INDEX

Explanations

references to the film industry and related terms

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilities

-0.79

cffff

-0.69

attled

-0.68

RESULTS

-0.67

rian

-0.67

ciating

-0.65

regulated

-0.65

iang

-0.64

tnc

-0.64

ords

-0.64

POSITIVE LOGITS

 theaters

1.08

 premie

1.05

 theater

1.04

goers

1.03

 studios

1.01

 theat

1.01

 adaptation

1.01

 theatre

1.00

 studio

0.98

 films

0.98

Activations Density 0.709%

references to the film industry and related terms

No Comments

No Known Activations