© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 791
    Prev
    Next
    INDEX
    Explanations

    Initial forms of words related to being inducted or induction

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "induct" and its variations related to formal recognitions

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

    No Comments

    Negative Logits
     PLAN
    -0.71
     Coastal
    -0.68
     Grimm
    -0.67
     REAL
    -0.67
     nerv
    -0.65
     Weed
    -0.63
     Dim
    -0.61
    mington
    -0.61
     JPMorgan
    -0.61
     interfering
    -0.60
    POSITIVE LOGITS
     induct
    1.39
    untarily
    1.01
     induction
    1.00
    naires
    0.87
    interstitial
    0.85
    naire
    0.82
    ãĤĭ
    0.82
    uably
    0.81
    idable
    0.81
    Reviewer
    0.79
    Activations Density 0.016%

    No Known Activations