INDEX

Explanations

technical terms or codes likely related to computer programming or policy regulations

oai_token-act-pair · gpt-3.5-turbo

terms related to legislation and policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

CHR

-0.73

¥µ

-0.71

Cyn

-0.70

 Streets

-0.68

oples

-0.67

borough

-0.66

Psy

-0.64

Urban

-0.62

church

-0.62

myra

-0.61

POSITIVE LOGITS

 inhibitors

0.94

 inhibitor

0.93

 binds

0.85

 activation

0.80

 activ

0.73

 bonded

0.73

 binding

0.72

 enabled

0.71

 blockers

0.70

 intervention

0.70

Activations Density 0.478%

technical terms or codes likely related to computer programming or policy regulations

terms related to legislation and policies

No Comments

No Known Activations