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. 41438
    Prev
    Next
    INDEX
    Explanations

    phrases that mention ages or durations expressed in years

    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
    208
    +0.11
    0.5%
    657
    +0.10
    0.5%
    601
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    657
    +0.11
    0.04
    361
    +0.10
    0.03
    198
    +0.10
    0.03
    Negative Logits
     ripple
    -0.63
    ulously
    -0.63
    illions
    -0.60
     ACTIONS
    -0.59
    iflower
    -0.59
    eus
    -0.58
    yrinth
    -0.55
     Dwell
    -0.54
     expansions
    -0.54
     contag
    -0.54
    POSITIVE LOGITS
     old
    1.10
     olds
    1.07
     ago
    0.96
    old
    0.90
     younger
    0.89
     Ago
    0.84
    Old
    0.82
     overdue
    0.80
     age
    0.79
    olds
    0.76
    Activations Density 0.041%

    No Known Activations