mahimairaja commited on
Commit
1625e80
1 Parent(s): 9391d28

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +9 -23
config.json CHANGED
@@ -1,21 +1,6 @@
1
- {'vocab_size': 32000,
2
- 'max_position_embeddings': 4096,
3
- 'hidden_size': 4096,
4
- 'intermediate_size': 11008,
5
- 'num_hidden_layers': 32,
6
- 'num_attention_heads': 32,
7
- 'num_key_value_heads': 32,
8
- 'hidden_act': 'silu',
9
- 'initializer_range': 0.02,
10
- 'rms_norm_eps': 1e-05,
11
- 'pretraining_tp': 1,
12
- 'use_cache': False,
13
- 'rope_scaling': None,
14
- 'return_dict': True,
15
- 'output_hidden_states': False,
16
- 'output_attentions': False,
17
  'torchscript': False,
18
- 'torch_dtype': 'float16',
19
  'use_bfloat16': False,
20
  'tf_legacy_loss': False,
21
  'pruned_heads': {},
@@ -51,10 +36,10 @@
51
  'exponential_decay_length_penalty': None,
52
  'suppress_tokens': None,
53
  'begin_suppress_tokens': None,
54
- 'architectures': ['LlamaForCausalLM'],
55
  'finetuning_task': None,
56
- 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'},
57
- 'label2id': {'LABEL_0': 0, 'LABEL_1': 1},
58
  'tokenizer_class': None,
59
  'prefix': None,
60
  'bos_token_id': 1,
@@ -64,9 +49,9 @@
64
  'decoder_start_token_id': None,
65
  'task_specific_params': None,
66
  'problem_type': None,
67
- '_name_or_path': 'mahimairaja/tweet-summarization-llama-2-finetuned',
68
  'transformers_version': '4.32.1',
69
- 'model_type': 'llama',
70
  'quantization_config': {'quant_method': <QuantizationMethod.BITS_AND_BYTES: 'bitsandbytes'>,
71
  'load_in_8bit': False,
72
  'load_in_4bit': True,
@@ -76,4 +61,5 @@
76
  'llm_int8_has_fp16_weight': False,
77
  'bnb_4bit_quant_type': 'nf4',
78
  'bnb_4bit_use_double_quant': False,
79
- 'bnb_4bit_compute_dtype': 'float16'}}
 
 
1
+ {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  'torchscript': False,
3
+ 'torch_dtype': "float16",
4
  'use_bfloat16': False,
5
  'tf_legacy_loss': False,
6
  'pruned_heads': {},
 
36
  'exponential_decay_length_penalty': None,
37
  'suppress_tokens': None,
38
  'begin_suppress_tokens': None,
39
+ 'architectures': ["LlamaForCausalLM"],
40
  'finetuning_task': None,
41
+ 'id2label': {0: "LABEL_0", 1: "LABEL_1"},
42
+ 'label2id': {"LABEL_0": 0, "LABEL_1": 1},
43
  'tokenizer_class': None,
44
  'prefix': None,
45
  'bos_token_id': 1,
 
49
  'decoder_start_token_id': None,
50
  'task_specific_params': None,
51
  'problem_type': None,
52
+ '_name_or_path': "mahimairaja/tweet-summarization-llama-2-finetuned",
53
  'transformers_version': '4.32.1',
54
+ 'model_type': "llama",
55
  'quantization_config': {'quant_method': <QuantizationMethod.BITS_AND_BYTES: 'bitsandbytes'>,
56
  'load_in_8bit': False,
57
  'load_in_4bit': True,
 
61
  'llm_int8_has_fp16_weight': False,
62
  'bnb_4bit_quant_type': 'nf4',
63
  'bnb_4bit_use_double_quant': False,
64
+ 'bnb_4bit_compute_dtype': 'float16'}}
65
+ }