Upload folder using huggingface_hub
Browse files- README.md +10 -0
- adapter_config.json +83 -0
- adapter_model.bin +3 -0
- config.json +192 -0
- non_lora_trainables.bin +3 -0
- preprocessor_config.json +20 -0
- trainer_state.json +366 -0
README.md
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
### Framework versions
|
7 |
+
|
8 |
+
- PEFT 0.4.0
|
9 |
+
|
10 |
+
- PEFT 0.4.0
|
adapter_config.json
ADDED
@@ -0,0 +1,83 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "q-future/one-align",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 256,
|
11 |
+
"lora_dropout": 0.05,
|
12 |
+
"modules_to_save": null,
|
13 |
+
"peft_type": "LORA",
|
14 |
+
"r": 128,
|
15 |
+
"revision": null,
|
16 |
+
"target_modules": [
|
17 |
+
"model.layers.17.self_attn.q_proj",
|
18 |
+
"model.layers.10.self_attn.v_proj.multiway.1",
|
19 |
+
"model.layers.3.self_attn.q_proj",
|
20 |
+
"model.layers.6.self_attn.v_proj.multiway.1",
|
21 |
+
"model.layers.17.self_attn.v_proj.multiway.1",
|
22 |
+
"model.layers.20.self_attn.v_proj.multiway.1",
|
23 |
+
"model.layers.24.self_attn.v_proj.multiway.1",
|
24 |
+
"model.layers.13.self_attn.v_proj.multiway.1",
|
25 |
+
"model.layers.23.self_attn.q_proj",
|
26 |
+
"model.layers.21.self_attn.v_proj.multiway.1",
|
27 |
+
"model.layers.15.self_attn.v_proj.multiway.1",
|
28 |
+
"model.layers.2.self_attn.v_proj.multiway.1",
|
29 |
+
"model.layers.27.self_attn.v_proj.multiway.1",
|
30 |
+
"model.layers.15.self_attn.q_proj",
|
31 |
+
"model.layers.28.self_attn.v_proj.multiway.1",
|
32 |
+
"model.layers.30.self_attn.q_proj",
|
33 |
+
"model.layers.5.self_attn.q_proj",
|
34 |
+
"model.layers.22.self_attn.q_proj",
|
35 |
+
"model.layers.20.self_attn.q_proj",
|
36 |
+
"model.layers.22.self_attn.v_proj.multiway.1",
|
37 |
+
"model.layers.24.self_attn.q_proj",
|
38 |
+
"model.layers.25.self_attn.v_proj.multiway.1",
|
39 |
+
"model.layers.7.self_attn.q_proj",
|
40 |
+
"model.layers.18.self_attn.v_proj.multiway.1",
|
41 |
+
"model.layers.31.self_attn.q_proj",
|
42 |
+
"model.layers.0.self_attn.v_proj.multiway.1",
|
43 |
+
"model.layers.1.self_attn.v_proj.multiway.1",
|
44 |
+
"model.layers.10.self_attn.q_proj",
|
45 |
+
"model.layers.11.self_attn.q_proj",
|
46 |
+
"model.layers.30.self_attn.v_proj.multiway.1",
|
47 |
+
"model.layers.29.self_attn.v_proj.multiway.1",
|
48 |
+
"model.layers.27.self_attn.q_proj",
|
49 |
+
"model.layers.31.self_attn.v_proj.multiway.1",
|
50 |
+
"model.layers.16.self_attn.q_proj",
|
51 |
+
"model.layers.19.self_attn.q_proj",
|
52 |
+
"model.layers.5.self_attn.v_proj.multiway.1",
|
53 |
+
"model.layers.26.self_attn.v_proj.multiway.1",
|
54 |
+
"model.layers.8.self_attn.v_proj.multiway.1",
|
55 |
+
"model.layers.12.self_attn.q_proj",
|
56 |
+
"model.layers.7.self_attn.v_proj.multiway.1",
|
57 |
+
"model.layers.25.self_attn.q_proj",
|
58 |
+
"model.layers.13.self_attn.q_proj",
|
59 |
+
"model.layers.1.self_attn.q_proj",
|
60 |
+
"model.layers.8.self_attn.q_proj",
|
61 |
+
"model.layers.4.self_attn.v_proj.multiway.1",
|
62 |
+
"model.layers.14.self_attn.v_proj.multiway.1",
|
63 |
+
"model.layers.16.self_attn.v_proj.multiway.1",
|
64 |
+
"model.layers.2.self_attn.q_proj",
|
65 |
+
"model.layers.28.self_attn.q_proj",
|
66 |
+
"model.layers.9.self_attn.v_proj.multiway.1",
|
67 |
+
"model.layers.14.self_attn.q_proj",
|
68 |
+
"model.layers.3.self_attn.v_proj.multiway.1",
|
69 |
+
"model.layers.0.self_attn.q_proj",
|
70 |
+
