Hugging Face's logo

Jinxing1
/
TGS-Agent

Safetensors
Model card Files
xet
 Community
jasongief commited on
Commit
9fe31b8
·
1 Parent(s): 7290b07

Upload model files

Browse files
Files changed (11) hide show
  1. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/README.md +202 -0
  2. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/adapter_config.json +34 -0
  3. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/adapter_model.safetensors +3 -0
  4. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/checkpoint-551/config.json +28 -0
  5. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/checkpoint-551/finetune_weights.bin +3 -0
  6. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/config.json +28 -0
  7. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/model.txt +371 -0
  8. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/model_trainable_params.txt +612 -0
  9. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/non_lora_trainables.bin +3 -0
  10. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/saved_config.json +193 -0
  11. epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/trainer_state.json +3990 -0
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: pretrained_weights/Llama-2-7b-chat-hf
3
+ library_name: peft
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.12.0
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "pretrained_weights/Llama-2-7b-chat-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 16,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "down_proj",
24
+ "up_proj",
25
+ "v_proj",
26
+ "k_proj",
27
+ "o_proj",
28
+ "gate_proj",
29
+ "q_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5697e8ca8a749034d30acf5f2c674614f574636a30c5027b90e253c728bc70b2
3
+ size 40036488
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/checkpoint-551/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "pretrained_weights/Llama-2-7b-chat-hf",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "max_position_embeddings": 4096,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 32,
18
+ "num_key_value_heads": 32,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-05,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float16",
25
+ "transformers_version": "4.37.2",
26
+ "use_cache": false,
27
+ "vocab_size": 32021
28
+ }
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/checkpoint-551/finetune_weights.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3c6bff2146db97e50ac62aca7be45ce21d287625f09960673a2db5304419858
3
+ size 332839606
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "pretrained_weights/Llama-2-7b-chat-hf",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "max_position_embeddings": 4096,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 32,
18
+ "num_key_value_heads": 32,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-05,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float16",
25
+ "transformers_version": "4.37.2",
26
+ "use_cache": true,
27
+ "vocab_size": 32021
28
+ }
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/model.txt ADDED
@@ -0,0 +1,371 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ PeftModelForCausalLM(
2
+ (base_model): LoraModel(
3
+ (model): UnifiedForCausalLM(
4
+ (model): UnifiedModel(
5
+ (embed_tokens): Embedding(32021, 4096)
6
+ (layers): ModuleList(
7
+ (0-31): 32 x LlamaDecoderLayer(
8
+ (self_attn): LlamaSdpaAttention(
9
+ (q_proj): lora.Linear(
10
+ (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
11
+ (lora_dropout): ModuleDict(
12
+ (default): Dropout(p=0.05, inplace=False)
13
+ )
14
+ (lora_A): ModuleDict(
15
+ (default): Linear(in_features=4096, out_features=8, bias=False)
16
+ )
17
+ (lora_B): ModuleDict(
18
+ (default): Linear(in_features=8, out_features=4096, bias=False)
19
+ )
20
+ (lora_embedding_A): ParameterDict()
21
+ (lora_embedding_B): ParameterDict()
22
+ (lora_magnitude_vector): ModuleDict()
23
+ )
24
+ (k_proj): lora.Linear(
25
+ (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
26
+ (lora_dropout): ModuleDict(
27
+ (default): Dropout(p=0.05, inplace=False)
28
+ )
29
+ (lora_A): ModuleDict(
30
+ (default): Linear(in_features=4096, out_features=8, bias=False)
31
+ )
32
+ (lora_B): ModuleDict(
33
+ (default): Linear(in_features=8, out_features=4096, bias=False)
34
+ )
35
+ (lora_embedding_A): ParameterDict()
36
+ (lora_embedding_B): ParameterDict()
37
+ (lora_magnitude_vector): ModuleDict()
38
+ )
39
+ (v_proj): lora.Linear(
40
+ (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
41
+ (lora_dropout): ModuleDict(
42
+ (default): Dropout(p=0.05, inplace=False)
43
+ )
44
+ (lora_A): ModuleDict(
45
+ (default): Linear(in_features=4096, out_features=8, bias=False)
46
+ )
47
+ (lora_B): ModuleDict(
48
+ (default): Linear(in_features=8, out_features=4096, bias=False)
49
+ )
50
+ (lora_embedding_A): ParameterDict()
51
+ (lora_embedding_B): ParameterDict()
52
+ (lora_magnitude_vector): ModuleDict()
53
+ )
54
+ (o_proj): lora.Linear(
55
+ (base_layer): Linear(in_features=4096, out_features=4096, bias=False)
56
+ (lora_dropout): ModuleDict(
57
+ (default): Dropout(p=0.05, inplace=False)
58
+ )
59
+ (lora_A): ModuleDict(
60
+ (default): Linear(in_features=4096, out_features=8, bias=False)
61
+ )
62
+ (lora_B): ModuleDict(
63
+ (default): Linear(in_features=8, out_features=4096, bias=False)
64
+ )
65
+ (lora_embedding_A): ParameterDict()
66
+ (lora_embedding_B): ParameterDict()
67
+ (lora_magnitude_vector): ModuleDict()
68
+ )
69
+ (rotary_emb): LlamaRotaryEmbedding()
70
+ )
71
+ (mlp): LlamaMLP(
72
+ (gate_proj): lora.Linear(
73
+ (base_layer): Linear(in_features=4096, out_features=11008, bias=False)
74
+ (lora_dropout): ModuleDict(
75
+ (default): Dropout(p=0.05, inplace=False)
76
+ )
77
+ (lora_A): ModuleDict(
78
+ (default): Linear(in_features=4096, out_features=8, bias=False)
79
+ )
80
+ (lora_B): ModuleDict(
81
+ (default): Linear(in_features=8, out_features=11008, bias=False)
82
+ )
83
+ (lora_embedding_A): ParameterDict()
84
+ (lora_embedding_B): ParameterDict()
85
+ (lora_magnitude_vector): ModuleDict()
86
+ )
87
+ (up_proj): lora.Linear(
88
+ (base_layer): Linear(in_features=4096, out_features=11008, bias=False)
89
+ (lora_dropout): ModuleDict(
90
+ (default): Dropout(p=0.05, inplace=False)
91
+ )
92
+ (lora_A): ModuleDict(
93
+ (default): Linear(in_features=4096, out_features=8, bias=False)
94
+ )
95
+ (lora_B): ModuleDict(
96
+ (default): Linear(in_features=8, out_features=11008, bias=False)
97
+ )
98
+ (lora_embedding_A): ParameterDict()
99
+ (lora_embedding_B): ParameterDict()
100
+ (lora_magnitude_vector): ModuleDict()
101
+ )
102
+ (down_proj): lora.Linear(
103
+ (base_layer): Linear(in_features=11008, out_features=4096, bias=False)
104
+ (lora_dropout): ModuleDict(
105
+ (default): Dropout(p=0.05, inplace=False)
106
+ )
107
+ (lora_A): ModuleDict(
108
+ (default): Linear(in_features=11008, out_features=8, bias=False)
109
+ )
110
+ (lora_B): ModuleDict(
111
+ (default): Linear(in_features=8, out_features=4096, bias=False)
112
+ )
113
+ (lora_embedding_A): ParameterDict()
114
+ (lora_embedding_B): ParameterDict()
115
+ (lora_magnitude_vector): ModuleDict()
116
+ )
117
+ (act_fn): SiLU()
118
+ )
119
+ (input_layernorm): LlamaRMSNorm()
120
+ (post_attention_layernorm): LlamaRMSNorm()
121
+ )
122
+ )
123
+ (norm): LlamaRMSNorm()
124
+ (visual_encoder): VisualEncoder(
125
+ (vision_tower): CLIPVisionModel(
126
+ (vision_model): CLIPVisionTransformer(
127
+ (embeddings): CLIPVisionEmbeddings(
128
+ (patch_embedding): Conv2d(3, 1024, kernel_size=(14, 14), stride=(14, 14), bias=False)
129
+ (position_embedding): Embedding(257, 1024)
130
+ )
131
+ (pre_layrnorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
132
+ (encoder): CLIPEncoder(
133
+ (layers): ModuleList(
134
+ (0-23): 24 x CLIPEncoderLayer(
135
+ (self_attn): CLIPAttention(
136
+ (k_proj): Linear(in_features=1024, out_features=1024, bias=True)
137
+ (v_proj): Linear(in_features=1024, out_features=1024, bias=True)
138
+ (q_proj): Linear(in_features=1024, out_features=1024, bias=True)
139
+ (out_proj): Linear(in_features=1024, out_features=1024, bias=True)
140
+ )
141
+ (layer_norm1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
142
+ (mlp): CLIPMLP(
143
+ (activation_fn): QuickGELUActivation()
144
+ (fc1): Linear(in_features=1024, out_features=4096, bias=True)
145
+ (fc2): Linear(in_features=4096, out_features=1024, bias=True)
146
+ )
147
+ (layer_norm2): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
148
+ )
149
+ )
150
+ )
151
+ (post_layernorm): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
152
+ )
153
+ )
154
+ )
155
+ (vl_projector): VLProjector(
156
+ (visual_ln): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)
157
+ (visual_Qformer): BertLMHeadModel(
158
+ (bert): BertModel(
159
+ (embeddings): BertEmbeddings(
160
+ (word_embeddings): Embedding(30522, 768, padding_idx=0)
161
+ (position_embeddings): Embedding(512, 768)
162
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
163
+ (dropout): Dropout(p=0.1, inplace=False)
164
+ )
165
+ (encoder): BertEncoder(
166
+ (layer): ModuleList(
167
+ (0-1): 2 x BertLayer(
168
+ (attention): BertAttention(
169
+ (self): BertSelfAttention(
170
+ (query): Linear(in_features=768, out_features=768, bias=True)
171
+ (key): Linear(in_features=768, out_features=768, bias=True)
172
+ (value): Linear(in_features=768, out_features=768, bias=True)
173
+ (dropout): Dropout(p=0.1, inplace=False)
174
+ )
175
+ (output): BertSelfOutput(
176
+ (dense): Linear(in_features=768, out_features=768, bias=True)
177
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
178
+ (dropout): Dropout(p=0.1, inplace=False)
179
+ )
180
+ )
181
+ (crossattention): BertAttention(
182
+ (self): BertSelfAttention(
183
+ (query): Linear(in_features=768, out_features=768, bias=True)
184
+ (key): Linear(in_features=1024, out_features=768, bias=True)
185
+ (value): Linear(in_features=1024, out_features=768, bias=True)
186
+ (dropout): Dropout(p=0.1, inplace=False)
187
+ )
188
+ (output): BertSelfOutput(
189
+ (dense): Linear(in_features=768, out_features=768, bias=True)
190
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
191
+ (dropout): Dropout(p=0.1, inplace=False)
192
+ )
193
+ )
194
+ (intermediate): BertIntermediate(
195
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
196
+ (intermediate_act_fn): GELUActivation()
197
+ )
198
+ (output): BertOutput(
199
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
200
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
201
+ (dropout): Dropout(p=0.1, inplace=False)
202
+ )
203
+ (intermediate_query): BertIntermediate(
204
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
205
+ (intermediate_act_fn): GELUActivation()
206
+ )
207
+ (output_query): BertOutput(
208
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
209
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
210
+ (dropout): Dropout(p=0.1, inplace=False)
211
+ )
212
+ )
213
+ )
214
+ )
215
+ )
216
+ (cls): BertOnlyMLMHead(
217
+ (predictions): BertLMPredictionHead(
218
+ (transform): BertPredictionHeadTransform(
219
+ (dense): Linear(in_features=768, out_features=768, bias=True)
220
+ (transform_act_fn): GELUActivation()
221
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
222
+ )
223
+ (decoder): Linear(in_features=768, out_features=30522, bias=True)
224
+ )
225
+ )
226
+ )
227
+ (visual_proj): Sequential(
228
+ (0): Linear(in_features=768, out_features=4096, bias=True)
229
+ (1): GELU(approximate='none')
230
+ (2): Linear(in_features=4096, out_features=4096, bias=True)
231
+ )
232
+ )
233
+ (audio_encoder): AudioEncoder(
234
+ (audio_encoder): BEATs(
235
+ (post_extract_proj): Linear(in_features=512, out_features=768, bias=True)
236
+ (patch_embedding): Conv2d(1, 512, kernel_size=(16, 16), stride=(16, 16), bias=False)
237
+ (dropout_input): Dropout(p=0.0, inplace=False)
238
+ (encoder): TransformerEncoder(
239
+ (pos_conv): Sequential(
240
+ (0): Conv1d(768, 768, kernel_size=(128,), stride=(1,), padding=(64,), groups=16)
241
+ (1): SamePad()
242
+ (2): GELU(approximate='none')
243
+ )
244
+ (layers): ModuleList(
245
+ (0): TransformerSentenceEncoderLayer(
246
+ (self_attn): MultiheadAttention(
247
+ (dropout_module): Dropout(p=0.0, inplace=False)
248
+ (relative_attention_bias): Embedding(320, 12)
249
+ (k_proj): Linear(in_features=768, out_features=768, bias=True)
250
+ (v_proj): Linear(in_features=768, out_features=768, bias=True)
251
+ (q_proj): Linear(in_features=768, out_features=768, bias=True)
252
+ (out_proj): Linear(in_features=768, out_features=768, bias=True)
253
+ (grep_linear): Linear(in_features=64, out_features=8, bias=True)
254
+ )
255
+ (dropout1): Dropout(p=0.0, inplace=False)
256
+ (dropout2): Dropout(p=0.0, inplace=False)
257
+ (dropout3): Dropout(p=0.0, inplace=False)
258
+ (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
259
+ (fc1): Linear(in_features=768, out_features=3072, bias=True)
260
+ (fc2): Linear(in_features=3072, out_features=768, bias=True)
261
+ (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
262
+ )
263
+ (1-11): 11 x TransformerSentenceEncoderLayer(
264
+ (self_attn): MultiheadAttention(
265
+ (dropout_module): Dropout(p=0.0, inplace=False)
266
+ (k_proj): Linear(in_features=768, out_features=768, bias=True)
267
+ (v_proj): Linear(in_features=768, out_features=768, bias=True)
268
+ (q_proj): Linear(in_features=768, out_features=768, bias=True)
269
+ (out_proj): Linear(in_features=768, out_features=768, bias=True)
270
+ (grep_linear): Linear(in_features=64, out_features=8, bias=True)
271
+ (relative_attention_bias): Embedding(320, 12)
272
+ )
273
+ (dropout1): Dropout(p=0.0, inplace=False)
274
+ (dropout2): Dropout(p=0.0, inplace=False)
275
+ (dropout3): Dropout(p=0.0, inplace=False)
276
+ (self_attn_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
277
+ (fc1): Linear(in_features=768, out_features=3072, bias=True)
278
+ (fc2): Linear(in_features=3072, out_features=768, bias=True)
279
+ (final_layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
280
+ )
281
+ )
282
+ (layer_norm): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
283
+ )
284
+ (layer_norm): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
285
+ (predictor_dropout): Dropout(p=0.0, inplace=False)
286
+ (predictor): Linear(in_features=768, out_features=527, bias=True)
287
+ )
288
+ )
289
+ (al_projector): ALProjector(
290
+ (audio_ln): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
291
+ (audio_Qformer): BertLMHeadModel(
292
+ (bert): BertModel(
293
+ (embeddings): BertEmbeddings(
294
+ (word_embeddings): Embedding(30522, 768, padding_idx=0)
295
+ (position_embeddings): Embedding(512, 768)
296
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
297
+ (dropout): Dropout(p=0.1, inplace=False)
298
+ )
299
+ (encoder): BertEncoder(
300
+ (layer): ModuleList(
301
+ (0-1): 2 x BertLayer(
302
+ (attention): BertAttention(
303
+ (self): BertSelfAttention(
304
+ (query): Linear(in_features=768, out_features=768, bias=True)
305
+ (key): Linear(in_features=768, out_features=768, bias=True)
306
+ (value): Linear(in_features=768, out_features=768, bias=True)
307
+ (dropout): Dropout(p=0.1, inplace=False)
308
+ )
309
+ (output): BertSelfOutput(
310
+ (dense): Linear(in_features=768, out_features=768, bias=True)
311
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
312
+ (dropout): Dropout(p=0.1, inplace=False)
313
+ )
314
+ )
315
+ (crossattention): BertAttention(
316
+ (self): BertSelfAttention(
317
+ (query): Linear(in_features=768, out_features=768, bias=True)
318
+ (key): Linear(in_features=768, out_features=768, bias=True)
319
+ (value): Linear(in_features=768, out_features=768, bias=True)
320
+ (dropout): Dropout(p=0.1, inplace=False)
321
+ )
322
+ (output): BertSelfOutput(
323
+ (dense): Linear(in_features=768, out_features=768, bias=True)
324
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
325
+ (dropout): Dropout(p=0.1, inplace=False)
326
+ )
327
+ )
328
+ (intermediate): BertIntermediate(
329
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
330
+ (intermediate_act_fn): GELUActivation()
331
+ )
332
+ (output): BertOutput(
333
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
334
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
335
+ (dropout): Dropout(p=0.1, inplace=False)
336
+ )
337
+ (intermediate_query): BertIntermediate(
338
+ (dense): Linear(in_features=768, out_features=3072, bias=True)
339
+ (intermediate_act_fn): GELUActivation()
340
+ )
341
+ (output_query): BertOutput(
342
+ (dense): Linear(in_features=3072, out_features=768, bias=True)
343
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
344
+ (dropout): Dropout(p=0.1, inplace=False)
345
+ )
346
+ )
347
+ )
348
+ )
349
+ )
350
+ (cls): BertOnlyMLMHead(
351
+ (predictions): BertLMPredictionHead(
352
+ (transform): BertPredictionHeadTransform(
353
+ (dense): Linear(in_features=768, out_features=768, bias=True)
354
+ (transform_act_fn): GELUActivation()
355
+ (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
356
+ )
357
+ (decoder): Linear(in_features=768, out_features=30522, bias=True)
358
+ )
359
+ )
360
+ )
361
+ (audio_proj): Sequential(
362
+ (0): Linear(in_features=768, out_features=4096, bias=True)
363
+ (1): GELU(approximate='none')
364
+ (2): Linear(in_features=4096, out_features=4096, bias=True)
365
+ )
366
+ )
367
+ )
368
+ (lm_head): Linear(in_features=4096, out_features=32021, bias=False)
369
+ )
370
+ )
371
+ )
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/model_trainable_params.txt ADDED
@@ -0,0 +1,612 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ base_model.model.model.embed_tokens.weight torch.Size([32021, 4096])
3
+ base_model.model.model.layers.0.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
4
+ base_model.model.model.layers.0.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
5
+ base_model.model.model.layers.0.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
6
+ base_model.model.model.layers.0.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
7
+ base_model.model.model.layers.0.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
8
+ base_model.model.model.layers.0.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
9
+ base_model.model.model.layers.0.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
10
+ base_model.model.model.layers.0.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
11
+ base_model.model.model.layers.0.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
12
+ base_model.model.model.layers.0.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
13
+ base_model.model.model.layers.0.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
14
+ base_model.model.model.layers.0.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
15
+ base_model.model.model.layers.0.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
16
+ base_model.model.model.layers.0.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
17
+ base_model.model.model.layers.1.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
18
+ base_model.model.model.layers.1.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
19
+ base_model.model.model.layers.1.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
20
+ base_model.model.model.layers.1.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
21
+ base_model.model.model.layers.1.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
22
+ base_model.model.model.layers.1.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
23
+ base_model.model.model.layers.1.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
24
+ base_model.model.model.layers.1.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
25
+ base_model.model.model.layers.1.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
26
+ base_model.model.model.layers.1.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
27
+ base_model.model.model.layers.1.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
28
+ base_model.model.model.layers.1.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
29
+ base_model.model.model.layers.1.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
30
+ base_model.model.model.layers.1.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
31
