Add files using upload-large-folder tool

README.md

@@ -0,0 +1,49 @@
+---
+library_name: transformers
+pipeline_tag: text-generation
+tags:
+- olmo3
+- baseline
+- safetensors
+- sliding-window-attention
+---
+
+# OLMo 3 1B Baseline — Stage 4 Instruct SFT
+
+This repository contains the pure OLMo 3 1B baseline checkpoint from
+`o3b1b-instruct-sft-dolci-s32768-g32-m1-tp1-cp8-dp32-hsdp32-b2-lr3e5-min0-wd5e2-wu10pct-2ep-256npu-share-20260802-v1` at iteration `3252`. SiameseNorm and Depth-Attention
+are disabled.
+
+- Training sequence length: 32,768
+- Model context capacity: 65,536
+- Sliding-window size: 4,096
+- Attention pattern: `[SWA, SWA, SWA, Full]`
+- Vocabulary: 100,278 real tokens; 74 Megatron padding rows removed
+
+Stage 3/4 apply YaRN only to Full-Attention layers. OLMo 3 SWA layers use the
+original RoPE and retain their 4,096-token local window.
+
+## Loading
+
+`transformers>=4.57.6,<5` is required.
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+repo_id = "ArchSpace-Collection/OLMo3-1B-SiameseNorm-DepthAttention--baseline-stage4-instruct"
+tokenizer = AutoTokenizer.from_pretrained(
+    repo_id,
+    use_fast=True,
+    fix_mistral_regex=False,
+)
+model = AutoModelForCausalLM.from_pretrained(
+    repo_id,
+    dtype=torch.bfloat16,
+    attn_implementation="sdpa",
+)
+```
+
+`fix_mistral_regex=False` preserves the tokenizer behavior used for training.
+The checkpoint uses the official Transformers `Olmo3ForCausalLM`
+implementation and does not require remote code.

SHA256SUMS

@@ -0,0 +1,12 @@
+ed7d006ae5f389ce37fab28f2b63e3cdcd694666325cb6362ffda948323a87cb  README.md
+a3d6a7ad3fde8e26fa9953d80ac30682f7d56ac1d03f3428910a4239a4e23bef  chat_template.jinja
+0dbc55ce09a6ce495d6889efa2d31ef01cd14a1f4b70262fe4612b94ec57f99c  config.json
+7c74827ba88128607f0e857f9531eca193cb5ac57c4e635973392ae5ffbe22ae  conversion_manifest.json
+dbd4aa5dc6f7f1bdc0c181d9e5582f6571e1d2a22878dfcd98149324b9f5c4a2  generation_config.json
+8ce269f5615d3813862b83683e6dc96d26f263d0c5888ffcb45f3ebaf84822a5  hf_validation_report.json
+b6fe424e334903f7fb84d3a106d9730455f4744b9fe3c21ee136d97a00e72502  merges.txt
+197cee5740ba915cce8a470c25b5680c43c5b03bde6c0a310809858d2ec26244  model.safetensors
+78afb564e81264029b25f9caf24bda2521d5bdaeff5cd3fdbc01d3da2e8ce2f2  special_tokens_map.json
+7738a25c46a6043a0e2e605138851baa450269ce9d800af6ca9bf6192e45510c  tokenizer.json
+5f959b9c0cff767a63f22a0569e0554cce28214fb49e41479ba326f44169c5ec  tokenizer_config.json
+d8eb44d0d51f5e39c9b14825ca8bd10a2500f99c438ca81d4462f0263acd3b53  vocab.json

chat_template.jinja

@@ -0,0 +1,16 @@
+{%- set has_system = messages|selectattr('role', 'equalto', 'system')|list|length > 0 -%}{%- if not has_system -%}{{- '<|im_start|>system
+You are a helpful function-calling AI assistant. ' -}}{%- if tools is none or (tools | length) == 0 -%}{{- 'You do not currently have access to any functions. <functions></functions><|im_end|>
+' -}}{%- else -%}{{- 'You are provided with function signatures within <functions></functions> XML tags. You may call one or more functions to assist with the user query. Output any function calls within <function_calls></function_calls> XML tags. Do not make assumptions about what values to plug into functions.' -}}{{- '<functions>' -}}{{- tools | tojson -}}{{- '</functions><|im_end|>
+' -}}{%- endif -%}{%- endif -%}{%- for message in messages -%}{%- if message['role'] == 'system' -%}{{- '<|im_start|>system
+' + message['content'] -}}{%- if tools is not none -%}{{- '<functions>' -}}{{- tools | tojson -}}{{- '</functions>' -}}{%- elif message.get('functions', none) is not none -%}{{- ' <functions>' + message['functions'] + '</functions>' -}}{%- endif -%}{{- '<|im_end|>
+' -}}{%- elif message['role'] == 'user' -%}{{- '<|im_start|>user
+' + message['content'] + '<|im_end|>
+' -}}{%- elif message['role'] == 'assistant' -%}{{- '<|im_start|>assistant
+' -}}{%- if message.get('content', none) is not none -%}{{- message['content'] -}}{%- endif -%}{%- if message.get('function_calls', none) is not none and (message['function_calls'] | length) > 0 -%}{{- '<function_calls>' + message['function_calls'] + '</function_calls>' -}}{%- elif message.get('tool_calls', none) is not none and (message['tool_calls'] | length) > 0 -%}{{- '<function_calls>' -}}{%- for tool_call in message['tool_calls'] %}{%- if tool_call is mapping and tool_call.get('function', none) is not none %}{%- set args = tool_call['function']['arguments'] -%}{%- set ns = namespace(arguments_list=[]) -%}{%- for key, value in args.items() -%}{%- set ns.arguments_list = ns.arguments_list + [key ~ '=' ~ (value | tojson)] -%}{%- endfor -%}{%- set arguments = ns.arguments_list | join(', ') -%}{{- tool_call['function']['name'] + '(' + arguments + ')' -}}{%- if not loop.last -%}{{ '
