INDEX

Explanations

comments or evaluations from critics related to various media or entertainment products

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

Ĩ

-0.94

aires

-0.77

croft

-0.76

aneous

-0.75

TM

-0.75

 dearly

-0.75

Tx

-0.73

udos

-0.73

RAL

-0.73

eur

-0.73

POSITIVE LOGITS

vernment

0.88

Â·Â·

0.85

vironment

0.85

ailable

0.83

 Sabb

0.81

ilities

0.79

ascus

0.78

merce

0.78

/-

0.76

minded

0.76

Activations Density 0.320%

comments or evaluations from critics related to various media or entertainment products

No Comments

No Known Activations