+ base_model.model.model.layers.2.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
32
+ base_model.model.model.layers.2.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
33
+ base_model.model.model.layers.2.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
34
+ base_model.model.model.layers.2.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
35
+ base_model.model.model.layers.2.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
36
+ base_model.model.model.layers.2.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
37
+ base_model.model.model.layers.2.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
38
+ base_model.model.model.layers.2.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
39
+ base_model.model.model.layers.2.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
40
+ base_model.model.model.layers.2.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
41
+ base_model.model.model.layers.2.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
42
+ base_model.model.model.layers.2.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
43
+ base_model.model.model.layers.2.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
44
+ base_model.model.model.layers.2.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
45
+ base_model.model.model.layers.3.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
46
+ base_model.model.model.layers.3.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
47
+ base_model.model.model.layers.3.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
48
+ base_model.model.model.layers.3.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
49
+ base_model.model.model.layers.3.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
50
+ base_model.model.model.layers.3.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
51
+ base_model.model.model.layers.3.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
52
+ base_model.model.model.layers.3.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
53
+ base_model.model.model.layers.3.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
54
+ base_model.model.model.layers.3.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
55
+ base_model.model.model.layers.3.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
56
+ base_model.model.model.layers.3.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
57
+ base_model.model.model.layers.3.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
58
+ base_model.model.model.layers.3.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
59
+ base_model.model.model.layers.4.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
60
+ base_model.model.model.layers.4.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
61
+ base_model.model.model.layers.4.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
62
+ base_model.model.model.layers.4.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
63
+ base_model.model.model.layers.4.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
64
+ base_model.model.model.layers.4.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
65
+ base_model.model.model.layers.4.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
66
+ base_model.model.model.layers.4.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
67
+ base_model.model.model.layers.4.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
68
+ base_model.model.model.layers.4.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
69
+ base_model.model.model.layers.4.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
70
+ base_model.model.model.layers.4.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
71
+ base_model.model.model.layers.4.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
72
+ base_model.model.model.layers.4.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
73
+ base_model.model.model.layers.5.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
74
+ base_model.model.model.layers.5.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
75
+ base_model.model.model.layers.5.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
76
+ base_model.model.model.layers.5.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
77
+ base_model.model.model.layers.5.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
78
+ base_model.model.model.layers.5.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
79
+ base_model.model.model.layers.5.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
80
+ base_model.model.model.layers.5.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
81
+ base_model.model.model.layers.5.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
82
+ base_model.model.model.layers.5.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
83
+ base_model.model.model.layers.5.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
84
+ base_model.model.model.layers.5.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
85
+ base_model.model.model.layers.5.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
86
+ base_model.model.model.layers.5.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
87
+ base_model.model.model.layers.6.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
88
+ base_model.model.model.layers.6.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
89
+ base_model.model.model.layers.6.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
90
+ base_model.model.model.layers.6.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
91
+ base_model.model.model.layers.6.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
92
+ base_model.model.model.layers.6.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
93
+ base_model.model.model.layers.6.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
94
+ base_model.model.model.layers.6.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
95
+ base_model.model.model.layers.6.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
96
+ base_model.model.model.layers.6.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
97
+ base_model.model.model.layers.6.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
98
+ base_model.model.model.layers.6.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
99
+ base_model.model.model.layers.6.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
100
+ base_model.model.model.layers.6.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
101
+ base_model.model.model.layers.7.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
102
+ base_model.model.model.layers.7.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
103
+ base_model.model.model.layers.7.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
104
+ base_model.model.model.layers.7.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
105
+ base_model.model.model.layers.7.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
106
+ base_model.model.model.layers.7.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
107
+ base_model.model.model.layers.7.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
108
+ base_model.model.model.layers.7.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
109
+ base_model.model.model.layers.7.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
110
+ base_model.model.model.layers.7.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
111
+ base_model.model.model.layers.7.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
112
+ base_model.model.model.layers.7.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
113
+ base_model.model.model.layers.7.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
114
+ base_model.model.model.layers.7.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
115
+ base_model.model.model.layers.8.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
116
+ base_model.model.model.layers.8.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
117
+ base_model.model.model.layers.8.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
118
+ base_model.model.model.layers.8.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
119
+ base_model.model.model.layers.8.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
120
+ base_model.model.model.layers.8.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
121
+ base_model.model.model.layers.8.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
122
+ base_model.model.model.layers.8.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
123
+ base_model.model.model.layers.8.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
124
+ base_model.model.model.layers.8.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
125
+ base_model.model.model.layers.8.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
126
+ base_model.model.model.layers.8.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
127
+ base_model.model.model.layers.8.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
128
+ base_model.model.model.layers.8.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
129
+ base_model.model.model.layers.9.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
130
+ base_model.model.model.layers.9.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
131
+ base_model.model.model.layers.9.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
132
+ base_model.model.model.layers.9.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
133
+ base_model.model.model.layers.9.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
134
+ base_model.model.model.layers.9.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
135
+ base_model.model.model.layers.9.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
136
+ base_model.model.model.layers.9.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
137
+ base_model.model.model.layers.9.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
138
+ base_model.model.model.layers.9.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
139
+ base_model.model.model.layers.9.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
140
+ base_model.model.model.layers.9.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
141
+ base_model.model.model.layers.9.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
142
+ base_model.model.model.layers.9.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
143
+ base_model.model.model.layers.10.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
144
+ base_model.model.model.layers.10.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
145
+ base_model.model.model.layers.10.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
146
+ base_model.model.model.layers.10.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
147
+ base_model.model.model.layers.10.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
148
+ base_model.model.model.layers.10.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
149
+ base_model.model.model.layers.10.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
150
+ base_model.model.model.layers.10.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
151
+ base_model.model.model.layers.10.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
152
+ base_model.model.model.layers.10.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
153
+ base_model.model.model.layers.10.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
154
+ base_model.model.model.layers.10.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
155
+ base_model.model.model.layers.10.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
156
+ base_model.model.model.layers.10.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
157
+ base_model.model.model.layers.11.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
158
+ base_model.model.model.layers.11.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
159
+ base_model.model.model.layers.11.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
160
+ base_model.model.model.layers.11.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
161
+ base_model.model.model.layers.11.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
162
+ base_model.model.model.layers.11.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
163
+ base_model.model.model.layers.11.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
164
+ base_model.model.model.layers.11.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
165
+ base_model.model.model.layers.11.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
166
+ base_model.model.model.layers.11.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
167
+ base_model.model.model.layers.11.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
168
+ base_model.model.model.layers.11.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
169
+ base_model.model.model.layers.11.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
170
+ base_model.model.model.layers.11.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
171
+ base_model.model.model.layers.12.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
172
+ base_model.model.model.layers.12.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
173
+ base_model.model.model.layers.12.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
174
+ base_model.model.model.layers.12.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
175
+ base_model.model.model.layers.12.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
176
+ base_model.model.model.layers.12.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
177
+ base_model.model.model.layers.12.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
178
+ base_model.model.model.layers.12.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
179
+ base_model.model.model.layers.12.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
180
+ base_model.model.model.layers.12.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
181
+ base_model.model.model.layers.12.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
182
+ base_model.model.model.layers.12.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
183
+ base_model.model.model.layers.12.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
184
+ base_model.model.model.layers.12.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
185
+ base_model.model.model.layers.13.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
186
+ base_model.model.model.layers.13.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
187
+ base_model.model.model.layers.13.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
188
+ base_model.model.model.layers.13.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
189
+ base_model.model.model.layers.13.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
190
+ base_model.model.model.layers.13.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
191
+ base_model.model.model.layers.13.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
192
+ base_model.model.model.layers.13.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
193
+ base_model.model.model.layers.13.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
194
+ base_model.model.model.layers.13.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
195
+ base_model.model.model.layers.13.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
196
+ base_model.model.model.layers.13.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
197
+ base_model.model.model.layers.13.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
198
+ base_model.model.model.layers.13.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
199
+ base_model.model.model.layers.14.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
200
+ base_model.model.model.layers.14.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
201
+ base_model.model.model.layers.14.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
202
+ base_model.model.model.layers.14.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
203
+ base_model.model.model.layers.14.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
204
+ base_model.model.model.layers.14.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
205
+ base_model.model.model.layers.14.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
206
+ base_model.model.model.layers.14.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
207
+ base_model.model.model.layers.14.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
208
+ base_model.model.model.layers.14.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
209
+ base_model.model.model.layers.14.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
210
+ base_model.model.model.layers.14.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
211
+ base_model.model.model.layers.14.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
212
+ base_model.model.model.layers.14.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
213
+ base_model.model.model.layers.15.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
214
+ base_model.model.model.layers.15.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
215
+ base_model.model.model.layers.15.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
216
+ base_model.model.model.layers.15.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
217
+ base_model.model.model.layers.15.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
218
+ base_model.model.model.layers.15.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
219
+ base_model.model.model.layers.15.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
220
+ base_model.model.model.layers.15.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
221
+ base_model.model.model.layers.15.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
222
+ base_model.model.model.layers.15.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
223
+ base_model.model.model.layers.15.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
224
+ base_model.model.model.layers.15.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
225
+ base_model.model.model.layers.15.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
226
+ base_model.model.model.layers.15.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
227
+ base_model.model.model.layers.16.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
228
+ base_model.model.model.layers.16.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
229
+ base_model.model.model.layers.16.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
230
+ base_model.model.model.layers.16.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
231
+ base_model.model.model.layers.16.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
232
+ base_model.model.model.layers.16.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
233
+ base_model.model.model.layers.16.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
234
+ base_model.model.model.layers.16.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
235
+ base_model.model.model.layers.16.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
236
+ base_model.model.model.layers.16.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
237
+ base_model.model.model.layers.16.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
238
+ base_model.model.model.layers.16.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
239
+ base_model.model.model.layers.16.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
240
+ base_model.model.model.layers.16.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
241
+ base_model.model.model.layers.17.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
242
+ base_model.model.model.layers.17.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
243
+ base_model.model.model.layers.17.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
244
+ base_model.model.model.layers.17.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
245
+ base_model.model.model.layers.17.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
246
+ base_model.model.model.layers.17.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
247
+ base_model.model.model.layers.17.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
248
+ base_model.model.model.layers.17.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
249
+ base_model.model.model.layers.17.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
250
+ base_model.model.model.layers.17.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
251
