© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 19887
    Prev
    Next
    INDEX
    Explanations

    Japanese numerical expressions related to 'thousand'

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.11
    0.5%
    107
    +0.11
    0.5%
    365
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.11
    0.01
    107
    +0.11
    0.01
    683
    +0.11
    0.01
    Negative Logits
     guiActiveUn
    -0.74
    vana
    -0.73
     disenfranch
    -0.73
     bonded
    -0.72
    theless
    -0.69
     disarm
    -0.69
    orkshire
    -0.68
     unsus
    -0.67
     identifiable
    -0.67
     kne
    -0.66
    POSITIVE LOGITS
    Ī
    1.46
    ¤
    1.40
    ĺ
    1.35
    ¶
    1.31
    ¿
    1.26
    ij
    1.25
    ļ
    1.24
    IJ
    1.20
    Ķ
    1.20
    ¡
    1.18
    Activations Density 0.009%

    No Known Activations