INDEX

Explanations

phrases related to blaming others or taking responsibility

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

marked

-0.95

inational

-0.89

intend

-0.85

mark

-0.82

mental

-0.79

edom

-0.79

spect

-0.79

marks

-0.76

mens

-0.76

perties

-0.76

POSITIVE LOGITS

 blame

0.94

 blaming

0.89

 scapego

0.83

 blames

0.80

 blamed

0.76

 lash

0.76

oshop

0.76

 Gateway

0.75

 victim

0.74

 culprit

0.73

Activations Density 9.535%

phrases related to blaming others or taking responsibility

No Comments

No Known Activations