Hugging Face's logo

PrunaAI
/
bigscience-bloom-7b1-bnb-8bit-smashed

Text Generation
Transformers
Safetensors
bloom
pruna-ai
text-generation-inference
Inference Endpoints
8-bit precision
bitsandbytes
Model card Files Community
2
sharpenb commited on
Commit
3f3ff8e
•
1 Parent(s): e3cddc6

36e4f11b01e0fb82675328faa3601a274b6a4b8ac00e94bdbcbc7dfb4550f1c0

Browse files
Files changed (6) hide show
  1. README.md +81 -0
  2. config.json +46 -0
  3. generation_config.json +7 -0
  4. model.safetensors.index.json +492 -0
  5. plots.png +0 -0
  6. smash_config.json +27 -0
README.md ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: pruna-engine
3
+ thumbnail: "https://assets-global.website-files.com/646b351987a8d8ce158d1940/64ec9e96b4334c0e1ac41504_Logo%20with%20white%20text.svg"
4
+ metrics:
5
+ - memory_disk
6
+ - memory_inference
7
+ - inference_latency
8
+ - inference_throughput
9
+ - inference_CO2_emissions
10
+ - inference_energy_consumption
11
+ ---
12
+ <!-- header start -->
13
+ <!-- 200823 -->
14
+ <div style="width: auto; margin-left: auto; margin-right: auto">
15
+ <a href="https://www.pruna.ai/" target="_blank" rel="noopener noreferrer">
16
+ <img src="https://i.imgur.com/eDAlcgk.png" alt="PrunaAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
17
+ </a>
18
+ </div>
19
+ <!-- header end -->
20
+
21
+ [![Twitter](https://img.shields.io/twitter/follow/PrunaAI?style=social)](https://twitter.com/PrunaAI)
22
+ [![GitHub](https://img.shields.io/github/followers/PrunaAI?label=Follow%20%40PrunaAI&style=social)](https://github.com/PrunaAI)
23
+ [![LinkedIn](https://img.shields.io/badge/LinkedIn-Connect-blue)](https://www.linkedin.com/company/93832878/admin/feed/posts/?feedType=following)
24
+ [![Discord](https://img.shields.io/badge/Discord-Join%20Us-blue?style=social&logo=discord)](https://discord.gg/CP4VSgck)
25
+
26
+ # Simply make AI models cheaper, smaller, faster, and greener!
27
+
28
+ - Give a thumbs up if you like this model!
29
+ - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
30
+ - Request access to easily compress your *own* AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
31
+ - Read the documentations to know more [here](https://pruna-ai-pruna.readthedocs-hosted.com/en/latest/)
32
+ - Join Pruna AI community on Discord [here](https://discord.gg/CP4VSgck) to share feedback/suggestions or get help.
33
+
34
+ ## Results
35
+
36
+ ![image info](./plots.png)
37
+
38
+ **Frequently Asked Questions**
39
+ - ***How does the compression work?*** The model is compressed with llm-int8.
40
+ - ***How does the model quality change?*** The quality of the model output might vary compared to the base model.
41
+ - ***How is the model efficiency evaluated?*** These results were obtained on NVIDIA A100-PCIE-40GB with configuration described in `model/smash_config.json` and are obtained after a hardware warmup. The smashed model is directly compared to the original base model. Efficiency results may vary in other settings (e.g. other hardware, image size, batch size, ...). We recommend to directly run them in the use-case conditions to know if the smashed model can benefit you.
42
+ - ***What is the model format?*** We use safetensors.
43
+ - ***What is the naming convention for Pruna Huggingface models?*** We take the original model name and append "turbo", "tiny", or "green" if the smashed model has a measured inference speed, inference memory, or inference energy consumption which is less than 90% of the original base model.
44
+ - ***How to compress my own models?*** You can request premium access to more compression methods and tech support for your specific use-cases [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
45
+ - ***What are "first" metrics?*** Results mentioning "first" are obtained after the first run of the model. The first run might take more memory or be slower than the subsequent runs due cuda overheads.
46
+ - ***What are "Sync" and "Async" metrics?*** "Sync" metrics are obtained by syncing all GPU processes and stop measurement when all of them are executed. "Async" metrics are obtained without syncing all GPU processes and stop when the model output can be used by the CPU. We provide both metrics since both could be relevant depending on the use-case. We recommend to test the efficiency gains directly in your use-cases.
