Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 7474
    Prev
    Next
    INDEX
    Explanations

    complex terms related to mapping or categorization

    oai_token-act-pair · gpt-3.5-turbo

    terms related to effort and attention to detail

    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
     deviations
    -0.80
     redund
    -0.72
     concessions
    -0.72
     adaptations
    -0.72
     characterization
    -0.70
     involve
    -0.68
     reservations
    -0.68
     exceptions
    -0.68
     venues
    -0.68
     embodiments
    -0.67
    POSITIVE LOGITS
    Dutch
    0.79
    Supporters
    0.74
    Leaks
    0.73
    lesi
    0.72
    lee
    0.72
    atform
    0.71
    odox
    0.69
    anon
    0.68
    yssey
    0.68
    clinton
    0.67
    Activations Density 0.000%

    No Known Activations