+ base_model.model.model.layers.17.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
252
+ base_model.model.model.layers.17.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
253
+ base_model.model.model.layers.17.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
254
+ base_model.model.model.layers.17.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
255
+ base_model.model.model.layers.18.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
256
+ base_model.model.model.layers.18.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
257
+ base_model.model.model.layers.18.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
258
+ base_model.model.model.layers.18.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
259
+ base_model.model.model.layers.18.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
260
+ base_model.model.model.layers.18.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
261
+ base_model.model.model.layers.18.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
262
+ base_model.model.model.layers.18.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
263
+ base_model.model.model.layers.18.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
264
+ base_model.model.model.layers.18.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
265
+ base_model.model.model.layers.18.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
266
+ base_model.model.model.layers.18.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
267
+ base_model.model.model.layers.18.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
268
+ base_model.model.model.layers.18.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
269
+ base_model.model.model.layers.19.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
270
+ base_model.model.model.layers.19.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
271
+ base_model.model.model.layers.19.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
272
+ base_model.model.model.layers.19.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
273
+ base_model.model.model.layers.19.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
274
+ base_model.model.model.layers.19.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
275
+ base_model.model.model.layers.19.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
276
+ base_model.model.model.layers.19.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
277
+ base_model.model.model.layers.19.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
278
+ base_model.model.model.layers.19.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
279
+ base_model.model.model.layers.19.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
280
+ base_model.model.model.layers.19.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
281
+ base_model.model.model.layers.19.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
282
+ base_model.model.model.layers.19.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
283
+ base_model.model.model.layers.20.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
284
+ base_model.model.model.layers.20.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
285
+ base_model.model.model.layers.20.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
286
+ base_model.model.model.layers.20.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
287
+ base_model.model.model.layers.20.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
288
+ base_model.model.model.layers.20.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
289
+ base_model.model.model.layers.20.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
290
+ base_model.model.model.layers.20.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
291
+ base_model.model.model.layers.20.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
292
+ base_model.model.model.layers.20.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
293
+ base_model.model.model.layers.20.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
294
+ base_model.model.model.layers.20.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
295
+ base_model.model.model.layers.20.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
296
+ base_model.model.model.layers.20.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
297
+ base_model.model.model.layers.21.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
298
+ base_model.model.model.layers.21.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
299
+ base_model.model.model.layers.21.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
300
+ base_model.model.model.layers.21.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
301
+ base_model.model.model.layers.21.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
302
+ base_model.model.model.layers.21.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
303
+ base_model.model.model.layers.21.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
304
+ base_model.model.model.layers.21.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
305
+ base_model.model.model.layers.21.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
306
+ base_model.model.model.layers.21.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
307
+ base_model.model.model.layers.21.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
308
+ base_model.model.model.layers.21.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
309
+ base_model.model.model.layers.21.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
310
+ base_model.model.model.layers.21.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
311
+ base_model.model.model.layers.22.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
312
+ base_model.model.model.layers.22.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
313
+ base_model.model.model.layers.22.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
314
+ base_model.model.model.layers.22.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
315
+ base_model.model.model.layers.22.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
316
+ base_model.model.model.layers.22.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
317
+ base_model.model.model.layers.22.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
318
+ base_model.model.model.layers.22.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
319
+ base_model.model.model.layers.22.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
320
+ base_model.model.model.layers.22.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
321
+ base_model.model.model.layers.22.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
322
+ base_model.model.model.layers.22.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
323
+ base_model.model.model.layers.22.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
324
+ base_model.model.model.layers.22.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
325
+ base_model.model.model.layers.23.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
326
+ base_model.model.model.layers.23.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
327
+ base_model.model.model.layers.23.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
328
+ base_model.model.model.layers.23.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
329
+ base_model.model.model.layers.23.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
330
+ base_model.model.model.layers.23.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
331
+ base_model.model.model.layers.23.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
332
+ base_model.model.model.layers.23.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
333
+ base_model.model.model.layers.23.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
334
+ base_model.model.model.layers.23.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
335
+ base_model.model.model.layers.23.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
336
+ base_model.model.model.layers.23.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
337
+ base_model.model.model.layers.23.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
338
+ base_model.model.model.layers.23.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
339
+ base_model.model.model.layers.24.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
340
+ base_model.model.model.layers.24.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
341
+ base_model.model.model.layers.24.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
342
+ base_model.model.model.layers.24.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
343
+ base_model.model.model.layers.24.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
344
+ base_model.model.model.layers.24.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
345
+ base_model.model.model.layers.24.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
346
+ base_model.model.model.layers.24.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
347
+ base_model.model.model.layers.24.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
348
+ base_model.model.model.layers.24.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
349
+ base_model.model.model.layers.24.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
350
+ base_model.model.model.layers.24.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
351
+ base_model.model.model.layers.24.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
352
+ base_model.model.model.layers.24.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
353
+ base_model.model.model.layers.25.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
354
+ base_model.model.model.layers.25.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
355
+ base_model.model.model.layers.25.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
356
+ base_model.model.model.layers.25.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
357
+ base_model.model.model.layers.25.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
358
+ base_model.model.model.layers.25.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
359
+ base_model.model.model.layers.25.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
360
+ base_model.model.model.layers.25.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
361
+ base_model.model.model.layers.25.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
362
+ base_model.model.model.layers.25.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
363
+ base_model.model.model.layers.25.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
364
+ base_model.model.model.layers.25.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
365
+ base_model.model.model.layers.25.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
366
+ base_model.model.model.layers.25.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
367
+ base_model.model.model.layers.26.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
368
+ base_model.model.model.layers.26.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
369
+ base_model.model.model.layers.26.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
370
+ base_model.model.model.layers.26.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
371
+ base_model.model.model.layers.26.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
372
+ base_model.model.model.layers.26.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
373
+ base_model.model.model.layers.26.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
374
+ base_model.model.model.layers.26.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
375
+ base_model.model.model.layers.26.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
376
+ base_model.model.model.layers.26.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
377
+ base_model.model.model.layers.26.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
378
+ base_model.model.model.layers.26.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
379
+ base_model.model.model.layers.26.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
380
+ base_model.model.model.layers.26.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
381
+ base_model.model.model.layers.27.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
382
+ base_model.model.model.layers.27.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
383
+ base_model.model.model.layers.27.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
384
+ base_model.model.model.layers.27.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
385
+ base_model.model.model.layers.27.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
386
+ base_model.model.model.layers.27.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
387
+ base_model.model.model.layers.27.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
388
+ base_model.model.model.layers.27.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
389
+ base_model.model.model.layers.27.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
390
+ base_model.model.model.layers.27.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
391
+ base_model.model.model.layers.27.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
392
+ base_model.model.model.layers.27.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
393
+ base_model.model.model.layers.27.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
394
+ base_model.model.model.layers.27.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
395
+ base_model.model.model.layers.28.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
396
+ base_model.model.model.layers.28.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
397
+ base_model.model.model.layers.28.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
398
+ base_model.model.model.layers.28.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
399
+ base_model.model.model.layers.28.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
400
+ base_model.model.model.layers.28.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
401
+ base_model.model.model.layers.28.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
402
+ base_model.model.model.layers.28.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
403
+ base_model.model.model.layers.28.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
404
+ base_model.model.model.layers.28.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
405
+ base_model.model.model.layers.28.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
406
+ base_model.model.model.layers.28.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
407
+ base_model.model.model.layers.28.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
408
+ base_model.model.model.layers.28.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
409
+ base_model.model.model.layers.29.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
410
+ base_model.model.model.layers.29.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
411
+ base_model.model.model.layers.29.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
412
+ base_model.model.model.layers.29.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
413
+ base_model.model.model.layers.29.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
414
+ base_model.model.model.layers.29.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
415
+ base_model.model.model.layers.29.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
416
+ base_model.model.model.layers.29.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
417
+ base_model.model.model.layers.29.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
418
+ base_model.model.model.layers.29.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
419
+ base_model.model.model.layers.29.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
420
+ base_model.model.model.layers.29.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
421
+ base_model.model.model.layers.29.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
422
+ base_model.model.model.layers.29.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
423
+ base_model.model.model.layers.30.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
424
+ base_model.model.model.layers.30.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
425
+ base_model.model.model.layers.30.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
426
+ base_model.model.model.layers.30.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
427
+ base_model.model.model.layers.30.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
428
+ base_model.model.model.layers.30.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
429
+ base_model.model.model.layers.30.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
430
+ base_model.model.model.layers.30.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
431
+ base_model.model.model.layers.30.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
432
+ base_model.model.model.layers.30.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
433
+ base_model.model.model.layers.30.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
434
+ base_model.model.model.layers.30.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
435
+ base_model.model.model.layers.30.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
436
+ base_model.model.model.layers.30.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
437
+ base_model.model.model.layers.31.self_attn.q_proj.lora_A.default.weight torch.Size([8, 4096])
438
+ base_model.model.model.layers.31.self_attn.q_proj.lora_B.default.weight torch.Size([4096, 8])
439
+ base_model.model.model.layers.31.self_attn.k_proj.lora_A.default.weight torch.Size([8, 4096])
440
+ base_model.model.model.layers.31.self_attn.k_proj.lora_B.default.weight torch.Size([4096, 8])
441
+ base_model.model.model.layers.31.self_attn.v_proj.lora_A.default.weight torch.Size([8, 4096])
442
+ base_model.model.model.layers.31.self_attn.v_proj.lora_B.default.weight torch.Size([4096, 8])
443
+ base_model.model.model.layers.31.self_attn.o_proj.lora_A.default.weight torch.Size([8, 4096])
444
+ base_model.model.model.layers.31.self_attn.o_proj.lora_B.default.weight torch.Size([4096, 8])
445
+ base_model.model.model.layers.31.mlp.gate_proj.lora_A.default.weight torch.Size([8, 4096])
446
+ base_model.model.model.layers.31.mlp.gate_proj.lora_B.default.weight torch.Size([11008, 8])
447
+ base_model.model.model.layers.31.mlp.up_proj.lora_A.default.weight torch.Size([8, 4096])
448
+ base_model.model.model.layers.31.mlp.up_proj.lora_B.default.weight torch.Size([11008, 8])
449
+ base_model.model.model.layers.31.mlp.down_proj.lora_A.default.weight torch.Size([8, 11008])
450
+ base_model.model.model.layers.31.mlp.down_proj.lora_B.default.weight torch.Size([4096, 8])
451
+ base_model.model.model.vl_projector.visual_query_tokens torch.Size([1, 32, 768])
452
+ base_model.model.model.vl_projector.visual_ln.weight torch.Size([1024])
453
+ base_model.model.model.vl_projector.visual_ln.bias torch.Size([1024])
454
+ base_model.model.model.vl_projector.visual_Qformer.bert.embeddings.word_embeddings.weight torch.Size([30522, 768])
455
+ base_model.model.model.vl_projector.visual_Qformer.bert.embeddings.position_embeddings.weight torch.Size([512, 768])
456
+ base_model.model.model.vl_projector.visual_Qformer.bert.embeddings.LayerNorm.weight torch.Size([768])
457
+ base_model.model.model.vl_projector.visual_Qformer.bert.embeddings.LayerNorm.bias torch.Size([768])
458
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.query.weight torch.Size([768, 768])
459
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.query.bias torch.Size([768])
460
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.key.weight torch.Size([768, 768])
461
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.key.bias torch.Size([768])
462
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.value.weight torch.Size([768, 768])
463
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.self.value.bias torch.Size([768])
464
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.output.dense.weight torch.Size([768, 768])
465
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.output.dense.bias torch.Size([768])
466
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.output.LayerNorm.weight torch.Size([768])
467
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.attention.output.LayerNorm.bias torch.Size([768])
468
