INDEX

Explanations

mentions of additional or special features related to products or services

oai_token-act-pair · gpt-3.5-turbo

mentions of additional features or bonuses

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

iard

-0.81

SHIP

-0.72

izont

-0.69

Trend

-0.68

Cola

-0.68

ŃĶ

-0.67

 Capital

-0.67

NING

-0.66

Mos

-0.66

GP

-0.66

POSITIVE LOGITS

 extras

1.22

ordinary

0.95

 goodies

0.94

visor

0.82

verted

0.81

alon

0.74

ervatives

0.72

henko

0.71

liga

0.70

 Extras

0.70

Activations Density 0.009%

mentions of additional or special features related to products or services

mentions of additional features or bonuses

No Comments

No Known Activations