asparius commited on
Commit
a35dcc2
·
verified ·
1 Parent(s): 0a479c3

Upload folder using huggingface_hub

Browse files
chat_template.jinja ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {#- OLMo 3 Think SFT chat template (ChatML format).
2
+ Based on allenai/Olmo-3-32B-Think-SFT with {%- generation %} tags added
3
+ for completion_only_loss training support. -#}
4
+
5
+ {%- set has_system = messages | selectattr('role', 'equalto', 'system') | list | length > 0 -%}
6
+
7
+ {%- if not has_system -%}
8
+ {{- '<|im_start|>system\nYou are a helpful AI assistant.<|im_end|>\n' -}}
9
+ {%- endif -%}
10
+
11
+ {%- for message in messages -%}
12
+ {%- if message['role'] == 'system' -%}
13
+ {{- '<|im_start|>system\n' + message['content'] -}}
14
+ {%- if message.get('functions', none) is not none -%}
15
+ {{- ' <functions>' + message['functions'] + '</functions>' -}}
16
+ {%- endif -%}
17
+ {{- '<|im_end|>\n' -}}
18
+
19
+ {%- elif message['role'] == 'user' -%}
20
+ {{- '<|im_start|>user\n' + message['content'] -}}
21
+ {%- if message.get('functions', none) is not none -%}
22
+ {{- '\n<functions>' + message['functions'] + '</functions>' -}}
23
+ {%- endif -%}
24
+ {{- '<|im_end|>\n' -}}
25
+
26
+ {%- elif message['role'] == 'assistant' -%}
27
+ {{- '<|im_start|>assistant\n' -}}
28
+ {%- generation -%}
29
+ {%- if message.get('content', none) is not none -%}
30
+ {{- message['content'] -}}
31
+ {%- endif -%}
32
+ {%- if message.get('function_calls', none) is not none -%}
33
+ {{- '<function_calls>' + message['function_calls'] + '</function_calls>' -}}
34
+ {%- endif -%}
35
+ {%- if not loop.last -%}
36
+ {{- '<|im_end|>\n' -}}
37
+ {%- else -%}
38
+ {{- eos_token -}}
39
+ {%- endif -%}
40
+ {%- endgeneration -%}
41
+
42
+ {%- elif message['role'] == 'environment' -%}
43
+ {{- '<|im_start|>environment\n' + message['content'] + '<|im_end|>\n' -}}
44
+
45
+ {%- endif -%}
46
+
47
+ {%- if loop.last and add_generation_prompt -%}
48
+ {{- '<|im_start|>assistant\n<think>' -}}
49
+ {%- endif -%}
50
+ {%- endfor -%}
config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Olmo3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "dtype": "float32",
9
+ "eos_token_id": 100257,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "layer_types": [
15
+ "sliding_attention",
16
+ "sliding_attention",
17
+ "sliding_attention",
18
+ "full_attention",
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "full_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "full_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "full_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "full_attention"
47
+ ],
48
+ "max_position_embeddings": 65536,
49
+ "model_type": "olmo3",
50
+ "num_attention_heads": 32,
51
+ "num_hidden_layers": 32,
52
+ "num_key_value_heads": 32,
53
+ "pad_token_id": 100257,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_parameters": {
56
+ "attention_factor": 1.2079441541679836,
57
+ "beta_fast": 32,
58
+ "beta_slow": 1,
59
+ "factor": 8.0,
60
+ "original_max_position_embeddings": 8192,
61
+ "rope_theta": 500000,
62
+ "rope_type": "yarn"
63
+ },
64
+ "sliding_window": 4096,
65
+ "tie_word_embeddings": false,
66
+ "transformers_version": "5.3.0.dev0",
67
+ "use_cache": false,
68
+ "vocab_size": 100278
69
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 100257
5
+ ],
6
+ "pad_token_id": 100257,
7
+ "transformers_version": "5.3.0.dev0"
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac156948e8f5fe3678957c5f903a06277adfbac9f986241cbcedf8433f04d9f3
3
+ size 29192086152
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|endoftext|>",
7
+ "is_local": false,
8
+ "max_length": 8192,
9
+ "model_max_length": 65536,
10
+ "pad_token": "<|endoftext|>",
11
+ "stride": 0,
12
+ "tokenizer_class": "TokenizersBackend",
13
+ "truncation_side": "right",
14
+ "truncation_strategy": "longest_first",
15
+ "unk_token": "<|endoftext|>"
16
+ }