INDEX

Explanations

phrases related to giving or providing

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

ansas

-0.71

Sphere

-0.67

can

-0.63

tenance

-0.61

lander

-0.57

timer

-0.57

hetto

-0.57

ategy

-0.57

uder

-0.56

 Incarn

-0.55

POSITIVE LOGITS

 rise

1.02

 birth

0.99

aways

0.96

 refunds

0.85

 chase

0.85

us

0.83

away

0.82

 generously

0.77

AW

0.74

 precedence

0.74

Activations Density 0.468%

phrases related to giving or providing

No Comments

No Known Activations