© 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. 0-RES-JB
    11. 6363
    Prev
    Next
    INDEX
    Explanations

    the word "convergence" or its variations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "converge" and its derivatives, indicating discussions about merging or gathering

    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
    SHIP
    -0.76
     exemption
    -0.74
    hole
    -0.73
    onym
    -0.71
     KH
    -0.67
     ministry
    -0.64
     nickname
    -0.64
    prison
    -0.64
    bara
    -0.64
     Rita
    -0.64
    POSITIVE LOGITS
     conver
    4.20
     converge
    3.02
     convergence
    2.43
     Conver
    2.43
     converter
    1.97
     coales
    1.56
     dove
    1.17
     diver
    1.16
     divergence
    1.07
     conven
    1.02
    Activations Density 0.040%

    No Known Activations