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.query.weight torch.Size([768, 768])
469
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.query.bias torch.Size([768])
470
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.key.weight torch.Size([768, 1024])
471
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.key.bias torch.Size([768])
472
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.value.weight torch.Size([768, 1024])
473
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.self.value.bias torch.Size([768])
474
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.output.dense.weight torch.Size([768, 768])
475
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.output.dense.bias torch.Size([768])
476
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.weight torch.Size([768])
477
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.bias torch.Size([768])
478
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.intermediate.dense.weight torch.Size([3072, 768])
479
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.intermediate.dense.bias torch.Size([3072])
480
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output.dense.weight torch.Size([768, 3072])
481
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output.dense.bias torch.Size([768])
482
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output.LayerNorm.weight torch.Size([768])
483
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output.LayerNorm.bias torch.Size([768])
484
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.intermediate_query.dense.weight torch.Size([3072, 768])
485
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.intermediate_query.dense.bias torch.Size([3072])
486
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output_query.dense.weight torch.Size([768, 3072])
487
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output_query.dense.bias torch.Size([768])
488
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output_query.LayerNorm.weight torch.Size([768])
489
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.0.output_query.LayerNorm.bias torch.Size([768])
490
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.query.weight torch.Size([768, 768])
491
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.query.bias torch.Size([768])
492
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.key.weight torch.Size([768, 768])
493
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.key.bias torch.Size([768])
494
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.value.weight torch.Size([768, 768])
495
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.self.value.bias torch.Size([768])
496
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.output.dense.weight torch.Size([768, 768])
497
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.output.dense.bias torch.Size([768])
498
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.output.LayerNorm.weight torch.Size([768])
499
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.attention.output.LayerNorm.bias torch.Size([768])
500
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.query.weight torch.Size([768, 768])
501
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.query.bias torch.Size([768])
502
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.key.weight torch.Size([768, 1024])
503
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.key.bias torch.Size([768])
504
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.value.weight torch.Size([768, 1024])
505
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.self.value.bias torch.Size([768])
506
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.output.dense.weight torch.Size([768, 768])
507
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.output.dense.bias torch.Size([768])
508
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.weight torch.Size([768])
509
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.bias torch.Size([768])
510
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.intermediate.dense.weight torch.Size([3072, 768])
511
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.intermediate.dense.bias torch.Size([3072])
512
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output.dense.weight torch.Size([768, 3072])
513
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output.dense.bias torch.Size([768])
514
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output.LayerNorm.weight torch.Size([768])
515
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output.LayerNorm.bias torch.Size([768])
516
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.intermediate_query.dense.weight torch.Size([3072, 768])
517
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.intermediate_query.dense.bias torch.Size([3072])
518
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output_query.dense.weight torch.Size([768, 3072])
519
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output_query.dense.bias torch.Size([768])
520
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output_query.LayerNorm.weight torch.Size([768])
521
+ base_model.model.model.vl_projector.visual_Qformer.bert.encoder.layer.1.output_query.LayerNorm.bias torch.Size([768])
522
+ base_model.model.model.vl_projector.visual_Qformer.cls.predictions.bias torch.Size([30522])
523
+ base_model.model.model.vl_projector.visual_Qformer.cls.predictions.transform.dense.weight torch.Size([768, 768])
524
+ base_model.model.model.vl_projector.visual_Qformer.cls.predictions.transform.dense.bias torch.Size([768])
525
+ base_model.model.model.vl_projector.visual_Qformer.cls.predictions.transform.LayerNorm.weight torch.Size([768])
526
+ base_model.model.model.vl_projector.visual_Qformer.cls.predictions.transform.LayerNorm.bias torch.Size([768])
527
+ base_model.model.model.vl_projector.visual_proj.0.weight torch.Size([4096, 768])
528
+ base_model.model.model.vl_projector.visual_proj.0.bias torch.Size([4096])
529
+ base_model.model.model.vl_projector.visual_proj.2.weight torch.Size([4096, 4096])
530
+ base_model.model.model.vl_projector.visual_proj.2.bias torch.Size([4096])
531
+ base_model.model.model.al_projector.audio_query_tokens torch.Size([1, 32, 768])
532
+ base_model.model.model.al_projector.audio_ln.weight torch.Size([768])
533
+ base_model.model.model.al_projector.audio_ln.bias torch.Size([768])
534
+ base_model.model.model.al_projector.audio_Qformer.bert.embeddings.word_embeddings.weight torch.Size([30522, 768])
535
+ base_model.model.model.al_projector.audio_Qformer.bert.embeddings.position_embeddings.weight torch.Size([512, 768])
536
+ base_model.model.model.al_projector.audio_Qformer.bert.embeddings.LayerNorm.weight torch.Size([768])
537
+ base_model.model.model.al_projector.audio_Qformer.bert.embeddings.LayerNorm.bias torch.Size([768])
538
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.query.weight torch.Size([768, 768])
539
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.query.bias torch.Size([768])
540
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.key.weight torch.Size([768, 768])
541
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.key.bias torch.Size([768])
542
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.value.weight torch.Size([768, 768])
543
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.self.value.bias torch.Size([768])
544
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.output.dense.weight torch.Size([768, 768])
545
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.output.dense.bias torch.Size([768])
546
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.output.LayerNorm.weight torch.Size([768])
547
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.attention.output.LayerNorm.bias torch.Size([768])
548
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.query.weight torch.Size([768, 768])
549
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.query.bias torch.Size([768])
550
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.key.weight torch.Size([768, 768])
551
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.key.bias torch.Size([768])
552
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.value.weight torch.Size([768, 768])
553
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.self.value.bias torch.Size([768])
554
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.output.dense.weight torch.Size([768, 768])
555
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.output.dense.bias torch.Size([768])
556
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.weight torch.Size([768])
557
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.crossattention.output.LayerNorm.bias torch.Size([768])
558
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.intermediate.dense.weight torch.Size([3072, 768])
559
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.intermediate.dense.bias torch.Size([3072])
560
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output.dense.weight torch.Size([768, 3072])
561
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output.dense.bias torch.Size([768])
562
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output.LayerNorm.weight torch.Size([768])
563
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output.LayerNorm.bias torch.Size([768])
564
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.intermediate_query.dense.weight torch.Size([3072, 768])
565
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.intermediate_query.dense.bias torch.Size([3072])
566
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output_query.dense.weight torch.Size([768, 3072])
567
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output_query.dense.bias torch.Size([768])
568
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output_query.LayerNorm.weight torch.Size([768])
569
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.0.output_query.LayerNorm.bias torch.Size([768])
570
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.query.weight torch.Size([768, 768])
571
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.query.bias torch.Size([768])
572
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.key.weight torch.Size([768, 768])
573
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.key.bias torch.Size([768])
574
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.value.weight torch.Size([768, 768])
575
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.self.value.bias torch.Size([768])
576
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.output.dense.weight torch.Size([768, 768])
577
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.output.dense.bias torch.Size([768])
578
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.output.LayerNorm.weight torch.Size([768])
579
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.attention.output.LayerNorm.bias torch.Size([768])
580
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.query.weight torch.Size([768, 768])
581
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.query.bias torch.Size([768])
582
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.key.weight torch.Size([768, 768])
583
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.key.bias torch.Size([768])
584
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.value.weight torch.Size([768, 768])
585
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.self.value.bias torch.Size([768])
586
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.output.dense.weight torch.Size([768, 768])
587
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.output.dense.bias torch.Size([768])
588
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.weight torch.Size([768])
589
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.crossattention.output.LayerNorm.bias torch.Size([768])
590
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.intermediate.dense.weight torch.Size([3072, 768])
591
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.intermediate.dense.bias torch.Size([3072])
592
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output.dense.weight torch.Size([768, 3072])
593
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output.dense.bias torch.Size([768])
594
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output.LayerNorm.weight torch.Size([768])
595
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output.LayerNorm.bias torch.Size([768])
596
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.intermediate_query.dense.weight torch.Size([3072, 768])
597
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.intermediate_query.dense.bias torch.Size([3072])
598
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output_query.dense.weight torch.Size([768, 3072])
599
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output_query.dense.bias torch.Size([768])
600
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output_query.LayerNorm.weight torch.Size([768])
601
+ base_model.model.model.al_projector.audio_Qformer.bert.encoder.layer.1.output_query.LayerNorm.bias torch.Size([768])
602
+ base_model.model.model.al_projector.audio_Qformer.cls.predictions.bias torch.Size([30522])
603
+ base_model.model.model.al_projector.audio_Qformer.cls.predictions.transform.dense.weight torch.Size([768, 768])
604
+ base_model.model.model.al_projector.audio_Qformer.cls.predictions.transform.dense.bias torch.Size([768])
605
+ base_model.model.model.al_projector.audio_Qformer.cls.predictions.transform.LayerNorm.weight torch.Size([768])
606
+ base_model.model.model.al_projector.audio_Qformer.cls.predictions.transform.LayerNorm.bias torch.Size([768])
607
+ base_model.model.model.al_projector.audio_proj.0.weight torch.Size([4096, 768])
608
+ base_model.model.model.al_projector.audio_proj.0.bias torch.Size([4096])
609
+ base_model.model.model.al_projector.audio_proj.2.weight torch.Size([4096, 4096])
610
+ base_model.model.model.al_projector.audio_proj.2.bias torch.Size([4096])
611
+ base_model.model.lm_head.weight torch.Size([32021, 4096])
612
+ trainable_params: 428.626MB
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e8f27de50cca66cf8391e6d6f7da7d052fa90f5d8ebb0da7515bd92f278b637
3
+ size 817336054
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/saved_config.json ADDED
@@ -0,0 +1,193 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_args": {
3
+ "model_name_or_path": "pretrained_weights/Llama-2-7b-chat-hf",
4
+ "freeze_backbone": true,
5
+ "llm_name": "llama",
6
+ "vit_ckpt_path": "./pretrained_weights/clip-vit-large-patch14",
7
+ "select_feature": "patch",
8
+ "image_size": 224,
9
+ "patch_size": 14,
10
+ "visual_query_token_nums": 32,
11
+ "BEATs_ckpt_path": "./pretrained_weights/BEATs_iter3_plus_AS2M_finetuned_on_AS2M_cpt2.pt",
12
+ "audio_query_token_nums": 32,
13
+ "prompt_embed_dim": 256,
14
+ "mask_decoder_transformer_depth": 2,
15
+ "low_res_mask_size": 112,
16
+ "image_scale_nums": 2,
17
+ "token_nums_per_scale": 3,
18
+ "avs_query_num": 300,
19
+ "num_classes": 1,
20
+ "query_generator_num_layers": 2
21
+ },
22
+ "data_args": {
23
+ "video_frame_nums": 10,
24
+ "image_caption_task": false,
25
+ "video_caption_task": false,
26
+ "audio_caption_task": false,
27
+ "segmentation_task": false,
28
+ "avqa_task": false,
29
+ "ave_task": false,
30
+ "avvp_task": false,
31
+ "arig_task": false,
32
+ "ms3_task": false,
33
+ "s4_task": false,
34
+ "avss_task": false,
35
+ "avcap_task": false,
36
+ "ref_avs_task": true,
37
+ "multi_frames": false
38
+ },
39
+ "training_args": {
40
+ "output_dir": "results_real/ref_avs_real/epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05",
41
+ "overwrite_output_dir": false,
42
+ "do_train": false,
43
+ "do_eval": false,
44
+ "do_predict": false,
45
+ "evaluation_strategy": "no",
46
+ "prediction_loss_only": false,
47
+ "per_device_train_batch_size": 4,
48
+ "per_device_eval_batch_size": 4,
49
+ "per_gpu_train_batch_size": null,
50
+ "per_gpu_eval_batch_size": null,
51
+ "gradient_accumulation_steps": 8,
52
+ "eval_accumulation_steps": null,
53
+ "eval_delay": 0,
54
+ "learning_rate": 0.0001,
55
+ "weight_decay": 0.0,
56
+ "adam_beta1": 0.9,
57
+ "adam_beta2": 0.999,
58
+ "adam_epsilon": 1e-08,
59
+ "max_grad_norm": 1.0,
60
+ "num_train_epochs": 6.0,
61
+ "max_steps": -1,
62
+ "lr_scheduler_type": "cosine",
63
+ "lr_scheduler_kwargs": {},
64
+ "warmup_ratio": 0.03,
65
+ "warmup_steps": 0,
66
+ "log_level": "passive",
67
+ "log_level_replica": "warning",
68
+ "log_on_each_node": true,
69
+ "logging_dir": "results_real/ref_avs_real/epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/runs/Jun05_09-29-17_gpu-12",
70
+ "logging_strategy": "steps",
71
+ "logging_first_step": false,
72
+ "logging_steps": 1.0,
73
+ "logging_nan_inf_filter": true,
74
+ "save_strategy": "epoch",
75
+ "save_steps": -1.0,
76
+ "save_total_limit": 10,
77
+ "save_safetensors": true,
78
+ "save_on_each_node": false,
79
+ "save_only_model": false,
80
+ "no_cuda": false,
81
+ "use_cpu": false,
82
+ "use_mps_device": false,
83
+ "seed": 42,
84
+ "data_seed": null,
85
+ "jit_mode_eval": false,
86
+ "use_ipex": false,
87
+ "bf16": true,
88
+ "fp16": false,
89
+ "fp16_opt_level": "O1",
90
+ "half_precision_backend": "auto",
91
+ "bf16_full_eval": false,
92
+ "fp16_full_eval": false,
93
+ "tf32": false,
94
+ "local_rank": 0,
95
+ "ddp_backend": null,
96
+ "tpu_num_cores": null,
97
+ "tpu_metrics_debug": false,
98
+ "debug": [],
99
+ "dataloader_drop_last": false,
100
+ "eval_steps": null,
101
+ "dataloader_num_workers": 4,
102
+ "past_index": -1,
103
+ "run_name": "results_real/ref_avs_real/epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05",
104
+ "disable_tqdm": false,
105
+ "remove_unused_columns": false,
106
+ "label_names": null,
107
+ "load_best_model_at_end": false,
108
+ "metric_for_best_model": null,
109
+ "greater_is_better": null,
110
+ "ignore_data_skip": false,
111
+ "fsdp": [],
112
+ "fsdp_min_num_params": 0,
113
+ "fsdp_config": {
114
+ "min_num_params": 0,
115
+ "xla": false,
116
+ "xla_fsdp_grad_ckpt": false
117
+ },
118
+ "fsdp_transformer_layer_cls_to_wrap": null,
119
+ "deepspeed": "deepspeed/stage2-offload.json",
120
+ "label_smoothing_factor": 0.0,
121
+ "optim": "adamw_torch",
122
+ "optim_args": null,
123
+ "adafactor": false,
124
+ "group_by_length": false,
125
+ "length_column_name": "length",
126
+ "report_to": [
127
+ "tensorboard",
128
+ "wandb"
129
+ ],
130
+ "ddp_find_unused_parameters": true,
131
+ "ddp_bucket_cap_mb": null,
132
+ "ddp_broadcast_buffers": null,
133
+ "dataloader_pin_memory": true,
134
+ "dataloader_persistent_workers": false,
135
+ "skip_memory_metrics": true,
136
+ "use_legacy_prediction_loop": false,
137
+ "push_to_hub": false,
138
+ "resume_from_checkpoint": null,
139
+ "hub_model_id": null,
140
+ "hub_strategy": "every_save",
141
+ "hub_token": null,
142
+ "hub_private_repo": false,
143
+ "hub_always_push": false,
144
+ "gradient_checkpointing": true,
145
+ "gradient_checkpointing_kwargs": null,
146
+ "include_inputs_for_metrics": false,
147
+ "fp16_backend": "auto",
148
+ "push_to_hub_model_id": null,
149
+ "push_to_hub_organization": null,
150
+ "push_to_hub_token": null,
151
+ "_n_gpu": 1,
152
+ "mp_parameters": "",
153
+ "auto_find_batch_size": false,
154
+ "full_determinism": false,
155
+ "torchdynamo": null,
156
+ "ray_scope": "last",
157
+ "ddp_timeout": 1800,
158
+ "torch_compile": false,
159
+ "torch_compile_backend": null,
160
+ "torch_compile_mode": null,
161
+ "dispatch_batches": null,
162
+ "split_batches": false,
163
+ "include_tokens_per_second": false,
164
+ "include_num_input_tokens_seen": false,
165
+ "neftune_noise_alpha": null,
166
+ "mm_projector_lr": null,
167
+ "freeze_mm_mlp_adapter": false,
168
+ "cache_dir": null,
169
+ "group_by_modality_length": false,
170
+ "model_max_length": 512,
171
+ "double_quant": true,
172
+ "quant_type": "nf4",
173
+ "bits": 32,
174
+ "lora_enable": true,
175
+ "lora_r": 8,
176
+ "lora_alpha": 16,
177
+ "lora_dropout": 0.05,
178
+ "lora_weight_path": "",
179
+ "lora_bias": "none",
180
+ "ce_loss_weight": 1.0,
181
+ "dice_loss_weight": 0.5,
182
+ "bce_loss_weight": 1.0,
183
+ "audio_branch": true,
184
+ "visual_branch": true,
185
+ "seg_branch": false,
186
+ "pretrain_ckpt_dir": "./pretrained_weights",
187
+ "finetune_ckpt_dir": "",
