clejordan commited on
Commit
2e26d04
·
verified ·
1 Parent(s): 2ca2e8f

Initial commit: MNLP M3 quantized model

Browse files
checkpoint-11502/config.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen3",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 8,
20
+ "quantization_config": {
21
+ "config_groups": {
22
+ "group_0": {
23
+ "input_activations": {
24
+ "actorder": null,
25
+ "block_structure": null,
26
+ "dynamic": true,
27
+ "group_size": null,
28
+ "num_bits": 8,
29
+ "observer": null,
30
+ "observer_kwargs": {},
31
+ "strategy": "token",
32
+ "symmetric": true,
33
+ "type": "int"
34
+ },
35
+ "output_activations": null,
36
+ "targets": [
37
+ "Linear"
38
+ ],
39
+ "weights": {
40
+ "actorder": null,
41
+ "block_structure": null,
42
+ "dynamic": false,
43
+ "group_size": null,
44
+ "num_bits": 8,
45
+ "observer": "minmax",
46
+ "observer_kwargs": {},
47
+ "strategy": "channel",
48
+ "symmetric": true,
49
+ "type": "int"
50
+ }
51
+ }
52
+ },
53
+ "format": "int-quantized",
54
+ "global_compression_ratio": null,
55
+ "ignore": [
56
+ "lm_head"
57
+ ],
58
+ "kv_cache_scheme": null,
59
+ "quant_method": "compressed-tensors",
60
+ "quantization_status": "compressed",
61
+ "sparsity_config": {}
62
+ },
63
+ "rms_norm_eps": 1e-06,
64
+ "rope_scaling": null,
65
+ "rope_theta": 1000000,
66
+ "sliding_window": null,
67
+ "tie_word_embeddings": true,
68
+ "torch_dtype": "float16",
69
+ "transformers_version": "4.51.3",
70
+ "use_cache": true,
71
+ "use_sliding_window": false,
72
+ "vocab_size": 151936
73
+ }
checkpoint-11502/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.51.3"
6
+ }
checkpoint-11502/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb3d1bd95c53800ef1cb04bf0b4dca4678ae38f01058094213011b8ab1df2db9
3
+ size 1192845752
checkpoint-11502/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:79090d2c55e4e99e4ae06262c00e9f2ce1bf8182109e3a269113fdeacc963eb6
3
- size 40699434
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ea4ec19e10fcf879489137083c39125bcd2f58e8de40103fa46d5219e46c904
3
+ size 622337116
checkpoint-11502/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28736a80102fab43bcf5c1d1e9dc6a88947e46af459cca97238dc9bbbdf5bc4b
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d138cfe3a4adf21f048848ee35837c9a757a0a3616ff7adbb45b69aac247435
3
  size 14244
checkpoint-11502/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7bf0678c8431c32ee488ef33c659bd02feb2088d7e6423f432b09a05eab0958d
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0ad21997940ab64708a0aee1c86735a1db628310427e22f01fbc6ae661643ac
3
  size 1064
checkpoint-11502/trainer_state.json CHANGED
@@ -11,179 +11,179 @@
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
- "grad_norm": 0.2196940779685974,
15
- "learning_rate": 0.00019426186750130412,
16
- "loss": 0.2606,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
- "grad_norm": 0.2677164673805237,
22
- "learning_rate": 0.000188465774068278,
23
- "loss": 0.1763,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
- "grad_norm": 0.21263259649276733,
29
- "learning_rate": 0.00018266968063525185,
30
- "loss": 0.1714,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
- "grad_norm": 0.275687575340271,
36
- "learning_rate": 0.0001768735872022257,
37
- "loss": 0.1693,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
- "grad_norm": 0.21894076466560364,
43
- "learning_rate": 0.00017107749376919957,
44
- "loss": 0.1719,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
- "grad_norm": 0.23001301288604736,
50
- "learning_rate": 0.00016528140033617342,
51
- "loss": 0.1691,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
- "grad_norm": 0.23900070786476135,
57
- "learning_rate": 0.0001594853069031473,
58
- "loss": 0.167,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
- "grad_norm": 0.2820037603378296,
64
- "learning_rate": 0.00015368921347012115,
65
- "loss": 0.1678,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
- "grad_norm": 0.3206230700016022,
71
- "learning_rate": 0.00014789312003709503,
72
- "loss": 0.1675,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
- "grad_norm": 0.20400723814964294,
78
- "learning_rate": 0.00014209702660406887,
79
- "loss": 0.1658,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
