© 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. 26561
    Prev
    Next
    INDEX
    Explanations

    social media handles and hashtags

    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
    218
    +0.14
    0.7%
    314
    +0.14
    0.6%
    160
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.14
    0.01
    176
    +0.14
    0.01
    627
    +0.13
    0.01
    Negative Logits
     Advisory
    -0.69
     waived
    -0.68
     assess
    -0.66
     hardness
    -0.65
     gauge
    -0.64
     indicted
    -0.64
     "âĢ¦
    -0.62
     ambiguity
    -0.62
     Chao
    -0.61
     equival
    -0.61
    POSITIVE LOGITS
    endif
    1.12
    nob
    1.04
    mast
    1.04
    realDonaldTrump
    1.03
    marriage
    1.03
    photos
    1.01
    max
    1.01
    etc
    1.01
    RIP
    0.99
    tips
    0.98
    Activations Density 0.013%

    No Known Activations