47
+
48
+ ## Setup
49
+
50
+ You can run the smashed model with these steps:
51
+
52
+ 0. Check requirements from the original repo bigscience/bloom-7b1 installed. In particular, check python, cuda, and transformers versions.
53
+ 1. Make sure that you have installed quantization related packages.
54
+ ```bash
55
+ pip install transformers accelerate bitsandbytes>0.37.0
56
+ ```
57
+ 2. Load & run the model.
58
+ ```python
59
+ from transformers import AutoModelForCausalLM, AutoTokenizer
60
+
61
+ model = AutoModelForCausalLM.from_pretrained("PrunaAI/bigscience-bloom-7b1-bnb-8bit-smashed",
62
+ trust_remote_code=True)
63
+ tokenizer = AutoTokenizer.from_pretrained("bigscience/bloom-7b1")
64
+
65
+ input_ids = tokenizer("What is the color of prunes?,", return_tensors='pt').to(model.device)["input_ids"]
66
+
67
+ outputs = model.generate(input_ids, max_new_tokens=216)
68
+ ```
69
+
70
+ ## Configurations
71
+
72
+ The configuration info are in `smash_config.json`.
73
+
74
+ ## Credits & License
75
+
76
+ The license of the smashed model follows the license of the original model. Please check the license of the original model bigscience/bloom-7b1 before using this model which provided the base model. The license of the `pruna-engine` is [here](https://pypi.org/project/pruna-engine/) on Pypi.
77
+
78
+ ## Want to compress other models?
79
+
80
+ - Contact us and tell us which model to compress next [here](https://www.pruna.ai/contact).
81
+ - Request access to easily compress your own AI models [here](https://z0halsaff74.typeform.com/pruna-access?typeform-source=www.pruna.ai).
config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/tmp/tmp0yw42s4z",
3
+ "apply_residual_connection_post_layernorm": false,
4
+ "architectures": [
5
+ "BloomForCausalLM"
6
+ ],
7
+ "attention_dropout": 0.0,
8
+ "attention_softmax_in_fp32": true,
9
+ "bias_dropout_fusion": true,
10
+ "bos_token_id": 1,
11
+ "eos_token_id": 2,
12
+ "hidden_dropout": 0.0,
13
+ "hidden_size": 4096,
14
+ "initializer_range": 0.02,
15
+ "layer_norm_epsilon": 1e-05,
16
+ "masked_softmax_fusion": true,
17
+ "model_type": "bloom",
18
+ "n_head": 32,
19
+ "n_inner": null,
20
+ "n_layer": 30,
21
+ "offset_alibi": 100,
22
+ "pad_token_id": 3,
23
+ "pretraining_tp": 1,
24
+ "quantization_config": {
25
+ "bnb_4bit_compute_dtype": "bfloat16",
26
+ "bnb_4bit_quant_type": "fp4",
27
+ "bnb_4bit_use_double_quant": true,
28
+ "llm_int8_enable_fp32_cpu_offload": false,
29
+ "llm_int8_has_fp16_weight": false,
30
+ "llm_int8_skip_modules": [
31
+ "lm_head"
32
+ ],
33
+ "llm_int8_threshold": 6.0,
34
+ "load_in_4bit": false,
35
+ "load_in_8bit": true,
36
+ "quant_method": "bitsandbytes"
37
+ },
38
+ "skip_bias_add": true,
39
+ "skip_bias_add_qkv": false,
40
+ "slow_but_exact": false,
41
+ "torch_dtype": "float16",
42
+ "transformers_version": "4.37.1",
43
+ "unk_token_id": 0,
44
+ "use_cache": true,
45
+ "vocab_size": 250880
46
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 3,
6
+ "transformers_version": "4.37.1"
7
+ }
model.safetensors.index.json ADDED
@@ -0,0 +1,492 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 8102658048
4
+ },
5
+ "weight_map": {
6
+ "transformer.h.0.input_layernorm.bias": "model-00001-of-00002.safetensors",
7
+ "transformer.h.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
8
+ "transformer.h.0.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
9
+ "transformer.h.0.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
10
+ "transformer.h.0.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
11
+ "transformer.h.0.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
12
+ "transformer.h.0.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
13
+ "transformer.h.0.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
14
+ "transformer.h.0.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
15
+ "transformer.h.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
16
+ "transformer.h.0.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
17
+ "transformer.h.0.self_attention.dense.bias": "model-00001-of-00002.safetensors",
18
+ "transformer.h.0.self_attention.dense.weight": "model-00001-of-00002.safetensors",
19
+ "transformer.h.0.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
20
+ "transformer.h.0.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
21
+ "transformer.h.0.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
22
+ "transformer.h.1.input_layernorm.bias": "model-00001-of-00002.safetensors",
23
+ "transformer.h.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
24
+ "transformer.h.1.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
25
+ "transformer.h.1.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
26
+ "transformer.h.1.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
27
+ "transformer.h.1.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
28
+ "transformer.h.1.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
29
+ "transformer.h.1.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
30
+ "transformer.h.1.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
31
+ "transformer.h.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
32
+ "transformer.h.1.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
33
+ "transformer.h.1.self_attention.dense.bias": "model-00001-of-00002.safetensors",
34
+ "transformer.h.1.self_attention.dense.weight": "model-00001-of-00002.safetensors",
35
+ "transformer.h.1.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
36
+ "transformer.h.1.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
37
+ "transformer.h.1.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
38
+ "transformer.h.10.input_layernorm.bias": "model-00001-of-00002.safetensors",
39
+ "transformer.h.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
40
+ "transformer.h.10.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
41
+ "transformer.h.10.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
42
+ "transformer.h.10.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
43
+ "transformer.h.10.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
