INDEX

Explanations

numerical values or statistics related to items or attributes

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

ledged

-0.74

 guiIcon

-0.71

 fireball

-0.62

LINE

-0.60

mma

-0.54

 limp

-0.54

 craw

-0.54

 equals

-0.54

printed

-0.53

 flatt

-0.53

POSITIVE LOGITS

ieri

0.84

beck

0.81

inki

0.81

encia

0.77

uddin

0.77

aceae

0.75

addin

0.74

ahar

0.73

quin

0.72

anamo

0.71

Activations Density 0.064%

numerical values or statistics related to items or attributes

No Comments

No Known Activations