luzimu commited on
Commit
a1a8d9b
1 Parent(s): 5e1c8ef

first commit

Browse files
README.md CHANGED
@@ -1,3 +1,81 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: /mnt/cache/luzimu/rlhf_math/alignment-handbook/outs/Mistral-7B-v0.1-lce
3
+ tags:
4
+ - alignment-handbook
5
+ - generated_from_trainer
6
+ datasets:
7
+ - /mnt/cache/luzimu/rlhf_math/data/controled_steps_math_gsm8k_lce_dpo_ascend_lim2_lim3_add_dpo1x1
8
+ model-index:
9
+ - name: Mistral-7B-v0.1-lce_controled_steps_dpo_ascend_lim2_lim3_add_dpo1x1
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # Mistral-7B-v0.1-lce_controled_steps_dpo_ascend_lim2_lim3_add_dpo1x1
17
+
18
+ This model is a fine-tuned version of [/mnt/cache/luzimu/rlhf_math/alignment-handbook/outs/Mistral-7B-v0.1-lce](https://huggingface.co//mnt/cache/luzimu/rlhf_math/alignment-handbook/outs/Mistral-7B-v0.1-lce) on the /mnt/cache/luzimu/rlhf_math/data/controled_steps_math_gsm8k_lce_dpo_ascend_lim2_lim3_add_dpo1x1 dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.1793
21
+ - Rewards/chosen: 0.2587
22
+ - Rewards/rejected: -7.0301
23
+ - Rewards/accuracies: 0.8947
24
+ - Rewards/margins: 7.2889
25
+ - Logps/rejected: -253.7773
26
+ - Logps/chosen: -80.3105
27
+ - Logits/rejected: -2.3417
28
+ - Logits/chosen: -2.3846
29
+
30
+ ## Model description
31
+
32
+ More information needed
33
+
34
+ ## Intended uses & limitations
35
+
36
+ More information needed
37
+
38
+ ## Training and evaluation data
39
+
40
+ More information needed
41
+
42
+ ## Training procedure
43
+
44
+ ### Training hyperparameters
45
+
46
+ The following hyperparameters were used during training:
47
+ - learning_rate: 1e-07
48
+ - train_batch_size: 2
49
+ - eval_batch_size: 4
50
+ - seed: 42
51
+ - distributed_type: multi-GPU
52
+ - num_devices: 4
53
+ - gradient_accumulation_steps: 8
54
+ - total_train_batch_size: 64
55
+ - total_eval_batch_size: 16
56
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
57
+ - lr_scheduler_type: cosine
58
+ - lr_scheduler_warmup_ratio: 0.1
59
+ - num_epochs: 2
60
+
61
+ ### Training results
62
+
63
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
64
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
65
+ | 0.3963 | 0.21 | 100 | 0.3636 | 1.8634 | -0.1518 | 0.8816 | 2.0152 | -184.9944 | -64.2644 | -2.7112 | -2.7505 |
66
+ | 0.2849 | 0.43 | 200 | 0.2598 | 0.7706 | -3.7221 | 0.8816 | 4.4927 | -220.6974 | -75.1921 | -2.5067 | -2.5475 |
67
+ | 0.2496 | 0.64 | 300 | 0.2295 | 0.9323 | -4.2717 | 0.8684 | 5.2040 | -226.1934 | -73.5753 | -2.5080 | -2.5494 |
68
+ | 0.2331 | 0.86 | 400 | 0.2089 | 0.7871 | -4.8912 | 0.8684 | 5.6783 | -232.3884 | -75.0269 | -2.4967 | -2.5382 |
69
+ | 0.0874 | 1.07 | 500 | 0.1872 | 0.6345 | -5.7444 | 0.8816 | 6.3789 | -240.9202 | -76.5527 | -2.4323 | -2.4761 |
70
+ | 0.1217 | 1.28 | 600 | 0.1832 | 0.2282 | -6.6907 | 0.8684 | 6.9188 | -250.3827 | -80.6161 | -2.3741 | -2.4172 |
71
+ | 0.0966 | 1.5 | 700 | 0.1807 | 0.1849 | -7.0125 | 0.8816 | 7.1975 | -253.6012 | -81.0485 | -2.3503 | -2.3940 |
72
+ | 0.0755 | 1.71 | 800 | 0.1802 | 0.3224 | -6.9539 | 0.8947 | 7.2763 | -253.0150 | -79.6739 | -2.3437 | -2.3867 |
73
+ | 0.1177 | 1.93 | 900 | 0.1793 | 0.2587 | -7.0301 | 0.8947 | 7.2889 | -253.7773 | -80.3105 | -2.3417 | -2.3846 |
74
+
75
+
76
+ ### Framework versions
77
+
78
+ - Transformers 4.38.2
79
+ - Pytorch 2.1.2
80
+ - Datasets 2.14.6
81
+ - Tokenizers 0.15.2
added_tokens.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<|assistant|>": 32002,
3
+ "<|code|>": 32004,
4
+ "<|endofblock|>": 32006,
5
+ "<|endofmessage|>": 32007,
6
+ "<|execution|>": 32005,
7
+ "<|system|>": 32001,
8
+ "<|text|>": 32003,
9
+ "<|user|>": 32000
10
+ }
all_results.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "eval_logits/chosen": -2.3846213817596436,
4
+ "eval_logits/rejected": -2.3414511680603027,
5
+ "eval_logps/chosen": -80.37240600585938,
6
+ "eval_logps/rejected": -253.9702911376953,
7
+ "eval_loss": 0.17943651974201202,
8
+ "eval_rewards/accuracies": 0.8947368264198303,
9
+ "eval_rewards/chosen": 0.25255322456359863,
10
+ "eval_rewards/margins": 7.301982879638672,
11
+ "eval_rewards/rejected": -7.049429416656494,
12
+ "eval_runtime": 27.9087,
13
+ "eval_samples": 301,
14
+ "eval_samples_per_second": 10.785,
15
+ "eval_steps_per_second": 0.681,
16
+ "train_loss": 0.21607095603003493,
17
+ "train_runtime": 14362.1771,
18
+ "train_samples": 29893,
19
+ "train_samples_per_second": 4.163,
20
+ "train_steps_per_second": 0.065
21
+ }
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/mnt/cache/luzimu/rlhf_math/alignment-handbook/outs/Mistral-7B-v0.1-lce",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 4096,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "mistral",
15
+ "num_attention_heads": 32,
16
+ "num_hidden_layers": 32,
17
+ "num_key_value_heads": 8,
18
+ "rms_norm_eps": 1e-05,
19
+ "rope_theta": 10000.0,
20
+ "sliding_window": 4096,
21
+ "tie_word_embeddings": false,
22
+ "torch_dtype": "bfloat16",
23
+ "transformers_version": "4.38.2",
24
+ "use_cache": true,
25
+ "vocab_size": 32128
26
+ }
eval_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "eval_logits/chosen": -2.3846213817596436,
4
+ "eval_logits/rejected": -2.3414511680603027,
5
+ "eval_logps/chosen": -80.37240600585938,
6
+ "eval_logps/rejected": -253.9702911376953,
7
+ "eval_loss": 0.17943651974201202,
8
+ "eval_rewards/accuracies": 0.8947368264198303,
9
+ "eval_rewards/chosen": 0.25255322456359863,
10
+ "eval_rewards/margins": 7.301982879638672,
11
+ "eval_rewards/rejected": -7.049429416656494,
12
+ "eval_runtime": 27.9087,
13
+ "eval_samples": 301,
14
+ "eval_samples_per_second": 10.785,
15
+ "eval_steps_per_second": 0.681
16
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.38.2"
6
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:069598f45d8fd3590f09b131ccc53f5e465fcad0e0fa0311e573059bc11981dc
3
+ size 4944210912
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41642eb1352d6e8c6223d417778bbe985f677e0528884a6847c79e32de850e97
3
+ size 4999819336
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a65eeb14cb3c2bd4bd2460f5f56989d6d02df8b51446409f430afcfb4b042bda
3
+ size 4541564920
model.safetensors.index.json ADDED
@@ -0,0 +1,298 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 14485561344
4
+ },
5
+ "weight_map": {
6
+ "lm_head.weight": "model-00003-of-00003.safetensors",
7
+ "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
13
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
14
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
15
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
16
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
17
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
18
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
19
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
20
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
21
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
22
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
23
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
24
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
25
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
26
+ "model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
27
+ "model.layers.10.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
28
+ "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
29
+ "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
30
+ "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
31
+ "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
32
+ "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
33
+ "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
34
+ "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
35
+ "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
36
+ "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
37
+ "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
38
+ "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
39
+ "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
40
+ "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
41
+ "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
42
+ "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
43
+ "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
44
+ "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
45
+ "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
46
+ "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
47
+ "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
48
+ "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
49
+ "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
50
+ "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
51
+ "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
52
+ "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
53
+ "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
54
+ "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
55
+ "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
56
+ "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
57
+ "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
58
+ "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
59
+ "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
60
+ "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
61
+ "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
62
+ "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
63
+ "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
64
+ "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
65
+ "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
66
+ "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
67
+ "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
68
+ "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
69
+ "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
70
+ "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
71
+ "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
72
+ "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
73
+ "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
74
+ "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
75
+ "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
76
+ "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
77
+ "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
