BHARGAV REDDY commited on
Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/chat_template.jinja +4 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/config.json +32 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/generation_config.json +10 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/model.safetensors +3 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/tokenizer.json +3 -0
- full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/tokenizer_config.json +43 -0
- full_models/cpt_latest_raw_base_step_12000/chat_template.jinja +4 -0
- full_models/cpt_latest_raw_base_step_12000/config.json +32 -0
- full_models/cpt_latest_raw_base_step_12000/generation_config.json +10 -0
- full_models/cpt_latest_raw_base_step_12000/model.safetensors +3 -0
- full_models/cpt_latest_raw_base_step_12000/tokenizer.json +3 -0
- full_models/cpt_latest_raw_base_step_12000/tokenizer_config.json +43 -0
.gitattributes
CHANGED
|
@@ -36,3 +36,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 36 |
sample_models/2026-08-08_cpt_latest_and_sft_v18/cpt_latest_sample/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
full_models/base_25667/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
full_models/v13_cpt30518_r16_h100_final_merged_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 36 |
sample_models/2026-08-08_cpt_latest_and_sft_v18/cpt_latest_sample/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
full_models/base_25667/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
full_models/v13_cpt30518_r16_h100_final_merged_model/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
full_models/cpt_latest_raw_base_step_12000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/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 %}
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 2,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 5632,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 20,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"rope_theta": 500000.0,
|
| 26 |
+
"rope_type": "default"
|
| 27 |
+
},
|
| 28 |
+
"tie_word_embeddings": true,
|
| 29 |
+
"transformers_version": "5.14.1",
|
| 30 |
+
"use_cache": false,
|
| 31 |
+
"vocab_size": 131072
|
| 32 |
+
}
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"pad_token_id": 1,
|
| 8 |
+
"transformers_version": "5.14.1",
|
| 9 |
+
"use_cache": false
|
| 10 |
+
}
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac297ad44788e713909274e49253210e9d246378afd2a8fb1249776593843ff8
|
| 3 |
+
size 2424496664
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d254ca0e87091d1a9b56fb22414ecac1e807d011c76013c621336f2be81b88c
|
| 3 |
+
size 12183261
|
full_models/cpt_latest_raw_base_step_12000/_snap/project_source_phase1_phase2_20260707/cpt_32b_2k/runs/cpt_32b_2k_v3/checkpoint-step-12000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|bos|>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"extra_special_tokens": [
|
| 7 |
+
"<|eos|>",
|
| 8 |
+
"<|sep|>",
|
| 9 |
+
"<|mask|>",
|
| 10 |
+
"<|system|>",
|
| 11 |
+
"<|user|>",
|
| 12 |
+
"<|assistant|>",
|
| 13 |
+
"<|tool|>",
|
| 14 |
+
"<|end|>",
|
| 15 |
+
"<|tool_call|>",
|
| 16 |
+
"<|tool_result|>",
|
| 17 |
+
"<|fim_prefix|>",
|
| 18 |
+
"<|fim_suffix|>",
|
| 19 |
+
"<|fim_middle|>",
|
| 20 |
+
"<|image|>",
|
| 21 |
+
"<|audio|>",
|
| 22 |
+
"<|video|>",
|
| 23 |
+
"<|code|>",
|
| 24 |
+
"<|curl|>",
|
| 25 |
+
"<|email|>",
|
| 26 |
+
"<|cite|>",
|
| 27 |
+
"<|think|>",
|
| 28 |
+
"<|im_start|>",
|
| 29 |
+
"<|im_end|>",
|
| 30 |
+
"<|endofprompt|>",
|
| 31 |
+
"<|repo_name|>",
|
| 32 |
+
"<|file_sep|>",
|
| 33 |
+
"<|image_start|>",
|
| 34 |
+
"<|image_end|>",
|
| 35 |
+
"<|image_pad|>"
|
| 36 |
+
],
|
| 37 |
+
"is_local": true,
|
| 38 |
+
"local_files_only": false,
|
| 39 |
+
"model_max_length": 1048576,
|
| 40 |
+
"pad_token": "<|pad|>",
|
| 41 |
+
"tokenizer_class": "TokenizersBackend",
|
| 42 |
+
"unk_token": null
|
| 43 |
+
}
|
full_models/cpt_latest_raw_base_step_12000/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 %}
|
full_models/cpt_latest_raw_base_step_12000/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 2,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 0,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 5632,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 20,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"rope_theta": 500000.0,
|
| 26 |
+
"rope_type": "default"
|
| 27 |
+
},
|
| 28 |
+
"tie_word_embeddings": true,
|
| 29 |
+
"transformers_version": "5.14.1",
|
| 30 |
+
"use_cache": false,
|
| 31 |
+
"vocab_size": 131072
|
| 32 |
+
}
|
full_models/cpt_latest_raw_base_step_12000/generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"pad_token_id": 1,
|
| 8 |
+
"transformers_version": "5.14.1",
|
| 9 |
+
"use_cache": false
|
| 10 |
+
}
|
full_models/cpt_latest_raw_base_step_12000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac297ad44788e713909274e49253210e9d246378afd2a8fb1249776593843ff8
|
| 3 |
+
size 2424496664
|
full_models/cpt_latest_raw_base_step_12000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d254ca0e87091d1a9b56fb22414ecac1e807d011c76013c621336f2be81b88c
|
| 3 |
+
size 12183261
|
full_models/cpt_latest_raw_base_step_12000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|bos|>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"extra_special_tokens": [
|
| 7 |
+
"<|eos|>",
|
| 8 |
+
"<|sep|>",
|
| 9 |
+
"<|mask|>",
|
| 10 |
+
"<|system|>",
|
| 11 |
+
"<|user|>",
|
| 12 |
+
"<|assistant|>",
|
| 13 |
+
"<|tool|>",
|
| 14 |
+
"<|end|>",
|
| 15 |
+
"<|tool_call|>",
|
| 16 |
+
"<|tool_result|>",
|
| 17 |
+
"<|fim_prefix|>",
|
| 18 |
+
"<|fim_suffix|>",
|
| 19 |
+
"<|fim_middle|>",
|
| 20 |
+
"<|image|>",
|
| 21 |
+
"<|audio|>",
|
| 22 |
+
"<|video|>",
|
| 23 |
+
"<|code|>",
|
| 24 |
+
"<|curl|>",
|
| 25 |
+
"<|email|>",
|
| 26 |
+
"<|cite|>",
|
| 27 |
+
"<|think|>",
|
| 28 |
+
"<|im_start|>",
|
| 29 |
+
"<|im_end|>",
|
| 30 |
+
"<|endofprompt|>",
|
| 31 |
+
"<|repo_name|>",
|
| 32 |
+
"<|file_sep|>",
|
| 33 |
+
"<|image_start|>",
|
| 34 |
+
"<|image_end|>",
|
| 35 |
+
"<|image_pad|>"
|
| 36 |
+
],
|
| 37 |
+
"is_local": true,
|
| 38 |
+
"local_files_only": false,
|
| 39 |
+
"model_max_length": 1048576,
|
| 40 |
+
"pad_token": "<|pad|>",
|
| 41 |
+
"tokenizer_class": "TokenizersBackend",
|
| 42 |
+
"unk_token": null
|
| 43 |
+
}
|