+' }}{%- endif -%}{% else %}{{- tool_call -}}{%- endif %}{%- endfor %}{{- '</function_calls>' -}}{%- endif -%}{%- if not loop.last -%}{{- '<|im_end|>' + '
+' -}}{%- else -%}{{- eos_token -}}{%- endif -%}{%- elif message['role'] == 'environment' -%}{{- '<|im_start|>environment
+' + message['content'] + '<|im_end|>
+' -}}{%- elif message['role'] == 'tool' -%}{{- '<|im_start|>environment
+' + message['content'] + '<|im_end|>
+' -}}{%- endif -%}{%- if loop.last and add_generation_prompt -%}{{- '<|im_start|>assistant
+' -}}{%- endif -%}{%- endfor -%}

config.json

@@ -0,0 +1,66 @@
+{
+  "_name_or_path": "",
+  "architectures": [
+    "Olmo3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": null,
+  "eos_token_id": 100257,
+  "hidden_act": "silu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "layer_types": [
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 65536,
+  "max_sequence_length": 65536,
+  "mlp_bias": false,
+  "model_type": "olmo3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 16,
+  "olmo3_weight_decay": true,
+  "pad_token_id": 100277,
+  "padded_vocab_size": 100352,
+  "qk_norm": true,
+  "qk_norm_mode": "full_projection",
+  "rms_norm_eps": 1e-06,
+  "rope_full_precision": true,
+  "rope_scaling": {
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "factor": 8.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "yarn"
+  },
+  "rope_theta": 500000.0,
+  "scale_embeds": false,
+  "scale_output_layer_init": false,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.57.6",
+  "true_vocab_size": 100278,
+  "truncated_normal_factor": 3.0,
+  "use_cache": true,
+  "vocab_size": 100278,
+  "vocab_z_loss_coeff": 1e-05
+}

conversion_manifest.json

@@ -0,0 +1,230 @@
+{
+  "schema": "olmo3.base.hf_export/v1",
+  "source": {
+    "checkpoint_metadata_sha256": "4183274e33be47adcdbc5da77f50bd201d3e03256d23e3e2a508a6abfb3244e0",
+    "iteration": 3252,
+    "model_schema_sha256": "5be7d972eb3294c31abe5a6ea18bb1b2c34e2ca8225a32a9bf1673fe5d67e4fd",
+    "parameter_count_with_padded_vocab": 1484916736,
+    "run_id": "o3b1b-instruct-sft-dolci-s32768-g32-m1-tp1-cp8-dp32-hsdp32-b2-lr3e5-min0-wd5e2-wu10pct-2ep-256npu-share-20260802-v1"
+  },
+  "stage": "stage4_instruct",
+  "target": {
+    "format": "huggingface_olmo3_safetensors",
+    "padded_vocab_size": 100352,
+    "parameter_count": 1484613632,
+    "tensor_count": 179,
+    "tensor_sha256": {
+      "lm_head.weight": "7de64ec2c4d649dca15e9ec7ce7ffe68c92de85d938e76497bde216d65e404fd",
+      "model.embed_tokens.weight": "d2bb1c2c2c8015540d7496684f27eac53315446378899796ed0d3e4fb72da697",
+      "model.layers.0.mlp.down_proj.weight": "5279872a217b69c3ca9242486ab73f4bb6962aeb7896512c26331173ec7f1f74",
+      "model.layers.0.mlp.gate_proj.weight": "603ea36ec138cd7dcbfb56cd981ca329f89a9369b32be20b2df2d3c547ddf4af",
+      "model.layers.0.mlp.up_proj.weight": "f98a37cea9453c0f370be2110f0d1fe936728600d3bfcd38990b641ad577aff4",
+      "model.layers.0.post_attention_layernorm.weight": "dc2f5fd3ba0513b0cb701fb75b8e8af46e21cb2ed5df3f6e6e4c48b0fdede23f",
+      "model.layers.0.post_feedforward_layernorm.weight": "a0ed78d527a4ec5f2c4d5b094a6adc794add6772d4a4c0ea0bcf6289531ca580",
+      "model.layers.0.self_attn.k_norm.weight": "4dae5b4f931b81cc5c96f32a45baee002383aef39e72435d284bb57f8850e7d6",
+      "model.layers.0.self_attn.k_proj.weight": "7fda545b84110c845830ba0f795a25ee5b0d6d7984a4993aa7a6bcb0e726a3f9",
+      "model.layers.0.self_attn.o_proj.weight": "52839a154e4d79ad8101c118f3966b8e07b87633b23fa1d7606d6f0517deb733",
+      "model.layers.0.self_attn.q_norm.weight": "13321bdcb00bc687a087c0864660a650b81e8658685fe98526a8ad4dbe2372f7",
+      "model.layers.0.self_attn.q_proj.weight": "78be4504f25b2ef22b6876b462ffc77b9f36851075492ab3bb9b22c4708cefd1",
+      "model.layers.0.self_attn.v_proj.weight": "6aad563e420d63046c8aa6676ea92a5f4bf0617a169bf271ec5414dd703b60cd",
