© 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. 14188
    Prev
    Next
    INDEX
    Explanations

    mentions of TV shows, channels, or broadcasts

    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
    167
    +0.20
    0.7%
    360
    +0.17
    0.6%
    288
    +0.17
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    76
    +0.20
    0.05
    725
    +0.17
    0.04
    360
    +0.17
    0.05
    Negative Logits
    xual
    -1.24
    ussen
    -1.23
    wegian
    -1.01
    20439
    -0.99
    quez
    -0.98
    hammad
    -0.97
    forth
    -0.93
    heny
    -0.91
    ctuary
    -0.91
    ortium
    -0.91
    POSITIVE LOGITS
    trop
    1.31
    ILLE
    1.30
    plays
    1.29
     Trop
    1.22
     Shows
    1.20
    izoph
    1.19
    NZ
    1.12
    Guide
    1.08
     Everywhere
    1.07
    levision
    1.07
    Activations Density 0.509%

    No Known Activations