INDEX

Explanations

specific words related to rooms like "room", "roommate", "roomm," and "rooming."

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

HER

-1.00

indal

-0.93

ÂłÂłÂłÂłÂłÂłÂłÂł

-0.90

elson

-0.89

usterity

-0.89

 Accessed

-0.87

Uz

-0.86

TAG

-0.85

FF

-0.83

RBI

-0.82

POSITIVE LOGITS

room

1.52

rooms

1.39

 doors

1.34

 rooms

1.22

 upstairs

1.09

stairs

1.09

 room

1.08

 floor

1.06

idges

1.05

clock

1.03

Activations Density 0.576%

specific words related to rooms like "room", "roommate", "roomm," and "rooming."

No Comments

No Known Activations