Hugging Face's logo

twhitworth
/
gpt-oss-120b-fp16

Text Generation
Transformers
Safetensors
English
gpt_oss
fp16
dequantized
gpt-oss
mxfp4-upcast
conversational
Model card Files
xet
Community
3
twhitworth commited on
Commit
85d738f
·
verified ·
1 Parent(s): 3c48e5e

Upload 78 files

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +1 -0
  2. README.md +59 -0
  3. config.json +135 -0
  4. model-00001-of-00073.safetensors +3 -0
  5. model-00002-of-00073.safetensors +3 -0
  6. model-00003-of-00073.safetensors +3 -0
  7. model-00004-of-00073.safetensors +3 -0
  8. model-00005-of-00073.safetensors +3 -0
  9. model-00006-of-00073.safetensors +3 -0
  10. model-00007-of-00073.safetensors +3 -0
  11. model-00008-of-00073.safetensors +3 -0
  12. model-00009-of-00073.safetensors +3 -0
  13. model-00010-of-00073.safetensors +3 -0
  14. model-00011-of-00073.safetensors +3 -0
  15. model-00012-of-00073.safetensors +3 -0
  16. model-00013-of-00073.safetensors +3 -0
  17. model-00014-of-00073.safetensors +3 -0
  18. model-00015-of-00073.safetensors +3 -0
  19. model-00016-of-00073.safetensors +3 -0
  20. model-00017-of-00073.safetensors +3 -0
  21. model-00018-of-00073.safetensors +3 -0
  22. model-00019-of-00073.safetensors +3 -0
  23. model-00020-of-00073.safetensors +3 -0
  24. model-00021-of-00073.safetensors +3 -0
  25. model-00022-of-00073.safetensors +3 -0
  26. model-00023-of-00073.safetensors +3 -0
  27. model-00024-of-00073.safetensors +3 -0
  28. model-00025-of-00073.safetensors +3 -0
  29. model-00026-of-00073.safetensors +3 -0
  30. model-00027-of-00073.safetensors +3 -0
  31. model-00028-of-00073.safetensors +3 -0
  32. model-00029-of-00073.safetensors +3 -0
  33. model-00030-of-00073.safetensors +3 -0
  34. model-00031-of-00073.safetensors +3 -0
  35. model-00032-of-00073.safetensors +3 -0
  36. model-00033-of-00073.safetensors +3 -0
  37. model-00034-of-00073.safetensors +3 -0
  38. model-00035-of-00073.safetensors +3 -0
  39. model-00036-of-00073.safetensors +3 -0
  40. model-00037-of-00073.safetensors +3 -0
  41. model-00038-of-00073.safetensors +3 -0
  42. model-00039-of-00073.safetensors +3 -0
  43. model-00040-of-00073.safetensors +3 -0
  44. model-00041-of-00073.safetensors +3 -0
  45. model-00042-of-00073.safetensors +3 -0
  46. model-00043-of-00073.safetensors +3 -0
  47. model-00044-of-00073.safetensors +3 -0
  48. model-00045-of-00073.safetensors +3 -0
  49. model-00046-of-00073.safetensors +3 -0
  50. model-00047-of-00073.safetensors +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ## Precision: FP32 vs FP16 (and BF16)
2
+
3
+ This project saves dequantized checkpoints in **FP16** (default) or **BF16**.
4
+ Here’s what those formats mean and when you might choose each:
5
+
6
+ ### TL;DR
7
+
8
+ - **FP32 (single precision, 32-bit, 4 bytes/param)**
9
+ Reference/default precision in many frameworks. Highest numerical range/precision, **largest memory**.
10
+ - **FP16 (half precision, 16-bit, 2 bytes/param)**
11
+ Half the memory of FP32. Great for **inference** on modern GPUs; may underflow/overflow more easily than BF16.
12
+ - **BF16 (bfloat16, 16-bit, 2 bytes/param)**
13
+ Same memory as FP16, **wider exponent like FP32**, often more numerically robust than FP16; slightly less precision in mantissa.
14
+
15
+ > In this repo, output precision is **FP16** (default) or **BF16** via `--dtype`.
16
+ > **FP32 output is not offered** because it doubles disk/RAM vs FP16/BF16 with minimal inference benefit on modern hardware.
