© 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. 4-RES-JB
    11. 19360
    Prev
    Next
    INDEX
    Explanations

    phrases related to soil science and land management

    oai_token-act-pair · gpt-3.5-turbo

    references to soil

    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
    yright
    -0.79
     Pist
    -0.75
    versions
    -0.74
    phan
    -0.72
    sych
    -0.71
    Lear
    -0.71
    "]=>
    -0.71
    cial
    -0.70
    general
    -0.69
    pload
    -0.67
    POSITIVE LOGITS
     moisture
    1.04
     soil
    1.00
     soils
    0.94
     erosion
    0.89
     microbes
    0.86
     flora
    0.81
    scape
    0.79
     arsenic
    0.78
     fertility
    0.78
     bacter
    0.78
    Activations Density 0.016%

    No Known Activations