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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 0-RES-JB
    11. 1320
    Prev
    Next
    INDEX
    Explanations

    words related to adornment or embellishment

    oai_token-act-pair · gpt-3.5-turbo

    words related to fantasy or mythical themes

    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
     analyse
    -0.73
     favourable
    -0.72
     favour
    -0.70
     fulfil
    -0.69
    icum
    -0.68
     streams
    -0.66
     organising
    -0.66
    iyah
    -0.66
     encoding
    -0.63
    eston
    -0.62
    POSITIVE LOGITS
    eled
    3.49
    culture
    0.80
    ndra
    0.79
    eny
    0.78
    otics
    0.75
    els
    0.73
    eper
    0.73
     reeling
    0.72
    el
    0.72
    elle
    0.71
    Activations Density 0.017%

    No Known Activations