INDEX

Explanations

references to various types of limits

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

instances of limits or restrictions

oai_token-act-pair · gpt-3.5-turbo

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

ublic

-0.79

rious

-0.78

earth

-0.67

alky

-0.66

nian

-0.65

Corn

-0.65

event

-0.64

Cel

-0.64

story

-0.63

Ń

-0.63

POSITIVE LOGITS

 limits

1.07

 restricts

0.93

 limit

0.93

 restraints

0.85

 restriction

0.83

 amounts

0.81

 withdrawals

0.79

 limiting

0.78

 amount

0.78

lessness

0.77

Activations Density 0.013%

references to various types of limits

instances of limits or restrictions

No Comments

No Known Activations