Hugging Face's logo

uirev
/
MLX_OpenELM_1.1b_UiRev_finetuned_V3

MLX
Safetensors
openelm
trl
sft
Generated from Trainer
custom_code
Model card Files Community
1
Paramstr commited on
Commit
4c56cb5
•
1 Parent(s): ee4968b

9b735fbe7433c2277099e22854a378930dfcd6f5e2977151d85b01e20e53db7b

Browse files
Files changed (7) hide show
  1. README.md +29 -0
  2. config.json +128 -0
  3. model.safetensors +3 -0
  4. model.safetensors.index.json +459 -0
  5. special_tokens_map.json +23 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +41 -0
README.md ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ tags:
4
+ - trl
5
+ - sft
6
+ - generated_from_trainer
7
+ - mlx
8
+ base_model: apple/OpenELM-1_1B-Instruct
9
+ model-index:
10
+ - name: OpenELM_1.1b_UiRev_finetuned_V2
11
+ results: []
12
+ ---
13
+
14
+ # Paramstr/MLX_OpenELM_1.1b_UiRev_finetuned_V3
15
+
16
+ The Model [Paramstr/MLX_OpenELM_1.1b_UiRev_finetuned_V3](https://huggingface.co/Paramstr/MLX_OpenELM_1.1b_UiRev_finetuned_V3) was converted to MLX format from [Paramstr/OpenELM_1.1b_UiRev_finetuned_V2](https://huggingface.co/Paramstr/OpenELM_1.1b_UiRev_finetuned_V2) using mlx-lm version **0.14.2**.
17
+
18
+ ## Use with mlx
19
+
20
+ ```bash
21
+ pip install mlx-lm
22
+ ```
23
+
24
+ ```python
25
+ from mlx_lm import load, generate
26
+
27
+ model, tokenizer = load("Paramstr/MLX_OpenELM_1.1b_UiRev_finetuned_V3")
28
+ response = generate(model, tokenizer, prompt="hello", verbose=True)
29
+ ```
config.json ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_fn_name": "swish",
3
+ "architectures": [
4
+ "OpenELMForCausalLM"
5
+ ],
6
+ "auto_map": {
7
+ "AutoConfig": "apple/OpenELM-1_1B-Instruct--configuration_openelm.OpenELMConfig",
8
+ "AutoModelForCausalLM": "apple/OpenELM-1_1B-Instruct--modeling_openelm.OpenELMForCausalLM"
9
+ },
10
+ "bos_token_id": 1,
11
+ "eos_token_id": 2,
12
+ "ffn_dim_divisor": 256,
13
+ "ffn_multipliers": [
14
+ 0.5,
15
+ 0.63,
16
+ 0.76,
17
+ 0.89,
18
+ 1.02,
19
+ 1.15,
20
+ 1.28,
21
+ 1.41,
22
+ 1.54,
23
+ 1.67,
24
+ 1.8,
25
+ 1.93,
26
+ 2.06,
27
+ 2.19,
28
+ 2.31,
29
+ 2.44,
30
+ 2.57,
31
+ 2.7,
32
+ 2.83,
33
+ 2.96,
34
+ 3.09,
35
+ 3.22,
36
+ 3.35,
37
+ 3.48,
38
+ 3.61,
39
+ 3.74,
40
+ 3.87,
41
+ 4.0
42
+ ],
43
+ "ffn_with_glu": true,
44
+ "head_dim": 64,
45
+ "initializer_range": 0.02,
46
+ "max_context_length": 2048,
47
+ "model_dim": 2048,
48
+ "model_type": "openelm",
49
+ "normalization_layer_name": "rms_norm",
50
+ "normalize_qk_projections": true,
51
+ "num_gqa_groups": 4,
52
+ "num_kv_heads": [
53
+ 4,
54
+ 4,
55
+ 4,
56
+ 5,
57
+ 5,
58
+ 5,
59
+ 5,
60
+ 5,
61
+ 5,
62
+ 5,
63
+ 6,
64
+ 6,
65
+ 6,
66
+ 6,
67
+ 6,
68
+ 6,
69
+ 6,
70
+ 6,
71
+ 7,
72
+ 7,
73
+ 7,
74
+ 7,
75
+ 7,
76
+ 7,
77
+ 8,
78
+ 8,
79
+ 8,
80
+ 8
81
+ ],
82
+ "num_query_heads": [
83
+ 16,
84
+ 16,
85
+ 16,
86
+ 20,
87
+ 20,
88
+ 20,
89
+ 20,
90
+ 20,
91
+ 20,
92
+ 20,
93
+ 24,
94
+ 24,
95
+ 24,
96
+ 24,
97
+ 24,
98
+ 24,
99
+ 24,
100
+ 24,
101
+ 28,
102
+ 28,
103
+ 28,
104
+ 28,
105
+ 28,
106
+ 28,
107
+ 32,
108
+ 32,
109
+ 32,
110
+ 32
111
+ ],
112
+ "num_transformer_layers": 28,
113
+ "qkv_multipliers": [
114
+ 0.5,
115
+ 1.0
116
+ ],
117
+ "quantization": {
118
+ "group_size": 64,
119
+ "bits": 4
120
+ },
121
+ "rope_freq_constant": 10000,
122
+ "rope_max_length": 4096,
123
+ "share_input_output_layers": true,
124
+ "torch_dtype": "float32",
125
+ "transformers_version": "4.40.1",
126
+ "use_cache": true,
127
+ "vocab_size": 32000
128
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83851b0c2e163eea7b1c7f133a45b64b13b670ee4dc0ada792043eab5a0e2397
3
+ size 607662906
model.safetensors.index.json ADDED
@@ -0,0 +1,459 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 607611904
4
+ },
5
+ "weight_map": {
6
+ "transformer.layers.0.attn.k_norm.weight": "model.safetensors",
7
+ "transformer.layers.0.attn.out_proj.biases": "model.safetensors",
8
+ "transformer.layers.0.attn.out_proj.scales": "model.safetensors",
9
+ "transformer.layers.0.attn.out_proj.weight": "model.safetensors",
10
+ "transformer.layers.0.attn.q_norm.weight": "model.safetensors",
11
+ "transformer.layers.0.attn.qkv_proj.biases": "model.safetensors",
12
+ "transformer.layers.0.attn.qkv_proj.scales": "model.safetensors",
13
+ "transformer.layers.0.attn.qkv_proj.weight": "model.safetensors",
14
+ "transformer.layers.0.attn_norm.weight": "model.safetensors",
15
+ "transformer.layers.0.ffn.proj_1.biases": "model.safetensors",
16
+ "transformer.layers.0.ffn.proj_1.scales": "model.safetensors",
