ironrock commited on
Commit
f26d673
1 Parent(s): e453d7d

Training in progress, step 180, checkpoint

Browse files
checkpoint-180/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: Weni/WeniGPT-Agents-Mistral-1.0.0-SFT-merged
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 0.10.0
checkpoint-180/adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Weni/WeniGPT-Agents-Mistral-1.0.0-SFT-merged",
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": 32,
14
+ "lora_dropout": 0.1,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 16,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "o_proj",
24
+ "k_proj",
25
+ "v_proj",
26
+ "q_proj",
27
+ "down_proj",
28
+ "gate_proj",
29
+ "up_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
checkpoint-180/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c935526dd85271c2590178d6089dcfa08846fcfb31e4b0e6f8f15281df6785d
3
+ size 167832240
checkpoint-180/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42ca4fe2abcb485a8e726d28d38bb5b7a4472a364a1a880aad4bdfaa8429c60c
3
+ size 335922386
checkpoint-180/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae78313eb528c8d3695eebaf4de3539bd0a0bc6ee18c66af1ee183442f1758a0
3
+ size 15024
checkpoint-180/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b38031f60d9e88601d369ef46bcdcf2b5b03f2cb4ba93853bcb2328df7ebb7c
3
+ size 15024
checkpoint-180/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f58092375c93d237cd0e3149aecfbf83e2acdae46279e07a32920d01cb507e64
3
+ size 15024
checkpoint-180/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83cd4bbff9962da7ec6787fcea8d65df7096917f9a5902e249ba7aee8887fe5f
3
+ size 15024
checkpoint-180/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b47d7566b2e3ec8e96007d3ecf18c3cfe003d3d58d6c01ae710dc49dfbfb649
3
+ size 1064
checkpoint-180/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-180/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-180/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
checkpoint-180/tokenizer_config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'system' %}{{ '<<SYS>>\\n' + message['content'] + '\\n<</SYS>>\\n\\n' }}{% elif message['role'] == 'assistant' %}{{ ' ' + message['content'] + ' ' + eos_token }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "max_lenght": 8192,
37
+ "max_length": 8192,
38
+ "model_max_length": 1000000000000000019884624838656,
39
+ "pad_token": "<unk>",
40
+ "padding": true,
41
+ "sp_model_kwargs": {},
42
+ "spaces_between_special_tokens": false,
43
+ "stride": 0,
44
+ "tokenizer_class": "LlamaTokenizer",
45
+ "truncation_side": "right",
46
+ "truncation_strategy": "longest_first",
47
+ "unk_token": "<unk>",
48
+ "use_default_system_prompt": false
49
+ }
checkpoint-180/trainer_state.json ADDED
@@ -0,0 +1,387 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.040502406656742096,
3
+ "best_model_checkpoint": "./mistral/21-04-24-Weni-WeniGPT-Agents-Mistral-1.0.0-SFT-1.0.16-DPO_Experiment on DPO with other hyperparameters and best SFT model of WeniGPT-2_max_steps-180_batch_8_2024-04-21_ppid_9/checkpoint-180",
4
+ "epoch": 5.806451612903226,
5
+ "eval_steps": 30,
6
+ "global_step": 180,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.3225806451612903,
13
+ "grad_norm": 23.757038116455078,
14
+ "learning_rate": 4.971264367816092e-06,
15
