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. 25691
    Prev
    Next
    INDEX
    Explanations

    references to groups of people or stakeholders

    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
    657
    +0.11
    0.5%
    575
    +0.11
    0.5%
    756
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.11
    0.07
    537
    +0.11
    0.05
    657
    +0.10
    0.05
    Negative Logits
     Peel
    -0.62
    TEXTURE
    -0.61
    Ī
    -0.60
     Racer
    -0.56
    ASED
    -0.56
    ¥µ
    -0.55
    ãĤ¶
    -0.55
     Tropical
    -0.55
    RW
    -0.54
    Ħ¢
    -0.53
    POSITIVE LOGITS
    hip
    1.53
    hips
    1.47
    pace
    1.19
    '
    1.19
    paces
    1.18
    peak
    1.13
    ']
    1.05
    cript
    1.02
    erv
    1.00
    ourcing
    1.00
    Activations Density 0.243%

    No Known Activations