© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 26841
    Prev
    Next
    INDEX
    Explanations

    URLs and links related to social media and web addresses

    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
    373
    +0.17
    0.8%
    297
    +0.12
    0.6%
    383
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.17
    0.02
    373
    +0.12
    -0.00
    592
    +0.09
    0.02
    Negative Logits
     Lank
    -0.72
    ©¶æ
    -0.68
     Krugman
    -0.67
     Ramsey
    -0.65
     Erit
    -0.65
     Samoa
    -0.65
     Blumenthal
    -0.65
     Mayo
    -0.65
     Afgh
    -0.64
     unpublished
    -0.64
    POSITIVE LOGITS
    twitter
    1.10
    goo
    1.08
    www
    1.07
    twitch
    1.01
    facebook
    1.00
    youtube
    0.98
    localhost
    0.96
    tiny
    0.92
    mys
    0.88
    steam
    0.85
    Activations Density 0.021%

    No Known Activations