+      "model.layers.1.mlp.down_proj.weight": "1a43662bff8a0d372398ae538a87256370c786581010c2c773b9082cffe1f447",
+      "model.layers.1.mlp.gate_proj.weight": "dcdf4f8ccc1f0e502b8dace5e47f65af34789c7222acfb503a410a051491c298",
+      "model.layers.1.mlp.up_proj.weight": "81007bbd88fc1659fab35b5136f98632091ea74050b8df32b4db5aae03cf782f",
+      "model.layers.1.post_attention_layernorm.weight": "458cefca4b9f84516b59f5e5818e115cd5a59053eeace407e2f2d2524f380335",
+      "model.layers.1.post_feedforward_layernorm.weight": "7d7028cc6b93cc02a6a41c7b59ab6d38a7d66e317d7727979ada79eb91b65275",
+      "model.layers.1.self_attn.k_norm.weight": "55f07bf0f9d80019ae1bb0c772fc2d762a3b2d70b3e214cd5b328c4e5d208398",
+      "model.layers.1.self_attn.k_proj.weight": "0ebb15e8d51b1bb4c50e1752d766e91cbb6f91ac6f6d1c38419ba01decdace7c",
+      "model.layers.1.self_attn.o_proj.weight": "3c138179b484c479bf192ebbba71e38417c965ad5f7cd4970756d7eb5f61b4b0",
+      "model.layers.1.self_attn.q_norm.weight": "98ed75f62a65314bb9dcf6aff89c798fe546d2db3168f8b4ce134c83e14d82cd",
+      "model.layers.1.self_attn.q_proj.weight": "58f0d91cded7e14a8a53f91688d926af20405f5ff6f5eaa25908cd7b0d11a16f",
+      "model.layers.1.self_attn.v_proj.weight": "65c7a6aeef00257ee87823fb90fb42710b051dd02354c90effb83d25c2538815",
+      "model.layers.10.mlp.down_proj.weight": "2d017fd83db9581c4a0d66d891d7b505353cb783f1bf4c51c0328119230cd61d",
+      "model.layers.10.mlp.gate_proj.weight": "2e79969db8f886a0d361faba351134386a7971ccb95f2c79450bc552cddba432",
+      "model.layers.10.mlp.up_proj.weight": "6d29e9d154cc14db27a62a40a9e450568cd9080a113de4e19025eccb3bb50730",
+      "model.layers.10.post_attention_layernorm.weight": "be872e03f5f792b669be0d439364728ba25eeac86de36f46f7dc441e853a0c3e",
+      "model.layers.10.post_feedforward_layernorm.weight": "57eb4aad28bbbb40fb1a9db7fa1868f830ead2339d90d3295cf94a1a8f9a697b",
+      "model.layers.10.self_attn.k_norm.weight": "2b2d53d573bc36a8280c3a958a37a516aab36be48e871ddd6bca96eec8f44786",
+      "model.layers.10.self_attn.k_proj.weight": "e47c9578d817d2378b6d41aa097011b49490dbbb9d9dd29c0a931c3994b6cd5d",
+      "model.layers.10.self_attn.o_proj.weight": "a2037b59d722962c0096bdfe64206196723d86bea982ba5a9ccb599c9d2ffcfd",
+      "model.layers.10.self_attn.q_norm.weight": "173b8628f9b7b0786e43682efe1c77d3ca64468d84d1c2032ccb172c769d41d5",
+      "model.layers.10.self_attn.q_proj.weight": "c67157a0a50cb8ed0783b2b58e6805566192202b990803980d20b4b7da722f39",
+      "model.layers.10.self_attn.v_proj.weight": "283a4582608a6d1895d65ecc40e0a98f516de5a7c1a8604a2a12d0ae72ddfb52",
+      "model.layers.11.mlp.down_proj.weight": "0f4fd7474dd8da694ea93b807d21c96081f475f527bec23bdc66b1b2266d56be",
+      "model.layers.11.mlp.gate_proj.weight": "973b70f321ca30dfadb7a13b0861bd7a00f778a1d6a1f6d3619dfe1e907d590d",
+      "model.layers.11.mlp.up_proj.weight": "6da7ddf92811a807e9e7c14f145df81b5c8124ef5c3e9e827e14f112650374d4",
+      "model.layers.11.post_attention_layernorm.weight": "4021c99c0afe135c749b6d9f75864910f549a9887fe23e3518615af39eb8b284",
+      "model.layers.11.post_feedforward_layernorm.weight": "65dc99f1e1d729db5487740ac2b1c23056db4c84138324185aacec449a3561c1",
+      "model.layers.11.self_attn.k_norm.weight": "ee57476ea25e29647c4e640fa08ccc92d5cbf59774cceebf896827f3c16114ed",
+      "model.layers.11.self_attn.k_proj.weight": "199852e4b53b664711ef4c095e4f8c80405cd759a8e6c742c9d8086d6f0cd743",
+      "model.layers.11.self_attn.o_proj.weight": "6b48aa67deb1441ee89292df7c6b67e5f46b7e8f64a24c124708cc79e762be72",
+      "model.layers.11.self_attn.q_norm.weight": "9a36b54348ceb7ddd61c69aa928933442c64dbb67ed6ab402fdceb2235331481",
+      "model.layers.11.self_attn.q_proj.weight": "0673c3b6dc6cd096e23330ef9bc1cb9315e84efc98caf3f9abca849396342606",
+      "model.layers.11.self_attn.v_proj.weight": "331768d0389435d49da739b51294b8254c610339cb61368c8804b28d8438fe3f",
+      "model.layers.12.mlp.down_proj.weight": "a2ac16fca2c9aa9aa66ee5b7e28bf0ba76a938a73356259b63c5353db91450e3",
+      "model.layers.12.mlp.gate_proj.weight": "74b201f390acec311fc70868dc2faf095832b2080de0b3234a800fa6b67186db",
+      "model.layers.12.mlp.up_proj.weight": "9358169f8ee341f4e4e48569eada8cd6a8b82bcf89e359b4e241943e1eedbddc",
+      "model.layers.12.post_attention_layernorm.weight": "2b9151a352b1113bb5d2aab8851ce469d66190fbe8edfe4d6b52be9db8f3a2dc",