17
+ "transformer.layers.0.ffn.proj_1.weight": "model.safetensors",
18
+ "transformer.layers.0.ffn.proj_2.biases": "model.safetensors",
19
+ "transformer.layers.0.ffn.proj_2.scales": "model.safetensors",
20
+ "transformer.layers.0.ffn.proj_2.weight": "model.safetensors",
21
+ "transformer.layers.0.ffn_norm.weight": "model.safetensors",
22
+ "transformer.layers.1.attn.k_norm.weight": "model.safetensors",
23
+ "transformer.layers.1.attn.out_proj.biases": "model.safetensors",
24
+ "transformer.layers.1.attn.out_proj.scales": "model.safetensors",
25
+ "transformer.layers.1.attn.out_proj.weight": "model.safetensors",
26
+ "transformer.layers.1.attn.q_norm.weight": "model.safetensors",
27
+ "transformer.layers.1.attn.qkv_proj.biases": "model.safetensors",
28
+ "transformer.layers.1.attn.qkv_proj.scales": "model.safetensors",
29
+ "transformer.layers.1.attn.qkv_proj.weight": "model.safetensors",
30
+ "transformer.layers.1.attn_norm.weight": "model.safetensors",
31
+ "transformer.layers.1.ffn.proj_1.biases": "model.safetensors",
32
+ "transformer.layers.1.ffn.proj_1.scales": "model.safetensors",
33
+ "transformer.layers.1.ffn.proj_1.weight": "model.safetensors",
34
+ "transformer.layers.1.ffn.proj_2.biases": "model.safetensors",
35
+ "transformer.layers.1.ffn.proj_2.scales": "model.safetensors",
36
+ "transformer.layers.1.ffn.proj_2.weight": "model.safetensors",
37
+ "transformer.layers.1.ffn_norm.weight": "model.safetensors",
38
+ "transformer.layers.10.attn.k_norm.weight": "model.safetensors",
39
+ "transformer.layers.10.attn.out_proj.biases": "model.safetensors",
40
+ "transformer.layers.10.attn.out_proj.scales": "model.safetensors",
41
+ "transformer.layers.10.attn.out_proj.weight": "model.safetensors",
42
+ "transformer.layers.10.attn.q_norm.weight": "model.safetensors",
43
+ "transformer.layers.10.attn.qkv_proj.biases": "model.safetensors",
44
+ "transformer.layers.10.attn.qkv_proj.scales": "model.safetensors",
45
+ "transformer.layers.10.attn.qkv_proj.weight": "model.safetensors",
46
+ "transformer.layers.10.attn_norm.weight": "model.safetensors",
47
+ "transformer.layers.10.ffn.proj_1.biases": "model.safetensors",
48
+ "transformer.layers.10.ffn.proj_1.scales": "model.safetensors",
49
+ "transformer.layers.10.ffn.proj_1.weight": "model.safetensors",
50
+ "transformer.layers.10.ffn.proj_2.biases": "model.safetensors",
51
+ "transformer.layers.10.ffn.proj_2.scales": "model.safetensors",
52
+ "transformer.layers.10.ffn.proj_2.weight": "model.safetensors",
53
+ "transformer.layers.10.ffn_norm.weight": "model.safetensors",
54
+ "transformer.layers.11.attn.k_norm.weight": "model.safetensors",
55
+ "transformer.layers.11.attn.out_proj.biases": "model.safetensors",
56
+ "transformer.layers.11.attn.out_proj.scales": "model.safetensors",
57
+ "transformer.layers.11.attn.out_proj.weight": "model.safetensors",
58
+ "transformer.layers.11.attn.q_norm.weight": "model.safetensors",
59
+ "transformer.layers.11.attn.qkv_proj.biases": "model.safetensors",
60
+ "transformer.layers.11.attn.qkv_proj.scales": "model.safetensors",
61
+ "transformer.layers.11.attn.qkv_proj.weight": "model.safetensors",
62
+ "transformer.layers.11.attn_norm.weight": "model.safetensors",
63
+ "transformer.layers.11.ffn.proj_1.biases": "model.safetensors",
64
+ "transformer.layers.11.ffn.proj_1.scales": "model.safetensors",
65
+ "transformer.layers.11.ffn.proj_1.weight": "model.safetensors",
66
+ "transformer.layers.11.ffn.proj_2.biases": "model.safetensors",
67
+ "transformer.layers.11.ffn.proj_2.scales": "model.safetensors",
68
+ "transformer.layers.11.ffn.proj_2.weight": "model.safetensors",
69
+ "transformer.layers.11.ffn_norm.weight": "model.safetensors",
70
+ "transformer.layers.12.attn.k_norm.weight": "model.safetensors",
71
+ "transformer.layers.12.attn.out_proj.biases": "model.safetensors",
72
+ "transformer.layers.12.attn.out_proj.scales": "model.safetensors",
73
+ "transformer.layers.12.attn.out_proj.weight": "model.safetensors",
74
+ "transformer.layers.12.attn.q_norm.weight": "model.safetensors",
75
+ "transformer.layers.12.attn.qkv_proj.biases": "model.safetensors",
76
+ "transformer.layers.12.attn.qkv_proj.scales": "model.safetensors",
77
+ "transformer.layers.12.attn.qkv_proj.weight": "model.safetensors",
78
+ "transformer.layers.12.attn_norm.weight": "model.safetensors",
79
+ "transformer.layers.12.ffn.proj_1.biases": "model.safetensors",
80
+ "transformer.layers.12.ffn.proj_1.scales": "model.safetensors",
81
