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 1536
    5. 8-RES_FS1536-JB
    6. 584
    Prev
    Next
    INDEX
    Explanations

    information related to events or exhibitions

    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
    87
    +0.33
    2.0%
    447
    +0.24
    1.4%
    756
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.33
    -0.11
    459
    +0.24
    0.17
    756
    +0.11
    -0.13
    Negative Logits
    etheless
    -0.67
     attRot
    -0.52
     Canaver
    -0.51
    QUIRE
    -0.50
    also
    -0.48
    yssey
    -0.47
    )))
    -0.47
     GOODMAN
    -0.47
    RESULTS
    -0.44
    quickShipAvailable
    -0.43
    POSITIVE LOGITS
     etc
    0.52
     or
    0.49
     versus
    0.42
    -,
    0.41
     ('
    0.41
     =>
    0.39
     (=
    0.38
     ->
    0.38
     bombed
    0.37
     queues
    0.36
    Activations Density 13.189%

    No Known Activations