+ "logits/chosen": -1.8281418085098267,
16
+ "logits/rejected": -1.8474775552749634,
17
+ "logps/chosen": -128.32241821289062,
18
+ "logps/rejected": -168.0380401611328,
19
+ "loss": 0.6742,
20
+ "rewards/accuracies": 0.5,
21
+ "rewards/chosen": 0.024183962494134903,
22
+ "rewards/margins": 0.03792798891663551,
23
+ "rewards/rejected": -0.013744029216468334,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.6451612903225806,
28
+ "grad_norm": 14.753284454345703,
29
+ "learning_rate": 4.683908045977012e-06,
30
+ "logits/chosen": -1.7773618698120117,
31
+ "logits/rejected": -1.780083417892456,
32
+ "logps/chosen": -144.4004364013672,
33
+ "logps/rejected": -171.8882598876953,
34
+ "loss": 0.5367,
35
+ "rewards/accuracies": 0.8500000238418579,
36
+ "rewards/chosen": 0.282940536737442,
37
+ "rewards/margins": 0.31420594453811646,
38
+ "rewards/rejected": -0.03126540407538414,
39
+ "step": 20
40
+ },
41
+ {
42
+ "epoch": 0.967741935483871,
43
+ "grad_norm": 13.17676067352295,
44
+ "learning_rate": 4.396551724137931e-06,
45
+ "logits/chosen": -1.8475195169448853,
46
+ "logits/rejected": -1.8598411083221436,
47
+ "logps/chosen": -166.5201416015625,
48
+ "logps/rejected": -198.87669372558594,
49
+ "loss": 0.3994,
50
+ "rewards/accuracies": 0.949999988079071,
51
+ "rewards/chosen": 0.7265206575393677,
52
+ "rewards/margins": 1.12095308303833,
53
+ "rewards/rejected": -0.3944324254989624,
54
+ "step": 30
55
+ },
56
+ {
57
+ "epoch": 0.967741935483871,
58
+ "eval_logits/chosen": -1.7065293788909912,
59
+ "eval_logits/rejected": -1.7336915731430054,
60
+ "eval_logps/chosen": -112.41703033447266,
61
+ "eval_logps/rejected": -210.2833709716797,
62
+ "eval_loss": 0.339741975069046,
63
+ "eval_rewards/accuracies": 0.7142857313156128,
64
+ "eval_rewards/chosen": 0.7506826519966125,
65
+ "eval_rewards/margins": 0.770815372467041,
66
+ "eval_rewards/rejected": -0.020132770761847496,
67
+ "eval_runtime": 10.1682,
68
+ "eval_samples_per_second": 2.754,
69
+ "eval_steps_per_second": 0.688,
70
+ "step": 30
71
+ },
72
+ {
73
+ "epoch": 1.2903225806451613,
74
+ "grad_norm": 9.547523498535156,
75
+ "learning_rate": 4.1091954022988515e-06,
76
+ "logits/chosen": -1.7585010528564453,
77
+ "logits/rejected": -1.7726383209228516,
78
+ "logps/chosen": -123.3655014038086,
79
+ "logps/rejected": -181.35037231445312,
80
+ "loss": 0.2655,
81
+ "rewards/accuracies": 0.949999988079071,
82
+ "rewards/chosen": 0.981993556022644,
83
+ "rewards/margins": 1.3875294923782349,
84
+ "rewards/rejected": -0.4055357575416565,
85
+ "step": 40
86
+ },
87
+ {
88
+ "epoch": 1.6129032258064515,
89
+ "grad_norm": 6.552857398986816,
90
+ "learning_rate": 3.82183908045977e-06,
91
+ "logits/chosen": -1.7593097686767578,
92
+ "logits/rejected": -1.7726491689682007,
93
+ "logps/chosen": -79.5435562133789,
94
+ "logps/rejected": -169.55801391601562,
95
+ "loss": 0.2323,
96
+ "rewards/accuracies": 0.949999988079071,
97
+ "rewards/chosen": 1.1555945873260498,
98
+ "rewards/margins": 1.6152770519256592,
99
+ "rewards/rejected": -0.45968255400657654,
100
+ "step": 50
101
+ },
102
+ {
103
+ "epoch": 1.935483870967742,
104
+ "grad_norm": 5.082254409790039,
105
+ "learning_rate": 3.5344827586206898e-06,
106
+ "logits/chosen": -1.7814455032348633,
107
+ "logits/rejected": -1.8014984130859375,
108
+ "logps/chosen": -135.05734252929688,
109
