 Christian Medina
|
d698a32f0f
|
feat: skip already processed shards (resume support)
|
2026-07-02 23:04:52 -04:00 |
|
|
Christian Medina
|
d1f0c81ac4
|
fix: remove quant_states reference
|
2026-07-02 22:52:49 -04:00 |
|
|
Christian Medina
|
18ef0f73ef
|
fix: don't save quant_state objects (BnB rebuilds on load)
|
2026-07-02 22:48:09 -04:00 |
|
|
Christian Medina
|
8c13be4d1f
|
fix: free GPU memory after each layer quantization
|
2026-07-02 22:43:27 -04:00 |
|
|
Christian Medina
|
b9f6ddaf50
|
fix: free GPU memory after each quantization
|
2026-07-02 22:41:52 -04:00 |
|
|
Christian Medina
|
1751d33f69
|
fix: load shard to CPU, quantize on GPU
|
2026-07-02 22:40:31 -04:00 |
|
|
Christian Medina
|
d769d39865
|
feat: add single shard test script
|
2026-07-02 22:39:25 -04:00 |
|
|
Christian Medina
|
0f624e1e75
|
fix: use bitsandbytes.functional for expert layer quantization
|
2026-07-02 22:36:02 -04:00 |
|
|
Christian Medina
|
410718bc73
|
feat: add CPU offloading for 70GB → 17GB conversion
|
2026-07-02 22:33:36 -04:00 |
|
|
Christian Medina
|
bee6683396
|
feat: simple NF4 quantization with device_map=auto (proven method)
|
2026-07-02 22:32:38 -04:00 |
|
|
Christian Medina
|
bfa569f4e9
|
fix: remove streaming from output path name
|
2026-07-02 22:06:50 -04:00 |
|
|
Christian Medina
|
959c43d44c
|
feat: true streaming NF4 quantization with safetensors
|
2026-07-02 22:05:59 -04:00 |
|
|
Christian Medina
|
52b93837f9
|
fix: use NF4 naming convention for quantized model
|
2026-07-02 22:00:15 -04:00 |
|
|
Christian Medina
|
2f997377f5
|
feat: add simple BnB 4-bit quantization script with device_map=auto
|
2026-07-02 21:57:02 -04:00 |
|
|
Christian Medina
|
d41648d4fc
|
feat: load shards directly to GPU, no CPU bottleneck
|
2026-07-02 21:44:28 -04:00 |
|
|
Christian Medina
|
82677070ff
|
fix: weights_only=False for safetensors loading
|
2026-07-02 21:41:44 -04:00 |
|
|
Christian Medina
|
aec7f405fe
|
feat: use both GPUs in parallel for streaming quantization
|
2026-07-02 21:30:35 -04:00 |
|
|
Christian Medina
|
0ea51488a7
|
feat: add streaming quantization script (process one shard at a time)
|
2026-07-02 21:28:05 -04:00 |
|
|
Christian Medina
|
170243de25
|
fix: remove GPU test, just check quantization
|
2026-07-02 21:14:34 -04:00 |
|
|
Christian Medina
|
43f5426583
|
feat: test BnB 4-bit loading and GPU transfer
|
2026-07-02 21:10:57 -04:00 |
|
|
Christian Medina
|
04a95f362f
|
fix: load bf16 model directly with BnB 4-bit (single load)
|
2026-07-02 21:09:54 -04:00 |
|
|
Christian Medina
|
1f804c95cd
|
feat: load bf16 to CPU, quantize with BnB, move to GPU
|
2026-07-02 20:58:22 -04:00 |
|
|
Christian Medina
|
a9b066246d
|
feat: add model size check script
|
2026-07-02 20:47:11 -04:00 |
|
|
Christian Medina
|
fa2f21562f
|
fix: load BnB 4-bit model to single GPU (no distribution)
|
2026-07-02 20:42:05 -04:00 |
|
|
Christian Medina
|
a23ecc49f0
|
feat: use transformers built-in BnB loading with device_map=cpu
|
2026-07-02 20:27:06 -04:00 |
|
|
Christian Medina
|
11f9c3a56c
|
fix: force quantization with dummy forward pass
|
