© 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. 11812
    Prev
    Next
    INDEX
    Explanations

    tech specifications related to storage capacity

    oai_token-act-pair · gpt-3.5-turbo

    specifications related to computer hardware, particularly memory and storage capacities

    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
    icians
    -0.67
    ople
    -0.67
    neum
    -0.67
    eering
    -0.64
     discipl
    -0.63
    unct
    -0.63
    hurst
    -0.63
    gypt
    -0.61
     trending
    -0.60
    eer
    -0.60
    POSITIVE LOGITS
    abytes
    1.11
     GDDR
    1.08
     RAM
    0.99
    ASE
    0.88
    abyte
    0.88
    GB
    0.87
     SSD
    0.84
    bps
    0.83
     heap
    0.81
     DDR
    0.81
    Activations Density 0.023%

    No Known Activations