kurogane commited on
Commit
1917599
·
verified ·
1 Parent(s): 4bb2754

Upload 13 files

Browse files
chat_template.jinja ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ {%- set user_messages = messages | selectattr('role', 'equalto', 'user') | list %}
3
+ {%- macro output_available_tools(tools, message) %}
4
+ {%- if tools and (message == user_messages[-1]) %}
5
+ {{- '<|available_tools|>[' }}
6
+ {%- for tool in tools %}
7
+ {%- set tool = tool.function %}
8
+ {{- "{" }}
9
+ {%- for key, val in tool.items() if key != "return" %}
10
+ {%- if val is string %}
11
+ {{- "'" + key + "': '" + val + "'" }}
12
+ {%- else %}
13
+ {{- "'" + key + "': " + val|string }}
14
+ {%- endif %}
15
+ {%- if not loop.last %}
16
+ {{- ", " }}
17
+ {%- endif %}
18
+ {%- endfor %}
19
+ {{- "}" }}
20
+ {%- if not loop.last %}
21
+ {{- ", " }}
22
+ {%- else %}
23
+ {{- "]" }}
24
+ {%- endif %}
25
+ {%- endfor %}
26
+ {{- eos_token -}}
27
+ {%- endif %}
28
+ {%- endmacro %}
29
+
30
+ {%- macro output_tool_results(tool_results) %}
31
+ {{- '<|tool_results|>[' }}
32
+ {%- for tool_result in tool_results %}
33
+ {{- "{'content': " + tool_result.content|string + ", 'call_id': '" + tool_result.call_id + "'}" }}
34
+ {%- endfor %}
35
+ {{- ']' }}
36
+ {{- eos_token -}}
37
+ {%- endmacro %}
38
+
39
+ {%- macro output_tool_calls(tool_calls) %}
40
+ {{- '<|tool_calls|>[' }}
41
+ {%- for tool_call in tool_calls %}
42
+ {{- "{'id': '" + tool_call.id + "', 'name': '" + tool_call.name + "', 'arguments': " + tool_call.arguments|string + '}' }}
43
+ {%- endfor %}
44
+ {{- ']' }}
45
+ {%- endmacro %}
46
+
47
+ {%- for message in messages %}
48
+ {%- if message['role'] == 'user' %}
49
+ {%- if tools is defined %}
50
+ {{- output_available_tools(tools, message) }}
51
+ {%- endif %}
52
+ {{- '<|user|>' + message['content'] + eos_token -}}
53
+ {%- elif message['role'] == 'system' %}
54
+ {{- '<|system|>' + message['content'] + eos_token -}}
55
+ {%- elif message['role'] == 'assistant' %}
56
+ {% set assistant_content = "" %}
57
+ {%- if message.content is defined %}
58
+ {% set assistant_content = message.content %}
59
+ {%- endif %}
60
+ {%- if message.tool_calls is defined and message.tool_calls -%}
61
+ {{- '<|assistant|>' + assistant_content + output_tool_calls(message['tool_calls']) + eos_token -}}
62
+ {%- else %}
63
+ {{- '<|assistant|>' + assistant_content + eos_token }}
64
+ {%- endif %}
65
+ {%- elif message['role'] == 'tool_results' %}
66
+ {{- output_tool_results(message.tool_results) }}
67
+ {%- endif %}
68
+ {%- if loop.last and add_generation_prompt -%}
69
+ {{- '<|assistant|>' -}}
70
+ {%- endif -%}
71
+ {%- endfor %}
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Mamba2ForCausalLM"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "chunk_size": 256,
7
+ "conv_kernel": 4,
8
+ "eos_token_id": 2,
9
+ "expand": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.1,
14
+ "layer_norm_epsilon": 1e-05,
15
+ "max_position_embeddings": 512,
16
+ "model_type": "mamba2",
17
+ "n_groups": 4,
18
+ "num_heads": 32,
19
+ "num_hidden_layers": 32,
20
+ "pad_token_id": 3,
21
+ "rescale_prenorm_residual": false,
22
+ "residual_in_fp32": true,
23
+ "rms_norm": true,
24
+ "state_size": 128,
25
+ "tie_word_embeddings": false,
26
+ "time_step_floor": 0.0001,
27
+ "time_step_limit": [
28
+ 0.0,
29
+ Infinity
30
+ ],
31
+ "time_step_max": 0.1,
32
+ "time_step_min": 0.001,
33
+ "time_step_rank": 64,
34
+ "torch_dtype": "bfloat16",
35
+ "transformers_version": "4.55.4",
36
+ "use_bias": false,
37
+ "use_cache": false,
38
+ "use_conv_bias": true,
39
+ "vocab_size": 102400
40
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 3,
6
+ "transformers_version": "4.55.4",
7
+ "use_cache": false
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7472852cfdc63858a732278cd639daca485a5167b5ab1c98434c0e570003f6dd
3
+ size 892508496
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc9c7ed7cff466957b113e65e613292664b48a3fbe7baaf41f0895c3ddf195f9
3
+ size 1785126027
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:716d83e48b4ef0ecc739db256420344d8764b568cd2fec53ea128d8653e43804
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba530fd445749832f86cf59804574c56fa38a6eb8299a8f2b1ec24af867ba62d
3
+ size 1465