+ "transformer.layers.12.ffn.proj_1.weight": "model.safetensors",
82
+ "transformer.layers.12.ffn.proj_2.biases": "model.safetensors",
83
+ "transformer.layers.12.ffn.proj_2.scales": "model.safetensors",
84
+ "transformer.layers.12.ffn.proj_2.weight": "model.safetensors",
85
+ "transformer.layers.12.ffn_norm.weight": "model.safetensors",
86
+ "transformer.layers.13.attn.k_norm.weight": "model.safetensors",
87
+ "transformer.layers.13.attn.out_proj.biases": "model.safetensors",
88
+ "transformer.layers.13.attn.out_proj.scales": "model.safetensors",
89
+ "transformer.layers.13.attn.out_proj.weight": "model.safetensors",
90
+ "transformer.layers.13.attn.q_norm.weight": "model.safetensors",
91
+ "transformer.layers.13.attn.qkv_proj.biases": "model.safetensors",
92
+ "transformer.layers.13.attn.qkv_proj.scales": "model.safetensors",
93
+ "transformer.layers.13.attn.qkv_proj.weight": "model.safetensors",
94
+ "transformer.layers.13.attn_norm.weight": "model.safetensors",
95
+ "transformer.layers.13.ffn.proj_1.biases": "model.safetensors",
96
+ "transformer.layers.13.ffn.proj_1.scales": "model.safetensors",
97
+ "transformer.layers.13.ffn.proj_1.weight": "model.safetensors",
98
+ "transformer.layers.13.ffn.proj_2.biases": "model.safetensors",
99
+ "transformer.layers.13.ffn.proj_2.scales": "model.safetensors",
100
+ "transformer.layers.13.ffn.proj_2.weight": "model.safetensors",
101
+ "transformer.layers.13.ffn_norm.weight": "model.safetensors",
102
+ "transformer.layers.14.attn.k_norm.weight": "model.safetensors",
103
+ "transformer.layers.14.attn.out_proj.biases": "model.safetensors",
104
+ "transformer.layers.14.attn.out_proj.scales": "model.safetensors",
105
+ "transformer.layers.14.attn.out_proj.weight": "model.safetensors",
106
+ "transformer.layers.14.attn.q_norm.weight": "model.safetensors",
107
+ "transformer.layers.14.attn.qkv_proj.biases": "model.safetensors",
108
+ "transformer.layers.14.attn.qkv_proj.scales": "model.safetensors",
109
+ "transformer.layers.14.attn.qkv_proj.weight": "model.safetensors",
110
+ "transformer.layers.14.attn_norm.weight": "model.safetensors",
111
+ "transformer.layers.14.ffn.proj_1.biases": "model.safetensors",
112
+ "transformer.layers.14.ffn.proj_1.scales": "model.safetensors",
113
+ "transformer.layers.14.ffn.proj_1.weight": "model.safetensors",
114
+ "transformer.layers.14.ffn.proj_2.biases": "model.safetensors",
115
+ "transformer.layers.14.ffn.proj_2.scales": "model.safetensors",
116
+ "transformer.layers.14.ffn.proj_2.weight": "model.safetensors",
117
+ "transformer.layers.14.ffn_norm.weight": "model.safetensors",
118
+ "transformer.layers.15.attn.k_norm.weight": "model.safetensors",
119
+ "transformer.layers.15.attn.out_proj.biases": "model.safetensors",
120
+ "transformer.layers.15.attn.out_proj.scales": "model.safetensors",
121
+ "transformer.layers.15.attn.out_proj.weight": "model.safetensors",
122
+ "transformer.layers.15.attn.q_norm.weight": "model.safetensors",
123
+ "transformer.layers.15.attn.qkv_proj.biases": "model.safetensors",
124
+ "transformer.layers.15.attn.qkv_proj.scales": "model.safetensors",
125
+ "transformer.layers.15.attn.qkv_proj.weight": "model.safetensors",
126
+ "transformer.layers.15.attn_norm.weight": "model.safetensors",
127
+ "transformer.layers.15.ffn.proj_1.biases": "model.safetensors",
128
+ "transformer.layers.15.ffn.proj_1.scales": "model.safetensors",
129
+ "transformer.layers.15.ffn.proj_1.weight": "model.safetensors",
130
+ "transformer.layers.15.ffn.proj_2.biases": "model.safetensors",
131
+ "transformer.layers.15.ffn.proj_2.scales": "model.safetensors",
132
+ "transformer.layers.15.ffn.proj_2.weight": "model.safetensors",
133
+ "transformer.layers.15.ffn_norm.weight": "model.safetensors",
134
+ "transformer.layers.16.attn.k_norm.weight": "model.safetensors",
135
+ "transformer.layers.16.attn.out_proj.biases": "model.safetensors",
136
+ "transformer.layers.16.attn.out_proj.scales": "model.safetensors",
137
+ "transformer.layers.16.attn.out_proj.weight": "model.safetensors",
138
+ "transformer.layers.16.attn.q_norm.weight": "model.safetensors",
139
+ "transformer.layers.16.attn.qkv_proj.biases": "model.safetensors",
140
+ "transformer.layers.16.attn.qkv_proj.scales": "model.safetensors",
141
+ "transformer.layers.16.attn.qkv_proj.weight": "model.safetensors",
142
+ "transformer.layers.16.attn_norm.weight": "model.safetensors",
143
+ "transformer.layers.16.ffn.proj_1.biases": "model.safetensors",
144
+ "transformer.layers.16.ffn.proj_1.scales": "model.safetensors",
