INDEX

Explanations

phrases related to visual observation and experience

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

markets

-0.68

Newsletter

-0.66

ju

-0.63

thumbnails

-0.63

current

-0.62

 minimized

-0.60

nai

-0.59

redients

-0.59

Documents

-0.58

still

-0.57

POSITIVE LOGITS

 anything

0.89

 anybody

0.88

nor

0.81

 dime

0.79

 anyone

0.79

 anywhere

0.79

azo

0.75

 EVER

0.70

essage

0.69

anything

0.69

Activations Density 0.111%

phrases related to visual observation and experience

No Comments

No Known Activations