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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 502
    Prev
    Next
    INDEX
    Explanations

    terms related to immigration, deportation, illegal immigration, and foreign-born individuals

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    575
    +0.16
    0.7%
    343
    +0.12
    0.5%
    469
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.16
    0.09
    288
    +0.12
    0.09
    469
    +0.11
    0.05
    Negative Logits
    MpServer
    -0.81
    idges
    -0.76
    eenth
    -0.75
    olls
    -0.73
    MET
    -0.72
    çͰ
    -0.71
    @#&
    -0.71
    Ü
    -0.71
    POSE
    -0.70
    rys
    -0.68
    POSITIVE LOGITS
     visas
    1.13
     immigrants
    1.08
     detention
    1.05
     visa
    1.00
     immigration
    0.98
     deported
    0.98
     fleeing
    0.97
     deport
    0.94
     deportation
    0.93
     detain
    0.93
    Activations Density 0.828%

    No Known Activations