INDEX

Explanations

words related to discussions or analysis of a topic, possibly related to critics' reviews or evaluations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

interstitial

-0.82

now

-0.79

stan

-0.75

ctl

-0.73

emate

-0.72

raf

-0.71

aka

-0.70

now

-0.69

currently

-0.69

 remem

-0.69

POSITIVE LOGITS

 last

1.20

 yesterday

1.13

 earlier

1.02

 originally

0.98

terday

0.92

âĹ¼

0.91

ago

0.84

 recently

0.81

 previously

0.80

 beforehand

0.78

Activations Density 44.156%

words related to discussions or analysis of a topic, possibly related to critics' reviews or evaluations

No Comments

No Known Activations