© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7-RES-JB
    11. 3686
    Prev
    Next
    INDEX
    Explanations

    phrases related to allocation of resources or responsibilities

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of allocation, particularly in relation to resources or responsibilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    icho
    -0.86
    sonian
    -0.77
    bone
    -0.70
    faced
    -0.69
    jong
    -0.66
    breaking
    -0.66
    psc
    -0.65
     Polo
    -0.65
    Ther
    -0.64
    chers
    -0.64
    POSITIVE LOGITS
     allocations
    1.21
     allocation
    1.21
    entric
    1.01
    folios
    0.96
     allocated
    0.95
    itures
    0.94
     allocate
    0.92
    mble
    0.85
    eering
    0.83
    iture
    0.82
    Activations Density 0.027%

    No Known Activations