INDEX

Explanations

references to historical events and literature

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

hesda

-0.88

ourcing

-0.78

ONSORED

-0.77

nels

-0.73

hunt

-0.72

bucks

-0.71

FML

-0.70

roo

-0.70

agara

-0.69

seekers

-0.69

POSITIVE LOGITS

 aloud

1.34

 manuscript

1.24

 poems

1.22

 autobiography

1.19

 papers

1.18

 memoir

1.18

 poem

1.15

 books

1.15

 journals

1.14

 screenplay

1.14

Activations Density 1.163%

references to historical events and literature

No Comments

No Known Activations