INDEX

Explanations

terms related to primary elements or focal points

oai_token-act-pair · gpt-3.5-turbo

references to primary and secondary categories or classifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

estern

-0.80

INESS

-0.77

uge

-0.70

OTO

-0.70

ARDS

-0.68

EEK

-0.67

 Doodle

-0.66

UGE

-0.65

sung

-0.65

Afee

-0.64

POSITIVE LOGITS

 antagonist

0.90

tenance

0.87

 careg

0.85

 objective

0.83

 pivot

0.79

stay

0.78

 distingu

0.78

 beneficiary

0.76

 responsibility

0.76

 distinguishing

0.74

Activations Density 0.012%

terms related to primary elements or focal points

references to primary and secondary categories or classifications

No Comments

No Known Activations