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<cls>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "<sep>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
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 sha256:008293028e1a9d9a1038d9b63d989a2319797dfeaa03f171093a57b33a3a8277
3
+ size 1831879
tokenizer_config.json ADDED
@@ -0,0 +1,171 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_dummy_prefix_space": false,
4
+ "add_eos_token": false,
5
+ "add_prefix_space": false,
6
+ "added_tokens_decoder": {
7
+ "0": {
8
+ "content": "<unk>",
9
+ "lstrip": false,
10
+ "normalized": false,
11
+ "rstrip": false,
12
+ "single_word": false,
13
+ "special": true
14
+ },
15
+ "1": {
16
+ "content": "<s>",
17
+ "lstrip": false,
18
+ "normalized": false,
19
+ "rstrip": false,
20
+ "single_word": false,
21
+ "special": true
22
+ },
23
+ "2": {
24
+ "content": "</s>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false,
29
+ "special": true
30
+ },
31
+ "3": {
32
+ "content": "<pad>",
33
+ "lstrip": false,
34
+ "normalized": false,
35
+ "rstrip": false,
36
+ "single_word": false,
37
+ "special": true
38
+ },
39
+ "4": {
40
+ "content": "<sep>",
41
+ "lstrip": false,
42
+ "normalized": false,
43
+ "rstrip": false,
44
+ "single_word": false,
45
+ "special": true
46
+ },
47
+ "5": {
48
+ "content": "<mask>",
49
+ "lstrip": false,
50
+ "normalized": false,
51
+ "rstrip": false,
52
+ "single_word": false,
53
+ "special": true
54
+ },
55
+ "6": {
56
+ "content": "<cls>",
57
+ "lstrip": false,
58
+ "normalized": false,
59
+ "rstrip": false,
60
+ "single_word": false,
61
+ "special": true
62
+ },
63
+ "7": {
64
+ "content": "<|system|>",
65
+ "lstrip": false,
66
+ "normalized": false,
67
+ "rstrip": false,
68
+ "single_word": false,
69
+ "special": false
70
+ },
71
+ "8": {
72
+ "content": "<|assistant|>",
73
+ "lstrip": false,
74
+ "normalized": false,
75
+ "rstrip": false,
76
+ "single_word": false,
77
+ "special": false
78
+ },
79
+ "9": {
80
+ "content": "<|user|>",
81
+ "lstrip": false,
82
+ "normalized": false,
83
+ "rstrip": false,
84
+ "single_word": false,
85
+ "special": false
86
+ },
87
+ "10": {
88
+ "content": "<|available_tools|>",
89
+ "lstrip": false,
90
+ "normalized": false,
91
+ "rstrip": false,
92
+ "single_word": false,
93
+ "special": false
94
+ },
95
+ "11": {
96
+ "content": "<|tool_calls|>",
97
+ "lstrip": false,
98
+ "normalized": false,
99
+ "rstrip": false,
100
+ "single_word": false,
101
+ "special": false
102
+ },
103
+ "12": {
104
+ "content": "<|tool_results|>",
105
+ "lstrip": false,
106
+ "normalized": false,
107
+ "rstrip": false,
108
+ "single_word": false,
109
+ "special": false
110
+ },
111
+ "13": {
112
+ "content": "<|code|>",
113
+ "lstrip": false,
114
+ "normalized": false,
115
+ "rstrip": false,
116
+ "single_word": false,
117
+ "special": false
118
+ },
119
+ "14": {
120
+ "content": "<|file|>",
121
+ "lstrip": false,
122
+ "normalized": false,
123
+ "rstrip": false,
124
+ "single_word": false,
125
+ "special": false
126
+ },
127
+ "102397": {
128
+ "content": "<|prefix|>",
129
+ "lstrip": false,
130
+ "normalized": false,
131
+ "rstrip": false,
132
+ "single_word": false,
133
+ "special": false
134
+ },
135
+ "102398": {
136
+ "content": "<|suffix|>",
137
+ "lstrip": false,
138
+ "normalized": false,
139
+ "rstrip": false,
140
+ "single_word": false,
141
+ "special": false
142
+ },
143
+ "102399": {
144
+ "content": "<|middle|>",
145
+ "lstrip": false,
146
+ "normalized": false,
147
+ "rstrip": false,
148
+ "single_word": false,
149
+ "special": false
150
+ }
151
+ },
152
+ "bos_token": "<s>",
153
+ "clean_up_tokenization_spaces": false,
154
+ "cls_token": "<cls>",
155
+ "do_lower_case": false,
156
+ "eos_token": "</s>",
157
+ "extra_ids": 0,
158
+ "extra_special_tokens": {},
159
+ "keep_accents": true,
160
+ "legacy": false,
161
+ "mask_token": "<mask>",
162
+ "model_max_length": 1000000000000000019884624838656,
163
+ "pad_token": "<pad>",
164
+ "padding_side": "left",
165
+ "sep_token": "<sep>",
166
+ "sp_model_kwargs": {},
167
+ "spaces_between_special_tokens": false,
168
+ "tokenizer_class": "LlamaTokenizer",
169
+ "unk_token": "<unk>",
170
+ "use_default_system_prompt": false
171
+ }
trainer_state.json ADDED
