© 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
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 18758
    Prev
    Next
    INDEX
    Explanations

    verbs related to speculative or potential actions

    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
    314
    +0.16
    0.7%
    652
    +0.10
    0.5%
    219
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    624
    +0.16
    0.03
    688
    +0.10
    0.03
    22
    +0.09
    0.03
    Negative Logits
     Entered
    -0.68
     �
    -0.65
     Meter
    -0.61
     Lu
    -0.58
     Trace
    -0.58
     Shepard
    -0.58
     Jess
    -0.56
     stemming
    -0.56
     Bayer
    -0.55
     DM
    -0.55
    POSITIVE LOGITS
    theless
    1.58
    usterity
    1.42
    etheless
    1.40
    selves
    1.31
    terday
    1.29
    ecd
    1.22
    mosp
    1.21
    withstanding
    1.10
    self
    1.05
    yond
    1.05
    Activations Density 0.242%

    No Known Activations