188
+ "save_modules": "vl_projector,al_projector,lora",
189
+ "exp_desc": "exp",
190
+ "use_process": true,
191
+ "use_hyper_lora": true
192
+ }
193
+ }
epochs6_lr1e-4_bs4_gradacc8_lora_r8alpha16dropout0.05/trainer_state.json ADDED
@@ -0,0 +1,3990 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 5.979614949037373,
5
+ "eval_steps": 500,
6
+ "global_step": 660,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 0.0,
14
+ "loss": 3.1127,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 2.313782131597592e-05,
20
+ "loss": 2.3729,
21
+ "step": 2
22
+ },
23
+ {
24
+ "epoch": 0.03,
25
+ "learning_rate": 3.667257913420847e-05,
26
+ "loss": 2.3541,
27
+ "step": 3
28
+ },
29
+ {
30
+ "epoch": 0.04,
31
+ "learning_rate": 4.627564263195184e-05,
32
+ "loss": 2.1759,
33
+ "step": 4
34
+ },
35
+ {
36
+ "epoch": 0.05,
37
+ "learning_rate": 5.372435736804816e-05,
38
+ "loss": 1.9987,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 0.05,
43
+ "learning_rate": 5.981040045018439e-05,
44
+ "loss": 1.7451,
45
+ "step": 6
46
+ },
47
+ {
48
+ "epoch": 0.06,
49
+ "learning_rate": 6.495607655709434e-05,
50
+ "loss": 1.5037,
51
+ "step": 7
52
+ },
53
+ {
54
+ "epoch": 0.07,
55
+ "learning_rate": 6.941346394792775e-05,
56
+ "loss": 1.3287,
57
+ "step": 8
58
+ },
59
+ {
60
+ "epoch": 0.08,
61
+ "learning_rate": 7.334515826841694e-05,
62
+ "loss": 1.1541,
63
+ "step": 9
64
+ },
65
+ {
66
+ "epoch": 0.09,
67
+ "learning_rate": 7.68621786840241e-05,
68
+ "loss": 1.0282,
69
+ "step": 10
70
+ },
71
+ {
72
+ "epoch": 0.1,
73
+ "learning_rate": 8.004371064686714e-05,
74
+ "loss": 0.9036,
75
+ "step": 11
76
+ },
77
+ {
78
+ "epoch": 0.11,
79
+ "learning_rate": 8.29482217661603e-05,
80
+ "loss": 0.8083,
81
+ "step": 12
82
+ },
83
+ {
84
+ "epoch": 0.12,
85
+ "learning_rate": 8.562011298888889e-05,
86
+ "loss": 0.715,
87
+ "step": 13
88
+ },
89
+ {
90
+ "epoch": 0.13,
91
+ "learning_rate": 8.809389787307026e-05,
92
+ "loss": 0.6602,
93
+ "step": 14
94
+ },
95
+ {
96
+ "epoch": 0.14,
97
+ "learning_rate": 9.039693650225664e-05,
98
+ "loss": 0.5938,
99
+ "step": 15
100
+ },
101
+ {
102
+ "epoch": 0.14,
103
+ "learning_rate": 9.255128526390367e-05,
104
+ "loss": 0.5876,
105
+ "step": 16
106
+ },
107
+ {
108
+ "epoch": 0.15,
109
+ "learning_rate": 9.457498485654161e-05,
110
+ "loss": 0.5373,
111
+ "step": 17
112
+ },
113
+ {
114
+ "epoch": 0.16,
115
+ "learning_rate": 9.648297958439284e-05,
116
+ "loss": 0.5317,
117
+ "step": 18
118
+ },
119
+ {
120
+ "epoch": 0.17,
121
+ "learning_rate": 9.828778776927557e-05,
122
+ "loss": 0.5139,
123
+ "step": 19
124
+ },
125
+ {
126
+ "epoch": 0.18,
127
+ "learning_rate": 0.0001,
128
+ "loss": 0.4989,
129
+ "step": 20
130
+ },
131
+ {
132
+ "epoch": 0.19,
133
+ "learning_rate": 0.0001,
134
+ "loss": 0.4844,
135
+ "step": 21
136
+ },
137
+ {
138
+ "epoch": 0.2,
139
+ "learning_rate": 9.984375e-05,
140
+ "loss": 0.4535,
141
+ "step": 22
142
+ },
143
+ {
144
+ "epoch": 0.21,
145
+ "learning_rate": 9.96875e-05,
146
+ "loss": 0.4414,
147
+ "step": 23
148
+ },
149
+ {
150
+ "epoch": 0.22,
151
+ "learning_rate": 9.953125000000001e-05,
152
+ "loss": 0.4391,
153
+ "step": 24
154
+ },
155
+ {
156
+ "epoch": 0.23,
157
+ "learning_rate": 9.9375e-05,
158
+ "loss": 0.4229,
159
+ "step": 25
160
+ },
161
+ {
162
+ "epoch": 0.24,
163
+ "learning_rate": 9.921875000000001e-05,
164
+ "loss": 0.4291,
165
+ "step": 26
166
+ },
167
+ {
168
+ "epoch": 0.24,
169
+ "learning_rate": 9.90625e-05,
170
+ "loss": 0.433,
171
+ "step": 27
172
+ },
173
+ {
174
+ "epoch": 0.25,
175
+ "learning_rate": 9.890625e-05,
176
+ "loss": 0.4309,
177
+ "step": 28
178
+ },
179
+ {
180
+ "epoch": 0.26,
181
+ "learning_rate": 9.875000000000002e-05,
182
+ "loss": 0.3866,
183
+ "step": 29
184
+ },
185
+ {
186
+ "epoch": 0.27,
187
+ "learning_rate": 9.859375000000001e-05,
188
+ "loss": 0.4084,
189
+ "step": 30
190
+ },
191
+ {
192
+ "epoch": 0.28,
193
+ "learning_rate": 9.84375e-05,
194
+ "loss": 0.3987,
195
+ "step": 31
196
+ },
197
+ {
198
+ "epoch": 0.29,
199
+ "learning_rate": 9.828125000000001e-05,
200
+ "loss": 0.4156,
201
+ "step": 32
202
+ },
203
+ {
204
+ "epoch": 0.3,
205
+ "learning_rate": 9.8125e-05,
206
+ "loss": 0.3865,
207
+ "step": 33
208
+ },
209
+ {
210
+ "epoch": 0.31,
211
+ "learning_rate": 9.796875e-05,
212
+ "loss": 0.386,
213
+ "step": 34
214
+ },
215
+ {
216
+ "epoch": 0.32,
217
+ "learning_rate": 9.781250000000001e-05,
218
+ "loss": 0.4243,
219
+ "step": 35
220
+ },
221
+ {
222
+ "epoch": 0.33,
223
+ "learning_rate": 9.765625e-05,
224
+ "loss": 0.4054,
225
+ "step": 36
226
+ },
227
+ {
228
+ "epoch": 0.34,
229
+ "learning_rate": 9.75e-05,
230
+ "loss": 0.3999,
231
+ "step": 37
232
+ },
233
+ {
234
+ "epoch": 0.34,
235
+ "learning_rate": 9.734375e-05,
236
+ "loss": 0.387,
237
+ "step": 38
238
+ },
239
+ {
240
+ "epoch": 0.35,
241
+ "learning_rate": 9.718750000000001e-05,
242
+ "loss": 0.3874,
243
+ "step": 39
244
+ },
245
+ {
246
+ "epoch": 0.36,
247
+ "learning_rate": 9.703125e-05,
248
+ "loss": 0.3968,
249
+ "step": 40
250
+ },
251
+ {
252
+ "epoch": 0.37,
253
+ "learning_rate": 9.687500000000001e-05,
254
+ "loss": 0.3729,
255
+ "step": 41
256
+ },
257
+ {
258
+ "epoch": 0.38,
259
+ "learning_rate": 9.671875e-05,
260
+ "loss": 0.378,
261
+ "step": 42
262
+ },
263
+ {
264
+ "epoch": 0.39,
265
+ "learning_rate": 9.65625e-05,
266
+ "loss": 0.3787,
267
+ "step": 43
268
+ },
269
+ {
270
+ "epoch": 0.4,
271
+ "learning_rate": 9.640625000000002e-05,
272
+ "loss": 0.3927,
273
+ "step": 44
274
+ },
275
+ {
276
+ "epoch": 0.41,
277
+ "learning_rate": 9.625000000000001e-05,
278
+ "loss": 0.3707,
279
+ "step": 45
280
+ },
281
+ {
282
+ "epoch": 0.42,
283
+ "learning_rate": 9.609375e-05,
284
+ "loss": 0.3603,
285
+ "step": 46
286
+ },
287
+ {
288
+ "epoch": 0.43,
289
+ "learning_rate": 9.593750000000001e-05,
290
+ "loss": 0.3777,
291
+ "step": 47
292
+ },
293
+ {
294
+ "epoch": 0.43,
295
+ "learning_rate": 9.578125e-05,
296
+ "loss": 0.3594,
297
+ "step": 48
298
+ },
299
+ {
300
+ "epoch": 0.44,
301
+ "learning_rate": 9.562500000000001e-05,
302
+ "loss": 0.3608,
303
+ "step": 49
304
+ },
305
+ {
306
+ "epoch": 0.45,
307
+ "learning_rate": 9.546875000000001e-05,
308
+ "loss": 0.3663,
309
+ "step": 50
310
+ },
311
+ {
312
+ "epoch": 0.46,
313
+ "learning_rate": 9.53125e-05,
314
+ "loss": 0.3733,
315
+ "step": 51
316
+ },
317
+ {
318
+ "epoch": 0.47,
319
+ "learning_rate": 9.515625e-05,
320
+ "loss": 0.3707,
321
+ "step": 52
322
+ },
323
+ {
324
+ "epoch": 0.48,
325
+ "learning_rate": 9.5e-05,
326
+ "loss": 0.3698,
327
+ "step": 53
328
+ },
329
+ {
330
+ "epoch": 0.49,
331
+ "learning_rate": 9.484375000000001e-05,
332
+ "loss": 0.3645,
333
+ "step": 54
334
+ },
335
+ {
336
+ "epoch": 0.5,
337
+ "learning_rate": 9.46875e-05,
338
+ "loss": 0.3592,
339
+ "step": 55
340
+ },
341
+ {
342
+ "epoch": 0.51,
343
+ "learning_rate": 9.453125000000001e-05,
344
+ "loss": 0.3511,
345
+ "step": 56
346
+ },
347
+ {
348
+ "epoch": 0.52,
349
+ "learning_rate": 9.4375e-05,
350
+ "loss": 0.3498,
351
+ "step": 57
352
+ },
353
+ {
354
+ "epoch": 0.53,
355
+ "learning_rate": 9.421875e-05,
356
+ "loss": 0.3547,
357
+ "step": 58
358
+ },
359
+ {
360
+ "epoch": 0.53,
361
+ "learning_rate": 9.40625e-05,
362
+ "loss": 0.3374,
363
+ "step": 59
364
+ },
365
+ {
366
+ "epoch": 0.54,
367
+ "learning_rate": 9.390625000000001e-05,
368
+ "loss": 0.3456,
369
+ "step": 60
370
+ },
371
+ {
372
+ "epoch": 0.55,
373
+ "learning_rate": 9.375e-05,
374
+ "loss": 0.3447,
375
+ "step": 61
376
+ },
377
+ {
378
+ "epoch": 0.56,
379
+ "learning_rate": 9.359375000000001e-05,
380
+ "loss": 0.3422,
381
+ "step": 62
382
+ },
383
+ {
384
+ "epoch": 0.57,
385
+ "learning_rate": 9.34375e-05,
386
+ "loss": 0.3498,
387
+ "step": 63
388
+ },
389
+ {
390
+ "epoch": 0.58,
391
+ "learning_rate": 9.328125000000001e-05,
392
+ "loss": 0.3399,
393
+ "step": 64
394
+ },
395
+ {
396
+ "epoch": 0.59,
397
+ "learning_rate": 9.3125e-05,
398
+ "loss": 0.3495,
399
+ "step": 65
400
+ },
401
+ {
402
+ "epoch": 0.6,
403
+ "learning_rate": 9.296875e-05,
404
+ "loss": 0.3375,
405
+ "step": 66
406
+ },
407
+ {
408
+ "epoch": 0.61,
409
+ "learning_rate": 9.28125e-05,
410
+ "loss": 0.3337,
411
+ "step": 67
412
+ },
413
+ {
414
+ "epoch": 0.62,
415
+ "learning_rate": 9.265625e-05,
416
+ "loss": 0.3303,
417
+ "step": 68
418
+ },
419
+ {
420
+ "epoch": 0.63,
421
+ "learning_rate": 9.250000000000001e-05,
422
+ "loss": 0.3295,
423
+ "step": 69
424
+ },
425
+ {
426
+ "epoch": 0.63,
427
+ "learning_rate": 9.234375e-05,
428
+ "loss": 0.3278,
429
+ "step": 70
430
+ },
431
+ {
432
+ "epoch": 0.64,
433
+ "learning_rate": 9.21875e-05,
434
+ "loss": 0.3372,
435
+ "step": 71
436
+ },
437
+ {
438
+ "epoch": 0.65,
439
+ "learning_rate": 9.203125e-05,
440
+ "loss": 0.3471,
441
+ "step": 72
442
+ },
443
+ {
444
+ "epoch": 0.66,
445
+ "learning_rate": 9.1875e-05,
446
+ "loss": 0.3217,
447
+ "step": 73
448
+ },
449
+ {
450
+ "epoch": 0.67,
451
+ "learning_rate": 9.171875e-05,
452
+ "loss": 0.337,
453
+ "step": 74
454
+ },
455
+ {
456
+ "epoch": 0.68,
457
+ "learning_rate": 9.156250000000001e-05,
458
+ "loss": 0.3532,
459
+ "step": 75
460
+ },
461
+ {
462
+ "epoch": 0.69,
463
+ "learning_rate": 9.140625e-05,
464
+ "loss": 0.3342,
465
+ "step": 76
466
+ },
467
+ {
468
+ "epoch": 0.7,
469
+ "learning_rate": 9.125e-05,
470
+ "loss": 0.3245,
471
+ "step": 77
472
+ },
473
+ {
474
+ "epoch": 0.71,
475
+ "learning_rate": 9.109375e-05,
476
+ "loss": 0.3338,
477
+ "step": 78
478
+ },
479
+ {
480
+ "epoch": 0.72,
481
+ "learning_rate": 9.093750000000001e-05,
482
+ "loss": 0.3382,
483
+ "step": 79
484
+ },
485
+ {
486
+ "epoch": 0.72,
487
+ "learning_rate": 9.078125e-05,
488
+ "loss": 0.3295,
489
+ "step": 80
490
+ },
491
+ {
492
+ "epoch": 0.73,
493
+ "learning_rate": 9.062500000000001e-05,
494
+ "loss": 0.3207,
495
+ "step": 81
496
+ },
497
+ {
498
+ "epoch": 0.74,
499
+ "learning_rate": 9.046875e-05,
500
+ "loss": 0.3224,
501
+ "step": 82
502
+ },
503
+ {
504
+ "epoch": 0.75,
505
+ "learning_rate": 9.03125e-05,
506
+ "loss": 0.3181,
507
+ "step": 83
508
+ },
509
+ {
510
+ "epoch": 0.76,
511
+ "learning_rate": 9.015625000000001e-05,
512
+ "loss": 0.3572,
513
+ "step": 84
514
+ },
515
+ {
516
+ "epoch": 0.77,
517
+ "learning_rate": 9e-05,
518
+ "loss": 0.3373,
519
+ "step": 85
520
+ },
521
+ {
522
+ "epoch": 0.78,
523
+ "learning_rate": 8.984375e-05,
524
+ "loss": 0.3349,
525
+ "step": 86
526
+ },
527
+ {
528
+ "epoch": 0.79,
529
+ "learning_rate": 8.96875e-05,
530
+ "loss": 0.3491,
531
+ "step": 87
532
+ },
533
+ {
534
+ "epoch": 0.8,
535
+ "learning_rate": 8.953125e-05,
536
+ "loss": 0.3414,
537
+ "step": 88
538
+ },
539
+ {
540
+ "epoch": 0.81,
541
+ "learning_rate": 8.9375e-05,
542
+ "loss": 0.3345,
543
+ "step": 89
544
+ },
545
+ {
546
+ "epoch": 0.82,
547
+ "learning_rate": 8.921875000000001e-05,
548
+ "loss": 0.3328,
549
+ "step": 90
550
+ },
551
+ {
552
+ "epoch": 0.82,
553
+ "learning_rate": 8.90625e-05,
554
+ "loss": 0.3496,
555
+ "step": 91
556
+ },
557
+ {
558
+ "epoch": 0.83,
559
+ "learning_rate": 8.890625e-05,
560
+ "loss": 0.3458,
561
+ "step": 92
562
+ },
563
+ {
564
+ "epoch": 0.84,
565
+ "learning_rate": 8.875e-05,
566
+ "loss": 0.3126,
567
+ "step": 93
568
+ },
569
+ {
570
+ "epoch": 0.85,
571
+ "learning_rate": 8.859375000000001e-05,
572
+ "loss": 0.3129,
573
+ "step": 94
574
+ },
575
+ {
576
+ "epoch": 0.86,
577
+ "learning_rate": 8.84375e-05,
578
+ "loss": 0.3167,
579
+ "step": 95
580
+ },
581
+ {
582
+ "epoch": 0.87,
583
+ "learning_rate": 8.828125000000001e-05,
584
+ "loss": 0.3201,
585
+ "step": 96
586
+ },
587
+ {
588
+ "epoch": 0.88,
589
+ "learning_rate": 8.8125e-05,
590
+ "loss": 0.3392,
591
+ "step": 97
592
+ },
593
+ {
594
+ "epoch": 0.89,
595
+ "learning_rate": 8.796875e-05,
596
+ "loss": 0.3106,
597
+ "step": 98
598
+ },
599
+ {
600
+ "epoch": 0.9,
601
+ "learning_rate": 8.781250000000001e-05,
602
+ "loss": 0.3106,
603
+ "step": 99
604
+ },
605
+ {
606
+ "epoch": 0.91,
607
+ "learning_rate": 8.765625e-05,
608
+ "loss": 0.3173,
609
+ "step": 100
610
+ },
611
+ {
612
+ "epoch": 0.92,
613
+ "learning_rate": 8.75e-05,
614
+ "loss": 0.3196,
615
+ "step": 101
616
+ },
617
+ {
618
+ "epoch": 0.92,
619
+ "learning_rate": 8.734375e-05,
620
+ "loss": 0.3239,
621
+ "step": 102
622
+ },
623
+ {
624
+ "epoch": 0.93,
625
+ "learning_rate": 8.71875e-05,
626
+ "loss": 0.3266,
627
+ "step": 103
628
+ },
629
+ {
630
+ "epoch": 0.94,
631
+ "learning_rate": 8.703125e-05,
632
+ "loss": 0.3175,
633
+ "step": 104
634
+ },
635
+ {
636
+ "epoch": 0.95,
637
+ "learning_rate": 8.687500000000001e-05,
638
+ "loss": 0.3111,
639
+ "step": 105
640
+ },
641
+ {
642
+ "epoch": 0.96,
643
+ "learning_rate": 8.671875e-05,
644
+ "loss": 0.3138,
645
+ "step": 106
646
+ },
647
+ {
648
+ "epoch": 0.97,
649
+ "learning_rate": 8.65625e-05,
650
+ "loss": 0.3195,
651
+ "step": 107
652
+ },
653
+ {
654
+ "epoch": 0.98,
655
+ "learning_rate": 8.640625e-05,
656
+ "loss": 0.3125,
657
+ "step": 108
658
+ },
659
+ {
660
+ "epoch": 0.99,
661
+ "learning_rate": 8.625000000000001e-05,
662
+ "loss": 0.3038,
663
+ "step": 109
664
+ },
665
+ {
666
+ "epoch": 1.0,
667
+ "learning_rate": 8.609375e-05,
668
+ "loss": 0.3052,
669
+ "step": 110
670
+ },
671
+ {
672
+ "epoch": 1.01,
673
+ "learning_rate": 8.593750000000001e-05,
674
+ "loss": 0.3141,
675
+ "step": 111
676
+ },
677
+ {
678
+ "epoch": 1.01,
679
+ "learning_rate": 8.578125e-05,
680
+ "loss": 0.3072,
681
+ "step": 112
682
+ },
683
+ {
684
+ "epoch": 1.02,
685
+ "learning_rate": 8.5625e-05,
686
+ "loss": 0.2979,
687
+ "step": 113
688
+ },
689
+ {
690
+ "epoch": 1.03,
691
+ "learning_rate": 8.546875000000001e-05,
692
+ "loss": 0.2828,
693
+ "step": 114
694
+ },
695
+ {
696
+ "epoch": 1.04,
697
+ "learning_rate": 8.531250000000001e-05,
698
+ "loss": 0.3011,
699
+ "step": 115
700
+ },
701
+ {
702
+ "epoch": 1.05,
703
+ "learning_rate": 8.515625e-05,
704
+ "loss": 0.2888,
705
+ "step": 116
706
+ },
707
+ {
708
+ "epoch": 1.06,
709
+ "learning_rate": 8.5e-05,
710
+ "loss": 0.3104,
711
+ "step": 117
712
+ },
713
+ {
714
+ "epoch": 1.07,
715
+ "learning_rate": 8.484375e-05,
716
+ "loss": 0.3073,
717
+ "step": 118
718
+ },
719
+ {
720
+ "epoch": 1.08,
721
+ "learning_rate": 8.46875e-05,
722
+ "loss": 0.3097,
723
+ "step": 119
724
+ },
725
+ {
726
+ "epoch": 1.09,
727
+ "learning_rate": 8.453125000000001e-05,
728
+ "loss": 0.3033,
729
+ "step": 120
730
+ },
731
+ {
732
+ "epoch": 1.1,
733
+ "learning_rate": 8.4375e-05,
734
+ "loss": 0.3058,
735
+ "step": 121
736
+ },
737
+ {
738
+ "epoch": 1.11,
739
+ "learning_rate": 8.421875e-05,
740
+ "loss": 0.287,
741
+ "step": 122
742
+ },
743
+ {
744
+ "epoch": 1.11,
745
+ "learning_rate": 8.40625e-05,
746
+ "loss": 0.306,
747
+ "step": 123
748
+ },
749
+ {
750
+ "epoch": 1.12,
751
+ "learning_rate": 8.390625000000001e-05,
752
+ "loss": 0.298,
753
+ "step": 124
754
+ },
755
+ {
756
+ "epoch": 1.13,
757
+ "learning_rate": 8.375e-05,
758
+ "loss": 0.2873,
759
+ "step": 125
760
+ },
761
+ {
762
+ "epoch": 1.14,
763
+ "learning_rate": 8.359375000000001e-05,
764
+ "loss": 0.2895,
765
+ "step": 126
766
+ },
767
+ {
768
+ "epoch": 1.15,
769
+ "learning_rate": 8.34375e-05,
770
+ "loss": 0.2928,
771
+ "step": 127
772
+ },
773
+ {
774
+ "epoch": 1.16,
775
+ "learning_rate": 8.328125e-05,
776
+ "loss": 0.2875,
777
+ "step": 128
778
+ },
779
+ {
780
+ "epoch": 1.17,
781
+ "learning_rate": 8.312500000000001e-05,
782
+ "loss": 0.2918,
783
+ "step": 129
784
+ },
785
+ {
786
+ "epoch": 1.18,
787
+ "learning_rate": 8.296875000000001e-05,
788
+ "loss": 0.2857,
789
+ "step": 130
790
+ },
791
+ {
792
+ "epoch": 1.19,
793
+ "learning_rate": 8.28125e-05,
794
+ "loss": 0.3044,
795
+ "step": 131
796
+ },
797
+ {
798
+ "epoch": 1.2,
799
+ "learning_rate": 8.265625e-05,
800
+ "loss": 0.2899,
801
+ "step": 132
802
+ },
803
+ {
804
+ "epoch": 1.2,
805
+ "learning_rate": 8.25e-05,
806
+ "loss": 0.2871,
807
+ "step": 133
808
+ },
809
+ {
810
+ "epoch": 1.21,
811
+ "learning_rate": 8.234375e-05,
812
+ "loss": 0.3135,
813
+ "step": 134
814
+ },
815
+ {
816
+ "epoch": 1.22,
817
+ "learning_rate": 8.218750000000001e-05,
818
+ "loss": 0.3033,
819
+ "step": 135
820
+ },
821
+ {
822
+ "epoch": 1.23,
823
+ "learning_rate": 8.203125e-05,
824
+ "loss": 0.281,
825
+ "step": 136
826
+ },
827
+ {
828
+ "epoch": 1.24,
829
+ "learning_rate": 8.1875e-05,
830
+ "loss": 0.2852,
831
+ "step": 137
832
+ },
833
+ {
834
+ "epoch": 1.25,
835
+ "learning_rate": 8.171875e-05,
836
+ "loss": 0.298,
837
+ "step": 138
838
+ },
839
+ {
840
+ "epoch": 1.26,
841
+ "learning_rate": 8.156250000000001e-05,
842
+ "loss": 0.2895,
843
+ "step": 139
844
+ },
845
+ {
846
+ "epoch": 1.27,
847
+ "learning_rate": 8.140625e-05,
848
+ "loss": 0.3002,
849
+ "step": 140
850
+ },
851
+ {
852
+ "epoch": 1.28,
853
+ "learning_rate": 8.125000000000001e-05,
854
+ "loss": 0.2844,
855
+ "step": 141
856
+ },
857
+ {
858
+ "epoch": 1.29,
859
+ "learning_rate": 8.109375e-05,
860
+ "loss": 0.2832,
861
+ "step": 142
862
+ },
863
+ {
864
+ "epoch": 1.3,
865
+ "learning_rate": 8.09375e-05,
866
+ "loss": 0.2917,
867
+ "step": 143
868
+ },
869
+ {
870
+ "epoch": 1.3,
871
+ "learning_rate": 8.078125e-05,
872
+ "loss": 0.2907,
873
+ "step": 144
874
+ },
875
+ {
876
+ "epoch": 1.31,
877
+ "learning_rate": 8.062500000000001e-05,
878
+ "loss": 0.2828,
879
+ "step": 145
880
+ },
881
+ {
882
+ "epoch": 1.32,
883
+ "learning_rate": 8.046875e-05,
884
+ "loss": 0.2835,
885
+ "step": 146
886
+ },
887
+ {
888
+ "epoch": 1.33,
889
+ "learning_rate": 8.031250000000001e-05,
890
+ "loss": 0.2952,
891
+ "step": 147
892
+ },
893
+ {
894
+ "epoch": 1.34,
895
+ "learning_rate": 8.015625e-05,
896
+ "loss": 0.2912,
897
+ "step": 148
898
+ },
899
+ {
900
+ "epoch": 1.35,
901
+ "learning_rate": 8e-05,
902
+ "loss": 0.2848,
903
+ "step": 149
904
+ },
905
+ {
906
+ "epoch": 1.36,
907
+ "learning_rate": 7.984375e-05,
908
+ "loss": 0.2944,
909
+ "step": 150
910
+ },
911
+ {
912
+ "epoch": 1.37,
913
+ "learning_rate": 7.96875e-05,
914
+ "loss": 0.3008,
915
+ "step": 151
916
+ },
917
+ {
918
+ "epoch": 1.38,
919
+ "learning_rate": 7.953125e-05,
920
+ "loss": 0.2887,
921
+ "step": 152
922
+ },
923
+ {
924
+ "epoch": 1.39,
925
+ "learning_rate": 7.9375e-05,
926
+ "loss": 0.2976,
927
+ "step": 153
928
+ },
929
+ {
930
+ "epoch": 1.4,
931
+ "learning_rate": 7.921875000000001e-05,
932
+ "loss": 0.2856,
933
+ "step": 154
934
+ },
935
+ {
936
+ "epoch": 1.4,
937
+ "learning_rate": 7.90625e-05,
938
+ "loss": 0.2851,
939
+ "step": 155
940
+ },
941
+ {
942
+ "epoch": 1.41,
943
+ "learning_rate": 7.890625000000001e-05,
944
+ "loss": 0.2833,
945
+ "step": 156
946
+ },
947
+ {
948
+ "epoch": 1.42,
949
+ "learning_rate": 7.875e-05,
950
+ "loss": 0.292,
951
+ "step": 157
952
+ },
953
+ {
954
+ "epoch": 1.43,
955
+ "learning_rate": 7.859375e-05,
956
+ "loss": 0.291,
957
+ "step": 158
958
+ },
959
+ {
960
+ "epoch": 1.44,
961
+ "learning_rate": 7.84375e-05,
962
+ "loss": 0.2863,
963
+ "step": 159
964
+ },
965
+ {
966
+ "epoch": 1.45,
967
+ "learning_rate": 7.828125000000001e-05,
968
+ "loss": 0.288,
969
+ "step": 160
970
+ },
971
+ {
972
+ "epoch": 1.46,
973
+ "learning_rate": 7.8125e-05,
974
+ "loss": 0.2973,
975
+ "step": 161
976
+ },
977
+ {
978
+ "epoch": 1.47,
979
+ "learning_rate": 7.796875000000001e-05,
980
+ "loss": 0.2722,
981
+ "step": 162
982
+ },
983
+ {
984
+ "epoch": 1.48,
985
+ "learning_rate": 7.78125e-05,
986
+ "loss": 0.2972,
987
+ "step": 163
988
+ },
989
+ {
990
+ "epoch": 1.49,
991
+ "learning_rate": 7.765625000000001e-05,
992
+ "loss": 0.2844,
993
+ "step": 164
994
+ },
995
+ {
996
+ "epoch": 1.49,
997
+ "learning_rate": 7.75e-05,
998
+ "loss": 0.2878,
999
+ "step": 165
1000
+ },
1001
+ {
1002
+ "epoch": 1.5,
1003
+ "learning_rate": 7.734375e-05,
1004
+ "loss": 0.2882,
1005
+ "step": 166
1006
+ },
1007
+ {
1008
+ "epoch": 1.51,
1009
+ "learning_rate": 7.71875e-05,
1010
+ "loss": 0.2749,
1011
+ "step": 167
1012
+ },
1013
+ {
1014
+ "epoch": 1.52,
1015
+ "learning_rate": 7.703125e-05,
1016
+ "loss": 0.2823,
1017
+ "step": 168
1018
+ },
1019
+ {
1020
+ "epoch": 1.53,
1021
+ "learning_rate": 7.687500000000001e-05,
1022
+ "loss": 0.2783,
1023
+ "step": 169
1024
+ },
1025
+ {
1026
+ "epoch": 1.54,
1027
+ "learning_rate": 7.671875e-05,
1028
+ "loss": 0.2941,
1029
+ "step": 170
1030
+ },
1031
+ {
1032
+ "epoch": 1.55,
1033
+ "learning_rate": 7.65625e-05,
1034
+ "loss": 0.2789,
1035