- "grad_norm": 0.31783542037010193,
85
- "learning_rate": 0.00013630093317104272,
86
- "loss": 0.1643,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
- "eval_loss": 0.16573764383792877,
92
- "eval_runtime": 80.6587,
93
- "eval_samples_per_second": 31.689,
94
- "eval_steps_per_second": 7.922,
95
  "step": 5751
96
  },
97
  {
98
  "epoch": 1.0432968179447053,
99
- "grad_norm": 0.3012489676475525,
100
- "learning_rate": 0.0001305048397380166,
101
- "loss": 0.1588,
102
  "step": 6000
103
  },
104
  {
105
  "epoch": 1.1302382194400973,
106
- "grad_norm": 0.2865125238895416,
107
- "learning_rate": 0.00012470874630499042,
108
- "loss": 0.1474,
109
  "step": 6500
110
  },
111
  {
112
  "epoch": 1.2171796209354895,
113
- "grad_norm": 0.2503669857978821,
114
- "learning_rate": 0.0001189126528719643,
115
- "loss": 0.1489,
116
  "step": 7000
117
  },
118
  {
119
  "epoch": 1.3041210224308815,
120
- "grad_norm": 0.2573595643043518,
121
- "learning_rate": 0.00011311655943893815,
122
- "loss": 0.1488,
123
  "step": 7500
124
  },
125
  {
126
  "epoch": 1.3910624239262737,
127
- "grad_norm": 0.2882978916168213,
128
- "learning_rate": 0.00010732046600591201,
129
- "loss": 0.1465,
130
  "step": 8000
131
  },
132
  {
133
  "epoch": 1.4780038254216659,
134
- "grad_norm": 0.33109399676322937,
135
- "learning_rate": 0.00010152437257288587,
136
- "loss": 0.1503,
137
  "step": 8500
138
  },
139
  {
140
  "epoch": 1.5649452269170578,
141
- "grad_norm": 0.5740349888801575,
142
- "learning_rate": 9.572827913985975e-05,
143
- "loss": 0.1493,
144
  "step": 9000
145
  },
146
  {
147
  "epoch": 1.65188662841245,
148
- "grad_norm": 0.298555463552475,
149
- "learning_rate": 8.99321857068336e-05,
150
- "loss": 0.1483,
151
  "step": 9500
152
  },
153
  {
154
  "epoch": 1.7388280299078422,
155
- "grad_norm": 0.333500474691391,
156
- "learning_rate": 8.413609227380746e-05,
157
- "loss": 0.1469,
158
  "step": 10000
159
  },
160
  {
161
  "epoch": 1.8257694314032342,
162
- "grad_norm": 0.30120372772216797,
163
- "learning_rate": 7.833999884078131e-05,
164
- "loss": 0.1493,
165
  "step": 10500
166
  },
167
  {
168
  "epoch": 1.9127108328986262,
169
- "grad_norm": 0.28990960121154785,
170
- "learning_rate": 7.254390540775517e-05,
171
- "loss": 0.1458,
172
  "step": 11000
173
  },
174
  {
175
  "epoch": 1.9996522343940184,
176
- "grad_norm": 0.5824082493782043,
177
- "learning_rate": 6.674781197472903e-05,
178
- "loss": 0.1504,
179
  "step": 11500
180
  },
181
  {
182
  "epoch": 2.0,
183
- "eval_loss": 0.1634208858013153,
184
- "eval_runtime": 80.8075,
185
- "eval_samples_per_second": 31.631,
186
- "eval_steps_per_second": 7.908,
187
  "step": 11502
188
  }
189
  ],
@@ -204,7 +204,7 @@
204
  "attributes": {}
205
  }
206
  },
207
- "total_flos": 6.29646476401705e+16,
208
  "train_batch_size": 4,
209
  "trial_name": null,
210
  "trial_params": null
 
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
+ "grad_norm": NaN,
15
+ "learning_rate": 0.00019421549875383993,
16
+ "loss": 4.2823,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
+ "grad_norm": NaN,
22
+ "learning_rate": 0.0001884194053208138,
23
+ "loss": 0.0,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
+ "grad_norm": NaN,
29
+ "learning_rate": 0.00018262331188778763,
30
+ "loss": 0.0,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
+ "grad_norm": NaN,
36
+ "learning_rate": 0.0001768272184547615,
37
+ "loss": 0.0,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
+ "grad_norm": NaN,
43
+ "learning_rate": 0.00017103112502173535,
44
+ "loss": 0.0,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
+ "grad_norm": NaN,
50
+ "learning_rate": 0.0001652350315887092,
51
+ "loss": 0.0,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
+ "grad_norm": NaN,
57
+ "learning_rate": 0.00015943893815568308,
58
+ "loss": 0.0,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
+ "grad_norm": NaN,
64
+ "learning_rate": 0.00015364284472265693,
65
+ "loss": 0.0,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
+ "grad_norm": NaN,
71
+ "learning_rate": 0.0001478467512896308,
72
+ "loss": 0.0,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
+ "grad_norm": NaN,
78
+ "learning_rate": 0.00014205065785660465,
