© 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. 4-RES-JB
    11. 940
    Prev
    Next
    INDEX
    Explanations

    organizations or entities

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Con" and its variations, indicating the neuron is looking for connections or references throughout the text

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

    No Comments

    Negative Logits
    TPS
    -0.67
     Rover
    -0.66
    gypt
    -0.65
     practicable
    -0.63
    OHN
    -0.63
    terday
    -0.63
     wip
    -0.61
     hydra
    -0.59
    taker
    -0.59
     Wem
    -0.57
    POSITIVE LOGITS
    stant
    1.21
    ventions
    1.11
    clusions
    1.09
    ference
    1.05
    cept
    1.05
    secut
    1.04
    joined
    0.99
    crete
    0.99
    cerning
    0.99
    ventional
    0.99
    Activations Density 0.014%

    No Known Activations