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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 45444
    Prev
    Next
    INDEX
    Explanations

    mentions of safety-related issues and concerns in various contexts

    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
    +0.13
    0.6%
    288
    +0.12
    0.5%
    523
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.06
    724
    +0.12
    0.05
    225
    +0.12
    0.04
    Negative Logits
     Butcher
    -0.61
     Wand
    -0.58
     nude
    -0.57
     Rand
    -0.57
     cameo
    -0.56
    oret
    -0.56
     Naked
    -0.56
     shepherd
    -0.55
     Pound
    -0.55
     soundtrack
    -0.55
    POSITIVE LOGITS
     levels
    0.85
     barriers
    0.77
    itism
    0.76
     morale
    0.75
    flows
    0.75
     pathways
    0.73
    ahime
    0.73
     expectations
    0.72
     nationwide
    0.71
     among
    0.71
    Activations Density 0.177%

    No Known Activations