INDEX

Explanations

phrases related to advice on personal behavior or actions

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

Negative Logits

 DRAG

-0.64

haps

-0.62

FER

-0.60

fit

-0.59

çīĪ

-0.57

apest

-0.56

 converter

-0.56

repre

-0.56

CAST

-0.56

ouf

-0.55

POSITIVE LOGITS

 indefinitely

1.29

 until

1.05

 forever

1.00

 lest

0.94

 till

0.94

 uninterrupted

0.94

 intact

0.91

til

0.89

 whilst

0.89

lishes

0.86

Activations Density 0.226%

phrases related to advice on personal behavior or actions

No Known Activations