@@ -0,0 +1,916 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 49419,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010117566118294584,
14
+ "grad_norm": 1.3671875,
15
+ "learning_rate": 6.058276001618777e-05,
16
+ "loss": 11.5303,
17
+ "mean_token_accuracy": 0.095332110256124,
18
+ "num_tokens": 12016855.0,
19
+ "step": 500
20
+ },
21
+ {
22
+ "epoch": 0.020235132236589168,
23
+ "grad_norm": 1.1953125,
24
+ "learning_rate": 0.00012128692836908133,
25
+ "loss": 6.812,
26
+ "mean_token_accuracy": 0.21804794558882715,
27
+ "num_tokens": 24034987.0,
28
+ "step": 1000
29
+ },
30
+ {
31
+ "epoch": 0.030352698354883748,
32
+ "grad_norm": 0.80859375,
33
+ "learning_rate": 0.00018199109672197488,
34
+ "loss": 5.7925,
35
+ "mean_token_accuracy": 0.27103758984804155,
36
+ "num_tokens": 36048578.0,
37
+ "step": 1500
38
+ },
39
+ {
40
+ "epoch": 0.040470264473178336,
41
+ "grad_norm": 0.703125,
42
+ "learning_rate": 0.00024269526507486846,
43
+ "loss": 5.4354,
44
+ "mean_token_accuracy": 0.2889174829721451,
45
+ "num_tokens": 48062842.0,
46
+ "step": 2000
47
+ },
48
+ {
49
+ "epoch": 0.05058783059147291,
50
+ "grad_norm": 0.5546875,
51
+ "learning_rate": 0.0002999997367049147,
52
+ "loss": 5.2677,
53
+ "mean_token_accuracy": 0.29825611528754237,
54
+ "num_tokens": 60066067.0,
55
+ "step": 2500
56
+ },
57
+ {
58
+ "epoch": 0.060705396709767497,
59
+ "grad_norm": 0.421875,
60
+ "learning_rate": 0.000299906384128946,
61
+ "loss": 5.029,
62
+ "mean_token_accuracy": 0.307963959723711,
63
+ "num_tokens": 72080788.0,
64
+ "step": 3000
65
+ },
66
+ {
67
+ "epoch": 0.07082296282806208,
68
+ "grad_norm": 0.404296875,
69
+ "learning_rate": 0.00029964523417265877,
70
+ "loss": 4.8824,
71
+ "mean_token_accuracy": 0.3160543051958084,
72
+ "num_tokens": 84106466.0,
73
+ "step": 3500
74
+ },
75
+ {
76
+ "epoch": 0.08094052894635667,
77
+ "grad_norm": 0.4453125,
78
+ "learning_rate": 0.00029921657915368024,
79
+ "loss": 4.7702,
80
+ "mean_token_accuracy": 0.3228713305592537,
81
+ "num_tokens": 96125107.0,
82
+ "step": 4000
83
+ },
84
+ {
85
+ "epoch": 0.09105809506465125,
86
+ "grad_norm": 0.369140625,
87
+ "learning_rate": 0.0002986208988860602,
88
+ "loss": 4.6848,
89
+ "mean_token_accuracy": 0.32875489246845246,
90
+ "num_tokens": 108141921.0,
91
+ "step": 4500
92
+ },
93
+ {
94
+ "epoch": 0.10117566118294583,
95
+ "grad_norm": 0.396484375,
96
+ "learning_rate": 0.0002978588601431918,
97
+ "loss": 4.6171,
98
+ "mean_token_accuracy": 0.33351320880651475,
99
+ "num_tokens": 120150578.0,
100
+ "step": 5000
101
+ },
102
+ {
103
+ "epoch": 0.11129322730124042,
104
+ "grad_norm": 0.458984375,
105
+ "learning_rate": 0.0002969313159114605,
106
+ "loss": 4.5504,
107
+ "mean_token_accuracy": 0.3377925636172295,
108
+ "num_tokens": 132163036.0,
109
+ "step": 5500
110
+ },
111
+ {
112
+ "epoch": 0.12141079341953499,
113
+ "grad_norm": 0.345703125,
114
+ "learning_rate": 0.00029583930443545563,
115
+ "loss": 4.517,
116
+ "mean_token_accuracy": 0.34025603461265563,
117
+ "num_tokens": 144176271.0,
118
+ "step": 6000
119
+ },
120
+ {
121
+ "epoch": 0.13152835953782957,
122
+ "grad_norm": 0.3984375,
123
+ "learning_rate": 0.00029458404805581337,
124
+ "loss": 4.4534,
125
+ "mean_token_accuracy": 0.34516079604625705,
126
+ "num_tokens": 156188843.0,
127
+ "step": 6500
128
+ },
129
+ {
130
+ "epoch": 0.14164592565612416,
131
+ "grad_norm": 0.369140625,
132
+ "learning_rate": 0.00029316695184099267,
133
+ "loss": 4.4374,
134
+ "mean_token_accuracy": 0.34581841200590135,
135
+ "num_tokens": 168208403.0,
136
+ "step": 7000
137
+ },
138
+ {
139
+ "epoch": 0.15176349177441875,
140
+ "grad_norm": 0.392578125,
141
+ "learning_rate": 0.0002915896020145148,
142
+ "loss": 4.3998,
143
+ "mean_token_accuracy": 0.3480748535394669,
144
+ "num_tokens": 180235618.0,
145
+ "step": 7500
146
+ },
147
+ {
148
+ "epoch": 0.16188105789271334,
149
+ "grad_norm": 0.41796875,
150
+ "learning_rate": 0.00028985376417942787,
151
+ "loss": 4.3825,
152
+ "mean_token_accuracy": 0.34968917137384414,
153
+ "num_tokens": 192246640.0,
154
+ "step": 8000
155
+ },
156
+ {
157
+ "epoch": 0.1719986240110079,
158
+ "grad_norm": 0.345703125,
159
+ "learning_rate": 0.00028796138134198245,
160
+ "loss": 4.3497,
161
+ "mean_token_accuracy": 0.3521047782897949,
162