145
+ "transformer.layers.16.ffn.proj_1.weight": "model.safetensors",
146
+ "transformer.layers.16.ffn.proj_2.biases": "model.safetensors",
147
+ "transformer.layers.16.ffn.proj_2.scales": "model.safetensors",
148
+ "transformer.layers.16.ffn.proj_2.weight": "model.safetensors",
149
+ "transformer.layers.16.ffn_norm.weight": "model.safetensors",
150
+ "transformer.layers.17.attn.k_norm.weight": "model.safetensors",
151
+ "transformer.layers.17.attn.out_proj.biases": "model.safetensors",
152
+ "transformer.layers.17.attn.out_proj.scales": "model.safetensors",
153
+ "transformer.layers.17.attn.out_proj.weight": "model.safetensors",
154
+ "transformer.layers.17.attn.q_norm.weight": "model.safetensors",
155
+ "transformer.layers.17.attn.qkv_proj.biases": "model.safetensors",
156
+ "transformer.layers.17.attn.qkv_proj.scales": "model.safetensors",
157
+ "transformer.layers.17.attn.qkv_proj.weight": "model.safetensors",
158
+ "transformer.layers.17.attn_norm.weight": "model.safetensors",
159
+ "transformer.layers.17.ffn.proj_1.biases": "model.safetensors",
160
+ "transformer.layers.17.ffn.proj_1.scales": "model.safetensors",
161
+ "transformer.layers.17.ffn.proj_1.weight": "model.safetensors",
162
+ "transformer.layers.17.ffn.proj_2.biases": "model.safetensors",
163
+ "transformer.layers.17.ffn.proj_2.scales": "model.safetensors",
164
+ "transformer.layers.17.ffn.proj_2.weight": "model.safetensors",
165
+ "transformer.layers.17.ffn_norm.weight": "model.safetensors",
166
+ "transformer.layers.18.attn.k_norm.weight": "model.safetensors",
167
+ "transformer.layers.18.attn.out_proj.biases": "model.safetensors",
168
+ "transformer.layers.18.attn.out_proj.scales": "model.safetensors",
169
+ "transformer.layers.18.attn.out_proj.weight": "model.safetensors",
170
+ "transformer.layers.18.attn.q_norm.weight": "model.safetensors",
171
+ "transformer.layers.18.attn.qkv_proj.biases": "model.safetensors",
172
+ "transformer.layers.18.attn.qkv_proj.scales": "model.safetensors",
173
+ "transformer.layers.18.attn.qkv_proj.weight": "model.safetensors",
174
+ "transformer.layers.18.attn_norm.weight": "model.safetensors",
175
+ "transformer.layers.18.ffn.proj_1.biases": "model.safetensors",
176
+ "transformer.layers.18.ffn.proj_1.scales": "model.safetensors",
177
+ "transformer.layers.18.ffn.proj_1.weight": "model.safetensors",
178
+ "transformer.layers.18.ffn.proj_2.biases": "model.safetensors",
179
+ "transformer.layers.18.ffn.proj_2.scales": "model.safetensors",
180
+ "transformer.layers.18.ffn.proj_2.weight": "model.safetensors",
181
+ "transformer.layers.18.ffn_norm.weight": "model.safetensors",
182
+ "transformer.layers.19.attn.k_norm.weight": "model.safetensors",
183
+ "transformer.layers.19.attn.out_proj.biases": "model.safetensors",
184
+ "transformer.layers.19.attn.out_proj.scales": "model.safetensors",
185
+ "transformer.layers.19.attn.out_proj.weight": "model.safetensors",
186
+ "transformer.layers.19.attn.q_norm.weight": "model.safetensors",
187
+ "transformer.layers.19.attn.qkv_proj.biases": "model.safetensors",
188
+ "transformer.layers.19.attn.qkv_proj.scales": "model.safetensors",
189
+ "transformer.layers.19.attn.qkv_proj.weight": "model.safetensors",
190
+ "transformer.layers.19.attn_norm.weight": "model.safetensors",
191
+ "transformer.layers.19.ffn.proj_1.biases": "model.safetensors",
192
+ "transformer.layers.19.ffn.proj_1.scales": "model.safetensors",
193
+ "transformer.layers.19.ffn.proj_1.weight": "model.safetensors",
194
+ "transformer.layers.19.ffn.proj_2.biases": "model.safetensors",
195
+ "transformer.layers.19.ffn.proj_2.scales": "model.safetensors",
196
+ "transformer.layers.19.ffn.proj_2.weight": "model.safetensors",
197
+ "transformer.layers.19.ffn_norm.weight": "model.safetensors",
198
+ "transformer.layers.2.attn.k_norm.weight": "model.safetensors",
199
+ "transformer.layers.2.attn.out_proj.biases": "model.safetensors",
200
+ "transformer.layers.2.attn.out_proj.scales": "model.safetensors",
201
+ "transformer.layers.2.attn.out_proj.weight": "model.safetensors",
202
+ "transformer.layers.2.attn.q_norm.weight": "model.safetensors",
203
+ "transformer.layers.2.attn.qkv_proj.biases": "model.safetensors",
204
+ "transformer.layers.2.attn.qkv_proj.scales": "model.safetensors",
205
+ "transformer.layers.2.attn.qkv_proj.weight": "model.safetensors",
206
+ "transformer.layers.2.attn_norm.weight": "model.safetensors",
207
+ "transformer.layers.2.ffn.proj_1.biases": "model.safetensors",
208
