INDEX

Explanations

locations related to traveling or movement

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

 Nielsen

-0.65

 consistency

-0.62

Trend

-0.62

 durability

-0.59

gey

-0.59

 centr

-0.59

aska

-0.57

arth

-0.57

BIP

-0.56

worthiness

-0.55

POSITIVE LOGITS

 unaccompanied

0.83

 whence

0.82

 refuel

0.82

 greet

0.75

ezvous

0.72

 expecting

0.72

 where

0.68

 refurb

0.68

Ô

0.67

 smelling

0.66

Activations Density 4.120%

locations related to traveling or movement

No Comments

No Known Activations