INDEX

Explanations

details about historical events and military actions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rowling

-0.90

Consumer

-0.87

 partName

-0.77

 Blockchain

-0.76

Owner

-0.76

 Ethereum

-0.75

thereum

-0.75

properties

-0.73

Netflix

-0.73

weet

-0.72

POSITIVE LOGITS

 stationed

1.64

 battalion

1.59

 troop

1.55

 troops

1.54

 infantry

1.51

 Infantry

1.47

 garrison

1.46

 reconnaissance

1.44

 artillery

1.44

 soldiers

1.44

Activations Density 2.985%

details about historical events and military actions

No Comments

No Known Activations