RichardErkhov
commited on
Commit
•
93fc4de
1
Parent(s):
eb2f889
uploaded model
Browse files- config.json +64 -0
config.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "tiny-random-codegen2",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"CodeGenForCausalLM"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.0,
|
8 |
+
"auto_map": {
|
9 |
+
"AutoConfig": "configuration_codegen.CodeGenConfig",
|
10 |
+
"AutoModel": "modeling_codegen.CodeGenModel",
|
11 |
+
"AutoModelForCausalLM": "modeling_codegen.CodeGenForCausalLM"
|
12 |
+
},
|
13 |
+
"bos_token_id": 1,
|
14 |
+
"embd_pdrop": 0.0,
|
15 |
+
"eos_token_id": 2,
|
16 |
+
"gradient_checkpointing": false,
|
17 |
+
"head_dim": 32,
|
18 |
+
"initializer_range": 0.02,
|
19 |
+
"layer_norm_epsilon": 1e-05,
|
20 |
+
"model_type": "codegen",
|
21 |
+
"n_ctx": 512,
|
22 |
+
"n_embd": 32,
|
23 |
+
"n_head": 8,
|
24 |
+
"n_inner": null,
|
25 |
+
"n_layer": 2,
|
26 |
+
"n_positions": 512,
|
27 |
+
"quantization_config": {
|
28 |
+
"_load_in_4bit": false,
|
29 |
+
"_load_in_8bit": true,
|
30 |
+
"bnb_4bit_compute_dtype": "float32",
|
31 |
+
"bnb_4bit_quant_storage": "uint8",
|
32 |
+
"bnb_4bit_quant_type": "fp4",
|
33 |
+
"bnb_4bit_use_double_quant": false,
|
34 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
35 |
+
"llm_int8_has_fp16_weight": false,
|
36 |
+
"llm_int8_skip_modules": null,
|
37 |
+
"llm_int8_threshold": 6.0,
|
38 |
+
"load_in_4bit": false,
|
39 |
+
"load_in_8bit": true,
|
40 |
+
"quant_method": "bitsandbytes"
|
41 |
+
},
|
42 |
+
"resid_pdrop": 0.0,
|
43 |
+
"rotary_dim": 4,
|
44 |
+
"scale_attn_weights": true,
|
45 |
+
"summary_activation": null,
|
46 |
+
"summary_first_dropout": 0.1,
|
47 |
+
"summary_proj_to_labels": true,
|
48 |
+
"summary_type": "cls_index",
|
49 |
+
"summary_use_proj": true,
|
50 |
+
"task_specific_params": {
|
51 |
+
"text-generation": {
|
52 |
+
"do_sample": true,
|
53 |
+
"max_length": 50,
|
54 |
+
"temperature": 1.0
|
55 |
+
}
|
56 |
+
},
|
57 |
+
"tie_word_embeddings": false,
|
58 |
+
"tokenizer_class": "GPT2Tokenizer",
|
59 |
+
"torch_dtype": "float16",
|
60 |
+
"transformers_version": "4.44.2",
|
61 |
+
"trust_remote_code": true,
|
62 |
+
"use_cache": true,
|
63 |
+
"vocab_size": 51200
|
64 |
+
}
|