© 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. 12166
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a continuation in a series or story

    oai_token-act-pair · gpt-3.5-turbo

    repetitive phrases indicating continuity or ongoing status

    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
    代
    -0.77
    ocol
    -0.77
    ãĥ³ãĤ¸
    -0.75
    rule
    -0.69
     opposite
    -0.69
    é¾
    -0.68
    duct
    -0.68
    ultimate
    -0.67
     è£ıç
    -0.66
    %]
    -0.64
    POSITIVE LOGITS
    Enjoy
    0.76
    noon
    0.75
    birth
    0.75
     Nadu
    0.75
     retaining
    0.75
    heres
    0.74
     reeling
    0.73
     retains
    0.72
    ega
    0.67
     technically
    0.66
    Activations Density 0.018%

    No Known Activations