© 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. 6767
    Prev
    Next
    INDEX
    Explanations

    numerical phrases indicating a specific quantity such as "two and a half" or ranges like "between seven and ten"

    oai_token-act-pair · gpt-3.5-turbo

    phrases containing the conjunction "and" along with variations of the article "a."

    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
    ãĤ©
    -0.88
    ONSORED
    -0.75
     Cla
    -0.73
    76561
    -0.71
     Majesty
    -0.71
    ãĤ´
    -0.70
     shroud
    -0.69
     FTA
    -0.69
    ¯¯¯¯¯¯¯¯
    -0.68
    PDATE
    -0.68
    POSITIVE LOGITS
    doms
    0.84
    legged
    0.81
    assic
    0.77
    minute
    0.76
    undred
    0.74
    azeera
    0.74
    ggles
    0.72
    anium
    0.72
    ionic
    0.70
    few
    0.70
    Activations Density 0.196%

    No Known Activations