+      "model.layers.12.post_feedforward_layernorm.weight": "a028dc247be4c86f64fcc12dc6b22b5f05ad2729e9e21f50e2fd46582fafb71a",
+      "model.layers.12.self_attn.k_norm.weight": "217ac7ee58fdcf779eb81861ecd3f303e4619078f0c0252a4e42824186d82c4c",
+      "model.layers.12.self_attn.k_proj.weight": "8bf923721ea8026ecd378e8e6dae8f1c03b7e224ffdc099bbc44243ef63e1887",
+      "model.layers.12.self_attn.o_proj.weight": "faa2f3834906797925345e562483e29fb0c62487213957bfe77772aa288db4d2",
+      "model.layers.12.self_attn.q_norm.weight": "34e871e337f732511e9ad7df97a131445a8be1b1ce90aef72cc95a50a418722e",
+      "model.layers.12.self_attn.q_proj.weight": "c30035b2c059c5db7403da7eff3e06a1a040f43eb3b5469f3643dbf36e771262",
+      "model.layers.12.self_attn.v_proj.weight": "2ce91f62d939280672b58c674b9c9e7da6287bcf57dda71ce477700b560930c3",
+      "model.layers.13.mlp.down_proj.weight": "f5369b57b93ccb3af04cbb13b5cf29a0fe0adf057641bb17a8eb3a1a55341b69",
+      "model.layers.13.mlp.gate_proj.weight": "8ab65fa2f253b21a7d181584c554f338a85468b43922477669a3cf626227cb01",
+      "model.layers.13.mlp.up_proj.weight": "f7b2ab879fcd9c56d02ff009b82c874169a1748f86cf780952241d6cd38a236b",
+      "model.layers.13.post_attention_layernorm.weight": "8f581de684fb4b54d0de4f34da5a9e4b04109728d0a5e4d590448f50cfdc74a0",
+      "model.layers.13.post_feedforward_layernorm.weight": "fbdf7f00b9021f10839f79d879a93b1a4bfe290c840820652ac23aea5294ecbb",
+      "model.layers.13.self_attn.k_norm.weight": "b202d504e03b6b6a562be5e411db0498fba1c4c220a390b38ce3f804cac83791",
+      "model.layers.13.self_attn.k_proj.weight": "0ac30a6e1bd737e2e282a69cc7b4fae8804cd74055cb69257567d80ac4ee9629",
+      "model.layers.13.self_attn.o_proj.weight": "3fab891250b7c47c037f400d560a70aff62d7b85c449d23eddade92e182f7acb",
+      "model.layers.13.self_attn.q_norm.weight": "69343156ca8f83f9b9c2064741f1d2805efe33ac3a49c1dffb33af0e00b82319",
+      "model.layers.13.self_attn.q_proj.weight": "fa8fac628684c93728465c52602fb1709cd9663b7b129ca7056348458bd06a70",
+      "model.layers.13.self_attn.v_proj.weight": "af277d4cb8e4a05f68daf77e18cbebc8370ec2d3a8d4f9ec6349922d3257b626",
+      "model.layers.14.mlp.down_proj.weight": "f20b9466f7bfcf44198c4bbb528936e572675ed26ac0a1ca8a58f269002bdb05",
+      "model.layers.14.mlp.gate_proj.weight": "3e10da6d53a57429e265d3e3b2970ec2f4cc6409ba4bbfb7b1fe30a2ab36d7ca",
+      "model.layers.14.mlp.up_proj.weight": "d6e1990d736f9b59fb5de93904b4c9a10225abda50c2b906801513d9c44a73fe",
+      "model.layers.14.post_attention_layernorm.weight": "04eabd995d7b9b9c93e48c8e5c182841783a22052eadc59ff4a77235ae7cc5fb",
+      "model.layers.14.post_feedforward_layernorm.weight": "97a61651404d1112955c688550d05e730f16180b9357d3b63cef8aad6843559a",
+      "model.layers.14.self_attn.k_norm.weight": "3df963d7aa5051cb33c3174c3346ac09a55426e763034951e7338191e254556b",
+      "model.layers.14.self_attn.k_proj.weight": "c3178c43c6638df5c357fc9072eaca1b6f73057dce24731d02565139e08cfe27",
+      "model.layers.14.self_attn.o_proj.weight": "480534fe09dd55d540baa1bba4379932093e53014b7a023752ffbd83c5a104a1",
+      "model.layers.14.self_attn.q_norm.weight": "f1e6514964fc11d8e67fdf0d2ff9b668fb6e1913bbf3e092c76a1e7d282d1859",
+      "model.layers.14.self_attn.q_proj.weight": "f7b2faa7bc596c613bdee82ff5b23488255cafd3496054e24fd9a7247523bfa9",
+      "model.layers.14.self_attn.v_proj.weight": "d97d74104f15e64383e5fa3160c72b13307101688938cb583c8d8d4296ce4ab5",
+      "model.layers.15.mlp.down_proj.weight": "7e239d5a1824cff23db2b945a8986670e32caf8abe3e2ceec57d1ff492a168f3",
+      "model.layers.15.mlp.gate_proj.weight": "c4282cd59e7c86cd3b19170f31036bdd1de68772a0a01bf25061af88b30a4de6",
+      "model.layers.15.mlp.up_proj.weight": "4e8f50a5e9849de3d26c009879ef7daa5bbedc462091ff2d6b0ff047edf9f469",
+      "model.layers.15.post_attention_layernorm.weight": "1d8fb5e081890a6606cad41b3871976f29845721675034e72a6fbfcab17bf992",
+      "model.layers.15.post_feedforward_layernorm.weight": "387c08c490d99960def92b95f75c8e76aba7b4fa1e235143330cca82906e45f4",
+      "model.layers.15.self_attn.k_norm.weight": "ecb23adb9ae09f3882e7c881e92e85bb70b60eda9ebb7d931089f7a8c776b4f4",
+      "model.layers.15.self_attn.k_proj.weight": "c97a4b83b1b8b057cb40f50638e37dc9c1a905c93e4a15d050f29dbdafec2a1e",
+      "model.layers.15.self_attn.o_proj.weight": "1c28c4a68df62a2269225bfb7424c13eaf63fd6391866a4f94fd22ac1e0dc0b0",
