© 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. 488
    Prev
    Next
    INDEX
    Explanations

    the name "Ambrosie" or variations of this name

    oai_token-act-pair · gpt-3.5-turbo

    mentions of individuals, particularly those named "Ros" or related to that name

    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
     Helpful
    -0.65
     Semin
    -0.60
     CARE
    -0.59
     magnitude
    -0.59
     populous
    -0.57
     NIC
    -0.57
     Chinese
    -0.57
     special
    -0.57
     judges
    -0.56
     Citation
    -0.56
    POSITIVE LOGITS
    clerosis
    1.10
    ions
    1.01
    heet
    0.94
    boro
    0.92
    helle
    0.92
    ecast
    0.91
    orb
    0.90
    ktop
    0.89
    ingen
    0.89
    xual
    0.88
    Activations Density 0.008%

    No Known Activations