79
+ "loss": 0.0,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
+ "grad_norm": NaN,
85
+ "learning_rate": 0.0001362545644235785,
86
+ "loss": 0.0,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
+ "eval_loss": NaN,
92
+ "eval_runtime": 152.5081,
93
+ "eval_samples_per_second": 16.76,
94
+ "eval_steps_per_second": 4.19,
95
  "step": 5751
96
  },
97
  {
98
  "epoch": 1.0432968179447053,
99
+ "grad_norm": NaN,
100
+ "learning_rate": 0.00013045847099055238,
101
+ "loss": 0.0,
102
  "step": 6000
103
  },
104
  {
105
  "epoch": 1.1302382194400973,
106
+ "grad_norm": NaN,
107
+ "learning_rate": 0.00012466237755752623,
108
+ "loss": 0.0,
109
  "step": 6500
110
  },
111
  {
112
  "epoch": 1.2171796209354895,
113
+ "grad_norm": NaN,
114
+ "learning_rate": 0.00011886628412450009,
115
+ "loss": 0.0,
116
  "step": 7000
117
  },
118
  {
119
  "epoch": 1.3041210224308815,
120
+ "grad_norm": NaN,
121
+ "learning_rate": 0.00011307019069147395,
122
+ "loss": 0.0,
123
  "step": 7500
124
  },
125
  {
126
  "epoch": 1.3910624239262737,
127
+ "grad_norm": NaN,
128
+ "learning_rate": 0.00010727409725844781,
129
+ "loss": 0.0,
130
  "step": 8000
131
  },
132
  {
133
  "epoch": 1.4780038254216659,
134
+ "grad_norm": NaN,
135
+ "learning_rate": 0.00010147800382542168,
136
+ "loss": 0.0,
137
  "step": 8500
138
  },
139
  {
140
  "epoch": 1.5649452269170578,
141
+ "grad_norm": NaN,
142
+ "learning_rate": 9.568191039239554e-05,
143
+ "loss": 0.0,
144
  "step": 9000
145
  },
146
  {
147
  "epoch": 1.65188662841245,
148
+ "grad_norm": NaN,
149
+ "learning_rate": 8.988581695936939e-05,
150
+ "loss": 0.0,
151
  "step": 9500
152
  },
153
  {
154
  "epoch": 1.7388280299078422,
155
+ "grad_norm": NaN,
156
+ "learning_rate": 8.408972352634325e-05,
157
+ "loss": 0.0,
158
  "step": 10000
159
  },
160
  {
161
  "epoch": 1.8257694314032342,
162
+ "grad_norm": NaN,
163
+ "learning_rate": 7.82936300933171e-05,
164
+ "loss": 0.0,
165
  "step": 10500
166
  },
167
  {
168
  "epoch": 1.9127108328986262,
169
+ "grad_norm": NaN,
170
+ "learning_rate": 7.249753666029096e-05,
171
+ "loss": 0.0,
172
  "step": 11000
173
  },
174
  {
175
  "epoch": 1.9996522343940184,
176
+ "grad_norm": NaN,
177
+ "learning_rate": 6.670144322726482e-05,
178
+ "loss": 0.0,
179
  "step": 11500
180
  },
181
  {
182
  "epoch": 2.0,
183
+ "eval_loss": NaN,
184
+ "eval_runtime": 152.3781,
185
+ "eval_samples_per_second": 16.774,
186
+ "eval_steps_per_second": 4.194,
187
  "step": 11502
188
  }
189
  ],
 
204
  "attributes": {}
205
  }
206
  },
207
+ "total_flos": 6.230008255217664e+16,
208
  "train_batch_size": 4,
209
  "trial_name": null,
210
  "trial_params": null
checkpoint-11502/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfb590c3405bb321e9c77b1f2243981f8f41d35f037852c7fac86c1191f3b93d
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99fccf4927862cc4fd8283f3958f1ef8528c86594c8ff39225b274adde4416df
3
  size 5240
checkpoint-17253/config.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen3",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 8,
20
+ "quantization_config": {
21
+ "config_groups": {
22
+ "group_0": {
23
+ "input_activations": {
24
+ "actorder": null,
25
+ "block_structure": null,
26
+ "dynamic": true,
27
+ "group_size": null,
28
+ "num_bits": 8,
29
+ "observer": null,
30
+ "observer_kwargs": {},
31
+ "strategy": "token",
32
+ "symmetric": true,
33
+ "type": "int"
34
+ },
35
+ "output_activations": null,
36
+ "targets": [
37
+ "Linear"
38
+ ],
39
+ "weights": {
40
+ "actorder": null,
41
+ "block_structure": null,
42
+ "dynamic": false,
43
+ "group_size": null,
44
+ "num_bits": 8,
45
+ "observer": "minmax",
46
+ "observer_kwargs": {},
47
+ "strategy": "channel",
48
+ "symmetric": true,
49
+ "type": "int"
50
+ }
51
+ }
52
+ },
53
+ "format": "int-quantized",
54
+ "global_compression_ratio": null,
55
+ "ignore": [
56
+ "lm_head"
57
+ ],
58
+ "kv_cache_scheme": null,
59
+ "quant_method": "compressed-tensors",
60
+ "quantization_status": "compressed",
61
+ "sparsity_config": {}
62
+ },
63
+ "rms_norm_eps": 1e-06,
64
+ "rope_scaling": null,
65
+ "rope_theta": 1000000,
66
+ "sliding_window": null,
67
+ "tie_word_embeddings": true,
68
+ "torch_dtype": "float16",
69
+ "transformers_version": "4.51.3",
