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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11405
    Prev
    Next
    INDEX
    Explanations

    long phrases focusing on specific legal or criminal actions in a negative context

    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
    632
    +0.12
    0.5%
    544
    +0.11
    0.5%
    598
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    227
    +0.12
    0.03
    153
    +0.11
    0.03
    544
    +0.10
    0.03
    Negative Logits
    thumbnails
    -0.84
    ãĥ¯ãĥ³
    -0.82
    iton
    -0.76
    ãĥĵ
    -0.74
    Wikipedia
    -0.73
    Quant
    -0.73
    Ult
    -0.73
    TPP
    -0.72
     CET
    -0.72
    quant
    -0.72
    POSITIVE LOGITS
     daughter
    1.01
    aughters
    0.99
     Females
    0.93
     kid
    0.93
     Girl
    0.92
     girl
    0.91
     child
    0.90
     babe
    0.89
     youngest
    0.89
     childhood
    0.88
    Activations Density 0.217%

    No Known Activations