© 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. 1-RES-JB
    11. 9008
    Prev
    Next
    INDEX
    Explanations

    negative or unpleasant descriptions

    oai_token-act-pair · gpt-3.5-turbo

    negative descriptions and associations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    veland
    -0.90
    glas
    -0.82
    imates
    -0.81
    _>
    -0.78
    uay
    -0.74
    hene
    -0.73
    ynthesis
    -0.73
    imated
    -0.71
    oned
    -0.70
    avia
    -0.70
    POSITIVE LOGITS
     unpleasant
    1.08
     pse
    0.91
     cumbers
    0.89
     undesirable
    0.82
     unden
    0.81
     satell
    0.81
     unwelcome
    0.80
     undermin
    0.78
    ness
    0.77
     traged
    0.77
    Activations Density 0.012%

    No Known Activations