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 1536
    5. 8-RES_FS1536-JB
    6. 1464
    Prev
    Next
    INDEX
    Explanations

    words related to positive impact or approval

    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
    71
    +0.09
    0.4%
    583
    +0.09
    0.4%
    455
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.09
    0.05
    89
    +0.09
    0.04
    700
    +0.09
    0.05
    Negative Logits
    anos
    -0.65
    bley
    -0.62
    atten
    -0.59
    mare
    -0.58
    kas
    -0.58
    zan
    -0.55
    mur
    -0.55
    mber
    -0.55
    rys
    -0.55
    inity
    -0.55
    POSITIVE LOGITS
     as
    0.64
    },"
    0.63
    onse
    0.61
    ]);
    0.59
    Parameters
    0.59
     };
    0.58
    FTWARE
    0.58
    isSpecialOrderable
    0.57
    atically
    0.56
    .:
    0.56
    Activations Density 0.933%

    No Known Activations