Compare commits
20 Commits
v0.5.6
...
parth/set-
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
b4de2e9189 | ||
|
|
61a5254115 | ||
|
|
53d2cf37d2 | ||
|
|
75f88e7aac | ||
|
|
4982089c84 | ||
|
|
8c231b0826 | ||
|
|
16abd181a9 | ||
|
|
5c2f35d846 | ||
|
|
6de3227841 | ||
|
|
35e97db03b | ||
|
|
2ef3c803a1 | ||
|
|
453e4d090b | ||
|
|
ca2f9843c8 | ||
|
|
294b6f5a22 | ||
|
|
7bb356c680 | ||
|
|
021817e59a | ||
|
|
a420a453b4 | ||
|
|
42cf4db601 | ||
|
|
93a8daf285 | ||
|
|
a041b4df7c |
@@ -369,6 +369,7 @@ See the [API documentation](./docs/api.md) for all endpoints.
|
||||
- [Minima](https://github.com/dmayboroda/minima) (RAG with on-premises or fully local workflow)
|
||||
- [aidful-ollama-model-delete](https://github.com/AidfulAI/aidful-ollama-model-delete) (User interface for simplified model cleanup)
|
||||
- [Perplexica](https://github.com/ItzCrazyKns/Perplexica) (An AI-powered search engine & an open-source alternative to Perplexity AI)
|
||||
- [AI Toolkit for Visual Studio Code](https://aka.ms/ai-tooklit/ollama-docs) (Microsoft-official VSCode extension to chat, test, evaluate models with Ollama support, and use them in your AI applications.)
|
||||
|
||||
### Cloud
|
||||
|
||||
@@ -481,6 +482,7 @@ See the [API documentation](./docs/api.md) for all endpoints.
|
||||
- [GoLamify](https://github.com/prasad89/golamify)
|
||||
- [Ollama for Haskell](https://github.com/tusharad/ollama-haskell)
|
||||
- [multi-llm-ts](https://github.com/nbonamy/multi-llm-ts) (A Typescript/JavaScript library allowing access to different LLM in unified API)
|
||||
- [LlmTornado](https://github.com/lofcz/llmtornado) (C# library providing a unified interface for major FOSS & Commercial inference APIs)
|
||||
|
||||
### Mobile
|
||||
|
||||
@@ -539,4 +541,5 @@ See the [API documentation](./docs/api.md) for all endpoints.
|
||||
### Observability
|
||||
|
||||
- [OpenLIT](https://github.com/openlit/openlit) is an OpenTelemetry-native tool for monitoring Ollama Applications & GPUs using traces and metrics.
|
||||
- [HoneyHive](https://docs.honeyhive.ai/integrations/ollama) is an AI observability and evaluation platform for AI agents. Use HoneyHive to evaluate agent performance, interrogate failures, and monitor quality in production.
|
||||
- [HoneyHive](https://docs.honeyhive.ai/integrations/ollama) is an AI observability and evaluation platform for AI agents. Use HoneyHive to evaluate agent performance, interrogate failures, and monitor quality in production.
|
||||
- [Langfuse](https://langfuse.com/docs/integrations/ollama) is an open source LLM observability platform that enables teams to collaboratively monitor, evaluate and debug AI applications.
|
||||
|
||||
@@ -59,7 +59,7 @@ func getModelfileName(cmd *cobra.Command) (string, error) {
|
||||
|
||||
_, err = os.Stat(absName)
|
||||
if err != nil {
|
||||
return filename, err
|
||||
return "", err
|
||||
}
|
||||
|
||||
return absName, nil
|
||||
|
||||
@@ -279,7 +279,7 @@ func TestGetModelfileName(t *testing.T) {
|
||||
name: "no modelfile specified, no modelfile exists",
|
||||
modelfileName: "",
|
||||
fileExists: false,
|
||||
expectedName: "Modelfile",
|
||||
expectedName: "",
|
||||
expectedErr: os.ErrNotExist,
|
||||
},
|
||||
{
|
||||
@@ -293,7 +293,7 @@ func TestGetModelfileName(t *testing.T) {
|
||||
name: "modelfile specified, no modelfile exists",
|
||||
modelfileName: "crazyfile",
|
||||
fileExists: false,
|
||||
expectedName: "crazyfile",
|
||||
expectedName: "",
|
||||
expectedErr: os.ErrNotExist,
|
||||
},
|
||||
{
|
||||
|
||||
@@ -191,6 +191,8 @@ func ConvertModel(fsys fs.FS, ws io.WriteSeeker) error {
|
||||
conv = &qwen2Model{}
|
||||
case "BertModel":
|
||||
conv = &bertModel{}
|
||||
case "CohereForCausalLM":
|
||||
conv = &commandrModel{}
|
||||
default:
|
||||
return errors.New("unsupported architecture")
|
||||
}
|
||||
|
||||
76
convert/convert_commandr.go
Normal file
76
convert/convert_commandr.go
Normal file
@@ -0,0 +1,76 @@
|
||||
package convert
|
||||
|
||||
import (
|
||||
"cmp"
|
||||
|
||||
"github.com/ollama/ollama/llm"
|
||||
)
|
||||
|
||||
type commandrModel struct {
|
||||
ModelParameters
|
||||
MaxPositionEmbeddings uint32 `json:"max_position_embeddings"`
|
||||
HiddenSize uint32 `json:"hidden_size"`
|
||||
HiddenLayers uint32 `json:"num_hidden_layers"`
|
||||
IntermediateSize uint32 `json:"intermediate_size"`
|
||||
NumAttentionHeads uint32 `json:"num_attention_heads"`
|
||||
NumKeyValueHeads uint32 `json:"num_key_value_heads"`
|
||||
LayerNormEPS float32 `json:"layer_norm_eps"`
|
||||
RopeTheta float32 `json:"rope_theta"`
|
||||
UseQKNorm bool `json:"use_qk_norm"`
|
||||
MaxLength uint32 `json:"model_max_length"`
|
||||
LogitScale float32 `json:"logit_scale"`
|
||||
NCtx uint32 `json:"n_ctx"`
|
||||
}
|
||||
|
||||
var _ ModelConverter = (*commandrModel)(nil)
|
||||
|
||||
func (p *commandrModel) KV(t *Tokenizer) llm.KV {
|
||||
kv := p.ModelParameters.KV(t)
|
||||
kv["general.architecture"] = "command-r"
|
||||
kv["general.name"] = "command-r"
|
||||
kv["command-r.context_length"] = cmp.Or(p.MaxLength, p.MaxPositionEmbeddings, p.NCtx)
|
||||
kv["command-r.embedding_length"] = p.HiddenSize
|
||||
kv["command-r.block_count"] = p.HiddenLayers
|
||||
kv["command-r.feed_forward_length"] = p.IntermediateSize
|
||||
kv["command-r.attention.head_count"] = p.NumAttentionHeads
|
||||
kv["command-r.attention.head_count_kv"] = p.NumKeyValueHeads
|
||||
kv["command-r.attention.layer_norm_epsilon"] = p.LayerNormEPS
|
||||
kv["command-r.rope.freq_base"] = p.RopeTheta
|
||||
kv["command-r.max_position_embeddings"] = cmp.Or(p.MaxLength, p.MaxPositionEmbeddings)
|
||||
kv["command-r.logit_scale"] = p.LogitScale
|
||||
kv["command-r.rope.scaling.type"] = "none"
|
||||
|
||||
return kv
|
||||
}
|
||||
|
||||
func (p *commandrModel) Tensors(ts []Tensor) []llm.Tensor {
|
||||
var out []llm.Tensor
|
||||
for _, t := range ts {
|
||||
out = append(out, llm.Tensor{
|
||||
Name: t.Name(),
|
||||
Kind: t.Kind(),
|
||||
Shape: t.Shape(),
|
||||
WriterTo: t,
|
||||
})
|
||||
}
|
||||
|
||||
return out
|
||||
}
|
||||
|
||||
func (p *commandrModel) Replacements() []string {
|
||||
return []string{
|
||||
"self_attn.q_norm", "attn_q_norm",
|
||||
"self_attn.k_norm", "attn_k_norm",
|
||||
"model.layers", "blk",
|
||||
"input_layernorm", "attn_norm",
|
||||
"mlp.down_proj", "ffn_down",
|
||||
"mlp.gate_proj", "ffn_gate",
|
||||
"mlp.up_proj", "ffn_up",
|
||||
"self_attn.k_proj", "attn_k",
|
||||
"self_attn.o_proj", "attn_output",
|
||||
"self_attn.q_proj", "attn_q",
|
||||
"self_attn.v_proj", "attn_v",
|
||||
"model.norm", "output_norm",
|
||||
"model.embed_tokens", "token_embd",
|
||||
}
|
||||
}
|
||||
@@ -109,6 +109,7 @@ func TestConvertModel(t *testing.T) {
|
||||
"all-MiniLM-L6-v2",
|
||||
"gemma-2-9b-it",
|
||||
