INDEX

Explanations

topics related to critiques and evaluations, such as recognizing positive and negative aspects, noteworthy elements, and assessments of value or potential

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

ancies

-1.09

ensions

-0.99

rams

-0.96

ignt

-0.95

atism

-0.93

alks

-0.92

ells

-0.89

 timelines

-0.85

iencies

-0.85

landers

-0.85

POSITIVE LOGITS

 unto

1.04

 reel

1.01

 magnet

0.92

 starter

0.89

 breaker

0.88

 keeper

0.86

 opener

0.86

 reliever

0.85

 deterrent

0.84

 distraction

0.84

Activations Density 13.307%

topics related to critiques and evaluations, such as recognizing positive and negative aspects, noteworthy elements, and assessments of value or potential

No Comments

No Known Activations