+ "logps/rejected": -220.71621704101562,
110
+ "loss": 0.1771,
111
+ "rewards/accuracies": 1.0,
112
+ "rewards/chosen": 1.483880877494812,
113
+ "rewards/margins": 2.265058994293213,
114
+ "rewards/rejected": -0.7811781167984009,
115
+ "step": 60
116
+ },
117
+ {
118
+ "epoch": 1.935483870967742,
119
+ "eval_logits/chosen": -1.7119187116622925,
120
+ "eval_logits/rejected": -1.7429500818252563,
121
+ "eval_logps/chosen": -106.78651428222656,
122
+ "eval_logps/rejected": -215.82162475585938,
123
+ "eval_loss": 0.17449094355106354,
124
+ "eval_rewards/accuracies": 0.8571428656578064,
125
+ "eval_rewards/chosen": 1.3137341737747192,
126
+ "eval_rewards/margins": 1.8876904249191284,
127
+ "eval_rewards/rejected": -0.5739563703536987,
128
+ "eval_runtime": 10.1675,
129
+ "eval_samples_per_second": 2.754,
130
+ "eval_steps_per_second": 0.688,
131
+ "step": 60
132
+ },
133
+ {
134
+ "epoch": 2.258064516129032,
135
+ "grad_norm": 3.0963690280914307,
136
+ "learning_rate": 3.2758620689655175e-06,
137
+ "logits/chosen": -1.8137136697769165,
138
+ "logits/rejected": -1.831018090248108,
139
+ "logps/chosen": -127.5184555053711,
140
+ "logps/rejected": -193.3052215576172,
141
+ "loss": 0.144,
142
+ "rewards/accuracies": 0.949999988079071,
143
+ "rewards/chosen": 1.2200127840042114,
144
+ "rewards/margins": 2.3993964195251465,
145
+ "rewards/rejected": -1.179383635520935,
146
+ "step": 70
147
+ },
148
+ {
149
+ "epoch": 2.5806451612903225,
150
+ "grad_norm": 4.23032808303833,
151
+ "learning_rate": 2.988505747126437e-06,
152
+ "logits/chosen": -1.8039875030517578,
153
+ "logits/rejected": -1.8293946981430054,
154
+ "logps/chosen": -109.42149353027344,
155
+ "logps/rejected": -206.9613494873047,
156
+ "loss": 0.1129,
157
+ "rewards/accuracies": 1.0,
158
+ "rewards/chosen": 1.4697037935256958,
159
+ "rewards/margins": 2.851051092147827,
160
+ "rewards/rejected": -1.3813471794128418,
161
+ "step": 80
162
+ },
163
+ {
164
+ "epoch": 2.903225806451613,
165
+ "grad_norm": 6.72817850112915,
166
+ "learning_rate": 2.7011494252873567e-06,
167
+ "logits/chosen": -1.814522385597229,
168
+ "logits/rejected": -1.8382046222686768,
169
+ "logps/chosen": -186.84329223632812,
170
+ "logps/rejected": -260.76788330078125,
171
+ "loss": 0.1314,
172
+ "rewards/accuracies": 1.0,
173
+ "rewards/chosen": 1.1758986711502075,
174
+ "rewards/margins": 2.6401941776275635,
175
+ "rewards/rejected": -1.4642951488494873,
176
+ "step": 90
177
+ },
178
+ {
179
+ "epoch": 2.903225806451613,
180
+ "eval_logits/chosen": -1.7031118869781494,
181
+ "eval_logits/rejected": -1.7357984781265259,
182
+ "eval_logps/chosen": -105.27674102783203,
183
+ "eval_logps/rejected": -222.73390197753906,
184
+ "eval_loss": 0.11280059069395065,
185
+ "eval_rewards/accuracies": 0.8571428656578064,
186
+ "eval_rewards/chosen": 1.4647114276885986,
187
+ "eval_rewards/margins": 2.7298948764801025,
188
+ "eval_rewards/rejected": -1.2651835680007935,
189
+ "eval_runtime": 10.1661,
190
+ "eval_samples_per_second": 2.754,
191
+ "eval_steps_per_second": 0.689,
192
+ "step": 90
193
+ },
194
+ {
195
+ "epoch": 3.225806451612903,
196
+ "grad_norm": 3.741824150085449,
197
+ "learning_rate": 2.4137931034482762e-06,
198
+ "logits/chosen": -1.9151725769042969,
199
+ "logits/rejected": -1.929137945175171,
200
+ "logps/chosen": -195.36695861816406,
