Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 2862
    Prev
    Next
    INDEX
    Explanations

    verbs related to building, constructing, and creating various projects

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    636
    +0.14
    0.6%
    37
    +0.12
    0.5%
    367
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    636
    +0.14
    0.09
    36
    +0.12
    0.13
    362
    +0.10
    0.11
    Negative Logits
    enance
    -0.85
    ighed
    -0.81
    autions
    -0.78
    seek
    -0.76
     intercepted
    -0.69
    played
    -0.65
    rants
    -0.65
    esson
    -0.64
    hani
    -0.63
    kick
    -0.63
    POSITIVE LOGITS
     anew
    0.89
     makeshift
    0.86
     viable
    0.81
     prototype
    0.79
     believable
    0.79
     new
    0.77
     rapport
    0.74
     miniature
    0.74
     rudimentary
    0.73
     groundwork
    0.72
    Activations Density 2.014%

    No Known Activations