70
+ "use_cache": true,
71
+ "use_sliding_window": false,
72
+ "vocab_size": 151936
73
+ }
checkpoint-17253/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.51.3"
6
+ }
checkpoint-17253/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb3d1bd95c53800ef1cb04bf0b4dca4678ae38f01058094213011b8ab1df2db9
3
+ size 1192845752
checkpoint-17253/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ad9c11ab10a9a068eb7be4ba5d55d24764197c51d4356aa0299c1c02a573e367
3
- size 40699434
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2284b8fb6f020a6f67ebde200adfab3395d52ae16e59e099cedfaf0891362079
3
+ size 622337116
checkpoint-17253/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:82169739e50e874f2f26da7d4d29a03b185e25b3179c2c76acfb6c1d9adb1a97
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c062f7f375beded48b5337f5a3f3a5cb38807fa3e85dbf3e294c0ab6b627bfc2
3
  size 14244
checkpoint-17253/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:400d96f8001902fcdfb4263517a92d19497f1ed5d512b07f5da1ea66c847616f
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69d2cf8d68dbd5ab076c36e56ef45fa3ee927d877ac4ed725d8b92e0458ccf6c
3
  size 1064
checkpoint-17253/trainer_state.json CHANGED
@@ -11,264 +11,264 @@
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
- "grad_norm": 0.2196940779685974,
15
- "learning_rate": 0.00019426186750130412,
16
- "loss": 0.2606,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
- "grad_norm": 0.2677164673805237,
22
- "learning_rate": 0.000188465774068278,
23
- "loss": 0.1763,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
- "grad_norm": 0.21263259649276733,
29
- "learning_rate": 0.00018266968063525185,
30
- "loss": 0.1714,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
- "grad_norm": 0.275687575340271,
36
- "learning_rate": 0.0001768735872022257,
37
- "loss": 0.1693,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
- "grad_norm": 0.21894076466560364,
43
- "learning_rate": 0.00017107749376919957,
44
- "loss": 0.1719,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
- "grad_norm": 0.23001301288604736,
50
- "learning_rate": 0.00016528140033617342,
51
- "loss": 0.1691,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
- "grad_norm": 0.23900070786476135,
57
- "learning_rate": 0.0001594853069031473,
58
- "loss": 0.167,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
- "grad_norm": 0.2820037603378296,
64
- "learning_rate": 0.00015368921347012115,
65
- "loss": 0.1678,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
- "grad_norm": 0.3206230700016022,
71
- "learning_rate": 0.00014789312003709503,
72
- "loss": 0.1675,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
- "grad_norm": 0.20400723814964294,
78
- "learning_rate": 0.00014209702660406887,
79
- "loss": 0.1658,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
- "grad_norm": 0.31783542037010193,
85
- "learning_rate": 0.00013630093317104272,
86
- "loss": 0.1643,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
- "eval_loss": 0.16573764383792877,
92
- "eval_runtime": 80.6587,
93
- "eval_samples_per_second": 31.689,
94
- "eval_steps_per_second": 7.922,
95
  "step": 5751
96
  },
97
  {
98
  "epoch": 1.0432968179447053,
99
- "grad_norm": 0.3012489676475525,
100
- "learning_rate": 0.0001305048397380166,
101
- "loss": 0.1588,
102
  "step": 6000
103
  },
104
  {
105
  "epoch": 1.1302382194400973,
106
- "grad_norm": 0.2865125238895416,
107
- "learning_rate": 0.00012470874630499042,
108
- "loss": 0.1474,
109
  "step": 6500
110
  },
111
  {
112
  "epoch": 1.2171796209354895,
113
- "grad_norm": 0.2503669857978821,
114
- "learning_rate": 0.0001189126528719643,
115
- "loss": 0.1489,
116
  "step": 7000
117
  },
118
  {
119
  "epoch": 1.3041210224308815,
120
- "grad_norm": 0.2573595643043518,
121
- "learning_rate": 0.00011311655943893815,
122
- "loss": 0.1488,
123
  "step": 7500
124
  },
125
  {
126
  "epoch": 1.3910624239262737,
127
- "grad_norm": 0.2882978916168213,
128
- "learning_rate": 0.00010732046600591201,
129
- "loss": 0.1465,
130
  "step": 8000
131
  },
132
  {
133
  "epoch": 1.4780038254216659,
134
- "grad_norm": 0.33109399676322937,
135
- "learning_rate": 0.00010152437257288587,
136
- "loss": 0.1503,
137
  "step": 8500
138
  },
139
  {
140