78
+ "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
79
+ "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
80
+ "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
81
+ "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
82
+ "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
83
+ "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
84
+ "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
85
+ "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
86
+ "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
87
+ "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
88
+ "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
89
+ "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
90
+ "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
91
+ "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
92
+ "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
93
+ "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
94
+ "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
95
+ "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
96
+ "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
97
+ "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
98
+ "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
99
+ "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
100
+ "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
101
+ "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
102
+ "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
103
+ "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
104
+ "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
105
+ "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
106
+ "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
107
+ "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
108
+ "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
109
+ "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
110
+ "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
111
+ "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
112
+ "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
113
+ "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
114
+ "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
115
+ "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
116
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
117
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
118
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
119
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
120
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
121
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
122
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
123
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
124
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
125
+ "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
126
+ "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
127
+ "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
128
+ "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
129
+ "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
130
+ "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
131
+ "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
132
+ "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
133
+ "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
134
+ "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
135
+ "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
136
+ "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
137
+ "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
138
+ "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
139
+ "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
140
+ "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
141
+ "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
142
+ "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
143
+ "model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
144
+ "model.layers.22.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
145
+ "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
146
+ "model.layers.22.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
147
+ "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
148
+ "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
149
+ "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
150
+ "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
151
+ "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
152
+ "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
153
+ "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
154
+ "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
155
+ "model.layers.23.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
156
+ "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
157
+ "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
158
+ "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
159
+ "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
160
+ "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
161
+ "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
162
+ "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
163
+ "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
164
+ "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
165
+ "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
166
+ "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
167
+ "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
168
+ "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
169
+ "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
170
+ "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
171
+ "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
172
+ "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
173
+ "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
174
+ "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
175
+ "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
176
+ "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
177
+ "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
178
+ "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
179
+ "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
180
+ "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
181
+ "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
182
+ "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
183
+ "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
184
+ "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
185
+ "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
186
+ "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
187
+ "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
188
+ "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
189
+ "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
190
+ "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
191
+ "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
192
+ "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
193
+ "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
194
+ "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
195
+ "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
196
+ "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
197
+ "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
198
+ "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
199
+ "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
200
+ "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
201
+ "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
202
+ "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
203
+ "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
204
+ "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
205
+ "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
206
+ "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
207
+ "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
208
+ "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
209
+ "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
210
+ "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
211
+ "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
212
+ "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
213
+ "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
214
+ "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
215
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
216
+ "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
217
+ "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
218
+ "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
219
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
220
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
221
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
222
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
223
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
224
+ "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
225
+ "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
226
+ "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
227
+ "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
228
+ "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
229
+ "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
230
+ "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
231
+ "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
232
+ "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
233
+ "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
234
+ "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
235
+ "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
236
+ "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
237
+ "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
238
+ "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
239
+ "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
240
+ "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
241
+ "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
242
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
243
+ "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
244
+ "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
245
+ "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
246
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
