Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 6-RES_POST_32K-OAI
    6. 4208
    Prev
    Next
    INDEX
    Explanations

    mentions of starting positions or starting roles in sports contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    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
    735
    +0.09
    0.4%
    21
    +0.09
    0.4%
    421
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    62
    +0.09
    0.02
    122
    +0.09
    0.03
    182
    +0.09
    0.02
    Negative Logits
    ably
    -1.00
    ability
    -0.94
    ingly
    -0.89
    edy
    -0.84
    abilities
    -0.83
    efully
    -0.82
    eg
    -0.81
    able
    -0.79
    jee
    -0.78
    ily
    -0.77
    POSITIVE LOGITS
     salaries
    0.81
     XI
    0.79
     eleven
    0.76
     principle
    0.69
     pitcher
    0.69
     rosters
    0.67
     point
    0.67
     assumption
    0.66
     Wage
    0.66
     salary
    0.66
    Activations Density 0.177%

    No Known Activations