  "epoch": 1.5649452269170578,
141
- "grad_norm": 0.5740349888801575,
142
- "learning_rate": 9.572827913985975e-05,
143
- "loss": 0.1493,
144
  "step": 9000
145
  },
146
  {
147
  "epoch": 1.65188662841245,
148
- "grad_norm": 0.298555463552475,
149
- "learning_rate": 8.99321857068336e-05,
150
- "loss": 0.1483,
151
  "step": 9500
152
  },
153
  {
154
  "epoch": 1.7388280299078422,
155
- "grad_norm": 0.333500474691391,
156
- "learning_rate": 8.413609227380746e-05,
157
- "loss": 0.1469,
158
  "step": 10000
159
  },
160
  {
161
  "epoch": 1.8257694314032342,
162
- "grad_norm": 0.30120372772216797,
163
- "learning_rate": 7.833999884078131e-05,
164
- "loss": 0.1493,
165
  "step": 10500
166
  },
167
  {
168
  "epoch": 1.9127108328986262,
169
- "grad_norm": 0.28990960121154785,
170
- "learning_rate": 7.254390540775517e-05,
171
- "loss": 0.1458,
172
  "step": 11000
173
  },
174
  {
175
  "epoch": 1.9996522343940184,
176
- "grad_norm": 0.5824082493782043,
177
- "learning_rate": 6.674781197472903e-05,
178
- "loss": 0.1504,
179
  "step": 11500
180
  },
181
  {
182
  "epoch": 2.0,
183
- "eval_loss": 0.1634208858013153,
184
- "eval_runtime": 80.8075,
185
- "eval_samples_per_second": 31.631,
186
- "eval_steps_per_second": 7.908,
187
  "step": 11502
188
  },
189
  {
190
  "epoch": 2.0865936358894106,
191
- "grad_norm": 0.4085701107978821,
192
- "learning_rate": 6.0951718541702897e-05,
193
- "loss": 0.1307,
194
  "step": 12000
195
  },
196
  {
197
  "epoch": 2.1735350373848026,
198
- "grad_norm": 0.32403185963630676,
199
- "learning_rate": 5.516721729554281e-05,
200
- "loss": 0.1283,
201
  "step": 12500
202
  },
203
  {
204
  "epoch": 2.2604764388801946,
205
- "grad_norm": 0.5084943175315857,
206
- "learning_rate": 4.9371123862516664e-05,
207
- "loss": 0.1298,
208
  "step": 13000
209
  },
210
  {
211
  "epoch": 2.347417840375587,
212
- "grad_norm": 0.31282761693000793,
213
- "learning_rate": 4.357503042949053e-05,
214
- "loss": 0.1292,
215
  "step": 13500
216
  },
217
  {
218
  "epoch": 2.434359241870979,
219
- "grad_norm": 0.28461700677871704,
220
- "learning_rate": 3.777893699646439e-05,
221
- "loss": 0.1299,
222
  "step": 14000
223
  },
224
  {
225
  "epoch": 2.521300643366371,
226
- "grad_norm": 0.4293518364429474,
227
- "learning_rate": 3.1982843563438245e-05,
228
- "loss": 0.1312,
229
  "step": 14500
230
  },
231
  {
232
  "epoch": 2.608242044861763,
233
- "grad_norm": 0.27910229563713074,
234
- "learning_rate": 2.6198342317278157e-05,
235
- "loss": 0.1286,
236
  "step": 15000
237
  },
238
  {
239
  "epoch": 2.6951834463571553,
240
- "grad_norm": 0.3739106059074402,
241
- "learning_rate": 2.0402248884252016e-05,
242
- "loss": 0.1313,
243
  "step": 15500
244
  },
245
  {
246
  "epoch": 2.7821248478525473,
247
- "grad_norm": 0.38641801476478577,
248
- "learning_rate": 1.4606155451225875e-05,
249
- "loss": 0.1299,
250
  "step": 16000
251
  },
252
  {
253
  "epoch": 2.8690662493479397,
254
- "grad_norm": 0.25568658113479614,
255
- "learning_rate": 8.810062018199734e-06,
256
- "loss": 0.1294,
257
  "step": 16500
258
  },
259
  {
260
  "epoch": 2.9560076508433317,
261
- "grad_norm": 0.3838019073009491,
262
- "learning_rate": 3.013968585173593e-06,
263
- "loss": 0.1304,
264
  "step": 17000
265
  },
266
  {
267
  "epoch": 3.0,
268
- "eval_loss": 0.16723191738128662,
269
- "eval_runtime": 80.6877,
270
- "eval_samples_per_second": 31.678,
271
- "eval_steps_per_second": 7.919,
272
  "step": 17253
273
  }
274
  ],
@@ -289,7 +289,7 @@
289
  "attributes": {}
290
  }
291
  },
292
- "total_flos": 9.444697146025574e+16,
293
  "train_batch_size": 4,
294
  "trial_name": null,
295
  "trial_params": null
 
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
+ "grad_norm": NaN,
15
+ "learning_rate": 0.00019421549875383993,
16
+ "loss": 4.2823,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
+ "grad_norm": NaN,
22
+ "learning_rate": 0.0001884194053208138,
23
+ "loss": 0.0,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
+ "grad_norm": NaN,
29
+ "learning_rate": 0.00018262331188778763,
30
+ "loss": 0.0,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
+ "grad_norm": NaN,
36
+ "learning_rate": 0.0001768272184547615,
37
+ "loss": 0.0,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