247
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
248
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
249
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
250
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
251
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
252
+ "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
253
+ "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
254
+ "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
255
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
256
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
257
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
258
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
259
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
260
+ "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
261
+ "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
262
+ "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
263
+ "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
264
+ "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
265
+ "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
266
+ "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
267
+ "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
268
+ "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
269
+ "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
270
+ "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
271
+ "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
272
+ "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
273
+ "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
274
+ "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
275
+ "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
276
+ "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
277
+ "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
278
+ "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
279
+ "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
280
+ "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
281
+ "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
282
+ "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
283
+ "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
284
+ "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
285
+ "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
286
+ "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
287
+ "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
288
+ "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
289
+ "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
290
+ "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
291
+ "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
292
+ "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
293
+ "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
294
+ "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
295
+ "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
296
+ "model.norm.weight": "model-00003-of-00003.safetensors"
297
+ }
298
+ }
runs/Apr13_21-08-49_pt-4rcsf1vl-worker-0/events.out.tfevents.1713013930.pt-4rcsf1vl-worker-0.38.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3889c86e1e64be477579961cb4a6af53bfa372d24eec5019ffff3f30a33a9f0
3
+ size 76575
runs/Apr13_21-08-49_pt-4rcsf1vl-worker-0/events.out.tfevents.1713028337.pt-4rcsf1vl-worker-0.38.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c1c008b168faa818b02fe2fed64b827c767cf28e925a7a453187472ff8bb45d
3
+ size 828
special_tokens_map.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|user|>",
4
+ "<|system|>",
5
+ "<|assistant|>",
6
+ "<|text|>",
7
+ "<|code|>",
8
+ "<|execution|>",
9
+ "<|endofblock|>",
10
+ "<|endofmessage|>"
11
+ ],
12
+ "bos_token": {
13
+ "content": "<s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "eos_token": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "pad_token": {
27
+ "content": "</s>",
28
+ "lstrip": false,
29
+ "normalized": true,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ },
33
+ "unk_token": {
34
+ "content": "<unk>",
35
+ "lstrip": false,
36
+ "normalized": true,
37
+ "rstrip": false,
38
+ "single_word": false
39
+ }
40
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "32000": {
30
+ "content": "<|user|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "32001": {
38
+ "content": "<|system|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "32002": {
46
+ "content": "<|assistant|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "32003": {
54
+ "content": "<|text|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "32004": {
62
+ "content": "<|code|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "32005": {
70
+ "content": "<|execution|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "32006": {
78
+ "content": "<|endofblock|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "32007": {
86
+ "content": "<|endofmessage|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ }
93
+ },
94
+ "additional_special_tokens": [
95
+ "<|user|>",
96
+ "<|system|>",
97
+ "<|assistant|>",
98
+ "<|text|>",
99
+ "<|code|>",
100
+ "<|execution|>",
101
+ "<|endofblock|>",
102
+ "<|endofmessage|>"
103
+ ],
104
+ "bos_token": "<s>",
105
+ "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>' }}{% elif message['role'] == 'system' %}\n{{ '<|system|>' }}{% elif message['role'] == 'assistant_correct' %}\n{{ '<|assistant|>' }}{% endif %}\n{% for block in message['content'] %}\n{% if block['type'] == 'text' %}\n{{ '<|text|>' }}{% elif block['type'] == 'code' %}\n{{ '<|code|>' }}{% elif block['type'] == 'execution' %}\n{{ '<|execution|>' }}{% endif %}\n{{ block['content'] + '<|endofblock|>' }}{% endfor %}\n{% if message['role'] != 'assistant_correct' %}\n{{ '<|endofmessage|>' }}{% endif %}{% endfor %}",
106
+ "clean_up_tokenization_spaces": false,
107
+ "eos_token": "</s>",
108
+ "legacy": true,
109
+ "model_max_length": 2048,
110
+ "pad_token": "</s>",
111
+ "sp_model_kwargs": {},
112
+ "spaces_between_special_tokens": false,
113
+ "tokenizer_class": "LlamaTokenizer",
114
+ "unk_token": "<unk>",
115
+ "use_default_system_prompt": true
116
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "train_loss": 0.21607095603003493,
4
+ "train_runtime": 14362.1771,
5
+ "train_samples": 29893,
6
+ "train_samples_per_second": 4.163,
7
+ "train_steps_per_second": 0.065
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1584 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9994648113459994,
5
+ "eval_steps": 100,
6
+ "global_step": 934,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "grad_norm": 233.47004915999355,
14
+ "learning_rate": 1.0638297872340424e-09,
15
+ "logits/chosen": -2.9313061237335205,
16
+ "logits/rejected": -2.9551453590393066,
17
+ "logps/chosen": -84.83019256591797,
18
+ "logps/rejected": -122.72698974609375,
19
+ "loss": 0.6931,
20
+ "rewards/accuracies": 0.0,
21
+ "rewards/chosen": 0.0,
22
+ "rewards/margins": 0.0,
23
+ "rewards/rejected": 0.0,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.02,
28
+ "grad_norm": 260.5751075825045,
29
+ "learning_rate": 1.0638297872340425e-08,
30
+ "logits/chosen": -2.9027864933013916,
31
+ "logits/rejected": -2.9143683910369873,
32
+ "logps/chosen": -89.16300964355469,
33
+ "logps/rejected": -185.06263732910156,
34
+ "loss": 0.6955,
35
+ "rewards/accuracies": 0.4236111044883728,
36
+ "rewards/chosen": 0.003953192848712206,
37
+ "rewards/margins": -6.558249151566997e-05,
38
+ "rewards/rejected": 0.004018775187432766,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "grad_norm": 208.3552637601932,
44
+ "learning_rate": 2.127659574468085e-08,
45
+ "logits/chosen": -2.913933038711548,
46
+ "logits/rejected": -2.928837537765503,
47
+ "logps/chosen": -80.42648315429688,
48
+ "logps/rejected": -182.06158447265625,
49
+ "loss": 0.6916,
50
+ "rewards/accuracies": 0.5625,
51
+ "rewards/chosen": 0.014185728505253792,
52
+ "rewards/margins": 0.009901548735797405,
53
+ "rewards/rejected": 0.004284179303795099,
54
+ "step": 20
55
+ },
56
+ {
57
+ "epoch": 0.06,
58
+ "grad_norm": 230.6593263920013,
59
+ "learning_rate": 3.191489361702128e-08,
60
+ "logits/chosen": -2.9638733863830566,
61
+ "logits/rejected": -2.947378396987915,
62
+ "logps/chosen": -98.21806335449219,
63
+ "logps/rejected": -209.5995635986328,
64
+ "loss": 0.6746,
65
+ "rewards/accuracies": 0.6187499761581421,
66
+ "rewards/chosen": 0.06607931107282639,
67
+ "rewards/margins": 0.030630504712462425,
68
+ "rewards/rejected": 0.03544881194829941,
69
+ "step": 30
70
+ },
71
+ {
72
+ "epoch": 0.09,
73
+ "grad_norm": 229.5001403960732,
74
+ "learning_rate": 4.25531914893617e-08,
75
+ "logits/chosen": -2.8966455459594727,
76
+ "logits/rejected": -2.928231716156006,
77
+ "logps/chosen": -89.00160217285156,
78
+ "logps/rejected": -181.76254272460938,
79
+ "loss": 0.6351,
80
+ "rewards/accuracies": 0.737500011920929,
81
+ "rewards/chosen": 0.2816316783428192,
82
+ "rewards/margins": 0.13491405546665192,
83
+ "rewards/rejected": 0.1467176377773285,
84
+ "step": 40
85
+ },
86
+ {
87
+ "epoch": 0.11,
88
+ "grad_norm": 156.2561754732514,
89
+ "learning_rate": 5.3191489361702123e-08,
90
+ "logits/chosen": -2.900092601776123,
91
+ "logits/rejected": -2.8932037353515625,
92
+ "logps/chosen": -83.15284729003906,
93
+ "logps/rejected": -171.3411102294922,
94
+ "loss": 0.5936,
95
+ "rewards/accuracies": 0.7437499761581421,
96
+ "rewards/chosen": 0.5719220638275146,
97
+ "rewards/margins": 0.29314228892326355,
98
+ "rewards/rejected": 0.2787797451019287,
99
+ "step": 50
100
+ },
101
+ {
102
+ "epoch": 0.13,
103
+ "grad_norm": 132.07680968174566,
104
+ "learning_rate": 6.382978723404255e-08,
105
+ "logits/chosen": -2.891676425933838,
106
+ "logits/rejected": -2.9195587635040283,
107
+ "logps/chosen": -74.60004425048828,
108
+ "logps/rejected": -182.98202514648438,
109
+ "loss": 0.5168,
110
+ "rewards/accuracies": 0.824999988079071,
111
+ "rewards/chosen": 1.0686428546905518,
112
+ "rewards/margins": 0.6195236444473267,
113
+ "rewards/rejected": 0.4491192400455475,
114
+ "step": 60
115
+ },
116
+ {
117
+ "epoch": 0.15,
118
+ "grad_norm": 151.5773318247949,
119
+ "learning_rate": 7.446808510638298e-08,
120
+ "logits/chosen": -2.9197444915771484,
121
+ "logits/rejected": -2.9284892082214355,
122
+ "logps/chosen": -74.910888671875,