+ "num_tokens": 204257025.0,
163
+ "step": 8500
164
+ },
165
+ {
166
+ "epoch": 0.1821161901293025,
167
+ "grad_norm": 0.357421875,
168
+ "learning_rate": 0.00028591457173673235,
169
+ "loss": 4.3222,
170
+ "mean_token_accuracy": 0.3550558543205261,
171
+ "num_tokens": 216269157.0,
172
+ "step": 9000
173
+ },
174
+ {
175
+ "epoch": 0.1922337562475971,
176
+ "grad_norm": 0.408203125,
177
+ "learning_rate": 0.00028371562645549314,
178
+ "loss": 4.2894,
179
+ "mean_token_accuracy": 0.3578448301553726,
180
+ "num_tokens": 228272544.0,
181
+ "step": 9500
182
+ },
183
+ {
184
+ "epoch": 0.20235132236589165,
185
+ "grad_norm": 0.376953125,
186
+ "learning_rate": 0.0002813670068828134,
187
+ "loss": 4.2983,
188
+ "mean_token_accuracy": 0.355904599070549,
189
+ "num_tokens": 240295030.0,
190
+ "step": 10000
191
+ },
192
+ {
193
+ "epoch": 0.21246888848418624,
194
+ "grad_norm": 0.359375,
195
+ "learning_rate": 0.00027887134194082996,
196
+ "loss": 4.2844,
197
+ "mean_token_accuracy": 0.35704867881536484,
198
+ "num_tokens": 252314856.0,
199
+ "step": 10500
200
+ },
201
+ {
202
+ "epoch": 0.22258645460248083,
203
+ "grad_norm": 0.365234375,
204
+ "learning_rate": 0.0002762314251465891,
205
+ "loss": 4.2747,
206
+ "mean_token_accuracy": 0.35737198293209077,
207
+ "num_tokens": 264343569.0,
208
+ "step": 11000
209
+ },
210
+ {
211
+ "epoch": 0.23270402072077542,
212
+ "grad_norm": 0.390625,
213
+ "learning_rate": 0.0002734502114851296,
214
+ "loss": 4.2424,
215
+ "mean_token_accuracy": 0.36065280497074126,
216
+ "num_tokens": 276359037.0,
217
+ "step": 11500
218
+ },
219
+ {
220
+ "epoch": 0.24282158683906999,
221
+ "grad_norm": 0.396484375,
222
+ "learning_rate": 0.00027053081410182697,
223
+ "loss": 4.234,
224
+ "mean_token_accuracy": 0.36129101461172103,
225
+ "num_tokens": 288373491.0,
226
+ "step": 12000
227
+ },
228
+ {
229
+ "epoch": 0.2529391529573646,
230
+ "grad_norm": 0.365234375,
231
+ "learning_rate": 0.0002674765008177008,
232
+ "loss": 4.2163,
233
+ "mean_token_accuracy": 0.36271012741327285,
234
+ "num_tokens": 300383941.0,
235
+ "step": 12500
236
+ },
237
+ {
238
+ "epoch": 0.26305671907565914,
239
+ "grad_norm": 0.40234375,
240
+ "learning_rate": 0.00026429069047158657,
241
+ "loss": 4.2082,
242
+ "mean_token_accuracy": 0.36296020871400836,
243
+ "num_tokens": 312416834.0,
244
+ "step": 13000
245
+ },
246
+ {
247
+ "epoch": 0.27317428519395376,
248
+ "grad_norm": 0.333984375,
249
+ "learning_rate": 0.000260976949093266,
250
+ "loss": 4.1802,
251
+ "mean_token_accuracy": 0.36644324856996535,
252
+ "num_tokens": 324430305.0,
253
+ "step": 13500
254
+ },
255
+ {
256
+ "epoch": 0.2832918513122483,
257
+ "grad_norm": 0.345703125,
258
+ "learning_rate": 0.0002575389859118394,
259
+ "loss": 4.1717,
260
+ "mean_token_accuracy": 0.36681148010492326,
261
+ "num_tokens": 336432753.0,
262
+ "step": 14000
263
+ },
264
+ {
265
+ "epoch": 0.2934094174305429,
266
+ "grad_norm": 0.361328125,
267
+ "learning_rate": 0.00025398064920380834,
268
+ "loss": 4.1852,
269
+ "mean_token_accuracy": 0.36508618742227555,
270
+ "num_tokens": 348448508.0,
271
+ "step": 14500
272
+ },
273
+ {
274
+ "epoch": 0.3035269835488375,
275
+ "grad_norm": 0.37890625,
276
+ "learning_rate": 0.00025030592198551575,
277
+ "loss": 4.1387,
278
+ "mean_token_accuracy": 0.37016365003585816,
279
+ "num_tokens": 360455888.0,
280
+ "step": 15000
281
+ },
282
+ {
283
+ "epoch": 0.31364454966713207,
284
+ "grad_norm": 0.314453125,
285
+ "learning_rate": 0.0002465189175547654,
286
+ "loss": 4.147,
287
+ "mean_token_accuracy": 0.36943590980768204,
288
+ "num_tokens": 372469047.0,
289
+ "step": 15500
290
+ },
291
+ {
292
+ "epoch": 0.3237621157854267,
293
+ "grad_norm": 0.4453125,
294
+ "learning_rate": 0.00024262387488661118,
295
+ "loss": 4.2161,
296
+ "mean_token_accuracy": 0.3663038157224655,
297
+ "num_tokens": 384482227.0,
298
+ "step": 16000
299
+ },
300
+ {
301
+ "epoch": 0.33387968190372125,
302
+ "grad_norm": 0.431640625,
303
+ "learning_rate": 0.0002386251538884696,
304
+ "loss": 4.2708,
305
+ "mean_token_accuracy": 0.35985021191835403,
306
+ "num_tokens": 396507994.0,
307
+ "step": 16500
308
+ },
309
+ {
310
+ "epoch": 0.3439972480220158,
311
+ "grad_norm": 0.3984375,
312
+ "learning_rate": 0.0002345272305198671,
313
+ "loss": 4.2626,
