© 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. 4115
    Prev
    Next
    INDEX
    Explanations

    numbers related to financial transactions or statistics

    oai_token-act-pair · gpt-3.5-turbo

    numbers and values associated with statistics or data points

    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
    itud
    -0.78
    ]+
    -0.76
    inav
    -0.75
    dfx
    -0.72
     dummy
    -0.70
    ient
    -0.69
    ysis
    -0.69
    metic
    -0.69
    lder
    -0.67
    Tesla
    -0.65
    POSITIVE LOGITS
    920
    1.32
     Pigs
    0.68
     Fork
    0.68
    hap
    0.67
    iably
    0.66
     920
    0.63
     Grill
    0.62
     Pork
    0.61
    pins
    0.61
     Mei
    0.60
    Activations Density 0.003%

    No Known Activations