© 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. 12547
    Prev
    Next
    INDEX
    Explanations

    patterns where one thing is likened to another

    oai_token-act-pair · gpt-3.5-turbo

    instances of words related to similarity or resemblance

    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
    arta
    -0.82
    FT
    -0.79
    alloc
    -0.74
    imb
    -0.73
    oard
    -0.71
    ourse
    -0.69
    acht
    -0.68
    ttp
    -0.68
    gard
    -0.67
    Dispatch
    -0.66
    POSITIVE LOGITS
    lihood
    1.74
    lier
    0.94
    liness
    0.85
     likeness
    0.85
     resembling
    0.80
    liest
    0.78
     resembles
    0.76
     favorably
    0.76
    atile
    0.75
     resemblance
    0.75
    Activations Density 0.029%

    No Known Activations