© 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. 3-RES-JB
    11. 2846
    Prev
    Next
    INDEX
    Explanations

    numerical approximations expressed in text

    oai_token-act-pair · gpt-3.5-turbo

    terms related to approximations and estimates

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    LINE
    -0.78
    nen
    -0.76
     Bomber
    -0.73
    spons
    -0.71
    gae
    -0.69
    cloth
    -0.68
    belt
    -0.67
    Series
    -0.67
    glass
    -0.66
    stru
    -0.66
    POSITIVE LOGITS
    imates
    1.19
     approx
    1.09
    imated
    1.02
     approximate
    0.99
    imating
    0.97
     approximation
    0.88
    imate
    0.85
    urations
    0.80
    ITNESS
    0.79
    Ń·
    0.74
    Activations Density 0.016%

    No Known Activations