© 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 49152
    9. 8-RES_FS49152-JB
    11. 37375
    Prev
    Next
    INDEX
    Explanations

    references to specific organizations and locations within a community context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    481
    +0.25
    1.1%
    712
    +0.11
    0.5%
    314
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.25
    0.05
    325
    +0.11
    0.04
    537
    +0.10
    0.04
    Negative Logits
     silence
    -0.76
     ende
    -0.69
     physic
    -0.62
     Kinnikuman
    -0.61
    \":
    -0.61
    >[
    -0.61
    theless
    -0.60
     gadgets
    -0.59
     nonex
    -0.59
    acca
    -0.59
    POSITIVE LOGITS
     Facility
    0.89
    astery
    0.89
     Association
    0.80
     Authority
    0.80
    ominium
    0.79
    itent
    0.77
     Station
    0.75
     Resort
    0.75
     Festival
    0.71
     Estate
    0.71
    Activations Density 0.175%

    No Known Activations