+ "step": 171
1036
+ },
1037
+ {
1038
+ "epoch": 1.56,
1039
+ "learning_rate": 7.640625e-05,
1040
+ "loss": 0.2743,
1041
+ "step": 172
1042
+ },
1043
+ {
1044
+ "epoch": 1.57,
1045
+ "learning_rate": 7.625e-05,
1046
+ "loss": 0.2839,
1047
+ "step": 173
1048
+ },
1049
+ {
1050
+ "epoch": 1.58,
1051
+ "learning_rate": 7.609375e-05,
1052
+ "loss": 0.2698,
1053
+ "step": 174
1054
+ },
1055
+ {
1056
+ "epoch": 1.59,
1057
+ "learning_rate": 7.593750000000001e-05,
1058
+ "loss": 0.2925,
1059
+ "step": 175
1060
+ },
1061
+ {
1062
+ "epoch": 1.59,
1063
+ "learning_rate": 7.578125e-05,
1064
+ "loss": 0.283,
1065
+ "step": 176
1066
+ },
1067
+ {
1068
+ "epoch": 1.6,
1069
+ "learning_rate": 7.5625e-05,
1070
+ "loss": 0.2776,
1071
+ "step": 177
1072
+ },
1073
+ {
1074
+ "epoch": 1.61,
1075
+ "learning_rate": 7.546875e-05,
1076
+ "loss": 0.3031,
1077
+ "step": 178
1078
+ },
1079
+ {
1080
+ "epoch": 1.62,
1081
+ "learning_rate": 7.531250000000001e-05,
1082
+ "loss": 0.2823,
1083
+ "step": 179
1084
+ },
1085
+ {
1086
+ "epoch": 1.63,
1087
+ "learning_rate": 7.515625e-05,
1088
+ "loss": 0.2883,
1089
+ "step": 180
1090
+ },
1091
+ {
1092
+ "epoch": 1.64,
1093
+ "learning_rate": 7.500000000000001e-05,
1094
+ "loss": 0.2845,
1095
+ "step": 181
1096
+ },
1097
+ {
1098
+ "epoch": 1.65,
1099
+ "learning_rate": 7.484375e-05,
1100
+ "loss": 0.2804,
1101
+ "step": 182
1102
+ },
1103
+ {
1104
+ "epoch": 1.66,
1105
+ "learning_rate": 7.468749999999999e-05,
1106
+ "loss": 0.2747,
1107
+ "step": 183
1108
+ },
1109
+ {
1110
+ "epoch": 1.67,
1111
+ "learning_rate": 7.453125000000001e-05,
1112
+ "loss": 0.2822,
1113
+ "step": 184
1114
+ },
1115
+ {
1116
+ "epoch": 1.68,
1117
+ "learning_rate": 7.4375e-05,
1118
+ "loss": 0.2708,
1119
+ "step": 185
1120
+ },
1121
+ {
1122
+ "epoch": 1.69,
1123
+ "learning_rate": 7.421875e-05,
1124
+ "loss": 0.287,
1125
+ "step": 186
1126
+ },
1127
+ {
1128
+ "epoch": 1.69,
1129
+ "learning_rate": 7.40625e-05,
1130
+ "loss": 0.2785,
1131
+ "step": 187
1132
+ },
1133
+ {
1134
+ "epoch": 1.7,
1135
+ "learning_rate": 7.390625e-05,
1136
+ "loss": 0.285,
1137
+ "step": 188
1138
+ },
1139
+ {
1140
+ "epoch": 1.71,
1141
+ "learning_rate": 7.375e-05,
1142
+ "loss": 0.2963,
1143
+ "step": 189
1144
+ },
1145
+ {
1146
+ "epoch": 1.72,
1147
+ "learning_rate": 7.359375000000001e-05,
1148
+ "loss": 0.3,
1149
+ "step": 190
1150
+ },
1151
+ {
1152
+ "epoch": 1.73,
1153
+ "learning_rate": 7.34375e-05,
1154
+ "loss": 0.2738,
1155
+ "step": 191
1156
+ },
1157
+ {
1158
+ "epoch": 1.74,
1159
+ "learning_rate": 7.328125e-05,
1160
+ "loss": 0.2721,
1161
+ "step": 192
1162
+ },
1163
+ {
1164
+ "epoch": 1.75,
1165
+ "learning_rate": 7.3125e-05,
1166
+ "loss": 0.2762,
1167
+ "step": 193
1168
+ },
1169
+ {
1170
+ "epoch": 1.76,
1171
+ "learning_rate": 7.296875000000001e-05,
1172
+ "loss": 0.2878,
1173
+ "step": 194
1174
+ },
1175
+ {
1176
+ "epoch": 1.77,
1177
+ "learning_rate": 7.28125e-05,
1178
+ "loss": 0.2989,
1179
+ "step": 195
1180
+ },
1181
+ {
1182
+ "epoch": 1.78,
1183
+ "learning_rate": 7.265625000000001e-05,
1184
+ "loss": 0.2826,
1185
+ "step": 196
1186
+ },
1187
+ {
1188
+ "epoch": 1.78,
1189
+ "learning_rate": 7.25e-05,
1190
+ "loss": 0.2832,
1191
+ "step": 197
1192
+ },
1193
+ {
1194
+ "epoch": 1.79,
1195
+ "learning_rate": 7.234375e-05,
1196
+ "loss": 0.2834,
1197
+ "step": 198
1198
+ },
1199
+ {
1200
+ "epoch": 1.8,
1201
+ "learning_rate": 7.218750000000001e-05,
1202
+ "loss": 0.2859,
1203
+ "step": 199
1204
+ },
1205
+ {
1206
+ "epoch": 1.81,
1207
+ "learning_rate": 7.203125e-05,
1208
+ "loss": 0.2644,
1209
+ "step": 200
1210
+ },
1211
+ {
1212
+ "epoch": 1.82,
1213
+ "learning_rate": 7.1875e-05,
1214
+ "loss": 0.2879,
1215
+ "step": 201
1216
+ },
1217
+ {
1218
+ "epoch": 1.83,
1219
+ "learning_rate": 7.171875e-05,
1220
+ "loss": 0.2807,
1221
+ "step": 202
1222
+ },
1223
+ {
1224
+ "epoch": 1.84,
1225
+ "learning_rate": 7.15625e-05,
1226
+ "loss": 0.2845,
1227
+ "step": 203
1228
+ },
1229
+ {
1230
+ "epoch": 1.85,
1231
+ "learning_rate": 7.140625e-05,
1232
+ "loss": 0.2905,
1233
+ "step": 204
1234
+ },
1235
+ {
1236
+ "epoch": 1.86,
1237
+ "learning_rate": 7.125000000000001e-05,
1238
+ "loss": 0.278,
1239
+ "step": 205
1240
+ },
1241
+ {
1242
+ "epoch": 1.87,
1243
+ "learning_rate": 7.109375e-05,
1244
+ "loss": 0.2779,
1245
+ "step": 206
1246
+ },
1247
+ {
1248
+ "epoch": 1.88,
1249
+ "learning_rate": 7.09375e-05,
1250
+ "loss": 0.2801,
1251
+ "step": 207
1252
+ },
1253
+ {
1254
+ "epoch": 1.88,
1255
+ "learning_rate": 7.078125e-05,
1256
+ "loss": 0.2734,
1257
+ "step": 208
1258
+ },
1259
+ {
1260
+ "epoch": 1.89,
1261
+ "learning_rate": 7.062500000000001e-05,
1262
+ "loss": 0.2821,
1263
+ "step": 209
1264
+ },
1265
+ {
1266
+ "epoch": 1.9,
1267
+ "learning_rate": 7.046875e-05,
1268
+ "loss": 0.2712,
1269
+ "step": 210
1270
+ },
1271
+ {
1272
+ "epoch": 1.91,
1273
+ "learning_rate": 7.031250000000001e-05,
1274
+ "loss": 0.274,
1275
+ "step": 211
1276
+ },
1277
+ {
1278
+ "epoch": 1.92,
1279
+ "learning_rate": 7.015625e-05,
1280
+ "loss": 0.2911,
1281
+ "step": 212
1282
+ },
1283
+ {
1284
+ "epoch": 1.93,
1285
+ "learning_rate": 7e-05,
1286
+ "loss": 0.2726,
1287
+ "step": 213
1288
+ },
1289
+ {
1290
+ "epoch": 1.94,
1291
+ "learning_rate": 6.984375000000001e-05,
1292
+ "loss": 0.2684,
1293
+ "step": 214
1294
+ },
1295
+ {
1296
+ "epoch": 1.95,
1297
+ "learning_rate": 6.96875e-05,
1298
+ "loss": 0.2631,
1299
+ "step": 215
1300
+ },
1301
+ {
1302
+ "epoch": 1.96,
1303
+ "learning_rate": 6.953125e-05,
1304
+ "loss": 0.2726,
1305
+ "step": 216
1306
+ },
1307
+ {
1308
+ "epoch": 1.97,
1309
+ "learning_rate": 6.9375e-05,
1310
+ "loss": 0.2814,
1311
+ "step": 217
1312
+ },
1313
+ {
1314
+ "epoch": 1.98,
1315
+ "learning_rate": 6.921875e-05,
1316
+ "loss": 0.2667,
1317
+ "step": 218
1318
+ },
1319
+ {
1320
+ "epoch": 1.98,
1321
+ "learning_rate": 6.90625e-05,
1322
+ "loss": 0.2872,
1323
+ "step": 219
1324
+ },
1325
+ {
1326
+ "epoch": 1.99,
1327
+ "learning_rate": 6.890625000000001e-05,
1328
+ "loss": 0.2731,
1329
+ "step": 220
1330
+ },
1331
+ {
1332
+ "epoch": 2.0,
1333
+ "learning_rate": 6.875e-05,
1334
+ "loss": 0.2745,
1335
+ "step": 221
1336
+ },
1337
+ {
1338
+ "epoch": 2.01,
1339
+ "learning_rate": 6.859375e-05,
1340
+ "loss": 0.2555,
1341
+ "step": 222
1342
+ },
1343
+ {
1344
+ "epoch": 2.02,
1345
+ "learning_rate": 6.84375e-05,
1346
+ "loss": 0.2398,
1347
+ "step": 223
1348
+ },
1349
+ {
1350
+ "epoch": 2.03,
1351
+ "learning_rate": 6.828125000000001e-05,
1352
+ "loss": 0.2629,
1353
+ "step": 224
1354
+ },
1355
+ {
1356
+ "epoch": 2.04,
1357
+ "learning_rate": 6.8125e-05,
1358
+ "loss": 0.2591,
1359
+ "step": 225
1360
+ },
1361
+ {
1362
+ "epoch": 2.05,
1363
+ "learning_rate": 6.796875000000001e-05,
1364
+ "loss": 0.2513,
1365
+ "step": 226
1366
+ },
1367
+ {
1368
+ "epoch": 2.06,
1369
+ "learning_rate": 6.78125e-05,
1370
+ "loss": 0.2494,
1371
+ "step": 227
1372
+ },
1373
+ {
1374
+ "epoch": 2.07,
1375
+ "learning_rate": 6.765625e-05,
1376
+ "loss": 0.255,
1377
+ "step": 228
1378
+ },
1379
+ {
1380
+ "epoch": 2.07,
1381
+ "learning_rate": 6.750000000000001e-05,
1382
+ "loss": 0.242,
1383
+ "step": 229
1384
+ },
1385
+ {
1386
+ "epoch": 2.08,
1387
+ "learning_rate": 6.734375000000001e-05,
1388
+ "loss": 0.2589,
1389
+ "step": 230
1390
+ },
1391
+ {
1392
+ "epoch": 2.09,
1393
+ "learning_rate": 6.71875e-05,
1394
+ "loss": 0.2502,
1395
+ "step": 231
1396
+ },
1397
+ {
1398
+ "epoch": 2.1,
1399
+ "learning_rate": 6.703125e-05,
1400
+ "loss": 0.2569,
1401
+ "step": 232
1402
+ },
1403
+ {
1404
+ "epoch": 2.11,
1405
+ "learning_rate": 6.6875e-05,
1406
+ "loss": 0.2545,
1407
+ "step": 233
1408
+ },
1409
+ {
1410
+ "epoch": 2.12,
1411
+ "learning_rate": 6.671875e-05,
1412
+ "loss": 0.2567,
1413
+ "step": 234
1414
+ },
1415
+ {
1416
+ "epoch": 2.13,
1417
+ "learning_rate": 6.656250000000001e-05,
1418
+ "loss": 0.2472,
1419
+ "step": 235
1420
+ },
1421
+ {
1422
+ "epoch": 2.14,
1423
+ "learning_rate": 6.640625e-05,
1424
+ "loss": 0.2692,
1425
+ "step": 236
1426
+ },
1427
+ {
1428
+ "epoch": 2.15,
1429
+ "learning_rate": 6.625e-05,
1430
+ "loss": 0.2671,
1431
+ "step": 237
1432
+ },
1433
+ {
1434
+ "epoch": 2.16,
1435
+ "learning_rate": 6.609375e-05,
1436
+ "loss": 0.2447,
1437
+ "step": 238
1438
+ },
1439
+ {
1440
+ "epoch": 2.17,
1441
+ "learning_rate": 6.593750000000001e-05,
1442
+ "loss": 0.2592,
1443
+ "step": 239
1444
+ },
1445
+ {
1446
+ "epoch": 2.17,
1447
+ "learning_rate": 6.578125e-05,
1448
+ "loss": 0.2691,
1449
+ "step": 240
1450
+ },
1451
+ {
1452
+ "epoch": 2.18,
1453
+ "learning_rate": 6.562500000000001e-05,
1454
+ "loss": 0.2636,
1455
+ "step": 241
1456
+ },
1457
+ {
1458
+ "epoch": 2.19,
1459
+ "learning_rate": 6.546875e-05,
1460
+ "loss": 0.2459,
1461
+ "step": 242
1462
+ },
1463
+ {
1464
+ "epoch": 2.2,
1465
+ "learning_rate": 6.53125e-05,
1466
+ "loss": 0.2518,
1467
+ "step": 243
1468
+ },
1469
+ {
1470
+ "epoch": 2.21,
1471
+ "learning_rate": 6.515625e-05,
1472
+ "loss": 0.2559,
1473
+ "step": 244
1474
+ },
1475
+ {
1476
+ "epoch": 2.22,
1477
+ "learning_rate": 6.500000000000001e-05,
1478
+ "loss": 0.2547,
1479
+ "step": 245
1480
+ },
1481
+ {
1482
+ "epoch": 2.23,
1483
+ "learning_rate": 6.484375e-05,
1484
+ "loss": 0.2446,
1485
+ "step": 246
1486
+ },
1487
+ {
1488
+ "epoch": 2.24,
1489
+ "learning_rate": 6.468750000000001e-05,
1490
+ "loss": 0.2532,
1491
+ "step": 247
1492
+ },
1493
+ {
1494
+ "epoch": 2.25,
1495
+ "learning_rate": 6.453125e-05,
1496
+ "loss": 0.2639,
1497
+ "step": 248
1498
+ },
1499
+ {
1500
+ "epoch": 2.26,
1501
+ "learning_rate": 6.4375e-05,
1502
+ "loss": 0.2617,
1503
+ "step": 249
1504
+ },
1505
+ {
1506
+ "epoch": 2.27,
1507
+ "learning_rate": 6.421875e-05,
1508
+ "loss": 0.2463,
1509
+ "step": 250
1510
+ },
1511
+ {
1512
+ "epoch": 2.27,
1513
+ "learning_rate": 6.40625e-05,
1514
+ "loss": 0.2521,
1515
+ "step": 251
1516
+ },
1517
+ {
1518
+ "epoch": 2.28,
1519
+ "learning_rate": 6.390625e-05,
1520
+ "loss": 0.2525,
1521
+ "step": 252
1522
+ },
1523
+ {
1524
+ "epoch": 2.29,
1525
+ "learning_rate": 6.375e-05,
1526
+ "loss": 0.2546,
1527
+ "step": 253
1528
+ },
1529
+ {
1530
+ "epoch": 2.3,
1531
+ "learning_rate": 6.359375000000001e-05,
1532
+ "loss": 0.2534,
1533
+ "step": 254
1534
+ },
1535
+ {
1536
+ "epoch": 2.31,
1537
+ "learning_rate": 6.34375e-05,
1538
+ "loss": 0.2476,
1539
+ "step": 255
1540
+ },
1541
+ {
1542
+ "epoch": 2.32,
1543
+ "learning_rate": 6.328125e-05,
1544
+ "loss": 0.2567,
1545
+ "step": 256
1546
+ },
1547
+ {
1548
+ "epoch": 2.33,
1549
+ "learning_rate": 6.3125e-05,
1550
+ "loss": 0.267,
1551
+ "step": 257
1552
+ },
1553
+ {
1554
+ "epoch": 2.34,
1555
+ "learning_rate": 6.296875e-05,
1556
+ "loss": 0.2505,
1557
+ "step": 258
1558
+ },
1559
+ {
1560
+ "epoch": 2.35,
1561
+ "learning_rate": 6.28125e-05,
1562
+ "loss": 0.2561,
1563
+ "step": 259
1564
+ },
1565
+ {
1566
+ "epoch": 2.36,
1567
+ "learning_rate": 6.265625000000001e-05,
1568
+ "loss": 0.2375,
1569
+ "step": 260
1570
+ },
1571
+ {
1572
+ "epoch": 2.36,
1573
+ "learning_rate": 6.25e-05,
1574
+ "loss": 0.2598,
1575
+ "step": 261
1576
+ },
1577
+ {
1578
+ "epoch": 2.37,
1579
+ "learning_rate": 6.234375e-05,
1580
+ "loss": 0.2616,
1581
+ "step": 262
1582
+ },
1583
+ {
1584
+ "epoch": 2.38,
1585
+ "learning_rate": 6.21875e-05,
1586
+ "loss": 0.2544,
1587
+ "step": 263
1588
+ },
1589
+ {
1590
+ "epoch": 2.39,
1591
+ "learning_rate": 6.203125000000001e-05,
1592
+ "loss": 0.2606,
1593
+ "step": 264
1594
+ },
1595
+ {
1596
+ "epoch": 2.4,
1597
+ "learning_rate": 6.1875e-05,
1598
+ "loss": 0.2566,
1599
+ "step": 265
1600
+ },
1601
+ {
1602
+ "epoch": 2.41,
1603
+ "learning_rate": 6.171875e-05,
1604
+ "loss": 0.2487,
1605
+ "step": 266
1606
+ },
1607
+ {
1608
+ "epoch": 2.42,
1609
+ "learning_rate": 6.15625e-05,
1610
+ "loss": 0.2431,
1611
+ "step": 267
1612
+ },
1613
+ {
1614
+ "epoch": 2.43,
1615
+ "learning_rate": 6.140625e-05,
1616
+ "loss": 0.2499,
1617
+ "step": 268
1618
+ },
1619
+ {
1620
+ "epoch": 2.44,
1621
+ "learning_rate": 6.125000000000001e-05,
1622
+ "loss": 0.2485,
1623
+ "step": 269
1624
+ },
1625
+ {
1626
+ "epoch": 2.45,
1627
+ "learning_rate": 6.109375e-05,
1628
+ "loss": 0.2427,
1629
+ "step": 270
1630
+ },
1631
+ {
1632
+ "epoch": 2.46,
1633
+ "learning_rate": 6.0937500000000004e-05,
1634
+ "loss": 0.2595,
1635
+ "step": 271
1636
+ },
1637
+ {
1638
+ "epoch": 2.46,
1639
+ "learning_rate": 6.0781250000000004e-05,
1640
+ "loss": 0.256,
1641
+ "step": 272
1642
+ },
1643
+ {
1644
+ "epoch": 2.47,
1645
+ "learning_rate": 6.0624999999999996e-05,
1646
+ "loss": 0.245,
1647
+ "step": 273
1648
+ },
1649
+ {
1650
+ "epoch": 2.48,
1651
+ "learning_rate": 6.046875000000001e-05,
1652
+ "loss": 0.2568,
1653
+ "step": 274
1654
+ },
1655
+ {
1656
+ "epoch": 2.49,
1657
+ "learning_rate": 6.03125e-05,
1658
+ "loss": 0.2616,
1659
+ "step": 275
1660
+ },
1661
+ {
1662
+ "epoch": 2.5,
1663
+ "learning_rate": 6.015625e-05,
1664
+ "loss": 0.2478,
1665
+ "step": 276
1666
+ },
1667
+ {
1668
+ "epoch": 2.51,
1669
+ "learning_rate": 6e-05,
1670
+ "loss": 0.2443,
1671
+ "step": 277
1672
+ },
1673
+ {
1674
+ "epoch": 2.52,
1675
+ "learning_rate": 5.984375e-05,
1676
+ "loss": 0.2574,
1677
+ "step": 278
1678
+ },
1679
+ {
1680
+ "epoch": 2.53,
1681
+ "learning_rate": 5.968750000000001e-05,
1682
+ "loss": 0.2408,
1683
+ "step": 279
1684
+ },
1685
+ {
1686
+ "epoch": 2.54,
1687
+ "learning_rate": 5.953125000000001e-05,
1688
+ "loss": 0.2521,
1689
+ "step": 280
1690
+ },
1691
+ {
1692
+ "epoch": 2.55,
1693
+ "learning_rate": 5.9375e-05,
1694
+ "loss": 0.2471,
1695
+ "step": 281
1696
+ },
1697
+ {
1698
+ "epoch": 2.55,
1699
+ "learning_rate": 5.921875e-05,
1700
+ "loss": 0.2564,
1701
+ "step": 282
1702
+ },
1703
+ {
1704
+ "epoch": 2.56,
1705
+ "learning_rate": 5.90625e-05,
1706
+ "loss": 0.2396,
1707
+ "step": 283
1708
+ },
1709
+ {
1710
+ "epoch": 2.57,
1711
+ "learning_rate": 5.8906250000000005e-05,
1712
+ "loss": 0.2537,
1713
+ "step": 284
1714
+ },
1715
+ {
1716
+ "epoch": 2.58,
1717
+ "learning_rate": 5.8750000000000005e-05,
1718
+ "loss": 0.2625,
1719
+ "step": 285
1720
+ },
1721
+ {
1722
+ "epoch": 2.59,
1723
+ "learning_rate": 5.8593750000000005e-05,
1724
+ "loss": 0.2481,
1725
+ "step": 286
1726
+ },
1727
+ {
1728
+ "epoch": 2.6,
1729
+ "learning_rate": 5.84375e-05,
1730
+ "loss": 0.2458,
1731
+ "step": 287
1732
+ },
1733
+ {
1734
+ "epoch": 2.61,
1735
+ "learning_rate": 5.828125e-05,
1736
+ "loss": 0.2324,
1737
+ "step": 288
1738
+ },
1739
+ {
1740
+ "epoch": 2.62,
1741
+ "learning_rate": 5.812500000000001e-05,
1742
+ "loss": 0.2439,
1743
+ "step": 289
1744
+ },
1745
+ {
1746
+ "epoch": 2.63,
1747
+ "learning_rate": 5.796875e-05,
1748
+ "loss": 0.2569,
1749
+ "step": 290
1750
+ },
1751
+ {
1752
+ "epoch": 2.64,
1753
+ "learning_rate": 5.78125e-05,
1754
+ "loss": 0.2571,
1755
+ "step": 291
1756
+ },
1757
+ {
1758
+ "epoch": 2.65,
1759
+ "learning_rate": 5.765625e-05,
1760
+ "loss": 0.2387,
1761
+ "step": 292
1762
+ },
1763
+ {
1764
+ "epoch": 2.65,
1765
+ "learning_rate": 5.7499999999999995e-05,
1766
+ "loss": 0.2635,
1767
+ "step": 293
1768
+ },
1769
+ {
1770
+ "epoch": 2.66,
1771
+ "learning_rate": 5.734375000000001e-05,
1772
+ "loss": 0.2511,
1773
+ "step": 294
1774
+ },
1775
+ {
1776
+ "epoch": 2.67,
1777
+ "learning_rate": 5.718750000000001e-05,
1778
+ "loss": 0.248,
1779
+ "step": 295
1780
+ },
1781
+ {
1782
+ "epoch": 2.68,
1783
+ "learning_rate": 5.703125e-05,
1784
+ "loss": 0.2533,
1785
+ "step": 296
1786
+ },
1787
+ {
1788
+ "epoch": 2.69,
1789
+ "learning_rate": 5.6875e-05,
1790
+ "loss": 0.2519,
1791
+ "step": 297
1792
+ },
1793
+ {
1794
+ "epoch": 2.7,
1795
+ "learning_rate": 5.671875e-05,
1796
+ "loss": 0.2478,
1797
+ "step": 298
1798
+ },
1799
+ {
1800
+ "epoch": 2.71,
1801
+ "learning_rate": 5.6562500000000006e-05,
1802
+ "loss": 0.2517,
1803
+ "step": 299
1804
+ },
1805
+ {
1806
+ "epoch": 2.72,
1807
+ "learning_rate": 5.6406250000000006e-05,
1808
+ "loss": 0.2414,
1809
+ "step": 300
1810
+ },
1811
+ {
1812
+ "epoch": 2.73,
1813
+ "learning_rate": 5.6250000000000005e-05,
1814
+ "loss": 0.2484,
1815
+ "step": 301
1816
+ },
1817
+ {
1818
+ "epoch": 2.74,
1819
+ "learning_rate": 5.609375e-05,
1820
+ "loss": 0.2366,
1821
+ "step": 302
1822
+ },
1823
+ {
1824
+ "epoch": 2.75,
1825
+ "learning_rate": 5.59375e-05,
1826
+ "loss": 0.2403,
1827
+ "step": 303
1828
+ },
1829
+ {
1830
+ "epoch": 2.75,
1831
+ "learning_rate": 5.5781250000000004e-05,
1832
+ "loss": 0.2389,
1833
+ "step": 304
1834
+ },
1835
+ {
1836
+ "epoch": 2.76,
1837
+ "learning_rate": 5.5625000000000004e-05,
1838
+ "loss": 0.2597,
1839
+ "step": 305
1840
+ },
1841
+ {
1842
+ "epoch": 2.77,
1843
+ "learning_rate": 5.546875e-05,
1844
+ "loss": 0.2375,
1845
+ "step": 306
1846
+ },
1847
+ {
1848
+ "epoch": 2.78,
1849
+ "learning_rate": 5.53125e-05,
1850
+ "loss": 0.2479,
1851
+ "step": 307
1852
+ },
1853
+ {
1854
+ "epoch": 2.79,
1855
+ "learning_rate": 5.5156249999999996e-05,
1856
+ "loss": 0.2686,
1857
+ "step": 308
1858
+ },
1859
+ {
1860
+ "epoch": 2.8,
1861
+ "learning_rate": 5.500000000000001e-05,
1862
+ "loss": 0.2445,
1863
+ "step": 309
1864
+ },
1865
+ {
1866
+ "epoch": 2.81,
1867
+ "learning_rate": 5.484375e-05,
1868
+ "loss": 0.2515,
1869
+ "step": 310
1870
+ },
1871
+ {
1872
+ "epoch": 2.82,
1873
+ "learning_rate": 5.46875e-05,
1874
+ "loss": 0.2632,
1875
+ "step": 311
1876
+ },
1877
+ {
1878
+ "epoch": 2.83,
1879
+ "learning_rate": 5.453125e-05,
1880
+ "loss": 0.264,
1881
+ "step": 312
1882
+ },
1883
+ {
1884
+ "epoch": 2.84,
1885
+ "learning_rate": 5.4375e-05,
1886
+ "loss": 0.2499,
1887
+ "step": 313
1888
+ },
1889
+ {
1890
+ "epoch": 2.84,
1891
+ "learning_rate": 5.421875000000001e-05,
1892
+ "loss": 0.2515,
1893
+ "step": 314
1894
+ },
1895
+ {
1896
+ "epoch": 2.85,
1897
+ "learning_rate": 5.4062500000000006e-05,
1898
+ "loss": 0.2601,
1899
+ "step": 315
1900
+ },
1901
+ {
1902
+ "epoch": 2.86,
1903
+ "learning_rate": 5.3906250000000006e-05,
1904
+ "loss": 0.2491,
1905
+ "step": 316
1906
+ },
1907
+ {
1908
+ "epoch": 2.87,
1909
+ "learning_rate": 5.375e-05,
1910
+ "loss": 0.2536,
1911
+ "step": 317
1912
+ },
1913
+ {
1914
+ "epoch": 2.88,
1915
+ "learning_rate": 5.359375e-05,
1916
+ "loss": 0.2413,
1917
+ "step": 318
1918
+ },
1919
+ {
1920
+ "epoch": 2.89,
1921
+ "learning_rate": 5.3437500000000005e-05,
1922
+ "loss": 0.2491,
1923
+ "step": 319
1924
+ },
1925
+ {
1926
+ "epoch": 2.9,
1927
+ "learning_rate": 5.3281250000000004e-05,
1928
+ "loss": 0.2569,
1929
+ "step": 320
1930
+ },
1931
+ {
1932
+ "epoch": 2.91,
1933
+ "learning_rate": 5.3125000000000004e-05,
1934
+ "loss": 0.2464,
1935
+ "step": 321
1936
+ },
1937
+ {
1938
+ "epoch": 2.92,
1939
+ "learning_rate": 5.2968750000000003e-05,
1940
+ "loss": 0.2401,
1941
+ "step": 322
1942
+ },
1943
+ {
1944
+ "epoch": 2.93,
1945
+ "learning_rate": 5.2812499999999996e-05,
1946
+ "loss": 0.248,
1947
+ "step": 323
1948
+ },
1949
+ {
1950
+ "epoch": 2.94,
1951
+ "learning_rate": 5.265625000000001e-05,
1952
+ "loss": 0.2458,
1953
+ "step": 324
1954
+ },
1955
+ {
1956
+ "epoch": 2.94,
1957
+ "learning_rate": 5.25e-05,
1958
+ "loss": 0.2559,
1959
+ "step": 325
1960
+ },
1961
+ {
1962
+ "epoch": 2.95,
1963
+ "learning_rate": 5.234375e-05,
1964
+ "loss": 0.2405,
1965
+ "step": 326
1966
+ },
1967
+ {
1968
+ "epoch": 2.96,
1969
+ "learning_rate": 5.21875e-05,
1970
+ "loss": 0.2463,
1971
+ "step": 327
1972
+ },
1973
+ {
1974
+ "epoch": 2.97,
1975
+ "learning_rate": 5.203125e-05,
1976
+ "loss": 0.247,
1977
+ "step": 328
1978
+ },
1979
+ {
1980
+ "epoch": 2.98,
1981
+ "learning_rate": 5.187500000000001e-05,
1982
+ "loss": 0.2456,
1983
+ "step": 329
1984
+ },
1985
+ {
1986
+ "epoch": 2.99,
1987
+ "learning_rate": 5.171875000000001e-05,
1988
+ "loss": 0.2368,
1989
+ "step": 330
1990
+ },
1991
+ {
1992
+ "epoch": 3.0,
1993
+ "learning_rate": 5.15625e-05,
1994
+ "loss": 0.2357,
1995
+ "step": 331
1996
+ },
1997
+ {
1998
+ "epoch": 3.01,
1999
+ "learning_rate": 5.140625e-05,
2000
+ "loss": 0.2384,
2001
+ "step": 332
2002
+ },
2003
+ {
2004
+ "epoch": 3.02,
2005
+ "learning_rate": 5.125e-05,
2006
+ "loss": 0.2318,
2007
+ "step": 333
2008
+ },
2009
+ {
2010
+ "epoch": 3.03,
2011
+ "learning_rate": 5.1093750000000005e-05,
2012
+ "loss": 0.2302,
2013
+ "step": 334
2014
+ },
2015
+ {
2016
+ "epoch": 3.04,
2017
+ "learning_rate": 5.0937500000000005e-05,
2018
+ "loss": 0.2122,
2019