"model.layers.26.self_attn.q_proj",
|
71 |
+
"model.layers.12.self_attn.v_proj.multiway.1",
|
72 |
+
"model.layers.19.self_attn.v_proj.multiway.1",
|
73 |
+
"model.layers.6.self_attn.q_proj",
|
74 |
+
"model.layers.11.self_attn.v_proj.multiway.1",
|
75 |
+
"model.layers.4.self_attn.q_proj",
|
76 |
+
"model.layers.21.self_attn.q_proj",
|
77 |
+
"model.layers.9.self_attn.q_proj",
|
78 |
+
"model.layers.23.self_attn.v_proj.multiway.1",
|
79 |
+
"model.layers.29.self_attn.q_proj",
|
80 |
+
"model.layers.18.self_attn.q_proj"
|
81 |
+
],
|
82 |
+
"task_type": "CAUSAL_LM"
|
83 |
+
}
|
adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ced90ec0bc072c12d9792c22e5bff526134b04ba1075fe7474915b8725ea47ac
|
3 |
+
size 134262669
|
config.json
ADDED
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "q-future/one-align",
|
3 |
+
"architectures": [
|
4 |
+
"MPLUGOwl2LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "q-future/one-align--configuration_mplug_owl2.MPLUGOwl2Config",
|
10 |
+
"AutoModel": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM",
|
11 |
+
"AutoModelForCausalLM": "q-future/one-align--modeling_mplug_owl2.MPLUGOwl2LlamaForCausalLM"
|
12 |
+
},
|
13 |
+
"bos_token_id": 1,
|
14 |
+
"eos_token_id": 2,
|
15 |
+
"freeze_vision_model": false,
|
16 |
+
"hidden_act": "silu",
|
17 |
+
"hidden_size": 4096,
|
18 |
+
"image_aspect_ratio": "pad",
|
19 |
+
"image_grid_pinpoints": null,
|
20 |
+
"initializer_range": 0.02,
|
21 |
+
"intermediate_size": 11008,
|
22 |
+
"max_position_embeddings": 2048,
|
23 |
+
"model_type": "mplug_owl2",
|
24 |
+
"num_attention_heads": 32,
|
25 |
+
"num_hidden_layers": 32,
|
26 |
+
"num_key_value_heads": 32,
|
27 |
+
"pretraining_tp": 1,
|
28 |
+
"rms_norm_eps": 1e-06,
|
29 |
+
"rope_scaling": null,
|
30 |
+
"rope_theta": 10000.0,
|
31 |
+
"tie_word_embeddings": false,
|
32 |
+
"torch_dtype": "bfloat16",
|
33 |
+
"transformers_version": "4.36.1",
|
34 |
+
"tune_visual_abstractor": true,
|
35 |
+
"use_cache": true,
|
36 |
+
"visual_abstractor_lr": 2e-05,
|
37 |
+
"visual_config": {
|
38 |
+
"visual_abstractor": {
|
39 |
+
"_name_or_path": "",
|
40 |
+
"add_cross_attention": false,
|
41 |
+
"architectures": null,
|
42 |
+
"attention_probs_dropout_prob": 0.0,
|
43 |
+
"bad_words_ids": null,
|
44 |
+
"begin_suppress_tokens": null,
|
45 |
+
"bos_token_id": null,
|
46 |
+
"chunk_size_feed_forward": 0,
|
47 |
+
"cross_attention_hidden_size": null,
|
48 |
+
"decoder_start_token_id": null,
|
49 |
+
"diversity_penalty": 0.0,
|
50 |
+
"do_sample": false,
|
51 |
+
"early_stopping": false,
|
52 |
+
"encoder_hidden_size": 1024,
|
53 |
+
"encoder_no_repeat_ngram_size": 0,
|
54 |
+
"eos_token_id": null,
|
55 |
+
"exponential_decay_length_penalty": null,
|
56 |
+
"finetuning_task": null,
|
57 |
+
"forced_bos_token_id": null,
|
58 |
+
"forced_eos_token_id": null,
|
59 |
+
"grid_size": 32,
|
60 |
+
"hidden_size": 1024,
|
61 |
+
"id2label": {
|
62 |
+
"0": "LABEL_0",
|
63 |
+
"1": "LABEL_1"
|
64 |
+
},
|
65 |
+
"initializer_range": 0.02,
|
66 |
+
"intermediate_size": 2816,
|
67 |
+
"is_decoder": false,
|
68 |
+
"is_encoder_decoder": false,
|
69 |
+
"label2id": {
|
70 |
+
"LABEL_0": 0,
|
71 |
+
"LABEL_1": 1
|
72 |
+
},
|
73 |
+
"layer_norm_eps": 1e-06,
|
74 |
+
"length_penalty": 1.0,
|
75 |
+
"max_length": 20,
|
76 |
+
"min_length": 0,
|
77 |
+
"model_type": "mplug_owl_visual_abstract",
|
78 |
+
"no_repeat_ngram_size": 0,
|
79 |
+
"num_attention_heads": 16,
|
80 |
+
