© 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. 17045
    Prev
    Next
    INDEX
    Explanations

    instances where a person or entity performs an action in response to a specific trigger

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "when" in various contexts

    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
    gem
    -0.85
    rolet
    -0.79
    bold
    -0.77
    elman
    -0.73
    icker
    -0.73
    yre
    -0.72
    thal
    -0.69
    esi
    -0.69
    lez
    -0.68
    olphin
    -0.68
    POSITIVE LOGITS
    soever
    1.10
     necessary
    0.99
     needed
    0.97
     warranted
    0.94
     possible
    0.90
     emergencies
    0.89
     bored
    0.86
     convenient
    0.84
     requested
    0.83
     they
    0.82
    Activations Density 0.164%

    No Known Activations