17
+
18
+ ### Memory math (example: 120B parameters)
19
+
20
+ Each parameter stores one number:
21
+
22
+ | Format | Bits | Bytes/param | Approx size for 120B params |
23
+ |-------:|-----:|-------------:|-----------------------------:|
24
+ | FP32 | 32 | 4 | ~ **447 GiB** |
25
+ | FP16 | 16 | 2 | ~ **224 GiB** |
26
+ | BF16 | 16 | 2 | ~ **224 GiB** |
27
+
28
+ > Calculation (GiB): `params * bytes_per_param / 1024^3`
29
+ > For 120,000,000,000 params:
30
+ > FP32: 480e9 B ≈ 447.03 GiB
31
+ > FP16/BF16: 240e9 B ≈ 223.52 GiB
32
+
33
+ ### When to use which
34
+
35
+ - **Inference on modern NVIDIA GPUs (Turing+/Ampere+/Ada/Hopper):**
36
+ Use **FP16** (default here) or **BF16**. You’ll get large memory savings and typically **equal or faster** throughput than FP32 thanks to tensor cores.
37
+
38
+ - **Training / Finetuning:**
39
+ Use **mixed precision** (BF16 or FP16 compute with an FP32 master copy of weights/optimizer states).
40
+ If your GPU supports BF16 well (e.g., A100/H100), **BF16** is preferred for numeric stability.
41
+ (This tool focuses on exporting dequantized checkpoints, not training loops.)
42
+
43
+ - **If you hit numeric issues in FP16:**
44
+ Try **BF16** (`--dtype bf16`). Same size as FP16 but usually more stable due to FP32-like exponent range.
45
+
46
+ ### Notes
47
+
48
+ - **FP32** remains the gold standard for numeric headroom and deterministic baselines, but for **inference** it’s typically unnecessary and **costly** (2× memory vs FP16/BF16).
49
+ - **Tensor cores** accelerate FP16/BF16 GEMMs on most modern NVIDIA GPUs; FP32 is often slower and more memory-bound.
50
+ - If a downstream runtime expects a specific dtype, export to that: FP16 for speed/memory, BF16 for robustness.
51
+
52
+ ---
53
+
54
+ ### WIP
55
+
56
+ - Upcoming models: cleaned FP16 release (uniform fp16 with fp32 LayerNorms), compressed variants (W8A8, W4A16, mixed experts), and smaller distilled checkpoints.
57
+ - Evals: MMLU, HellaSwag, TruthfulQA, GSM8K, BBH, MT‑Bench; plus latency/throughput and memory footprint on 3090/A100.
58
+ - Extras: scripted upload tooling, detailed model cards, and reproducible Docker workflows.
59
+
config.json ADDED
@@ -0,0 +1,135 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 201088,
3
+ "hidden_size": 2880,
4
+ "intermediate_size": 2880,
5
+ "num_hidden_layers": 36,
6
+ "num_attention_heads": 64,
7
+ "num_local_experts": 128,
8
+ "sliding_window": 128,
9
+ "num_experts_per_tok": 4,
10
+ "num_key_value_heads": 8,
11
+ "hidden_act": "silu",
12
+ "initializer_range": 0.02,
13
+ "rms_norm_eps": 0.00001,
14
+ "rope_theta": 150000,
15
+ "rope_scaling": {
16
+ "beta_fast": 32.0,
17
+ "beta_slow": 1.0,
18
+ "factor": 32.0,
19
+ "original_max_position_embeddings": 4096,
20
+ "rope_type": "yarn",
21
+ "truncate": false
22
+ },
23
+ "attention_dropout": 0.0,
24
+ "head_dim": 64,
25
+ "layer_types": [
26
+ "sliding_attention",
27
+ "full_attention",
28
+ "sliding_attention",
29
+ "full_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "full_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "full_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "full_attention",
48
+ "sliding_attention",
49
+ "full_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "full_attention",
54
+ "sliding_attention",
55
+ "full_attention",
56
+ "sliding_attention",
57
+ "full_attention",
58
+ "sliding_attention",
59
+ "full_attention",
60
+ "sliding_attention",
61
+ "full_attention"
62
+ ],
63
+ "attention_bias": true,
64
+ "max_position_embeddings": 131072,
65
+ "router_aux_loss_coef": 0.9,
66
+ "output_router_logits": false,
67
+ "use_cache": true,
68
+ "return_dict": true,
69
+ "output_hidden_states": false,
70
+ "torchscript": false,
71
+ "torch_dtype": "float16",
72
+ "pruned_heads": {},
73
+ "tie_word_embeddings": false,
74
+ "chunk_size_feed_forward": 0,
75
+ "is_encoder_decoder": false,
76
+ "is_decoder": false,
77
+ "cross_attention_hidden_size": null,
78
+ "add_cross_attention": false,
79
+ "tie_encoder_decoder": false,
80
+ "architectures": [
81
+ "GptOssForCausalLM"
82
+ ],
83
+ "finetuning_task": null,
84
+ "id2label": {
85
+ "0": "LABEL_0",
86
+ "1": "LABEL_1"
87
+ },
88
+ "label2id": {
89
+ "LABEL_0": 0,
