vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

gguf-dump.py: use standard tensor name lookup. Also add tensor ID field

Browse source
This commit is contained in:
brian khuu 2024-06-10 22:19:56 +10:00
parent 3363405f13
commit b000526741
1 changed files with 39 additions and 16 deletions
Download patch file Download diff file Expand all files Collapse all files

55
gguf-py/scripts/gguf-dump.py
View file

@ -130,22 +130,44 @@ def element_count_rounded_notation(count: int) -> str:
def translate_tensor_name(name): def translate_tensor_name(name):
import re words = name.split(".")
words = re.split(r"[._]", name)
# Source: https://github.com/ggerganov/ggml/blob/master/docs/gguf.md#standardized-tensor-names
abbreviation_dictionary = { abbreviation_dictionary = {
'ffn' : 'Feed Forward', 'token_embd': 'Token embedding',
'attn' : 'Attention', 'pos_embd': 'Position embedding',
'blk' : 'Block', 'output_norm': 'Output normalization',
'norm' : 'Normalization', 'output': 'Output',
'embd' : 'Embedding', 'attn_norm': 'Attention normalization',
'attn_norm_2': 'Attention normalization',
'attn_qkv': 'Attention query-key-value',
'attn_q': 'Attention query',
'attn_k': 'Attention key',
'attn_v': 'Attention value',
'attn_output': 'Attention output',
'ffn_norm': 'Feed-forward network normalization',
'ffn_up': 'Feed-forward network "up"',
'ffn_gate': 'Feed-forward network "gate"',
'ffn_down': 'Feed-forward network "down"',
'ffn_gate_inp': 'Expert-routing layer for the Feed-forward network in Mixture of Expert models',
'ffn_gate_exp': 'Feed-forward network "gate" layer per expert in Mixture of Expert models',
'ffn_down_exp': 'Feed-forward network "down" layer per expert in Mixture of Expert models',
'ffn_up_exp': 'Feed-forward network "up" layer per expert in Mixture of Expert models',
'ssm_in': 'State space model input projections',
'ssm_conv1d': 'State space model rolling/shift',
'ssm_x': 'State space model selective parametrization',
'ssm_a': 'State space model state compression',
'ssm_d': 'State space model skip connection',
'ssm_dt': 'State space model time step',
'ssm_out': 'State space model output projection',
'blk': 'Block'
} }
expanded_words = [] expanded_words = []
for word in words: for word in words:
word_norm = word.strip().lower() word_norm = word.strip().lower()
if word_norm in abbreviation_dictionary: if word_norm in abbreviation_dictionary:
expanded_words.append(abbreviation_dictionary[word_norm]) expanded_words.append(abbreviation_dictionary[word_norm].title())
else: else:
expanded_words.append(word.title()) expanded_words.append(word.title())
@ -187,13 +209,14 @@ def dump_markdown_metadata(reader: GGUFReader, args: argparse.Namespace) -> None
if not args.no_tensors: if not args.no_tensors:
# Group tensors by their prefix and maintain order # Group tensors by their prefix and maintain order
tensor_prefix_order = [] tensor_prefix_order = []
tensor_name_to_key = {}
tensor_groups = {} tensor_groups = {}
total_elements = sum(tensor.n_elements for tensor in reader.tensors) total_elements = sum(tensor.n_elements for tensor in reader.tensors)
for tensor in reader.tensors: for key, tensor in enumerate(reader.tensors):
tensor_name = tensor.name.replace(".weight", "") tensor_components = tensor.name.split('.')
tensor_components = tensor_name.split('.')
tensor_prefix = tensor_components[0] tensor_prefix = tensor_components[0]
if tensor_prefix == 'blk': if tensor_prefix == 'blk':
tensor_prefix = f"{tensor_components[0]}.{tensor_components[1]}" tensor_prefix = f"{tensor_components[0]}.{tensor_components[1]}"
@ -201,6 +224,7 @@ def dump_markdown_metadata(reader: GGUFReader, args: argparse.Namespace) -> None
tensor_groups[tensor_prefix] = [] tensor_groups[tensor_prefix] = []
tensor_prefix_order.append(tensor_prefix) tensor_prefix_order.append(tensor_prefix)
tensor_name_to_key[tensor.name] = key
tensor_groups[tensor_prefix].append(tensor) tensor_groups[tensor_prefix].append(tensor)
# Generate Markdown metadata # Generate Markdown metadata
@ -217,14 +241,13 @@ def dump_markdown_metadata(reader: GGUFReader, args: argparse.Namespace) -> None
group_elements = sum(tensor.n_elements for tensor in tensors) group_elements = sum(tensor.n_elements for tensor in tensors)
group_percentage = group_elements / total_elements * 100 group_percentage = group_elements / total_elements * 100
markdown_content += f"### {translate_tensor_name(group)} Tensor Group : {element_count_rounded_notation(group_elements)} Elements <a name=\"{group.replace('.', '_')}\"></a>\n" markdown_content += f"### {translate_tensor_name(group)} Tensor Group : {element_count_rounded_notation(group_elements)} Elements <a name=\"{group.replace('.', '_')}\"></a>\n"
markdown_content += "| Tensor Name | Human Friendly Name | Elements | Shape | Type |\n" markdown_content += "| T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type |\n"
markdown_content += "|----------------------|-------------------------------------|----------------|---------------------------------|------|\n" markdown_content += "|------|---------------------------|----------------------------------------------------|----------------|---------------------------------|------|\n"
for tensor in tensors: for tensor in tensors:
tensor_name = tensor.name.replace(".weight", "") human_friendly_name = translate_tensor_name(tensor.name.replace(".weight", ".(W)").replace(".bias", ".(B)"))
human_friendly_name = translate_tensor_name(tensor.name.replace(".weight", ""))
prettydims = ' x '.join('{0:^5}'.format(d) for d in list(tensor.shape) + [1] * (4 - len(tensor.shape))) prettydims = ' x '.join('{0:^5}'.format(d) for d in list(tensor.shape) + [1] * (4 - len(tensor.shape)))
markdown_content += f"| {tensor_name:20} | {human_friendly_name:35} | ({element_count_rounded_notation(tensor.n_elements):>4}) {tensor.n_elements:7} | [{prettydims:29}] | {tensor.tensor_type.name:4} |\n" markdown_content += f"| {tensor_name_to_key[tensor.name]:4} | {tensor.name:25} | {human_friendly_name:50} | ({element_count_rounded_notation(tensor.n_elements):>4}) {tensor.n_elements:7} | [{prettydims:29}] | {tensor.tensor_type.name:4} |\n"
markdown_content += "\n" markdown_content += "\n"
markdown_content += f"- Total elements in {group}: ({element_count_rounded_notation(group_elements):>4}) {group_elements}\n" markdown_content += f"- Total elements in {group}: ({element_count_rounded_notation(group_elements):>4}) {group_elements}\n"
markdown_content += f"- Percentage of total elements: {group_percentage:.2f}%\n" markdown_content += f"- Percentage of total elements: {group_percentage:.2f}%\n"