INDEX

Explanations

references to individuals named "Murray" with a particularly strong focus on someone named "Murray" as well as mentions of various incidents involving them

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ulatory

-0.77

ulation

-0.74

umper

-0.73

iaries

-0.73

ptroller

-0.71

--------------------------------------------------------

-0.68

itious

-0.67

opher

-0.66

reek

-0.65

ened

-0.64

POSITIVE LOGITS

cloth

0.74

hyde

0.74

hoe

0.71

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.69

 Ferdinand

0.66

fur

0.63

 Murray

0.62

vy

0.62

wu

0.61

anyahu

0.61

Activations Density 7.135%

references to individuals named "Murray" with a particularly strong focus on someone named "Murray" as well as mentions of various incidents involving them

No Comments

No Known Activations