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

    Neuronpedia

    Jacobian LensNEW
    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. 42986
    Prev
    Next
    INDEX
    Explanations

    the use of the word "ef" and variations of the informal term “f” in various contexts

    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
    +0.20
    0.9%
    594
    +0.10
    0.5%
    88
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.20
    -0.00
    568
    +0.10
    0.01
    374
    +0.10
    0.01
    Negative Logits
     Zur
    -0.60
     archaic
    -0.57
     Ambro
    -0.56
     extant
    -0.56
     exclusion
    -0.54
    "},"
    -0.54
     Trafford
    -0.53
     Harrington
    -0.53
    }}}
    -0.53
     Hyde
    -0.52
    POSITIVE LOGITS
    otos
    0.93
    istas
    0.76
    amily
    0.74
    prot
    0.74
    erves
    0.73
    uly
    0.72
    acs
    0.71
    rites
    0.71
    isher
    0.70
    emia
    0.70
    Activations Density 0.013%

    No Known Activations