© 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. 8-RES-JB
    11. 781
    Prev
    Next
    INDEX
    Explanations

    names of locations, particularly cities and countries

    oai_token-act-pair · gpt-3.5-turbo

    terms related to organizations or music groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    orer
    -0.71
    fusc
    -0.68
    owicz
    -0.65
    ubis
    -0.62
    roman
    -0.60
    ttp
    -0.60
    ificate
    -0.60
    idae
    -0.59
    ieve
    -0.59
    Redditor
    -0.59
    POSITIVE LOGITS
    VILLE
    1.47
     CITY
    1.24
    FIELD
    1.18
    LAND
    1.12
    VIEW
    1.11
     COUNTY
    1.10
    STON
    1.10
    BUR
    1.07
     SPR
    1.06
     PARK
    1.05
    Activations Density 0.123%

    No Known Activations