+ "transformer.layers.2.ffn.proj_1.scales": "model.safetensors",
209
+ "transformer.layers.2.ffn.proj_1.weight": "model.safetensors",
210
+ "transformer.layers.2.ffn.proj_2.biases": "model.safetensors",
211
+ "transformer.layers.2.ffn.proj_2.scales": "model.safetensors",
212
+ "transformer.layers.2.ffn.proj_2.weight": "model.safetensors",
213
+ "transformer.layers.2.ffn_norm.weight": "model.safetensors",
214
+ "transformer.layers.20.attn.k_norm.weight": "model.safetensors",
215
+ "transformer.layers.20.attn.out_proj.biases": "model.safetensors",
216
+ "transformer.layers.20.attn.out_proj.scales": "model.safetensors",
217
+ "transformer.layers.20.attn.out_proj.weight": "model.safetensors",
218
+ "transformer.layers.20.attn.q_norm.weight": "model.safetensors",
219
+ "transformer.layers.20.attn.qkv_proj.biases": "model.safetensors",
220
+ "transformer.layers.20.attn.qkv_proj.scales": "model.safetensors",
221
+ "transformer.layers.20.attn.qkv_proj.weight": "model.safetensors",
222
+ "transformer.layers.20.attn_norm.weight": "model.safetensors",
223
+ "transformer.layers.20.ffn.proj_1.biases": "model.safetensors",
224
+ "transformer.layers.20.ffn.proj_1.scales": "model.safetensors",
225
+ "transformer.layers.20.ffn.proj_1.weight": "model.safetensors",
226
+ "transformer.layers.20.ffn.proj_2.biases": "model.safetensors",
227
+ "transformer.layers.20.ffn.proj_2.scales": "model.safetensors",
228
+ "transformer.layers.20.ffn.proj_2.weight": "model.safetensors",
229
+ "transformer.layers.20.ffn_norm.weight": "model.safetensors",
230
+ "transformer.layers.21.attn.k_norm.weight": "model.safetensors",
231
+ "transformer.layers.21.attn.out_proj.biases": "model.safetensors",
232
+ "transformer.layers.21.attn.out_proj.scales": "model.safetensors",
233
+ "transformer.layers.21.attn.out_proj.weight": "model.safetensors",
234
+ "transformer.layers.21.attn.q_norm.weight": "model.safetensors",
235
+ "transformer.layers.21.attn.qkv_proj.biases": "model.safetensors",
236
+ "transformer.layers.21.attn.qkv_proj.scales": "model.safetensors",
237
+ "transformer.layers.21.attn.qkv_proj.weight": "model.safetensors",
238
+ "transformer.layers.21.attn_norm.weight": "model.safetensors",
239
+ "transformer.layers.21.ffn.proj_1.biases": "model.safetensors",
240
+ "transformer.layers.21.ffn.proj_1.scales": "model.safetensors",
241
+ "transformer.layers.21.ffn.proj_1.weight": "model.safetensors",
242
+ "transformer.layers.21.ffn.proj_2.biases": "model.safetensors",
243
+ "transformer.layers.21.ffn.proj_2.scales": "model.safetensors",
244
+ "transformer.layers.21.ffn.proj_2.weight": "model.safetensors",
245
+ "transformer.layers.21.ffn_norm.weight": "model.safetensors",
246
+ "transformer.layers.22.attn.k_norm.weight": "model.safetensors",
247
+ "transformer.layers.22.attn.out_proj.biases": "model.safetensors",
248
+ "transformer.layers.22.attn.out_proj.scales": "model.safetensors",
249
+ "transformer.layers.22.attn.out_proj.weight": "model.safetensors",
250
+ "transformer.layers.22.attn.q_norm.weight": "model.safetensors",
251
+ "transformer.layers.22.attn.qkv_proj.biases": "model.safetensors",
252
+ "transformer.layers.22.attn.qkv_proj.scales": "model.safetensors",
253
+ "transformer.layers.22.attn.qkv_proj.weight": "model.safetensors",
254
+ "transformer.layers.22.attn_norm.weight": "model.safetensors",
255
+ "transformer.layers.22.ffn.proj_1.biases": "model.safetensors",
256
+ "transformer.layers.22.ffn.proj_1.scales": "model.safetensors",
257
+ "transformer.layers.22.ffn.proj_1.weight": "model.safetensors",
258
+ "transformer.layers.22.ffn.proj_2.biases": "model.safetensors",
259
+ "transformer.layers.22.ffn.proj_2.scales": "model.safetensors",
260
+ "transformer.layers.22.ffn.proj_2.weight": "model.safetensors",
261
+ "transformer.layers.22.ffn_norm.weight": "model.safetensors",
262
+ "transformer.layers.23.attn.k_norm.weight": "model.safetensors",
263
+ "transformer.layers.23.attn.out_proj.biases": "model.safetensors",
264
+ "transformer.layers.23.attn.out_proj.scales": "model.safetensors",
265
+ "transformer.layers.23.attn.out_proj.weight": "model.safetensors",
266
+ "transformer.layers.23.attn.q_norm.weight": "model.safetensors",
267
+ "transformer.layers.23.attn.qkv_proj.biases": "model.safetensors",
268
+ "transformer.layers.23.attn.qkv_proj.scales": "model.safetensors",
269
+ "transformer.layers.23.attn.qkv_proj.weight": "model.safetensors",
270
+ "transformer.layers.23.attn_norm.weight": "model.safetensors",
271
+ "transformer.layers.23.ffn.proj_1.biases": "model.safetensors",
