add adapter cwm_arith_wholetrace

.gitattributes

@@ -45,3 +45,4 @@ cwm_mixed_expanded/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 cwm_oop_expanded/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 cwm_arith_gold/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 cwm_arith_drift/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+cwm_arith_wholetrace/tokenizer.json filter=lfs diff=lfs merge=lfs -text

cwm_arith_wholetrace/adapter_config.json

@@ -0,0 +1,48 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "/home/t-nagupta/.cache/huggingface/hub/models--facebook--cwm/snapshots/1b572b04e7f9613bb8ea902fedff4a1ed887c573/",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "v_proj",
+    "k_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

cwm_arith_wholetrace/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39422f211b9e1216224b453bb2943cd21b70c941eccffe88d058ca430b0111cd
+size 499243216

cwm_arith_wholetrace/chat_template.jinja

@@ -0,0 +1,13 @@
+{% set loop_messages=messages %}{%- set ns=namespace(last_query_index=-1) %}{%- for message in loop_messages[::-1] %}{%- set index=(loop_messages|length-1)-loop.index0 %}{%- if message.role=='user' %}{%- set ns.last_query_index=index %}{%- break %}{%- endif %}{%- endfor %}{% for message in loop_messages %}{%- set loop_index=loop.index0 %}{%- set content=message['content']|trim %}{%- if message.role=='assistant' %}{%- set reasoning_content='' %}{%- if '<think>
+' in content and '</think>' in content %}{%- set reasoning_content=content.split('</think>')[0].split('<think>
+')[-1].strip() %}{%- set content=content.split('</think>')[-1].lstrip() %}{%- endif %}{%- if loop_index>ns.last_query_index or preserve_previous_think is not defined or preserve_previous_think is true %}{%- if loop_index==0 %}{{ bos_token }}{%- endif %}{{ '<|start_header_id|>assistant<|end_header_id|>
+
+' }}{%- if reasoning_content %}{{ '<think>
+' }}{{ reasoning_content }}{{ '</think>' }}{%- endif %}{{ content }}<|eot_id|>{%- else %}{%- if loop_index==0 %}{{ bos_token }}{%- endif %}{{ '<|start_header_id|>assistant<|end_header_id|>
+
+' }}{{ content }}<|eot_id|>{%- endif %}{%- else %}{%- if loop_index==0 %}{{ bos_token }}{%- endif %}<|start_header_id|>{{ message.role }}{{ '<|end_header_id|>
+
+' }}{{ content }}<|eot_id|>{%- endif %}{%- endfor %}{%- if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>
+
+' }}{%- if enable_thinking is not defined or enable_thinking is true %}{{ '<think>
+' }}{%- endif %}{%- endif %}

cwm_arith_wholetrace/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ce6c1030d1247437cf3e2d197c37f3ab4dad08113c7a9956dae6db15479e0eb
+size 17208464

cwm_arith_wholetrace/tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|end_of_text|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "tokenizer_class": "TokenizersBackend"
+}