© 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. 7-RES-JB
    11. 8827
    Prev
    Next
    INDEX
    Explanations

    phrases related to movement or actions involving different locations

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "and" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    odi
    -0.76
    intel
    -0.70
    gest
    -0.68
    zip
    -0.65
    itans
    -0.62
    microsoft
    -0.61
    ource
    -0.60
    rain
    -0.59
    ieth
    -0.58
    uta
    -0.58
    POSITIVE LOGITS
     thereby
    1.25
     consequently
    1.14
     thus
    1.13
     thence
    1.09
     thereafter
    0.96
     hence
    0.95
     subsequently
    0.93
     onwards
    0.89
     eventually
    0.89
     onward
    0.87
    Activations Density 0.316%

    No Known Activations