© 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. 2883
    Prev
    Next
    INDEX
    Explanations

    phrases related to a specific mythical creature called "Hoopa"

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific entity or subject, particularly involving "opa"

    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
     fits
    -0.76
     Coh
    -0.72
     fit
    -0.69
     lashes
    -0.66
     Reilly
    -0.63
     explodes
    -0.63
     seams
    -0.62
     âĩ
    -0.60
    atern
    -0.60
     Blades
    -0.60
    POSITIVE LOGITS
    opa
    4.98
    OPA
    3.07
    uca
    1.29
    OSE
    1.11
    idon
    1.11
    agos
    0.97
    aquin
    0.96
    odi
    0.95
    opal
    0.94
    appa
    0.94
    Activations Density 0.027%

    No Known Activations