© 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. 1-RES-JB
    11. 19958
    Prev
    Next
    INDEX
    Explanations

    the article "a" followed by an adjective, particularly one that conveys positive sentiment

    oai_token-act-pair · gpt-3.5-turbo

    the indefinite article "a" as well as other occurrences of the letter "a"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chuk
    -0.68
     Jagu
    -0.64
     Vaugh
    -0.63
    Edit
    -0.60
    Ont
    -0.58
    Catal
    -0.58
    Allow
    -0.57
    âĿ
    -0.54
    =[
    -0.54
     admits
    -0.54
    POSITIVE LOGITS
     huge
    0.82
     lot
    0.82
     bunch
    0.80
     tremend
    0.77
    merce
    0.76
     plethora
    0.76
     lengthy
    0.75
    rouse
    0.74
    cknowled
    0.73
     sizeable
    0.73
    Activations Density 0.617%

    No Known Activations