© 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 EvalsExportsinterp-engineNEWAPI 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. 16395
    Prev
    Next
    INDEX
    Explanations

    words related to segments or parts, such as "part," "week," "game," and their numbers

    oai_token-act-pair · gpt-3.5-turbo

    terms related to divisions or segments of time, particularly in conjunction with parts or durations of events

    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
    estern
    -0.69
    Dialogue
    -0.67
    geist
    -0.60
    ELL
    -0.60
     Cth
    -0.59
     Andromeda
    -0.58
    lishes
    -0.58
    urus
    -0.57
    ONEY
    -0.57
    culus
    -0.57
    POSITIVE LOGITS
    ousand
    0.88
    undred
    0.87
     increments
    0.80
    tenance
    0.78
    clusive
    0.77
    enger
    0.76
    strument
    0.74
    holding
    0.74
    acters
    0.72
    iversary
    0.72
    Activations Density 0.123%

    No Known Activations