"Qwen2.5-0.5B-Instruct",
|
||||
"c4ai-command-r-v01",
|
||||
}
|
||||
|
||||
for i := range cases {
|
||||
|
||||
344
convert/testdata/c4ai-command-r-v01.json
vendored
Normal file
344
convert/testdata/c4ai-command-r-v01.json
vendored
Normal file
@@ -0,0 +1,344 @@
|
||||
{
|
||||
"general.architecture": "command-r",
|
||||
"general.name": "command-r",
|
||||
"command-r.attention.head_count": "64",
|
||||
"command-r.attention.head_count_kv": "64",
|
||||
"command-r.attention.layer_norm_epsilon": "1e-05",
|
||||
"command-r.block_count": "40",
|
||||
"command-r.context_length": "131072",
|
||||
"command-r.embedding_length": "8192",
|
||||
"command-r.feed_forward_length": "22528",
|
||||
"command-r.logit_scale": "0.0625",
|
||||
"command-r.rope.freq_base": "8e+06",
|
||||
"command-r.rope.scaling.type": "none",
|
||||
"tokenizer.ggml.add_bos_token": "true",
|
||||
"tokenizer.ggml.add_eos_token": "false",
|
||||
"tokenizer.ggml.bos_token_id": "5",
|
||||
"tokenizer.ggml.eos_token_id": "255001",
|
||||
"tokenizer.ggml.merges": "902a060cac8884a5793d2a857dd2e53a259de46c8d08c4deb243c239671e1350",
|
||||
"tokenizer.ggml.model": "gpt2",
|
||||
"tokenizer.ggml.padding_token_id": "0",
|
||||
"tokenizer.ggml.token_type": "b7a352ccd1c99d4413bcf452c2db707b0526d0e1216616b865560fab80296462",
|
||||
"tokenizer.ggml.tokens": "815ac90ff23565081522d7258f46648c8a0619eb847a9c7c31b238a9b984e4ae",
|
||||
"blk.0.attn_k.weight": "6fcfdb466f9ceb1229404ce4ec4e480751b8d00da12707a11783dad7256cb864",
|
||||
"blk.0.attn_norm.weight": "6063317f731371864049c7704a70772f1eb632194201ebdc2ed0f8e483507c72",
|
||||
"blk.0.attn_output.weight": "920f49716a1e2fc73b6794ec777947f1c122701e63ed302422ac89e90f06e9da",
|
||||
"blk.0.attn_q.weight": "ddbcd7cde197e632564ac58e4f25d9e3a8ca52917329eeb6081eb41a797932ab",
|
||||
"blk.0.attn_v.weight": "318fc02a189d87420f0cbf57f47f11e00c21ec1ed472ce0a2a895b44f7fa0fca",
|
||||
"blk.0.ffn_down.weight": "aa71975b6eb1f4c77b03d2ac4a194cf8d95718efac741bb12f0f3ff79a27f9bc",
|
||||
"blk.0.ffn_gate.weight": "42967702fa0bc738b88dc50007ace26dbe74a5a9e0978124dd093f818241a9e1",
|
||||
"blk.0.ffn_up.weight": "5282c8788b086bd30f46525e7995a17464882a72703fd27165491afdd8bfd4af",
|
||||
"blk.1.attn_k.weight": "cd248882e64fd2c3402c44790ebe12440133dc671b6893fdad0564c461973adc",
|
||||
"blk.1.attn_norm.weight": "ba84e1c8fd30af6ec94208db4078befac8c921aad3acb887812887f3282ea2be",
|
||||
"blk.1.attn_output.weight": "2efa3ef7c5666ccceb05e339b83ad680cc0d2c3ec78203f5da5959f23a80e14f",
|
||||
"blk.1.attn_q.weight": "5106f2e255358a1303c22e8b5f0ec044852bb30a866c52cabefd30017a7a6b7d",
|
||||
"blk.1.attn_v.weight": "a211a634a1a5df1d5f973645438be0461dd922210f9747c6b04e386c7f1ebe95",
|
||||
"blk.1.ffn_down.weight": "37093afe48d32c578ec956c9ed85242cd000d6aa979e60526aafa10c822dbb10",
|
||||
"blk.1.ffn_gate.weight": "469860819e9159caefb1aad0bc66db790f3393f05fd87b08e52256a7ed256543",
|
||||
"blk.1.ffn_up.weight": "736742c97d35d1a011f9cafd3c0ce947ad559bb2fba6da73c816f6bfd0fa9aeb",
|
||||
"blk.2.attn_k.weight": "92c219d92804d832ab404bd6dc7339c90877bb7cf405dd030c121f8b27757739",
|
||||
"blk.2.attn_norm.weight": "61e4466069474b76b6d1e702566420eb669faf3556b00ff7b824784aca13a2d6",
|
||||
"blk.2.attn_output.weight": "d2fb38a2b2171fd91caf037faa585a62225819aa232d86fd4f7f9d2c3c8a45e9",
|
||||
"blk.2.attn_q.weight": "f6faf5cc6844e3daa4f9f68d90f5458c64879de68a7728860e38374e30c3429d",
|
||||
"blk.2.attn_v.weight": "f340ef8f7341d987a6f37c0e9afe0aef5be67be00c0ce5f57612daf73319cce1",
|
||||
"blk.2.ffn_down.weight": "c7be61a701d779860b621b143fb6365b607bf99ec7c0f153b07908ac8120885a",
|
||||
"blk.2.ffn_gate.weight": "b64f0878187bd3392abfa4c3e8ad2f8b4c133903e54246747ff8f3b4639ad83e",
|
||||
"blk.2.ffn_up.weight": "50b11c712652e90ee7428dbb45cffebb80662ac982bc72bd9eafff361b5eb5a8",
|
||||
"blk.3.attn_k.weight": "2b7bcbe9ee5c9c630c8c8d7483887e78b73581016f4cbb6933db2a147a25f431",
|
||||
"blk.3.attn_norm.weight": "0181dac7f4eee7252980323e8032cf339bef2046ce0a16c0fd72af7c98a8a37b",
|
||||
"blk.3.attn_output.weight": "aef8843b636ce231da9e7c9acbee197883cc15df0e2887709324c6a50f16da7b",
|
||||
"blk.3.attn_q.weight": "55404130fa10e81322d33eb378aa0de31a92990ce7730f1338c0ace0406bb1b1",
|
||||
"blk.3.attn_v.weight": "76f7fb8040d82b957d689ce34fea2302a6640ad5bbaa0052ad2b7ebce270c33d",
|
||||
"blk.3.ffn_down.weight": "648628933eff3b357c3729c33c5b1ae51c28e59b9c19acd1601a2ff7c5d5d9a5",
|
||||
"blk.3.ffn_gate.weight": "6a588885d16e98d5f50ebed05af089154f680085ca9c97691e5b489088630a4a",
|
||||
"blk.3.ffn_up.weight": "e12455a1d702f4986e1a663493e3d5102b367af74d45557522002a35d63ecac2",
|
||||
"blk.4.attn_k.weight": "40d943380a8a85e4eab147934bf6e16f23cc8ab753f6636526382c074d182288",
|
||||
"blk.4.attn_norm.weight": "4ab2c098983d4599fe540eef624c4df954adb7473faebda7471ef0ba4134814c",
|
||||
"blk.4.attn_output.weight": "d14b91e40f58bf4a3c8c2eca0b12bb541de406574af39027d56f6c588a147082",
|
||||
"blk.4.attn_q.weight": "e1224960a3562107488589f883fa32414bae41712fa8dbd47c5f3e3a7801452f",
|
||||
"blk.4.attn_v.weight": "063f297bc4aa6e709fc32c4c32e35af7d07d80e83cb939b76adbba858006c03d",
|
||||
"blk.4.ffn_down.weight": "f88a18020c5e1caaa29596895eb348e76ee5bfad27ed57651a86cd8cd1f9b5aa",
|
||||
"blk.4.ffn_gate.weight": "48e7e1eed3fb52e92e61d3557dd0ec002418327090e034ce4322fd68542266f8",
|
||||
"blk.4.ffn_up.weight": "1ca8a7aa17355b6ce0d9ad5539fdad3899fa47fd359c285fbfb31f19f47bf073",
|
||||
"blk.5.attn_k.weight": "2bdf15f8e73d068d972380f25d207004cf0bf3b5bfa46946803ba6fba07d9175",
|
||||
"blk.5.attn_norm.weight": "60448d7cde6e1b6467aa31bdea012e39cdb08c88081cee7d102dca4f93f766ef",
|
||||
"blk.5.attn_output.weight": "f9f687d7c457537f9fca8a4087a59f1c3bebfaf5537b94e42c831a13224f7799",
|
||||
"blk.5.attn_q.weight": "987db7a2ad68657a92625e1980effbb1f79697c2183f2b9f3b3a0570c51b0ab9",
|
||||
"blk.5.attn_v.weight": "cf696891148f3e4783ad1d20f93462ae091eb8651c656bba9b662253b6263e02",
|
||||
"blk.5.ffn_down.weight": "c0662b0bd0929136005fb9d691fdd9b2c33867d9ce9622339a6a456b720b059a",
|
||||
"blk.5.ffn_gate.weight": "200bbdfab615d7a3a84719b6ced7751e3ce52757ef212d96f87798bc1de5e987",
|
||||
"blk.5.ffn_up.weight": "df5d23e7e035fb1b9d163da7ddfdfe38da6a37e86e96534dc02ad20f011b55b3",
|
||||
"blk.6.attn_k.weight": "c0dae2d272a7c5a2fa004bbb8475dbab362fc1f6d008e73d5a4434a9382ac6ba",
|
||||
"blk.6.attn_norm.weight": "51c57ac8b55e04354d5dca6bb9c0cf4177639d3b038e80209e33036209688f64",
|
||||
"blk.6.attn_output.weight": "229d97892c62f85bcdf431675250e01c976ad69ffa450b01fb543bf88f14a2fb",
|
||||
"blk.6.attn_q.weight": "c20e49621821bd46ed156e6823864a5bda4f317750e71ab8dc54e44eb48cf7c2",
|
||||
"blk.6.attn_v.weight": "53ceb1a2ee43fce3c7b5b33c58a9fc5ee7f44dc1c6f29bc9dbefc37582102dc9",
|
||||
"blk.6.ffn_down.weight": "7923c943b7629d560a032d1efa210d1d75c6692140f1be94464ee7ed24f44ed0",
|
||||
