© 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. 3354
    Prev
    Next
    INDEX
    Explanations

    words related to predictions or forecasts

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating predictions or forecasts

    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
    unal
    -0.74
    atha
    -0.74
    tha
    -0.73
    Bio
    -0.72
    xia
    -0.72
    ste
    -0.70
    zanne
    -0.70
    tan
    -0.69
    adra
    -0.68
    kat
    -0.67
    POSITIVE LOGITS
     predicts
    1.15
     predicted
    1.14
     predict
    1.03
     predictions
    1.02
     forecasts
    0.95
     predicting
    0.95
     prediction
    0.93
     doom
    0.88
     Prediction
    0.87
     forecast
    0.84
    Activations Density 0.007%

    No Known Activations