INDEX

Explanations

verbs related to action and observation

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rahman

-0.69

abiding

-0.68

LY

-0.67

audi

-0.65

frames

-0.65

 behav

-0.64

Statement

-0.63

 Beng

-0.63

CI

-0.63

 evid

-0.61

POSITIVE LOGITS

 grips

1.05

asted

0.99

wered

0.97

pload

0.84

otom

0.80

itiveness

0.77

asty

0.77

tted

0.77

iled

0.76

 settle

0.75

Activations Density 10.719%

verbs related to action and observation

No Comments

No Known Activations