2026-07-02 20:25:26 -04:00 |
|
|
Christian Medina
|
b624e90b02
|
fix: use BnB quantize_batch to actually quantize weights
|
2026-07-02 20:22:39 -04:00 |
|
|
Christian Medina
|
ec2b1be864
|
chore: clean up temp files
|
2026-07-02 20:18:13 -04:00 |
|
|
Christian Medina
|
3716051abc
|
fix: remove PEFT prepare (causes OOM)
|
2026-07-02 20:17:23 -04:00 |
|
|
Christian Medina
|
ac891fe25b
|
fix: replace with proper BnB 4-bit quantization script
|
2026-07-02 20:13:19 -04:00 |
|
|
Christian Medina
|
7984ae93e7
|
feat: proper BnB 4-bit quantization with PEFT prepare
|
2026-07-02 20:11:38 -04:00 |
|
|
Christian Medina
|
57b169f202
|
feat: use BnB 4-bit model for QLoRA training
|
2026-07-02 20:04:55 -04:00 |
|
|
Christian Medina
|
76ff09b140
|
feat: add BnB 4-bit quantization script
|
2026-07-02 20:04:44 -04:00 |
|
|
Christian Medina
|
3c31e4dfb0
|
feat: load bf16 model, quantize with BnB on CPU, move to GPU
|
2026-07-02 19:55:01 -04:00 |
|
|
Christian Medina
|
a0f4f644b0
|
fix: load 4-bit model with device_map=auto (let transformers distribute)
|
2026-07-02 19:33:37 -04:00 |
|
|
Christian Medina
|
20626e7a78
|
feat: manual 4-bit quantization with PEFT prepare + Linear4bit
|
2026-07-02 18:46:05 -04:00 |
|
|
Christian Medina
|
42c25962b2
|
feat: load model to CPU with BnB 4-bit, then move to GPU
|
2026-07-02 18:23:39 -04:00 |
|
|
Christian Medina
|
1c7e73e927
|
fix: load already-quantized 4-bit model directly (no BnB)
|
2026-07-02 18:20:43 -04:00 |
|
|
Christian Medina
|
199176c8d0
|
feat: load model with BnB 4-bit quantization (fit in single GPU)
|
2026-07-02 18:05:27 -04:00 |
|
|
Christian Medina
|
5a84b0eee5
|
feat: add CPU offload for optimizer states
|
2026-07-02 17:53:56 -04:00 |
|
|
Christian Medina
|
a51a744d7c
|
fix: reduce seq_length to 512 and increase gradient_accumulation
|
2026-07-02 17:53:11 -04:00 |
|
|
Christian Medina
|
691e5d3611
|
feat: use 8-bit AdamW optimizer to save VRAM
|
2026-07-02 17:53:05 -04:00 |
|
|
Christian Medina
|
92999b7783
|
fix: load model to single GPU (MoE requires all layers on same device)
|
2026-07-02 17:50:01 -04:00 |
|
|
Christian Medina
|
0456d55c8a
|
fix: disable gradient checkpointing (device mismatch with distributed MoE)
|
2026-07-02 17:29:19 -04:00 |
|
|
Christian Medina
|
d9c3781242
|
fix: quote 'no' in eval_strategy (YAML boolean issue)
|
2026-07-02 17:09:40 -04:00 |
|
|
Christian Medina
|
565ddd416e
|
fix: disable eval (no test split in dataset)
|
2026-07-02 16:49:34 -04:00 |
|
|
Christian Medina
|
dd86a4df37
|
fix: rename dataset column to 'text' for SFTTrainer compatibility
|
2026-07-02 16:39:21 -04:00 |
|
|
Christian Medina
|
36fe2c8d45
|
fix: pass dataset_text_field to SFTTrainer from config
|
2026-07-02 16:24:34 -04:00 |
|
|
Christian Medina
|
728073ef6a
|
fix: use 'output' column (not 'text')
|
2026-07-02 16:14:53 -04:00 |
|
|
Christian Medina
|
0752c03b83
|
fix: use absolute path for dataset (no ~ expansion)
|
2026-07-02 16:04:21 -04:00 |
|
|
Christian Medina
|
9e1c5120fb
|
fix: single-process training, remove FSDP (model pre-distributed via device_map)
|
2026-07-02 15:53:51 -04:00 |
|
