© 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. 17993
    Prev
    Next
    INDEX
    Explanations

    words related to transportation and commuting, specifically mentioning stops and stations

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the "RT" station and associated references

    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
    iasis
    -0.66
     Krug
    -0.66
    tained
    -0.65
    hots
    -0.63
    vasive
    -0.63
     [|
    -0.62
    ãĤ©
    -0.62
    BIP
    -0.61
    lihood
    -0.59
    rities
    -0.59
    POSITIVE LOGITS
    Ãī
    1.08
    ional
    0.94
    RT
    0.90
    uner
    0.84
    TY
    0.83
    TE
    0.80
    imes
    0.79
    ework
    0.79
    irtual
    0.78
    atcher
    0.76
    Activations Density 0.007%

    No Known Activations