44
+ "transformer.h.10.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
45
+ "transformer.h.10.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
46
+ "transformer.h.10.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
47
+ "transformer.h.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
48
+ "transformer.h.10.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
49
+ "transformer.h.10.self_attention.dense.bias": "model-00001-of-00002.safetensors",
50
+ "transformer.h.10.self_attention.dense.weight": "model-00001-of-00002.safetensors",
51
+ "transformer.h.10.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
52
+ "transformer.h.10.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
53
+ "transformer.h.10.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
54
+ "transformer.h.11.input_layernorm.bias": "model-00001-of-00002.safetensors",
55
+ "transformer.h.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
56
+ "transformer.h.11.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
57
+ "transformer.h.11.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
58
+ "transformer.h.11.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
59
+ "transformer.h.11.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
60
+ "transformer.h.11.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
61
+ "transformer.h.11.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
62
+ "transformer.h.11.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
63
+ "transformer.h.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
64
+ "transformer.h.11.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
65
+ "transformer.h.11.self_attention.dense.bias": "model-00001-of-00002.safetensors",
66
+ "transformer.h.11.self_attention.dense.weight": "model-00001-of-00002.safetensors",
67
+ "transformer.h.11.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
68
+ "transformer.h.11.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
69
+ "transformer.h.11.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
70
+ "transformer.h.12.input_layernorm.bias": "model-00001-of-00002.safetensors",
71
+ "transformer.h.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
72
+ "transformer.h.12.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
73
+ "transformer.h.12.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
74
+ "transformer.h.12.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
75
+ "transformer.h.12.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
76
+ "transformer.h.12.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
77
+ "transformer.h.12.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
78
+ "transformer.h.12.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
79
+ "transformer.h.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
80
+ "transformer.h.12.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
81
+ "transformer.h.12.self_attention.dense.bias": "model-00001-of-00002.safetensors",
82
+ "transformer.h.12.self_attention.dense.weight": "model-00001-of-00002.safetensors",
83
+ "transformer.h.12.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
84
+ "transformer.h.12.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
85
+ "transformer.h.12.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
86
+ "transformer.h.13.input_layernorm.bias": "model-00001-of-00002.safetensors",
87
+ "transformer.h.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
88
+ "transformer.h.13.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
89
+ "transformer.h.13.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
90
+ "transformer.h.13.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
91
+ "transformer.h.13.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
92
+ "transformer.h.13.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
93
+ "transformer.h.13.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
94
+ "transformer.h.13.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
95
+ "transformer.h.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
96
+ "transformer.h.13.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
97
+ "transformer.h.13.self_attention.dense.bias": "model-00001-of-00002.safetensors",
98
+ "transformer.h.13.self_attention.dense.weight": "model-00001-of-00002.safetensors",
99
+ "transformer.h.13.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
100
+ "transformer.h.13.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
101
+ "transformer.h.13.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
102
+ "transformer.h.14.input_layernorm.bias": "model-00001-of-00002.safetensors",
103
+ "transformer.h.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
104
+ "transformer.h.14.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
105
+ "transformer.h.14.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
106
+ "transformer.h.14.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
107
+ "transformer.h.14.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
108
+ "transformer.h.14.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
109
+ "transformer.h.14.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
110
+ "transformer.h.14.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
111
+ "transformer.h.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
112
+ "transformer.h.14.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
113
+ "transformer.h.14.self_attention.dense.bias": "model-00001-of-00002.safetensors",
114
+ "transformer.h.14.self_attention.dense.weight": "model-00001-of-00002.safetensors",
115
+ "transformer.h.14.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
116
+ "transformer.h.14.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
117
+ "transformer.h.14.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
118
+ "transformer.h.15.input_layernorm.bias": "model-00002-of-00002.safetensors",
119
+ "transformer.h.15.input_layernorm.weight": "model-00002-of-00002.safetensors",
120
+ "transformer.h.15.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