272
+ "transformer.layers.23.ffn.proj_1.scales": "model.safetensors",
273
+ "transformer.layers.23.ffn.proj_1.weight": "model.safetensors",
274
+ "transformer.layers.23.ffn.proj_2.biases": "model.safetensors",
275
+ "transformer.layers.23.ffn.proj_2.scales": "model.safetensors",
276
+ "transformer.layers.23.ffn.proj_2.weight": "model.safetensors",
277
+ "transformer.layers.23.ffn_norm.weight": "model.safetensors",
278
+ "transformer.layers.24.attn.k_norm.weight": "model.safetensors",
279
+ "transformer.layers.24.attn.out_proj.biases": "model.safetensors",
280
+ "transformer.layers.24.attn.out_proj.scales": "model.safetensors",
281
+ "transformer.layers.24.attn.out_proj.weight": "model.safetensors",
282
+ "transformer.layers.24.attn.q_norm.weight": "model.safetensors",
283
+ "transformer.layers.24.attn.qkv_proj.biases": "model.safetensors",
284
+ "transformer.layers.24.attn.qkv_proj.scales": "model.safetensors",
285
+ "transformer.layers.24.attn.qkv_proj.weight": "model.safetensors",
286
+ "transformer.layers.24.attn_norm.weight": "model.safetensors",
287
+ "transformer.layers.24.ffn.proj_1.biases": "model.safetensors",
288
+ "transformer.layers.24.ffn.proj_1.scales": "model.safetensors",
289
+ "transformer.layers.24.ffn.proj_1.weight": "model.safetensors",
290
+ "transformer.layers.24.ffn.proj_2.biases": "model.safetensors",
291
+ "transformer.layers.24.ffn.proj_2.scales": "model.safetensors",
292
+ "transformer.layers.24.ffn.proj_2.weight": "model.safetensors",
293
+ "transformer.layers.24.ffn_norm.weight": "model.safetensors",
294
+ "transformer.layers.25.attn.k_norm.weight": "model.safetensors",
295
+ "transformer.layers.25.attn.out_proj.biases": "model.safetensors",
296
+ "transformer.layers.25.attn.out_proj.scales": "model.safetensors",
297
+ "transformer.layers.25.attn.out_proj.weight": "model.safetensors",
298
+ "transformer.layers.25.attn.q_norm.weight": "model.safetensors",
299
+ "transformer.layers.25.attn.qkv_proj.biases": "model.safetensors",
300
+ "transformer.layers.25.attn.qkv_proj.scales": "model.safetensors",
301
+ "transformer.layers.25.attn.qkv_proj.weight": "model.safetensors",
302
+ "transformer.layers.25.attn_norm.weight": "model.safetensors",
303
+ "transformer.layers.25.ffn.proj_1.biases": "model.safetensors",
304
+ "transformer.layers.25.ffn.proj_1.scales": "model.safetensors",
305
+ "transformer.layers.25.ffn.proj_1.weight": "model.safetensors",
306
+ "transformer.layers.25.ffn.proj_2.biases": "model.safetensors",
307
+ "transformer.layers.25.ffn.proj_2.scales": "model.safetensors",
308
+ "transformer.layers.25.ffn.proj_2.weight": "model.safetensors",
309
+ "transformer.layers.25.ffn_norm.weight": "model.safetensors",
310
+ "transformer.layers.26.attn.k_norm.weight": "model.safetensors",
311
+ "transformer.layers.26.attn.out_proj.biases": "model.safetensors",
312
+ "transformer.layers.26.attn.out_proj.scales": "model.safetensors",
313
+ "transformer.layers.26.attn.out_proj.weight": "model.safetensors",
314
+ "transformer.layers.26.attn.q_norm.weight": "model.safetensors",
315
+ "transformer.layers.26.attn.qkv_proj.biases": "model.safetensors",
316
+ "transformer.layers.26.attn.qkv_proj.scales": "model.safetensors",
317
+ "transformer.layers.26.attn.qkv_proj.weight": "model.safetensors",
318
+ "transformer.layers.26.attn_norm.weight": "model.safetensors",
319
+ "transformer.layers.26.ffn.proj_1.biases": "model.safetensors",
320
+ "transformer.layers.26.ffn.proj_1.scales": "model.safetensors",
321
+ "transformer.layers.26.ffn.proj_1.weight": "model.safetensors",
322
+ "transformer.layers.26.ffn.proj_2.biases": "model.safetensors",
323
+ "transformer.layers.26.ffn.proj_2.scales": "model.safetensors",
324
+ "transformer.layers.26.ffn.proj_2.weight": "model.safetensors",
325
+ "transformer.layers.26.ffn_norm.weight": "model.safetensors",
326
+ "transformer.layers.27.attn.k_norm.weight": "model.safetensors",
327
+ "transformer.layers.27.attn.out_proj.biases": "model.safetensors",
328
+ "transformer.layers.27.attn.out_proj.scales": "model.safetensors",
329
+ "transformer.layers.27.attn.out_proj.weight": "model.safetensors",
330
+ "transformer.layers.27.attn.q_norm.weight": "model.safetensors",
331
+ "transformer.layers.27.attn.qkv_proj.biases": "model.safetensors",
332
+ "transformer.layers.27.attn.qkv_proj.scales": "model.safetensors",
333
+ "transformer.layers.27.attn.qkv_proj.weight": "model.safetensors",
334
+ "transformer.layers.27.attn_norm.weight": "model.safetensors",