+ "step": 335
2020
+ },
2021
+ {
2022
+ "epoch": 3.04,
2023
+ "learning_rate": 5.0781250000000004e-05,
2024
+ "loss": 0.2197,
2025
+ "step": 336
2026
+ },
2027
+ {
2028
+ "epoch": 3.05,
2029
+ "learning_rate": 5.0625e-05,
2030
+ "loss": 0.2179,
2031
+ "step": 337
2032
+ },
2033
+ {
2034
+ "epoch": 3.06,
2035
+ "learning_rate": 5.046875e-05,
2036
+ "loss": 0.2252,
2037
+ "step": 338
2038
+ },
2039
+ {
2040
+ "epoch": 3.07,
2041
+ "learning_rate": 5.031250000000001e-05,
2042
+ "loss": 0.2215,
2043
+ "step": 339
2044
+ },
2045
+ {
2046
+ "epoch": 3.08,
2047
+ "learning_rate": 5.015625e-05,
2048
+ "loss": 0.2264,
2049
+ "step": 340
2050
+ },
2051
+ {
2052
+ "epoch": 3.09,
2053
+ "learning_rate": 5e-05,
2054
+ "loss": 0.2301,
2055
+ "step": 341
2056
+ },
2057
+ {
2058
+ "epoch": 3.1,
2059
+ "learning_rate": 4.984375e-05,
2060
+ "loss": 0.2261,
2061
+ "step": 342
2062
+ },
2063
+ {
2064
+ "epoch": 3.11,
2065
+ "learning_rate": 4.96875e-05,
2066
+ "loss": 0.2185,
2067
+ "step": 343
2068
+ },
2069
+ {
2070
+ "epoch": 3.12,
2071
+ "learning_rate": 4.953125e-05,
2072
+ "loss": 0.2261,
2073
+ "step": 344
2074
+ },
2075
+ {
2076
+ "epoch": 3.13,
2077
+ "learning_rate": 4.937500000000001e-05,
2078
+ "loss": 0.2301,
2079
+ "step": 345
2080
+ },
2081
+ {
2082
+ "epoch": 3.13,
2083
+ "learning_rate": 4.921875e-05,
2084
+ "loss": 0.2214,
2085
+ "step": 346
2086
+ },
2087
+ {
2088
+ "epoch": 3.14,
2089
+ "learning_rate": 4.90625e-05,
2090
+ "loss": 0.2154,
2091
+ "step": 347
2092
+ },
2093
+ {
2094
+ "epoch": 3.15,
2095
+ "learning_rate": 4.8906250000000006e-05,
2096
+ "loss": 0.2229,
2097
+ "step": 348
2098
+ },
2099
+ {
2100
+ "epoch": 3.16,
2101
+ "learning_rate": 4.875e-05,
2102
+ "loss": 0.207,
2103
+ "step": 349
2104
+ },
2105
+ {
2106
+ "epoch": 3.17,
2107
+ "learning_rate": 4.8593750000000005e-05,
2108
+ "loss": 0.225,
2109
+ "step": 350
2110
+ },
2111
+ {
2112
+ "epoch": 3.18,
2113
+ "learning_rate": 4.8437500000000005e-05,
2114
+ "loss": 0.2327,
2115
+ "step": 351
2116
+ },
2117
+ {
2118
+ "epoch": 3.19,
2119
+ "learning_rate": 4.828125e-05,
2120
+ "loss": 0.2185,
2121
+ "step": 352
2122
+ },
2123
+ {
2124
+ "epoch": 3.2,
2125
+ "learning_rate": 4.8125000000000004e-05,
2126
+ "loss": 0.2171,
2127
+ "step": 353
2128
+ },
2129
+ {
2130
+ "epoch": 3.21,
2131
+ "learning_rate": 4.7968750000000004e-05,
2132
+ "loss": 0.2226,
2133
+ "step": 354
2134
+ },
2135
+ {
2136
+ "epoch": 3.22,
2137
+ "learning_rate": 4.7812500000000003e-05,
2138
+ "loss": 0.2392,
2139
+ "step": 355
2140
+ },
2141
+ {
2142
+ "epoch": 3.23,
2143
+ "learning_rate": 4.765625e-05,
2144
+ "loss": 0.2243,
2145
+ "step": 356
2146
+ },
2147
+ {
2148
+ "epoch": 3.23,
2149
+ "learning_rate": 4.75e-05,
2150
+ "loss": 0.2244,
2151
+ "step": 357
2152
+ },
2153
+ {
2154
+ "epoch": 3.24,
2155
+ "learning_rate": 4.734375e-05,
2156
+ "loss": 0.2264,
2157
+ "step": 358
2158
+ },
2159
+ {
2160
+ "epoch": 3.25,
2161
+ "learning_rate": 4.71875e-05,
2162
+ "loss": 0.2188,
2163
+ "step": 359
2164
+ },
2165
+ {
2166
+ "epoch": 3.26,
2167
+ "learning_rate": 4.703125e-05,
2168
+ "loss": 0.2282,
2169
+ "step": 360
2170
+ },
2171
+ {
2172
+ "epoch": 3.27,
2173
+ "learning_rate": 4.6875e-05,
2174
+ "loss": 0.2259,
2175
+ "step": 361
2176
+ },
2177
+ {
2178
+ "epoch": 3.28,
2179
+ "learning_rate": 4.671875e-05,
2180
+ "loss": 0.2252,
2181
+ "step": 362
2182
+ },
2183
+ {
2184
+ "epoch": 3.29,
2185
+ "learning_rate": 4.65625e-05,
2186
+ "loss": 0.2294,
2187
+ "step": 363
2188
+ },
2189
+ {
2190
+ "epoch": 3.3,
2191
+ "learning_rate": 4.640625e-05,
2192
+ "loss": 0.2233,
2193
+ "step": 364
2194
+ },
2195
+ {
2196
+ "epoch": 3.31,
2197
+ "learning_rate": 4.6250000000000006e-05,
2198
+ "loss": 0.2262,
2199
+ "step": 365
2200
+ },
2201
+ {
2202
+ "epoch": 3.32,
2203
+ "learning_rate": 4.609375e-05,
2204
+ "loss": 0.2172,
2205
+ "step": 366
2206
+ },
2207
+ {
2208
+ "epoch": 3.33,
2209
+ "learning_rate": 4.59375e-05,
2210
+ "loss": 0.2203,
2211
+ "step": 367
2212
+ },
2213
+ {
2214
+ "epoch": 3.33,
2215
+ "learning_rate": 4.5781250000000005e-05,
2216
+ "loss": 0.2225,
2217
+ "step": 368
2218
+ },
2219
+ {
2220
+ "epoch": 3.34,
2221
+ "learning_rate": 4.5625e-05,
2222
+ "loss": 0.2334,
2223
+ "step": 369
2224
+ },
2225
+ {
2226
+ "epoch": 3.35,
2227
+ "learning_rate": 4.5468750000000004e-05,
2228
+ "loss": 0.2295,
2229
+ "step": 370
2230
+ },
2231
+ {
2232
+ "epoch": 3.36,
2233
+ "learning_rate": 4.5312500000000004e-05,
2234
+ "loss": 0.2177,
2235
+ "step": 371
2236
+ },
2237
+ {
2238
+ "epoch": 3.37,
2239
+ "learning_rate": 4.515625e-05,
2240
+ "loss": 0.2229,
2241
+ "step": 372
2242
+ },
2243
+ {
2244
+ "epoch": 3.38,
2245
+ "learning_rate": 4.5e-05,
2246
+ "loss": 0.2211,
2247
+ "step": 373
2248
+ },
2249
+ {
2250
+ "epoch": 3.39,
2251
+ "learning_rate": 4.484375e-05,
2252
+ "loss": 0.2264,
2253
+ "step": 374
2254
+ },
2255
+ {
2256
+ "epoch": 3.4,
2257
+ "learning_rate": 4.46875e-05,
2258
+ "loss": 0.2244,
2259
+ "step": 375
2260
+ },
2261
+ {
2262
+ "epoch": 3.41,
2263
+ "learning_rate": 4.453125e-05,
2264
+ "loss": 0.221,
2265
+ "step": 376
2266
+ },
2267
+ {
2268
+ "epoch": 3.42,
2269
+ "learning_rate": 4.4375e-05,
2270
+ "loss": 0.2177,
2271
+ "step": 377
2272
+ },
2273
+ {
2274
+ "epoch": 3.42,
2275
+ "learning_rate": 4.421875e-05,
2276
+ "loss": 0.2134,
2277
+ "step": 378
2278
+ },
2279
+ {
2280
+ "epoch": 3.43,
2281
+ "learning_rate": 4.40625e-05,
2282
+ "loss": 0.2134,
2283
+ "step": 379
2284
+ },
2285
+ {
2286
+ "epoch": 3.44,
2287
+ "learning_rate": 4.390625000000001e-05,
2288
+ "loss": 0.2167,
2289
+ "step": 380
2290
+ },
2291
+ {
2292
+ "epoch": 3.45,
2293
+ "learning_rate": 4.375e-05,
2294
+ "loss": 0.2227,
2295
+ "step": 381
2296
+ },
2297
+ {
2298
+ "epoch": 3.46,
2299
+ "learning_rate": 4.359375e-05,
2300
+ "loss": 0.218,
2301
+ "step": 382
2302
+ },
2303
+ {
2304
+ "epoch": 3.47,
2305
+ "learning_rate": 4.3437500000000006e-05,
2306
+ "loss": 0.2215,
2307
+ "step": 383
2308
+ },
2309
+ {
2310
+ "epoch": 3.48,
2311
+ "learning_rate": 4.328125e-05,
2312
+ "loss": 0.225,
2313
+ "step": 384
2314
+ },
2315
+ {
2316
+ "epoch": 3.49,
2317
+ "learning_rate": 4.3125000000000005e-05,
2318
+ "loss": 0.2111,
2319
+ "step": 385
2320
+ },
2321
+ {
2322
+ "epoch": 3.5,
2323
+ "learning_rate": 4.2968750000000004e-05,
2324
+ "loss": 0.225,
2325
+ "step": 386
2326
+ },
2327
+ {
2328
+ "epoch": 3.51,
2329
+ "learning_rate": 4.28125e-05,
2330
+ "loss": 0.2306,
2331
+ "step": 387
2332
+ },
2333
+ {
2334
+ "epoch": 3.52,
2335
+ "learning_rate": 4.2656250000000003e-05,
2336
+ "loss": 0.2225,
2337
+ "step": 388
2338
+ },
2339
+ {
2340
+ "epoch": 3.52,
2341
+ "learning_rate": 4.25e-05,
2342
+ "loss": 0.2276,
2343
+ "step": 389
2344
+ },
2345
+ {
2346
+ "epoch": 3.53,
2347
+ "learning_rate": 4.234375e-05,
2348
+ "loss": 0.2286,
2349
+ "step": 390
2350
+ },
2351
+ {
2352
+ "epoch": 3.54,
2353
+ "learning_rate": 4.21875e-05,
2354
+ "loss": 0.2198,
2355
+ "step": 391
2356
+ },
2357
+ {
2358
+ "epoch": 3.55,
2359
+ "learning_rate": 4.203125e-05,
2360
+ "loss": 0.2236,
2361
+ "step": 392
2362
+ },
2363
+ {
2364
+ "epoch": 3.56,
2365
+ "learning_rate": 4.1875e-05,
2366
+ "loss": 0.221,
2367
+ "step": 393
2368
+ },
2369
+ {
2370
+ "epoch": 3.57,
2371
+ "learning_rate": 4.171875e-05,
2372
+ "loss": 0.2157,
2373
+ "step": 394
2374
+ },
2375
+ {
2376
+ "epoch": 3.58,
2377
+ "learning_rate": 4.156250000000001e-05,
2378
+ "loss": 0.2331,
2379
+ "step": 395
2380
+ },
2381
+ {
2382
+ "epoch": 3.59,
2383
+ "learning_rate": 4.140625e-05,
2384
+ "loss": 0.2218,
2385
+ "step": 396
2386
+ },
2387
+ {
2388
+ "epoch": 3.6,
2389
+ "learning_rate": 4.125e-05,
2390
+ "loss": 0.2366,
2391
+ "step": 397
2392
+ },
2393
+ {
2394
+ "epoch": 3.61,
2395
+ "learning_rate": 4.1093750000000006e-05,
2396
+ "loss": 0.2226,
2397
+ "step": 398
2398
+ },
2399
+ {
2400
+ "epoch": 3.61,
2401
+ "learning_rate": 4.09375e-05,
2402
+ "loss": 0.2303,
2403
+ "step": 399
2404
+ },
2405
+ {
2406
+ "epoch": 3.62,
2407
+ "learning_rate": 4.0781250000000005e-05,
2408
+ "loss": 0.217,
2409
+ "step": 400
2410
+ },
2411
+ {
2412
+ "epoch": 3.63,
2413
+ "learning_rate": 4.0625000000000005e-05,
2414
+ "loss": 0.217,
2415
+ "step": 401
2416
+ },
2417
+ {
2418
+ "epoch": 3.64,
2419
+ "learning_rate": 4.046875e-05,
2420
+ "loss": 0.2244,
2421
+ "step": 402
2422
+ },
2423
+ {
2424
+ "epoch": 3.65,
2425
+ "learning_rate": 4.0312500000000004e-05,
2426
+ "loss": 0.2235,
2427
+ "step": 403
2428
+ },
2429
+ {
2430
+ "epoch": 3.66,
2431
+ "learning_rate": 4.0156250000000004e-05,
2432
+ "loss": 0.2192,
2433
+ "step": 404
2434
+ },
2435
+ {
2436
+ "epoch": 3.67,
2437
+ "learning_rate": 4e-05,
2438
+ "loss": 0.2422,
2439
+ "step": 405
2440
+ },
2441
+ {
2442
+ "epoch": 3.68,
2443
+ "learning_rate": 3.984375e-05,
2444
+ "loss": 0.2214,
2445
+ "step": 406
2446
+ },
2447
+ {
2448
+ "epoch": 3.69,
2449
+ "learning_rate": 3.96875e-05,
2450
+ "loss": 0.2209,
2451
+ "step": 407
2452
+ },
2453
+ {
2454
+ "epoch": 3.7,
2455
+ "learning_rate": 3.953125e-05,
2456
+ "loss": 0.2208,
2457
+ "step": 408
2458
+ },
2459
+ {
2460
+ "epoch": 3.71,
2461
+ "learning_rate": 3.9375e-05,
2462
+ "loss": 0.2281,
2463
+ "step": 409
2464
+ },
2465
+ {
2466
+ "epoch": 3.71,
2467
+ "learning_rate": 3.921875e-05,
2468
+ "loss": 0.2142,
2469
+ "step": 410
2470
+ },
2471
+ {
2472
+ "epoch": 3.72,
2473
+ "learning_rate": 3.90625e-05,
2474
+ "loss": 0.2306,
2475
+ "step": 411
2476
+ },
2477
+ {
2478
+ "epoch": 3.73,
2479
+ "learning_rate": 3.890625e-05,
2480
+ "loss": 0.2209,
2481
+ "step": 412
2482
+ },
2483
+ {
2484
+ "epoch": 3.74,
2485
+ "learning_rate": 3.875e-05,
2486
+ "loss": 0.2183,
2487
+ "step": 413
2488
+ },
2489
+ {
2490
+ "epoch": 3.75,
2491
+ "learning_rate": 3.859375e-05,
2492
+ "loss": 0.2211,
2493
+ "step": 414
2494
+ },
2495
+ {
2496
+ "epoch": 3.76,
2497
+ "learning_rate": 3.8437500000000006e-05,
2498
+ "loss": 0.2134,
2499
+ "step": 415
2500
+ },
2501
+ {
2502
+ "epoch": 3.77,
2503
+ "learning_rate": 3.828125e-05,
2504
+ "loss": 0.2095,
2505
+ "step": 416
2506
+ },
2507
+ {
2508
+ "epoch": 3.78,
2509
+ "learning_rate": 3.8125e-05,
2510
+ "loss": 0.2199,
2511
+ "step": 417
2512
+ },
2513
+ {
2514
+ "epoch": 3.79,
2515
+ "learning_rate": 3.7968750000000005e-05,
2516
+ "loss": 0.2185,
2517
+ "step": 418
2518
+ },
2519
+ {
2520
+ "epoch": 3.8,
2521
+ "learning_rate": 3.78125e-05,
2522
+ "loss": 0.2241,
2523
+ "step": 419
2524
+ },
2525
+ {
2526
+ "epoch": 3.81,
2527
+ "learning_rate": 3.7656250000000004e-05,
2528
+ "loss": 0.2184,
2529
+ "step": 420
2530
+ },
2531
+ {
2532
+ "epoch": 3.81,
2533
+ "learning_rate": 3.7500000000000003e-05,
2534
+ "loss": 0.2205,
2535
+ "step": 421
2536
+ },
2537
+ {
2538
+ "epoch": 3.82,
2539
+ "learning_rate": 3.7343749999999996e-05,
2540
+ "loss": 0.2225,
2541
+ "step": 422
2542
+ },
2543
+ {
2544
+ "epoch": 3.83,
2545
+ "learning_rate": 3.71875e-05,
2546
+ "loss": 0.2214,
2547
+ "step": 423
2548
+ },
2549
+ {
2550
+ "epoch": 3.84,
2551
+ "learning_rate": 3.703125e-05,
2552
+ "loss": 0.2139,
2553
+ "step": 424
2554
+ },
2555
+ {
2556
+ "epoch": 3.85,
2557
+ "learning_rate": 3.6875e-05,
2558
+ "loss": 0.2122,
2559
+ "step": 425
2560
+ },
2561
+ {
2562
+ "epoch": 3.86,
2563
+ "learning_rate": 3.671875e-05,
2564
+ "loss": 0.2242,
2565
+ "step": 426
2566
+ },
2567
+ {
2568
+ "epoch": 3.87,
2569
+ "learning_rate": 3.65625e-05,
2570
+ "loss": 0.2165,
2571
+ "step": 427
2572
+ },
2573
+ {
2574
+ "epoch": 3.88,
2575
+ "learning_rate": 3.640625e-05,
2576
+ "loss": 0.2284,
2577
+ "step": 428
2578
+ },
2579
+ {
2580
+ "epoch": 3.89,
2581
+ "learning_rate": 3.625e-05,
2582
+ "loss": 0.2237,
2583
+ "step": 429
2584
+ },
2585
+ {
2586
+ "epoch": 3.9,
2587
+ "learning_rate": 3.6093750000000007e-05,
2588
+ "loss": 0.2216,
2589
+ "step": 430
2590
+ },
2591
+ {
2592
+ "epoch": 3.9,
2593
+ "learning_rate": 3.59375e-05,
2594
+ "loss": 0.2222,
2595
+ "step": 431
2596
+ },
2597
+ {
2598
+ "epoch": 3.91,
2599
+ "learning_rate": 3.578125e-05,
2600
+ "loss": 0.2228,
2601
+ "step": 432
2602
+ },
2603
+ {
2604
+ "epoch": 3.92,
2605
+ "learning_rate": 3.5625000000000005e-05,
2606
+ "loss": 0.2146,
2607
+ "step": 433
2608
+ },
2609
+ {
2610
+ "epoch": 3.93,
2611
+ "learning_rate": 3.546875e-05,
2612
+ "loss": 0.2327,
2613
+ "step": 434
2614
+ },
2615
+ {
2616
+ "epoch": 3.94,
2617
+ "learning_rate": 3.5312500000000005e-05,
2618
+ "loss": 0.2288,
2619
+ "step": 435
2620
+ },
2621
+ {
2622
+ "epoch": 3.95,
2623
+ "learning_rate": 3.5156250000000004e-05,
2624
+ "loss": 0.212,
2625
+ "step": 436
2626
+ },
2627
+ {
2628
+ "epoch": 3.96,
2629
+ "learning_rate": 3.5e-05,
2630
+ "loss": 0.2223,
2631
+ "step": 437
2632
+ },
2633
+ {
2634
+ "epoch": 3.97,
2635
+ "learning_rate": 3.484375e-05,
2636
+ "loss": 0.2119,
2637
+ "step": 438
2638
+ },
2639
+ {
2640
+ "epoch": 3.98,
2641
+ "learning_rate": 3.46875e-05,
2642
+ "loss": 0.2102,
2643
+ "step": 439
2644
+ },
2645
+ {
2646
+ "epoch": 3.99,
2647
+ "learning_rate": 3.453125e-05,
2648
+ "loss": 0.2165,
2649
+ "step": 440
2650
+ },
2651
+ {
2652
+ "epoch": 4.0,
2653
+ "learning_rate": 3.4375e-05,
2654
+ "loss": 0.2255,
2655
+ "step": 441
2656
+ },
2657
+ {
2658
+ "epoch": 4.0,
2659
+ "learning_rate": 3.421875e-05,
2660
+ "loss": 0.2115,
2661
+ "step": 442
2662
+ },
2663
+ {
2664
+ "epoch": 4.01,
2665
+ "learning_rate": 3.40625e-05,
2666
+ "loss": 0.1965,
2667
+ "step": 443
2668
+ },
2669
+ {
2670
+ "epoch": 4.02,
2671
+ "learning_rate": 3.390625e-05,
2672
+ "loss": 0.1968,
2673
+ "step": 444
2674
+ },
2675
+ {
2676
+ "epoch": 4.03,
2677
+ "learning_rate": 3.375000000000001e-05,
2678
+ "loss": 0.1938,
2679
+ "step": 445
2680
+ },
2681
+ {
2682
+ "epoch": 4.04,
2683
+ "learning_rate": 3.359375e-05,
2684
+ "loss": 0.2004,
2685
+ "step": 446
2686
+ },
2687
+ {
2688
+ "epoch": 4.05,
2689
+ "learning_rate": 3.34375e-05,
2690
+ "loss": 0.1899,
2691
+ "step": 447
2692
+ },
2693
+ {
2694
+ "epoch": 4.06,
2695
+ "learning_rate": 3.3281250000000006e-05,
2696
+ "loss": 0.1927,
2697
+ "step": 448
2698
+ },
2699
+ {
2700
+ "epoch": 4.07,
2701
+ "learning_rate": 3.3125e-05,
2702
+ "loss": 0.1937,
2703
+ "step": 449
2704
+ },
2705
+ {
2706
+ "epoch": 4.08,
2707
+ "learning_rate": 3.2968750000000005e-05,
2708
+ "loss": 0.1958,
2709
+ "step": 450
2710
+ },
2711
+ {
2712
+ "epoch": 4.09,
2713
+ "learning_rate": 3.2812500000000005e-05,
2714
+ "loss": 0.1998,
2715
+ "step": 451
2716
+ },
2717
+ {
2718
+ "epoch": 4.1,
2719
+ "learning_rate": 3.265625e-05,
2720
+ "loss": 0.193,
2721
+ "step": 452
2722
+ },
2723
+ {
2724
+ "epoch": 4.1,
2725
+ "learning_rate": 3.2500000000000004e-05,
2726
+ "loss": 0.1983,
2727
+ "step": 453
2728
+ },
2729
+ {
2730
+ "epoch": 4.11,
2731
+ "learning_rate": 3.2343750000000004e-05,
2732
+ "loss": 0.1998,
2733
+ "step": 454
2734
+ },
2735
+ {
2736
+ "epoch": 4.12,
2737
+ "learning_rate": 3.21875e-05,
2738
+ "loss": 0.1976,
2739
+ "step": 455
2740
+ },
2741
+ {
2742
+ "epoch": 4.13,
2743
+ "learning_rate": 3.203125e-05,
2744
+ "loss": 0.202,
2745
+ "step": 456
2746
+ },
2747
+ {
2748
+ "epoch": 4.14,
2749
+ "learning_rate": 3.1875e-05,
2750
+ "loss": 0.1915,
2751
+ "step": 457
2752
+ },
2753
+ {
2754
+ "epoch": 4.15,
2755
+ "learning_rate": 3.171875e-05,
2756
+ "loss": 0.2,
2757
+ "step": 458
2758
+ },
2759
+ {
2760
+ "epoch": 4.16,
2761
+ "learning_rate": 3.15625e-05,
2762
+ "loss": 0.2018,
2763
+ "step": 459
2764
+ },
2765
+ {
2766
+ "epoch": 4.17,
2767
+ "learning_rate": 3.140625e-05,
2768
+ "loss": 0.1951,
2769
+ "step": 460
2770
+ },
2771
+ {
2772
+ "epoch": 4.18,
2773
+ "learning_rate": 3.125e-05,
2774
+ "loss": 0.1956,
2775
+ "step": 461
2776
+ },
2777
+ {
2778
+ "epoch": 4.19,
2779
+ "learning_rate": 3.109375e-05,
2780
+ "loss": 0.2104,
2781
+ "step": 462
2782
+ },
2783
+ {
2784
+ "epoch": 4.19,
2785
+ "learning_rate": 3.09375e-05,
2786
+ "loss": 0.194,
2787
+ "step": 463
2788
+ },
2789
+ {
2790
+ "epoch": 4.2,
2791
+ "learning_rate": 3.078125e-05,
2792
+ "loss": 0.2056,
2793
+ "step": 464
2794
+ },
2795
+ {
2796
+ "epoch": 4.21,
2797
+ "learning_rate": 3.0625000000000006e-05,
2798
+ "loss": 0.1899,
2799
+ "step": 465
2800
+ },
2801
+ {
2802
+ "epoch": 4.22,
2803
+ "learning_rate": 3.0468750000000002e-05,
2804
+ "loss": 0.1991,
2805
+ "step": 466
2806
+ },
2807
+ {
2808
+ "epoch": 4.23,
2809
+ "learning_rate": 3.0312499999999998e-05,
2810
+ "loss": 0.1982,
2811
+ "step": 467
2812
+ },
2813
+ {
2814
+ "epoch": 4.24,
2815
+ "learning_rate": 3.015625e-05,
2816
+ "loss": 0.1881,
2817
+ "step": 468
2818
+ },
2819
+ {
2820
+ "epoch": 4.25,
2821
+ "learning_rate": 3e-05,
2822
+ "loss": 0.1963,
2823
+ "step": 469
2824
+ },
2825
+ {
2826
+ "epoch": 4.26,
2827
+ "learning_rate": 2.9843750000000004e-05,
2828
+ "loss": 0.2032,
2829
+ "step": 470
2830
+ },
2831
+ {
2832
+ "epoch": 4.27,
2833
+ "learning_rate": 2.96875e-05,
2834
+ "loss": 0.2019,
2835
+ "step": 471
2836
+ },
2837
+ {
2838
+ "epoch": 4.28,
2839
+ "learning_rate": 2.953125e-05,
2840
+ "loss": 0.1884,
2841
+ "step": 472
2842
+ },
2843
+ {
2844
+ "epoch": 4.29,
2845
+ "learning_rate": 2.9375000000000003e-05,
2846
+ "loss": 0.1904,
2847
+ "step": 473
2848
+ },
2849
+ {
2850
+ "epoch": 4.29,
2851
+ "learning_rate": 2.921875e-05,
2852
+ "loss": 0.1952,
2853
+ "step": 474
2854
+ },
2855
+ {
2856
+ "epoch": 4.3,
2857
+ "learning_rate": 2.9062500000000005e-05,
2858
+ "loss": 0.1976,
2859
+ "step": 475
2860
+ },
2861
+ {
2862
+ "epoch": 4.31,
2863
+ "learning_rate": 2.890625e-05,
2864
+ "loss": 0.2013,
2865
+ "step": 476
2866
+ },
2867
+ {
2868
+ "epoch": 4.32,
2869
+ "learning_rate": 2.8749999999999997e-05,
2870
+ "loss": 0.1916,
2871
+ "step": 477
2872
+ },
2873
+ {
2874
+ "epoch": 4.33,
2875
+ "learning_rate": 2.8593750000000004e-05,
2876
+ "loss": 0.196,
2877
+ "step": 478
2878
+ },
2879
+ {
2880
+ "epoch": 4.34,
2881
+ "learning_rate": 2.84375e-05,
2882
+ "loss": 0.1932,
2883
+ "step": 479
2884
+ },
2885
+ {
2886
+ "epoch": 4.35,
2887
+ "learning_rate": 2.8281250000000003e-05,
2888
+ "loss": 0.2052,
2889
+ "step": 480
2890
+ },
2891
+ {
2892
+ "epoch": 4.36,
2893
+ "learning_rate": 2.8125000000000003e-05,
2894
+ "loss": 0.2001,
2895
+ "step": 481
2896
+ },
2897
+ {
2898
+ "epoch": 4.37,
2899
+ "learning_rate": 2.796875e-05,
2900
+ "loss": 0.205,
2901
+ "step": 482
2902
+ },
2903
+ {
2904
+ "epoch": 4.38,
2905
+ "learning_rate": 2.7812500000000002e-05,
2906
+ "loss": 0.1903,
2907
+ "step": 483
2908
+ },
2909
+ {
2910
+ "epoch": 4.39,
2911
+ "learning_rate": 2.765625e-05,
2912
+ "loss": 0.1879,
2913
+ "step": 484
2914
+ },
2915
+ {
2916
+ "epoch": 4.39,
2917
+ "learning_rate": 2.7500000000000004e-05,
2918
+ "loss": 0.1959,
2919
+ "step": 485
2920
+ },
2921
+ {
2922
+ "epoch": 4.4,
2923
+ "learning_rate": 2.734375e-05,
2924
+ "loss": 0.1959,
2925
+ "step": 486
2926
+ },
2927
+ {
2928
+ "epoch": 4.41,
2929
+ "learning_rate": 2.71875e-05,
2930
+ "loss": 0.2061,
2931
+ "step": 487
2932
+ },
2933
+ {
2934
+ "epoch": 4.42,
2935
+ "learning_rate": 2.7031250000000003e-05,
2936
+ "loss": 0.1899,
2937
+ "step": 488
2938
+ },
2939
+ {
2940
+ "epoch": 4.43,
2941
+ "learning_rate": 2.6875e-05,
2942
+ "loss": 0.2051,
2943
+ "step": 489
2944
+ },
2945
+ {
2946
+ "epoch": 4.44,
2947
+ "learning_rate": 2.6718750000000002e-05,
2948
+ "loss": 0.1923,
2949
+ "step": 490
2950
+ },
2951
+ {
2952
+ "epoch": 4.45,
2953
+ "learning_rate": 2.6562500000000002e-05,
2954
+ "loss": 0.2023,
2955
+ "step": 491
2956
+ },
2957
+ {
2958
+ "epoch": 4.46,
2959
+ "learning_rate": 2.6406249999999998e-05,
2960
+ "loss": 0.1866,
2961
+ "step": 492
2962
+ },
2963
+ {
2964
+ "epoch": 4.47,
2965
+ "learning_rate": 2.625e-05,
2966
+ "loss": 0.194,
2967
+ "step": 493
2968
+ },
2969
+ {
2970
+ "epoch": 4.48,
2971
+ "learning_rate": 2.609375e-05,
2972
+ "loss": 0.1918,
2973
+ "step": 494
2974
+ },
2975
+ {
2976
+ "epoch": 4.48,
2977
+ "learning_rate": 2.5937500000000004e-05,
2978
+ "loss": 0.1911,
2979
+ "step": 495
2980
+ },
2981
+ {
2982
+ "epoch": 4.49,
2983
+ "learning_rate": 2.578125e-05,
2984
+ "loss": 0.1922,
2985
+ "step": 496
2986
+ },
2987
+ {
2988
+ "epoch": 4.5,
2989
+ "learning_rate": 2.5625e-05,
2990
+ "loss": 0.182,
2991
+ "step": 497
2992
+ },
2993
+ {
2994
+ "epoch": 4.51,
2995
+ "learning_rate": 2.5468750000000002e-05,
2996
+ "loss": 0.186,
2997
+ "step": 498
2998
+ },
2999
+ {
3000
+ "epoch": 4.52,
3001
+ "learning_rate": 2.53125e-05,
3002
+ "loss": 0.2017,
3003
+ "step": 499
3004
+ },
3005
+ {