121
+ "transformer.h.15.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
122
+ "transformer.h.15.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
123
+ "transformer.h.15.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
124
+ "transformer.h.15.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
125
+ "transformer.h.15.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
126
+ "transformer.h.15.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
127
+ "transformer.h.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
128
+ "transformer.h.15.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
129
+ "transformer.h.15.self_attention.dense.bias": "model-00002-of-00002.safetensors",
130
+ "transformer.h.15.self_attention.dense.weight": "model-00002-of-00002.safetensors",
131
+ "transformer.h.15.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
132
+ "transformer.h.15.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
133
+ "transformer.h.15.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
134
+ "transformer.h.16.input_layernorm.bias": "model-00002-of-00002.safetensors",
135
+ "transformer.h.16.input_layernorm.weight": "model-00002-of-00002.safetensors",
136
+ "transformer.h.16.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
137
+ "transformer.h.16.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
138
+ "transformer.h.16.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
139
+ "transformer.h.16.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
140
+ "transformer.h.16.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
141
+ "transformer.h.16.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
142
+ "transformer.h.16.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
143
+ "transformer.h.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
144
+ "transformer.h.16.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
145
+ "transformer.h.16.self_attention.dense.bias": "model-00002-of-00002.safetensors",
146
+ "transformer.h.16.self_attention.dense.weight": "model-00002-of-00002.safetensors",
147
+ "transformer.h.16.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
148
+ "transformer.h.16.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
149
+ "transformer.h.16.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
150
+ "transformer.h.17.input_layernorm.bias": "model-00002-of-00002.safetensors",
151
+ "transformer.h.17.input_layernorm.weight": "model-00002-of-00002.safetensors",
152
+ "transformer.h.17.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
153
+ "transformer.h.17.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
154
+ "transformer.h.17.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
155
+ "transformer.h.17.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
156
+ "transformer.h.17.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
157
+ "transformer.h.17.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
158
+ "transformer.h.17.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
159
+ "transformer.h.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
160
+ "transformer.h.17.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
161
+ "transformer.h.17.self_attention.dense.bias": "model-00002-of-00002.safetensors",
162
+ "transformer.h.17.self_attention.dense.weight": "model-00002-of-00002.safetensors",
163
+ "transformer.h.17.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
164
+ "transformer.h.17.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
165
+ "transformer.h.17.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
166
+ "transformer.h.18.input_layernorm.bias": "model-00002-of-00002.safetensors",
167
+ "transformer.h.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
168
+ "transformer.h.18.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
169
+ "transformer.h.18.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
170
+ "transformer.h.18.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
171
+ "transformer.h.18.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
172
+ "transformer.h.18.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
173
+ "transformer.h.18.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
174
+ "transformer.h.18.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
175
+ "transformer.h.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
176
+ "transformer.h.18.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
177
+ "transformer.h.18.self_attention.dense.bias": "model-00002-of-00002.safetensors",
178
+ "transformer.h.18.self_attention.dense.weight": "model-00002-of-00002.safetensors",
179
+ "transformer.h.18.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
180
+ "transformer.h.18.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
181
+ "transformer.h.18.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
182
+ "transformer.h.19.input_layernorm.bias": "model-00002-of-00002.safetensors",
183
+ "transformer.h.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
184
+ "transformer.h.19.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
185
+ "transformer.h.19.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
186
+ "transformer.h.19.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
187
+ "transformer.h.19.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
188
+ "transformer.h.19.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
189
+ "transformer.h.19.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
190
+ "transformer.h.19.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
191
+ "transformer.h.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
192
+ "transformer.h.19.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
193
+ "transformer.h.19.self_attention.dense.bias": "model-00002-of-00002.safetensors",
194
+ "transformer.h.19.self_attention.dense.weight": "model-00002-of-00002.safetensors",
195
+ "transformer.h.19.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
196
+ "transformer.h.19.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
197
