© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4964
    Prev
    Next
    INDEX
    Explanations

    references to data structures and their operations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    332
    +0.09
    0.4%
    551
    +0.08
    0.4%
    336
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    237
    +0.09
    0.01
    568
    +0.08
    0.01
    316
    +0.08
    0.01
    Negative Logits
     lockout
    -0.64
     Benny
    -0.64
     traveller
    -0.61
     Steelers
    -0.60
    dayName
    -0.59
     robbers
    -0.58
    balls
    -0.58
     Lob
    -0.58
     Vanderbilt
    -0.58
     Bulg
    -0.58
    POSITIVE LOGITS
    ide
    0.98
    isions
    0.93
    imal
    0.92
    arian
    0.82
    ided
    0.80
    apor
    0.80
    ault
    0.79
    ority
    0.78
    idem
    0.77
    iform
    0.76
    Activations Density 0.027%

    No Known Activations