INDEX

Explanations

phrases related to the concept of 'usual suspects'

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

mented

-0.79

acus

-0.79

mentation

-0.77

kamp

-0.76

OPE

-0.74

asus

-0.68

enced

-0.68

Reviewed

-0.67

formerly

-0.66

ole

-0.65

POSITIVE LOGITS

 suspects

1.25

 caveats

1.16

 disclaimer

1.14

 disclaim

1.03

 assortment

1.03

 caveat

0.89

 tropes

0.87

 wisdom

0.83

 deviations

0.82

 refrain

0.82

Activations Density 0.057%

phrases related to the concept of 'usual suspects'

No Comments

No Known Activations