© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 1-RES-JB
    11. 3554
    Prev
    Next
    INDEX
    Explanations

    terrain-related words and phrases

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of terrain

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    erd
    -0.79
    ership
    -0.75
    ulate
    -0.74
    monds
    -0.74
    aghetti
    -0.72
    conn
    -0.72
    Discussion
    -0.71
    umption
    -0.71
    ever
    -0.70
    ofi
    -0.69
    POSITIVE LOGITS
     terrain
    1.14
    schild
    0.94
     vanquished
    0.78
     guiActiveUn
    0.77
     conquered
    0.76
     inhabited
    0.74
     airst
    0.73
     scenery
    0.71
     hardened
    0.70
     Conditions
    0.70
    Activations Density 0.007%

    No Known Activations