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. 23812
    Prev
    Next
    INDEX
    Explanations

    instances of the word "turn" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered 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
    373
    +0.12
    0.5%
    195
    +0.11
    0.5%
    45
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    45
    +0.12
    0.02
    128
    +0.11
    0.02
    113
    +0.11
    0.02
    Negative Logits
    è¦ļéĨĴ
    -0.75
     Mellon
    -0.75
    ropolitan
    -0.73
    aimon
    -0.63
     franc
    -0.62
    ording
    -0.61
    gdala
    -0.59
    utical
    -0.59
     challenger
    -0.59
    heon
    -0.58
    POSITIVE LOGITS
    around
    0.89
    buck
    0.82
    wheel
    0.82
    coat
    0.80
    overs
    0.79
    about
    0.76
     Tables
    0.75
    over
    0.73
    ips
    0.73
    bull
    0.73
    Activations Density 0.031%

    No Known Activations