"blk.6.ffn_gate.weight": "57593d350361af753a6a39f53b066282634c0fb44f396f6f2966a574b01d8f8c",
|
||||
"blk.6.ffn_up.weight": "327b6a7a387098b8899d3ded04a4d4e7c658ca61b80d4e7b17594be232721602",
|
||||
"blk.7.attn_k.weight": "9ca48b87a10116fd8868e62b76f211d4bb91f166096be9061439ee2e1c3a5c20",
|
||||
"blk.7.attn_norm.weight": "cd56cfcc4e2ad6b96e23ea7b0d32b4caf236107d99a0b22c56760b62e63c8cfd",
|
||||
"blk.7.attn_output.weight": "7352b509a03cae2491ffc060e577d189341a0f861233f18c96f9d275dc4234bf",
|
||||
"blk.7.attn_q.weight": "2b3791c8c008c33ddbe12bedba8191322ceea2dcce5cf0eb7a93d40ad254e672",
|
||||
"blk.7.attn_v.weight": "3ae721d52466487a3d48150581e57f6d64ea1e83ab929f23b28c3d777422eeb6",
|
||||
"blk.7.ffn_down.weight": "3b6fa8ececdb3c34af3a5363863d6f94289c1c95bf47fce3a3ddcf184c5f0848",
|
||||
"blk.7.ffn_gate.weight": "dbd7df6c5ae5eb4adb859f0d36453813a4e289a359a1ba8f72d67fcbf21c3e22",
|
||||
"blk.7.ffn_up.weight": "de68380a334b4c5cfd4c318b0e9854aec59bd79aa0f0c30af3f56414f83482b0",
|
||||
"blk.8.attn_k.weight": "7303c4e4480abc72a7ee271811311199245fb5c2ea27a2bd3b8cad3a53a03c27",
|
||||
"blk.8.attn_norm.weight": "2e3d1921898d1b943ce1a1b6818546c8b471d6d542da24f51a8b514b8c3dd4ef",
|
||||
"blk.8.attn_output.weight": "30421520887b66bf97a18dbcdc283bc8d0b60590b612fd638a319a6eae923227",
|
||||
"blk.8.attn_q.weight": "73e064d5433c9b500068a1c31744dbd53f4ade298fb450a0e8c97f62cf1f8a8d",
|
||||
"blk.8.attn_v.weight": "27e21f8b9a9a8533e8178ca34a72aa1d786393d57302b7806dcdf3e51de511a8",
|
||||
"blk.8.ffn_down.weight": "bf694bd8e00047982108000e7b3dee7b225db8b19abc595e5697b6bbefd92e7c",
|
||||
"blk.8.ffn_gate.weight": "d55fdbf8606d9141b774b0500c58944fd1253b9e69d1f765eaa9a680b9f2ca40",
|
||||
"blk.8.ffn_up.weight": "1ae3f580655e7c8e8dd6c34fa4ac574fdfc5e3f1a8536da0c5442d3a2976f0e7",
|
||||
"blk.9.attn_k.weight": "b18080626012d8aabcf78542d6c7bf31c712bf55a70172fbfe173fcf34481036",
|
||||
"blk.9.attn_norm.weight": "2e3620620dc09998c6d3063a7d5de5433fbbae8c11e5b00d13f145d39140e162",
|
||||
"blk.9.attn_output.weight": "69c3c0e27ef1c0fc933eeb7b612b70909f18cde238873c0d576a2ba9714ef174",
|
||||
"blk.9.attn_q.weight": "68330e5aa28a28873c9a6e67f032186ef651df2df5844e0f27094ba349fbe4ab",
|
||||
"blk.9.attn_v.weight": "3df8d45a102be082d0793a51cb82aa62a43cd0e9d047ba4115ca0f2414b39325",
|
||||
"blk.9.ffn_down.weight": "1d6cc162b73745b135b4f040a0aac3c06d5135a3dc5b2421e7ee2af48662fd7f",
|
||||
"blk.9.ffn_gate.weight": "034a9d40fb1e32b534b45f4bccd65cbe43c4a6a3f5d01132bd245ca0005de5fc",
|
||||
"blk.9.ffn_up.weight": "c838c38d0e1a0ac0da17eb2a66023ed31929f07d8fcfe1cc546df26096c91f0c",
|
||||
"blk.10.attn_k.weight": "a78507cb72f744b86ceaa032596e74e5571c822d0226d334881169addb32cbd5",
|
||||
"blk.10.attn_norm.weight": "35f48d0b28ee0e6b4cad4e983925737562d64824be5b168b3e26df3d6b260cf1",
|
||||
"blk.10.attn_output.weight": "53712db06796de39b131323e7abf9a58551b6d52da6db66a471580386d396252",
|
||||
"blk.10.attn_q.weight": "efe08429ba196026b81cd1c471e1c7418afd9e966659feb3936b674aa0803b58",
|
||||
"blk.10.attn_v.weight": "7ec6055e134f89da0cbe79ec9f13ef2e442ac584b1f03c3e13e7d0cdad0078bd",
|
||||
"blk.10.ffn_down.weight": "37e66af4bcd1f3079e841e892255b8255070655901864ea3a8c602a7f681a640",
|
||||
"blk.10.ffn_gate.weight": "1825282bc34830d371c6edcc3c1e73e6ecc1e10f4aea0122dbb7acc1d6f7b1bc",
|
||||
"blk.10.ffn_up.weight": "819b3b276a4d4c14a35ed6682d5ef18a5e8ed468e5ce3f12e8c75ec18ac20ec4",
|
||||
"blk.11.attn_k.weight": "5327e6a2af82dfff0619a14971f5864a15553c36fead84e1af42c7630f2729c6",
|
||||
"blk.11.attn_norm.weight": "fec363b3c4a43036d2c635fb8aa9e122dd87ee79811839f2f6cd955be3373e7b",
|
||||
"blk.11.attn_output.weight": "ccf7b38f18ee8798b8a6a35018e2df3eb3e007de62876befb68025dd66c79763",
|
||||
"blk.11.attn_q.weight": "da8c4a1c824ffe174e39f126cd72f7ef83c56aff1259d452a1212de80f98f5e9",
|
||||
"blk.11.attn_v.weight": "d17ae6bb77f03982b55d341eb67acb5969e9ad3da5994b96eafc09793dcfe3a0",
|
||||
"blk.11.ffn_down.weight": "a6bac521e2791345f22c57205fa1c2f2f687794dfd24d0e98d50ae0d0eb6088a",
|
||||
"blk.11.ffn_gate.weight": "5ed902c488cb51ba5635f3df08258c5f84f31a679a00211ea5f9d8b824ef6d9d",
|
||||
"blk.11.ffn_up.weight": "ee9f1437eb890d2cf9df2574afa1cecf20aafdd847cd75b152d7eb74419afd34",
|
||||
"blk.12.attn_k.weight": "5a069c06e1019b0f889088e67458f7a11ec77fa190ada6069e46211f62219947",
|
||||
"blk.12.attn_norm.weight": "194d7e5fcc8c49aea62daf1940532419cf3c505afdce6be377286b677db5db8f",
|
||||
"blk.12.attn_output.weight": "6534995fd4d6fecb55e317add4b1723aba4d825e1e9471d0b08813dfdc247176",
|
||||
"blk.12.attn_q.weight": "4ab51ca519b5995581fa34f846276feca3b907ef2b51f192f6cc0b3263c3f5a2",
|
||||
"blk.12.attn_v.weight": "5652ca3fa81ef9a1ac1543d71fc6813f8517f8ec54b25c701f6f98061614830f",
|
||||
"blk.12.ffn_down.weight": "4b2c263f54c88516b8eb273bb8d9615b01c5c8b484dc70358adb91b50b300edd",
|
||||
"blk.12.ffn_gate.weight": "8f50c3c3e3e8568991d6c1b0e74b500cf4f208e7700bbb8e87c3f6a6d359b6b5",
|
||||
"blk.12.ffn_up.weight": "1c1a581fec1fbe959e1427fa513f400100b5e1ee9d83932630be9905fb49c231",
|
||||
"blk.13.attn_k.weight": "efd7a38c46f08d8376d82974f33c644e3a02220e142d63b1704718699a8a884c",
|
||||
"blk.13.attn_norm.weight": "d28fa4f1bd75abbd063b0e622e08f579c89cd0c0c5ce63c1952ec9f944f8ee13",
|
||||
"blk.13.attn_output.weight": "71e0068a639288718bdb70a6cfdefd50bc8b3ec3993347a65129e70001ca5827",
|
||||
"blk.13.attn_q.weight": "b97077adc92cff07a2e07d80ee38f214ad8713571c69cd5c70ebd43dc501ac87",
|
||||
"blk.13.attn_v.weight": "79b3e2749ab4b459c81e96e322b215f1e8af645eb346e176c326bd00cf6ed2fd",
|
||||
"blk.13.ffn_down.weight": "9f8687d11effa1db7cfecf7bec5631734bcf2962aad74a9f519144491e08ec85",
|
||||
"blk.13.ffn_gate.weight": "7d14dfa0543852e7777fe8fff29ca533744cbcf1ebcf10067e5adfc4eb345e65",
|
||||
"blk.13.ffn_up.weight": "852b9527b97fdab211ff3f832a660ee1d93ccb56906144c50f01319a6e8ee615",
|
||||
"blk.14.attn_k.weight": "79e926b20f36f66d58226cb358881f2f68ae7b468787d33cafae5110287a14a0",
|
||||
"blk.14.attn_norm.weight": "97d481b63deb0df6142c2c6cd23043720c62eb609e390f47a7113751c79974ec",
|
||||
"blk.14.attn_output.weight": "aa6e94d7176d5c79fbb89b96e5f13ce75702ce3dd23ee52986446da436a6c3d6",
|
||||
"blk.14.attn_q.weight": "214becb6d1bb460da9fb8ace0f99b9a5afa9edf7aa7acc19606c7401b11d6305",
|
||||
"blk.14.attn_v.weight": "488b0e6d7f1a7a2ed0972aaa6d10ef9c775ee5373460324efcf5b3e3da9311df",
|
||||
"blk.14.ffn_down.weight": "29c7ad16cf9542e30996a1a01ab95b844533b28051f04cc7949c371afb796471",
|
||||
"blk.14.ffn_gate.weight": "b7ef208f2b054803665b377f5a5980c122c026841809cf855c6ba06d1c3a885a",
|
||||
"blk.14.ffn_up.weight": "76a5cc28100748d79c4398ce7b9176aab4d661548b6293a82f99144812e5b70e",
|
||||
"blk.15.attn_k.weight": "a6b8f9e98ab878fa7ebc5d080978ebf2d050acc2ab2fa8ea9188eb10e27702c8",
|
||||
"blk.15.attn_norm.weight": "a26d07a9752d6dccb68e3a8a2a49fd0752cdd0a415e05547819bc37d9ba63d5e",
|
||||