3006
+ "epoch": 4.53,
3007
+ "learning_rate": 2.5156250000000005e-05,
3008
+ "loss": 0.195,
3009
+ "step": 500
3010
+ },
3011
+ {
3012
+ "epoch": 4.54,
3013
+ "learning_rate": 2.5e-05,
3014
+ "loss": 0.1876,
3015
+ "step": 501
3016
+ },
3017
+ {
3018
+ "epoch": 4.55,
3019
+ "learning_rate": 2.484375e-05,
3020
+ "loss": 0.1914,
3021
+ "step": 502
3022
+ },
3023
+ {
3024
+ "epoch": 4.56,
3025
+ "learning_rate": 2.4687500000000004e-05,
3026
+ "loss": 0.1952,
3027
+ "step": 503
3028
+ },
3029
+ {
3030
+ "epoch": 4.57,
3031
+ "learning_rate": 2.453125e-05,
3032
+ "loss": 0.2089,
3033
+ "step": 504
3034
+ },
3035
+ {
3036
+ "epoch": 4.58,
3037
+ "learning_rate": 2.4375e-05,
3038
+ "loss": 0.1887,
3039
+ "step": 505
3040
+ },
3041
+ {
3042
+ "epoch": 4.58,
3043
+ "learning_rate": 2.4218750000000003e-05,
3044
+ "loss": 0.1899,
3045
+ "step": 506
3046
+ },
3047
+ {
3048
+ "epoch": 4.59,
3049
+ "learning_rate": 2.4062500000000002e-05,
3050
+ "loss": 0.1982,
3051
+ "step": 507
3052
+ },
3053
+ {
3054
+ "epoch": 4.6,
3055
+ "learning_rate": 2.3906250000000002e-05,
3056
+ "loss": 0.1978,
3057
+ "step": 508
3058
+ },
3059
+ {
3060
+ "epoch": 4.61,
3061
+ "learning_rate": 2.375e-05,
3062
+ "loss": 0.1909,
3063
+ "step": 509
3064
+ },
3065
+ {
3066
+ "epoch": 4.62,
3067
+ "learning_rate": 2.359375e-05,
3068
+ "loss": 0.1975,
3069
+ "step": 510
3070
+ },
3071
+ {
3072
+ "epoch": 4.63,
3073
+ "learning_rate": 2.34375e-05,
3074
+ "loss": 0.1954,
3075
+ "step": 511
3076
+ },
3077
+ {
3078
+ "epoch": 4.64,
3079
+ "learning_rate": 2.328125e-05,
3080
+ "loss": 0.2066,
3081
+ "step": 512
3082
+ },
3083
+ {
3084
+ "epoch": 4.65,
3085
+ "learning_rate": 2.3125000000000003e-05,
3086
+ "loss": 0.1919,
3087
+ "step": 513
3088
+ },
3089
+ {
3090
+ "epoch": 4.66,
3091
+ "learning_rate": 2.296875e-05,
3092
+ "loss": 0.1987,
3093
+ "step": 514
3094
+ },
3095
+ {
3096
+ "epoch": 4.67,
3097
+ "learning_rate": 2.28125e-05,
3098
+ "loss": 0.1942,
3099
+ "step": 515
3100
+ },
3101
+ {
3102
+ "epoch": 4.67,
3103
+ "learning_rate": 2.2656250000000002e-05,
3104
+ "loss": 0.1981,
3105
+ "step": 516
3106
+ },
3107
+ {
3108
+ "epoch": 4.68,
3109
+ "learning_rate": 2.25e-05,
3110
+ "loss": 0.1819,
3111
+ "step": 517
3112
+ },
3113
+ {
3114
+ "epoch": 4.69,
3115
+ "learning_rate": 2.234375e-05,
3116
+ "loss": 0.1989,
3117
+ "step": 518
3118
+ },
3119
+ {
3120
+ "epoch": 4.7,
3121
+ "learning_rate": 2.21875e-05,
3122
+ "loss": 0.1903,
3123
+ "step": 519
3124
+ },
3125
+ {
3126
+ "epoch": 4.71,
3127
+ "learning_rate": 2.203125e-05,
3128
+ "loss": 0.19,
3129
+ "step": 520
3130
+ },
3131
+ {
3132
+ "epoch": 4.72,
3133
+ "learning_rate": 2.1875e-05,
3134
+ "loss": 0.1921,
3135
+ "step": 521
3136
+ },
3137
+ {
3138
+ "epoch": 4.73,
3139
+ "learning_rate": 2.1718750000000003e-05,
3140
+ "loss": 0.2014,
3141
+ "step": 522
3142
+ },
3143
+ {
3144
+ "epoch": 4.74,
3145
+ "learning_rate": 2.1562500000000002e-05,
3146
+ "loss": 0.1928,
3147
+ "step": 523
3148
+ },
3149
+ {
3150
+ "epoch": 4.75,
3151
+ "learning_rate": 2.140625e-05,
3152
+ "loss": 0.1922,
3153
+ "step": 524
3154
+ },
3155
+ {
3156
+ "epoch": 4.76,
3157
+ "learning_rate": 2.125e-05,
3158
+ "loss": 0.1905,
3159
+ "step": 525
3160
+ },
3161
+ {
3162
+ "epoch": 4.77,
3163
+ "learning_rate": 2.109375e-05,
3164
+ "loss": 0.194,
3165
+ "step": 526
3166
+ },
3167
+ {
3168
+ "epoch": 4.77,
3169
+ "learning_rate": 2.09375e-05,
3170
+ "loss": 0.1898,
3171
+ "step": 527
3172
+ },
3173
+ {
3174
+ "epoch": 4.78,
3175
+ "learning_rate": 2.0781250000000004e-05,
3176
+ "loss": 0.197,
3177
+ "step": 528
3178
+ },
3179
+ {
3180
+ "epoch": 4.79,
3181
+ "learning_rate": 2.0625e-05,
3182
+ "loss": 0.2145,
3183
+ "step": 529
3184
+ },
3185
+ {
3186
+ "epoch": 4.8,
3187
+ "learning_rate": 2.046875e-05,
3188
+ "loss": 0.1894,
3189
+ "step": 530
3190
+ },
3191
+ {
3192
+ "epoch": 4.81,
3193
+ "learning_rate": 2.0312500000000002e-05,
3194
+ "loss": 0.1879,
3195
+ "step": 531
3196
+ },
3197
+ {
3198
+ "epoch": 4.82,
3199
+ "learning_rate": 2.0156250000000002e-05,
3200
+ "loss": 0.2029,
3201
+ "step": 532
3202
+ },
3203
+ {
3204
+ "epoch": 4.83,
3205
+ "learning_rate": 2e-05,
3206
+ "loss": 0.196,
3207
+ "step": 533
3208
+ },
3209
+ {
3210
+ "epoch": 4.84,
3211
+ "learning_rate": 1.984375e-05,
3212
+ "loss": 0.1875,
3213
+ "step": 534
3214
+ },
3215
+ {
3216
+ "epoch": 4.85,
3217
+ "learning_rate": 1.96875e-05,
3218
+ "loss": 0.192,
3219
+ "step": 535
3220
+ },
3221
+ {
3222
+ "epoch": 4.86,
3223
+ "learning_rate": 1.953125e-05,
3224
+ "loss": 0.2006,
3225
+ "step": 536
3226
+ },
3227
+ {
3228
+ "epoch": 4.87,
3229
+ "learning_rate": 1.9375e-05,
3230
+ "loss": 0.1805,
3231
+ "step": 537
3232
+ },
3233
+ {
3234
+ "epoch": 4.87,
3235
+ "learning_rate": 1.9218750000000003e-05,
3236
+ "loss": 0.1982,
3237
+ "step": 538
3238
+ },
3239
+ {
3240
+ "epoch": 4.88,
3241
+ "learning_rate": 1.90625e-05,
3242
+ "loss": 0.1869,
3243
+ "step": 539
3244
+ },
3245
+ {
3246
+ "epoch": 4.89,
3247
+ "learning_rate": 1.890625e-05,
3248
+ "loss": 0.1886,
3249
+ "step": 540
3250
+ },
3251
+ {
3252
+ "epoch": 4.9,
3253
+ "learning_rate": 1.8750000000000002e-05,
3254
+ "loss": 0.1803,
3255
+ "step": 541
3256
+ },
3257
+ {
3258
+ "epoch": 4.91,
3259
+ "learning_rate": 1.859375e-05,
3260
+ "loss": 0.1922,
3261
+ "step": 542
3262
+ },
3263
+ {
3264
+ "epoch": 4.92,
3265
+ "learning_rate": 1.84375e-05,
3266
+ "loss": 0.1903,
3267
+ "step": 543
3268
+ },
3269
+ {
3270
+ "epoch": 4.93,
3271
+ "learning_rate": 1.828125e-05,
3272
+ "loss": 0.1951,
3273
+ "step": 544
3274
+ },
3275
+ {
3276
+ "epoch": 4.94,
3277
+ "learning_rate": 1.8125e-05,
3278
+ "loss": 0.1912,
3279
+ "step": 545
3280
+ },
3281
+ {
3282
+ "epoch": 4.95,
3283
+ "learning_rate": 1.796875e-05,
3284
+ "loss": 0.196,
3285
+ "step": 546
3286
+ },
3287
+ {
3288
+ "epoch": 4.96,
3289
+ "learning_rate": 1.7812500000000003e-05,
3290
+ "loss": 0.1853,
3291
+ "step": 547
3292
+ },
3293
+ {
3294
+ "epoch": 4.96,
3295
+ "learning_rate": 1.7656250000000002e-05,
3296
+ "loss": 0.1935,
3297
+ "step": 548
3298
+ },
3299
+ {
3300
+ "epoch": 4.97,
3301
+ "learning_rate": 1.75e-05,
3302
+ "loss": 0.1955,
3303
+ "step": 549
3304
+ },
3305
+ {
3306
+ "epoch": 4.98,
3307
+ "learning_rate": 1.734375e-05,
3308
+ "loss": 0.19,
3309
+ "step": 550
3310
+ },
3311
+ {
3312
+ "epoch": 4.99,
3313
+ "learning_rate": 1.71875e-05,
3314
+ "loss": 0.1787,
3315
+ "step": 551
3316
+ },
3317
+ {
3318
+ "epoch": 5.0,
3319
+ "learning_rate": 1.703125e-05,
3320
+ "loss": 0.1967,
3321
+ "step": 552
3322
+ },
3323
+ {
3324
+ "epoch": 5.01,
3325
+ "learning_rate": 1.6875000000000004e-05,
3326
+ "loss": 0.1653,
3327
+ "step": 553
3328
+ },
3329
+ {
3330
+ "epoch": 5.02,
3331
+ "learning_rate": 1.671875e-05,
3332
+ "loss": 0.1715,
3333
+ "step": 554
3334
+ },
3335
+ {
3336
+ "epoch": 5.03,
3337
+ "learning_rate": 1.65625e-05,
3338
+ "loss": 0.1592,
3339
+ "step": 555
3340
+ },
3341
+ {
3342
+ "epoch": 5.04,
3343
+ "learning_rate": 1.6406250000000002e-05,
3344
+ "loss": 0.1616,
3345
+ "step": 556
3346
+ },
3347
+ {
3348
+ "epoch": 5.05,
3349
+ "learning_rate": 1.6250000000000002e-05,
3350
+ "loss": 0.1779,
3351
+ "step": 557
3352
+ },
3353
+ {
3354
+ "epoch": 5.06,
3355
+ "learning_rate": 1.609375e-05,
3356
+ "loss": 0.162,
3357
+ "step": 558
3358
+ },
3359
+ {
3360
+ "epoch": 5.06,
3361
+ "learning_rate": 1.59375e-05,
3362
+ "loss": 0.1779,
3363
+ "step": 559
3364
+ },
3365
+ {
3366
+ "epoch": 5.07,
3367
+ "learning_rate": 1.578125e-05,
3368
+ "loss": 0.1666,
3369
+ "step": 560
3370
+ },
3371
+ {
3372
+ "epoch": 5.08,
3373
+ "learning_rate": 1.5625e-05,
3374
+ "loss": 0.1804,
3375
+ "step": 561
3376
+ },
3377
+ {
3378
+ "epoch": 5.09,
3379
+ "learning_rate": 1.546875e-05,
3380
+ "loss": 0.1712,
3381
+ "step": 562
3382
+ },
3383
+ {
3384
+ "epoch": 5.1,
3385
+ "learning_rate": 1.5312500000000003e-05,
3386
+ "loss": 0.1692,
3387
+ "step": 563
3388
+ },
3389
+ {
3390
+ "epoch": 5.11,
3391
+ "learning_rate": 1.5156249999999999e-05,
3392
+ "loss": 0.167,
3393
+ "step": 564
3394
+ },
3395
+ {
3396
+ "epoch": 5.12,
3397
+ "learning_rate": 1.5e-05,
3398
+ "loss": 0.182,
3399
+ "step": 565
3400
+ },
3401
+ {
3402
+ "epoch": 5.13,
3403
+ "learning_rate": 1.484375e-05,
3404
+ "loss": 0.178,
3405
+ "step": 566
3406
+ },
3407
+ {
3408
+ "epoch": 5.14,
3409
+ "learning_rate": 1.4687500000000001e-05,
3410
+ "loss": 0.1805,
3411
+ "step": 567
3412
+ },
3413
+ {
3414
+ "epoch": 5.15,
3415
+ "learning_rate": 1.4531250000000003e-05,
3416
+ "loss": 0.179,
3417
+ "step": 568
3418
+ },
3419
+ {
3420
+ "epoch": 5.16,
3421
+ "learning_rate": 1.4374999999999999e-05,
3422
+ "loss": 0.1755,
3423
+ "step": 569
3424
+ },
3425
+ {
3426
+ "epoch": 5.16,
3427
+ "learning_rate": 1.421875e-05,
3428
+ "loss": 0.167,
3429
+ "step": 570
3430
+ },
3431
+ {
3432
+ "epoch": 5.17,
3433
+ "learning_rate": 1.4062500000000001e-05,
3434
+ "loss": 0.1668,
3435
+ "step": 571
3436
+ },
3437
+ {
3438
+ "epoch": 5.18,
3439
+ "learning_rate": 1.3906250000000001e-05,
3440
+ "loss": 0.1657,
3441
+ "step": 572
3442
+ },
3443
+ {
3444
+ "epoch": 5.19,
3445
+ "learning_rate": 1.3750000000000002e-05,
3446
+ "loss": 0.1718,
3447
+ "step": 573
3448
+ },
3449
+ {
3450
+ "epoch": 5.2,
3451
+ "learning_rate": 1.359375e-05,
3452
+ "loss": 0.1765,
3453
+ "step": 574
3454
+ },
3455
+ {
3456
+ "epoch": 5.21,
3457
+ "learning_rate": 1.34375e-05,
3458
+ "loss": 0.1786,
3459
+ "step": 575
3460
+ },
3461
+ {
3462
+ "epoch": 5.22,
3463
+ "learning_rate": 1.3281250000000001e-05,
3464
+ "loss": 0.1773,
3465
+ "step": 576
3466
+ },
3467
+ {
3468
+ "epoch": 5.23,
3469
+ "learning_rate": 1.3125e-05,
3470
+ "loss": 0.1708,
3471
+ "step": 577
3472
+ },
3473
+ {
3474
+ "epoch": 5.24,
3475
+ "learning_rate": 1.2968750000000002e-05,
3476
+ "loss": 0.1702,
3477
+ "step": 578
3478
+ },
3479
+ {
3480
+ "epoch": 5.25,
3481
+ "learning_rate": 1.28125e-05,
3482
+ "loss": 0.1764,
3483
+ "step": 579
3484
+ },
3485
+ {
3486
+ "epoch": 5.25,
3487
+ "learning_rate": 1.265625e-05,
3488
+ "loss": 0.1707,
3489
+ "step": 580
3490
+ },
3491
+ {
3492
+ "epoch": 5.26,
3493
+ "learning_rate": 1.25e-05,
3494
+ "loss": 0.1783,
3495
+ "step": 581
3496
+ },
3497
+ {
3498
+ "epoch": 5.27,
3499
+ "learning_rate": 1.2343750000000002e-05,
3500
+ "loss": 0.1663,
3501
+ "step": 582
3502
+ },
3503
+ {
3504
+ "epoch": 5.28,
3505
+ "learning_rate": 1.21875e-05,
3506
+ "loss": 0.1698,
3507
+ "step": 583
3508
+ },
3509
+ {
3510
+ "epoch": 5.29,
3511
+ "learning_rate": 1.2031250000000001e-05,
3512
+ "loss": 0.1694,
3513
+ "step": 584
3514
+ },
3515
+ {
3516
+ "epoch": 5.3,
3517
+ "learning_rate": 1.1875e-05,
3518
+ "loss": 0.1745,
3519
+ "step": 585
3520
+ },
3521
+ {
3522
+ "epoch": 5.31,
3523
+ "learning_rate": 1.171875e-05,
3524
+ "loss": 0.1646,
3525
+ "step": 586
3526
+ },
3527
+ {
3528
+ "epoch": 5.32,
3529
+ "learning_rate": 1.1562500000000002e-05,
3530
+ "loss": 0.1673,
3531
+ "step": 587
3532
+ },
3533
+ {
3534
+ "epoch": 5.33,
3535
+ "learning_rate": 1.140625e-05,
3536
+ "loss": 0.172,
3537
+ "step": 588
3538
+ },
3539
+ {
3540
+ "epoch": 5.34,
3541
+ "learning_rate": 1.125e-05,
3542
+ "loss": 0.1726,
3543
+ "step": 589
3544
+ },
3545
+ {
3546
+ "epoch": 5.35,
3547
+ "learning_rate": 1.109375e-05,
3548
+ "loss": 0.1788,
3549
+ "step": 590
3550
+ },
3551
+ {
3552
+ "epoch": 5.35,
3553
+ "learning_rate": 1.09375e-05,
3554
+ "loss": 0.1734,
3555
+ "step": 591
3556
+ },
3557
+ {
3558
+ "epoch": 5.36,
3559
+ "learning_rate": 1.0781250000000001e-05,
3560
+ "loss": 0.1757,
3561
+ "step": 592
3562
+ },
3563
+ {
3564
+ "epoch": 5.37,
3565
+ "learning_rate": 1.0625e-05,
3566
+ "loss": 0.1623,
3567
+ "step": 593
3568
+ },
3569
+ {
3570
+ "epoch": 5.38,
3571
+ "learning_rate": 1.046875e-05,
3572
+ "loss": 0.1666,
3573
+ "step": 594
3574
+ },
3575
+ {
3576
+ "epoch": 5.39,
3577
+ "learning_rate": 1.03125e-05,
3578
+ "loss": 0.1778,
3579
+ "step": 595
3580
+ },
3581
+ {
3582
+ "epoch": 5.4,
3583
+ "learning_rate": 1.0156250000000001e-05,
3584
+ "loss": 0.17,
3585
+ "step": 596
3586
+ },
3587
+ {
3588
+ "epoch": 5.41,
3589
+ "learning_rate": 1e-05,
3590
+ "loss": 0.1676,
3591
+ "step": 597
3592
+ },
3593
+ {
3594
+ "epoch": 5.42,
3595
+ "learning_rate": 9.84375e-06,
3596
+ "loss": 0.1761,
3597
+ "step": 598
3598
+ },
3599
+ {
3600
+ "epoch": 5.43,
3601
+ "learning_rate": 9.6875e-06,
3602
+ "loss": 0.1758,
3603
+ "step": 599
3604
+ },
3605
+ {
3606
+ "epoch": 5.44,
3607
+ "learning_rate": 9.53125e-06,
3608
+ "loss": 0.1772,
3609
+ "step": 600
3610
+ },
3611
+ {
3612
+ "epoch": 5.45,
3613
+ "learning_rate": 9.375000000000001e-06,
3614
+ "loss": 0.176,
3615
+ "step": 601
3616
+ },
3617
+ {
3618
+ "epoch": 5.45,
3619
+ "learning_rate": 9.21875e-06,
3620
+ "loss": 0.1604,
3621
+ "step": 602
3622
+ },
3623
+ {
3624
+ "epoch": 5.46,
3625
+ "learning_rate": 9.0625e-06,
3626
+ "loss": 0.1694,
3627
+ "step": 603
3628
+ },
3629
+ {
3630
+ "epoch": 5.47,
3631
+ "learning_rate": 8.906250000000001e-06,
3632
+ "loss": 0.1756,
3633
+ "step": 604
3634
+ },
3635
+ {
3636
+ "epoch": 5.48,
3637
+ "learning_rate": 8.75e-06,
3638
+ "loss": 0.1692,
3639
+ "step": 605
3640
+ },
3641
+ {
3642
+ "epoch": 5.49,
3643
+ "learning_rate": 8.59375e-06,
3644
+ "loss": 0.1808,
3645
+ "step": 606
3646
+ },
3647
+ {
3648
+ "epoch": 5.5,
3649
+ "learning_rate": 8.437500000000002e-06,
3650
+ "loss": 0.1706,
3651
+ "step": 607
3652
+ },
3653
+ {
3654
+ "epoch": 5.51,
3655
+ "learning_rate": 8.28125e-06,
3656
+ "loss": 0.168,
3657
+ "step": 608
3658
+ },
3659
+ {
3660
+ "epoch": 5.52,
3661
+ "learning_rate": 8.125000000000001e-06,
3662
+ "loss": 0.173,
3663
+ "step": 609
3664
+ },
3665
+ {
3666
+ "epoch": 5.53,
3667
+ "learning_rate": 7.96875e-06,
3668
+ "loss": 0.1612,
3669
+ "step": 610
3670
+ },
3671
+ {
3672
+ "epoch": 5.54,
3673
+ "learning_rate": 7.8125e-06,
3674
+ "loss": 0.1684,
3675
+ "step": 611
3676
+ },
3677
+ {
3678
+ "epoch": 5.54,
3679
+ "learning_rate": 7.656250000000001e-06,
3680
+ "loss": 0.1775,
3681
+ "step": 612
3682
+ },
3683
+ {
3684
+ "epoch": 5.55,
3685
+ "learning_rate": 7.5e-06,
3686
+ "loss": 0.1638,
3687
+ "step": 613
3688
+ },
3689
+ {
3690
+ "epoch": 5.56,
3691
+ "learning_rate": 7.343750000000001e-06,
3692
+ "loss": 0.1701,
3693
+ "step": 614
3694
+ },
3695
+ {
3696
+ "epoch": 5.57,
3697
+ "learning_rate": 7.187499999999999e-06,
3698
+ "loss": 0.1641,
3699
+ "step": 615
3700
+ },
3701
+ {
3702
+ "epoch": 5.58,
3703
+ "learning_rate": 7.031250000000001e-06,
3704
+ "loss": 0.165,
3705
+ "step": 616
3706
+ },
3707
+ {
3708
+ "epoch": 5.59,
3709
+ "learning_rate": 6.875000000000001e-06,
3710
+ "loss": 0.1806,
3711
+ "step": 617
3712
+ },
3713
+ {
3714
+ "epoch": 5.6,
3715
+ "learning_rate": 6.71875e-06,
3716
+ "loss": 0.1793,
3717
+ "step": 618
3718
+ },
3719
+ {
3720
+ "epoch": 5.61,
3721
+ "learning_rate": 6.5625e-06,
3722
+ "loss": 0.1625,
3723
+ "step": 619
3724
+ },
3725
+ {
3726
+ "epoch": 5.62,
3727
+ "learning_rate": 6.40625e-06,
3728
+ "loss": 0.1632,
3729
+ "step": 620
3730
+ },
3731
+ {
3732
+ "epoch": 5.63,
3733
+ "learning_rate": 6.25e-06,
3734
+ "loss": 0.1629,
3735
+ "step": 621
3736
+ },
3737
+ {
3738
+ "epoch": 5.64,
3739
+ "learning_rate": 6.09375e-06,
3740
+ "loss": 0.183,
3741
+ "step": 622
3742
+ },
3743
+ {
3744
+ "epoch": 5.64,
3745
+ "learning_rate": 5.9375e-06,
3746
+ "loss": 0.1768,
3747
+ "step": 623
3748
+ },
3749
+ {
3750
+ "epoch": 5.65,
3751
+ "learning_rate": 5.781250000000001e-06,
3752
+ "loss": 0.1605,
3753
+ "step": 624
3754
+ },
3755
+ {
3756
+ "epoch": 5.66,
3757
+ "learning_rate": 5.625e-06,
3758
+ "loss": 0.1749,
3759
+ "step": 625
3760
+ },
3761
+ {
3762
+ "epoch": 5.67,
3763
+ "learning_rate": 5.46875e-06,
3764
+ "loss": 0.1625,
3765
+ "step": 626
3766
+ },
3767
+ {
3768
+ "epoch": 5.68,
3769
+ "learning_rate": 5.3125e-06,
3770
+ "loss": 0.1538,
3771
+ "step": 627
3772
+ },
3773
+ {
3774
+ "epoch": 5.69,
3775
+ "learning_rate": 5.15625e-06,
3776
+ "loss": 0.174,
3777
+ "step": 628
3778
+ },
3779
+ {
3780
+ "epoch": 5.7,
3781
+ "learning_rate": 5e-06,
3782
+ "loss": 0.1686,
3783
+ "step": 629
3784
+ },
3785
+ {
3786
+ "epoch": 5.71,
3787
+ "learning_rate": 4.84375e-06,
3788
+ "loss": 0.1706,
3789
+ "step": 630
3790
+ },
3791
+ {
3792
+ "epoch": 5.72,
3793
+ "learning_rate": 4.6875000000000004e-06,
3794
+ "loss": 0.164,
3795
+ "step": 631
3796
+ },
3797
+ {
3798
+ "epoch": 5.73,
3799
+ "learning_rate": 4.53125e-06,
3800
+ "loss": 0.1794,
3801
+ "step": 632
3802
+ },
3803
+ {
3804
+ "epoch": 5.73,
3805
+ "learning_rate": 4.375e-06,
3806
+ "loss": 0.1601,
3807
+ "step": 633
3808
+ },
3809
+ {
3810
+ "epoch": 5.74,
3811
+ "learning_rate": 4.218750000000001e-06,
3812
+ "loss": 0.1754,
3813
+ "step": 634
3814
+ },
3815
+ {
3816
+ "epoch": 5.75,
3817
+ "learning_rate": 4.0625000000000005e-06,
3818
+ "loss": 0.1711,
3819
+ "step": 635
3820
+ },
3821
+ {
3822
+ "epoch": 5.76,
3823
+ "learning_rate": 3.90625e-06,
3824
+ "loss": 0.1777,
3825
+ "step": 636
3826
+ },
3827
+ {
3828
+ "epoch": 5.77,
3829
+ "learning_rate": 3.75e-06,
3830
+ "loss": 0.1695,
3831
+ "step": 637
3832
+ },
3833
+ {
3834
+ "epoch": 5.78,
3835
+ "learning_rate": 3.5937499999999997e-06,
3836
+ "loss": 0.1651,
3837
+ "step": 638
3838
+ },
3839
+ {
3840
+ "epoch": 5.79,
3841
+ "learning_rate": 3.4375000000000005e-06,
3842
+ "loss": 0.1683,
3843
+ "step": 639
3844
+ },
3845
+ {
3846
+ "epoch": 5.8,
3847
+ "learning_rate": 3.28125e-06,
3848
+ "loss": 0.1625,
3849
+ "step": 640
3850
+ },
3851
+ {
3852
+ "epoch": 5.81,
3853
+ "learning_rate": 3.125e-06,
3854
+ "loss": 0.1713,
3855
+ "step": 641
3856
+ },
3857
+ {
3858
+ "epoch": 5.82,
3859
+ "learning_rate": 2.96875e-06,
3860
+ "loss": 0.1692,
3861
+ "step": 642
3862
+ },
3863
+ {
3864
+ "epoch": 5.83,
3865
+ "learning_rate": 2.8125e-06,
3866
+ "loss": 0.1628,
3867
+ "step": 643
3868
+ },
3869
+ {
3870
+ "epoch": 5.83,
3871
+ "learning_rate": 2.65625e-06,
3872
+ "loss": 0.1639,
3873
+ "step": 644
3874
+ },
3875
+ {
3876
+ "epoch": 5.84,
3877
+ "learning_rate": 2.5e-06,
3878
+ "loss": 0.1804,
3879
+ "step": 645
3880
+ },
3881
+ {
3882
+ "epoch": 5.85,
3883
+ "learning_rate": 2.3437500000000002e-06,
3884
+ "loss": 0.1732,
3885
+ "step": 646
3886
+ },
3887
+ {
3888
+ "epoch": 5.86,
3889
+ "learning_rate": 2.1875e-06,
3890
+ "loss": 0.1632,
3891
+ "step": 647
3892
+ },
3893
+ {
3894
+ "epoch": 5.87,
3895
+ "learning_rate": 2.0312500000000002e-06,
3896
+ "loss": 0.1735,
3897
+ "step": 648
3898
+ },
3899
+ {
3900
+ "epoch": 5.88,
3901
+ "learning_rate": 1.875e-06,
3902
+ "loss": 0.169,
3903
+ "step": 649
3904
+ },
3905
+ {
3906
+ "epoch": 5.89,
3907
+ "learning_rate": 1.7187500000000003e-06,
3908
+ "loss": 0.1671,
3909
+ "step": 650
3910
+ },
3911
+ {
3912
+ "epoch": 5.9,
3913
+ "learning_rate": 1.5625e-06,
3914
+ "loss": 0.1578,
3915
+ "step": 651
3916
+ },
3917
+ {
3918
+ "epoch": 5.91,
3919
+ "learning_rate": 1.40625e-06,
3920
+ "loss": 0.1715,
3921
+ "step": 652
3922
+ },
3923
+ {
3924
+ "epoch": 5.92,
3925
+ "learning_rate": 1.25e-06,
3926
+ "loss": 0.1694,
3927
+ "step": 653
3928
+ },
3929
+ {
3930
+ "epoch": 5.93,
3931
+ "learning_rate": 1.09375e-06,
3932
+ "loss": 0.1688,
3933
+ "step": 654
3934
+ },
3935
+ {
3936
+ "epoch": 5.93,
3937
+ "learning_rate": 9.375e-07,
3938
+ "loss": 0.1688,
3939
+ "step": 655
3940
+ },
3941
+ {
3942
+ "epoch": 5.94,
3943
+ "learning_rate": 7.8125e-07,
3944
+ "loss": 0.1766,
3945
+ "step": 656
3946
+ },
3947
+ {
3948
+ "epoch": 5.95,
3949
+ "learning_rate": 6.25e-07,
3950
+ "loss": 0.1725,
3951
+ "step": 657
3952
+ },
3953
+ {
3954
+ "epoch": 5.96,
3955
+ "learning_rate": 4.6875e-07,
3956
+ "loss": 0.163,
3957
+ "step": 658
3958
+ },
3959
+ {
3960
+ "epoch": 5.97,
3961
+ "learning_rate": 3.125e-07,
3962
+ "loss": 0.184,
3963
+ "step": 659
3964
+ },
3965
+ {
3966
+ "epoch": 5.98,
3967
+ "learning_rate": 1.5625e-07,
3968
+ "loss": 0.1672,
3969
+ "step": 660
3970
+ },
3971
+ {
3972
+ "epoch": 5.98,
3973
+ "step": 660,
3974
+ "total_flos": 0.0,
3975
+ "train_loss": 0.27463054072224735,
3976
+ "train_runtime": 10229.9582,
3977
+ "train_samples_per_second": 8.277,
3978
+ "train_steps_per_second": 0.065
3979
+ }
3980
+ ],
3981
+ "logging_steps": 1.0,
3982
+ "max_steps": 660,
3983
+ "num_input_tokens_seen": 0,
3984
+ "num_train_epochs": 6,
3985
+ "save_steps": -660,
3986
+ "total_flos": 0.0,
3987
+ "train_batch_size": 4,
3988
+ "trial_name": null,
3989
+ "trial_params": null
3990
+ }