Commit
•
400c71a
1
Parent(s):
82c9ac4
Upload config
Browse files- README.md +1 -2
- config.json +6 -3
README.md
CHANGED
|
@@ -1,13 +1,12 @@
|
|
| 1 |
---
|
| 2 |
language:
|
| 3 |
- en
|
| 4 |
-
license: apache-2.0
|
| 5 |
library_name: transformers
|
|
|
|
| 6 |
tags:
|
| 7 |
- unsloth
|
| 8 |
- transformers
|
| 9 |
- tinyllama
|
| 10 |
-
|
| 11 |
---
|
| 12 |
|
| 13 |
# Finetune Mistral, Gemma, Llama 2-5x faster with 70% less memory via Unsloth!
|
|
|
|
| 1 |
---
|
| 2 |
language:
|
| 3 |
- en
|
|
|
|
| 4 |
library_name: transformers
|
| 5 |
+
license: apache-2.0
|
| 6 |
tags:
|
| 7 |
- unsloth
|
| 8 |
- transformers
|
| 9 |
- tinyllama
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
# Finetune Mistral, Gemma, Llama 2-5x faster with 70% less memory via Unsloth!
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
@@ -12,14 +12,17 @@
|
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"intermediate_size": 5632,
|
| 14 |
"max_position_embeddings": 2048,
|
|
|
|
| 15 |
"model_type": "llama",
|
| 16 |
"num_attention_heads": 32,
|
| 17 |
"num_hidden_layers": 22,
|
| 18 |
"num_key_value_heads": 4,
|
| 19 |
-
"pad_token_id": 0,
|
| 20 |
"pretraining_tp": 1,
|
| 21 |
"quantization_config": {
|
|
|
|
|
|
|
| 22 |
"bnb_4bit_compute_dtype": "bfloat16",
|
|
|
|
| 23 |
"bnb_4bit_quant_type": "nf4",
|
| 24 |
"bnb_4bit_use_double_quant": true,
|
| 25 |
"llm_int8_enable_fp32_cpu_offload": false,
|
|
@@ -35,7 +38,7 @@
|
|
| 35 |
"rope_theta": 10000.0,
|
| 36 |
"tie_word_embeddings": false,
|
| 37 |
"torch_dtype": "bfloat16",
|
| 38 |
-
"transformers_version": "4.
|
| 39 |
"use_cache": true,
|
| 40 |
"vocab_size": 32000
|
| 41 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "TinyLlama/TinyLlama_v1.1",
|
| 3 |
"architectures": [
|
| 4 |
"LlamaForCausalLM"
|
| 5 |
],
|
|
|
|
| 12 |
"initializer_range": 0.02,
|
| 13 |
"intermediate_size": 5632,
|
| 14 |
"max_position_embeddings": 2048,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
"model_type": "llama",
|
| 17 |
"num_attention_heads": 32,
|
| 18 |
"num_hidden_layers": 22,
|
| 19 |
"num_key_value_heads": 4,
|
|
|
|
| 20 |
"pretraining_tp": 1,
|
| 21 |
"quantization_config": {
|
| 22 |
+
"_load_in_4bit": true,
|
| 23 |
+
"_load_in_8bit": false,
|
| 24 |
"bnb_4bit_compute_dtype": "bfloat16",
|
| 25 |
+
"bnb_4bit_quant_storage": "uint8",
|
| 26 |
"bnb_4bit_quant_type": "nf4",
|
| 27 |
"bnb_4bit_use_double_quant": true,
|
| 28 |
"llm_int8_enable_fp32_cpu_offload": false,
|
|
|
|
| 38 |
"rope_theta": 10000.0,
|
| 39 |
"tie_word_embeddings": false,
|
| 40 |
"torch_dtype": "bfloat16",
|
| 41 |
+
"transformers_version": "4.44.0",
|
| 42 |
"use_cache": true,
|
| 43 |
"vocab_size": 32000
|
| 44 |
}
|