123
+ "logps/rejected": -172.590087890625,
124
+ "loss": 0.5056,
125
+ "rewards/accuracies": 0.824999988079071,
126
+ "rewards/chosen": 1.4046766757965088,
127
+ "rewards/margins": 0.7043676376342773,
128
+ "rewards/rejected": 0.7003089189529419,
129
+ "step": 70
130
+ },
131
+ {
132
+ "epoch": 0.17,
133
+ "grad_norm": 130.65989464011713,
134
+ "learning_rate": 8.51063829787234e-08,
135
+ "logits/chosen": -2.9318323135375977,
136
+ "logits/rejected": -2.897862195968628,
137
+ "logps/chosen": -67.76197052001953,
138
+ "logps/rejected": -189.8668670654297,
139
+ "loss": 0.4526,
140
+ "rewards/accuracies": 0.84375,
141
+ "rewards/chosen": 1.649476408958435,
142
+ "rewards/margins": 1.2739518880844116,
143
+ "rewards/rejected": 0.3755245506763458,
144
+ "step": 80
145
+ },
146
+ {
147
+ "epoch": 0.19,
148
+ "grad_norm": 135.89857787297694,
149
+ "learning_rate": 9.574468085106384e-08,
150
+ "logits/chosen": -2.870614528656006,
151
+ "logits/rejected": -2.8664841651916504,
152
+ "logps/chosen": -72.87008666992188,
153
+ "logps/rejected": -173.2789764404297,
154
+ "loss": 0.4347,
155
+ "rewards/accuracies": 0.8125,
156
+ "rewards/chosen": 1.8745962381362915,
157
+ "rewards/margins": 1.2003813982009888,
158
+ "rewards/rejected": 0.674214780330658,
159
+ "step": 90
160
+ },
161
+ {
162
+ "epoch": 0.21,
163
+ "grad_norm": 147.9607240979239,
164
+ "learning_rate": 9.998741174712532e-08,
165
+ "logits/chosen": -2.8332231044769287,
166
+ "logits/rejected": -2.846297264099121,
167
+ "logps/chosen": -74.4647445678711,
168
+ "logps/rejected": -185.64999389648438,
169
+ "loss": 0.3963,
170
+ "rewards/accuracies": 0.856249988079071,
171
+ "rewards/chosen": 1.7484337091445923,
172
+ "rewards/margins": 1.455592393875122,
173
+ "rewards/rejected": 0.29284122586250305,
174
+ "step": 100
175
+ },
176
+ {
177
+ "epoch": 0.21,
178
+ "eval_logits/chosen": -2.750455856323242,
179
+ "eval_logits/rejected": -2.711240530014038,
180
+ "eval_logps/chosen": -64.26441192626953,
181
+ "eval_logps/rejected": -184.994384765625,
182
+ "eval_loss": 0.36364680528640747,
183
+ "eval_rewards/accuracies": 0.8815789222717285,
184
+ "eval_rewards/chosen": 1.8633519411087036,
185
+ "eval_rewards/margins": 2.015190839767456,
186
+ "eval_rewards/rejected": -0.15183834731578827,
187
+ "eval_runtime": 27.7804,
188
+ "eval_samples_per_second": 10.835,
189
+ "eval_steps_per_second": 0.684,
190
+ "step": 100
191
+ },
192
+ {
193
+ "epoch": 0.24,
194
+ "grad_norm": 144.1908317487022,
195
+ "learning_rate": 9.991050648838675e-08,
196
+ "logits/chosen": -2.8495163917541504,
197
+ "logits/rejected": -2.819291353225708,
198
+ "logps/chosen": -62.78534698486328,
199
+ "logps/rejected": -203.07852172851562,
200
+ "loss": 0.3763,
201
+ "rewards/accuracies": 0.918749988079071,
202
+ "rewards/chosen": 1.7807117700576782,
203
+ "rewards/margins": 2.012101888656616,
204
+ "rewards/rejected": -0.231390118598938,
205
+ "step": 110
206
+ },
207
+ {
208
+ "epoch": 0.26,
209
+ "grad_norm": 126.7835721201166,
210
+ "learning_rate": 9.976379687325631e-08,
211
+ "logits/chosen": -2.7876718044281006,
212
+ "logits/rejected": -2.7896816730499268,
213
+ "logps/chosen": -63.9537353515625,
214
+ "logps/rejected": -183.58328247070312,
215
+ "loss": 0.3381,
216
+ "rewards/accuracies": 0.893750011920929,
217
+ "rewards/chosen": 1.7092759609222412,
218
+ "rewards/margins": 2.222316265106201,
219
+ "rewards/rejected": -0.5130404233932495,
220
+ "step": 120
221
+ },
222
+ {
223
+ "epoch": 0.28,
224
+ "grad_norm": 224.70875384146225,
225
+ "learning_rate": 9.954748808839674e-08,
226
+ "logits/chosen": -2.799767017364502,
227
+ "logits/rejected": -2.79805064201355,
228
+ "logps/chosen": -81.58202362060547,
229
+ "logps/rejected": -194.84872436523438,
230
+ "loss": 0.3302,
231
+ "rewards/accuracies": 0.84375,
232
+ "rewards/chosen": 1.278132438659668,
233
+ "rewards/margins": 2.2866179943084717,
234
+ "rewards/rejected": -1.0084853172302246,
235
+ "step": 130
236
+ },
237
+ {
238
+ "epoch": 0.3,
239
+ "grad_norm": 152.12671314001338,
240
+ "learning_rate": 9.926188266120296e-08,
241
+ "logits/chosen": -2.763793706893921,
242
+ "logits/rejected": -2.727785348892212,
243
+ "logps/chosen": -75.18443298339844,
244
+ "logps/rejected": -176.11270141601562,
245
+ "loss": 0.3357,
246
+ "rewards/accuracies": 0.8500000238418579,
247
+ "rewards/chosen": 1.239394187927246,
248
+ "rewards/margins": 2.4168992042541504,
249
+ "rewards/rejected": -1.1775051355361938,
250
+ "step": 140
251
+ },
252
+ {
253
+ "epoch": 0.32,
254
+ "grad_norm": 175.5036541136684,
255
+ "learning_rate": 9.890738003669028e-08,
256
+ "logits/chosen": -2.7513179779052734,
257
+ "logits/rejected": -2.7361416816711426,
258
+ "logps/chosen": -81.78472900390625,
259
+ "logps/rejected": -219.29556274414062,
260
+ "loss": 0.2906,
261
+ "rewards/accuracies": 0.8812500238418579,
262
+ "rewards/chosen": 0.717334508895874,
263
+ "rewards/margins": 3.1458559036254883,
264
+ "rewards/rejected": -2.4285213947296143,
265
+ "step": 150
266
+ },
267
+ {
268
+ "epoch": 0.34,
269
+ "grad_norm": 170.33413139097263,
270
+ "learning_rate": 9.848447601883434e-08,
271
+ "logits/chosen": -2.6835436820983887,
272
+ "logits/rejected": -2.68257737159729,
273
+ "logps/chosen": -84.51789855957031,
274
+ "logps/rejected": -208.9514923095703,
275
+ "loss": 0.3133,
276
+ "rewards/accuracies": 0.893750011920929,
277
+ "rewards/chosen": 0.4816192090511322,
278
+ "rewards/margins": 2.897939443588257,
279
+ "rewards/rejected": -2.416320323944092,
280
+ "step": 160
281
+ },
282
+ {
283
+ "epoch": 0.36,
284
+ "grad_norm": 162.15836168812638,
285
+ "learning_rate": 9.799376207714445e-08,
286
+ "logits/chosen": -2.645550489425659,
287
+ "logits/rejected": -2.6841909885406494,
288
+ "logps/chosen": -82.82842254638672,
289
+ "logps/rejected": -195.99717712402344,
290
+ "loss": 0.3403,
291
+ "rewards/accuracies": 0.893750011920929,
292
+ "rewards/chosen": 0.4788424074649811,
293
+ "rewards/margins": 2.944272518157959,
294
+ "rewards/rejected": -2.46543025970459,
295
+ "step": 170
296
+ },
297
+ {
298
+ "epoch": 0.39,
299
+ "grad_norm": 119.29724112251517,
300
+ "learning_rate": 9.743592451943998e-08,
301
+ "logits/chosen": -2.687215805053711,
302
+ "logits/rejected": -2.6954283714294434,
303
+ "logps/chosen": -82.87139129638672,
304
+ "logps/rejected": -229.29202270507812,
305
+ "loss": 0.2932,
306
+ "rewards/accuracies": 0.8812500238418579,
307
+ "rewards/chosen": 0.5700880289077759,
308
+ "rewards/margins": 3.488856792449951,
309
+ "rewards/rejected": -2.9187686443328857,
310
+ "step": 180
311
+ },
312
+ {
313
+ "epoch": 0.41,
314
+ "grad_norm": 141.87377453577582,
315
+ "learning_rate": 9.681174353198685e-08,
316
+ "logits/chosen": -2.6703782081604004,
317
+ "logits/rejected": -2.6757917404174805,
318
+ "logps/chosen": -82.6243667602539,
319
+ "logps/rejected": -188.33477783203125,
320
+ "loss": 0.2942,
321
+ "rewards/accuracies": 0.8374999761581421,
322
+ "rewards/chosen": 0.5440769791603088,
323
+ "rewards/margins": 2.693603515625,
324
+ "rewards/rejected": -2.149526834487915,
325
+ "step": 190
326
+ },
327
+ {
328
+ "epoch": 0.43,
329
+ "grad_norm": 222.8132498459904,
330
+ "learning_rate": 9.612209208833647e-08,
331
+ "logits/chosen": -2.649033784866333,
332
+ "logits/rejected": -2.6577847003936768,
333
+ "logps/chosen": -78.76862335205078,
334
+ "logps/rejected": -221.69357299804688,
335
+ "loss": 0.2849,
336
+ "rewards/accuracies": 0.925000011920929,
337
+ "rewards/chosen": 0.6831364631652832,
338
+ "rewards/margins": 3.6835713386535645,
339
+ "rewards/rejected": -3.000434398651123,
340
+ "step": 200
341
+ },
342
+ {
343
+ "epoch": 0.43,
344
+ "eval_logits/chosen": -2.5474884510040283,
345
+ "eval_logits/rejected": -2.5067243576049805,
346
+ "eval_logps/chosen": -75.19207763671875,
347
+ "eval_logps/rejected": -220.6973876953125,
348
+ "eval_loss": 0.2597656548023224,
349
+ "eval_rewards/accuracies": 0.8815789222717285,
350
+ "eval_rewards/chosen": 0.7705855369567871,
351
+ "eval_rewards/margins": 4.492726802825928,
352
+ "eval_rewards/rejected": -3.7221410274505615,
353
+ "eval_runtime": 27.7534,
354
+ "eval_samples_per_second": 10.846,
355
+ "eval_steps_per_second": 0.685,
356
+ "step": 200
357
+ },
358
+ {
359
+ "epoch": 0.45,
360
+ "grad_norm": 159.31516163900298,
361
+ "learning_rate": 9.536793472839323e-08,
362
+ "logits/chosen": -2.6742613315582275,
363
+ "logits/rejected": -2.6248650550842285,
364
+ "logps/chosen": -83.2725830078125,
365
+ "logps/rejected": -208.2786865234375,
366
+ "loss": 0.2658,
367
+ "rewards/accuracies": 0.8999999761581421,
368
+ "rewards/chosen": 0.7041572332382202,
369
+ "rewards/margins": 3.5826714038848877,
370
+ "rewards/rejected": -2.878513813018799,
371
+ "step": 210
372
+ },
373
+ {
374
+ "epoch": 0.47,
375
+ "grad_norm": 179.27000301013558,
376
+ "learning_rate": 9.455032620941839e-08,
377
+ "logits/chosen": -2.6498050689697266,
378
+ "logits/rejected": -2.676229953765869,
379
+ "logps/chosen": -89.39515686035156,
380
+ "logps/rejected": -219.5927276611328,
381
+ "loss": 0.3068,
382
+ "rewards/accuracies": 0.8687499761581421,
383
+ "rewards/chosen": 0.4539889395236969,
384
+ "rewards/margins": 3.6276097297668457,
385
+ "rewards/rejected": -3.1736209392547607,
386
+ "step": 220
387
+ },
388
+ {
389
+ "epoch": 0.49,
390
+ "grad_norm": 128.89336085629637,
391
+ "learning_rate": 9.367041003085648e-08,
392
+ "logits/chosen": -2.6379408836364746,
393
+ "logits/rejected": -2.672508716583252,
394
+ "logps/chosen": -81.60795593261719,
395
+ "logps/rejected": -219.58493041992188,
396
+ "loss": 0.247,
397
+ "rewards/accuracies": 0.8999999761581421,
398
+ "rewards/chosen": 0.5703738927841187,
399
+ "rewards/margins": 4.038177013397217,
400
+ "rewards/rejected": -3.4678032398223877,
401
+ "step": 230
402
+ },
403
+ {
404
+ "epoch": 0.51,
405
+ "grad_norm": 131.59397339189383,
406
+ "learning_rate": 9.272941683504808e-08,
407
+ "logits/chosen": -2.6790764331817627,
408
+ "logits/rejected": -2.6560444831848145,
409
+ "logps/chosen": -80.53921508789062,
410
+ "logps/rejected": -245.55868530273438,
411
+ "loss": 0.2733,
412
+ "rewards/accuracies": 0.8999999761581421,
413
+ "rewards/chosen": 0.6836718320846558,
414
+ "rewards/margins": 5.092911720275879,
415
+ "rewards/rejected": -4.409239768981934,
416
+ "step": 240
417