+ "transformer.h.19.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
198
+ "transformer.h.2.input_layernorm.bias": "model-00001-of-00002.safetensors",
199
+ "transformer.h.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
200
+ "transformer.h.2.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
201
+ "transformer.h.2.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
202
+ "transformer.h.2.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
203
+ "transformer.h.2.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
204
+ "transformer.h.2.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
205
+ "transformer.h.2.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
206
+ "transformer.h.2.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
207
+ "transformer.h.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
208
+ "transformer.h.2.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
209
+ "transformer.h.2.self_attention.dense.bias": "model-00001-of-00002.safetensors",
210
+ "transformer.h.2.self_attention.dense.weight": "model-00001-of-00002.safetensors",
211
+ "transformer.h.2.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
212
+ "transformer.h.2.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
213
+ "transformer.h.2.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
214
+ "transformer.h.20.input_layernorm.bias": "model-00002-of-00002.safetensors",
215
+ "transformer.h.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
216
+ "transformer.h.20.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
217
+ "transformer.h.20.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
218
+ "transformer.h.20.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
219
+ "transformer.h.20.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
220
+ "transformer.h.20.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
221
+ "transformer.h.20.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
222
+ "transformer.h.20.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
223
+ "transformer.h.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
224
+ "transformer.h.20.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
225
+ "transformer.h.20.self_attention.dense.bias": "model-00002-of-00002.safetensors",
226
+ "transformer.h.20.self_attention.dense.weight": "model-00002-of-00002.safetensors",
227
+ "transformer.h.20.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
228
+ "transformer.h.20.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
229
+ "transformer.h.20.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
230
+ "transformer.h.21.input_layernorm.bias": "model-00002-of-00002.safetensors",
231
+ "transformer.h.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
232
+ "transformer.h.21.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
233
+ "transformer.h.21.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
234
+ "transformer.h.21.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
235
+ "transformer.h.21.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
236
+ "transformer.h.21.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
237
+ "transformer.h.21.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
238
+ "transformer.h.21.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
239
+ "transformer.h.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
240
+ "transformer.h.21.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
241
+ "transformer.h.21.self_attention.dense.bias": "model-00002-of-00002.safetensors",
242
+ "transformer.h.21.self_attention.dense.weight": "model-00002-of-00002.safetensors",
243
+ "transformer.h.21.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
244
+ "transformer.h.21.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
245
+ "transformer.h.21.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
246
+ "transformer.h.22.input_layernorm.bias": "model-00002-of-00002.safetensors",
247
+ "transformer.h.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
248
+ "transformer.h.22.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
249
+ "transformer.h.22.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
250
+ "transformer.h.22.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
251
+ "transformer.h.22.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
252
+ "transformer.h.22.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
253
+ "transformer.h.22.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
254
+ "transformer.h.22.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
255
+ "transformer.h.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
256
+ "transformer.h.22.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
257
+ "transformer.h.22.self_attention.dense.bias": "model-00002-of-00002.safetensors",
258
+ "transformer.h.22.self_attention.dense.weight": "model-00002-of-00002.safetensors",
259
+ "transformer.h.22.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
260
+ "transformer.h.22.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
261
+ "transformer.h.22.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
262
+ "transformer.h.23.input_layernorm.bias": "model-00002-of-00002.safetensors",
263
+ "transformer.h.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
264
+ "transformer.h.23.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
265
+ "transformer.h.23.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
266
+ "transformer.h.23.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
267
+ "transformer.h.23.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
268
+ "transformer.h.23.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
269
+ "transformer.h.23.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
270
+ "transformer.h.23.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
271
+ "transformer.h.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
272
+ "transformer.h.23.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
273
+ "transformer.h.23.self_attention.dense.bias": "model-00002-of-00002.safetensors",