314
+ "mean_token_accuracy": 0.35938486641645434,
315
+ "num_tokens": 408527074.0,
316
+ "step": 17000
317
+ },
318
+ {
319
+ "epoch": 0.35411481414031043,
320
+ "grad_norm": 0.4375,
321
+ "learning_rate": 0.00023033469178228457,
322
+ "loss": 4.2604,
323
+ "mean_token_accuracy": 0.35944406408071516,
324
+ "num_tokens": 420551611.0,
325
+ "step": 17500
326
+ },
327
+ {
328
+ "epoch": 0.364232380258605,
329
+ "grad_norm": 0.4296875,
330
+ "learning_rate": 0.0002260522305847074,
331
+ "loss": 4.259,
332
+ "mean_token_accuracy": 0.3585598506331444,
333
+ "num_tokens": 432578604.0,
334
+ "step": 18000
335
+ },
336
+ {
337
+ "epoch": 0.37434994637689956,
338
+ "grad_norm": 0.458984375,
339
+ "learning_rate": 0.00022168464049062824,
340
+ "loss": 4.2539,
341
+ "mean_token_accuracy": 0.359400195479393,
342
+ "num_tokens": 444606807.0,
343
+ "step": 18500
344
+ },
345
+ {
346
+ "epoch": 0.3844675124951942,
347
+ "grad_norm": 0.46484375,
348
+ "learning_rate": 0.0002172368103523822,
349
+ "loss": 4.2396,
350
+ "mean_token_accuracy": 0.3610253592133522,
351
+ "num_tokens": 456622578.0,
352
+ "step": 19000
353
+ },
354
+ {
355
+ "epoch": 0.39458507861348874,
356
+ "grad_norm": 0.4921875,
357
+ "learning_rate": 0.0002127137188388207,
358
+ "loss": 4.2317,
359
+ "mean_token_accuracy": 0.3614521362781525,
360
+ "num_tokens": 468626335.0,
361
+ "step": 19500
362
+ },
363
+ {
364
+ "epoch": 0.4047026447317833,
365
+ "grad_norm": 0.431640625,
366
+ "learning_rate": 0.0002081204288624496,
367
+ "loss": 4.1995,
368
+ "mean_token_accuracy": 0.3652240701317787,
369
+ "num_tokens": 480636609.0,
370
+ "step": 20000
371
+ },
372
+ {
373
+ "epoch": 0.4148202108500779,
374
+ "grad_norm": 0.4375,
375
+ "learning_rate": 0.00020346208191226927,
376
+ "loss": 4.1966,
377
+ "mean_token_accuracy": 0.36518134355545045,
378
+ "num_tokens": 492648597.0,
379
+ "step": 20500
380
+ },
381
+ {
382
+ "epoch": 0.4249377769683725,
383
+ "grad_norm": 0.48046875,
384
+ "learning_rate": 0.0001987438922986602,
385
+ "loss": 4.1926,
386
+ "mean_token_accuracy": 0.365411285161972,
387
+ "num_tokens": 504658765.0,
388
+ "step": 21000
389
+ },
390
+ {
391
+ "epoch": 0.43505534308666705,
392
+ "grad_norm": 0.5625,
393
+ "learning_rate": 0.0001939711413167559,
394
+ "loss": 4.1985,
395
+ "mean_token_accuracy": 0.36408871501684187,
396
+ "num_tokens": 516665859.0,
397
+ "step": 21500
398
+ },
399
+ {
400
+ "epoch": 0.44517290920496166,
401
+ "grad_norm": 0.404296875,
402
+ "learning_rate": 0.0001891491713348375,
403
+ "loss": 4.2043,
404
+ "mean_token_accuracy": 0.3634589037895203,
405
+ "num_tokens": 528675596.0,
406
+ "step": 22000
407
+ },
408
+ {
409
+ "epoch": 0.4552904753232562,
410
+ "grad_norm": 0.423828125,
411
+ "learning_rate": 0.0001842833798143648,
412
+ "loss": 4.1872,
413
+ "mean_token_accuracy": 0.36537654942274095,
414
+ "num_tokens": 540694481.0,
415
+ "step": 22500
416
+ },
417
+ {
418
+ "epoch": 0.46540804144155085,
419
+ "grad_norm": 0.439453125,
420
+ "learning_rate": 0.00017937921326834042,
421
+ "loss": 4.1642,
422
+ "mean_token_accuracy": 0.36855128794908526,
423
+ "num_tokens": 552703293.0,
424
+ "step": 23000
425
+ },
426
+ {
427
+ "epoch": 0.4755256075598454,
428
+ "grad_norm": 0.455078125,
429
+ "learning_rate": 0.0001744421611647669,
430
+ "loss": 4.1161,
431
+ "mean_token_accuracy": 0.37240500724315645,
432
+ "num_tokens": 564716126.0,
433
+ "step": 23500
434
+ },
435
+ {
436
+ "epoch": 0.48564317367813997,
437
+ "grad_norm": 0.451171875,
438
+ "learning_rate": 0.00016947774978202324,
439
+ "loss": 4.114,
440
+ "mean_token_accuracy": 0.3716895794272423,
441
+ "num_tokens": 576739422.0,
442
+ "step": 24000
443
+ },
444
+ {
445
+ "epoch": 0.4957607397964346,
446
+ "grad_norm": 0.46875,
447
+ "learning_rate": 0.00016449153602303716,
448
+ "loss": 4.127,
449
+ "mean_token_accuracy": 0.37084434121847154,
450
+ "num_tokens": 588758561.0,
451
+ "step": 24500
452
+ },
453
+ {
454
+ "epoch": 0.5058783059147292,
455
+ "grad_norm": 0.455078125,
456
+ "learning_rate": 0.00015948910119517801,
457
+ "loss": 4.1131,
458
+ "mean_token_accuracy": 0.3723976674079895,
459
+ "num_tokens": 600780001.0,
460
+ "step": 25000
461
+ },
462
+ {
463
+ "epoch": 0.5159958720330238,
464
+ "grad_norm": 0.44140625,
465
