INDEX

Explanations

references to specific terms or abbreviations related to health or safety

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ately

-0.76

ators

-0.74

istically

-0.73

uary

-0.71

osity

-0.70

ator

-0.69

uably

-0.69

ãĥ¼ãĥĨ

-0.69

atives

-0.67

naire

-0.67

POSITIVE LOGITS

TPS

1.38

TL

0.98

TY

0.91

RA

0.90

EN

0.90

LV

0.90

BUR

0.86

ECH

0.85

PC

0.85

ERSON

0.82

Activations Density 0.005%

references to specific terms or abbreviations related to health or safety

No Comments

No Known Activations