© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 4745
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "ise" at various positions within words

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ise."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    etheless
    -0.83
    padding
    -0.68
    dies
    -0.66
    ajo
    -0.65
    女
    -0.61
    soDeliveryDate
    -0.60
    cffff
    -0.60
    é¾
    -0.59
    ļéĨĴ
    -0.59
    è¦ļéĨĴ
    -0.59
    POSITIVE LOGITS
    ppe
    0.90
    vich
    0.86
    veland
    0.83
    olate
    0.83
    vic
    0.77
    lda
    0.75
    ptic
    0.73
    vered
    0.72
    ller
    0.71
    ise
    0.70
    Activations Density 0.017%

    No Known Activations