201
+ "logps/rejected": -271.27081298828125,
202
+ "loss": 0.1269,
203
+ "rewards/accuracies": 1.0,
204
+ "rewards/chosen": 1.3293390274047852,
205
+ "rewards/margins": 3.3184096813201904,
206
+ "rewards/rejected": -1.9890711307525635,
207
+ "step": 100
208
+ },
209
+ {
210
+ "epoch": 3.5483870967741935,
211
+ "grad_norm": 3.6807944774627686,
212
+ "learning_rate": 2.1264367816091954e-06,
213
+ "logits/chosen": -1.7386070489883423,
214
+ "logits/rejected": -1.766031265258789,
215
+ "logps/chosen": -120.89439392089844,
216
+ "logps/rejected": -234.75289916992188,
217
+ "loss": 0.08,
218
+ "rewards/accuracies": 1.0,
219
+ "rewards/chosen": 1.6623188257217407,
220
+ "rewards/margins": 4.070118427276611,
221
+ "rewards/rejected": -2.4078001976013184,
222
+ "step": 110
223
+ },
224
+ {
225
+ "epoch": 3.870967741935484,
226
+ "grad_norm": 4.077445030212402,
227
+ "learning_rate": 1.839080459770115e-06,
228
+ "logits/chosen": -1.8381446599960327,
229
+ "logits/rejected": -1.848647117614746,
230
+ "logps/chosen": -162.987548828125,
231
+ "logps/rejected": -245.94497680664062,
232
+ "loss": 0.1016,
233
+ "rewards/accuracies": 1.0,
234
+ "rewards/chosen": 1.698729157447815,
235
+ "rewards/margins": 3.5379676818847656,
236
+ "rewards/rejected": -1.8392385244369507,
237
+ "step": 120
238
+ },
239
+ {
240
+ "epoch": 3.870967741935484,
241
+ "eval_logits/chosen": -1.6965323686599731,
242
+ "eval_logits/rejected": -1.729313850402832,
243
+ "eval_logps/chosen": -104.65281677246094,
244
+ "eval_logps/rejected": -229.02908325195312,
245
+ "eval_loss": 0.06442112475633621,
246
+ "eval_rewards/accuracies": 1.0,
247
+ "eval_rewards/chosen": 1.527103066444397,
248
+ "eval_rewards/margins": 3.4218084812164307,
249
+ "eval_rewards/rejected": -1.8947057723999023,
250
+ "eval_runtime": 10.1682,
251
+ "eval_samples_per_second": 2.754,
252
+ "eval_steps_per_second": 0.688,
253
+ "step": 120
254
+ },
255
+ {
256
+ "epoch": 4.193548387096774,
257
+ "grad_norm": 2.3105103969573975,
258
+ "learning_rate": 1.5517241379310346e-06,
259
+ "logits/chosen": -1.8013200759887695,
260
+ "logits/rejected": -1.817307472229004,
261
+ "logps/chosen": -132.51260375976562,
262
+ "logps/rejected": -271.40521240234375,
263
+ "loss": 0.0683,
264
+ "rewards/accuracies": 1.0,
265
+ "rewards/chosen": 1.6829872131347656,
266
+ "rewards/margins": 4.633041858673096,
267
+ "rewards/rejected": -2.950054407119751,
268
+ "step": 130
269
+ },
270
+ {
271
+ "epoch": 4.516129032258064,
272
+ "grad_norm": 2.9849331378936768,
273
+ "learning_rate": 1.2643678160919542e-06,
274
+ "logits/chosen": -1.8443206548690796,
275
+ "logits/rejected": -1.8669426441192627,
276
+ "logps/chosen": -173.345947265625,
277
+ "logps/rejected": -228.7808380126953,
278
+ "loss": 0.0546,
279
+ "rewards/accuracies": 1.0,
280
+ "rewards/chosen": 1.7071704864501953,
281
+ "rewards/margins": 3.6590206623077393,
282
+ "rewards/rejected": -1.9518499374389648,
283
+ "step": 140
284
+ },
285
+ {
286
+ "epoch": 4.838709677419355,
287
+ "grad_norm": 1.6962618827819824,
288
+ "learning_rate": 9.770114942528738e-07,
289
+ "logits/chosen": -1.8463163375854492,
290
+ "logits/rejected": -1.8704414367675781,
291
+ "logps/chosen": -127.43244934082031,
292
+ "logps/rejected": -244.4454803466797,
293
+ "loss": 0.0633,
294
+ "rewards/accuracies": 1.0,
295
