© 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. 8-RES-JB
    11. 11797
    Prev
    Next
    INDEX
    Explanations

    phrases related to seeking, improving, or evaluating things

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to continuous improvement and exploration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ptoms
    -0.65
    assi
    -0.65
    dm
    -0.64
    ettlement
    -0.64
    DIT
    -0.63
     CLSID
    -0.60
    FORMATION
    -0.60
     Extension
    -0.59
     Reps
    -0.58
    hab
    -0.58
    POSITIVE LOGITS
    ministic
    0.74
    andro
    0.71
    oin
    0.66
    oglu
    0.64
     underdog
    0.64
     pursu
    0.64
    enos
    0.63
     surprises
    0.63
     innocence
    0.62
    ãĤ¿
    0.62
    Activations Density 0.201%

    No Known Activations