© 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. 0-RES-JB
    11. 22363
    Prev
    Next
    INDEX
    Explanations

    names or terms related to people or entities

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Hor" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Pak
    -0.78
     bikini
    -0.73
    ribution
    -0.70
    pack
    -0.69
     cc
    -0.66
     prep
    -0.65
    cast
    -0.65
     api
    -0.65
     prepaid
    -0.64
     correctness
    -0.64
    POSITIVE LOGITS
     Hor
    3.80
    Hor
    2.71
     hor
    2.03
    hor
    1.87
     HOR
    1.82
     Horror
    1.23
     Vertical
    1.18
     Horowitz
    1.13
     Vert
    1.10
     Ter
    1.00
    Activations Density 0.021%

    No Known Activations