© 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. 20760
    Prev
    Next
    INDEX
    Explanations

    references to individuals with the last name "Val" or words containing "val" as a partial match

    oai_token-act-pair · gpt-3.5-turbo

    mentions of various individuals or entities

    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
    pread
    -0.71
    ModLoader
    -0.71
    washer
    -0.70
     EntityItem
    -0.68
    ħĭ
    -0.68
     plur
    -0.64
    Wan
    -0.63
    Publisher
    -0.63
    riad
    -0.61
    clips
    -0.61
    POSITIVE LOGITS
    uation
    1.14
    uations
    1.10
    uable
    0.99
    val
    0.94
    anches
    0.93
    uating
    0.91
    anche
    0.91
    uates
    0.91
    entin
    0.91
    uated
    0.89
    Activations Density 0.009%

    No Known Activations