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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 4-RES-JB
    6. 22744
    Prev
    Next
    INDEX
    Explanations

    phrases related to housing and residence

    oai_token-act-pair · gpt-3.5-turbo

    references to dwelling places or the act of dwelling

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

    No Comments

    Negative Logits
    enic
    -0.73
     athlet
    -0.72
    xon
    -0.69
    AAF
    -0.67
    boa
    -0.66
    channelAvailability
    -0.66
    maxwell
    -0.65
    apa
    -0.65
    avez
    -0.65
    henko
    -0.64
    POSITIVE LOGITS
     Dwell
    0.98
    ers
    0.98
     dwelling
    0.97
     dwell
    0.95
    erers
    0.89
    lords
    0.89
    erness
    0.83
    rite
    0.80
    erer
    0.80
    ants
    0.76
    Activations Density 0.016%

    No Known Activations