+ "learning_rate": 0.00015447604476283297,
466
+ "loss": 4.1103,
467
+ "mean_token_accuracy": 0.37234303742647173,
468
+ "num_tokens": 612790687.0,
469
+ "step": 25500
470
+ },
471
+ {
472
+ "epoch": 0.5261134381513183,
473
+ "grad_norm": 0.439453125,
474
+ "learning_rate": 0.00014945797807965866,
475
+ "loss": 4.0859,
476
+ "mean_token_accuracy": 0.3740212817192078,
477
+ "num_tokens": 624818330.0,
478
+ "step": 26000
479
+ },
480
+ {
481
+ "epoch": 0.5362310042696129,
482
+ "grad_norm": 0.447265625,
483
+ "learning_rate": 0.00014444051810752503,
484
+ "loss": 4.0808,
485
+ "mean_token_accuracy": 0.3752990872859955,
486
+ "num_tokens": 636836789.0,
487
+ "step": 26500
488
+ },
489
+ {
490
+ "epoch": 0.5463485703879075,
491
+ "grad_norm": 0.455078125,
492
+ "learning_rate": 0.00013942928112918113,
493
+ "loss": 4.0793,
494
+ "mean_token_accuracy": 0.37516462814807894,
495
+ "num_tokens": 648846886.0,
496
+ "step": 27000
497
+ },
498
+ {
499
+ "epoch": 0.556466136506202,
500
+ "grad_norm": 0.44140625,
501
+ "learning_rate": 0.0001344298764616819,
502
+ "loss": 4.0674,
503
+ "mean_token_accuracy": 0.37648630076646805,
504
+ "num_tokens": 660859222.0,
505
+ "step": 27500
506
+ },
507
+ {
508
+ "epoch": 0.5665837026244966,
509
+ "grad_norm": 0.455078125,
510
+ "learning_rate": 0.00012944790017761106,
511
+ "loss": 4.0677,
512
+ "mean_token_accuracy": 0.37691014271974566,
513
+ "num_tokens": 672873174.0,
514
+ "step": 28000
515
+ },
516
+ {
517
+ "epoch": 0.5767012687427913,
518
+ "grad_norm": 0.46484375,
519
+ "learning_rate": 0.00012448892884112992,
520
+ "loss": 4.041,
521
+ "mean_token_accuracy": 0.3792866112589836,
522
+ "num_tokens": 684869774.0,
523
+ "step": 28500
524
+ },
525
+ {
526
+ "epoch": 0.5868188348610858,
527
+ "grad_norm": 2.34375,
528
+ "learning_rate": 0.00011955851326586234,
529
+ "loss": 4.0607,
530
+ "mean_token_accuracy": 0.3807999837398529,
531
+ "num_tokens": 696881301.0,
532
+ "step": 29000
533
+ },
534
+ {
535
+ "epoch": 0.5969364009793804,
536
+ "grad_norm": 0.55078125,
537
+ "learning_rate": 0.00011466217230160404,
538
+ "loss": 4.0344,
539
+ "mean_token_accuracy": 0.3827848986387253,
540
+ "num_tokens": 708895273.0,
541
+ "step": 29500
542
+ },
543
+ {
544
+ "epoch": 0.607053967097675,
545
+ "grad_norm": 0.53125,
546
+ "learning_rate": 0.00010980538665680985,
547
+ "loss": 4.0099,
548
+ "mean_token_accuracy": 0.38436073118448255,
549
+ "num_tokens": 720912243.0,
550
+ "step": 30000
551
+ },
552
+ {
553
+ "epoch": 0.6171715332159695,
554
+ "grad_norm": 0.54296875,
555
+ "learning_rate": 0.00010499359276377534,
556
+ "loss": 3.9881,
557
+ "mean_token_accuracy": 0.386309523165226,
558
+ "num_tokens": 732929445.0,
559
+ "step": 30500
560
+ },
561
+ {
562
+ "epoch": 0.6272890993342641,
563
+ "grad_norm": 0.5390625,
564
+ "learning_rate": 0.00010023217669337769,
565
+ "loss": 3.9851,
566
+ "mean_token_accuracy": 0.386250978410244,
567
+ "num_tokens": 744956030.0,
568
+ "step": 31000
569
+ },
570
+ {
571
+ "epoch": 0.6374066654525588,
572
+ "grad_norm": 0.55859375,
573
+ "learning_rate": 9.552646812618973e-05,
574
+ "loss": 3.9912,
575
+ "mean_token_accuracy": 0.38541185545921325,
576
+ "num_tokens": 756973559.0,
577
+ "step": 31500
578
+ },
579
+ {
580
+ "epoch": 0.6475242315708534,
581
+ "grad_norm": 0.5703125,
582
+ "learning_rate": 9.088173438671332e-05,
583
+ "loss": 3.9618,
584
+ "mean_token_accuracy": 0.3886643843054771,
585
+ "num_tokens": 768993412.0,
586
+ "step": 32000
587
+ },
588
+ {
589
+ "epoch": 0.6576417976891479,
590
+ "grad_norm": 0.5625,
591
+ "learning_rate": 8.630317454741159e-05,
592
+ "loss": 3.9566,
593
+ "mean_token_accuracy": 0.38894479793310166,
594
+ "num_tokens": 781002910.0,
595
+ "step": 32500
596
+ },
597
+ {
598
+ "epoch": 0.6677593638074425,
599
+ "grad_norm": 0.5390625,
600
+ "learning_rate": 8.17959136091384e-05,
601
+ "loss": 3.9384,
602
+ "mean_token_accuracy": 0.3909342797398567,
603
+ "num_tokens": 793011499.0,
604
+ "step": 33000
605
+ },
606
+ {
607
+ "epoch": 0.6778769299257371,
608
+ "grad_norm": 0.5625,
609
+ "learning_rate": 7.736499676448012e-05,
610
+ "loss": 3.9304,
611
+ "mean_token_accuracy": 0.3920640903711319,
612
+ "num_tokens": 805027429.0,
613
+ "step": 33500
614
+ },
615
+ {
616
+ "epoch": 0.6879944960440316,
