© 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. 2-RES-JB
    11. 452
    Prev
    Next
    INDEX
    Explanations

    the word "verse"

    oai_token-act-pair · gpt-4-turbo

    references to different versions of a particular setting or concept

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "verse," indicating a focus on different contexts or narratives within a fictional or creative framework

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    urated
    -0.95
    annis
    -0.74
    bered
    -0.73
    ICH
    -0.72
    istered
    -0.71
    dyl
    -0.69
    akings
    -0.68
    lished
    -0.66
    unders
    -0.65
    OUNT
    -0.65
    POSITIVE LOGITS
    verse
    1.21
    theless
    0.92
    lihood
    0.88
    terday
    0.86
    engers
    0.73
    backer
    0.71
    cliffe
    0.69
     Sins
    0.69
    itudinal
    0.68
     Whedon
    0.63
    Activations Density 0.016%

    No Known Activations