INDEX

Explanations

mentions of staff members, particularly those in certain situations or locations

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

Fem

-0.72

 arous

-0.70

-0.68

Pac

-0.63

ICO

-0.63

 Bout

-0.63

ventions

-0.63

 Roosevelt

-0.62

displayText

-0.61

CEPT

-0.61

POSITIVE LOGITS

member

1.08

 members

1.07

members

1.05

 sergeant

1.00

 member

0.99

atcher

0.88

0.85

ord

0.84

 Writer

0.83

ords

0.81

Activations Density 0.028%

mentions of staff members, particularly those in certain situations or locations

No Comments

No Known Activations