eval_steps_per_second": 0.996,
101
  "step": 80
102
  },
103
  {
104
- "epoch": 0.45,
105
- "eval_loss": 0.39267557859420776,
106
- "eval_runtime": 180.6936,
107
- "eval_samples_per_second": 15.867,
108
- "eval_steps_per_second": 0.996,
109
  "step": 90
110
  }
111
  ],
@@ -114,7 +114,7 @@
114
  "num_input_tokens_seen": 0,
115
  "num_train_epochs": 1,
116
  "save_steps": 10,
117
- "total_flos": 4.080622701075497e+17,
118
  "train_batch_size": 16,
119
  "trial_name": null,
120
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.3798506557941437,
3
  "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-90",
4
+ "epoch": 0.8910891089108911,
5
  "eval_steps": 10,
6
  "global_step": 90,
7
  "is_hyper_param_search": false,
 
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.1,
13
+ "eval_loss": 0.6981944441795349,
14
+ "eval_runtime": 94.6462,
15
+ "eval_samples_per_second": 30.292,
16
+ "eval_steps_per_second": 0.951,
17
  "step": 10
18
  },
19
  {
20
+ "epoch": 0.2,
21
+ "grad_norm": 0.2654201090335846,
22
+ "learning_rate": 0.0001961261695938319,
23
+ "loss": 1.0016,
24
  "step": 20
25
  },
26
  {
27
+ "epoch": 0.2,
28
+ "eval_loss": 0.467332661151886,
29
+ "eval_runtime": 94.5373,
30
+ "eval_samples_per_second": 30.327,
31
+ "eval_steps_per_second": 0.952,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.3,
36
+ "eval_loss": 0.434685617685318,
37
+ "eval_runtime": 94.5899,
38
+ "eval_samples_per_second": 30.31,
39
+ "eval_steps_per_second": 0.951,
40
  "step": 30
41
  },
42
  {
43
+ "epoch": 0.4,
44
+ "grad_norm": 1.3763986825942993,
45
+ "learning_rate": 0.0001559192903470747,
46
+ "loss": 0.4357,
47
  "step": 40
48
  },
49
  {
50
+ "epoch": 0.4,
51
+ "eval_loss": 0.4082697033882141,
52
+ "eval_runtime": 94.551,
53
+ "eval_samples_per_second": 30.322,
54
+ "eval_steps_per_second": 0.952,
55
  "step": 40
56
  },
57
  {
58
+ "epoch": 0.5,
59
+ "eval_loss": 0.40642818808555603,
60
+ "eval_runtime": 94.5831,
61
+ "eval_samples_per_second": 30.312,
62
+ "eval_steps_per_second": 0.952,
63
  "step": 50
64
  },
65
  {
66
+ "epoch": 0.59,
67
+ "grad_norm": 0.16300354897975922,
68
+ "learning_rate": 8.954715367323468e-05,
69
+ "loss": 0.4012,
70
  "step": 60
71
  },
72
  {
73
+ "epoch": 0.59,
74
+ "eval_loss": 0.39164847135543823,
75
+ "eval_runtime": 94.6074,
76
+ "eval_samples_per_second": 30.304,
77
+ "eval_steps_per_second": 0.951,
78
  "step": 60
79
  },
80
  {
81
+ "epoch": 0.69,
82
+ "eval_loss": 0.3856147527694702,
83
+ "eval_runtime": 94.5883,
84
+ "eval_samples_per_second": 30.31,
85
+ "eval_steps_per_second": 0.951,
86
  "step": 70
87
  },
88
  {
89
+ "epoch": 0.79,
90
+ "grad_norm": 0.14875428378582,
91
+ "learning_rate": 2.8066019966134904e-05,
92
+ "loss": 0.3846,
93
  "step": 80
94
  },
95
  {
96
+ "epoch": 0.79,
97
+ "eval_loss": 0.38140159845352173,
98
+ "eval_runtime": 94.5138,
99
+ "eval_samples_per_second": 30.334,
100
+ "eval_steps_per_second": 0.952,
101
  "step": 80
102
  },
103
  {
104
+ "epoch": 0.89,
105
+ "eval_loss": 0.3798506557941437,
106
+ "eval_runtime": 94.6456,
107
+ "eval_samples_per_second": 30.292,
108
+ "eval_steps_per_second": 0.951,
109
  "step": 90
110
  }
111
  ],
 
114
  "num_input_tokens_seen": 0,
115
  "num_train_epochs": 1,
116
  "save_steps": 10,
117
+ "total_flos": 8.129633583858975e+17,
118
  "train_batch_size": 16,
119
  "trial_name": null,
120
  "trial_params": null
checkpoint-90/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
3
  size 5176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
3
  size 5176