© 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. 0-RES-JB
    11. 13150
    Prev
    Next
    INDEX
    Explanations

    words related to specific types of materials, particularly polystyrene

    oai_token-act-pair · gpt-3.5-turbo

    repeated occurrences of the suffix "rene" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     UM
    -0.70
     SB
    -0.67
     Inf
    -0.65
     Chick
    -0.64
     haw
    -0.63
     differential
    -0.62
     Vaults
    -0.61
     Vari
    -0.61
     dat
    -0.60
     Level
    -0.60
    POSITIVE LOGITS
    rene
    4.61
    ren
    1.67
    lene
    1.51
    vana
    1.12
    rane
    1.11
    rina
    1.09
    rend
    1.04
    lde
    1.03
    rien
    1.02
    hene
    1.02
    Activations Density 0.013%

    No Known Activations