+ },
418
+ {
419
+ "epoch": 0.54,
420
+ "grad_norm": 172.8756884288556,
421
+ "learning_rate": 9.172866268606513e-08,
422
+ "logits/chosen": -2.6560401916503906,
423
+ "logits/rejected": -2.6558754444122314,
424
+ "logps/chosen": -83.98664855957031,
425
+ "logps/rejected": -211.24819946289062,
426
+ "loss": 0.2742,
427
+ "rewards/accuracies": 0.862500011920929,
428
+ "rewards/chosen": 0.5306154489517212,
429
+ "rewards/margins": 3.6272902488708496,
430
+ "rewards/rejected": -3.096675157546997,
431
+ "step": 250
432
+ },
433
+ {
434
+ "epoch": 0.56,
435
+ "grad_norm": 133.84123014944902,
436
+ "learning_rate": 9.066954722907638e-08,
437
+ "logits/chosen": -2.6467976570129395,
438
+ "logits/rejected": -2.6493191719055176,
439
+ "logps/chosen": -81.35044860839844,
440
+ "logps/rejected": -215.6734161376953,
441
+ "loss": 0.2596,
442
+ "rewards/accuracies": 0.862500011920929,
443
+ "rewards/chosen": 0.13885298371315002,
444
+ "rewards/margins": 3.929053783416748,
445
+ "rewards/rejected": -3.790201187133789,
446
+ "step": 260
447
+ },
448
+ {
449
+ "epoch": 0.58,
450
+ "grad_norm": 170.45984672890785,
451
+ "learning_rate": 8.955355173281707e-08,
452
+ "logits/chosen": -2.6520111560821533,
453
+ "logits/rejected": -2.662503719329834,
454
+ "logps/chosen": -92.13319396972656,
455
+ "logps/rejected": -233.948486328125,
456
+ "loss": 0.2782,
457
+ "rewards/accuracies": 0.8999999761581421,
458
+ "rewards/chosen": 0.021317768841981888,
459
+ "rewards/margins": 4.14536190032959,
460
+ "rewards/rejected": -4.124044418334961,
461
+ "step": 270
462
+ },
463
+ {
464
+ "epoch": 0.6,
465
+ "grad_norm": 229.80467359056394,
466
+ "learning_rate": 8.838223701790055e-08,
467
+ "logits/chosen": -2.6390814781188965,
468
+ "logits/rejected": -2.622087240219116,
469
+ "logps/chosen": -81.07563781738281,
470
+ "logps/rejected": -205.11767578125,
471
+ "loss": 0.2882,
472
+ "rewards/accuracies": 0.887499988079071,
473
+ "rewards/chosen": 0.6619526147842407,
474
+ "rewards/margins": 3.7605528831481934,
475
+ "rewards/rejected": -3.098600387573242,
476
+ "step": 280
477
+ },
478
+ {
479
+ "epoch": 0.62,
480
+ "grad_norm": 141.1175297189335,
481
+ "learning_rate": 8.715724127386971e-08,
482
+ "logits/chosen": -2.631877899169922,
483
+ "logits/rejected": -2.6188888549804688,
484
+ "logps/chosen": -77.27508544921875,
485
+ "logps/rejected": -202.60848999023438,
486
+ "loss": 0.2766,
487
+ "rewards/accuracies": 0.887499988079071,
488
+ "rewards/chosen": 0.7706214189529419,
489
+ "rewards/margins": 3.9430625438690186,
490
+ "rewards/rejected": -3.1724414825439453,
491
+ "step": 290
492
+ },
493
+ {
494
+ "epoch": 0.64,
495
+ "grad_norm": 128.57223622967007,
496
+ "learning_rate": 8.588027776804058e-08,
497
+ "logits/chosen": -2.659540891647339,
498
+ "logits/rejected": -2.6711039543151855,
499
+ "logps/chosen": -79.66828918457031,
500
+ "logps/rejected": -224.89675903320312,
501
+ "loss": 0.2496,
502
+ "rewards/accuracies": 0.9312499761581421,
503
+ "rewards/chosen": 0.8538252711296082,
504
+ "rewards/margins": 4.59223747253418,
505
+ "rewards/rejected": -3.7384121417999268,
506
+ "step": 300
507
+ },
508
+ {
509
+ "epoch": 0.64,
510
+ "eval_logits/chosen": -2.54941725730896,
511
+ "eval_logits/rejected": -2.507993698120117,
512
+ "eval_logps/chosen": -73.57530975341797,
513
+ "eval_logps/rejected": -226.1934356689453,
514
+ "eval_loss": 0.22951094806194305,
515
+ "eval_rewards/accuracies": 0.8684210777282715,
516
+ "eval_rewards/chosen": 0.9322623014450073,
517
+ "eval_rewards/margins": 5.204004287719727,
518
+ "eval_rewards/rejected": -4.271742343902588,
519
+ "eval_runtime": 27.7467,
520
+ "eval_samples_per_second": 10.848,
521
+ "eval_steps_per_second": 0.685,
522
+ "step": 300
523
+ },
524
+ {
525
+ "epoch": 0.66,
526
+ "grad_norm": 124.55913875585814,
527
+ "learning_rate": 8.455313244934323e-08,
528
+ "logits/chosen": -2.6776740550994873,
529
+ "logits/rejected": -2.6694130897521973,
530
+ "logps/chosen": -77.09771728515625,
531
+ "logps/rejected": -221.2405548095703,
532
+ "loss": 0.2385,
533
+ "rewards/accuracies": 0.887499988079071,
534
+ "rewards/chosen": 0.8464420437812805,
535
+ "rewards/margins": 4.465517520904541,
536
+ "rewards/rejected": -3.619075059890747,
537
+ "step": 310
538
+ },
539
+ {
540
+ "epoch": 0.69,
541
+ "grad_norm": 126.82250398979915,
542
+ "learning_rate": 8.317766145051056e-08,
543
+ "logits/chosen": -2.655475616455078,
544
+ "logits/rejected": -2.646566867828369,
545
+ "logps/chosen": -90.94697570800781,
546
+ "logps/rejected": -239.02603149414062,
547
+ "loss": 0.2527,
548
+ "rewards/accuracies": 0.856249988079071,
549
+ "rewards/chosen": 0.3390585482120514,
550
+ "rewards/margins": 4.169254302978516,
551
+ "rewards/rejected": -3.830195903778076,
552
+ "step": 320
553
+ },
554
+ {
555
+ "epoch": 0.71,
556
+ "grad_norm": 128.9818135312364,
557
+ "learning_rate": 8.175578849210893e-08,
558
+ "logits/chosen": -2.628675699234009,
559
+ "logits/rejected": -2.63059401512146,
560
+ "logps/chosen": -79.80452728271484,
561
+ "logps/rejected": -214.23886108398438,
562
+ "loss": 0.2346,
563
+ "rewards/accuracies": 0.8812500238418579,
564
+ "rewards/chosen": 0.7122211456298828,
565
+ "rewards/margins": 4.485437393188477,
566
+ "rewards/rejected": -3.7732162475585938,
567
+ "step": 330
568
+ },
569
+ {
570
+ "epoch": 0.73,
571
+ "grad_norm": 170.62278854131603,
572
+ "learning_rate": 8.0289502192041e-08,
573
+ "logits/chosen": -2.6830954551696777,
574
+ "logits/rejected": -2.639918804168701,
575
+ "logps/chosen": -84.37899780273438,
576
+ "logps/rejected": -245.4980926513672,
577
+ "loss": 0.2358,
578
+ "rewards/accuracies": 0.893750011920929,
579
+ "rewards/chosen": 0.7535432577133179,
580
+ "rewards/margins": 4.955996990203857,
581
+ "rewards/rejected": -4.20245361328125,
582
+ "step": 340
583
+ },
584
+ {
585
+ "epoch": 0.75,
586
+ "grad_norm": 208.2810878095979,
587
+ "learning_rate": 7.878085328428368e-08,
588
+ "logits/chosen": -2.6215579509735107,
589
+ "logits/rejected": -2.6440749168395996,
590
+ "logps/chosen": -88.97381591796875,
591
+ "logps/rejected": -217.13919067382812,
592
+ "loss": 0.239,
593
+ "rewards/accuracies": 0.84375,
594
+ "rewards/chosen": 0.4338725507259369,
595
+ "rewards/margins": 3.943838596343994,
596
+ "rewards/rejected": -3.5099663734436035,
597
+ "step": 350
598
+ },
599
+ {
600
+ "epoch": 0.77,
601
+ "grad_norm": 129.5194204672045,
602
+ "learning_rate": 7.723195175075135e-08,
603
+ "logits/chosen": -2.633841037750244,
604
+ "logits/rejected": -2.6651291847229004,
605
+ "logps/chosen": -85.39970397949219,
606
+ "logps/rejected": -230.7396240234375,
607
+ "loss": 0.2339,
608
+ "rewards/accuracies": 0.925000011920929,
609
+ "rewards/chosen": 0.3009163737297058,
610
+ "rewards/margins": 4.652897357940674,
611
+ "rewards/rejected": -4.351980686187744,
612
+ "step": 360
613
+ },
614
+ {
615
+ "epoch": 0.79,
616
+ "grad_norm": 164.1373245285756,
617
+ "learning_rate": 7.56449638702953e-08,
618
+ "logits/chosen": -2.614220142364502,
619
+ "logits/rejected": -2.6127521991729736,
620
+ "logps/chosen": -81.96704864501953,
621
+ "logps/rejected": -249.83847045898438,
622
+ "loss": 0.2353,
623
+ "rewards/accuracies": 0.893750011920929,
624
+ "rewards/chosen": 0.21380162239074707,
625
+ "rewards/margins": 4.993688106536865,
626
+ "rewards/rejected": -4.779886245727539,
627
+ "step": 370
628
+ },
629
+ {
630
+ "epoch": 0.81,
631
+ "grad_norm": 145.70422295065282,
632
+ "learning_rate": 7.402210918896688e-08,
633
+ "logits/chosen": -2.6128079891204834,
634
+ "logits/rejected": -2.6537294387817383,
635
+ "logps/chosen": -92.5042953491211,
636
+ "logps/rejected": -237.99649047851562,
637
+ "loss": 0.2447,
638
+ "rewards/accuracies": 0.9125000238418579,
639
+ "rewards/chosen": -0.36580491065979004,
640
+ "rewards/margins": 4.624101638793945,
641
+ "rewards/rejected": -4.9899067878723145,
642
+ "step": 380
643
+ },
644
+ {
645
+ "epoch": 0.83,
646
+ "grad_norm": 187.73130049718608,
647
+ "learning_rate": 7.236565741578162e-08,
648
+ "logits/chosen": -2.6431612968444824,
649
+ "logits/rejected": -2.640349864959717,
650
+ "logps/chosen": -100.9793472290039,
651
+ "logps/rejected": -250.1385498046875,
652
+ "loss": 0.2144,
653
+ "rewards/accuracies": 0.856249988079071,
654
+ "rewards/chosen": -0.3357706665992737,
655
+ "rewards/margins": 4.967101097106934,
656
+ "rewards/rejected": -5.3028717041015625,
657
+ "step": 390
658
+ },
659
+ {
660
+ "epoch": 0.86,
661
+ "grad_norm": 149.7762972401691,
662
+ "learning_rate": 7.067792524832604e-08,
663
+ "logits/chosen": -2.6969475746154785,
664
+ "logits/rejected": -2.7058839797973633,
665
+ "logps/chosen": -87.51020812988281,
666
+ "logps/rejected": -230.3865966796875,
667
+ "loss": 0.2331,
668
+ "rewards/accuracies": 0.9125000238418579,
669
+ "rewards/chosen": 0.4354115426540375,
670
+ "rewards/margins": 4.516859531402588,
671
+ "rewards/rejected": -4.081448554992676,
672
+ "step": 400
673
+ },
674
+ {
675
+ "epoch": 0.86,
676
+ "eval_logits/chosen": -2.538165807723999,
677
+ "eval_logits/rejected": -2.4967117309570312,
678
+ "eval_logps/chosen": -75.02692413330078,
679
+ "eval_logps/rejected": -232.38836669921875,
680
+ "eval_loss": 0.2088860273361206,
681
+ "eval_rewards/accuracies": 0.8684210777282715,
682
+ "eval_rewards/chosen": 0.7871010899543762,
683
+ "eval_rewards/margins": 5.678337574005127,
684
+ "eval_rewards/rejected": -4.891236782073975,
685
+ "eval_runtime": 27.7296,
686
+ "eval_samples_per_second": 10.855,
687
+ "eval_steps_per_second": 0.685,
688
+ "step": 400
689
+ },
690
+ {
691
+ "epoch": 0.88,
692
+ "grad_norm": 156.8019073660933,
693
+ "learning_rate": 6.896127313264642e-08,
694
+ "logits/chosen": -2.64673113822937,
695
+ "logits/rejected": -2.7038655281066895,
696
+ "logps/chosen": -80.8321533203125,
697
+ "logps/rejected": -215.33200073242188,
698
+ "loss": 0.2199,
699
+ "rewards/accuracies": 0.8999999761581421,
700
+ "rewards/chosen": 0.6177021861076355,
701
+ "rewards/margins": 4.641745090484619,
702
+ "rewards/rejected": -4.02404260635376,
703
+ "step": 410
704
+ },
705
+ {
706
+ "epoch": 0.9,
707
+ "grad_norm": 159.07138805831667,
708
+ "learning_rate": 6.721810196195174e-08,
709
+ "logits/chosen": -2.6594576835632324,
710
+ "logits/rejected": -2.6509957313537598,
711
+ "logps/chosen": -85.29691314697266,
712
