INDEX

Explanations

keywords related to balance and neutrality

oai_token-act-pair · gpt-3.5-turbo

references to balance in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.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

OLOG

-0.77

ABE

-0.75

TN

-0.74

clips

-0.73

olog

-0.70

Assembly

-0.68

ACH

-0.67

uber

-0.64

 Offic

-0.64

Jere

-0.63

POSITIVE LOGITS

balanced

1.07

 balanced

0.98

 imbalance

0.95

balance

0.91

 balance

0.90

 balancing

0.86

 equilibrium

0.82

 balances

0.80

 Balanced

0.76

ament

0.73

Activations Density 0.013%

keywords related to balance and neutrality

references to balance in various contexts

No Comments

No Known Activations