© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 2402
    Prev
    Next
    INDEX
    Explanations

    names or terms related to locations, potentially with a mix of first names and surnames

    oai_token-act-pair · gpt-3.5-turbo

    references to specific names or locations, particularly involving "San" as a prefix

    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
     enactment
    -0.72
    glers
    -0.67
     bluff
    -0.66
     variance
    -0.65
     authorization
    -0.59
     tilt
    -0.59
     Cly
    -0.59
     date
    -0.58
     Ballistic
    -0.57
     Overt
    -0.57
    POSITIVE LOGITS
    ij士
    0.83
    ahar
    0.81
    ctic
    0.78
    abet
    0.75
    uay
    0.74
    itary
    0.74
    ieri
    0.72
    ãĥĻ
    0.72
    aney
    0.72
    anch
    0.70
    Activations Density 0.152%

    No Known Activations