INDEX

Explanations

phrases related to actions, situations, and strategies in a narrative context

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

 Expend

-0.71

Atk

-0.69

antine

-0.66

 Mankind

-0.65

 Fist

-0.63

 Ethiopia

-0.62

Ao

-0.61

 Disability

-0.60

address

-0.60

MSN

-0.60

POSITIVE LOGITS

way

1.06

 hardest

0.82

yond

0.81

uchs

0.80

vez

0.80

til

0.79

ocally

0.79

anova

0.78

 furthe

0.77

atically

0.75

Activations Density 7.379%

phrases related to actions, situations, and strategies in a narrative context

No Comments

No Known Activations