+ "logps/rejected": -228.0242156982422,
713
+ "loss": 0.2228,
714
+ "rewards/accuracies": 0.9125000238418579,
715
+ "rewards/chosen": 0.5911294221878052,
716
+ "rewards/margins": 4.615528583526611,
717
+ "rewards/rejected": -4.024399280548096,
718
+ "step": 420
719
+ },
720
+ {
721
+ "epoch": 0.92,
722
+ "grad_norm": 168.14810806211554,
723
+ "learning_rate": 6.545084971874738e-08,
724
+ "logits/chosen": -2.6609866619110107,
725
+ "logits/rejected": -2.6674180030822754,
726
+ "logps/chosen": -82.54444885253906,
727
+ "logps/rejected": -218.54458618164062,
728
+ "loss": 0.242,
729
+ "rewards/accuracies": 0.90625,
730
+ "rewards/chosen": 0.23233623802661896,
731
+ "rewards/margins": 4.690705299377441,
732
+ "rewards/rejected": -4.458369255065918,
733
+ "step": 430
734
+ },
735
+ {
736
+ "epoch": 0.94,
737
+ "grad_norm": 168.9123818369519,
738
+ "learning_rate": 6.3661988065096e-08,
739
+ "logits/chosen": -2.653125286102295,
740
+ "logits/rejected": -2.6496918201446533,
741
+ "logps/chosen": -90.27217102050781,
742
+ "logps/rejected": -240.4198455810547,
743
+ "loss": 0.2216,
744
+ "rewards/accuracies": 0.918749988079071,
745
+ "rewards/chosen": 0.1593785583972931,
746
+ "rewards/margins": 4.545279502868652,
747
+ "rewards/rejected": -4.38590145111084,
748
+ "step": 440
749
+ },
750
+ {
751
+ "epoch": 0.96,
752
+ "grad_norm": 148.89537306570088,
753
+ "learning_rate": 6.185401888577488e-08,
754
+ "logits/chosen": -2.658726930618286,
755
+ "logits/rejected": -2.6689276695251465,
756
+ "logps/chosen": -80.66201782226562,
757
+ "logps/rejected": -221.6832733154297,
758
+ "loss": 0.2592,
759
+ "rewards/accuracies": 0.8500000238418579,
760
+ "rewards/chosen": 0.2890837788581848,
761
+ "rewards/margins": 4.448688507080078,
762
+ "rewards/rejected": -4.159604549407959,
763
+ "step": 450
764
+ },
765
+ {
766
+ "epoch": 0.98,
767
+ "grad_norm": 149.76478012255194,
768
+ "learning_rate": 6.002947078916364e-08,
769
+ "logits/chosen": -2.590343713760376,
770
+ "logits/rejected": -2.6157240867614746,
771
+ "logps/chosen": -88.16920471191406,
772
+ "logps/rejected": -197.1310577392578,
773
+ "loss": 0.2411,
774
+ "rewards/accuracies": 0.875,
775
+ "rewards/chosen": -0.010756964795291424,
776
+ "rewards/margins": 3.987351179122925,
777
+ "rewards/rejected": -3.998108386993408,
778
+ "step": 460
779
+ },
780
+ {
781
+ "epoch": 1.01,
782
+ "grad_norm": 69.73653024843325,
783
+ "learning_rate": 5.8190895570756884e-08,
784
+ "logits/chosen": -2.6571004390716553,
785
+ "logits/rejected": -2.6537766456604004,
786
+ "logps/chosen": -91.45221710205078,
787
+ "logps/rejected": -221.7407684326172,
788
+ "loss": 0.1735,
789
+ "rewards/accuracies": 0.8999999761581421,
790
+ "rewards/chosen": 0.4764205813407898,
791
+ "rewards/margins": 5.178285121917725,
792
+ "rewards/rejected": -4.701864242553711,
793
+ "step": 470
794
+ },
795
+ {
796
+ "epoch": 1.03,
797
+ "grad_norm": 74.69175649941268,
798
+ "learning_rate": 5.634086464424742e-08,
799
+ "logits/chosen": -2.629711866378784,
800
+ "logits/rejected": -2.628819227218628,
801
+ "logps/chosen": -81.51197052001953,
802
+ "logps/rejected": -247.3602294921875,
803
+ "loss": 0.1179,
804
+ "rewards/accuracies": 0.9937499761581421,
805
+ "rewards/chosen": 0.6573706865310669,
806
+ "rewards/margins": 5.840112209320068,
807
+ "rewards/rejected": -5.182741641998291,
808
+ "step": 480
809
+ },
810
+ {
811
+ "epoch": 1.05,
812
+ "grad_norm": 81.17366180243518,
813
+ "learning_rate": 5.4481965445171675e-08,
814
+ "logits/chosen": -2.6255042552948,
815
+ "logits/rejected": -2.6369516849517822,
816
+ "logps/chosen": -92.74832153320312,
817
+ "logps/rejected": -253.36770629882812,
818
+ "loss": 0.1127,
819
+ "rewards/accuracies": 0.987500011920929,
820
+ "rewards/chosen": 0.6330462098121643,
821
+ "rewards/margins": 5.925755500793457,
822
+ "rewards/rejected": -5.2927093505859375,
823
+ "step": 490
824
+ },
825
+ {
826
+ "epoch": 1.07,
827
+ "grad_norm": 42.75765323253819,
828
+ "learning_rate": 5.261679781214719e-08,
829
+ "logits/chosen": -2.615849018096924,
830
+ "logits/rejected": -2.583253860473633,
831
+ "logps/chosen": -81.4043197631836,
832
+ "logps/rejected": -224.62063598632812,
833
+ "loss": 0.0874,
834
+ "rewards/accuracies": 0.987500011920929,
835
+ "rewards/chosen": 0.7663254737854004,
836
+ "rewards/margins": 5.37575101852417,
837
+ "rewards/rejected": -4.6094255447387695,
838
+ "step": 500
839
+ },
840
+ {
841
+ "epoch": 1.07,
842
+ "eval_logits/chosen": -2.476076364517212,
843
+ "eval_logits/rejected": -2.4322872161865234,
844
+ "eval_logps/chosen": -76.55270385742188,
845
+ "eval_logps/rejected": -240.9202117919922,
846
+ "eval_loss": 0.1871984452009201,
847
+ "eval_rewards/accuracies": 0.8815789222717285,
848
+ "eval_rewards/chosen": 0.6345227956771851,
849
+ "eval_rewards/margins": 6.378945350646973,
850
+ "eval_rewards/rejected": -5.744422435760498,
851
+ "eval_runtime": 27.7245,
852
+ "eval_samples_per_second": 10.857,
853
+ "eval_steps_per_second": 0.685,
854
+ "step": 500
855
+ },
856
+ {
857
+ "epoch": 1.09,
858
+ "grad_norm": 80.40588664774708,
859
+ "learning_rate": 5.074797035076318e-08,
860
+ "logits/chosen": -2.600032091140747,
861
+ "logits/rejected": -2.6238975524902344,
862
+ "logps/chosen": -80.40613555908203,
863
+ "logps/rejected": -222.7124786376953,
864
+ "loss": 0.1237,
865
+ "rewards/accuracies": 0.9750000238418579,
866
+ "rewards/chosen": 0.6139314770698547,
867
+ "rewards/margins": 5.547031402587891,
868
+ "rewards/rejected": -4.93310022354126,
869
+ "step": 510
870
+ },
871
+ {
872
+ "epoch": 1.11,
873
+ "grad_norm": 144.58586359448068,
874
+ "learning_rate": 4.887809678520975e-08,
875
+ "logits/chosen": -2.5908000469207764,
876
+ "logits/rejected": -2.598961114883423,
877
+ "logps/chosen": -84.61862182617188,
878
+ "logps/rejected": -278.3133239746094,
879
+ "loss": 0.1324,
880
+ "rewards/accuracies": 0.9750000238418579,
881
+ "rewards/chosen": 0.5282140970230103,
882
+ "rewards/margins": 6.792176723480225,
883
+ "rewards/rejected": -6.263962745666504,
884
+ "step": 520
885
+ },
886
+ {
887
+ "epoch": 1.13,
888
+ "grad_norm": 93.29727808293761,
889
+ "learning_rate": 4.7009792302748285e-08,
890
+ "logits/chosen": -2.5732998847961426,
891
+ "logits/rejected": -2.5646510124206543,
892
+ "logps/chosen": -93.49433898925781,
893
+ "logps/rejected": -266.72454833984375,
894
+ "loss": 0.1078,
895
+ "rewards/accuracies": 0.9624999761581421,
896
+ "rewards/chosen": 0.5050237774848938,
897
+ "rewards/margins": 6.757340908050537,
898
+ "rewards/rejected": -6.252317428588867,
899
+ "step": 530
900
+ },
901
+ {
902
+ "epoch": 1.16,
903
+ "grad_norm": 100.11556094998792,
904
+ "learning_rate": 4.514566989613559e-08,
905
+ "logits/chosen": -2.5525121688842773,
906
+ "logits/rejected": -2.5795490741729736,
907
+ "logps/chosen": -82.06108856201172,
908
+ "logps/rejected": -232.47549438476562,
909
+ "loss": 0.0966,
910
+ "rewards/accuracies": 0.9750000238418579,
911
+ "rewards/chosen": 0.49808913469314575,
912
+ "rewards/margins": 5.77323579788208,
913
+ "rewards/rejected": -5.275146484375,
914
+ "step": 540
915
+ },
916
+ {
917
+ "epoch": 1.18,
918
+ "grad_norm": 42.560167765291645,
919
+ "learning_rate": 4.328833670911724e-08,
920
+ "logits/chosen": -2.6239736080169678,
921
+ "logits/rejected": -2.5731518268585205,
922
+ "logps/chosen": -87.47773742675781,
923
+ "logps/rejected": -260.68145751953125,
924
+ "loss": 0.1202,
925
+ "rewards/accuracies": 0.96875,
926
+ "rewards/chosen": 0.3656201958656311,
927
+ "rewards/margins": 6.3458123207092285,
928
+ "rewards/rejected": -5.980191707611084,
929
+ "step": 550
930
+ },
931
+ {
932
+ "epoch": 1.2,
933
+ "grad_norm": 93.5777273080985,
934
+ "learning_rate": 4.144039039010124e-08,
935
+ "logits/chosen": -2.5428924560546875,
936
+ "logits/rejected": -2.573913097381592,
937
+ "logps/chosen": -83.42646789550781,
938
+ "logps/rejected": -236.0640106201172,
939
+ "loss": 0.1036,
940
+ "rewards/accuracies": 0.9375,
941
+ "rewards/chosen": 0.4360501170158386,
942
+ "rewards/margins": 5.618575096130371,
943
+ "rewards/rejected": -5.182524681091309,
944
+ "step": 560
945
+ },
946
+ {
947
+ "epoch": 1.22,
948
+ "grad_norm": 131.2737748435812,
949
+ "learning_rate": 3.960441545911204e-08,
950
+ "logits/chosen": -2.494786024093628,
951
+ "logits/rejected": -2.546558380126953,
952
+ "logps/chosen": -83.13180541992188,
953
+ "logps/rejected": -227.36331176757812,
954
+ "loss": 0.0986,
955
+ "rewards/accuracies": 0.96875,
956
+ "rewards/chosen": 0.48235464096069336,
957
+ "rewards/margins": 5.769989013671875,
958
+ "rewards/rejected": -5.28763484954834,
959
+ "step": 570
960
+ },
961
+ {
962
+ "epoch": 1.24,
963
+ "grad_norm": 111.6736032390892,
964
+ "learning_rate": 3.778297969310529e-08,
965
+ "logits/chosen": -2.5785977840423584,
966
+ "logits/rejected": -2.6077938079833984,
967
+ "logps/chosen": -99.91156768798828,
968
+ "logps/rejected": -274.0140075683594,
969
+ "loss": 0.1194,
970
+ "rewards/accuracies": 0.96875,
971
+ "rewards/chosen": 0.2391074150800705,
972
+ "rewards/margins": 6.7299485206604,
973
+ "rewards/rejected": -6.490840911865234,
974
+ "step": 580
975
+ },
976
+ {
977
+ "epoch": 1.26,
978
+ "grad_norm": 85.48495941913924,
979
+ "learning_rate": 3.5978630534699866e-08,
980
+ "logits/chosen": -2.5449576377868652,
981
+ "logits/rejected": -2.5609145164489746,
982
+ "logps/chosen": -79.68949127197266,
983
+ "logps/rejected": -262.533203125,
984
+ "loss": 0.1049,
985
+ "rewards/accuracies": 0.9750000238418579,
986
+ "rewards/chosen": 0.3723651170730591,
987
+ "rewards/margins": 6.530053615570068,
988
+ "rewards/rejected": -6.157688617706299,
989
+ "step": 590
990
+ },
991
+ {
992
+ "epoch": 1.28,
993
+ "grad_norm": 113.71617513097341,
994
+ "learning_rate": 3.41938915293488e-08,
995
+ "logits/chosen": -2.5033059120178223,
996
+ "logits/rejected": -2.535037040710449,
997
+ "logps/chosen": -72.401123046875,
998
+ "logps/rejected": -242.56362915039062,
999
+ "loss": 0.1217,
1000
+ "rewards/accuracies": 0.9750000238418579,
1001
+ "rewards/chosen": 0.5488067865371704,
1002
+ "rewards/margins": 6.211435794830322,
1003
+ "rewards/rejected": -5.662629127502441,
1004
+ "step": 600
1005
+ },
1006
+ {
1007
+ "epoch": 1.28,
1008
+ "eval_logits/chosen": -2.417189359664917,
1009
+ "eval_logits/rejected": -2.374096632003784,
1010