+ "grad_norm": NaN,
43
+ "learning_rate": 0.00017103112502173535,
44
+ "loss": 0.0,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
+ "grad_norm": NaN,
50
+ "learning_rate": 0.0001652350315887092,
51
+ "loss": 0.0,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
+ "grad_norm": NaN,
57
+ "learning_rate": 0.00015943893815568308,
58
+ "loss": 0.0,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
+ "grad_norm": NaN,
64
+ "learning_rate": 0.00015364284472265693,
65
+ "loss": 0.0,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
+ "grad_norm": NaN,
71
+ "learning_rate": 0.0001478467512896308,
72
+ "loss": 0.0,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
+ "grad_norm": NaN,
78
+ "learning_rate": 0.00014205065785660465,
79
+ "loss": 0.0,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
+ "grad_norm": NaN,
85
+ "learning_rate": 0.0001362545644235785,
86
+ "loss": 0.0,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
+ "eval_loss": NaN,
92
+ "eval_runtime": 152.5081,
93
+ "eval_samples_per_second": 16.76,
94
+ "eval_steps_per_second": 4.19,
95
  "step": 5751
96
  },
97
  {
98
  "epoch": 1.0432968179447053,
99
+ "grad_norm": NaN,
100
+ "learning_rate": 0.00013045847099055238,
101
+ "loss": 0.0,
102
  "step": 6000
103
  },
104
  {
105
  "epoch": 1.1302382194400973,
106
+ "grad_norm": NaN,
107
+ "learning_rate": 0.00012466237755752623,
108
+ "loss": 0.0,
109
  "step": 6500
110
  },
111
  {
112
  "epoch": 1.2171796209354895,
113
+ "grad_norm": NaN,
114
+ "learning_rate": 0.00011886628412450009,
115
+ "loss": 0.0,
116
  "step": 7000
117
  },
118
  {
119
  "epoch": 1.3041210224308815,
120
+ "grad_norm": NaN,
121
+ "learning_rate": 0.00011307019069147395,
122
+ "loss": 0.0,
123
  "step": 7500
124
  },
125
  {
126
  "epoch": 1.3910624239262737,
127
+ "grad_norm": NaN,
128
+ "learning_rate": 0.00010727409725844781,
129
+ "loss": 0.0,
130
  "step": 8000
131
  },
132
  {
133
  "epoch": 1.4780038254216659,
134
+ "grad_norm": NaN,
135
+ "learning_rate": 0.00010147800382542168,
136
+ "loss": 0.0,
137
  "step": 8500
138
  },
139
  {
140
  "epoch": 1.5649452269170578,
141
+ "grad_norm": NaN,
142
+ "learning_rate": 9.568191039239554e-05,
143
+ "loss": 0.0,
144
  "step": 9000
145
  },
146
  {
147
  "epoch": 1.65188662841245,
148
+ "grad_norm": NaN,
149
+ "learning_rate": 8.988581695936939e-05,
150
+ "loss": 0.0,
151
  "step": 9500
152
  },
153
  {
154
  "epoch": 1.7388280299078422,
155
+ "grad_norm": NaN,
156
+ "learning_rate": 8.408972352634325e-05,
157
+ "loss": 0.0,
158
  "step": 10000
159
  },
160
  {
161
  "epoch": 1.8257694314032342,
162
+ "grad_norm": NaN,
163
+ "learning_rate": 7.82936300933171e-05,
164
+ "loss": 0.0,
165
  "step": 10500
166
  },
167
  {
168
  "epoch": 1.9127108328986262,
169
+ "grad_norm": NaN,
170
+ "learning_rate": 7.249753666029096e-05,
171
+ "loss": 0.0,
172
  "step": 11000
173
  },
174
  {
175
  "epoch": 1.9996522343940184,
176
+ "grad_norm": NaN,
177
+ "learning_rate": 6.670144322726482e-05,
178
+ "loss": 0.0,
179
  "step": 11500
180
  },
181
  {
182
  "epoch": 2.0,
183
+ "eval_loss": NaN,
184
+ "eval_runtime": 152.3781,
185
+ "eval_samples_per_second": 16.774,
186
+ "eval_steps_per_second": 4.194,
187
  "step": 11502
188
  },
189
  {
190
  "epoch": 2.0865936358894106,
191
+ "grad_norm": NaN,
192
+ "learning_rate": 6.0905349794238687e-05,
193
+ "loss": 0.0,
194
  "step": 12000
195
  },
196
  {
197
  "epoch": 2.1735350373848026,
198
+ "grad_norm": NaN,
199
+ "learning_rate": 5.510925636121255e-05,
200
+ "loss": 0.0,
201
  "step": 12500
202
  },
203
  {
204
  "epoch": 2.2604764388801946,
205
+ "grad_norm": NaN,
206
+ "learning_rate": 4.9313162928186405e-05,
207
+ "loss": 0.0,
208
  "step": 13000
209
  },
210
  {
211
  "epoch": 2.347417840375587,
212
+ "grad_norm": NaN,
213
+ "learning_rate": 4.351706949516026e-05,
214
+ "loss": 0.0,
215
  "step": 13500
216
  },
217
  {
218
  "epoch": 2.434359241870979,
219
+ "grad_norm": NaN,
220
+ "learning_rate": 3.7720976062134123e-05,
221
+ "loss": 0.0,
222
  "step": 14000
223
  },
224
  {
225
  "epoch": 2.521300643366371,
226
+ "grad_norm": NaN,
227
+ "learning_rate": 3.1924882629107986e-05,
228
+ "loss": 0.0,
229
  "step": 14500
230
  },
231
  {
232
  "epoch": 2.608242044861763,
233
