INDEX

Explanations

phrases related to dedication and loyalty

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

ailable

-0.76

ku

-0.66

umo

-0.64

CTV

-0.63

abet

-0.63

ids

-0.63

inguishable

-0.62

etus

-0.60

DT

-0.60

iver

-0.59

POSITIVE LOGITS

 waivers

0.83

 intermedi

0.82

 channels

0.77

 stages

0.77

 clenched

0.74

 various

0.73

lining

0.69

 phases

0.68

 Explorer

0.67

 disciplines

0.67

Activations Density 3.340%

phrases related to dedication and loyalty

No Comments

No Known Activations