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 24576
    5. 8-RES_FS24576-JB
    6. 11443
    Prev
    Next
    INDEX
    Explanations

    keywords related to rules, regulations, and requirements

    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
    497
    +0.11
    0.5%
    39
    +0.09
    0.4%
    495
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.11
    -0.01
    566
    +0.09
    0.04
    570
    +0.09
    0.01
    Negative Logits
    soDeliveryDate
    -0.76
    Ö¼
    -0.70
    cffffcc
    -0.67
    soType
    -0.64
     embold
    -0.62
    Years
    -0.59
    ternity
    -0.57
     angered
    -0.57
    ester
    -0.56
    cil
    -0.56
    POSITIVE LOGITS
     except
    1.02
     equally
    0.91
    except
    0.82
     individually
    0.81
     Same
    0.80
     identical
    0.78
     HTTPS
    0.78
    same
    0.77
     simultaneously
    0.75
     separately
    0.75
    Activations Density 0.360%

    No Known Activations