© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 1597
    Prev
    Next
    INDEX
    Explanations

    Twitter-related content and social media accounts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    706
    +0.11
    0.5%
    294
    +0.11
    0.5%
    13
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    396
    +0.11
    0.04
    13
    +0.11
    0.04
    23
    +0.10
    0.04
    Negative Logits
    cised
    -0.73
    cision
    -0.71
     Scandinavian
    -0.66
     Samoa
    -0.65
    cise
    -0.64
     refriger
    -0.64
    olphin
    -0.61
    entially
    -0.61
     triangles
    -0.61
    istries
    -0.60
    POSITIVE LOGITS
     hashtag
    0.96
     hasht
    0.88
     (@
    0.88
    @@@@@@@@
    0.85
     "@
    0.83
    storms
    0.82
    TW
    0.80
     handles
    0.80
     feeds
    0.80
    username
    0.80
    Activations Density 0.047%

    No Known Activations