© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18823
    Prev
    Next
    INDEX
    Explanations

    references to scientific breakthroughs and achievements

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    528
    +0.11
    0.5%
    624
    +0.10
    0.5%
    393
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    669
    +0.11
    0.04
    549
    +0.10
    0.03
    755
    +0.10
    0.03
    Negative Logits
    ittees
    -1.01
    ittee
    -0.89
    BuyableInstoreAndOnline
    -0.84
    DragonMagazine
    -0.80
    sic
    -0.78
    falls
    -0.76
    chers
    -0.71
    ups
    -0.70
    NPR
    -0.70
    ersen
    -0.68
    POSITIVE LOGITS
     causation
    1.02
     relativity
    1.01
     caus
    0.97
     morality
    0.89
     cognition
    0.88
     warfare
    0.88
     epist
    0.86
     human
    0.85
     rationality
    0.85
     phenomena
    0.84
    Activations Density 0.185%

    No Known Activations