+      "model.layers.15.self_attn.q_norm.weight": "73f9e46f7d690a3c68b73214858d28d63dccc5cca4a45b4f2bb4eca540f13d7d",
+      "model.layers.15.self_attn.q_proj.weight": "afdb72cd38663d52b58f6cd868724e211058863e221b4b7b4e4b96899b9200ec",
+      "model.layers.15.self_attn.v_proj.weight": "5300d4c7bf013d524ddcd1efe099bb7c528796c903038105e217e128f364eac1",
+      "model.layers.2.mlp.down_proj.weight": "c7bc47f900a2be8b75ee70132246ada2acf8d5c1a80025e9427a7fd818312eeb",
+      "model.layers.2.mlp.gate_proj.weight": "d179057f74f500a71932b021d7ae1c23bef53094dc530821deec06427b1612ca",
+      "model.layers.2.mlp.up_proj.weight": "fdef2ff28f773069798beabae96a7d5e00e9b9fae40d5f33c8a545153affdbdf",
+      "model.layers.2.post_attention_layernorm.weight": "b7d04221fa6d680338d8b8a441feacdce666d55d2f5c61c537d32fbff84ea6dc",
+      "model.layers.2.post_feedforward_layernorm.weight": "cf46f97da255782bfbdbddcc09d5b6b99dc2717dfe1c3170be91d3bdda75b799",
+      "model.layers.2.self_attn.k_norm.weight": "14adcccf719fa4fb594e51179c93627fab2f421773158918a88c202ba22c0c32",
+      "model.layers.2.self_attn.k_proj.weight": "30af8a1e527385d80e9ec805d44b9eacd1c5797224945f60161bf771e5b0b979",
+      "model.layers.2.self_attn.o_proj.weight": "49787916aaf9ba7787a62d84552e6206774f5e707d904c1e3a24788e13c13872",
+      "model.layers.2.self_attn.q_norm.weight": "6728ad4a50540ff0f6c0c2b117cf679bc2483f52a2050deca5202b060472a7c8",
+      "model.layers.2.self_attn.q_proj.weight": "94dbaebe3824fd552245e5ad20230c1478b9799bf4e507095db9bc1542779dd5",
+      "model.layers.2.self_attn.v_proj.weight": "b74a6c47b24a445375f8acaca3f48abc2886b038b773b5be993910e9d7c0fc5b",
+      "model.layers.3.mlp.down_proj.weight": "984ff7a04355899e8c364ae04cb8404b29e7541c324b56a392dfbb55f4c1ad82",
+      "model.layers.3.mlp.gate_proj.weight": "1266958886c7aa649c625ce1297110188d1b578fb7a3c7b2c3e6cdfc5615295c",
+      "model.layers.3.mlp.up_proj.weight": "5dbb791ecc547e50829fab66fede89944d00827cf8746e0c8b726584136254c5",
+      "model.layers.3.post_attention_layernorm.weight": "3c982f62e29a768a78f42a3b4d0d6736163f0f07fcb9a8d65ef993feae1268a9",
+      "model.layers.3.post_feedforward_layernorm.weight": "c05698ade6e81bd27733371198833f7faf04bfe3ca884d7f184a88e1227bbe56",
+      "model.layers.3.self_attn.k_norm.weight": "d3f9ec1a405a9aec0afbab27268ea99366fa08adc4d726f23c287c4a6d0d4b3c",
+      "model.layers.3.self_attn.k_proj.weight": "4a1736e983b195ead58b0ad6d6bdfd4023c22dedfe2d1f76faa177fac55751d5",
+      "model.layers.3.self_attn.o_proj.weight": "21b848c8fa20f0ffbf75a3e0f6ff77c4aadec4b89785224b0843e8cb44fc484f",
+      "model.layers.3.self_attn.q_norm.weight": "555a12a87b6b166e67a2ce0214ee39053f05e85011ab424b6a519555842af941",
+      "model.layers.3.self_attn.q_proj.weight": "b0b415f95b1df9d517dd64bc3ba11bfa6d74dba7384a692646bf768484611cbc",
+      "model.layers.3.self_attn.v_proj.weight": "1ed839641d1bcaff50577011f5afb05373d42c46e31c49b8e40f0dfa1bf92f7b",
+      "model.layers.4.mlp.down_proj.weight": "590dbe74677d9da9f9ddf8347bd0f630a2fbf6361813c8bd5aede6602444c84c",
+      "model.layers.4.mlp.gate_proj.weight": "1c1ce3810650e902c9db633b2a15d5f027b63d0def1629bfa44445d3b14e82d8",
+      "model.layers.4.mlp.up_proj.weight": "ab7620fbfa127e7350dbcbb58ae4843922f1acb042c0a82a9e786b77599fe16e",
+      "model.layers.4.post_attention_layernorm.weight": "40a57a6fe2b194c2ce25afc3a9b5c2f8b5cfce27baa10c1a6e672f1a29307737",
+      "model.layers.4.post_feedforward_layernorm.weight": "b551275e53afb20d33fa94211e184ad30930729c66c93de67fa34b0865ff46f1",
+      "model.layers.4.self_attn.k_norm.weight": "d57f3b8b29fa95bc495fdcc2278f353be718bceb136478ddecb68e9b7548bf89",
+      "model.layers.4.self_attn.k_proj.weight": "32ad2adff51aa4b1f02be59e122e95af2fb7790ec3d799c31bbb6d5b530770bd",
+      "model.layers.4.self_attn.o_proj.weight": "6d185cdf9eb554736371051266e604616035f8e3f634825238c4a126aef04d1f",
+      "model.layers.4.self_attn.q_norm.weight": "fc2ce34d6b04388f79d0897d827f2abd133d2c78c60609bc620d54ed0b45ccf8",
+      "model.layers.4.self_attn.q_proj.weight": "72be60943c66ba12f28c28caf6986eb13d4e62956b0eda5e1a6ce75aaa802a6f",
+      "model.layers.4.self_attn.v_proj.weight": "f4560de3425bde3685c992350459c5722d490f08e429e7d3d7e783ad74799e40",
+      "model.layers.5.mlp.down_proj.weight": "ab85a739d6d9e1e68bde90819826ce9f9f9632a2bfcb16c26c93f7a56d807952",