"blk.15.attn_output.weight": "c63616c69048ccbee801e05be4f56d21fda21aa0cc470f41d57c31b4d9283a4d",
|
||||
"blk.15.attn_q.weight": "fd595a67bf96c6ba16eb148a9d02fa52fa3c1d33ed10be28a08f851409fd6e64",
|
||||
"blk.15.attn_v.weight": "1c5c9d33fa07c05d5f4ed0032c6c4aa83d863f0d31c94a66109d239dcd03cea3",
|
||||
"blk.15.ffn_down.weight": "585ea62ab8aff7d7d212ea5c1a03226fda6b68370c890b776834af70c948dcbc",
|
||||
"blk.15.ffn_gate.weight": "a13c63f86f879b03a573d5dd2a25cfd1f4dc73e8132e6454ecc23e538b4cdf6f",
|
||||
"blk.15.ffn_up.weight": "f7112450f57c12fcd511f049e0dc0b541625a107a7901c3261ed9e984299f65c",
|
||||
"blk.16.attn_k.weight": "2d2c8b11dd71fba6d1c106aa1673c113a5448653cca7eab897c8739212ed5003",
|
||||
"blk.16.attn_norm.weight": "95c2ec7be9469690e18a9a1779684acb3e9da44b13e263a0da840305646fbf8a",
|
||||
"blk.16.attn_output.weight": "31a65046e677f54dae654ded4e733479fcc0f7283d83076b7dc7cbcae8528230",
|
||||
"blk.16.attn_q.weight": "bfc6292b9c6d49b7118d08060242a138182eb182d136ba5dfaf469437c16081d",
|
||||
"blk.16.attn_v.weight": "68f81d037340217d87c7853ff4d6edfbc46d9e827ee6d5bff7c3f6238e3a95ad",
|
||||
"blk.16.ffn_down.weight": "bbd6629691950cef4d5113e1c6670e91b216a9b872cb92cee02dfda4d6c4f7b8",
|
||||
"blk.16.ffn_gate.weight": "63cb56f282b7401ed6c76e5bb6fdf1bf68a64f9af0c82c014209b55bcb5191d0",
|
||||
"blk.16.ffn_up.weight": "b54f39a2541063cbfb6f713aa81c3b69a04100e999aa2ebbeec195dc382eceec",
|
||||
"blk.17.attn_k.weight": "3d9ba49799cc56664ec30a002bcad61eb651294212a68c3ddb573eb042aef5a4",
|
||||
"blk.17.attn_norm.weight": "42ee0db4b9d63257bca0012a30b12737ead1caafeb5ed3d93c8f48ffec4b46de",
|
||||
"blk.17.attn_output.weight": "a38fd100f05c9041c592bc739e287de0b10d08ef2bda41a879225bdca9002f71",
|
||||
"blk.17.attn_q.weight": "8a3bee285b0180a9eb35662e449ee4cbe16d992bdd48fb3a94bc4a347728cfa2",
|
||||
"blk.17.attn_v.weight": "d7f8f1b8b863494ed4392a1656775912e9b264ad36016547b12e832a1d6757d6",
|
||||
"blk.17.ffn_down.weight": "bb7ee58f61da8630972e25b621996fbe8ec06f4dc9ab1e268ab5b120c526ca28",
|
||||
"blk.17.ffn_gate.weight": "6b652dbf167fee09a45ebfd78d500ff6548fb2756dbe5343ffec3f7e6207179f",
|
||||
"blk.17.ffn_up.weight": "3b67f727e55e742715de978fab80457781e7a3762bc48f79d13b45dcb8de664c",
|
||||
"blk.18.attn_k.weight": "ff7fe57c57b90c6fcc0aefc39ec24593c3a7d1ea1c23770480075a015450e0f5",
|
||||
"blk.18.attn_norm.weight": "1d40faca082d2633ef0ccf19e121870dd6c7c3e2154607c7f3543fa96e99cb2d",
|
||||
"blk.18.attn_output.weight": "9adfecaaa397a92db4687efd5fcabfa0daef9e6b0493763b7ff5ebc185c43a6c",
|
||||
"blk.18.attn_q.weight": "ad1803eb9b291948639277afe981e666b07167eb3fcae903ba5b73bf86d8f50b",
|
||||
"blk.18.attn_v.weight": "308cf23399adccf27401a4ab60d74dac6fb9d4cd4b9c5940d9145118d1881b34",
|
||||
"blk.18.ffn_down.weight": "7de4ac9a561fb580619b745687dfd7ca8a69ef70471dee978741b80e9ff7bead",
|
||||
"blk.18.ffn_gate.weight": "0c66970f696b33bd5ee8f1f2fbcb41fd78fa5ccabdc927e11a4d5a4089f19c69",
|
||||
"blk.18.ffn_up.weight": "66a42e988e8a1f468fabf976c48e9e4bb045eaac6916ef16555ac101cd674abc",
|
||||
"blk.19.attn_k.weight": "a928ab50390bacbcebe2e4b66922498134ce22d7b93beaa87d6cf4ab52eb7174",
|
||||
"blk.19.attn_norm.weight": "b4a02c55b46c2a96aec9c64a254087cf48e6c1d4b6f31782c77a46fc4daebad1",
|
||||
"blk.19.attn_output.weight": "b768319c641dff1eac5d1f8ceb960c9899c795bf2b24c1d6bf70aa24fda45f77",
|
||||
"blk.19.attn_q.weight": "79ef3f57d187d3954a26362096e1b6c222d76f537dff73e034d6e9999935b8bc",
|
||||
"blk.19.attn_v.weight": "ce13d6b13e24fcb2d5bc6a2662e5bd295b31b12db10a6d0307f86cf29b8d5001",
|
||||
"blk.19.ffn_down.weight": "cf90d7e2137482cfd50934a8223ad774621d08554969da80a9712df5e6227eb0",
|
||||
"blk.19.ffn_gate.weight": "71ce30150f003b6eeb3bf7464e05b6ae615f135110d8e47f0a47fd973e537c0f",
|
||||
"blk.19.ffn_up.weight": "7f92aca0cc29866633feec701ec01a85a8ee2fd4e2b9630173a6cffb1d9d50ee",
|
||||
"blk.20.attn_k.weight": "a2df23159d6fb74ef28e14b61028fe8b00a693a2fc9234a980be74f20b958682",
|
||||
"blk.20.attn_norm.weight": "c6cd5f1b096fc5efa4eb59ca1c8c4bd28730f3dcedd59a63601663eccc6724ed",
|
||||
"blk.20.attn_output.weight": "896a8a166d0f006d4b09867ae4345426303cbc3fb13a18d3d4e1bde00f16dbdf",
|
||||
"blk.20.attn_q.weight": "01eb79588fe61baea0da43e99f4dc5939590e1bafd01e12dadb8326f102bfea2",
|
||||
"blk.20.attn_v.weight": "bd39630fdd5a7c859ac1addaf53e63faf524c3f32f5f4896d86b6e746b1d5c06",
|
||||
"blk.20.ffn_down.weight": "0304a5d39957a0e3f031c4bcc4549a135d396c8d97c8d276fd1c823ce86560c2",
|
||||
"blk.20.ffn_gate.weight": "117b79d595b1dca0c8b37586beaecc4d84411507276212dc286cde7fc36c9bef",
|
||||
"blk.20.ffn_up.weight": "6e799346db145c125f01783539749d3828fcc451cd4f10c5352f047a47e28714",
|
||||
"blk.21.attn_k.weight": "1c37e4c0664147e775bb006b226b9553e3421140cd96288ea755f81731ab80ba",
|
||||
"blk.21.attn_norm.weight": "00ae783a29000ccda5e4bdbff03df0752fb82805dc3f9b987500ebd80714476e",
|
||||
"blk.21.attn_output.weight": "7588b84f9fb19f15095b5265c60b4a4e7ae74bcc47d4607dfa5d0bfab6f136cb",
|
||||
"blk.21.attn_q.weight": "a65f1c0dd06d45bb97532d3e932689c1eecfe7359089b39174a96a149335cbc1",
|
||||
"blk.21.attn_v.weight": "4220b77e7d5e8709b4eef33a679b5dad11f297085ef44c9977f9e54ef08f7a2d",
|
||||
"blk.21.ffn_down.weight": "b8c082a0530d4b5328e67db0df84c5498f2af956de23c639fa0198ffea853950",
|
||||
"blk.21.ffn_gate.weight": "cd1b656ee72d00e9835ef667c19ef89a88de261eb8eb7c0e936e0f9ddf83ef9f",
|
||||
"blk.21.ffn_up.weight": "dc445f73e36ec7a3bd86884186b728f8e0187f32848c3b8b69d4d41f8571bf31",
|
||||
"blk.22.attn_k.weight": "e37cf0b893ec8b9ee8c78dd139b8d9c45cb997a3bc0c3d93a70ca1c3f6af8859",
|
||||
"blk.22.attn_norm.weight": "248a27838d3c46cc03a5c312facc84e2e0e2c990ef8401e93da25918497f88d1",
|
||||
"blk.22.attn_output.weight": "fc191a18f6d18332c66761f7ab28008bfe295dd1f5c8741a2488442f9e00d0f5",
|
||||
"blk.22.attn_q.weight": "4b193a2ab8bc2b085db18f2bf3eeba26e02b537b2cdd738160c8f14b165d0f5a",
|
||||
"blk.22.attn_v.weight": "7a60ce5ccac7e045e55ba1e1e85bd2a0f93f8c781daee96c5223665e22f0c666",
|
||||
"blk.22.ffn_down.weight": "e0a34fb4244e2c7168f3dbaa1904c15d339ec39999cdf27128bbaf619ee0a237",
|
||||
"blk.22.ffn_gate.weight": "8bac872d4b8549c8812f927efa309f1792b524f33601095fff61b826de5a5615",
|
||||
"blk.22.ffn_up.weight": "b67fa2b94dd901b6ec64c0853ce8ca2d86fe9cb1cc6d2f15fbbbe0e691c0c648",
|
||||
"blk.23.attn_k.weight": "2c32e66ad01942b819ac09a197c71579fe66f02226a264fdd72ad1e02c67a27e",
|
||||
"blk.23.attn_norm.weight": "825fdc94deb439cb93c713eeb077c1052b90ed658d6d464fc4ad3d611e911d48",
|
||||
"blk.23.attn_output.weight": "95ca6707a95b8750b0c7c5d379d368f0f2e7ebef631954e7d4d8ec0f41f13a3a",
|
||||
"blk.23.attn_q.weight": "6eccc84faca5fac015d1b26e2854501edcfd292a302228fe14cf99f5eb59a34b",
|
||||
"blk.23.attn_v.weight": "b343ac3d226040f1033ee049668aa1d89b1774bc18431965682e5dbdce78ccdc",
|
||||
"blk.23.ffn_down.weight": "9fc599befea8d3b1e342d564a110074f66d2542df406c4b90b6bdc5828fbb2b2",