"num_beam_groups": 1,
|
81 |
+
"num_beams": 1,
|
82 |
+
"num_hidden_layers": 6,
|
83 |
+
"num_learnable_queries": 64,
|
84 |
+
"num_return_sequences": 1,
|
85 |
+
"output_attentions": false,
|
86 |
+
"output_hidden_states": false,
|
87 |
+
"output_scores": false,
|
88 |
+
"pad_token_id": null,
|
89 |
+
"prefix": null,
|
90 |
+
"problem_type": null,
|
91 |
+
"pruned_heads": {},
|
92 |
+
"remove_invalid_values": false,
|
93 |
+
"repetition_penalty": 1.0,
|
94 |
+
"return_dict": true,
|
95 |
+
"return_dict_in_generate": false,
|
96 |
+
"sep_token_id": null,
|
97 |
+
"suppress_tokens": null,
|
98 |
+
"task_specific_params": null,
|
99 |
+
"temperature": 1.0,
|
100 |
+
"tf_legacy_loss": false,
|
101 |
+
"tie_encoder_decoder": false,
|
102 |
+
"tie_word_embeddings": true,
|
103 |
+
"tokenizer_class": null,
|
104 |
+
"top_k": 50,
|
105 |
+
"top_p": 1.0,
|
106 |
+
"torch_dtype": null,
|
107 |
+
"torchscript": false,
|
108 |
+
"transformers_version": "4.28.1",
|
109 |
+
"typical_p": 1.0,
|
110 |
+
"use_bfloat16": false
|
111 |
+
},
|
112 |
+
"visual_model": {
|
113 |
+
"_name_or_path": "",
|
114 |
+
"add_cross_attention": false,
|
115 |
+
"architectures": null,
|
116 |
+
"attention_dropout": 0.0,
|
117 |
+
"bad_words_ids": null,
|
118 |
+
"begin_suppress_tokens": null,
|
119 |
+
"bos_token_id": null,
|
120 |
+
"chunk_size_feed_forward": 0,
|
121 |
+
"cross_attention_hidden_size": null,
|
122 |
+
"decoder_start_token_id": null,
|
123 |
+
"diversity_penalty": 0.0,
|
124 |
+
"do_sample": false,
|
125 |
+
"early_stopping": false,
|
126 |
+
"encoder_no_repeat_ngram_size": 0,
|
127 |
+
"eos_token_id": null,
|
128 |
+
"exponential_decay_length_penalty": null,
|
129 |
+
"finetuning_task": null,
|
130 |
+
"forced_bos_token_id": null,
|
131 |
+
"forced_eos_token_id": null,
|
132 |
+
"hidden_act": "quick_gelu",
|
133 |
+
"hidden_size": 1024,
|
134 |
+
"id2label": {
|
135 |
+
"0": "LABEL_0",
|
136 |
+
"1": "LABEL_1"
|
137 |
+
},
|
138 |
+
"image_size": 448,
|
139 |
+
"initializer_factor": 1.0,
|
140 |
+
"initializer_range": 0.02,
|
141 |
+
"intermediate_size": 4096,
|
142 |
+
"is_decoder": false,
|
143 |
+
"is_encoder_decoder": false,
|
144 |
+
"label2id": {
|
145 |
+
"LABEL_0": 0,
|
146 |
+
"LABEL_1": 1
|
147 |
+
},
|
148 |
+
"layer_norm_eps": 1e-06,
|
149 |
+
"length_penalty": 1.0,
|
150 |
+
"max_length": 20,
|
151 |
+
"min_length": 0,
|
152 |
+
"model_type": "mplug_owl_vision_model",
|
153 |
+
"no_repeat_ngram_size": 0,
|
154 |
+
"num_attention_heads": 16,
|
155 |
+
"num_beam_groups": 1,
|
156 |
+
"num_beams": 1,
|
157 |
+
"num_channels": 3,
|
158 |
+
"num_hidden_layers": 24,
|
159 |
+
"num_return_sequences": 1,
|
160 |
+
"output_attentions": false,
|
161 |
+
"output_hidden_states": false,
|
162 |
+
"output_scores": false,
|
163 |
+
"pad_token_id": null,
|
164 |
+
"patch_size": 14,
|
165 |
+
"prefix": null,
|
166 |
+
"problem_type": null,
|
167 |
+
"projection_dim": 768,
|
168 |
+
"pruned_heads": {},
|
169 |
+
"remove_invalid_values": false,
|
170 |
+
"repetition_penalty": 1.0,
|
171 |
+
"return_dict": true,
|
172 |
+
"return_dict_in_generate": false,
|
173 |
+
"sep_token_id": null,
|
174 |
+
"suppress_tokens": null,
|
175 |
+
"task_specific_params": null,
|
176 |
+
"temperature": 1.0,
|
177 |
+
"tf_legacy_loss": false,
|
178 |
+
"tie_encoder_decoder": false,
|
179 |
+
"tie_word_embeddings": true,
|
180 |
+
"tokenizer_class": null,
|
181 |
+
"top_k": 50,
|
182 |
+
"top_p": 1.0,
|
183 |
+