+      "model.layers.5.mlp.gate_proj.weight": "20c07126905442badf6f861d96ec235733b465b7d517ade376e3b89042dde07b",
+      "model.layers.5.mlp.up_proj.weight": "7910a49790ffa6036e286f443e5999401ac0745d1d80207e4fc1c4e3bc111c5a",
+      "model.layers.5.post_attention_layernorm.weight": "9710174afaecbe7a37169a72ca781601c4b5deab7643acf3bf2c04466da2c94f",
+      "model.layers.5.post_feedforward_layernorm.weight": "9a4eb76a94ab98d5dac58a17d192fb5ea09d584372ca4fcf9b695f400e76fa99",
+      "model.layers.5.self_attn.k_norm.weight": "588211ca8418a2ae2541667613120cedf494e1a292d3b7d4851b921f0c7e8fc2",
+      "model.layers.5.self_attn.k_proj.weight": "b016404d3656095268385e22fed4f042217dcd46edef513d3d3e94568f926f1e",
+      "model.layers.5.self_attn.o_proj.weight": "973e8515973d04482c148a69aa4cd2900e2d679ec1c5e390b85af679a035c98b",
+      "model.layers.5.self_attn.q_norm.weight": "06316e9a494f2a6bb02782879af347ba72501a011571c19f638fd75bf9a5e621",
+      "model.layers.5.self_attn.q_proj.weight": "559b7152a3d93953a50d1d10cc20d6888a07243c1465d401ad6445337345d2ee",
+      "model.layers.5.self_attn.v_proj.weight": "7e1494cdd8bc41ccb12fc37f954aa35ab2bb07c0a72e491c55fac391c688c208",
+      "model.layers.6.mlp.down_proj.weight": "209c454e9ca222de4f2309527b3827afe2784b7ea71a14f534d58b258de46857",
+      "model.layers.6.mlp.gate_proj.weight": "f8504cc83a0228cdb196f666ecb5ff8446fe128cc1e3f1ac9b1d7ba0393906ed",
+      "model.layers.6.mlp.up_proj.weight": "df3c564be868531365d25c4fea3306154899981115cb64c234b6bc3371086024",
+      "model.layers.6.post_attention_layernorm.weight": "1978871f71a0a94a3c6ca312801687a30d9fd2b972c5e53c43e23694dd641d82",
+      "model.layers.6.post_feedforward_layernorm.weight": "eadf2cd625cae0730e5ba52bdb8b420dc93a132dcf7e6897eae1694f7ca9f337",
+      "model.layers.6.self_attn.k_norm.weight": "d574b38f5abbd90404f9bc7a99808fec8c07f88388fe223c0b986d86cd29f948",
+      "model.layers.6.self_attn.k_proj.weight": "3103e596bfe7e5c99b1e6bfbfc908746526e34ab9d4d47fa64de57e8cba1441f",
+      "model.layers.6.self_attn.o_proj.weight": "04170e543483ae1969560cd3b51720b519ea61b6b45483b9428f845c2d47e841",
+      "model.layers.6.self_attn.q_norm.weight": "30949308cec0936fcb8c43cc3b3055d85c40f5761fe28e41b01894a684f1c178",
+      "model.layers.6.self_attn.q_proj.weight": "f6bb45531c724a824b880a5995eb599c6b4be71faba9ed2063ed4b2a8f3827b6",
+      "model.layers.6.self_attn.v_proj.weight": "2eee4ab92e520f2f726b148e410a63d76daa40b6fcccae51f06cc2f1adced5e3",
+      "model.layers.7.mlp.down_proj.weight": "dbefed2c916ffb36a2ac6771e7505972fce99f471127bf75cd52d8ed2c104fc2",
+      "model.layers.7.mlp.gate_proj.weight": "3c66aa656a2d694a9417607e41cab4a7bcb0d5941413a6dea2c529e88bb9c791",
+      "model.layers.7.mlp.up_proj.weight": "b14503a16faeefcfccd8ca0f7c555ab75707506bb2285e29231f66f1ad488184",
+      "model.layers.7.post_attention_layernorm.weight": "3ce945eccd57face9dd7b3394102d601c13b9dfe65eae79a606295c4cffbec5d",
+      "model.layers.7.post_feedforward_layernorm.weight": "ab371324782a2aedb757559ed3406fdcf51dd5453f8eda727add8e0d66c7a31c",
+      "model.layers.7.self_attn.k_norm.weight": "2c815345b37d357488cd21d0d43e0814d5b816158930afd27817dbbebb327b27",
+      "model.layers.7.self_attn.k_proj.weight": "0580f3ff0d1724dbb3e949f5cc87f9e6d0914f140e77355d2d63b2fde3a64958",
+      "model.layers.7.self_attn.o_proj.weight": "7a4d3e572eeea2c373c8f10e7108f72e742f765d9817406f3f442abd72f6025c",
+      "model.layers.7.self_attn.q_norm.weight": "115fd2a690420d71ba010613e7e6c318488248dc757c32f9b66c8aa4ade3cbba",
+      "model.layers.7.self_attn.q_proj.weight": "84e583527548945ecd47edb13fc2fa6f6b02cf1034719e4af7965d88cd0f37c4",
+      "model.layers.7.self_attn.v_proj.weight": "6e4cb2f3c0d16dee585d195d73269ad39f24a50c12904a54c7925d23c2a7e06f",
+      "model.layers.8.mlp.down_proj.weight": "57caaaf1e143b2953e1bc74e5ec86e2fca5849d09ccfc438e9c47af3a941847b",
+      "model.layers.8.mlp.gate_proj.weight": "73450ecb42e549f7c8a3634ee8d7d2dfe2e9c2daba6f786ca080739de388b292",
+      "model.layers.8.mlp.up_proj.weight": "2f8450500a3b16cb3e7f31e19d675d1c381dc71adb099253706494d82aeaa8bd",
+      "model.layers.8.post_attention_layernorm.weight": "6e4db38ee086ed05a733486e646ec9dd55ff0e60a7c1bb89dd466826688e1538",
+      "model.layers.8.post_feedforward_layernorm.weight": "b350a0ccde073d7743f752801eafb0618df1c6d1a7aeefc6bfaec1b8dee72dbc",