|
||||
"blk.23.ffn_gate.weight": "488556c1b0c9f0b20b0c99b4bac2e0f4046b81edb601d7b91e7e5b3bab47d667",
|
||||
"blk.23.ffn_up.weight": "1088e291d7008dd9c7c2dd6830af686a8a84b724d123a016209bd5156d6898f1",
|
||||
"blk.24.attn_k.weight": "a923fbe35e61e009a53927d7828818e0592bb737d6a1106c4b0b5a1efc367e07",
|
||||
"blk.24.attn_norm.weight": "9b51aaaa939cefafdd9b13a7e5b74ac7fa2d603427e55a16a909d6f3f353750a",
|
||||
"blk.24.attn_output.weight": "1beb2baba56f8409466434b037771248c2f620ec5f53e15f44c271d5a2d9ecf4",
|
||||
"blk.24.attn_q.weight": "4b0194fe5bfae0c6bf6131dcf8cb6e2b994f6ea10b27cb03574f0f4f8cc0c950",
|
||||
"blk.24.attn_v.weight": "6ac34b1ab0f66226d85bca1194a7c212cd93d384ecbc8b8395de48aec0970a61",
|
||||
"blk.24.ffn_down.weight": "5508f74cb732a662c2936b32ac5e90742d172b9f961a747b0e5cba0e5906a89d",
|
||||
"blk.24.ffn_gate.weight": "095e39b8584403835f9bb1ac33e0e81f54175575e4800273d281b845bff381e7",
|
||||
"blk.24.ffn_up.weight": "2d43ec21637dda12973de367b0113ee9840b0d815bf6fce042f7c3f270b0b530",
|
||||
"blk.25.attn_k.weight": "9e2aee029f3d2c7f67dfc7926e72c8228fb978382c8e5a4701bbf82c93801419",
|
||||
"blk.25.attn_norm.weight": "220cd7164fb4cdbe22d26058e4153b26c27c7b5ce2bec8e95bf2c0ea08d23103",
|
||||
"blk.25.attn_output.weight": "a17f4a5dc6aa51f03dbd75602d98e9491767c205cdc2c3a5f8667fc54bbf7c64",
|
||||
"blk.25.attn_q.weight": "f60827496835c440c794bf57ce9780704d10a59d8229886bf75ebb18900ba4ef",
|
||||
"blk.25.attn_v.weight": "9cac217e9e9f4f4c85f14ee51165a77c580165bd4a34b202389169bbe61a1ced",
|
||||
"blk.25.ffn_down.weight": "a0f36949b663e80849581dfb71e7babcc73580793bbcb0c80ab26d5a6e000359",
|
||||
"blk.25.ffn_gate.weight": "df4d1be4d50d6afe5ad3ef0d0e0fac76a33e85c963dea769641d612dd53e7d13",
|
||||
"blk.25.ffn_up.weight": "992da76be762632e25ebc5ef4d03728eece1b43f7c4e31827df19ca724aea694",
|
||||
"blk.26.attn_k.weight": "34199ff856ac32a500c754539d070258574192a34ecba87a182897cb59fdff52",
|
||||
"blk.26.attn_norm.weight": "a8e9dfb2dae5d22b5c0aec5f3675991c0e3c3e6a44153db2579136b73f456e00",
|
||||
"blk.26.attn_output.weight": "1c4f257ffb0d7db0f11cfb275e38b4af736917b43ad82de1badce3f1d227da4d",
|
||||
"blk.26.attn_q.weight": "33d55786274c2e718cf61e8fbecf3dfa5ee0c208f0b716d42b061f55459acb3c",
|
||||
"blk.26.attn_v.weight": "684b636939cd4ffcfec5a6238a0790ffa43d853c95783af9b9e8275e74071a7a",
|
||||
"blk.26.ffn_down.weight": "89d0bf066db154e6d312b5433aed1714f6a28b40f4c52e3e1530ee07703303c8",
|
||||
"blk.26.ffn_gate.weight": "393d649bebe5e2940e1b043649f6c860b4b8b9f380f30e9da1744a830f358156",
|
||||
"blk.26.ffn_up.weight": "179edc85ababd9d8440cc6093eecd1004290aa1cb96434b26ecf7585b6cca17b",
|
||||
"blk.27.attn_k.weight": "334841445a7f1e14731b08f56eb0b1f0938c63823d28bc6d078c4c5f05b36f19",
|
||||
"blk.27.attn_norm.weight": "57344471bbda2e9deffdfdb2dd05a07aa47f8761e24de53525588639145bf551",
|
||||
"blk.27.attn_output.weight": "506126af9ee54b535d49f97e36f630e74834f480329f098d6d62e96246d8d65a",
|
||||
"blk.27.attn_q.weight": "dd984df1acb4783849e25ba7ae378bfd385cd9efc540fb798cd5bdd873f0118f",
|
||||
"blk.27.attn_v.weight": "b4b3fe9a4455d34c297ff20a2f537b647cef424741d840a747b265f23d320ac0",
|
||||
"blk.27.ffn_down.weight": "621fdb185ba0d35ba5476dae73d2c81ec1482a0e878d5bfd5c3b29fe837af013",
|
||||
"blk.27.ffn_gate.weight": "e4fbab45f2ec506fa374103251a0bdb7baa6f576080bdd796f3e9db92098e08f",
|
||||
"blk.27.ffn_up.weight": "a0c57e463e988002bbd6a6c6792baa21a65e6f89ae303a2c301951b0ae6e4bbe",
|
||||
"blk.28.attn_k.weight": "bac36cbd52ec5056841663865e1291ddab4b47ef9a2544dd285d4503bfb0e4a0",
|
||||
"blk.28.attn_norm.weight": "5774a9df2bbb2e86d1f70179c7b92d81e1f401160148b3328fb64db6646a5425",
|
||||
"blk.28.attn_output.weight": "e8712622d1569557000c75f26c3f55fad267fd300463c2c2cfe3afbfa1c8f908",
|
||||
"blk.28.attn_q.weight": "11677751fddee52cc739699c02836f7be54d96038be4240be5d4f53d00161608",
|
||||
"blk.28.attn_v.weight": "e5ee459b8958d65e1445997b9aa1e90e2f5d17761ebcf5357313119a45322507",
|
||||
"blk.28.ffn_down.weight": "3934518f9f85292da8475fe38a8edcbfc4e24ac56c351b472d6351f98750871e",
|
||||
"blk.28.ffn_gate.weight": "6ba735d57e98d0847e487f25ffaa25256deaa8abec76f428cb70bd9774279d83",
|
||||
"blk.28.ffn_up.weight": "977fae6e1e5353114fc645dd98429464749758765cbc6e6457593d596e57850c",
|
||||
"blk.29.attn_k.weight": "8122a457307d580ad6f1e0acea09a2f593d97f595ba0d6737f5fea16d2433642",
|
||||
"blk.29.attn_norm.weight": "d626f721e05aa1202439b01027031d4caf1adace61ed37870a277cb6297c77cc",
|
||||
"blk.29.attn_output.weight": "7fb7122ab1b6b1e6615ca746897da27bc52c92cb70d3147183cdde61795b72b3",
|
||||
"blk.29.attn_q.weight": "be43e94ff6b6e391024dc824101efa0ddf4005d5b002ac26cb03765c0c73c2fa",
|
||||
"blk.29.attn_v.weight": "af93c85ebff908f74f9935b81bde0516ca487c84139868a1ce079c3ae20036b1",
|
||||
"blk.29.ffn_down.weight": "39dae12340ed3120bd19c495fe0872b559613641e41fde69d02d8631900b84c0",
|
||||
"blk.29.ffn_gate.weight": "36fd482439840ef197c9f3b8905d86acfcea49bcf018544106ca465d4bf8d5c7",
|
||||
"blk.29.ffn_up.weight": "5243fbdfdc1e2a1dd84b6210a9869d18a014db9088897e345240cdc99990bd5d",
|
||||
"blk.30.attn_k.weight": "948f263616bd3788b2b968baafd69b9c5bd1b77578665f096c4b7e247b4cea42",
|
||||
"blk.30.attn_norm.weight": "e168df981e744874ff303faf2eb470e5f6868c2040ba5f383f6c5148669975e7",
|
||||
"blk.30.attn_output.weight": "4cf0ccca04b792573b756655a24fc89cfb1f272da8305633f0bc66ef14990b93",
|
||||
"blk.30.attn_q.weight": "21e07d6cba6c50d65350289258209717174a13c42be57e8141d69712cbaf32c1",
|
||||
"blk.30.attn_v.weight": "65a8ca29c7237b3182ccf03e2fc94e84f9a53d0e160fb679ab401c853170dd9c",
|
||||
"blk.30.ffn_down.weight": "8b00500a6d00d84058f6658ee1d6f06fb4fcae2f90d4341792259362923b3c13",
|
||||
"blk.30.ffn_gate.weight": "5bc0e19ab7a31b50ac2118ad1b36e31055271a322cd8ff661d47c3ac0210703c",
|
||||
"blk.30.ffn_up.weight": "f37a0561955725bd59ee2d064fa9f4e00a12a1b620b624db3bc3add5330bc321",
|
||||
"blk.31.attn_k.weight": "9a5663edda227f5d87533897146764f8e8a7481b9e71fae197c39204f8463221",
|
||||
"blk.31.attn_norm.weight": "060a4f438a1ee5e220b5b5278ad2f5c085a428bf38c515766781815597c87529",
|
||||
"blk.31.attn_output.weight": "6ada5d3cad9dea4780ffbb43302bb6ccc2f24eddd0fc4f5f84c9ce0fc0c6e5dd",
|
||||
"blk.31.attn_q.weight": "bb5d08c08603907981ad388d5d8b70fcc9b98034ba264b8474c8890cc0297af0",
|
||||
"blk.31.attn_v.weight": "e01b4252ea9c6a889c32b21144b441a347464d04536ef4f6572425be55759796",
|
||||
"blk.31.ffn_down.weight": "8ba4d679c36e93ba65ba03180385ef35ea86b3b7cdf2fded9df59369f1c09630",
|
||||
"blk.31.ffn_gate.weight": "e5b41dc93645f8b5e8eebae3ada3ea43a18f97ce2654228655170b07b463ccb0",
|
||||
"blk.31.ffn_up.weight": "25b88cdddc8b547af294ed107d3d1312e90b983cae87936fa6062ecd8ea02539",
|
||||
"blk.32.attn_k.weight": "4bcf86dc0858c8ca2fbdf6aa76674d43eb698f78979fdc1a38f556a7af1facc4",
|
||||
"blk.32.attn_norm.weight": "cdcc12f3b8b9773c6722736bfb748a2729230b21478cbcc4104859d3148df815",