+ "eval_logps/chosen": -80.61611938476562,
1011
+ "eval_logps/rejected": -250.38265991210938,
1012
+ "eval_loss": 0.1832417994737625,
1013
+ "eval_rewards/accuracies": 0.8684210777282715,
1014
+ "eval_rewards/chosen": 0.22818052768707275,
1015
+ "eval_rewards/margins": 6.918849945068359,
1016
+ "eval_rewards/rejected": -6.690669536590576,
1017
+ "eval_runtime": 27.7688,
1018
+ "eval_samples_per_second": 10.84,
1019
+ "eval_steps_per_second": 0.684,
1020
+ "step": 600
1021
+ },
1022
+ {
1023
+ "epoch": 1.31,
1024
+ "grad_norm": 72.12098063143571,
1025
+ "learning_rate": 3.2431258795932856e-08,
1026
+ "logits/chosen": -2.5755138397216797,
1027
+ "logits/rejected": -2.567202568054199,
1028
+ "logps/chosen": -85.8843002319336,
1029
+ "logps/rejected": -263.94622802734375,
1030
+ "loss": 0.0791,
1031
+ "rewards/accuracies": 0.987500011920929,
1032
+ "rewards/chosen": 0.24374854564666748,
1033
+ "rewards/margins": 6.808441162109375,
1034
+ "rewards/rejected": -6.564692497253418,
1035
+ "step": 610
1036
+ },
1037
+ {
1038
+ "epoch": 1.33,
1039
+ "grad_norm": 84.30497343810121,
1040
+ "learning_rate": 3.0693197535712687e-08,
1041
+ "logits/chosen": -2.556939125061035,
1042
+ "logits/rejected": -2.553295373916626,
1043
+ "logps/chosen": -93.59156036376953,
1044
+ "logps/rejected": -255.7102813720703,
1045
+ "loss": 0.104,
1046
+ "rewards/accuracies": 0.9750000238418579,
1047
+ "rewards/chosen": -0.050883639603853226,
1048
+ "rewards/margins": 6.550814628601074,
1049
+ "rewards/rejected": -6.601698398590088,
1050
+ "step": 620
1051
+ },
1052
+ {
1053
+ "epoch": 1.35,
1054
+ "grad_norm": 97.92510560815347,
1055
+ "learning_rate": 2.8982138584521732e-08,
1056
+ "logits/chosen": -2.523881435394287,
1057
+ "logits/rejected": -2.5281319618225098,
1058
+ "logps/chosen": -87.27870178222656,
1059
+ "logps/rejected": -281.7742919921875,
1060
+ "loss": 0.0988,
1061
+ "rewards/accuracies": 0.96875,
1062
+ "rewards/chosen": 0.02099490724503994,
1063
+ "rewards/margins": 6.995011329650879,
1064
+ "rewards/rejected": -6.974015712738037,
1065
+ "step": 630
1066
+ },
1067
+ {
1068
+ "epoch": 1.37,
1069
+ "grad_norm": 144.6029547590291,
1070
+ "learning_rate": 2.730047501302266e-08,
1071
+ "logits/chosen": -2.511509656906128,
1072
+ "logits/rejected": -2.491812229156494,
1073
+ "logps/chosen": -80.01902770996094,
1074
+ "logps/rejected": -249.27377319335938,
1075
+ "loss": 0.1017,
1076
+ "rewards/accuracies": 0.9750000238418579,
1077
+ "rewards/chosen": 0.18679390847682953,
1078
+ "rewards/margins": 6.189632415771484,
1079
+ "rewards/rejected": -6.002839088439941,
1080
+ "step": 640
1081
+ },
1082
+ {
1083
+ "epoch": 1.39,
1084
+ "grad_norm": 51.26390219844907,
1085
+ "learning_rate": 2.5650558779781632e-08,
1086
+ "logits/chosen": -2.468006134033203,
1087
+ "logits/rejected": -2.5272440910339355,
1088
+ "logps/chosen": -92.99039459228516,
1089
+ "logps/rejected": -263.6413269042969,
1090
+ "loss": 0.1057,
1091
+ "rewards/accuracies": 0.96875,
1092
+ "rewards/chosen": 0.007961297407746315,
1093
+ "rewards/margins": 6.804115295410156,
1094
+ "rewards/rejected": -6.7961530685424805,
1095
+ "step": 650
1096
+ },
1097
+ {
1098
+ "epoch": 1.41,
1099
+ "grad_norm": 74.26596525719256,
1100
+ "learning_rate": 2.403469744184154e-08,
1101
+ "logits/chosen": -2.4938857555389404,
1102
+ "logits/rejected": -2.4678053855895996,
1103
+ "logps/chosen": -83.55183410644531,
1104
+ "logps/rejected": -240.530517578125,
1105
+ "loss": 0.0828,
1106
+ "rewards/accuracies": 0.987500011920929,
1107
+ "rewards/chosen": 0.22306349873542786,
1108
+ "rewards/margins": 7.2999396324157715,
1109
+ "rewards/rejected": -7.076876640319824,
1110
+ "step": 660
1111
+ },
1112
+ {
1113
+ "epoch": 1.43,
1114
+ "grad_norm": 67.9457822191306,
1115
+ "learning_rate": 2.2455150927394877e-08,
1116
+ "logits/chosen": -2.5248875617980957,
1117
+ "logits/rejected": -2.524660587310791,
1118
+ "logps/chosen": -91.45790100097656,
1119
+ "logps/rejected": -274.40240478515625,
1120
+ "loss": 0.1147,
1121
+ "rewards/accuracies": 0.956250011920929,
1122
+ "rewards/chosen": -0.18505340814590454,
1123
+ "rewards/margins": 6.406013488769531,
1124
+ "rewards/rejected": -6.591066837310791,
1125
+ "step": 670
1126
+ },
1127
+ {
1128
+ "epoch": 1.46,
1129
+ "grad_norm": 93.08918649012821,
1130
+ "learning_rate": 2.0914128375069722e-08,
1131
+ "logits/chosen": -2.503322124481201,
1132
+ "logits/rejected": -2.5299384593963623,
1133
+ "logps/chosen": -87.551025390625,
1134
+ "logps/rejected": -258.9871520996094,
1135
+ "loss": 0.0997,
1136
+ "rewards/accuracies": 0.96875,
1137
+ "rewards/chosen": 0.09187181293964386,
1138
+ "rewards/margins": 6.872333526611328,
1139
+ "rewards/rejected": -6.78046178817749,
1140
+ "step": 680
1141
+ },
1142
+ {
1143
+ "epoch": 1.48,
1144
+ "grad_norm": 47.52507218664577,
1145
+ "learning_rate": 1.9413785044249675e-08,
1146
+ "logits/chosen": -2.4908030033111572,
1147
+ "logits/rejected": -2.4941422939300537,
1148
+ "logps/chosen": -92.09065246582031,
1149
+ "logps/rejected": -250.3391876220703,
1150
+ "loss": 0.0975,
1151
+ "rewards/accuracies": 0.987500011920929,
1152
+ "rewards/chosen": 0.06340374052524567,
1153
+ "rewards/margins": 6.6621413230896,
1154
+ "rewards/rejected": -6.598738193511963,
1155
+ "step": 690
1156
+ },
1157
+ {
1158
+ "epoch": 1.5,
1159
+ "grad_norm": 165.96896499860316,
1160
+ "learning_rate": 1.795621930074879e-08,
1161
+ "logits/chosen": -2.4552981853485107,
1162
+ "logits/rejected": -2.4628846645355225,
1163
+ "logps/chosen": -87.13786315917969,
1164
+ "logps/rejected": -213.85501098632812,
1165
+ "loss": 0.0966,
1166
+ "rewards/accuracies": 0.9750000238418579,
1167
+ "rewards/chosen": 0.18398265540599823,
1168
+ "rewards/margins": 5.881737232208252,
1169
+ "rewards/rejected": -5.697754383087158,
1170
+ "step": 700
1171
+ },
1172
+ {
1173
+ "epoch": 1.5,
1174
+ "eval_logits/chosen": -2.393958330154419,
1175
+ "eval_logits/rejected": -2.3503360748291016,
1176
+ "eval_logps/chosen": -81.04850006103516,
1177
+ "eval_logps/rejected": -253.6011962890625,
1178
+ "eval_loss": 0.18069878220558167,
1179
+ "eval_rewards/accuracies": 0.8815789222717285,
1180
+ "eval_rewards/chosen": 0.18494324386119843,
1181
+ "eval_rewards/margins": 7.197464942932129,
1182
+ "eval_rewards/rejected": -7.012521743774414,
1183
+ "eval_runtime": 27.7257,
1184
+ "eval_samples_per_second": 10.856,
1185
+ "eval_steps_per_second": 0.685,
1186
+ "step": 700
1187
+ },
1188
+ {
1189
+ "epoch": 1.52,
1190
+ "grad_norm": 159.91079268011805,
1191
+ "learning_rate": 1.6543469682057103e-08,
1192
+ "logits/chosen": -2.5185043811798096,
1193
+ "logits/rejected": -2.4764022827148438,
1194
+ "logps/chosen": -84.46337890625,
1195
+ "logps/rejected": -264.3829345703125,
1196
+ "loss": 0.0926,
1197
+ "rewards/accuracies": 0.96875,
1198
+ "rewards/chosen": 0.18679778277873993,
1199
+ "rewards/margins": 7.148007392883301,
1200
+ "rewards/rejected": -6.961209297180176,
1201
+ "step": 710
1202
+ },
1203
+ {
1204
+ "epoch": 1.54,
1205
+ "grad_norm": 48.047361074216745,
1206
+ "learning_rate": 1.5177512046261666e-08,
1207
+ "logits/chosen": -2.484215259552002,
1208
+ "logits/rejected": -2.507033109664917,
1209
+ "logps/chosen": -84.62954711914062,
1210
+ "logps/rejected": -256.36737060546875,
1211
+ "loss": 0.0983,
1212
+ "rewards/accuracies": 0.9750000238418579,
1213
+ "rewards/chosen": 0.39748048782348633,
1214
+ "rewards/margins": 7.362854957580566,
1215
+ "rewards/rejected": -6.965374946594238,
1216
+ "step": 720
1217
+ },
1218
+ {
1219
+ "epoch": 1.56,
1220
+ "grad_norm": 112.9192400527363,
1221
+ "learning_rate": 1.3860256808630427e-08,
1222
+ "logits/chosen": -2.555844783782959,
1223
+ "logits/rejected": -2.478036403656006,
1224
+ "logps/chosen": -88.8625259399414,
1225
+ "logps/rejected": -237.5943145751953,
1226
+ "loss": 0.107,
1227
+ "rewards/accuracies": 0.9750000238418579,
1228
+ "rewards/chosen": 0.22078856825828552,
1229
+ "rewards/margins": 6.411069393157959,
1230
+ "rewards/rejected": -6.190280914306641,
1231
+ "step": 730
1232
+ },
1233
+ {
1234
+ "epoch": 1.58,
1235
+ "grad_norm": 80.75197492702765,
1236
+ "learning_rate": 1.2593546269723649e-08,
1237
+ "logits/chosen": -2.4870190620422363,
1238
+ "logits/rejected": -2.470374584197998,
1239
+ "logps/chosen": -87.91392517089844,
1240
+ "logps/rejected": -268.05450439453125,
1241
+ "loss": 0.103,
1242
+ "rewards/accuracies": 0.9624999761581421,
1243
+ "rewards/chosen": 0.023371051996946335,
1244
+ "rewards/margins": 7.2978057861328125,
1245
+ "rewards/rejected": -7.274435520172119,
1246
+ "step": 740
1247
+ },
1248
+ {
1249
+ "epoch": 1.61,
1250
+ "grad_norm": 108.88349531511021,
1251
+ "learning_rate": 1.1379152038770029e-08,
1252
+ "logits/chosen": -2.498399019241333,
1253
+ "logits/rejected": -2.522373676300049,
1254
+ "logps/chosen": -79.47743225097656,
1255
+ "logps/rejected": -267.198486328125,
1256
+ "loss": 0.1003,
1257
+ "rewards/accuracies": 0.9624999761581421,
1258
+ "rewards/chosen": 0.3667542338371277,
1259
+ "rewards/margins": 7.25054931640625,
1260
+ "rewards/rejected": -6.883795261383057,
1261
+ "step": 750
1262
+ },
1263
+ {
1264
+ "epoch": 1.63,
1265
+ "grad_norm": 106.04443890243027,
1266
+ "learning_rate": 1.0218772555910953e-08,
1267
+ "logits/chosen": -2.49924898147583,
1268
+ "logits/rejected": -2.533095121383667,
1269
+ "logps/chosen": -88.90888977050781,
1270
+ "logps/rejected": -216.93911743164062,
1271
+ "loss": 0.1093,
1272
+ "rewards/accuracies": 0.949999988079071,
1273
+ "rewards/chosen": 0.10971729457378387,
1274
+ "rewards/margins": 5.331303596496582,
1275
+ "rewards/rejected": -5.221585750579834,
1276
+ "step": 760
1277
+ },
1278
+ {
1279
+ "epoch": 1.65,
1280
+ "grad_norm": 119.24007089314645,
1281
+ "learning_rate": 9.114030716778432e-09,
1282
+ "logits/chosen": -2.522010087966919,
1283
+ "logits/rejected": -2.510510206222534,
1284
+ "logps/chosen": -85.21174621582031,
1285
+ "logps/rejected": -258.8554992675781,
1286
+ "loss": 0.0854,
1287
+ "rewards/accuracies": 0.981249988079071,
1288
+ "rewards/chosen": 0.3466641306877136,
1289
+ "rewards/margins": 7.279857635498047,
1290
+ "rewards/rejected": -6.933194160461426,
1291
+ "step": 770
1292
+ },
1293
+ {
1294
+ "epoch": 1.67,
1295
+ "grad_norm": 134.5502526352079,
1296
+ "learning_rate": 8.066471602728803e-09,
1297
