INDEX

Explanations

specific phrases with quantifying language like "few" or "several."

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

 Akin

-0.81

matter

-0.72

 Amon

-0.64

 Ezek

-0.64

complex

-0.63

 Dirk

-0.60

 Quantity

-0.60

borough

-0.60

Emb

-0.59

Lov

-0.59

POSITIVE LOGITS

 quir

0.68

 creations

0.67

 favorites

0.66

 finalists

0.65

 additions

0.63

 manifestations

0.62

quished

0.62

 inspir

0.61

 remaining

0.61

 contenders

0.61

Activations Density 9.264%

specific phrases with quantifying language like "few" or "several."

No Comments

No Known Activations