+      "model.layers.8.self_attn.k_norm.weight": "416b3ec47499a736dd6833f6cce057442e12179350ddd8893790442b327da638",
+      "model.layers.8.self_attn.k_proj.weight": "7331023e511854870e87730d91a21543aeb5d6086721c06f6ff58426f31dc9e0",
+      "model.layers.8.self_attn.o_proj.weight": "8475eac1690069efc6601d8b08ce622fe70b86ebe67343526f6582fbd64e2bb4",
+      "model.layers.8.self_attn.q_norm.weight": "47769bf0f27f29745fb6e6d9842b0ebf47f68a3380446fdcd1e3a8edda083684",
+      "model.layers.8.self_attn.q_proj.weight": "ea0c27b1ad2e1f2ab9544da6b951bad6d56b3e9f8718ab6fb2d15365a10b7490",
+      "model.layers.8.self_attn.v_proj.weight": "077de9ab22ef6d686b9864eae6b7dedeacc3ae613114e1825ba9b5b4df48f0ed",
+      "model.layers.9.mlp.down_proj.weight": "44bacbe9ccd26e93a80d158b72d7c14d5228520ace44c196ec585f36bcbd0779",
+      "model.layers.9.mlp.gate_proj.weight": "89cf1ac0bcec3b17e467aba1a2affdd975a1c694c61bdb8e99530277835bf290",
+      "model.layers.9.mlp.up_proj.weight": "b1115b90da640a6584a84474f0c068d9483fad7ca70dccde2babc6398cf0dc72",
+      "model.layers.9.post_attention_layernorm.weight": "cdf20fed19c6b6f2bdc02dcc76e88b7f43205ab7af3104c0894256b7ada1b663",
+      "model.layers.9.post_feedforward_layernorm.weight": "ca6192476248828679b8344ad57795b1adda191fb448e352a815676ae49befd8",
+      "model.layers.9.self_attn.k_norm.weight": "68233cf240106d360dd7cce1de4086c8188b19981837a1c962e0d66005a9cf7d",
+      "model.layers.9.self_attn.k_proj.weight": "9c0329f9cc48f485a1805a0109b2f2a1a1be32f702f880a0061e4c078476f4ef",
+      "model.layers.9.self_attn.o_proj.weight": "3919f6f9437b491ab89b48fcb72333c13c01791f695ba5af1446112c47d061f0",
+      "model.layers.9.self_attn.q_norm.weight": "286239e7a4052004cb4c480d5bd6919bcf049cbd8f61b2ecdbdfe7ab4c600b92",
+      "model.layers.9.self_attn.q_proj.weight": "22541491bc4e4415ea71e36728156ad9213a1659f1b0ee7971b0a9b3d3907339",
+      "model.layers.9.self_attn.v_proj.weight": "52214dc7a06cc9a80205cf65e8900b3cca5cb1c8e00eab13a786bc28e3efbe25",
+      "model.norm.weight": "70aed28683df7ab71d811c91e975f323a89a1e05637131eec518dabd015647bf"
+    },
+    "tokenizer_files": [
+      "tokenizer.json",
+      "tokenizer_config.json",
+      "special_tokens_map.json",
+      "chat_template.jinja",
+      "vocab.json",
+      "merges.txt"
+    ],
+    "transformers": ">=4.57.6,<5",
+    "true_vocab_size": 100278,
+    "vocab_crop": {
+      "embedding": {
+        "exported_rows": 100278,
+        "removed_rows": 74,
+        "removed_tail_sha256": "9478ec4d7366618d530922bea53340f7ff004f464119c3635b5a0a59457b3818",
+        "source_rows": 100352
+      },
+      "lm_head": {
+        "exported_rows": 100278,
+        "removed_rows": 74,
+        "removed_tail_sha256": "7e2c5b9f2fb7ae84e16943b8ca2ffc9e5176991df0d4620405dbf3fc4642b212",
+        "source_rows": 100352
+      }
+    }
+  },
+  "validation": {
+    "fc1_split_repack": "bitwise",
+    "finite_model_tensors": true,
+    "qkv_split_repack": "bitwise",
+    "safetensors_reload": "bitwise",
+    "source_key_coverage": "exact",
+    "source_shapes": "exact"
+  }
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": null,
+  "eos_token_id": 100257,
+  "pad_token_id": 100277,
+  "transformers_version": "4.57.6"
+}

hf_validation_report.json

@@ -0,0 +1,20 @@
+{
+  "forward": {
+    "cache_no_cache_max_abs": 0.0,
+    "logits_finite": true,
+    "logits_shape": [
+      1,
+      4,
+      100278
+    ],
+    "post_8192_yarn_finite": true
+  },
+  "parameter_count": 1484613632,
+  "schema": "olmo3.base.hf_validation/v1",
+  "stage": "stage4_instruct",
+  "status": "pass",
+  "tensor_count": 179,
+  "tokenizer_size": 100278,
+  "torch_version": "2.6.0+cpu",
+  "transformers_version": "4.57.6"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:197cee5740ba915cce8a470c25b5680c43c5b03bde6c0a310809858d2ec26244
+size 2969248016

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,191 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "100256": {
+      "content": "<|extra_id_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100257": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100258": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100259": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100260": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100261": {
+      "content": "|||PHONE_NUMBER|||",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100262": {
