© 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. 8-RES-JB
    11. 949
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial investments or business portfolios

    oai_token-act-pair · gpt-3.5-turbo

    references to "portfolio" and "lineup" in the context of investments or product collections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nai
    -1.06
    abad
    -0.81
    ween
    -0.72
    rupt
    -0.71
    tale
    -0.70
    tarians
    -0.69
    OTO
    -0.68
    ode
    -0.68
    ths
    -0.66
    tein
    -0.65
    POSITIVE LOGITS
     mates
    0.89
     consisting
    0.85
    selection
    0.78
     comprising
    0.76
     comprises
    0.73
     choices
    0.73
     portfolio
    0.72
     INCLUD
    0.72
     spans
    0.72
     spanning
    0.71
    Activations Density 0.117%

    No Known Activations