INDEX

Explanations

mentions of retirement

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PLIED

-0.78

anon

-0.68

smith

-0.64

 evidenced

-0.64

 Provided

-0.62

sers

-0.62

 eroded

-0.61

 argued

-0.60

 taught

-0.59

kat

-0.59

POSITIVE LOGITS

 intention

1.05

 impending

0.96

 intentions

0.95

 candidacy

0.94

 tentative

0.92

 allegiance

0.92

 plans

0.86

 arrival

0.85

 partnership

0.85

 moratorium

0.84

Activations Density 0.148%

mentions of retirement

No Comments

No Known Activations