© 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. 14316
    Prev
    Next
    INDEX
    Explanations

    words related to locations or territories described as 'held' by a certain group

    oai_token-act-pair · gpt-3.5-turbo

    instances of areas or regions that are under control or influence by rebel groups

    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
    Phys
    -0.64
     Express
    -0.62
    xxx
    -0.58
    ADD
    -0.58
     Typh
    -0.57
     impress
    -0.57
     opportun
    -0.56
     Traffic
    -0.55
    IDENT
    -0.55
    cler
    -0.55
    POSITIVE LOGITS
    held
    1.58
    scl
    1.08
    sburg
    0.98
    quartered
    0.97
    erness
    0.97
    hold
    0.97
    holding
    0.93
    terness
    0.91
    quarters
    0.88
     Held
    0.87
    Activations Density 0.008%

    No Known Activations