INDEX

Explanations

words related to uncertainty or degrees of certainty

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

ires

-0.76

ocracy

-0.69

 Pony

-0.68

ETS

-0.66

ocratic

-0.63

ocrats

-0.60

efer

-0.60

estro

-0.59

 Wast

-0.57

eval

-0.57

POSITIVE LOGITS

chard

1.61

acle

1.49

acles

1.48

Else

1.46

chid

1.41

nam

1.40

ifice

1.39

acular

1.33

 alternatively

1.32

 else

1.21

Activations Density 1.052%

words related to uncertainty or degrees of certainty

No Comments

No Known Activations