617
+ "grad_norm": 0.58203125,
618
+ "learning_rate": 7.301538375043028e-05,
619
+ "loss": 3.927,
620
+ "mean_token_accuracy": 0.39186790603399274,
621
+ "num_tokens": 817043838.0,
622
+ "step": 34000
623
+ },
624
+ {
625
+ "epoch": 0.6981120621623262,
626
+ "grad_norm": 0.52734375,
627
+ "learning_rate": 6.875194329671924e-05,
628
+ "loss": 3.9135,
629
+ "mean_token_accuracy": 0.39424380254745484,
630
+ "num_tokens": 829051457.0,
631
+ "step": 34500
632
+ },
633
+ {
634
+ "epoch": 0.7082296282806209,
635
+ "grad_norm": 0.59375,
636
+ "learning_rate": 6.457944767601184e-05,
637
+ "loss": 3.9194,
638
+ "mean_token_accuracy": 0.3936538654565811,
639
+ "num_tokens": 841063558.0,
640
+ "step": 35000
641
+ },
642
+ {
643
+ "epoch": 0.7183471943989154,
644
+ "grad_norm": 0.478515625,
645
+ "learning_rate": 6.0502567362074895e-05,
646
+ "loss": 3.9167,
647
+ "mean_token_accuracy": 0.39285322940349576,
648
+ "num_tokens": 853081842.0,
649
+ "step": 35500
650
+ },
651
+ {
652
+ "epoch": 0.72846476051721,
653
+ "grad_norm": 0.59765625,
654
+ "learning_rate": 5.6525865801892425e-05,
655
+ "loss": 3.8827,
656
+ "mean_token_accuracy": 0.3979086767435074,
657
+ "num_tokens": 865088195.0,
658
+ "step": 36000
659
+ },
660
+ {
661
+ "epoch": 0.7385823266355046,
662
+ "grad_norm": 0.55078125,
663
+ "learning_rate": 5.265379430758167e-05,
664
+ "loss": 3.8683,
665
+ "mean_token_accuracy": 0.3997677757143974,
666
+ "num_tokens": 877099653.0,
667
+ "step": 36500
668
+ },
669
+ {
670
+ "epoch": 0.7486998927537991,
671
+ "grad_norm": 0.51953125,
672
+ "learning_rate": 4.889068707382631e-05,
673
+ "loss": 3.8626,
674
+ "mean_token_accuracy": 0.4014857310652733,
675
+ "num_tokens": 889098774.0,
676
+ "step": 37000
677
+ },
678
+ {
679
+ "epoch": 0.7588174588720937,
680
+ "grad_norm": 0.5703125,
681
+ "learning_rate": 4.524075632640548e-05,
682
+ "loss": 3.8701,
683
+ "mean_token_accuracy": 0.3988613831400871,
684
+ "num_tokens": 901122305.0,
685
+ "step": 37500
686
+ },
687
+ {
688
+ "epoch": 0.7689350249903883,
689
+ "grad_norm": 0.58984375,
690
+ "learning_rate": 4.1708087607247995e-05,
691
+ "loss": 3.869,
692
+ "mean_token_accuracy": 0.3993934275507927,
693
+ "num_tokens": 913141836.0,
694
+ "step": 38000
695
+ },
696
+ {
697
+ "epoch": 0.7790525911086829,
698
+ "grad_norm": 0.5703125,
699
+ "learning_rate": 3.829663520129017e-05,
700
+ "loss": 3.8759,
701
+ "mean_token_accuracy": 0.3983539624810219,
702
+ "num_tokens": 925159236.0,
703
+ "step": 38500
704
+ },
705
+ {
706
+ "epoch": 0.7891701572269775,
707
+ "grad_norm": 0.54296875,
708
+ "learning_rate": 3.5010217710255475e-05,
709
+ "loss": 3.8641,
710
+ "mean_token_accuracy": 0.3992738093137741,
711
+ "num_tokens": 937190539.0,
712
+ "step": 39000
713
+ },
714
+ {
715
+ "epoch": 0.7992877233452721,
716
+ "grad_norm": 0.5703125,
717
+ "learning_rate": 3.185251377831133e-05,
718
+ "loss": 3.8512,
719
+ "mean_token_accuracy": 0.4009958416223526,
720
+ "num_tokens": 949212097.0,
721
+ "step": 39500
722
+ },
723
+ {
724
+ "epoch": 0.8094052894635666,
725
+ "grad_norm": 0.55078125,
726
+ "learning_rate": 2.8827057974386658e-05,
727
+ "loss": 3.8534,
728
+ "mean_token_accuracy": 0.40131270414590836,
729
+ "num_tokens": 961210545.0,
730
+ "step": 40000
731
+ },
732
+ {
733
+ "epoch": 0.8195228555818612,
734
+ "grad_norm": 0.53515625,
735
+ "learning_rate": 2.5937236835760272e-05,
736
+ "loss": 3.8699,
737
+ "mean_token_accuracy": 0.3994129166603088,
738
+ "num_tokens": 973234104.0,
739
+ "step": 40500
740
+ },
741
+ {
742
+ "epoch": 0.8296404217001558,
743
+ "grad_norm": 0.55859375,
744
+ "learning_rate": 2.318628507734748e-05,
745
+ "loss": 3.8392,
746
+ "mean_token_accuracy": 0.40251973223686216,
747
+ "num_tokens": 985244716.0,
748
+ "step": 41000
749
+ },
750
+ {
751
+ "epoch": 0.8397579878184503,
752
+ "grad_norm": 0.54296875,
753
+ "learning_rate": 2.0577281970929332e-05,
754
+ "loss": 3.8332,
755
+ "mean_token_accuracy": 0.40371298098564146,
756
+ "num_tokens": 997252495.0,
757
+ "step": 41500
758
+ },
759
+ {
760
+ "epoch": 0.849875553936745,
761
+ "grad_norm": 0.53125,
762
+ "learning_rate": 1.8113147898376435e-05,
763
+ "loss": 3.8686,
764
+ "mean_token_accuracy": 0.39953106695413587,
765
+ "num_tokens": 1009277702.0,
766
+ "step": 42000
