Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 9-RES-JB
    6. 22196
    Prev
    Next
    INDEX
    Explanations

    instances where events and actions are taking place in a political or organizational context

    oai_token-act-pair · gpt-3.5-turbo

    statements regarding the legal status or events surrounding individuals, particularly relating to their release or conviction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    classes
    -0.79
     sexes
    -0.79
     Ort
    -0.75
     invent
    -0.75
     backgrounds
    -0.74
     notebooks
    -0.72
     inhab
    -0.71
     capacities
    -0.70
     profiles
    -0.70
     classes
    -0.68
    POSITIVE LOGITS
     imminent
    1.18
     inevitable
    1.13
     bitters
    1.11
     postponed
    1.01
     instantaneous
    0.97
     irreversible
    0.95
     herald
    0.95
     abrupt
    0.94
     delayed
    0.94
     effected
    0.90
    Activations Density 0.454%

    No Known Activations