+ "rewards/chosen": 1.2895654439926147,
296
+ "rewards/margins": 4.082577705383301,
297
+ "rewards/rejected": -2.7930119037628174,
298
+ "step": 150
299
+ },
300
+ {
301
+ "epoch": 4.838709677419355,
302
+ "eval_logits/chosen": -1.692850947380066,
303
+ "eval_logits/rejected": -1.7251230478286743,
304
+ "eval_logps/chosen": -105.38704681396484,
305
+ "eval_logps/rejected": -233.88169860839844,
306
+ "eval_loss": 0.04773273691534996,
307
+ "eval_rewards/accuracies": 1.0,
308
+ "eval_rewards/chosen": 1.4536798000335693,
309
+ "eval_rewards/margins": 3.8336431980133057,
310
+ "eval_rewards/rejected": -2.3799636363983154,
311
+ "eval_runtime": 10.1649,
312
+ "eval_samples_per_second": 2.755,
313
+ "eval_steps_per_second": 0.689,
314
+ "step": 150
315
+ },
316
+ {
317
+ "epoch": 5.161290322580645,
318
+ "grad_norm": 1.0977122783660889,
319
+ "learning_rate": 6.896551724137931e-07,
320
+ "logits/chosen": -1.854543924331665,
321
+ "logits/rejected": -1.8709430694580078,
322
+ "logps/chosen": -124.8465347290039,
323
+ "logps/rejected": -214.95779418945312,
324
+ "loss": 0.0374,
325
+ "rewards/accuracies": 1.0,
326
+ "rewards/chosen": 1.6399686336517334,
327
+ "rewards/margins": 4.3664116859436035,
328
+ "rewards/rejected": -2.72644305229187,
329
+ "step": 160
330
+ },
331
+ {
332
+ "epoch": 5.483870967741936,
333
+ "grad_norm": 0.5631425976753235,
334
+ "learning_rate": 4.0229885057471266e-07,
335
+ "logits/chosen": -1.882641077041626,
336
+ "logits/rejected": -1.8940929174423218,
337
+ "logps/chosen": -163.1198272705078,
338
+ "logps/rejected": -238.10739135742188,
339
+ "loss": 0.0451,
340
+ "rewards/accuracies": 1.0,
341
+ "rewards/chosen": 1.1276298761367798,
342
+ "rewards/margins": 3.9089503288269043,
343
+ "rewards/rejected": -2.781320333480835,
344
+ "step": 170
345
+ },
346
+ {
347
+ "epoch": 5.806451612903226,
348
+ "grad_norm": 1.9172598123550415,
349
+ "learning_rate": 1.1494252873563219e-07,
350
+ "logits/chosen": -1.7833986282348633,
351
+ "logits/rejected": -1.8089481592178345,
352
+ "logps/chosen": -177.43270874023438,
353
+ "logps/rejected": -286.78851318359375,
354
+ "loss": 0.0497,
355
+ "rewards/accuracies": 1.0,
356
+ "rewards/chosen": 1.6389026641845703,
357
+ "rewards/margins": 4.935039520263672,
358
+ "rewards/rejected": -3.2961373329162598,
359
+ "step": 180
360
+ },
361
+ {
362
+ "epoch": 5.806451612903226,
363
+ "eval_logits/chosen": -1.6861690282821655,
364
+ "eval_logits/rejected": -1.7177609205245972,
365
+ "eval_logps/chosen": -105.73729705810547,
366
+ "eval_logps/rejected": -235.87796020507812,
367
+ "eval_loss": 0.040502406656742096,
368
+ "eval_rewards/accuracies": 1.0,
369
+ "eval_rewards/chosen": 1.4186561107635498,
370
+ "eval_rewards/margins": 3.9982476234436035,
371
+ "eval_rewards/rejected": -2.579591751098633,
372
+ "eval_runtime": 10.1675,
373
+ "eval_samples_per_second": 2.754,
374
+ "eval_steps_per_second": 0.688,
375
+ "step": 180
376
+ }
377
+ ],
378
+ "logging_steps": 10,
379
+ "max_steps": 180,
380
+ "num_input_tokens_seen": 0,
381
+ "num_train_epochs": 6,
382
+ "save_steps": 90,
383
+ "total_flos": 0.0,
384
+ "train_batch_size": 1,
385
+ "trial_name": null,
386
+ "trial_params": null
387
+ }
checkpoint-180/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:254388b8f07517916d64a90fed86a2932ed3a112a76ae3c0534722034f9a6905
3
+ size 5432