© 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. 41807
    Prev
    Next
    INDEX
    Explanations

    numbers and attributes associated with specific entities or actions

    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
    756
    +0.08
    0.3%
    383
    +0.07
    0.3%
    125
    +0.07
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    190
    +0.08
    0.00
    111
    +0.07
    0.00
    729
    +0.07
    0.00
    Negative Logits
     Donation
    -0.74
    ISON
    -0.69
    bull
    -0.68
     Traps
    -0.68
    Ire
    -0.68
    atra
    -0.67
    uala
    -0.66
     Rath
    -0.66
    ouf
    -0.65
     Guards
    -0.64
    POSITIVE LOGITS
     altern
    0.75
     futures
    0.65
     expiration
    0.63
     serving
    0.62
     clock
    0.62
     karma
    0.59
     repl
    0.59
     expire
    0.58
     emanc
    0.58
     reincarn
    0.57
    Activations Density 0.002%

    No Known Activations