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. 933
    Prev
    Next
    INDEX
    Explanations

    personal pronouns in conjunction with actions directed towards a specific individual

    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
    15
    +0.11
    0.5%
    762
    +0.10
    0.5%
    471
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.11
    0.13
    139
    +0.10
    0.11
    15
    +0.10
    0.12
    Negative Logits
    izabeth
    -0.71
     Gems
    -0.68
    andel
    -0.65
    mary
    -0.63
    ocument
    -0.61
    urban
    -0.60
     pregn
    -0.60
    services
    -0.60
    ylon
    -0.60
    USE
    -0.59
    POSITIVE LOGITS
     Majesty
    1.24
    panic
    1.17
    zbollah
    1.01
    resy
    1.00
    sing
    0.98
    ading
    0.92
    eded
    0.92
     Emin
    0.88
    sed
    0.85
    pan
    0.83
    Activations Density 1.397%

    No Known Activations