90
+ "LABEL_1": 1
91
+ },
92
+ "task_specific_params": null,
93
+ "problem_type": null,
94
+ "tokenizer_class": null,
95
+ "prefix": null,
96
+ "bos_token_id": null,
97
+ "pad_token_id": 199999,
98
+ "eos_token_id": 200002,
99
+ "sep_token_id": null,
100
+ "decoder_start_token_id": null,
101
+ "max_length": 20,
102
+ "min_length": 0,
103
+ "do_sample": false,
104
+ "early_stopping": false,
105
+ "num_beams": 1,
106
+ "num_beam_groups": 1,
107
+ "diversity_penalty": 0.0,
108
+ "temperature": 1.0,
109
+ "top_k": 50,
110
+ "top_p": 1.0,
111
+ "typical_p": 1.0,
112
+ "repetition_penalty": 1.0,
113
+ "length_penalty": 1.0,
114
+ "no_repeat_ngram_size": 0,
115
+ "encoder_no_repeat_ngram_size": 0,
116
+ "bad_words_ids": null,
117
+ "num_return_sequences": 1,
118
+ "output_scores": false,
119
+ "return_dict_in_generate": false,
120
+ "forced_bos_token_id": null,
121
+ "forced_eos_token_id": null,
122
+ "remove_invalid_values": false,
123
+ "exponential_decay_length_penalty": null,
124
+ "suppress_tokens": null,
125
+ "begin_suppress_tokens": null,
126
+ "_name_or_path": "openai/gpt-oss-120b",
127
+ "transformers_version": "4.55.1",
128
+ "experts_per_token": 4,
129
+ "initial_context_length": 4096,
130
+ "model_type": "gpt_oss",
131
+ "swiglu_limit": 7.0,
132
+ "tf_legacy_loss": false,
133
+ "use_bfloat16": false,
134
+ "output_attentions": false
135
+ }
model-00001-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d2cbeb91ab681dacd107aaabeebdaa7ce658189f486954f66c91be68ed51eae
3
+ size 1212106024
model-00002-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a106d6a1cbfedad04ba58fe55341c5d0662479348151c683f70152ed5b7295f
3
+ size 4248207608
model-00003-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20e43143de2a41a7f5b4ac3e16fb94ba45a7bfbbd45ba4ed9e838779010caf79
3
+ size 2177966200
model-00004-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f490a24b4ef54c1b81da8ca86689e9962956cebaca6e585a08d80fd435d5c7e5
3
+ size 4248207608
model-00005-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:000240ccc86960cfea8dc7c52bf82375b4b07c22201d5c0dc0c184e84387ee80
3
+ size 2177966200
model-00006-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:746c2890ddbb9d176d6660916ba68d37bde3eca64cb203f7a28e539dd354a078
3
+ size 4248207608
model-00007-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe87b82b49dd8101577e5a68927a0a87f0b8811562362bc1311d0739eec047da
3
+ size 2177966200
model-00008-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:384dd1a38790cf44ace59438fd4ffdeebdfa90d0025b936cf99039c33de0fda2
3
+ size 4248207608
model-00009-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:168709ddd68dfeda0827e9a10d4e5f57505f23f4886693ddf6174f6de7106493
3
+ size 2177966200
model-00010-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0389fca580334c41f6ab048cc8a89dc43ae4bf3b69d8498d86c443af5611938
3
+ size 4248207608
model-00011-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:893b5dbb553a442a1bbb0ddfe86820d94b9a02bdc6d2df8ce0f05a21947f54f4
3
+ size 2177966200
model-00012-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c27f5622cf1c5f2264f278c2f06653fd9acdbb04c01b1956e44317a1c9755c3
3
+ size 4248207608
model-00013-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c909b19ef07269c000f667cdea42df693a04fbdf46e3b724285dd6385963b14
3
+ size 2177966200
model-00014-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:95f5f32613862b01373a5b0fb79c828db876e53a06a396f5beac60ed3e0cd789
3
+ size 4248207608
model-00015-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e57df02381f6981b212c17cf6db3540ab61df6d9d5d44ac1d25af70b4b1cf373
3
+ size 2177966200
model-00016-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:882418dc94e57875150aa6f0dda3bc039b9cec42bc693ce540c483d9ce9d0089
3
+ size 4248207608
model-00017-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee6e015481ae64c8863b9ec7b5f6e88a93c5758d5971f3850dfaadbbce065e59
3
+ size 2177966200
model-00018-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:209b8e4122ca03062ba1134d1763f7dfe4c7e2d6a85a0fcda1d8700ebeb58c1b
3
+ size 4248207608
model-00019-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6dd629c13b37f3a8ac58dbe4c5b00ce85e0b3c4e70197f0ac7ca8092e5e33f3
3