|
||||
"blk.32.attn_output.weight": "d43f1196822995ed89a9365c97054753a8b30ce20b6e273c8edcc42673a1e141",
|
||||
"blk.32.attn_q.weight": "ebf2972bb3865cbc5be4840113a322089752038344beab2a0122c7cb4fb399b6",
|
||||
"blk.32.attn_v.weight": "714db81704ff34fa137512903c1013acee7877467473e46600728b9240582eb7",
|
||||
"blk.32.ffn_down.weight": "2cde3da1258bb170a79d5d3cdfe10c86a71eb34b77da46b74c5ed71e7f4fe274",
|
||||
"blk.32.ffn_gate.weight": "c7e1ed792532613ff9d4e5834b6536e2e0f47df2303bc0fdaa90aac0c1f4e8db",
|
||||
"blk.32.ffn_up.weight": "d8d6f13fe66a716e28f79101a29817f0c0d6f99969a6f017d51bafd1a16c600c",
|
||||
"blk.33.attn_k.weight": "a0a28f6cbca88da00cab2ca37094d9b0503bf9defdae77b91895b911c408cbb6",
|
||||
"blk.33.attn_norm.weight": "0251200c24cc8445607ace6dc8c5aa0566567997262b7cca53a11ac23cc564b2",
|
||||
"blk.33.attn_output.weight": "b2423205bdf6a1096d43c44d8d12f1a84fcd4e1bb70fcf6dc8542b8b8a71a13c",
|
||||
"blk.33.attn_q.weight": "00b425c3ef71065ce5e0234e702bf38143b4952da78a85f52ab2c2e3073d97ab",
|
||||
"blk.33.attn_v.weight": "035edd2335df816c42c765a5e66b9d9b9e15a822a8dc1863508145499c942c14",
|
||||
"blk.33.ffn_down.weight": "4894a923a3db75bae4496ba3ce5f28796ad31fe33996a066271fb8654964310e",
|
||||
"blk.33.ffn_gate.weight": "8f6c819b8bbfbe3357fae89e1ac5a3d58be85b3b04be3bacf7b62775869046ff",
|
||||
"blk.33.ffn_up.weight": "257c3544b5b544fd5d839665bf5caf107a329b59dbc3751efcaa24ae63c56179",
|
||||
"blk.34.attn_k.weight": "b6cd8bba892e38dac4a2ebc3ba1bce49e71b967fc436fde30c6d76f54a18935f",
|
||||
"blk.34.attn_norm.weight": "2b3c8e60a064cba9955752bbbbdd92c71ba5c2f1bd721097bdbe88b5abc68787",
|
||||
"blk.34.attn_output.weight": "8cc272551c9aaca9db5a660c6927bab94a0243d74a30b2bc165f06bd577714ea",
|
||||
"blk.34.attn_q.weight": "74b561eb4792484e6a94b58fe2583848c3ae28ff2f1bf3d02939a0cfdfa49990",
|
||||
"blk.34.attn_v.weight": "dba19e24ff05154dc5a1f55c023729303a583d13d68732ce22ea74d4410dc8f0",
|
||||
"blk.34.ffn_down.weight": "76eca5dfeb274c35774e0bf9f22ee420ed9085c8e99aa2cd5a236e4918b44c61",
|
||||
"blk.34.ffn_gate.weight": "9af0862d5fcbc24732846488e653db8242a467765c0cdbc00332b3a40256b4a6",
|
||||
"blk.34.ffn_up.weight": "2a03126bf73587eaba99ece2066103d12e47bcd4ce30ff6c17b2f383b81d40df",
|
||||
"blk.35.attn_k.weight": "52513fc0cd4e997a842729af7d21dd09399bce0a339558374738be266d0fa2f0",
|
||||
"blk.35.attn_norm.weight": "e5281fa911964263ccf1630b14762edbd41d0b9472d6ec695fc600fed4892c35",
|
||||
"blk.35.attn_output.weight": "b391d6705d5dc6f48326b5fd16573f679edf64109d86fb729a498819676590ca",
|
||||
"blk.35.attn_q.weight": "d16446921966db9b0e0539626ad22a2511ace780e59379d6a4162d8c5441440b",
|
||||
"blk.35.attn_v.weight": "9d8cdf23ffdb0c5c74106843390b94b24c9f33ef0eb9998d39f78c73390101ea",
|
||||
"blk.35.ffn_down.weight": "938eb6301f7bbf162d7dd965682a5ed11d0a4a530c6fedd7e5469ce80012fc17",
|
||||
"blk.35.ffn_gate.weight": "5ad84f5a0c8edcfea1ecf1a3e3d21d85ceda0c4ad9e3c6ca68885eeff8ed3c2f",
|
||||
"blk.35.ffn_up.weight": "1c4330d9dc71bf4c98812c34356c51f520f47610a534152aa6d29284b758090d",
|
||||
"blk.36.attn_k.weight": "ef720655e5ca2465f13db2dfc4732fb4ef2c9d53acde52f514fd4f301e974081",
|
||||
"blk.36.attn_norm.weight": "88f4b9310b3c8c2644e3029160cd35678c79dfa59280430e03f5c29a6fe84a58",
|
||||
"blk.36.attn_output.weight": "aec6f915fffd7bb72cd783273e871b4f09605950089d45e72059d1316b6c4b01",
|
||||
"blk.36.attn_q.weight": "72f9408a2405d42f8db6ce5fcf1d26a3660b6f225fc60e77d0277109cfcb82ed",
|
||||
"blk.36.attn_v.weight": "0f3b3d851dc44b3893ef53f6cca5b4acc9658bacfe1cc2d13c3d704ddd409b67",
|
||||
"blk.36.ffn_down.weight": "470aec48ce8c5129a6654d9fd26fcae72776f9fc1429a8bb05818072a876475d",
|
||||
"blk.36.ffn_gate.weight": "7f5f296d09cf55679767b5d15de3eff489c456782119f25204be4b1647f18dcf",
|
||||
"blk.36.ffn_up.weight": "b7ef74a1f7ffb4982711d93f1787be3a70edc3d2358d5203c41d8900508037d4",
|
||||
"blk.37.attn_k.weight": "c4ffa5412e4ff2dcfe1aed991c1f54169fd171a4c7638e4b9f21a1ca64c5e1d6",
|
||||
"blk.37.attn_norm.weight": "4eb6c888d841cccfacf5b963f8611120f6ff24b84af0b5714fd9ab36dcda422f",
|
||||
"blk.37.attn_output.weight": "db2a7bbf9682f9f6eea672dae8e150738f1bf74dbc80edc7022017a3f040c8ac",
|
||||
"blk.37.attn_q.weight": "e38c0462aff139afcbab289189823527e453abc9e541154adde5e7af88cacf0b",
|
||||
"blk.37.attn_v.weight": "952eb2492ed452a72f96bcc12d4b2affad9dfdf46ee39ce4a5d7b57a5dc301e5",
|
||||
"blk.37.ffn_down.weight": "25f23a8fbc44febf6dc4848fd7fe03a580e2822bd3b3b5a51f4990826bfe3e4e",
|
||||
"blk.37.ffn_gate.weight": "707da5eb40118b035305d3262444382351f170a20a537386a70e90c5a83a7817",
|
||||
"blk.37.ffn_up.weight": "d2d2ba5cfc4ef47338dd7384219e22bf030a5a2209e0354d88f5bbaaafd20e87",
|
||||
"blk.38.attn_k.weight": "abc4bb189dedf7ce661e79028427623a4f91ac091c2cd60e31b58bc62b1cda71",
|
||||
"blk.38.attn_norm.weight": "9f4803a7d03fd40fcb83d85f84eb1d5682ea4e5bb084f210c02850675d804c3d",
|
||||
"blk.38.attn_output.weight": "77cb66007f1a41df7135d0e7f900ceb499c2f667dfc3f1a6ac01a3203bbd3ccf",
|
||||
"blk.38.attn_q.weight": "d94a8b26cd375bf2bcaa76597e314aa8268ee50a479d00931e5e0e021feadb5d",
|
||||
"blk.38.attn_v.weight": "660c907888bc5016dc69b7d35fe6f55c7ded697c93be0e2d332a2f17aff88758",
|
||||
"blk.38.ffn_down.weight": "6f06173bae5b00ffaf88ef383619a8b9c6a8d0d5c6494695d17f6c1de1a68a13",
|
||||
"blk.38.ffn_gate.weight": "89f99be149d03f116527bfcabe073c50001c874de40fb6e817f6619027f3cd05",
|
||||
"blk.38.ffn_up.weight": "8d57557c8d5e2d2688b73f01dddf1ce8d5194990cda6358153320aea88aac7f8",
|
||||
"blk.39.attn_k.weight": "21be09c988b46c8393e6c2ec9230f3b5136eb7607dd1953ba92d0811c2f0dd75",
|
||||
"blk.39.attn_norm.weight": "ba7c1912dd1c4e2d16917201f62396fd0600e4a451137eaddff255548c209abd",
|
||||
"blk.39.attn_output.weight": "acfaf4abb3fd27fd899b5563c3877f176b597d8f6cdb2f2fd3f3a0bd4da15ed6",
|
||||
"blk.39.attn_q.weight": "e8adbc140d4c8f0db2a27ca584c5531d5b1e080555fe627e34d80d0814a92bed",
|
||||
"blk.39.attn_v.weight": "92f96b0e1f724e73a0f90a76c145654418844c04a6d4b14c05eb5af8a62bf8dc",
|
||||
"blk.39.ffn_down.weight": "4d9ee7c65fc16fe95d10c47b79ac6a525741947600a64b5fcea5d300a82c50de",
|
||||
"blk.39.ffn_gate.weight": "7e18507989f39b32191133d2657c2ee3b74f42f070579204d727eb72215793d1",
|
||||
"blk.39.ffn_up.weight": "22cda752269c9757ba918abede1df95bb0f83a5c772dea13c8deea3d5f2723d9",
|
||||
"output_norm.weight": "2858cf0e39d32caf52b7861378ace076000241e147f10b9eb21d8a5cd149e3cb"
|
||||
}
|
||||
@@ -2,7 +2,7 @@
|
||||
|
||||
### Getting Started
|
||||
* [Quickstart](../README.md#quickstart)
|
||||
* [Examples](../examples)
|
||||
* [Examples](./examples.md)
|
||||
* [Importing models](./import.md)
|
||||
* [Linux Documentation](./linux.md)
|
||||
* [Windows Documentation](./windows.md)
|
||||
|
||||
@@ -38,7 +38,7 @@ Numeric IDs may be used, however ordering may vary, so UUIDs are more reliable.