"torch_dtype": null,
|
184 |
+
"torchscript": false,
|
185 |
+
"transformers_version": "4.28.1",
|
186 |
+
"typical_p": 1.0,
|
187 |
+
"use_bfloat16": false,
|
188 |
+
"use_flash_attn": false
|
189 |
+
}
|
190 |
+
},
|
191 |
+
"vocab_size": 32000
|
192 |
+
}
|
non_lora_trainables.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4f9a6301cac7ed0e6cefe54e66ef8d1c29415f91308bfc8012be40ddd4ac356c
|
3 |
+
size 162998185
|
preprocessor_config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"crop_size": 448,
|
3 |
+
"do_center_crop": true,
|
4 |
+
"do_normalize": true,
|
5 |
+
"do_resize": true,
|
6 |
+
"feature_extractor_type": "CLIPFeatureExtractor",
|
7 |
+
"image_mean": [
|
8 |
+
0.48145466,
|
9 |
+
0.4578275,
|
10 |
+
0.40821073
|
11 |
+
],
|
12 |
+
"image_std": [
|
13 |
+
0.26862954,
|
14 |
+
0.26130258,
|
15 |
+
0.27577711
|
16 |
+
],
|
17 |
+
"resample": 3,
|
18 |
+
"size": 448
|
19 |
+
}
|
20 |
+
|
trainer_state.json
ADDED
@@ -0,0 +1,366 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9846153846153847,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 56,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.02,
|
13 |
+
"learning_rate": 0.0001,
|
14 |
+
"loss": 0.1065,
|
15 |
+
"step": 1
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.04,
|
19 |
+
"learning_rate": 0.0002,
|
20 |
+
"loss": 0.1299,
|
21 |
+
"step": 2
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.05,
|
25 |
+
"learning_rate": 0.00019983081582712685,
|
26 |
+
"loss": 0.145,
|
27 |
+
"step": 3
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.07,
|
31 |
+
"learning_rate": 0.00019932383577419432,
|
32 |
+
"loss": 0.1167,
|
33 |
+
"step": 4
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.09,
|
37 |
+
"learning_rate": 0.00019848077530122083,
|
38 |
+
"loss": 0.1202,
|
39 |
+
"step": 5
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.11,
|
43 |
+
"learning_rate": 0.00019730448705798239,
|
44 |
+
"loss": 0.1051,
|
45 |
+
"step": 6
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.12,
|
49 |
+
"learning_rate": 0.0001957989512315489,
|
50 |
+
"loss": 0.1127,
|
51 |
+
"step": 7
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.14,
|
55 |
+
"learning_rate": 0.00019396926207859084,
|
56 |
+
"loss": 0.0987,
|
57 |
+
"step": 8
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.16,
|
61 |
+
"learning_rate": 0.00019182161068802741,
|
62 |
+
"loss": 0.0992,
|
63 |
+
"step": 9
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.18,
|
67 |
+
"learning_rate": 0.00018936326403234125,
|
68 |
+
"loss": 0.129,
|
69 |
+
"step": 10
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.19,
|
73 |
+
"learning_rate": 0.00018660254037844388,
|
74 |
+
"loss": 0.0898,
|
75 |
+
"step": 11
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.21,
|
79 |
+
"learning_rate": 0.00018354878114129367,
|
80 |
+
"loss": 0.1101,
|
81 |
+
"step": 12
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.23,
|
85 |
+
"learning_rate": 0.0001802123192755044,
|
86 |
+
"loss": 0.0951,
|
87 |
+
"step": 13
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.25,
|
91 |
+
"learning_rate": 0.0001766044443118978,
|
92 |
+
"loss": 0.0916,
|
93 |
+
"step": 14
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.26,
|
97 |
+
"learning_rate": 0.00017273736415730488,
|
98 |
+
"loss": 0.0969,
|
99 |
+
"step": 15
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.28,
|
103 |
+
"learning_rate": 0.0001686241637868734,
