INDEX

Explanations

words related to strong negative emotions, especially anger or frustration

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

soDeliveryDate

-0.66

Nos

-0.57

çĦ

-0.56

ãģ¯

-0.55

ãĤ©

-0.54

zzo

-0.53

adr

-0.53

 Vert

-0.53

soType

-0.52

phies

-0.51

POSITIVE LOGITS

rid

0.77

agascar

0.69

mad

0.61

iated

0.60

ly

0.59

ulously

0.58

cia

0.58

 dash

0.58

owl

0.57

 maniac

0.57

Activations Density 4.290%

words related to strong negative emotions, especially anger or frustration

No Comments

No Known Activations