INDEX

Explanations

technical details and specifications mentioned in a document

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

ItemImage

-0.71

perty

-0.67

matically

-0.65

regor

-0.61

cerning

-0.60

akov

-0.60

 Spears

-0.59

APH

-0.59

lav

-0.57

 conduc

-0.56

POSITIVE LOGITS

xff

1.26

xd

0.92

xa

0.92

xe

0.86

xb

0.86

xc

0.85

0.80

MQ

0.80

0.79

Activations Density 1.627%

technical details and specifications mentioned in a document

No Comments

No Known Activations