INDEX

Explanations

words related to assuming or belief in a context of statements and descriptions

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

psey

-0.84

gom

-0.83

cgi

-0.80

sung

-0.76

turn

-0.76

waters

-0.76

agra

-0.75

HCR

-0.75

encer

-0.73

odes

-0.71

POSITIVE LOGITS

 assume

0.80

 assumed

0.77

 infall

0.76

 ownership

0.74

LY

0.74

 incorrectly

0.73

ially

0.72

UME

0.71

olate

0.70

 Cortana

0.69

Activations Density 12.117%

words related to assuming or belief in a context of statements and descriptions

No Comments

No Known Activations