767
+ },
768
+ {
769
+ "epoch": 0.8599931200550396,
770
+ "grad_norm": 0.53125,
771
+ "learning_rate": 1.5796641082725953e-05,
772
+ "loss": 3.846,
773
+ "mean_token_accuracy": 0.4027084149122238,
774
+ "num_tokens": 1021271039.0,
775
+ "step": 42500
776
+ },
777
+ {
778
+ "epoch": 0.8701106861733341,
779
+ "grad_norm": 0.53125,
780
+ "learning_rate": 1.3630354500770884e-05,
781
+ "loss": 3.8402,
782
+ "mean_token_accuracy": 0.4029158518910408,
783
+ "num_tokens": 1033279547.0,
784
+ "step": 43000
785
+ },
786
+ {
787
+ "epoch": 0.8802282522916287,
788
+ "grad_norm": 0.5546875,
789
+ "learning_rate": 1.1616712980616954e-05,
790
+ "loss": 3.8334,
791
+ "mean_token_accuracy": 0.4026625081896782,
792
+ "num_tokens": 1045302954.0,
793
+ "step": 43500
794
+ },
795
+ {
796
+ "epoch": 0.8903458184099233,
797
+ "grad_norm": 0.56640625,
798
+ "learning_rate": 9.757970487456834e-06,
799
+ "loss": 3.846,
800
+ "mean_token_accuracy": 0.4019183788895607,
801
+ "num_tokens": 1057312737.0,
802
+ "step": 44000
803
+ },
804
+ {
805
+ "epoch": 0.900463384528218,
806
+ "grad_norm": 0.53125,
807
+ "learning_rate": 8.056207600599047e-06,
808
+ "loss": 3.8371,
809
+ "mean_token_accuracy": 0.40209768468141555,
810
+ "num_tokens": 1069335133.0,
811
+ "step": 44500
812
+ },
813
+ {
814
+ "epoch": 0.9105809506465125,
815
+ "grad_norm": 0.5546875,
816
+ "learning_rate": 6.513329184576232e-06,
817
+ "loss": 3.8209,
818
+ "mean_token_accuracy": 0.40574184638261795,
819
+ "num_tokens": 1081330922.0,
820
+ "step": 45000
821
+ },
822
+ {
823
+ "epoch": 0.9206985167648071,
824
+ "grad_norm": 0.55859375,
825
+ "learning_rate": 5.131062256939128e-06,
826
+ "loss": 3.847,
827
+ "mean_token_accuracy": 0.40258895909786224,
828
+ "num_tokens": 1093337188.0,
829
+ "step": 45500
830
+ },
831
+ {
832
+ "epoch": 0.9308160828831017,
833
+ "grad_norm": 0.5234375,
834
+ "learning_rate": 3.910954055123322e-06,
835
+ "loss": 3.856,
836
+ "mean_token_accuracy": 0.4003161287903786,
837
+ "num_tokens": 1105361310.0,
838
+ "step": 46000
839
+ },
840
+ {
841
+ "epoch": 0.9409336490013962,
842
+ "grad_norm": 0.58984375,
843
+ "learning_rate": 2.8543703045524167e-06,
844
+ "loss": 3.8275,
845
+ "mean_token_accuracy": 0.40424364000558854,
846
+ "num_tokens": 1117367893.0,
847
+ "step": 46500
848
+ },
849
+ {
850
+ "epoch": 0.9510512151196908,
851
+ "grad_norm": 0.59375,
852
+ "learning_rate": 1.9624936899163945e-06,
853
+ "loss": 3.8295,
854
+ "mean_token_accuracy": 0.4040605016946793,
855
+ "num_tokens": 1129375583.0,
856
+ "step": 47000
857
+ },
858
+ {
859
+ "epoch": 0.9611687812379854,
860
+ "grad_norm": 0.54296875,
861
+ "learning_rate": 1.2363225313359758e-06,
862
+ "loss": 3.8356,
863
+ "mean_token_accuracy": 0.40258765465021135,
864
+ "num_tokens": 1141387823.0,
865
+ "step": 47500
866
+ },
867
+ {
868
+ "epoch": 0.9712863473562799,
869
+ "grad_norm": 0.56640625,
870
+ "learning_rate": 6.766696668952853e-07,
871
+ "loss": 3.855,
872
+ "mean_token_accuracy": 0.4004232720732689,
873
+ "num_tokens": 1153399040.0,
874
+ "step": 48000
875
+ },
876
+ {
877
+ "epoch": 0.9814039134745746,
878
+ "grad_norm": 0.55078125,
879
+ "learning_rate": 2.8416154279330417e-07,
880
+ "loss": 3.8378,
881
+ "mean_token_accuracy": 0.40229264545440674,
882
+ "num_tokens": 1165414940.0,
883
+ "step": 48500
884
+ },
885
+ {
886
+ "epoch": 0.9915214795928692,
887
+ "grad_norm": 0.57421875,
888
+ "learning_rate": 5.923751213269823e-08,
889
+ "loss": 3.8603,
890
+ "mean_token_accuracy": 0.4007369539737701,
891
+ "num_tokens": 1177449058.0,
892
+ "step": 49000
893
+ }
894
+ ],
895
+ "logging_steps": 500,
896
+ "max_steps": 49419,
897
+ "num_input_tokens_seen": 0,
898
+ "num_train_epochs": 1,
899
+ "save_steps": 5000,
900
+ "stateful_callbacks": {
901
+ "TrainerControl": {
902
+ "args": {
903
+ "should_epoch_stop": false,
904
+ "should_evaluate": false,
905
+ "should_log": false,
906
+ "should_save": true,
907
+ "should_training_stop": true
908
+ },
909
+ "attributes": {}
910
+ }
911
+ },
912
+ "total_flos": 2.4876879400717517e+18,
913
+ "train_batch_size": 48,
914
+ "trial_name": null,
915
+ "trial_params": null
916
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13ae268dac2d87d4215d24239b45ef2e07bc07264fed3802401b6c060b1314b3
3
+ size 6225