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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 7641
    Prev
    Next
    INDEX
    Explanations

    instances of words related to setting up or establishing something, particularly installations and devices being put in place

    oai_token-act-pair · gpt-3.5-turbo

    terms related to the installation of equipment or systems

    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
    sem
    -0.73
    fax
    -0.72
    furt
    -0.69
    west
    -0.67
    zin
    -0.66
    ombs
    -0.66
    icz
    -0.64
    sweet
    -0.64
    externalActionCode
    -0.63
    vier
    -0.62
    POSITIVE LOGITS
     cameras
    0.80
     benches
    0.75
     installation
    0.75
     electrodes
    0.74
     installations
    0.71
     fixtures
    0.69
     sensors
    0.69
    igators
    0.69
     PV
    0.67
     fences
    0.67
    Activations Density 0.048%

    No Known Activations