INDEX

Explanations

relationships and associations between variables, particularly focusing on causation and correlations

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

imm

-0.57

anas

-0.57

sw

-0.56

mbuds

-0.56

scrib

-0.55

wagen

-0.55

thur

-0.54

stall

-0.54

 Citiz

-0.54

hoff

-0.54

POSITIVE LOGITS

 coefficient

1.14

 coefficients

1.02

SHIP

0.83

 correlation

0.80

 correlate

0.72

 negatively

0.71

 correlations

0.70

 positively

0.67

 linking

0.65

abolic

0.64

Activations Density 11.749%

relationships and associations between variables, particularly focusing on causation and correlations

No Comments

No Known Activations