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. 0-RES-JB
    6. 22954
    Prev
    Next
    INDEX
    Explanations

    words related to stairs or staircases

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "stairs" and related terms indicating vertical movement or structures

    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
    qqa
    -0.75
     Cameroon
    -0.75
    Tag
    -0.71
    ann
    -0.67
    doms
    -0.66
    >[
    -0.66
    Pos
    -0.66
    uster
    -0.65
    iod
    -0.64
     Hav
    -0.63
    POSITIVE LOGITS
     stairs
    3.72
     stair
    2.79
     staircase
    2.71
    stairs
    2.31
     hallway
    1.80
     upstairs
    1.75
     elevator
    1.74
     balcony
    1.72
     downstairs
    1.70
     escal
    1.66
    Activations Density 0.017%

    No Known Activations