335
+ "transformer.layers.27.ffn.proj_1.biases": "model.safetensors",
336
+ "transformer.layers.27.ffn.proj_1.scales": "model.safetensors",
337
+ "transformer.layers.27.ffn.proj_1.weight": "model.safetensors",
338
+ "transformer.layers.27.ffn.proj_2.biases": "model.safetensors",
339
+ "transformer.layers.27.ffn.proj_2.scales": "model.safetensors",
340
+ "transformer.layers.27.ffn.proj_2.weight": "model.safetensors",
341
+ "transformer.layers.27.ffn_norm.weight": "model.safetensors",
342
+ "transformer.layers.3.attn.k_norm.weight": "model.safetensors",
343
+ "transformer.layers.3.attn.out_proj.biases": "model.safetensors",
344
+ "transformer.layers.3.attn.out_proj.scales": "model.safetensors",
345
+ "transformer.layers.3.attn.out_proj.weight": "model.safetensors",
346
+ "transformer.layers.3.attn.q_norm.weight": "model.safetensors",
347
+ "transformer.layers.3.attn.qkv_proj.biases": "model.safetensors",
348
+ "transformer.layers.3.attn.qkv_proj.scales": "model.safetensors",
349
+ "transformer.layers.3.attn.qkv_proj.weight": "model.safetensors",
350
+ "transformer.layers.3.attn_norm.weight": "model.safetensors",
351
+ "transformer.layers.3.ffn.proj_1.biases": "model.safetensors",
352
+ "transformer.layers.3.ffn.proj_1.scales": "model.safetensors",
353
+ "transformer.layers.3.ffn.proj_1.weight": "model.safetensors",
354
+ "transformer.layers.3.ffn.proj_2.biases": "model.safetensors",
355
+ "transformer.layers.3.ffn.proj_2.scales": "model.safetensors",
356
+ "transformer.layers.3.ffn.proj_2.weight": "model.safetensors",
357
+ "transformer.layers.3.ffn_norm.weight": "model.safetensors",
358
+ "transformer.layers.4.attn.k_norm.weight": "model.safetensors",
359
+ "transformer.layers.4.attn.out_proj.biases": "model.safetensors",
360
+ "transformer.layers.4.attn.out_proj.scales": "model.safetensors",
361
+ "transformer.layers.4.attn.out_proj.weight": "model.safetensors",
362
+ "transformer.layers.4.attn.q_norm.weight": "model.safetensors",
363
+ "transformer.layers.4.attn.qkv_proj.biases": "model.safetensors",
364
+ "transformer.layers.4.attn.qkv_proj.scales": "model.safetensors",
365
+ "transformer.layers.4.attn.qkv_proj.weight": "model.safetensors",
366
+ "transformer.layers.4.attn_norm.weight": "model.safetensors",
367
+ "transformer.layers.4.ffn.proj_1.biases": "model.safetensors",
368
+ "transformer.layers.4.ffn.proj_1.scales": "model.safetensors",
369
+ "transformer.layers.4.ffn.proj_1.weight": "model.safetensors",
370
+ "transformer.layers.4.ffn.proj_2.biases": "model.safetensors",
371
+ "transformer.layers.4.ffn.proj_2.scales": "model.safetensors",
372
+ "transformer.layers.4.ffn.proj_2.weight": "model.safetensors",
373
+ "transformer.layers.4.ffn_norm.weight": "model.safetensors",
374
+ "transformer.layers.5.attn.k_norm.weight": "model.safetensors",
375
+ "transformer.layers.5.attn.out_proj.biases": "model.safetensors",
376
+ "transformer.layers.5.attn.out_proj.scales": "model.safetensors",
377
+ "transformer.layers.5.attn.out_proj.weight": "model.safetensors",
378
+ "transformer.layers.5.attn.q_norm.weight": "model.safetensors",
379
+ "transformer.layers.5.attn.qkv_proj.biases": "model.safetensors",
380
+ "transformer.layers.5.attn.qkv_proj.scales": "model.safetensors",
381
+ "transformer.layers.5.attn.qkv_proj.weight": "model.safetensors",
382
+ "transformer.layers.5.attn_norm.weight": "model.safetensors",
383
+ "transformer.layers.5.ffn.proj_1.biases": "model.safetensors",
384
+ "transformer.layers.5.ffn.proj_1.scales": "model.safetensors",
385
+ "transformer.layers.5.ffn.proj_1.weight": "model.safetensors",
386
+ "transformer.layers.5.ffn.proj_2.biases": "model.safetensors",
387
+ "transformer.layers.5.ffn.proj_2.scales": "model.safetensors",
388
+ "transformer.layers.5.ffn.proj_2.weight": "model.safetensors",
389
+ "transformer.layers.5.ffn_norm.weight": "model.safetensors",
390
+ "transformer.layers.6.attn.k_norm.weight": "model.safetensors",
391
+ "transformer.layers.6.attn.out_proj.biases": "model.safetensors",
392
+ "transformer.layers.6.attn.out_proj.scales": "model.safetensors",
393
+ "transformer.layers.6.attn.out_proj.weight": "model.safetensors",
394
+ "transformer.layers.6.attn.q_norm.weight": "model.safetensors",
395
+ "transformer.layers.6.attn.qkv_proj.biases": "model.safetensors",
396
+ "transformer.layers.6.attn.qkv_proj.scales": "model.safetensors",
397
+ "transformer.layers.6.attn.qkv_proj.weight": "model.safetensors",
398
+ "transformer.layers.6.attn_norm.weight": "model.safetensors",
