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. 1191
    Prev
    Next
    INDEX
    Explanations

    words related to fleeing or escaping from a place

    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
    238
    +0.14
    0.6%
    410
    +0.13
    0.6%
    207
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.14
    0.02
    410
    +0.13
    0.02
    238
    +0.10
    0.02
    Negative Logits
    essee
    -0.95
    wcsstore
    -0.66
    arist
    -0.65
    çͰ
    -0.64
    Deal
    -0.62
    inite
    -0.62
     MFT
    -0.61
    å¦
    -0.59
    umeric
    -0.59
    atsu
    -0.58
    POSITIVE LOGITS
     abroad
    0.82
     captivity
    0.71
    ring
    0.70
    ce
    0.68
     fleeing
    0.67
    rily
    0.67
     Mosul
    0.67
    itably
    0.67
     peacefully
    0.67
     Calais
    0.67
    Activations Density 0.030%

    No Known Activations