© 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. 4-RES-JB
    11. 2868
    Prev
    Next
    INDEX
    Explanations

    terms related to financial market performance such as "highs" and "lows."

    oai_token-act-pair · gpt-3.5-turbo

    terms related to financial metrics, specifically peaks and troughs

    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
    oute
    -0.79
    Fac
    -0.71
    OTS
    -0.70
    nee
    -0.70
    Reply
    -0.67
    bus
    -0.67
     QR
    -0.65
     Galile
    -0.64
    Translation
    -0.63
    fac
    -0.63
    POSITIVE LOGITS
    terday
    1.11
    terness
    1.09
     lows
    1.03
    mith
    1.02
     highs
    0.99
    poons
    0.89
     heights
    0.88
    leep
    0.83
     mathemat
    0.83
    hift
    0.80
    Activations Density 0.025%

    No Known Activations