© 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. 4-RES-JB
    11. 20164
    Prev
    Next
    INDEX
    Explanations

    phrases related to directions and positions, particularly terms indicating movement in a certain direction (upstream, downstream, uphill, downhill)

    oai_token-act-pair · gpt-3.5-turbo

    references to directional terms related to movement or flow, particularly focusing on upstream and downstream

    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
     Manson
    -0.77
    ggle
    -0.75
    phe
    -0.74
    andum
    -0.72
    ander
    -0.72
    pmwiki
    -0.71
    iasis
    -0.70
    chin
    -0.70
    abilia
    -0.69
    "},"
    -0.69
    POSITIVE LOGITS
     upstream
    1.10
     downstream
    1.08
    side
    0.75
    ward
    0.73
    é¾
    0.71
    raint
    0.71
    stream
    0.71
    wards
    0.69
    edIn
    0.68
    raints
    0.66
    Activations Density 0.008%

    No Known Activations