© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 7231
    Prev
    Next
    INDEX
    Explanations

    words related to criminal activities and illegal actions, especially involving counterfeiting and immigration

    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
    496
    +0.16
    0.7%
    289
    +0.14
    0.7%
    7
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    588
    +0.16
    0.02
    289
    +0.14
    0.01
    284
    +0.12
    0.01
    Negative Logits
     Lisbon
    -0.69
    lihood
    -0.68
     Longh
    -0.63
     Dangerous
    -0.63
     Senegal
    -0.61
    Constructed
    -0.59
     Niger
    -0.59
     Oo
    -0.59
    houses
    -0.58
     Schumer
    -0.57
    POSITIVE LOGITS
    onial
    1.22
    inals
    1.07
    inally
    1.06
    inatory
    1.03
    inately
    0.99
    inating
    0.99
    ming
    0.98
    rim
    0.96
    etric
    0.96
    inant
    0.91
    Activations Density 0.023%

    No Known Activations