© 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 49152
    9. 8-RES_FS49152-JB
    11. 26577
    Prev
    Next
    INDEX
    Explanations

    geographical locations and cities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    481
    +0.14
    0.6%
    288
    +0.12
    0.5%
    447
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    201
    +0.14
    0.05
    757
    +0.12
    0.04
    245
    +0.12
    0.05
    Negative Logits
     intro
    -0.72
     explanation
    -0.66
     deduction
    -0.64
     replay
    -0.61
    idious
    -0.61
    IU
    -0.61
    letter
    -0.61
     rapport
    -0.60
    ascript
    -0.59
     ICO
    -0.59
    POSITIVE LOGITS
     Airport
    1.15
     Boulevard
    1.08
     Avenue
    1.05
     Station
    1.00
     airport
    0.99
     Blvd
    0.96
     Harbor
    0.94
     Harbour
    0.93
     shores
    0.92
     Junction
    0.91
    Activations Density 0.159%

    No Known Activations