Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 7497
    Prev
    Next
    INDEX
    Explanations

    instances of the word "to" within various contexts

    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
    608
    +0.17
    0.8%
    718
    +0.15
    0.7%
    49
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    49
    +0.17
    0.04
    718
    +0.15
    0.04
    608
    +0.12
    0.04
    Negative Logits
    LAN
    -0.65
    skinned
    -0.63
    UR
    -0.62
    Mur
    -0.62
    Child
    -0.61
    pie
    -0.60
    grain
    -0.60
    expected
    -0.59
    calling
    -0.59
    majority
    -0.58
    POSITIVE LOGITS
     demonstrate
    1.27
     educate
    1.25
     minimize
    1.21
     conserve
    1.21
     ensure
    1.20
     facilitate
    1.19
     satisfy
    1.19
     preserve
    1.18
     avoid
    1.18
     enhance
    1.18
    Activations Density 0.098%

    No Known Activations