+ "grad_norm": NaN,
234
+ "learning_rate": 2.6128789196081842e-05,
235
+ "loss": 0.0,
236
  "step": 15000
237
  },
238
  {
239
  "epoch": 2.6951834463571553,
240
+ "grad_norm": NaN,
241
+ "learning_rate": 2.03326957630557e-05,
242
+ "loss": 0.0,
243
  "step": 15500
244
  },
245
  {
246
  "epoch": 2.7821248478525473,
247
+ "grad_norm": NaN,
248
+ "learning_rate": 1.4536602330029562e-05,
249
+ "loss": 0.0,
250
  "step": 16000
251
  },
252
  {
253
  "epoch": 2.8690662493479397,
254
+ "grad_norm": NaN,
255
+ "learning_rate": 8.74050889700342e-06,
256
+ "loss": 0.0,
257
  "step": 16500
258
  },
259
  {
260
  "epoch": 2.9560076508433317,
261
+ "grad_norm": NaN,
262
+ "learning_rate": 2.9444154639772795e-06,
263
+ "loss": 0.0,
264
  "step": 17000
265
  },
266
  {
267
  "epoch": 3.0,
268
+ "eval_loss": NaN,
269
+ "eval_runtime": 152.6977,
270
+ "eval_samples_per_second": 16.739,
271
+ "eval_steps_per_second": 4.185,
272
  "step": 17253
273
  }
274
  ],
 
289
  "attributes": {}
290
  }
291
  },
292
+ "total_flos": 9.345012382826496e+16,
293
  "train_batch_size": 4,
294
  "trial_name": null,
295
  "trial_params": null
checkpoint-17253/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfb590c3405bb321e9c77b1f2243981f8f41d35f037852c7fac86c1191f3b93d
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99fccf4927862cc4fd8283f3958f1ef8528c86594c8ff39225b274adde4416df
3
  size 5240
checkpoint-5751/config.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 32768,
15
+ "max_window_layers": 28,
16
+ "model_type": "qwen3",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 28,
19
+ "num_key_value_heads": 8,
20
+ "quantization_config": {
21
+ "config_groups": {
22
+ "group_0": {
23
+ "input_activations": {
24
+ "actorder": null,
25
+ "block_structure": null,
26
+ "dynamic": true,
27
+ "group_size": null,
28
+ "num_bits": 8,
29
+ "observer": null,
30
+ "observer_kwargs": {},
31
+ "strategy": "token",
32
+ "symmetric": true,
33
+ "type": "int"
34
+ },
35
+ "output_activations": null,
36
+ "targets": [
37
+ "Linear"
38
+ ],
39
+ "weights": {
40
+ "actorder": null,
41
+ "block_structure": null,
42
+ "dynamic": false,
43
+ "group_size": null,
44
+ "num_bits": 8,
45
+ "observer": "minmax",
46
+ "observer_kwargs": {},
47
+ "strategy": "channel",
48
+ "symmetric": true,
49
+ "type": "int"
50
+ }
51
+ }
52
+ },
53
+ "format": "int-quantized",
54
+ "global_compression_ratio": null,
55
+ "ignore": [
56
+ "lm_head"
57
+ ],
58
+ "kv_cache_scheme": null,
59
+ "quant_method": "compressed-tensors",
60
+ "quantization_status": "compressed",
61
+ "sparsity_config": {}
62
+ },
63
+ "rms_norm_eps": 1e-06,
64
+ "rope_scaling": null,
65
+ "rope_theta": 1000000,
66
+ "sliding_window": null,
67
+ "tie_word_embeddings": true,
68
+ "torch_dtype": "float16",
69
+ "transformers_version": "4.51.3",
70
+ "use_cache": true,
71
+ "use_sliding_window": false,
72
+ "vocab_size": 151936
73
+ }
checkpoint-5751/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "eos_token_id": 151643,
4
+ "max_new_tokens": 2048,
5
+ "transformers_version": "4.51.3"
6
+ }
checkpoint-5751/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb3d1bd95c53800ef1cb04bf0b4dca4678ae38f01058094213011b8ab1df2db9
3
+ size 1192845752
checkpoint-5751/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:73d0682019e9bed4623ac62035f013ab3f9476d0f384d4fd945108357de6f423
3
- size 40699434
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f80e3bfb0a586f776698e79ea6153229419cd4d77b77c70d4fa9d4c9ead70a5
3
+ size 622337116
checkpoint-5751/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4394136fb64eb98f651f75ef940681d780caab4c23a847d316298109499168bd
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
3
  size 14244
checkpoint-5751/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:005bbd780391869e7961f515cfb7223f41a929b03950c9ee31e174d339bd5ada
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:903ca14f005e51a1e28984de583841d853b88748f2e6d3a88d37904a28596d8e
3
  size 1064
checkpoint-5751/trainer_state.json CHANGED
@@ -11,87 +11,87 @@
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
- "grad_norm": 0.2196940779685974,
15
- "learning_rate": 0.00019426186750130412,
16
- "loss": 0.2606,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
