INDEX

Explanations

words related to causing action or influence

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

FACE

-0.69

ALD

-0.66

ITIES

-0.66

 Antiqu

-0.63

Ô

-0.63

ļéĨĴ

-0.61

HCR

-0.61

 Clyde

-0.60

zan

-0.59

PORT

-0.59

POSITIVE LOGITS

igators

1.31

igator

1.21

igated

1.12

igation

1.09

igating

1.03

iated

0.97

igate

0.97

ipl

0.92

rave

0.91

iation

0.89

Activations Density 0.018%

words related to causing action or influence

No Comments

No Known Activations