© 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. 2-RES-JB
    11. 667
    Prev
    Next
    INDEX
    Explanations

    words that contain the substring "hol"

    oai_token-act-pair · gpt-3.5-turbo

    words containing the sequence "hol"

    oai_token-act-pair · gpt-4-turbo

    references to the concept of "holidays."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FUL
    -0.71
    EED
    -0.68
     unauthorized
    -0.67
    INS
    -0.66
     fierce
    -0.65
    AGES
    -0.59
     tooth
    -0.58
    XL
    -0.57
    hips
    -0.56
     Drain
    -0.56
    POSITIVE LOGITS
    idays
    1.24
    ocaust
    1.21
    iday
    1.06
    stery
    1.05
    ocene
    0.96
    sten
    0.92
    mes
    0.91
    nob
    0.90
    inar
    0.90
    phe
    0.89
    Activations Density 0.007%

    No Known Activations