+ "logits/chosen": -2.4917855262756348,
1298
+ "logits/rejected": -2.505199670791626,
1299
+ "logps/chosen": -90.26045989990234,
1300
+ "logps/rejected": -252.13491821289062,
1301
+ "loss": 0.085,
1302
+ "rewards/accuracies": 0.956250011920929,
1303
+ "rewards/chosen": 0.11037313938140869,
1304
+ "rewards/margins": 6.754474639892578,
1305
+ "rewards/rejected": -6.644102573394775,
1306
+ "step": 780
1307
+ },
1308
+ {
1309
+ "epoch": 1.69,
1310
+ "grad_norm": 70.79957881261483,
1311
+ "learning_rate": 7.077560319906695e-09,
1312
+ "logits/chosen": -2.4431352615356445,
1313
+ "logits/rejected": -2.495988368988037,
1314
+ "logps/chosen": -79.27693176269531,
1315
+ "logps/rejected": -225.5204620361328,
1316
+ "loss": 0.0872,
1317
+ "rewards/accuracies": 0.96875,
1318
+ "rewards/chosen": 0.49195781350135803,
1319
+ "rewards/margins": 6.213925361633301,
1320
+ "rewards/rejected": -5.721967697143555,
1321
+ "step": 790
1322
+ },
1323
+ {
1324
+ "epoch": 1.71,
1325
+ "grad_norm": 111.87130217318148,
1326
+ "learning_rate": 6.148679950161672e-09,
1327
+ "logits/chosen": -2.475398302078247,
1328
+ "logits/rejected": -2.5179803371429443,
1329
+ "logps/chosen": -89.04031372070312,
1330
+ "logps/rejected": -270.52117919921875,
1331
+ "loss": 0.0755,
1332
+ "rewards/accuracies": 0.981249988079071,
1333
+ "rewards/chosen": 0.1288180649280548,
1334
+ "rewards/margins": 7.451496124267578,
1335
+ "rewards/rejected": -7.322678565979004,
1336
+ "step": 800
1337
+ },
1338
+ {
1339
+ "epoch": 1.71,
1340
+ "eval_logits/chosen": -2.3867335319519043,
1341
+ "eval_logits/rejected": -2.3436965942382812,
1342
+ "eval_logps/chosen": -79.67388153076172,
1343
+ "eval_logps/rejected": -253.01498413085938,
1344
+ "eval_loss": 0.18022477626800537,
1345
+ "eval_rewards/accuracies": 0.8947368264198303,
1346
+ "eval_rewards/chosen": 0.32240524888038635,
1347
+ "eval_rewards/margins": 7.276304721832275,
1348
+ "eval_rewards/rejected": -6.953899383544922,
1349
+ "eval_runtime": 27.7976,
1350
+ "eval_samples_per_second": 10.828,
1351
+ "eval_steps_per_second": 0.684,
1352
+ "step": 800
1353
+ },
1354
+ {
1355
+ "epoch": 1.73,
1356
+ "grad_norm": 110.49991026961725,
1357
+ "learning_rate": 5.281129616683166e-09,
1358
+ "logits/chosen": -2.5127198696136475,
1359
+ "logits/rejected": -2.512899160385132,
1360
+ "logps/chosen": -86.99423217773438,
1361
+ "logps/rejected": -254.9462432861328,
1362
+ "loss": 0.0962,
1363
+ "rewards/accuracies": 0.9437500238418579,
1364
+ "rewards/chosen": 0.06987112015485764,
1365
+ "rewards/margins": 6.7237701416015625,
1366
+ "rewards/rejected": -6.6538987159729,
1367
+ "step": 810
1368
+ },
1369
+ {
1370
+ "epoch": 1.76,
1371
+ "grad_norm": 108.21879194257498,
1372
+ "learning_rate": 4.476122667059206e-09,
1373
+ "logits/chosen": -2.4863157272338867,
1374
+ "logits/rejected": -2.4990665912628174,
1375
+ "logps/chosen": -86.436279296875,
1376
+ "logps/rejected": -256.197998046875,
1377
+ "loss": 0.1138,
1378
+ "rewards/accuracies": 0.981249988079071,
1379
+ "rewards/chosen": 0.21532371640205383,
1380
+ "rewards/margins": 6.994332313537598,
1381
+ "rewards/rejected": -6.779008388519287,
1382
+ "step": 820
1383
+ },
1384
+ {
1385
+ "epoch": 1.78,
1386
+ "grad_norm": 76.19186603851071,
1387
+ "learning_rate": 3.734784976300165e-09,
1388
+ "logits/chosen": -2.4668314456939697,
1389
+ "logits/rejected": -2.4680819511413574,
1390
+ "logps/chosen": -89.12982940673828,
1391
+ "logps/rejected": -259.98974609375,
1392
+ "loss": 0.1034,
1393
+ "rewards/accuracies": 0.956250011920929,
1394
+ "rewards/chosen": 0.2635418772697449,
1395
+ "rewards/margins": 6.681604862213135,
1396
+ "rewards/rejected": -6.418063163757324,
1397
+ "step": 830
1398
+ },
1399
+ {
1400
+ "epoch": 1.8,
1401
+ "grad_norm": 89.59280146601071,
1402
+ "learning_rate": 3.0581533722006947e-09,
1403
+ "logits/chosen": -2.521101236343384,
1404
+ "logits/rejected": -2.5018861293792725,
1405
+ "logps/chosen": -85.35905456542969,
1406
+ "logps/rejected": -227.835205078125,
1407
+ "loss": 0.082,
1408
+ "rewards/accuracies": 0.9937499761581421,
1409
+ "rewards/chosen": 0.3482588827610016,
1410
+ "rewards/margins": 6.087993144989014,
1411
+ "rewards/rejected": -5.739733695983887,
1412
+ "step": 840
1413
+ },
1414
+ {
1415
+ "epoch": 1.82,
1416
+ "grad_norm": 132.74610726242307,
1417
+ "learning_rate": 2.4471741852423233e-09,
1418
+ "logits/chosen": -2.4352290630340576,
1419
+ "logits/rejected": -2.490058422088623,
1420
+ "logps/chosen": -89.67268371582031,
1421
+ "logps/rejected": -224.5243682861328,
1422
+ "loss": 0.1031,
1423
+ "rewards/accuracies": 0.9750000238418579,
1424
+ "rewards/chosen": 0.25209230184555054,
1425
+ "rewards/margins": 6.445871829986572,
1426
+ "rewards/rejected": -6.1937785148620605,
1427
+ "step": 850
1428
+ },
1429
+ {
1430
+ "epoch": 1.84,
1431
+ "grad_norm": 95.32823058547511,
1432
+ "learning_rate": 1.9027019250647036e-09,
1433
+ "logits/chosen": -2.5332539081573486,
1434
+ "logits/rejected": -2.539168357849121,
1435
+ "logps/chosen": -89.87360382080078,
1436
+ "logps/rejected": -267.892822265625,
1437
+ "loss": 0.0952,
1438
+ "rewards/accuracies": 0.981249988079071,
1439
+ "rewards/chosen": 0.2068900316953659,
1440
+ "rewards/margins": 6.962863922119141,
1441
+ "rewards/rejected": -6.755973815917969,
1442
+ "step": 860
1443
+ },
1444
+ {
1445
+ "epoch": 1.86,
1446
+ "grad_norm": 145.27755044968825,
1447
+ "learning_rate": 1.4254980853566246e-09,
1448
+ "logits/chosen": -2.4520225524902344,
1449
+ "logits/rejected": -2.4698426723480225,
1450
+ "logps/chosen": -85.50424194335938,
1451
+ "logps/rejected": -255.6893768310547,
1452
+ "loss": 0.0956,
1453
+ "rewards/accuracies": 0.9624999761581421,
1454
+ "rewards/chosen": 0.2579875588417053,
1455
+ "rewards/margins": 7.30988073348999,
1456
+ "rewards/rejected": -7.0518927574157715,
1457
+ "step": 870
1458
+ },
1459
+ {
1460
+ "epoch": 1.88,
1461
+ "grad_norm": 87.4062367020671,
1462
+ "learning_rate": 1.0162300788382261e-09,
1463
+ "logits/chosen": -2.488208770751953,
1464
+ "logits/rejected": -2.515857458114624,
1465
+ "logps/chosen": -90.32164001464844,
1466
+ "logps/rejected": -226.85171508789062,
1467
+ "loss": 0.0942,
1468
+ "rewards/accuracies": 0.96875,
1469
+ "rewards/chosen": 0.23069357872009277,
1470
+ "rewards/margins": 6.096683502197266,
1471
+ "rewards/rejected": -5.865989685058594,
1472
+ "step": 880
1473
+ },
1474
+ {
1475
+ "epoch": 1.91,
1476
+ "grad_norm": 129.60497880498312,
1477
+ "learning_rate": 6.754703038239329e-10,
1478
+ "logits/chosen": -2.4546942710876465,
1479
+ "logits/rejected": -2.498014450073242,
1480
+ "logps/chosen": -93.2281494140625,
1481
+ "logps/rejected": -241.6513671875,
1482
+ "loss": 0.0995,
1483
+ "rewards/accuracies": 0.987500011920929,
1484
+ "rewards/chosen": 0.2960784137248993,
1485
+ "rewards/margins": 6.222127914428711,
1486
+ "rewards/rejected": -5.926049709320068,
1487
+ "step": 890
1488
+ },
1489
+ {
1490
+ "epoch": 1.93,
1491
+ "grad_norm": 189.4146678200498,
1492
+ "learning_rate": 4.0369534367168944e-10,
1493
+ "logits/chosen": -2.4957008361816406,
1494
+ "logits/rejected": -2.4675281047821045,
1495
+ "logps/chosen": -77.89246368408203,
1496
+ "logps/rejected": -257.5621643066406,
1497
+ "loss": 0.1177,
1498
+ "rewards/accuracies": 0.9624999761581421,
1499
+ "rewards/chosen": 0.49517494440078735,
1500
+ "rewards/margins": 7.060118198394775,
1501
+ "rewards/rejected": -6.564943790435791,
1502
+ "step": 900
1503
+ },
1504
+ {
1505
+ "epoch": 1.93,
1506
+ "eval_logits/chosen": -2.384641647338867,
1507
+ "eval_logits/rejected": -2.3416695594787598,
1508
+ "eval_logps/chosen": -80.31053924560547,
1509
+ "eval_logps/rejected": -253.77731323242188,
1510
+ "eval_loss": 0.17933423817157745,
1511
+ "eval_rewards/accuracies": 0.8947368264198303,
1512
+ "eval_rewards/chosen": 0.25873973965644836,
1513
+ "eval_rewards/margins": 7.288874626159668,
1514
+ "eval_rewards/rejected": -7.030135154724121,
1515
+ "eval_runtime": 27.722,
1516
+ "eval_samples_per_second": 10.858,
1517
+ "eval_steps_per_second": 0.685,
1518
+ "step": 900
1519
+ },
1520
+ {
1521
+ "epoch": 1.95,
1522
+ "grad_norm": 122.05899501763088,
1523
+ "learning_rate": 2.0128530023804656e-10,
1524
+ "logits/chosen": -2.496187448501587,
1525
+ "logits/rejected": -2.4914169311523438,
1526
+ "logps/chosen": -88.0294189453125,
1527
+ "logps/rejected": -257.91558837890625,
1528
+ "loss": 0.1067,
1529
+ "rewards/accuracies": 0.9624999761581421,
1530
+ "rewards/chosen": 0.12927184998989105,
1531
+ "rewards/margins": 6.56699275970459,
1532
+ "rewards/rejected": -6.437721252441406,
1533
+ "step": 910
1534
+ },
1535
+ {
1536
+ "epoch": 1.97,
1537
+ "grad_norm": 65.9343080763714,
1538
+ "learning_rate": 6.852326227130833e-11,
1539
+ "logits/chosen": -2.521149158477783,
1540
+ "logits/rejected": -2.5130820274353027,
1541
+ "logps/chosen": -84.37642669677734,
1542
+ "logps/rejected": -246.3243865966797,
1543
+ "loss": 0.0899,
1544
+ "rewards/accuracies": 0.9937499761581421,
1545
+ "rewards/chosen": 0.4820326268672943,
1546
+ "rewards/margins": 6.372977256774902,
1547
+ "rewards/rejected": -5.890944957733154,
1548
+ "step": 920
1549
+ },
1550
+ {
1551
+ "epoch": 1.99,
1552
+ "grad_norm": 93.45308900990692,
1553
+ "learning_rate": 5.594909486328347e-12,
1554
+ "logits/chosen": -2.450763702392578,
1555
+ "logits/rejected": -2.467440128326416,
1556
+ "logps/chosen": -78.51066589355469,
1557
+ "logps/rejected": -229.6824188232422,
1558
+ "loss": 0.0926,
1559
+ "rewards/accuracies": 0.9624999761581421,
1560
+ "rewards/chosen": 0.3496243357658386,
1561
+ "rewards/margins": 5.816868305206299,
1562
+ "rewards/rejected": -5.4672441482543945,
1563
+ "step": 930
1564
+ },
1565
+ {
1566
+ "epoch": 2.0,
1567
+ "step": 934,
1568
+ "total_flos": 0.0,
1569
+ "train_loss": 0.21607095603003493,
1570
+ "train_runtime": 14362.1771,
1571
+ "train_samples_per_second": 4.163,
1572
+ "train_steps_per_second": 0.065
1573
+ }
1574
+ ],
1575
+ "logging_steps": 10,
1576
+ "max_steps": 934,
1577
+ "num_input_tokens_seen": 0,
1578
+ "num_train_epochs": 2,
1579
+ "save_steps": 50,
1580
+ "total_flos": 0.0,
1581
+ "train_batch_size": 2,
1582
+ "trial_name": null,
1583
+ "trial_params": null
1584
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:671b65610da39dbaa29fbb27313c65bc3ff72cb5c681dbc518d28d001dda8d40
3
+ size 6456