274
+ "transformer.h.23.self_attention.dense.weight": "model-00002-of-00002.safetensors",
275
+ "transformer.h.23.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
276
+ "transformer.h.23.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
277
+ "transformer.h.23.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
278
+ "transformer.h.24.input_layernorm.bias": "model-00002-of-00002.safetensors",
279
+ "transformer.h.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
280
+ "transformer.h.24.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
281
+ "transformer.h.24.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
282
+ "transformer.h.24.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
283
+ "transformer.h.24.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
284
+ "transformer.h.24.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
285
+ "transformer.h.24.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
286
+ "transformer.h.24.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
287
+ "transformer.h.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
288
+ "transformer.h.24.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
289
+ "transformer.h.24.self_attention.dense.bias": "model-00002-of-00002.safetensors",
290
+ "transformer.h.24.self_attention.dense.weight": "model-00002-of-00002.safetensors",
291
+ "transformer.h.24.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
292
+ "transformer.h.24.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
293
+ "transformer.h.24.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
294
+ "transformer.h.25.input_layernorm.bias": "model-00002-of-00002.safetensors",
295
+ "transformer.h.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
296
+ "transformer.h.25.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
297
+ "transformer.h.25.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
298
+ "transformer.h.25.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
299
+ "transformer.h.25.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
300
+ "transformer.h.25.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
301
+ "transformer.h.25.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
302
+ "transformer.h.25.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
303
+ "transformer.h.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
304
+ "transformer.h.25.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
305
+ "transformer.h.25.self_attention.dense.bias": "model-00002-of-00002.safetensors",
306
+ "transformer.h.25.self_attention.dense.weight": "model-00002-of-00002.safetensors",
307
+ "transformer.h.25.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
308
+ "transformer.h.25.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
309
+ "transformer.h.25.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
310
+ "transformer.h.26.input_layernorm.bias": "model-00002-of-00002.safetensors",
311
+ "transformer.h.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
312
+ "transformer.h.26.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
313
+ "transformer.h.26.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
314
+ "transformer.h.26.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
315
+ "transformer.h.26.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
316
+ "transformer.h.26.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
317
+ "transformer.h.26.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
318
+ "transformer.h.26.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
319
+ "transformer.h.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
320
+ "transformer.h.26.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
321
+ "transformer.h.26.self_attention.dense.bias": "model-00002-of-00002.safetensors",
322
+ "transformer.h.26.self_attention.dense.weight": "model-00002-of-00002.safetensors",
323
+ "transformer.h.26.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
324
+ "transformer.h.26.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
325
+ "transformer.h.26.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
326
+ "transformer.h.27.input_layernorm.bias": "model-00002-of-00002.safetensors",
327
+ "transformer.h.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
328
+ "transformer.h.27.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
329
+ "transformer.h.27.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
330
+ "transformer.h.27.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
331
+ "transformer.h.27.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
332
+ "transformer.h.27.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
333
+ "transformer.h.27.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
334
+ "transformer.h.27.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
335
+ "transformer.h.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
336
+ "transformer.h.27.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
337
+ "transformer.h.27.self_attention.dense.bias": "model-00002-of-00002.safetensors",
338
+ "transformer.h.27.self_attention.dense.weight": "model-00002-of-00002.safetensors",
339
+ "transformer.h.27.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
340
+ "transformer.h.27.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
341
+ "transformer.h.27.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
342
+ "transformer.h.28.input_layernorm.bias": "model-00002-of-00002.safetensors",
343
+ "transformer.h.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
344
+ "transformer.h.28.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
345
+ "transformer.h.28.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
346
+ "transformer.h.28.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
347
+ "transformer.h.28.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
348
+ "transformer.h.28.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
349
+ "transformer.h.28.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
350
+ "transformer.h.28.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
