Views
No views yet
0.4.01base_model: mistralai/Mistral-7B-v0.1
2model_type: MistralForCausalLM
3tokenizer_type: LlamaTokenizer
4is_mistral_derived_model: true
5
6load_in_8bit: false
7load_in_4bit: false
8strict: false
9
10chat_template: chatml
11datasets:
12 - path: data/merged_all.json
13 ds_type: json
14 type: alpaca
15 conversation: chatml
16
17 - path: data/capybara_sharegpt.json
18 ds_type: json
19 type: sharegpt
20 conversation: chatml
21
22 - path: data/synthia-v1.3_sharegpt_12500.json
23 ds_type: json
24 type: sharegpt
25 conversation: chatml
26
27 - path: data/cot_alpaca_gpt4_extracted_openhermes_2.5_sharegpt.json
28 ds_type: json
29 type: sharegpt
30 conversation: chatml
31
32 - path: data/slimorca_dedup_filtered_95k_sharegpt.json
33 ds_type: json
34 type: sharegpt
35 conversation: chatml
36
37 - path: data/airoboros_3.2_without_contextual_slimorca_orca_sharegpt.json
38 ds_type: json
39 type: sharegpt
40 conversation: chatml
41
42dataset_prepared_path: last_run_prepared
43val_set_size: 0.005
44output_dir: ./Einstein-v4-model
45
46sequence_len: 8192
47sample_packing: true
48pad_to_sequence_len: true
49eval_sample_packing: false
50
51wandb_project: Einstein
52wandb_entity:
53wandb_watch:
54wandb_name:
55wandb_log_model:
56hub_model_id: Weyaxi/Einstein-v4-7B
57
58save_safetensors: true
59
60gradient_accumulation_steps: 4
61micro_batch_size: 1
62num_epochs: 1.5
63optimizer: adamw_bnb_8bit
64lr_scheduler: cosine
65learning_rate: 0.000005
66
67train_on_inputs: false
68group_by_length: false
69bf16: true
70fp16: false
71tf32: false
72
73gradient_checkpointing: true
74early_stopping_patience:
75resume_from_checkpoint:
76local_rank:
77logging_steps: 1
78xformers_attention:
79flash_attention: true
80
81warmup_steps: 10
82evals_per_epoch: 2 # changed
83eval_table_size:
84eval_table_max_new_tokens: 128
85saves_per_epoch: 4
86debug:
87
88deepspeed: zero3_bf16.json
89weight_decay: 0.0
90fsdp:
91fsdp_config:
92special_tokens:
93 bos_token: "<s>"
94 eos_token: "<|im_end|>"
95 unk_token: "<unk>"
96tokens:
97 - "<|im_start|>"
98
99resume_from_checkpoint: Einstein-v4-model/checkpoint-521
100<|im_start|>system
{system}<|im_end|>
<|im_start|>user
{user}<|im_end|>
<|im_start|>assistant
{asistant}<|im_end|>tokenizer.apply_chat_template() method:1messages = [
2 {"role": "system", "content": "You are helpful AI asistant."},
3 {"role": "user", "content": "Hello!"}
4]
5gen_input = tokenizer.apply_chat_template(message, return_tensors="pt")
6model.generate(**gen_input)| Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
|---|---|---|---|---|---|---|
| 8_0 | 8.0 | 8.0 | 8.4 GB | 9.8 GB | 11.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
| 6_5 | 6.5 | 8.0 | 7.2 GB | 8.6 GB | 10.6 GB | Very similar to 8.0, good tradeoff of size vs performance, recommended. |
| 5_0 | 5.0 | 6.0 | 6.0 GB | 7.4 GB | 9.4 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
| 4_25 | 4.25 | 6.0 | 5.3 GB | 6.7 GB | 8.7 GB | GPTQ equivalent bits per weight, slightly higher quality. |
| 3_5 | 3.5 | 6.0 | 4.7 GB | 6.1 GB | 8.1 GB | Lower quality, only use if you have to. |
| Metric | Value |
|---|---|
| Avg. | 66.62 |
| AI2 Reasoning Challenge (25-Shot) | 64.68 |
| HellaSwag (10-Shot) | 83.75 |
| MMLU (5-Shot) | 62.31 |
| TruthfulQA (0-shot) | 55.15 |
| Winogrande (5-shot) | 76.24 |
| GSM8k (5-shot) | 57.62 |
