© 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 12288
    9. 8-RES_FS12288-JB
    11. 10115
    Prev
    Next
    INDEX
    Explanations

    historical information

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    373
    +0.14
    0.6%
    241
    +0.11
    0.5%
    702
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    702
    +0.14
    0.04
    432
    +0.11
    0.04
    632
    +0.10
    0.03
    Negative Logits
    plet
    -0.68
    milo
    -0.65
     manif
    -0.64
    gren
    -0.63
    wcs
    -0.61
     servicing
    -0.61
     lobb
    -0.60
    ciating
    -0.60
     incarcer
    -0.60
    urat
    -0.59
    POSITIVE LOGITS
     edit
    0.75
    geist
    0.72
    Spoiler
    0.69
     onwards
    0.68
    onal
    0.67
     ]
    0.64
     âĨij
    0.64
     Edit
    0.63
     Blizzard
    0.62
     Khal
    0.62
    Activations Density 0.026%

    No Known Activations