INDEX

Explanations

highlights or characteristics that define a person, place, or thing

oai_token-act-pair · gpt-3.5-turbo

phrases that question or describe the reasons behind qualities or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

aimon

-0.82

nery

-0.71

Sov

-0.69

duty

-0.67

jury

-0.67

loader

-0.66

earances

-0.65

jab

-0.63

imation

-0.62

nown

-0.61

POSITIVE LOGITS

us

0.96

 them

0.95

him

0.87

me

0.85

 Difference

0.78

 these

0.72

 humanity

0.69

 difference

0.69

 people

0.68

 Amen

0.68

Activations Density 0.072%

highlights or characteristics that define a person, place, or thing

phrases that question or describe the reasons behind qualities or characteristics

No Comments

No Known Activations