© 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 12288
    9. 8-RES_FS12288-JB
    11. 10169
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology and software updates

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    72
    +0.10
    0.4%
    244
    +0.10
    0.4%
    108
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    72
    +0.10
    0.03
    312
    +0.10
    0.02
    423
    +0.10
    0.03
    Negative Logits
     Secondly
    -0.68
     EntityItem
    -0.68
    azeera
    -0.68
     Cosponsors
    -0.66
    asus
    -0.65
    ammy
    -0.65
    assic
    -0.62
    resa
    -0.61
     secondly
    -0.60
    esar
    -0.60
    POSITIVE LOGITS
     anyway
    0.86
     though
    0.76
     sake
    0.71
    here
    0.71
     suffice
    0.71
     anyways
    0.69
    adays
    0.68
     unim
    0.67
     purposes
    0.67
     scen
    0.65
    Activations Density 0.090%

    No Known Activations