INDEX

Explanations

instances of the word "single" followed by a numeric value or the word "one"

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

eln

-0.78

ello

-0.76

iosyn

-0.73

olas

-0.73

aba

-0.68

enaries

-0.68

 srfAttach

-0.68

HCR

-0.67

ÃĥÃĤ

-0.66

ctr

-0.66

POSITIVE LOGITS

THING

1.22

 imaginable

1.04

 inch

0.96

 conceivable

0.96

WHERE

0.96

 penny

0.96

 thing

0.96

 goddamn

0.92

 ounce

0.91

 facet

0.90

Activations Density 0.030%

instances of the word "single" followed by a numeric value or the word "one"

No Comments

No Known Activations