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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41407
    Prev
    Next
    INDEX
    Explanations

    terms related to accountability and missing persons or entities

    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
    288
    +0.12
    0.5%
    688
    +0.10
    0.4%
    541
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    688
    +0.12
    0.01
    100
    +0.10
    0.01
    503
    +0.09
    0.01
    Negative Logits
     DRAG
    -0.71
    ĺħ
    -0.70
     injunction
    -0.65
    USE
    -0.63
    OPLE
    -0.63
     Angels
    -0.63
     slicing
    -0.58
     Devils
    -0.58
    ÏĢ
    -0.58
    creen
    -0.56
    POSITIVE LOGITS
    able
    1.14
    ledged
    1.07
    ability
    1.06
    atile
    1.04
    ables
    1.01
    ably
    1.01
    ibly
    0.99
    ibility
    0.99
    ivable
    0.97
    ationally
    0.97
    Activations Density 0.003%

    No Known Activations