|
||||
You can discover the UUID of your GPUs by running `nvidia-smi -L` If you want to
|
||||
ignore the GPUs and force CPU usage, use an invalid GPU ID (e.g., "-1")
|
||||
|
||||
### Laptop Suspend Resume
|
||||
### Linux Suspend Resume
|
||||
|
||||
On linux, after a suspend/resume cycle, sometimes Ollama will fail to discover
|
||||
your NVIDIA GPU, and fallback to running on the CPU. You can workaround this
|
||||
|
||||
@@ -67,8 +67,6 @@ To use this:
|
||||
3. `ollama run choose-a-model-name`
|
||||
4. Start using the model!
|
||||
|
||||
More examples are available in the [examples directory](../examples).
|
||||
|
||||
To view the Modelfile of a given model, use the `ollama show --modelfile` command.
|
||||
|
||||
```bash
|
||||
@@ -155,7 +153,6 @@ PARAMETER <parameter> <parametervalue>
|
||||
| temperature | The temperature of the model. Increasing the temperature will make the model answer more creatively. (Default: 0.8) | float | temperature 0.7 |
|
||||
| seed | Sets the random number seed to use for generation. Setting this to a specific number will make the model generate the same text for the same prompt. (Default: 0) | int | seed 42 |
|
||||
| stop | Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return. Multiple stop patterns may be set by specifying multiple separate `stop` parameters in a modelfile. | string | stop "AI assistant:" |
|
||||
| tfs_z | Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0) will reduce the impact more, while a value of 1.0 disables this setting. (default: 1) | float | tfs_z 1 |
|
||||
| num_predict | Maximum number of tokens to predict when generating text. (Default: -1, infinite generation) | int | num_predict 42 |
|
||||
| top_k | Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a lower value (e.g. 10) will be more conservative. (Default: 40) | int | top_k 40 |
|
||||
| top_p | Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9) | float | top_p 0.9 |
|
||||
|
||||
@@ -204,6 +204,45 @@ curl http://localhost:11434/v1/embeddings \
|
||||
}'
|
||||
```
|
||||
|
||||
## Extra arguments
|
||||
|
||||
### Setting context length
|
||||
- `context_length` parameter can be used to set the context length for the model
|
||||
|
||||
#### OpenAI python library
|
||||
- OpenAI python library does not support setting context length, however this can be set for Ollama through the `extra_body` parameter
|
||||
|
||||
```py
|
||||
completion = client.chat.completions.create(
|
||||
model="llama3.1:8b",
|
||||
messages=[{"role": "user", "content": "Say this is a test"}],
|
||||
extra_body={"context_length": 4096},
|
||||
)
|
||||
```
|
||||
|
||||
#### OpenAI JavaScript library
|
||||
- OpenAI JavaScript library does not support setting context length, however this can be set for Ollama by passing `context_length` directly with a `@ts-expect-error` as an undocumented parameter in the OpenAI JavaScript library. [See documentation here](https://github.com/openai/openai-node?tab=readme-ov-file#making-customundocumented-requests)
|
||||
|
||||
```ts
|
||||
const chatCompletion = await openai.chat.completions.create({
|
||||
messages: [{ role: 'user', content: 'Say this is a test' }],
|
||||
model: 'llama3.2',
|
||||
// @ts-expect-error context_length is an additional parameter
|
||||
context_length: 4096,
|
||||
})
|
||||
```
|
||||
|
||||
#### `curl`
|
||||
```shell
|
||||
curl http://localhost:11434/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "llama3.2",
|
||||
"messages": [{"role": "user", "content": "Say this is a test"}],
|
||||
"context_length": 4096
|
||||
}'
|
||||
```
|
||||
|
||||
## Endpoints
|
||||
|
||||
### `/v1/chat/completions`
|
||||
@@ -213,6 +252,7 @@ curl http://localhost:11434/v1/embeddings \
|
||||
- [x] Chat completions
|
||||
- [x] Streaming
|
||||
- [x] JSON mode
|
||||
- [x] Structured outputs
|
||||
- [x] Reproducible outputs
|
||||
- [x] Vision
|
||||
- [x] Tools
|
||||
@@ -339,27 +379,3 @@ curl http://localhost:11434/v1/chat/completions \
|
||||
}'
|
||||
```
|
||||
|
||||
### Setting the context size
|
||||
|
||||
The OpenAI API does not have a way of setting the context size for a model. If you need to change the context size, create a `Modelfile` which looks like:
|
||||
|
||||
```modelfile
|
||||
FROM <some model>
|
||||
PARAMETER num_ctx <context size>
|
||||
```
|
||||
|
||||
Use the `ollama create mymodel` command to create a new model with the updated context size. Call the API with the updated model name:
|
||||
|
||||
```shell
|
||||
curl http://localhost:11434/v1/chat/completions \
|
||||
-H "Content-Type: application/json" \
|
||||
-d '{
|
||||
"model": "mymodel",
|
||||
"messages": [
|
||||
{
|
||||
"role": "user",
|
||||
"content": "Hello!"
|
||||
}
|
||||
]
|
||||
}'
|
||||
```
|
||||
|
||||
@@ -80,10 +80,12 @@ type StreamOptions struct {
|
||||
}
|
||||
|
||||
type ChatCompletionRequest struct {
|
||||
Model string `json:"model"`
|
||||
Messages []Message `json:"messages"`
|
||||
Stream bool `json:"stream"`
|
||||
StreamOptions *StreamOptions `json:"stream_options"`
|
||||
Model string `json:"model"`
|
||||
Messages []Message `json:"messages"`
|
||||
Stream bool `json:"stream"`
|
||||
StreamOptions *StreamOptions `json:"stream_options"`
|
||||
MaxCompletionTokens *int `json:"max_completion_tokens"`
|
||||
// Deprecated: Use [ChatCompletionRequest.MaxCompletionTokens]
|
||||
MaxTokens *int `json:"max_tokens"`
|
||||
Seed *int `json:"seed"`
|
||||
Stop any `json:"stop"`
|
||||
@@ -93,6 +95,7 @@ type ChatCompletionRequest struct {
|
||||
TopP *float64 `json:"top_p"`
|
||||
ResponseFormat *ResponseFormat `json:"response_format"`
|
||||
Tools []api.Tool `json:"tools"`
|
||||
ContextLength *int `json:"context_length"`
|
||||
}
|
||||
|
||||
type ChatCompletion struct {
|
||||
@@ -475,8 +478,17 @@ func fromChatRequest(r ChatCompletionRequest) (*api.ChatRequest, error) {
|
||||
options["stop"] = stops
|
||||
}
|
||||
|
||||
if r.ContextLength != nil {
|
||||
options["num_ctx"] = *r.ContextLength
|
||||
}
|
||||
|
||||
// Deprecated: MaxTokens is deprecated, use MaxCompletionTokens instead
|
||||
if r.MaxTokens != nil {
|
||||
options["num_predict"] = *r.MaxTokens
|
||||
r.MaxCompletionTokens = r.MaxTokens
|
||||
}
|
||||
|
||||
if r.MaxCompletionTokens != nil {
|
||||
options["num_predict"] = *r.MaxCompletionTokens
|
||||
}
|
||||
|
||||
if r.Temperature != nil {
|
||||
@@ -962,6 +974,7 @@ func ChatMiddleware() gin.HandlerFunc {
|
||||
c.AbortWithStatusJSON(http.StatusBadRequest, NewError(http.StatusBadRequest, err.Error()))
|
||||
return
|
||||
}
|
||||
slog.Info("num_ctx", "num_ctx", chatReq.Options["num_ctx"])
|
||||
|
||||
if err := json.NewEncoder(&b).Encode(chatReq); err != nil {
|
||||
c.AbortWithStatusJSON(http.StatusInternalServerError, NewError(http.StatusInternalServerError, err.Error()))
|
||||
|
||||
@@ -7,7 +7,6 @@ import (
|
||||
"io"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"reflect"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
@@ -315,6 +314,42 @@ func TestChatMiddleware(t *testing.T) {
|
||||
Stream: &True,
|
||||
},
|
||||
},
|
||||
{
|
||||
name: "chat handler with context_length",
|
||||
body: `{
|
||||
"model": "test-model",
|
||||
"messages": [{"role": "user", "content": "Hello"}],
|
||||
"context_length": 4096
|
||||
}`,
|
||||
req: api.ChatRequest{
|
||||
Model: "test-model",
|
||||
Messages: []api.Message{{Role: "user", Content: "Hello"}},
|
||||
Options: map[string]any{
|
||||
"num_ctx": 4096.0, // float because JSON doesn't distinguish between float and int
|
||||
"temperature": 1.0,
|
||||
"top_p": 1.0,
|
||||
},
|
||||
Stream: &False,
|
||||
},
|
||||
},
|
||||
{
|
||||
name: "chat handler with max_completion_tokens",
|
||||
body: `{
|
||||
"model": "test-model",
|
||||
"messages": [{"role": "user", "content": "Hello"}],
|
||||
"max_completion_tokens": 2
|
||||
}`,
|
||||
req: api.ChatRequest{
|
||||
Model: "test-model",
|
||||
Messages: []api.Message{{Role: "user", Content: "Hello"}},
|
||||
Options: map[string]any{
|
||||
"num_predict": 2.0, // float because JSON doesn't distinguish between float and int
|
||||
"temperature": 1.0,
|
||||
"top_p": 1.0,
|
||||
},
|
||||
Stream: &False,
|
||||
},
|
||||
},
|
||||
{
|
||||
name: "chat handler error forwarding",
|
||||
body: `{
|
||||
@@ -359,7 +394,7 @@ func TestChatMiddleware(t *testing.T) {
|
||||
return
|
||||
}
|
||||
if diff := cmp.Diff(&tc.req, capturedRequest); diff != "" {
|
||||
t.Fatalf("requests did not match: %+v", diff)
|
||||
t.Fatalf("requests did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
if diff := cmp.Diff(tc.err, errResp); diff != "" {
|
||||
t.Fatalf("errors did not match for %s:\n%s", tc.name, diff)
|
||||
@@ -493,12 +528,14 @@ func TestCompletionsMiddleware(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
if capturedRequest != nil && !reflect.DeepEqual(tc.req, *capturedRequest) {
|
||||
t.Fatal("requests did not match")
|
||||
if capturedRequest != nil {
|
||||
if diff := cmp.Diff(tc.req, *capturedRequest); diff != "" {
|
||||
t.Fatalf("requests did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
}
|
||||
|
||||
if !reflect.DeepEqual(tc.err, errResp) {
|
||||
t.Fatal("errors did not match")
|
||||
if diff := cmp.Diff(tc.err, errResp); diff != "" {
|
||||
t.Fatalf("errors did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
|
||||
capturedRequest = nil
|
||||
@@ -577,12 +614,14 @@ func TestEmbeddingsMiddleware(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