|
Christian Medina
|
b8214f64cb
|
fix: use 4-bit model, remove DeepSpeed, fix dataset path
|
2026-07-02 15:23:26 -04:00 |
|
|
Christian Medina
|
4348323116
|
feat: use accelerate device_map for DISTRIBUTED model loading (Test 7 method)
|
2026-07-02 15:19:04 -04:00 |
|
|
Christian Medina
|
48e44cdf1c
|
feat: add Test 11 - PEFT prepare + manual 4-bit quantization
|
2026-07-02 14:30:19 -04:00 |
|
|
Christian Medina
|
da8658ae73
|
fix: switch to CompressedTensors 4-bit checkpoint (BnB on CPU not working)
|
2026-07-02 14:17:52 -04:00 |
|
|
Christian Medina
|
81dea9b9c8
|
docs: add model class verification to test output
|
2026-07-02 13:59:25 -04:00 |
|
|
Christian Medina
|
f608c7656f
|
feat: dynamically detect layer names from model config
|
2026-07-02 13:56:46 -04:00 |
|
|
Christian Medina
|
a55c39c818
|
fix: use device_map=cpu with BnB config for quantization
|
2026-07-02 13:54:19 -04:00 |
|
|
Christian Medina
|
3480dd9fbd
|
docs: add model inspection script and comment failing tests
|
2026-07-02 13:53:45 -04:00 |
|
|
Christian Medina
|
048e68f91a
|
refactor: comment out failing tests, add Test 7-10 variations
|
2026-07-02 13:45:17 -04:00 |
|
|
Christian Medina
|
e2b351e04c
|
fix: actually quantize model with quantize_4bit()
|
2026-07-02 13:34:33 -04:00 |
|
|
Christian Medina
|
aff91c990b
|
feat: add Test 6 - bf16 to CPU → BnB 4-bit quantize → GPU
|
2026-07-02 13:26:50 -04:00 |
|
|
Christian Medina
|
ffd1d29c35
|
fix: use bf16 model with BnB 4-bit (ON-THE-FLY quantization)
|
2026-07-02 13:20:26 -04:00 |
|
|
Christian Medina
|
48a2518b4c
|
feat: comprehensive test of 5 loading strategies
|
2026-07-02 12:32:59 -04:00 |
|
|
Christian Medina
|
3892ea7fec
|
feat: test with BnB 4-bit quantization to force proper loading
|
2026-07-02 12:23:34 -04:00 |
|
|
Christian Medina
|
c5a3b87eed
|
feat: test both strategies - device_map=auto and FSDP
|
2026-07-02 12:19:05 -04:00 |
|
|
Christian Medina
|
7f498134f2
|
fix: FSDP loads model to GPU first, then shards across GPUs
|
2026-07-02 12:17:44 -04:00 |
|
|
Christian Medina
|
f9c748706f
|
fix: use already-quantized 4-bit model for training (no BnB needed)
|
2026-07-02 12:09:59 -04:00 |
|
|
Christian Medina
|
3e26d7d37e
|
fix: use already-quantized 4-bit model for test
|
2026-07-02 12:09:49 -04:00 |
|
|
Christian Medina
|
e3ea60e6c6
|
fix: clarify distribution pattern detection in test script
|
2026-07-02 12:04:54 -04:00 |
|
|
Christian Medina
|
3049aa9b0a
|
feat: add test script to verify model loading and GPU distribution
|
2026-07-02 12:04:04 -04:00 |
|
|
Christian Medina
|
14ef1a07c4
|
fix: remove sync_module_states=True (model is on CPU)
|
2026-07-02 11:55:21 -04:00 |
|
|
Christian Medina
|
5a13ca1d1c
|
feat: FSDP training failures now fallback to next strategy
|
2026-07-02 11:48:03 -04:00 |
|
|
Christian Medina
|
ac1417567c
|
fix: fix another indentation error
|
2026-07-02 11:38:54 -04:00 |
|
|
Christian Medina
|
243223d899
|
fix: fix indentation error in strategy 3
|
2026-07-02 11:38:37 -04:00 |
|
|
Christian Medina
|
1b3f678b50
|
feat: add Strategy 1 - 4-bit QLoRA with device_map=auto (distributed across GPUs)