399
+ "transformer.layers.6.ffn.proj_1.biases": "model.safetensors",
400
+ "transformer.layers.6.ffn.proj_1.scales": "model.safetensors",
401
+ "transformer.layers.6.ffn.proj_1.weight": "model.safetensors",
402
+ "transformer.layers.6.ffn.proj_2.biases": "model.safetensors",
403
+ "transformer.layers.6.ffn.proj_2.scales": "model.safetensors",
404
+ "transformer.layers.6.ffn.proj_2.weight": "model.safetensors",
405
+ "transformer.layers.6.ffn_norm.weight": "model.safetensors",
406
+ "transformer.layers.7.attn.k_norm.weight": "model.safetensors",
407
+ "transformer.layers.7.attn.out_proj.biases": "model.safetensors",
408
+ "transformer.layers.7.attn.out_proj.scales": "model.safetensors",
409
+ "transformer.layers.7.attn.out_proj.weight": "model.safetensors",
410
+ "transformer.layers.7.attn.q_norm.weight": "model.safetensors",
411
+ "transformer.layers.7.attn.qkv_proj.biases": "model.safetensors",
412
+ "transformer.layers.7.attn.qkv_proj.scales": "model.safetensors",
413
+ "transformer.layers.7.attn.qkv_proj.weight": "model.safetensors",
414
+ "transformer.layers.7.attn_norm.weight": "model.safetensors",
415
+ "transformer.layers.7.ffn.proj_1.biases": "model.safetensors",
416
+ "transformer.layers.7.ffn.proj_1.scales": "model.safetensors",
417
+ "transformer.layers.7.ffn.proj_1.weight": "model.safetensors",
418
+ "transformer.layers.7.ffn.proj_2.biases": "model.safetensors",
419
+ "transformer.layers.7.ffn.proj_2.scales": "model.safetensors",
420
+ "transformer.layers.7.ffn.proj_2.weight": "model.safetensors",
421
+ "transformer.layers.7.ffn_norm.weight": "model.safetensors",
422
+ "transformer.layers.8.attn.k_norm.weight": "model.safetensors",
423
+ "transformer.layers.8.attn.out_proj.biases": "model.safetensors",
424
+ "transformer.layers.8.attn.out_proj.scales": "model.safetensors",
425
+ "transformer.layers.8.attn.out_proj.weight": "model.safetensors",
426
+ "transformer.layers.8.attn.q_norm.weight": "model.safetensors",
427
+ "transformer.layers.8.attn.qkv_proj.biases": "model.safetensors",
428
+ "transformer.layers.8.attn.qkv_proj.scales": "model.safetensors",
429
+ "transformer.layers.8.attn.qkv_proj.weight": "model.safetensors",
430
+ "transformer.layers.8.attn_norm.weight": "model.safetensors",
431
+ "transformer.layers.8.ffn.proj_1.biases": "model.safetensors",
432
+ "transformer.layers.8.ffn.proj_1.scales": "model.safetensors",
433
+ "transformer.layers.8.ffn.proj_1.weight": "model.safetensors",
434
+ "transformer.layers.8.ffn.proj_2.biases": "model.safetensors",
435
+ "transformer.layers.8.ffn.proj_2.scales": "model.safetensors",
436
+ "transformer.layers.8.ffn.proj_2.weight": "model.safetensors",
437
+ "transformer.layers.8.ffn_norm.weight": "model.safetensors",
438
+ "transformer.layers.9.attn.k_norm.weight": "model.safetensors",
439
+ "transformer.layers.9.attn.out_proj.biases": "model.safetensors",
440
+ "transformer.layers.9.attn.out_proj.scales": "model.safetensors",
441
+ "transformer.layers.9.attn.out_proj.weight": "model.safetensors",
442
+ "transformer.layers.9.attn.q_norm.weight": "model.safetensors",
443
+ "transformer.layers.9.attn.qkv_proj.biases": "model.safetensors",
444
+ "transformer.layers.9.attn.qkv_proj.scales": "model.safetensors",
445
+ "transformer.layers.9.attn.qkv_proj.weight": "model.safetensors",
446
+ "transformer.layers.9.attn_norm.weight": "model.safetensors",
447
+ "transformer.layers.9.ffn.proj_1.biases": "model.safetensors",
448
+ "transformer.layers.9.ffn.proj_1.scales": "model.safetensors",
449
+ "transformer.layers.9.ffn.proj_1.weight": "model.safetensors",
450
+ "transformer.layers.9.ffn.proj_2.biases": "model.safetensors",
451
+ "transformer.layers.9.ffn.proj_2.scales": "model.safetensors",
452
+ "transformer.layers.9.ffn.proj_2.weight": "model.safetensors",
453
+ "transformer.layers.9.ffn_norm.weight": "model.safetensors",
454
+ "transformer.norm.weight": "model.safetensors",
455
+ "transformer.token_embeddings.biases": "model.safetensors",
456
+ "transformer.token_embeddings.scales": "model.safetensors",
457
+ "transformer.token_embeddings.weight": "model.safetensors"
458
+ }
459
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "unk_token": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": false,
34
+ "model_max_length": 1000000000000000019884624838656,
35
+ "pad_token": null,
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "tokenizer_class": "LlamaTokenizer",
39
+ "unk_token": "<unk>",
40
+ "use_default_system_prompt": false
41
+ }