INDEX

Explanations

phrases related to persistence or continuity

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

insula

-0.91

ãĥ³ãĤ¸

-0.76

psey

-0.69

Pause

-0.68

emis

-0.66

Winter

-0.65

ESE

-0.65

stice

-0.65

 Inventory

-0.62

 Hunts

-0.60

POSITIVE LOGITS

 technically

0.86

 manages

0.80

 intact

0.80

birth

0.78

 usable

0.77

 theoretically

0.75

 managed

0.75

 retains

0.72

 somehow

0.71

 viable

0.71

Activations Density 0.050%

phrases related to persistence or continuity

No Comments

No Known Activations