351
+ "transformer.h.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
352
+ "transformer.h.28.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
353
+ "transformer.h.28.self_attention.dense.bias": "model-00002-of-00002.safetensors",
354
+ "transformer.h.28.self_attention.dense.weight": "model-00002-of-00002.safetensors",
355
+ "transformer.h.28.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
356
+ "transformer.h.28.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
357
+ "transformer.h.28.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
358
+ "transformer.h.29.input_layernorm.bias": "model-00002-of-00002.safetensors",
359
+ "transformer.h.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
360
+ "transformer.h.29.mlp.dense_4h_to_h.SCB": "model-00002-of-00002.safetensors",
361
+ "transformer.h.29.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
362
+ "transformer.h.29.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
363
+ "transformer.h.29.mlp.dense_h_to_4h.SCB": "model-00002-of-00002.safetensors",
364
+ "transformer.h.29.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
365
+ "transformer.h.29.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
366
+ "transformer.h.29.post_attention_layernorm.bias": "model-00002-of-00002.safetensors",
367
+ "transformer.h.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
368
+ "transformer.h.29.self_attention.dense.SCB": "model-00002-of-00002.safetensors",
369
+ "transformer.h.29.self_attention.dense.bias": "model-00002-of-00002.safetensors",
370
+ "transformer.h.29.self_attention.dense.weight": "model-00002-of-00002.safetensors",
371
+ "transformer.h.29.self_attention.query_key_value.SCB": "model-00002-of-00002.safetensors",
372
+ "transformer.h.29.self_attention.query_key_value.bias": "model-00002-of-00002.safetensors",
373
+ "transformer.h.29.self_attention.query_key_value.weight": "model-00002-of-00002.safetensors",
374
+ "transformer.h.3.input_layernorm.bias": "model-00001-of-00002.safetensors",
375
+ "transformer.h.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
376
+ "transformer.h.3.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
377
+ "transformer.h.3.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
378
+ "transformer.h.3.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
379
+ "transformer.h.3.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
380
+ "transformer.h.3.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
381
+ "transformer.h.3.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
382
+ "transformer.h.3.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
383
+ "transformer.h.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
384
+ "transformer.h.3.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
385
+ "transformer.h.3.self_attention.dense.bias": "model-00001-of-00002.safetensors",
386
+ "transformer.h.3.self_attention.dense.weight": "model-00001-of-00002.safetensors",
387
+ "transformer.h.3.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
388
+ "transformer.h.3.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
389
+ "transformer.h.3.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
390
+ "transformer.h.4.input_layernorm.bias": "model-00001-of-00002.safetensors",
391
+ "transformer.h.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
392
+ "transformer.h.4.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
393
+ "transformer.h.4.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
394
+ "transformer.h.4.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
395
+ "transformer.h.4.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
396
+ "transformer.h.4.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
397
+ "transformer.h.4.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
398
+ "transformer.h.4.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
399
+ "transformer.h.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
400
+ "transformer.h.4.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
401
+ "transformer.h.4.self_attention.dense.bias": "model-00001-of-00002.safetensors",
402
+ "transformer.h.4.self_attention.dense.weight": "model-00001-of-00002.safetensors",
403
+ "transformer.h.4.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
404
+ "transformer.h.4.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
405
+ "transformer.h.4.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
406
+ "transformer.h.5.input_layernorm.bias": "model-00001-of-00002.safetensors",
407
+ "transformer.h.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
408
+ "transformer.h.5.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
409
+ "transformer.h.5.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
410
+ "transformer.h.5.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
411
+ "transformer.h.5.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
412
+ "transformer.h.5.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
413
+ "transformer.h.5.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
414
+ "transformer.h.5.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
415
+ "transformer.h.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
416
+ "transformer.h.5.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
417
+ "transformer.h.5.self_attention.dense.bias": "model-00001-of-00002.safetensors",
418
+ "transformer.h.5.self_attention.dense.weight": "model-00001-of-00002.safetensors",
419
+ "transformer.h.5.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
420
+ "transformer.h.5.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
421
+ "transformer.h.5.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
422
+ "transformer.h.6.input_layernorm.bias": "model-00001-of-00002.safetensors",
423
+ "transformer.h.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
424
+ "transformer.h.6.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
425
+ "transformer.h.6.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
426
+ "transformer.h.6.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
