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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 11161
    Prev
    Next
    INDEX
    Explanations

    expressions related to personal desires and actions

    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
    53
    +0.11
    0.5%
    43
    +0.10
    0.5%
    644
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.11
    0.03
    49
    +0.10
    0.02
    431
    +0.10
    0.02
    Negative Logits
     precinct
    -0.66
     plank
    -0.61
     transgress
    -0.60
     remarkably
    -0.57
     strikingly
    -0.57
     indec
    -0.56
     extremes
    -0.56
     disbelief
    -0.56
    ftime
    -0.56
     violating
    -0.55
    POSITIVE LOGITS
     someday
    0.91
    hma
    0.77
    htt
    0.76
    bley
    0.70
     existed
    0.70
     listened
    0.69
    instead
    0.68
    itia
    0.68
    innie
    0.67
    ucket
    0.66
    Activations Density 0.086%

    No Known Activations