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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9504
    Prev
    Next
    INDEX
    Explanations

    phrases related to academic institutions or research organizations

    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
    647
    +0.12
    0.5%
    313
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    730
    +0.12
    0.02
    146
    +0.11
    0.02
    313
    +0.10
    0.02
    Negative Logits
    theless
    -0.99
    mere
    -0.72
     tame
    -0.71
    erous
    -0.70
     Bundy
    -0.70
     sled
    -0.68
    dry
    -0.65
    taboola
    -0.64
    engers
    -0.64
    MET
    -0.64
    POSITIVE LOGITS
    itute
    0.78
    una
    0.78
     Herz
    0.72
    Research
    0.71
     Research
    0.71
    ollah
    0.67
    emi
    0.66
    uity
    0.66
    plementation
    0.64
     Tend
    0.64
    Activations Density 0.025%

    No Known Activations