INDEX

Explanations

information regarding facts or knowledge about a topic

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

 mates

-0.63

 Apostles

-0.62

 coincides

-0.62

 contrace

-0.60

 Perse

-0.60

 carriage

-0.59

Pse

-0.58

bies

-0.58

 Ples

-0.58

 Advertisement

-0.58

POSITIVE LOGITS

 worth

0.87

 happening

0.84

ername

0.84

needed

0.82

ģ«

0.79

ŃĶ

0.79

aker

0.78

ritten

0.77

 underway

0.77

hes

0.76

Activations Density 12.464%

information regarding facts or knowledge about a topic

No Comments

No Known Activations