427
+ "transformer.h.6.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
428
+ "transformer.h.6.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
429
+ "transformer.h.6.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
430
+ "transformer.h.6.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
431
+ "transformer.h.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
432
+ "transformer.h.6.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
433
+ "transformer.h.6.self_attention.dense.bias": "model-00001-of-00002.safetensors",
434
+ "transformer.h.6.self_attention.dense.weight": "model-00001-of-00002.safetensors",
435
+ "transformer.h.6.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
436
+ "transformer.h.6.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
437
+ "transformer.h.6.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
438
+ "transformer.h.7.input_layernorm.bias": "model-00001-of-00002.safetensors",
439
+ "transformer.h.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
440
+ "transformer.h.7.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
441
+ "transformer.h.7.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
442
+ "transformer.h.7.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
443
+ "transformer.h.7.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
444
+ "transformer.h.7.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
445
+ "transformer.h.7.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
446
+ "transformer.h.7.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
447
+ "transformer.h.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
448
+ "transformer.h.7.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
449
+ "transformer.h.7.self_attention.dense.bias": "model-00001-of-00002.safetensors",
450
+ "transformer.h.7.self_attention.dense.weight": "model-00001-of-00002.safetensors",
451
+ "transformer.h.7.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
452
+ "transformer.h.7.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
453
+ "transformer.h.7.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
454
+ "transformer.h.8.input_layernorm.bias": "model-00001-of-00002.safetensors",
455
+ "transformer.h.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
456
+ "transformer.h.8.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
457
+ "transformer.h.8.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
458
+ "transformer.h.8.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
459
+ "transformer.h.8.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
460
+ "transformer.h.8.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
461
+ "transformer.h.8.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
462
+ "transformer.h.8.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
463
+ "transformer.h.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
464
+ "transformer.h.8.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
465
+ "transformer.h.8.self_attention.dense.bias": "model-00001-of-00002.safetensors",
466
+ "transformer.h.8.self_attention.dense.weight": "model-00001-of-00002.safetensors",
467
+ "transformer.h.8.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
468
+ "transformer.h.8.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
469
+ "transformer.h.8.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
470
+ "transformer.h.9.input_layernorm.bias": "model-00001-of-00002.safetensors",
471
+ "transformer.h.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
472
+ "transformer.h.9.mlp.dense_4h_to_h.SCB": "model-00001-of-00002.safetensors",
473
+ "transformer.h.9.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
474
+ "transformer.h.9.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
475
+ "transformer.h.9.mlp.dense_h_to_4h.SCB": "model-00001-of-00002.safetensors",
476
+ "transformer.h.9.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
477
+ "transformer.h.9.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
478
+ "transformer.h.9.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
479
+ "transformer.h.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
480
+ "transformer.h.9.self_attention.dense.SCB": "model-00001-of-00002.safetensors",
481
+ "transformer.h.9.self_attention.dense.bias": "model-00001-of-00002.safetensors",
482
+ "transformer.h.9.self_attention.dense.weight": "model-00001-of-00002.safetensors",
483
+ "transformer.h.9.self_attention.query_key_value.SCB": "model-00001-of-00002.safetensors",
484
+ "transformer.h.9.self_attention.query_key_value.bias": "model-00001-of-00002.safetensors",
485
+ "transformer.h.9.self_attention.query_key_value.weight": "model-00001-of-00002.safetensors",
486
+ "transformer.ln_f.bias": "model-00002-of-00002.safetensors",
487
+ "transformer.ln_f.weight": "model-00002-of-00002.safetensors",
488
+ "transformer.word_embeddings.weight": "model-00001-of-00002.safetensors",
489
+ "transformer.word_embeddings_layernorm.bias": "model-00001-of-00002.safetensors",
490
+ "transformer.word_embeddings_layernorm.weight": "model-00001-of-00002.safetensors"
491
+ }
492
+ }
plots.png ADDED
smash_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "api_key": null,
3
+ "verify_url": "http://johnrachwan.pythonanywhere.com",
4
+ "smash_config": {
5
+ "pruners": "None",
6
+ "factorizers": "None",
7
+ "quantizers": "['llm-int8']",
8
+ "compilers": "None",
9
+ "task": "text_text_generation",
10
+ "device": "cuda",
11
+ "cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsygm4y5gb",
12
+ "batch_size": 1,
13
+ "model_name": "bigscience/bloom-7b1",
14
+ "pruning_ratio": 0.0,
15
+ "n_quantization_bits": 8,
16
+ "output_deviation": 0.005,
17
+ "max_batch_size": 1,
18
+ "qtype_weight": "torch.qint8",
19
+ "qtype_activation": "torch.quint8",
20
+ "qobserver": "<class 'torch.ao.quantization.observer.MinMaxObserver'>",
21
+ "qscheme": "torch.per_tensor_symmetric",
22
+ "qconfig": "x86",
23
+ "group_size": 128,
24
+ "damp_percent": 0.1,
25
+ "save_load_fn": "bitsandbytes"
26
+ }
27
+ }