INDEX

Explanations

phrases indicating readiness or willingness to take action

oai_token-act-pair · gpt-4o-mini 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

AMA

-0.78

VERTISEMENT

-0.76

adish

-0.71

aples

-0.69

MRI

-0.66

eatures

-0.64

uo

-0.63

ophone

-0.61

 Gems

-0.61

tein

-0.58

POSITIVE LOGITS

 enough

0.82

nesses

0.75

ptin

0.74

to

0.72

 gladly

0.72

tc

0.68

going

0.68

 trusting

0.68

tarian

0.67

£ı

0.66

Activations Density 0.043%

phrases indicating readiness or willingness to take action

No Comments

No Known Activations