© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 5379
    Prev
    Next
    INDEX
    Explanations

    text related to characters and events featuring a strong and powerful figure referred to as "Garg" or "Gig" with a particular focus on specific traits and details about this character's actions and surroundings

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    587
    +0.05
    0.6%
    544
    +0.04
    0.5%
    465
    +0.04
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.05
    0.07
    326
    +0.04
    0.05
    48
    +0.04
    0.04
    Negative Logits
    xual
    -0.34
    66666666
    -0.28
    amac
    -0.28
    merce
    -0.28
    sonian
    -0.26
    arya
    -0.26
    road
    -0.25
    esty
    -0.25
    owa
    -0.24
    nington
    -0.23
    POSITIVE LOGITS
    antes
    0.29
    ante
    0.28
    abyte
    0.27
    ente
    0.27
    entric
    0.26
    abytes
    0.26
    oyle
    0.26
    inas
    0.24
    gent
    0.24
     pigeon
    0.22
    Activations Density 0.437%

    No Known Activations