if capturedRequest != nil && !reflect.DeepEqual(tc.req, *capturedRequest) {
|
||||
t.Fatal("requests did not match")
|
||||
if capturedRequest != nil {
|
||||
if diff := cmp.Diff(tc.req, *capturedRequest); diff != "" {
|
||||
t.Fatalf("requests did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
}
|
||||
|
||||
if !reflect.DeepEqual(tc.err, errResp) {
|
||||
t.Fatal("errors did not match")
|
||||
if diff := cmp.Diff(tc.err, errResp); diff != "" {
|
||||
t.Fatalf("errors did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
|
||||
capturedRequest = nil
|
||||
@@ -656,8 +695,8 @@ func TestListMiddleware(t *testing.T) {
|
||||
t.Fatalf("failed to unmarshal actual response: %v", err)
|
||||
}
|
||||
|
||||
if !reflect.DeepEqual(expected, actual) {
|
||||
t.Errorf("responses did not match\nExpected: %+v\nActual: %+v", expected, actual)
|
||||
if diff := cmp.Diff(expected, actual); diff != "" {
|
||||
t.Errorf("responses did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -722,8 +761,8 @@ func TestRetrieveMiddleware(t *testing.T) {
|
||||
t.Fatalf("failed to unmarshal actual response: %v", err)
|
||||
}
|
||||
|
||||
if !reflect.DeepEqual(expected, actual) {
|
||||
t.Errorf("responses did not match\nExpected: %+v\nActual: %+v", expected, actual)
|
||||
if diff := cmp.Diff(expected, actual); diff != "" {
|
||||
t.Errorf("responses did not match (-want +got):\n%s", diff)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -62,7 +62,13 @@ func (f Modelfile) CreateRequest(relativeDir string) (*api.CreateRequest, error)
|
||||
return nil, err
|
||||
}
|
||||
|
||||
req.Files = digestMap
|
||||
if req.Files == nil {
|
||||
req.Files = digestMap
|
||||
} else {
|
||||
for k, v := range digestMap {
|
||||
req.Files[k] = v
|
||||
}
|
||||
}
|
||||
case "adapter":
|
||||
path, err := expandPath(c.Args, relativeDir)
|
||||
if err != nil {
|
||||
|
||||
@@ -490,7 +490,6 @@ func TestParseFileParameters(t *testing.T) {
|
||||
"top_k 1": {"top_k", "1"},
|
||||
"top_p 1.0": {"top_p", "1.0"},
|
||||
"min_p 0.05": {"min_p", "0.05"},
|
||||
"tfs_z 1.0": {"tfs_z", "1.0"},
|
||||
"typical_p 1.0": {"typical_p", "1.0"},
|
||||
"repeat_last_n 1": {"repeat_last_n", "1"},
|
||||
"temperature 1.0": {"temperature", "1.0"},
|
||||
@@ -793,15 +792,20 @@ func createBinFile(t *testing.T, kv map[string]any, ti []llm.Tensor) (string, st
|
||||
}
|
||||
|
||||
func TestCreateRequestFiles(t *testing.T) {
|
||||
name, digest := createBinFile(t, nil, nil)
|
||||
n1, d1 := createBinFile(t, nil, nil)
|
||||
n2, d2 := createBinFile(t, map[string]any{"foo": "bar"}, nil)
|
||||
|
||||
cases := []struct {
|
||||
input string
|
||||
expected *api.CreateRequest
|
||||
}{
|
||||
{
|
||||
fmt.Sprintf("FROM %s", name),
|
||||
&api.CreateRequest{Files: map[string]string{name: digest}},
|
||||
fmt.Sprintf("FROM %s", n1),
|
||||
&api.CreateRequest{Files: map[string]string{n1: d1}},
|
||||
},
|
||||
{
|
||||
fmt.Sprintf("FROM %s\nFROM %s", n1, n2),
|
||||
&api.CreateRequest{Files: map[string]string{n1: d1, n2: d2}},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
67
template/command-r.gotmpl
Normal file
67
template/command-r.gotmpl
Normal file
@@ -0,0 +1,67 @@
|
||||
{{- if or .Tools .System }}<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>
|
||||
{{- if .Tools }}# Safety Preamble
|
||||
The instructions in this section override those in the task description and style guide sections. Don't answer questions that are harmful or immoral.
|
||||
|
||||
# System Preamble
|
||||
## Basic Rules
|
||||
You are a powerful conversational AI trained by Cohere to help people. You are augmented by a number of tools, and your job is to use and consume the output of these tools to best help the user. You will see a conversation history between yourself and a user, ending with an utterance from the user. You will then see a specific instruction instructing you what kind of response to generate. When you answer the user's requests, you cite your sources in your answers, according to those instructions.
|
||||
|
||||
{{ if .System }}# User Preamble
|
||||
{{ .System }}
|
||||
{{- end }}
|
||||
|
||||
## Available Tools
|
||||
Here is a list of tools that you have available to you:
|
||||
{{- range .Tools }}
|
||||
|
||||
```python
|
||||
def {{ .Function.Name }}(
|
||||
{{- range $name, $property := .Function.Parameters.Properties }}{{ $name }}: {{ $property.Type }}, {{ end }}) -> List[Dict]:
|
||||
'''{{ .Function.Description }}
|
||||
|
||||
{{- if .Function.Parameters.Properties }}
|
||||
|
||||
Args:
|
||||
{{- range $name, $property := .Function.Parameters.Properties }}
|
||||
{{ $name }} ({{ $property.Type }}): {{ $property.Description }}
|
||||
{{- end }}
|
||||
{{- end }}
|
||||
'''
|
||||
pass
|
||||
```
|
||||
{{- end }}
|
||||
{{- else if .System }}{{ .System }}
|
||||
{{- end }}<|END_OF_TURN_TOKEN|>
|
||||
{{- end }}
|
||||
{{- range .Messages }}
|
||||
{{- if eq .Role "system" }}
|
||||
{{- continue }}
|
||||
{{- end }}<|START_OF_TURN_TOKEN|>
|
||||
{{- if eq .Role "user" }}<|USER_TOKEN|>{{ .Content }}
|
||||
{{- if $.Tools }}<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>Write 'Action:' followed by a json-formatted list of actions that you want to perform in order to produce a good response to the user's last input. You can use any of the supplied tools any number of times, but you should aim to execute the minimum number of necessary actions for the input. You should use the `directly-answer` tool if calling the other tools is unnecessary. The list of actions you want to call should be formatted as a list of json objects, for example:
|
||||
```json
|
||||
[
|
||||
{
|
||||
"tool_name": title of the tool in the specification,
|
||||
"parameters": a dict of parameters to input into the tool as they are defined in the specs, or {} if it takes no parameters
|
||||
}
|
||||
]```
|
||||
{{- end }}
|
||||
{{- else if eq .Role "assistant" }}<|CHATBOT_TOKEN|>
|
||||
{{- if .Content }}{{ .Content }}
|
||||
{{- else if .ToolCalls }}
|
||||
Action: ```json
|
||||
[
|
||||
{{- range .ToolCalls }}
|
||||
{
|
||||
"tool_name": "{{ .Function.Name }}",
|
||||
"parameters": {{ .Function.Arguments }}
|
||||
}
|
||||
{{- end }}
|
||||
]```
|
||||
{{- end }}
|
||||
{{- else if eq .Role "tool" }}<|SYSTEM_TOKEN|><results>
|
||||
console_output: {{ .Content }}
|
||||
</results>
|
||||
{{- end }}<|END_OF_TURN_TOKEN|>
|
||||
{{- end }}<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
|
||||
6
template/command-r.json
Normal file
6
template/command-r.json
Normal file
@@ -0,0 +1,6 @@
|
||||
{
|
||||
"stop": [
|
||||
"<|START_OF_TURN_TOKEN|>",
|
||||
"<|END_OF_TURN_TOKEN|>"
|
||||
]
|
||||
}
|
||||
@@ -138,5 +138,9 @@
|
||||
{
|
||||
"template": "{% for message in messages %}{% if message['role'] == 'system' %}{% if message['content']%}{{'### System:\n' + message['content']+'\n\n'}}{% endif %}{% elif message['role'] == 'user' %}{{'### User:\n' + message['content']+'\n\n'}}{% elif message['role'] == 'assistant' %}{{'### Assistant:\n' + message['content']}}{% endif %}{% if loop.last and add_generation_prompt %}{{ '### Assistant:\n' }}{% endif %}{% endfor %}",
|
||||
"name": "solar-instruct"
|
||||
},
|
||||
{
|
||||
"template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% elif false == true %}{% set loop_messages = messages %}{% set system_message = 'You are Command-R, a brilliant, sophisticated, AI-assistant trained to assist human users by providing thorough responses. You are trained by Cohere.' %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% if system_message != false %}{{ '<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>' + system_message + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|START_OF_TURN_TOKEN|><|USER_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% elif message['role'] == 'assistant' %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' + content.strip() + '<|END_OF_TURN_TOKEN|>' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>' }}{% endif %}",
|
||||
"name": "command-r"
|
||||
}
|
||||
]
|
||||
|
||||
1
template/testdata/command-r.gotmpl/system-user-assistant-user
vendored
Normal file
1
template/testdata/command-r.gotmpl/system-user-assistant-user
vendored
Normal file
@@ -0,0 +1 @@
|
||||
<|START_OF_TURN_TOKEN|><|SYSTEM_TOKEN|>You are a helpful assistant.<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>I'm doing great. How can I help you today?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>I'd like to show off how chat templating works!<|END_OF_TURN_TOKEN|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
|
||||
1
template/testdata/command-r.gotmpl/user
vendored
Normal file
1
template/testdata/command-r.gotmpl/user
vendored
Normal file
@@ -0,0 +1 @@
|
||||
<|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
|
||||
1
template/testdata/command-r.gotmpl/user-assistant-user
vendored
Normal file
1
template/testdata/command-r.gotmpl/user-assistant-user
vendored
Normal file
@@ -0,0 +1 @@
|
||||
<|START_OF_TURN_TOKEN|><|USER_TOKEN|>Hello, how are you?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>I'm doing great. How can I help you today?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|USER_TOKEN|>I'd like to show off how chat templating works!<|END_OF_TURN_TOKEN|><|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
|
||||
Reference in New Issue
Block a user