INDEX

Explanations

terms related to geology and medical-related terms with a focus on psychological and humanitarian aspects;

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

tower

-0.92

lain

-0.87

lessly

-0.83

hov

-0.81

ortium

-0.79

etts

-0.79

ered

-0.78

xual

-0.78

ertodd

-0.76

plin

-0.76

POSITIVE LOGITS

ities

1.21

ized

1.17

ization

1.17

ity

1.16

izing

1.09

 significance

1.09

ised

1.05

 prowess

1.02

 impossibility

1.01

isation

1.01

Activations Density 2.218%

terms related to geology and medical-related terms with a focus on psychological and humanitarian aspects;

No Comments

No Known Activations