|
2026-07-02 11:35:09 -04:00 |
|
|
Christian Medina
|
93ac7391dc
|
fix: clarify strategy naming + init distributed process group for FSDP
|
2026-07-02 11:30:34 -04:00 |
|
|
Christian Medina
|
8cabc0e986
|
fix: correct transformer_auto_wrap_policy API signature
|
2026-07-02 11:12:31 -04:00 |
|
|
Christian Medina
|
651965e844
|
feat: manually wrap model with FSDP on CPU before trainer
|
2026-07-02 11:03:41 -04:00 |
|
|
Christian Medina
|
b056ec0306
|
fix: force FSDP1 with string value to avoid FSDP2 gather spike
|
2026-07-02 10:54:17 -04:00 |
|
|
Christian Medina
|
f1e016c8af
|
fix: set gradient_checkpointing=False (FSDP handles it)
|
2026-07-02 09:38:51 -04:00 |
|
|
Christian Medina
|
361a58addc
|
fix: simplify FSDP config, comment out mixed_precision
|
2026-07-02 09:38:10 -04:00 |
|
|
Christian Medina
|
ab1988705b
|
fix: add use_orig_params=True for LoRA/PEFT compatibility
|
2026-07-02 09:37:37 -04:00 |
|
|
Christian Medina
|
b7d680966e
|
feat: restore FSDP with SHARD_GRAD_OP + sync_module_states
|
2026-07-02 09:36:08 -04:00 |
|
|
Christian Medina
|
7b39fc3a1b
|
fix: add --mixed_precision bf16 to accelerate launch
|
2026-07-02 09:26:37 -04:00 |
|
|
Christian Medina
|
fe7df7c92f
|
fix: disable FSDP, use standard accelerate data parallelism
|
2026-07-02 09:22:28 -04:00 |
|
|
Christian Medina
|
42d61c15b0
|
fix: disable sync_module_states + add mixed precision for FSDP
|
2026-07-02 09:14:18 -04:00 |
|
|
Christian Medina
|
4eb06a625e
|
fix: add bnb_4bit_quant_storage for FSDP sharding of 4-bit weights
|
2026-07-02 09:05:00 -04:00 |
|
|
Christian Medina
|
4c9522072e
|
fix: add BitsAndBytesConfig import + remove bf16 GPU strategy
|
2026-07-02 08:54:41 -04:00 |
|
|
Christian Medina
|
70a2bf0dd6
|
fix: use FSDP1 with sync_module_states to avoid GPU gather spike
|
2026-07-02 08:47:56 -04:00 |
|
|
Christian Medina
|
c607888d8f
|
fix: load QLoRA model to CPU first, FSDP shards later
|
2026-07-02 08:26:23 -04:00 |
|
|
Christian Medina
|
dbc97138e9
|
fix: use SHARD_GRAD_OP instead of FULL_SHARD for 2-GPU setup
|
2026-07-02 00:58:41 -04:00 |
|
|
Christian Medina
|
e3ea45fe96
|
fix: remove gradient_checkpointing (FSDP activation_checkpointing handles it)
|
2026-07-02 00:41:57 -04:00 |
|
|
Christian Medina
|
dfe85c3c82
|
fix: fsdp=True instead of list + add activation_checkpointing
|
2026-07-02 00:16:33 -04:00 |
|
|
Christian Medina
|
8747111671
|
fix: update config comment to reflect Ornith-35B
|
2026-07-02 00:11:10 -04:00 |
|
|
Christian Medina
|
d5ccd28e39
|
chore: remove temporary debugging code
|
2026-07-01 23:46:34 -04:00 |
|
|
Christian Medina
|
8a5edb25e0
|
fix: specify correct transformer layer class for FSDP auto-wrap
|
2026-07-01 23:45:05 -04:00 |
|
|
Christian Medina
|
05cbd7b6b2
|
show layer class names
|
2026-07-01 23:27:25 -04:00 |
|
|
Christian Medina
|
e64515e3ca
|
refactor: switch from DeepSpeed ZeRO-3 to FSDP for QLoRA compatibility
|
2026-07-01 22:33:06 -04:00 |
|
|
Christian Medina
|
a655040db2
|
feat: use DeepSpeed ZeRO-3 for proper model sharding across GPUs
|
2026-07-01 22:08:37 -04:00 |
|