INDEX

Explanations

phrases related to decision-making and personal experiences

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

currently

-0.78

 presently

-0.77

Dialogue

-0.70

now

-0.69

 anymore

-0.68

now

-0.67

 currently

-0.66

ethy

-0.65

arta

-0.65

ulum

-0.64

POSITIVE LOGITS

 originally

1.18

 yesterday

1.05

 last

1.03

 earlier

1.02

 previously

0.98

 initially

0.92

hes

0.91

 recently

0.87

wolves

0.87

 unsuccessful

0.82

Activations Density 2.929%

phrases related to decision-making and personal experiences

No Comments

No Known Activations