© 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 1536
    9. 8-RES_FS1536-JB
    11. 1119
    Prev
    Next
    INDEX
    Explanations

    verbs related to being capable or proficient at something

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    447
    +0.26
    1.3%
    480
    +0.25
    1.2%
    481
    +0.19
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    464
    +0.26
    0.18
    480
    +0.25
    0.04
    266
    +0.19
    0.12
    Negative Logits
     Niet
    -0.68
     Borders
    -0.62
     Grail
    -0.62
     Seym
    -0.60
     Azerb
    -0.59
     Clarkson
    -0.56
     Frie
    -0.56
     Nare
    -0.55
     Oval
    -0.54
     prest
    -0.53
    POSITIVE LOGITS
     ][
    0.91
     ]
    0.78
    redients
    0.69
     ::
    0.68
     ];
    0.67
    _
    0.67
     ].
    0.67
     )
    0.66
     );
    0.65
    actionDate
    0.62
    Activations Density 2.187%

    No Known Activations