|
104 |
+
"loss": 0.082,
|
105 |
+
"step": 16
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.3,
|
109 |
+
"learning_rate": 0.00016427876096865394,
|
110 |
+
"loss": 0.0842,
|
111 |
+
"step": 17
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.32,
|
115 |
+
"learning_rate": 0.00015971585917027862,
|
116 |
+
"loss": 0.0923,
|
117 |
+
"step": 18
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.33,
|
121 |
+
"learning_rate": 0.0001549508978070806,
|
122 |
+
"loss": 0.0867,
|
123 |
+
"step": 19
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.35,
|
127 |
+
"learning_rate": 0.00015000000000000001,
|
128 |
+
"loss": 0.1179,
|
129 |
+
"step": 20
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.37,
|
133 |
+
"learning_rate": 0.00014487991802004623,
|
134 |
+
"loss": 0.1062,
|
135 |
+
"step": 21
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.39,
|
139 |
+
"learning_rate": 0.0001396079766039157,
|
140 |
+
"loss": 0.1287,
|
141 |
+
"step": 22
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.4,
|
145 |
+
"learning_rate": 0.00013420201433256689,
|
146 |
+
"loss": 0.0968,
|
147 |
+
"step": 23
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 0.42,
|
151 |
+
"learning_rate": 0.00012868032327110904,
|
152 |
+
"loss": 0.0906,
|
153 |
+
"step": 24
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 0.44,
|
157 |
+
"learning_rate": 0.00012306158707424403,
|
158 |
+
"loss": 0.0819,
|
159 |
+
"step": 25
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.46,
|
163 |
+
"learning_rate": 0.00011736481776669306,
|
164 |
+
"loss": 0.0878,
|
165 |
+
"step": 26
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.47,
|
169 |
+
"learning_rate": 0.00011160929141252303,
|
170 |
+
"loss": 0.0793,
|
171 |
+
"step": 27
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.49,
|
175 |
+
"learning_rate": 0.00010581448289104758,
|
176 |
+
"loss": 0.099,
|
177 |
+
"step": 28
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.51,
|
181 |
+
"learning_rate": 0.0001,
|
182 |
+
"loss": 0.1027,
|
183 |
+
"step": 29
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 0.53,
|
187 |
+
"learning_rate": 9.418551710895243e-05,
|
188 |
+
"loss": 0.093,
|
189 |
+
"step": 30
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.55,
|
193 |
+
"learning_rate": 8.839070858747697e-05,
|
194 |
+
"loss": 0.0949,
|
195 |
+
"step": 31
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.56,
|
199 |
+
"learning_rate": 8.263518223330697e-05,
|
200 |
+
"loss": 0.0938,
|
201 |
+
"step": 32
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 0.58,
|
205 |
+
"learning_rate": 7.693841292575598e-05,
|
206 |
+
"loss": 0.0996,
|
207 |
+
"step": 33
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 0.6,
|
211 |
+
"learning_rate": 7.131967672889101e-05,
|
212 |
+
"loss": 0.0976,
|
213 |
+
"step": 34
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 0.62,
|
217 |
+
"learning_rate": 6.579798566743314e-05,
|
218 |
+
"loss": 0.1112,
|
219 |
+
"step": 35
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.63,
|
223 |
+
"learning_rate": 6.039202339608432e-05,
|
224 |
+
"loss": 0.0932,
|
225 |
+
"step": 36
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.65,
|
229 |
+
"learning_rate": 5.5120081979953785e-05,
|
230 |
+
"loss": 0.1011,
|
231 |
+
"step": 37
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 0.67,
|
235 |
+
"learning_rate": 5.000000000000002e-05,
|
236 |
