© 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. 10-RES-JB
    11. 885
    Prev
    Next
    INDEX
    Explanations

    phrases related to being done or achieved with a particular focus or purpose

    oai_token-act-pair · gpt-3.5-turbo

    expressions emphasizing exclusivity or singularity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    itely
    -0.67
     ASAP
    -0.64
    LI
    -0.63
    most
    -0.60
     repeatedly
    -0.60
     promptly
    -0.60
    immer
    -0.59
     downright
    -0.58
    ĺħ
    -0.56
    umm
    -0.56
    POSITIVE LOGITS
     reliant
    0.93
     focused
    0.87
     relying
    0.77
     responsible
    0.75
     comprised
    0.74
     foc
    0.73
     benefiting
    0.73
     concerned
    0.73
     devoted
    0.72
    rative
    0.71
    Activations Density 0.067%

    No Known Activations