© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 14866
    Prev
    Next
    INDEX
    Explanations

    words related to cutting or reducing

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "slash" and its variations, indicating cuts or reductions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    otti
    -0.74
    sett
    -0.70
     Wonders
    -0.69
     Comfort
    -0.68
     Cons
    -0.67
    beh
    -0.67
    enn
    -0.66
     Observer
    -0.66
     observ
    -0.65
    Jerry
    -0.65
    POSITIVE LOGITS
     slash
    3.79
     slashed
    3.13
     slashing
    2.94
     Slash
    2.39
     slicing
    1.40
     slit
    1.27
     cuts
    1.26
     axe
    1.24
     cut
    1.20
     slug
    1.13
    Activations Density 0.030%

    No Known Activations