© 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. 1-RES-JB
    11. 16338
    Prev
    Next
    INDEX
    Explanations

    Roman numerals, specifically the number "III"

    oai_token-act-pair · gpt-3.5-turbo

    references to the Roman numeral "III," indicating the third in a series or a number of important entities or events associated with that designation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    poll
    -0.76
    pal
    -0.70
    organ
    -0.69
    \":
    -0.66
    ãĤ±
    -0.65
    ãĥ¡
    -0.65
    arching
    -0.65
    medium
    -0.65
    machine
    -0.65
    Station
    -0.64
    POSITIVE LOGITS
     III
    1.30
    III
    1.30
    II
    1.10
     IV
    0.87
     XIII
    0.84
     II
    0.83
     Jr
    0.82
    abba
    0.81
     VIII
    0.80
    rd
    0.78
    Activations Density 0.006%

    No Known Activations