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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1248
    Prev
    Next
    INDEX
    Explanations

    discussions related to policies, human rights, and societal issues

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    373
    +0.12
    0.6%
    21
    +0.10
    0.5%
    234
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.12
    0.11
    145
    +0.10
    0.11
    680
    +0.10
    0.09
    Negative Logits
    tumblr
    -0.87
     stret
    -0.75
     masturb
    -0.69
     screws
    -0.69
    SPONSORED
    -0.69
     proved
    -0.69
    ?!
    -0.67
    MSN
    -0.65
     nig
    -0.65
    Synopsis
    -0.64
    POSITIVE LOGITS
     analyst
    1.02
    earcher
    0.94
     researcher
    0.93
     sociology
    0.90
    heastern
    0.89
     Liberties
    0.89
     consultancy
    0.87
    earch
    0.87
     spokeswoman
    0.86
     Consulting
    0.84
    Activations Density 3.273%

    No Known Activations