+ size 2177966200
model-00020-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a08672de7209f0a375991ee4ca9cfb89817e4ae309b510190e5a556fc825db44
3
+ size 4248207608
model-00021-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a92acdaf093342d2ac369b872b1b83dd9de4aa25f777d0d11a82494a709113d
3
+ size 2177966152
model-00022-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6280c4f71600f0d8464c5372123ec5ba0e44c97ac688fa533e609a100eef35db
3
+ size 4248207608
model-00023-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fef9ded36a25b51c66bdadde514ba58cb133abc91ee71506c91fd3b2f861229
3
+ size 2177966216
model-00024-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1cfb393370bf9259f926ecd35a9f384da00b994380d8817f3105dcd57747f8b
3
+ size 4248207608
model-00025-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7c17e9181eae9c67dfc7549e9496a0e2e671be8cb6d3de025927238c95daeef
3
+ size 2177966216
model-00026-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d42e377e00438cc5f12ade59fc5e3effba7dfcaf5b18e9653f90da670a0d1af
3
+ size 4248207608
model-00027-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50b154ebc25911c44468f2f44961ea7c559cadb5c5feb8f5575d16c2517431d1
3
+ size 2177966216
model-00028-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5314c859821cc46eda633aca96bae5d512eb35bfbd7d25b157c971e04e52ca69
3
+ size 4248207608
model-00029-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a32ed91720cda3ec9c7a7fcf27ce78e4225b15fa50430a0f788808234d0775c2
3
+ size 2177966216
model-00030-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d06097e01bb51703a688f1e86838779903ba117a104810da64ced95fe462af72
3
+ size 4248207608
model-00031-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0ac34e3e886e0f05ceaab5864b8ea5b97629c2e9d76fe6ca946ef01694df6b8
3
+ size 2177966216
model-00032-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e16fbe7fddd34ca5f8270f627be113d8c4eb53db14ba238b2128fa4c9b3addbd
3
+ size 4248207608
model-00033-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08f466aa0f45c1680fe817ef426e5dfcbdb6a49dbf88cc444dd1c49482967fda
3
+ size 2177966216
model-00034-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da4efac69476ee725226941c7505cdec52a2be88603e1ec08f356e6a1ad228a0
3
+ size 4248207608
model-00035-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:325a31858596fd5b1ec971300c097d1ca0dfafff0953b2622739c4503456cbc7
3
+ size 2177966216
model-00036-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e0b23972d8a42c05557793a837ce46350b0bf3b1383110ebdaecc059e25b0bd
3
+ size 4248207608
model-00037-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6784d0a32a398ac929e875b703d033f517f408415eb3705e6770b5320a2206b8
3
+ size 2177966216
model-00038-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5088b51769655642be1d7521f6e054659a4a5dbeee75668ddd2f8fa2c63faf3
3
+ size 4248207608
model-00039-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2638fd7664b8ebefdd33cc53f25a6d3aa283defbdf9ef649fd0f3739a1685e6b
3
+ size 2177966216
model-00040-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:737c35d33842a328fec4a998c11eb041964dc65ff2da5887a088b36aefa1ebfd
3
+ size 4248207608
model-00041-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a96999e1a529f0421389416aaf14d4857070a34e4764b5e4e3519325ec99e71
3
+ size 2177966216
model-00042-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0d719182ffb2b82beba426c2f56e6f8f8f854169ab189c45c2bc9fb2271556b
3
+ size 4248207608
model-00043-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe1340d9628d1829ec086e1d21855b3d04e3e137fe44f9ba698b8cb6fa600091
3
+ size 2177966216
model-00044-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1ad8e6e5d06780f9c6a09e991a71f0978ba9e88b7224f146a5933c13f1f319c
3
+ size 4248207608
model-00045-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21c53d8e8a7827388018fec44f0de55f7079493ef9a5ec697f2c3d3258cc1565
3
+ size 2177966216
model-00046-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3650773042b703b7b4974e3877f4eb25152f4e9b27500788357bc55651bc8c86
3
+ size 4248207608
model-00047-of-00073.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54f3dcfa2f275c5271776797922415c966a944710f791f8c11788ebef072ec06
3
+ size 2177966216