INDEX

Explanations

terms related to the concept of 'clearance' or 'being cleared'

oai_token-act-pair · gpt-3.5-turbo

the concept of something being authorized or made permissible

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

egal

-0.78

ENTS

-0.77

Bey

-0.74

ENC

-0.68

lav

-0.68

ingo

-0.68

INA

-0.67

 Kardash

-0.66

ENT

-0.65

ADD

-0.65

POSITIVE LOGITS

 clears

1.15

 cleared

0.99

 clearing

0.88

cloth

0.81

houses

0.81

0.76

 Clear

0.72

utics

0.71

housing

0.71

stocks

0.71

Activations Density 0.017%

terms related to the concept of 'clearance' or 'being cleared'

the concept of something being authorized or made permissible

No Comments

No Known Activations