- "grad_norm": 0.2677164673805237,
22
- "learning_rate": 0.000188465774068278,
23
- "loss": 0.1763,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
- "grad_norm": 0.21263259649276733,
29
- "learning_rate": 0.00018266968063525185,
30
- "loss": 0.1714,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
- "grad_norm": 0.275687575340271,
36
- "learning_rate": 0.0001768735872022257,
37
- "loss": 0.1693,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
- "grad_norm": 0.21894076466560364,
43
- "learning_rate": 0.00017107749376919957,
44
- "loss": 0.1719,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
- "grad_norm": 0.23001301288604736,
50
- "learning_rate": 0.00016528140033617342,
51
- "loss": 0.1691,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
- "grad_norm": 0.23900070786476135,
57
- "learning_rate": 0.0001594853069031473,
58
- "loss": 0.167,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
- "grad_norm": 0.2820037603378296,
64
- "learning_rate": 0.00015368921347012115,
65
- "loss": 0.1678,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
- "grad_norm": 0.3206230700016022,
71
- "learning_rate": 0.00014789312003709503,
72
- "loss": 0.1675,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
- "grad_norm": 0.20400723814964294,
78
- "learning_rate": 0.00014209702660406887,
79
- "loss": 0.1658,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
- "grad_norm": 0.31783542037010193,
85
- "learning_rate": 0.00013630093317104272,
86
- "loss": 0.1643,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
- "eval_loss": 0.16573764383792877,
92
- "eval_runtime": 80.6587,
93
- "eval_samples_per_second": 31.689,
94
- "eval_steps_per_second": 7.922,
95
  "step": 5751
96
  }
97
  ],
@@ -112,7 +112,7 @@
112
  "attributes": {}
113
  }
114
  },
115
- "total_flos": 3.148232382008525e+16,
116
  "train_batch_size": 4,
117
  "trial_name": null,
118
  "trial_params": null
 
11
  "log_history": [
12
  {
13
  "epoch": 0.0869414014953921,
14
+ "grad_norm": NaN,
15
+ "learning_rate": 0.00019421549875383993,
16
+ "loss": 4.2823,
17
  "step": 500
18
  },
19
  {
20
  "epoch": 0.1738828029907842,
21
+ "grad_norm": NaN,
22
+ "learning_rate": 0.0001884194053208138,
23
+ "loss": 0.0,
24
  "step": 1000
25
  },
26
  {
27
  "epoch": 0.2608242044861763,
28
+ "grad_norm": NaN,
29
+ "learning_rate": 0.00018262331188778763,
30
+ "loss": 0.0,
31
  "step": 1500
32
  },
33
  {
34
  "epoch": 0.3477656059815684,
35
+ "grad_norm": NaN,
36
+ "learning_rate": 0.0001768272184547615,
37
+ "loss": 0.0,
38
  "step": 2000
39
  },
40
  {
41
  "epoch": 0.4347070074769605,
42
+ "grad_norm": NaN,
43
+ "learning_rate": 0.00017103112502173535,
44
+ "loss": 0.0,
45
  "step": 2500
46
  },
47
  {
48
  "epoch": 0.5216484089723527,
49
+ "grad_norm": NaN,
50
+ "learning_rate": 0.0001652350315887092,
51
+ "loss": 0.0,
52
  "step": 3000
53
  },
54
  {
55
  "epoch": 0.6085898104677447,
56
+ "grad_norm": NaN,
57
+ "learning_rate": 0.00015943893815568308,
58
+ "loss": 0.0,
59
  "step": 3500
60
  },
61
  {
62
  "epoch": 0.6955312119631368,
63
+ "grad_norm": NaN,
64
+ "learning_rate": 0.00015364284472265693,
65
+ "loss": 0.0,
66
  "step": 4000
67
  },
68
  {
69
  "epoch": 0.7824726134585289,
70
+ "grad_norm": NaN,
71
+ "learning_rate": 0.0001478467512896308,
72
+ "loss": 0.0,
73
  "step": 4500
74
  },
75
  {
76
  "epoch": 0.869414014953921,
77
+ "grad_norm": NaN,
78
+ "learning_rate": 0.00014205065785660465,
79
+ "loss": 0.0,
80
  "step": 5000
81
  },
82
  {
83
  "epoch": 0.9563554164493132,
84
+ "grad_norm": NaN,
85
+ "learning_rate": 0.0001362545644235785,
86
+ "loss": 0.0,
87
  "step": 5500
88
  },
89
  {
90
  "epoch": 1.0,
91
+ "eval_loss": NaN,
92
+ "eval_runtime": 152.5081,
93
+ "eval_samples_per_second": 16.76,
94
+ "eval_steps_per_second": 4.19,
95
  "step": 5751
96
  }
97
  ],
 
112
  "attributes": {}
113
  }
114
  },
115
+ "total_flos": 3.115004127608832e+16,
116
  "train_batch_size": 4,
117
  "trial_name": null,
118
  "trial_params": null
checkpoint-5751/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bfb590c3405bb321e9c77b1f2243981f8f41d35f037852c7fac86c1191f3b93d
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99fccf4927862cc4fd8283f3958f1ef8528c86594c8ff39225b274adde4416df
3
  size 5240