Add Olmo-3-7B thinking-SFT checkpoint
Browse files- chat_template.jinja +4 -0
- config.json +69 -0
- generation_config.json +4 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +13 -0
chat_template.jinja
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- for message in messages -%}{{- '<|im_start|>' + message['role'] + '
|
| 2 |
+
' + message['content'] + '<|im_end|>' + '
|
| 3 |
+
' -}}{%- endfor -%}{%- if add_generation_prompt -%}{{- '<|im_start|>assistant
|
| 4 |
+
' -}}{%- endif -%}
|
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": "bfloat16",
|
| 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": 100277,
|
| 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": "5.7.0",
|
| 67 |
+
"use_cache": false,
|
| 68 |
+
"vocab_size": 100278
|
| 69 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "5.7.0"
|
| 4 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7369f6757f88f5fb8d0da9b4578434aab56204f51d22d861614c9c74b7c14512
|
| 3 |
+
size 14596063960
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 65536,
|
| 10 |
+
"pad_token": "<|pad|>",
|
| 11 |
+
"tokenizer_class": "TokenizersBackend",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|