+      "content": "|||EMAIL_ADDRESS|||",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100263": {
+      "content": "|||IP_ADDRESS|||",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100264": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100265": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100266": {
+      "content": "<functions>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100267": {
+      "content": "</functions>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100268": {
+      "content": "<function_calls>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100269": {
+      "content": "</function_calls>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100270": {
+      "content": "<|extra_id_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100271": {
+      "content": "<|extra_id_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100272": {
+      "content": "<|extra_id_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100273": {
+      "content": "<|extra_id_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100274": {
+      "content": "<|extra_id_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100275": {
+      "content": "<|extra_id_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100276": {
+      "content": "<|endofprompt|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100277": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "chat_template": "{%- set has_system = messages|selectattr('role', 'equalto', 'system')|list|length > 0 -%}{%- if not has_system -%}{{- '<|im_start|>system\nYou are a helpful function-calling AI assistant. ' -}}{%- if tools is none or (tools | length) == 0 -%}{{- 'You do not currently have access to any functions. <functions></functions><|im_end|>\n' -}}{%- else -%}{{- 'You are provided with function signatures within <functions></functions> XML tags. You may call one or more functions to assist with the user query. Output any function calls within <function_calls></function_calls> XML tags. Do not make assumptions about what values to plug into functions.' -}}{{- '<functions>' -}}{{- tools | tojson -}}{{- '</functions><|im_end|>\n' -}}{%- endif -%}{%- endif -%}{%- for message in messages -%}{%- if message['role'] == 'system' -%}{{- '<|im_start|>system\n' + message['content'] -}}{%- if tools is not none -%}{{- '<functions>' -}}{{- tools | tojson -}}{{- '</functions>' -}}{%- elif message.get('functions', none) is not none -%}{{- ' <functions>' + message['functions'] + '</functions>' -}}{%- endif -%}{{- '<|im_end|>\n' -}}{%- elif message['role'] == 'user' -%}{{- '<|im_start|>user\n' + message['content'] + '<|im_end|>\n' -}}{%- elif message['role'] == 'assistant' -%}{{- '<|im_start|>assistant\n' -}}{%- if message.get('content', none) is not none -%}{{- message['content'] -}}{%- endif -%}{%- if message.get('function_calls', none) is not none and (message['function_calls'] | length) > 0 -%}{{- '<function_calls>' + message['function_calls'] + '</function_calls>' -}}{%- elif message.get('tool_calls', none) is not none and (message['tool_calls'] | length) > 0 -%}{{- '<function_calls>' -}}{%- for tool_call in message['tool_calls'] %}{%- if tool_call is mapping and tool_call.get('function', none) is not none %}{%- set args = tool_call['function']['arguments'] -%}{%- set ns = namespace(arguments_list=[]) -%}{%- for key, value in args.items() -%}{%- set ns.arguments_list = ns.arguments_list + [key ~ '=' ~ (value | tojson)] -%}{%- endfor -%}{%- set arguments = ns.arguments_list | join(', ') -%}{{- tool_call['function']['name'] + '(' + arguments + ')' -}}{%- if not loop.last -%}{{ '\n' }}{%- endif -%}{% else %}{{- tool_call -}}{%- endif %}{%- endfor %}{{- '</function_calls>' -}}{%- endif -%}{%- if not loop.last -%}{{- '<|im_end|>' + '\n' -}}{%- else -%}{{- eos_token -}}{%- endif -%}{%- elif message['role'] == 'environment' -%}{{- '<|im_start|>environment\n' + message['content'] + '<|im_end|>\n' -}}{%- elif message['role'] == 'tool' -%}{{- '<|im_start|>environment\n' + message['content'] + '<|im_end|>\n' -}}{%- endif -%}{%- if loop.last and add_generation_prompt -%}{{- '<|im_start|>assistant\n' -}}{%- endif -%}{%- endfor -%}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "fix_mistral_regex": false,
+  "model_max_length": 65536,
+  "pad_token": "<|pad|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}