+
"loss": 0.0911,
|
237 |
+
"step": 38
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 0.69,
|
241 |
+
"learning_rate": 4.50491021929194e-05,
|
242 |
+
"loss": 0.0997,
|
243 |
+
"step": 39
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"epoch": 0.7,
|
247 |
+
"learning_rate": 4.028414082972141e-05,
|
248 |
+
"loss": 0.0987,
|
249 |
+
"step": 40
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 0.72,
|
253 |
+
"learning_rate": 3.5721239031346066e-05,
|
254 |
+
"loss": 0.0935,
|
255 |
+
"step": 41
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 0.74,
|
259 |
+
"learning_rate": 3.137583621312665e-05,
|
260 |
+
"loss": 0.1054,
|
261 |
+
"step": 42
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.76,
|
265 |
+
"learning_rate": 2.7262635842695127e-05,
|
266 |
+
"loss": 0.0869,
|
267 |
+
"step": 43
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.77,
|
271 |
+
"learning_rate": 2.339555568810221e-05,
|
272 |
+
"loss": 0.095,
|
273 |
+
"step": 44
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 0.79,
|
277 |
+
"learning_rate": 1.9787680724495617e-05,
|
278 |
+
"loss": 0.1,
|
279 |
+
"step": 45
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 0.81,
|
283 |
+
"learning_rate": 1.6451218858706374e-05,
|
284 |
+
"loss": 0.0985,
|
285 |
+
"step": 46
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 0.83,
|
289 |
+
"learning_rate": 1.339745962155613e-05,
|
290 |
+
"loss": 0.088,
|
291 |
+
"step": 47
|
292 |
+
},
|
293 |
+
{
|
294 |
+
"epoch": 0.84,
|
295 |
+
"learning_rate": 1.0636735967658784e-05,
|
296 |
+
"loss": 0.0777,
|
297 |
+
"step": 48
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 0.86,
|
301 |
+
"learning_rate": 8.178389311972612e-06,
|
302 |
+
"loss": 0.0844,
|
303 |
+
"step": 49
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.88,
|
307 |
+
"learning_rate": 6.030737921409169e-06,
|
308 |
+
"loss": 0.0971,
|
309 |
+
"step": 50
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.9,
|
313 |
+
"learning_rate": 4.20104876845111e-06,
|
314 |
+
"loss": 0.0946,
|
315 |
+
"step": 51
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 0.91,
|
319 |
+
"learning_rate": 2.6955129420176196e-06,
|
320 |
+
"loss": 0.0853,
|
321 |
+
"step": 52
|
322 |
+
},
|
323 |
+
{
|
324 |
+
"epoch": 0.93,
|
325 |
+
"learning_rate": 1.5192246987791981e-06,
|
326 |
+
"loss": 0.0889,
|
327 |
+
"step": 53
|
328 |
+
},
|
329 |
+
{
|
330 |
+
"epoch": 0.95,
|
331 |
+
"learning_rate": 6.761642258056978e-07,
|
332 |
+
"loss": 0.0927,
|
333 |
+
"step": 54
|
334 |
+
},
|
335 |
+
{
|
336 |
+
"epoch": 0.97,
|
337 |
+
"learning_rate": 1.6918417287318245e-07,
|
338 |
+
"loss": 0.0971,
|
339 |
+
"step": 55
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 0.98,
|
343 |
+
"learning_rate": 0.0,
|
344 |
+
"loss": 0.1002,
|
345 |
+
"step": 56
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.98,
|
349 |
+
"step": 56,
|
350 |
+
"total_flos": 1401921699840.0,
|
351 |
+
"train_loss": 0.0989542007446289,
|
352 |
+
"train_runtime": 4591.4133,
|
353 |
+
"train_samples_per_second": 0.791,
|
354 |
+
"train_steps_per_second": 0.012
|
355 |
+
}
|
356 |
+
],
|
357 |
+
"logging_steps": 1.0,
|
358 |
+
"max_steps": 56,
|
359 |
+
"num_input_tokens_seen": 0,
|
360 |
+
"num_train_epochs": 1,
|
361 |
+
"save_steps": 800,
|
362 |
+
"total_flos": 1401921699840.0,
|
363 |
+
"train_batch_size": 4,
|
364 |
+
"trial_name": null,
|
365 |
+
"trial_params": null
|
366 |
+
}
|